The Cuckoos

Bird Families of the World A series of authoritative, illustrated handbooks of which this is the 15th volume to be published

Series editors ROBERT B. PAYNE Chief editor MICHAEL D. SORENSON KAREN KLITZ JOHN MEGAHAN Robert Payne is an ornithologist at the University of Michigan, where he is Professor of Zoology in the Department of Ecology and Evolutionary Biology, and Curator of Birds in the Museum of Zoology. Earlier ornithological positions include two years postdoc experience at the Fitzpatrick Institute of Ornithology at the University of Cape Town, in a study of African brood-parasitic cuckoos and finches, and three years teaching at the University of Oklahoma in the Great Plains. He is interested in parental care and brood parasitism of birds, evolution of the brood-parasitic finches and cuckoos and their hosts, nestling mimicry of the parental hosts by the brood parasites, behavioral imprinting and how this may lead to the origin of new species of brood-parasitic birds, and the biology of bird song. Over time his interests have developed through a mix of travel to watch birds in the field, experimental studies to ask questions about behavior development, and systematics using the museum collections. He has about 190 publications on birds, many on brood parasitism and population biology. He works with others who work well with the molecular systematics of birds. In field research he has studied the variation of the broodparasitic finches and their host species associations in Africa, and after this book is completed he plans to travel with his wife Laura to observe cuckoos and record their songs in the field. Karen Klitz began painting birds in Africa in 1965. She has been a staff illustrator in the Museums of Zoology and Paleontology at the University of Michigan and is currently Principal Illustrator for the Museum of Vertebrate Zoology at the University of California, Berkeley.

Bird Families of the World 1. The Hornbills Alan Kemp 2. The Penguins Tony D.Williams 3. The Megapodes Darryl N. Jones, René W. R. J. Dekker, and Cees S. Roselaar 4. Fairy-wrens and Grasswrens Ian Rowley and Eleanor Russell 5. The Auks Anthony J. Gaston and Ian L. Jones 6. The Birds of Paradise Clifford B. Frith and Bruce Beehler 7. The Nightjars and their Allies D.T. Holyoak 8. Toucans, Barbets and Honeyguides Lester L. Short and Jennifer F. M. Horne 9. Ratites and Tinamous S. J. J. F. Davies 10. The Bowerbirds Clifford B. Frith and Dawn W. Frith 11. Albatrosses and Petrels across the World Michael Brooke 12. The Grebes Jon Fjeldså 13. The Hawaiian Honeycreepers H. Douglas Pratt 14. Herons James A. Kushlan and James A. Hancock 15. The Cuckoos Robert B. Payne 16. Ducks, Geese and Swans Edited by Janet Kear

Bird Families of the World

The Cuckoos Robert B. Payne with a molecular genetic analysis of cuckoo phylogeny by

Michael D. Sorenson and Robert B. Payne Color plates by

Karen Klitz Black and white illustrations by

John Megahan

Great Clarendon Street, Oxford OX2 6DP Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide in Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Taipei Toronto Shanghai With offices in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan South Korea Poland Portugal Singapore Switzerland Thailand Turkey Ukraine Vietnam Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries Published in the United States by Oxford University Press Inc., New York Illustrations © Oxford University Press, 2005 Text © Robert B. Payne except Text of Molecular Systematics Chapter © Michael D. Sorenson and Robert D. Payne, 2005 The moral rights of the authors have been asserted Database right Oxford University Press (maker) First published 2005 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, or under terms agreed with the appropriate reprographics rights organization. Enquiries concerning reproduction outside the scope of the above should be sent to the Rights Department, Oxford University Press, at the address above You must not circulate this book in any other binding or cover and you must impose this same condition on any acquirer A catalogue record for this title is available from the British Library British Library Cataloging in Publication Data Data available ISBN 0 19 850213 3 10 9 8 7 6 5 4 3 2 1 Typeset by Macmillan India Limited Printed in China

Acknowledgments

My father Russell Payne and Sao Saimong Mongrai introduced me to cuckoos in the Shan States in Burma in 1951, where I watched cuckoos at Taunggyi, a site of classic cuckoo studies in the 1930s. Undergraduate work at the University of Michigan Museum of Zoology and a talk by Herbert Friedmann on brood-parasitic birds at Cranbrook Institute of Science were important in turning attention toward the cuckoos. Graduate work at the Museum of Vertebrate Zoology, University of California, Berkeley, supported a field study of broodparasitic cowbirds. A National Science Foundation postdoctoral research fellowship in 1965 allowed two years in the field with brood-parasitic cuckoos in Africa, with the help of Herbert Friedmann and the Percy FitzPatrick Institute of African Ornithology at the University of Cape Town, with fieldwork in South Africa, Botswana, Rhodesia, Zambia, Malawi and Kenya. In southern and east Africa, Karen Klitz and I worked with cuckoos in the 1960s, guided by Jack Winterbottom, Richard Liversidge, Jack Skead and Gordon Ranger in South Africa, Michael Irwin in Zimbabwe, and John Williams in Kenya. My wife Laura Payne and I watched Neotropical cuckoos in Gayana and Curaçao, we next shared fieldwork with cuckoos and fairy-wrens in the 1980s in Western Australia with the CSIRO research team of Ian Rowley, Eleanor Russell, Michael Brooker and Graeme Chapman on Gooseberry Hill, and with Ian and Eleanor and Dick and Molly Brown in the Manjimup forests. Later we worked with Nedra Klein in Cameroon, Mike Sorenson in Zimbabwe and Guinea, Pierre Reynaud in Senegal, Clive Barlow in Guinea, Senegal, Gambia and Mali, Ben King in India, Phil Round in Thailand,Wang Luan Keng

and Lim Kim Keang in Malaysia, Ir Darjono in Java, and Kathy Groschupf in Arizona. In other field studies I worked with Geoffrey Field in Sierra Leone, Chris Risley in Ghana, Kathy Groschupf in Cameroon, and Roger Wilkinson, Neville Skinner and Mark Hopkins in Nigeria. In North America the University of Michigan Biological Station, the Matthai Botanical Gardens, and the E. S. George Reserve of the University of Michigan provided field support, as did the Oklahoma University Biological Station at Willis, Oklahoma. The second phase of work was examining the specimens in natural history museums.This gave a closer look at cuckoos than the birds themselves ever allowed in the field. The curators of these valuable collections allowed me to examine specimens during research visits or through loans. For permission to examine specimens and information from their collections, I thank the curators and staff of the following museums and live bird collections: Brian Gill (AM,Auckland); Dean Amadon, George Barrowclough, Joel Cracraft, Wesley Lanyon, Mary LeCroy, Pamela Beresford, Peter Capainolo, Shannon Kenney, Chris Milensky, Chris Vogel and Paul Sweet (AMNH, New York); David Agro and Leo Joseph (ANSP, Philadelphia); Robert PrysJones, Michael Walters, Frank Steinbrenner and Mark Adams (BMNH,Tring); Carla Kishinami (B. P. Bishop Museum, Honolulu); Kit Hustler and Michael Irwin (BWYO, Bulawayo); Phil Bruner (BYU, Laie, Hawaii); Lei Fu-min (CAS, Beijing); Ken Parkes, Brad Livezey and Robin Panza (CM, Pittsburgh); Robert S. Kennedy (CMNH, Cincinnati); Michael Brooke (Cambridge University Museum); Kevin McGowan and Charles Dardia

vi Acknowledgments (CU, Ithaca); Mike Brooke (CUM, Cambridge); David Allan and Philip Clancey (DM, Durban); Gene Hess and Jean Woods (DMNH, Wilmington); Dave Willard, John Bates, Shannon Hackett, Melvin Traylor, Peter Lowther and Amy Driskell (FMNH, Chicago); Dave Steadman and Tom Webber (FLMNH, Gainesville); René Verheyen (IRSNB, Brussels), Bob Zink, Scott Lanyon and John Klicka ( JFBM, Minneapolis); C. J. Skead (Kaffrarian Museum, King William’s Town); H. Schouteden, Anton De Roo and Michel Louette (KMMA, Tervuren); Rick Prum and Mark Robbins (KU, Lawrence); Donna Dittmann, Steve Cardiff, Rob Moyle, Van Remsen and Fred Sheldon (LSU, Baton Rouge); Karl Schuchmann, Renate van den Elzen and Hans Wolters (ZFMK, Bonn); Eulàlia Garcia Franquesa (MCNB, Barcelona); Doug Causey, Ray Paynter and Alison Pirie (MCZ, Cambridge); Robin Restall (MHNLS); Pavel Tomkovich (MMZ, Moscow); Christian Erard and C. Voisin (MNHN, Paris); Martin Nhlane (MNM, Blantyre); Giuliano Doria, Roberto Poggi and Enrico Borgo (MSNG, Genoa); Giorgio Chiozzi and Carlo Violani (MSNM, Milan); Laura Abraczinskas, Barb Lundrigan and Pamela Rasmussen (MSU, East Lansing); Ned Johnson, Carla Cicero and Karen Klitz (MVZ, Berkeley); S. Somadikarta and Ir Darjono (MZB Cibinong); Marta Poggesi (MZUS, Firenze); Tony Parker (NMGL, Liverpool); Chum Cunningham-van Someren (NMK, Nairobi); Rodolfo A. Caberoy (NMP, Manila); Ernst Baurenfeind and Herbert Schifter (NMW, Wien); Jarunjin Nabhitabhata (NSM, Bangkok); Gary Schnell (OU); Robin Restall (POC); Steve van Dyke (QM, Brisbane); René Dekker, Gerlof Mees and Peter Van Dam (RMNH, Leiden); Jon Barlow and Jim Dick (ROM, Toronto); John M. Winterbottom (South African Museum, Cape Town); Gerald Mayr (SMF, Frankfurt); Siegfried Eck (SMTD, Dresden), Maklarin Bin Lakim (SP, Kinabalu); James Mueller (SRSU); Keith Arnold (TCWC, College Station, Texas); Lakkana Pakkarnseree (TISTR, Bangkok); Fritz Hertel (UCLA, Los Angeles); Bob Dickerman (UNM, Albuquerque); Gary Voelcker (UNLV, Las Vegas); Mike Braun, Phil Angle, James Dean, Gary Graves, Helen James, Storrs Olson, Pamela Rasmussen and

Dick Zusi (USNM,Washington, D.C.); Scott Cutler (UTEP); Scott Edwards, Sievert Rohwer,Wang Luan Keng, Chris Wood, Gary Voelcker, Robert Faucett, Sharon Birks and Sergei Drovetski (UWBM, Seattle); Dieter Rinke (VW, Walsrode); Glen Storr (WAM, Perth); René Corrado (WFVZ, Camarillo); Jacques Gautier, Paul Whitfield and Kristof Zyskowski (YPM, New Haven); Juergen Fiebig (ZMB, Berlin); Jon Fjeldså (ZMUC, Copenhagen); Yang Chang Man, Darren C. J. Yeo and Lua Hui Keng (ZRC, Singapore); Josef Reichholf (ZSM, München); Kees Roselaar (ZMA,Amsterdam); and Pavel Tomkovich (ZMM, Moscow). Tom Huels hand-reared young roadrunners for developmental studies. Michael Walters measured eggs at BMNH. Dan Gaebel, Karen Kevelighan, Jeff Rebitzke, Sharon Reske and Angie Steffen, undergraduate students at the University of Michigan, examined skeletal specimens with me. George Kulesza made available the measurements of brain size in cuckoos from his comparative brain measurements from skeletons in UMMZ. Carlo Violani was helpful in arranging visits to natural history museums in Italy and in locating specimens and publications. Godfrey Symons gave access to photograph bird eggs in his collection; the collection is now in the Cathcart Museum, South Africa, and Jean Woods at DMNH loaned bird eggs to use in calibrating egg colors in the photographs. For photographs of museum specimens, I thank Patricia Escalante Pliego and Adolfo G. Navarro S. (Colección Nacional de Aves, Dept. Zoología, Instituto de Biología, UNAM, Mexico City), Leslie Jessop (NEWHM), Michel Louette (KMMA), Maklarin Bin Lakim (Sabah Parks), Ir Darjono (MZB) and Lei Fu-min (CAS), and for photographs of live cuckoos I thank Graeme Chapman, Carla Fontana, Johannes Foufopoulos, Karen Klitz, Mary LeCroy, Dale Lewis, Regina Macedo, Richard Peek, Cheong Weng Chuan and Kit Hustler, Ron Orenstein, Ole Post, Ong Kiem Sian, Trevor Price, Carol Ralph, Cheong Wan Chun, Pamela Rasmussen, Phil Round, Joe Strauch, Bob Thornton, Sandra Vehrencamp, Wang Luan Keng, Jongmin Yoon and Moo-Boo Yoon. Rob Moyle and Carl Vernon provided their photographs of forest habitat which we adapted into the background of the color plates.The British Library National Sound Archive (NSA, Richard

Acknowledgments vii Ranft), Cornell Library of Natural Sounds (LNS), Sandra Gaunt (OSU), Per Alström, Clive Barlow, Graeme Chapman, Jared Diamond, Chris Filardi, Lei Fumin, Paul Holt, Bob Kennedy, Ben King, Lindy McBride, Barry Morgan, Pamela Rasmussen, Phil Round, Pratap Singh, Deepal Warakagoda and Martha Whitson gave descriptions of songs and copies of their tape and digital recordings of cuckoo songs. Colleagues at the University of Michigan, including Lei Fumin, Janet Hinshaw, George Kulesza, David Mindell, John Mitani, Phil Myers, Robert Storer, Jean Woods and Tamaki Yuri, were helpful throughout the study. Material for molecular genetics analysis was provided by several museums and collections (AMNH, ANSP, BMNH, CM, CMNH, FMNH, LSU, QM, AM, DM, DMNH, MCZ, MNM, MVZ, RMNH, SMNS, UMMZ, UMMZ, USNM, UWBM, VW, WFVZ, YPM and ZMUC), and by Dinesh Bhatt (Gurukul Kangri University, Haridwar), Steve Latta (University of Missouri), Terry Root (Ann Arbor), Steve Schneider (Palo Alto) and Lei Fu-min (CAS, Beijing). Preliminary genetic analyses were conducted in David Mindell’s laboratory at the University of Michigan Museum of Zoology, for which the University of Michigan supported funding for DNA sequencing. Suzanne Ambs, Marina Ramon, Tracy Heath, Kristina Sefc and Elen Oneal assisted with the processing of DNA sequencing reactions at Boston University. Funding for DNA sequencing and a computer cluster used for the phylogenetic analyses were provided by Boston University. For bibliographic help I thank Joann Constantinides, Janet Hinshaw, Charlene Stachnick and Dorothy Riemenschneider (University of Michigan, Ann Arbor), Linda Birch (Alexander Library, EGI, Oxford University), Effie Wahr (BMNH, Tring), Mary Petersen (ZMUC, Copenhagen), Renate van den Elzen (ZFMK, Bonn) and the librarians of ZSM (Munich). David Seel of Bangor guided me to Pound Green Common, Worcestershire, where Edgar Chance observed cuckoos in the 1920s. Carla Fontana, Lydia Skrynnikova and Tamaki Yuri translated works in Portuguese, Russian and Japanese; and Ir Darjono, James Dean, Antonia Gorog and Paul Taylor interpreted the museum specimen labels of birds taken in Indonesia. David Allan, Wayne Arendt, Walter Boles, Mariana

Cariello, Mario Cohn-Haft, Nigel Collar, Jack Cox, Carla Fontana, Larry Heaney, Regina Macedo, Sergio Posso, Pamela Rasmussen, Danny Rogers, Burkhard Stephan and Robert L. Sutton kindly sent original data, observations and copies of their papers, and Burkhard Stephan allowed us to use his drawings of cuckoo wings. Clive Barlow, Pamela Beresford, Andy Berger, Bonnie Bowen, Michael Brooker, John Colebrook-Robjent, Richard Dean, Edward Dickinson, Jon Fjeldså, Steve Goodman, Kathy Groschupf, Bill Hamilton, Hiroyoshi Higuchi, Tom Huels, Kit Hustler, Klaus Immelmann, Bob Kennedy, Jiro Kikkawa, J. M. (Martjan) Lammetink, Mary LeCroy, Regina Macedo, Gerald Mauersberger, Ernst Mayr, Sergio Posso, Pamela Rasmussen, Manolo Soler, Richard Thorington, Carl Vernon, Carlo Violani, Michael Walters and Amotz Zahavi offered discussion during the study. For comments on the manuscript I thank Bonnie Bowen, Richard Dean, Edward Dickinson, Kathy Groschupf, Ben King, Karen Klitz, George Kulesza, David Lahti, Regina Macedo, Laura Payne, Rick Prum, Pamela Rasmussen, Robin Restall, Charlie Rosa, Mike Sorenson and J.Van Remsen. Karen Klitz painted the color plates of cuckoos. John Megahan drew the black and white illustrations of birds and bones and digitized the maps, the eggs and nestling photographs. Janet Hinshaw, Alec Lindsay, David Lahti, John Megahan, Laura Payne, Jean Woods and Tamaki Yuri helped prepare the illustrations. The National Science Foundation supported fieldwork on cuckoos and laboratory work on molecular genetics. Additional support was provided for field studies by the National Geographic Research Committee and the University of Michigan, and for molecular studies by the University of Michigan Museum of Zoology and Boston University. I thanks the series editors Professors Christopher Perrins,Walter Bock, and Jiro Kikkawa for their support, Judith May, formerly Senior Editor, for inviting the contribution, and the staff of Oxford University Press and Macmillan India, especially Stephen Benaim and Laura O’Neill, all for providing the freedom and encouragement to develop the book. I thank OUP for publishing this wonderful series on birds.

A brood parasite destroys its host and family—speaks the Fool, as Goneril evicts Lear: “For you know, nuncle, The hedge-sparrow fed the Cuckoo so long That it’s had its head bit off by its young. (King Lear, I, 4, 203–205),

“. . . the Cuckoos. In these the confusion is tremendous, and I don’t know if [I] will be able to make them out satisfactorily. I am afraid that I shall have not the time to go through them in a proper way.”— Tommaso Salvadori, 1875, letter to R. B. Sharpe, 12 Oct 1877 (Violani et al. 1997: 40–41).

Laying behavior of Little Bronze-cuckoo Chrysococcyx minutillus in Putrajaya Wetland, near Kuala Lumpur, Malay Peninsula. a, female cuckoo perches and peers at a pocket-shaped nest of a flyeater, Golden-bellied Gerygone Gerygone sulphurea. The cuckoo then perched closer to the nest, then moved to the nest. b, cuckoo laying in nest, perched on nest pocket rim, the head and cloacal region inserted into the nest entrance, the back feathers raised, the wings and tail spread outside the nest, wings grasping the nest and the tail supporting the bird (obscured by vegetation). The cuckoo remained in this posture for 3–4 seconds. After the cuckoo's visit, an unspotted greenish egg appeared in the nest — Little Bronze-cuckoos lay greenish eggs; the flyeaters lay spotted whitish eggs. Drawn from photos by Cheong Weng Chun. The laying behavior of this cuckoo is similar to that described in other bronze-cuckoo species in Australia (Brooker et al. 1988).

Contents

1 2 3 4 5 6 7 8 9 10 11

List of color plates List of abbreviations Plan of the book

xiv xv xix

PART I General chapters Introduction to the cuckoos Distribution, habitats and conservation status Behavior Morphology A molecular genetic analysis of cuckoo phylogeny Species Fossil and comparative evidence of cuckoo relationships Breeding biology and life histories Cooperative breeding Brood parasitism The evolution of brood parasitism in cuckoos

3 9 15 27 68 95 109 114 132 137 154

PART II Species accounts Crotophaginae Genus Guira Guira Cuckoo Genus Crotophaga Greater Ani Smooth-billed Ani Groove-billed Ani

169 169 169 172 172 174 178

Neomorphinae Genus Tapera American Striped Cuckoo Genus Dromococcyx Pheasant Cuckoo Pavonine Cuckoo Genus Morococcyx Lesser Ground-cuckoo Genus Geococcyx Greater Roadrunner Lesser Roadrunner

Guira guira Crotophaga major Crotophaga ani Crotophaga sulcirostris

Tapera naevia Dromococcyx phasianellus Dromococcyx pavoninus Morococcyx erythropygus Geococcyx californianus Geococcyx velox

183 183 183 187 187 189 191 191 193 193 198

x Contents Genus Neomorphus Banded Ground-cuckoo Rufous-winged Ground-cuckoo Red-billed Ground-cuckoo Rufous-vented Ground-cuckoo Centropodinae Genus Centropus Buff-headed Coucal Pied Coucal Greater Black Coucal Biak Coucal Rufous Coucal Green-billed Coucal Black-faced Coucal Black-hooded Coucal Short-toed Coucal Bay Coucal Gabon Coucal Black-throated Coucal Senegal Coucal Blue-headed Coucal Coppery-tailed Coucal White-browed Coucal Javan Coucal Greater Coucal Goliath Coucal Madagascar Coucal African Black Coucal Philippine Coucal Lesser Coucal Violaceous Coucal Lesser Black Coucal Pheasant Coucal Couinae Genus Carpococcyx Sumatran Ground-cuckoo Bornean Ground-cuckoo Coral-billed Ground-cuckoo Genus Coua Crested Coua Verreaux’s Coua Blue Coua Red-capped Coua Red-fronted Coua Coquerel’s Coua

radiolosus rufipennis pucheranii geoffroyi

200 200 202 203 204

milo ateralbus menbeki chalybeus unirufus chlororhynchos melanops steerii rectunguis celebensis anselli leucogaster senegalensis monachus cupreicaudus superciliosus nigrorufus sinensis goliath toulou grillii viridis bengalensis violaceus bernsteini phasianinus

208 208 208 210 211 213 214 215 216 218 219 221 222 224 226 228 231 233 236 238 242 244 246 248 250 254 255 257

Neomorphus Neomorphus Neomorphus Neomorphus

Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus Centropus

Carpococcyx viridis Carpococcyx radiatus Carpococcyx renauldi Coua Coua Coua Coua Coua Coua

cristata verreauxi caerulea ruficeps reynaudii coquereli

262 262 262 264 266 268 268 270 271 273 274 276

Contents xi Running Coua Giant Coua Snail-eating Coua Red-breasted Coua

Cuculinae Genus Rhinortha Raffles’s Malkoha Genus Ceuthmochares Whistling Yellowbill Chattering Yellowbill Genus Taccocua Sirkeer Malkoha Genus Zanclostomus Red-billed Malkoha Genus Phaenicophaeus Chestnut-breasted Malkoha Chestnut-bellied Malkoha Red-faced Malkoha Blue-faced Malkoha Black-bellied Malkoha Green-billed Malkoha Genus Rhamphococcyx Yellow-billed Malkoha Genus Dasylophus Rough-crested Cuckoo Scale-feathered Malkoha Genus Clamator Chestnut-winged Cuckoo Great Spotted Cuckoo Levaillant’s Cuckoo Jacobin Cuckoo Genus Coccycua Little Cuckoo Dwarf Cuckoo Ash-colored Cuckoo Genus Piaya Squirrel Cuckoo Black-bellied Cuckoo Genus Coccyzus Dark-billed Cuckoo Yellow-billed Cuckoo Pearly-breasted Cuckoo Mangrove Cuckoo Cocos Cuckoo Black-billed Cuckoo Gray-capped Cuckoo

Coua Coua Coua Coua

cursor gigas delalandei serriana

277 279 280 281

283 283 Rhinortha chlorophaea 283 285 Ceuthmochares australis 285 Ceuthmochares aereus 287 291 Taccocua leschenaultii 291 293 Zanclostomus javanicus 293 294 Phaenicophaeus curvirostris 294 Phaenicophaeus sumatranus 298 Phaenicophaeus pyrrhocephalus 299 Phaenicophaeus viridirostris 301 Phaenicophaeus diardi 302 Phaenicophaeus tristis 304 306 Rhamphococcyx calyorhynchus 306 308 Dasylophus superciliosus 308 Dasylophus cumingi 309 311 Clamator coromandus 311 Clamator glandarius 313 Clamator levaillantii 318 Clamator jacobinus 320 325 Coccycua minuta 325 Coccycua pumila 328 Coccycua cinerea 330 331 Piaya cayana 331 Piaya melanogaster 335 336 Coccyzus melacoryphus 336 Coccyzus americanus 339 Coccyzus euleri 345 Coccyzus minor 347 Coccyzus ferrugineus 350 Coccyzus erythropthalmus 351 Coccyzus lansbergi 354

xii Contents Chestnut-bellied Cuckoo Rufous-breasted Cuckoo Jamaican Lizard-cuckoo Puerto Rican Lizard-cuckoo Cuban Lizard-cuckoo Hispaniolan Lizard-cuckoo Genus Pachycoccyx Thick-billed Cuckoo Genus Microdynamis Dwarf Koel Genus Eudynamys Common Koel Genus Urodynamis Long-tailed Cuckoo Genus Scythrops Channel-billed Cuckoo Genus Chrysococcyx Asian Emerald Cuckoo Violet Cuckoo Diederik Cuckoo Klaas’s Cuckoo Yellow-throated Cuckoo African Emerald Cuckoo Long-billed Cuckoo Horsfield’s Bronze-cuckoo Black-eared Cuckoo Rufous-throated Bronze-cuckoo Shining Bronze-cuckoo White-eared Bronze-cuckoo Little Bronze-cuckoo Genus Cacomantis Pallid Cuckoo White-crowned Cuckoo Chestnut-breasted Cuckoo Fan-tailed Cuckoo Banded Bay Cuckoo Plaintive Cuckoo Gray-bellied Cuckoo Brush Cuckoo Genus Cercococcyx Dusky Long-tailed Cuckoo Olive Long-tailed Cuckoo Barred Long-tailed Cuckoo Genus Surniculus Fork-tailed Drongo-cuckoo Philippine Drongo-cuckoo Square-tailed Drongo-cuckoo

Coccyzus Coccyzus Coccyzus Coccyzus Coccyzus Coccyzus

pluvialis rufigularis vetula vieilloti merlini longirostris

Pachycoccyx audeberti Microdynamis parva Eudynamys scolopacea Urodynamis taitensis Scythrops novaehollandiae Chrysococcyx Chrysococcyx Chrysococcyx Chrysococcyx Chrysococcyx Chrysococcyx Chrysococcyx Chrysococcyx Chrysococcyx Chrysococcyx Chrysococcyx Chrysococcyx Chrysococcyx Cacomantis Cacomantis Cacomantis Cacomantis Cacomantis Cacomantis Cacomantis Cacomantis

maculatus xanthorhynchus caprius klaas flavigularis cupreus megarhynchus basalis osculans ruficollis lucidus meyeri minutillus

pallidus leucolophus castaneiventris flabelliformis sonneratii merulinus passerinus variolosus

Cercococcyx mechowi Cercococcyx olivinus Cercococcyx montanus Surniculus dicruroides Surniculus velutinus Surniculus lugubris

356 357 358 360 361 362 364 364 367 367 368 369 378 378 380 381 385 385 388 391 395 399 400 403 405 407 409 410 413 415 421 422 424 426 427 430 433 437 439 447 447 449 451 453 454 458 460

Contents xiii Moluccan Drongo-cuckoo Genus Hierococcyx Mustached Hawk-cuckoo Dark Hawk-cuckoo Large Hawk-cuckoo Common Hawk-cuckoo Rufous Hawk-cuckoo Philippine Hawk-cuckoo Javan Hawk-cuckoo Whistling Hawk-cuckoo Genus Cuculus Black Cuckoo Red-chested Cuckoo Asian Lesser Cuckoo Sulawesi Cuckoo Indian Cuckoo Madagascar Lesser Cuckoo African Cuckoo Oriental Cuckoo Himalayan Cuckoo Sunda Lesser Cuckoo Common Cuckoo

Surniculus musschenbroeki Hierococcyx Hierococcyx Hierococcyx Hierococcyx Hierococcyx Hierococcyx Hierococcyx Hierococcyx Cuculus Cuculus Cuculus Cuculus Cuculus Cuculus Cuculus Cuculus Cuculus Cuculus Cuculus

vagans bocki sparverioides varius hyperythrus pectoralis fugax nisicolor

clamosus solitarius poliocephalus crassirostris micropterus rochii gularis optatus saturatus lepidus canorus

463 464 465 466 468 471 473 475 477 479 481 481 485 488 491 492 496 498 500 505 508 510

Glossary

519

Bibliography

526

Index

607

Color plates Color plates fall between pages 166–167. Plate 1 Plate 2 Plate 3 Plate 4 Plate 5 Plate 6 Plate 7 Plate 8 Plate 9 Plate 10 Plate Plate Plate Plate Plate Plate Plate

11 12 13 14 15 16 17

Plate 18 Plate 19 Plate 20

New World cuckoos Crotophaga, Guira,Tapera, Dromococcyx and Morococcyx New World Geococcyx and Neomorphus Coucals Centropus milo, ateralbus, menbeki, chalybeus, unirufus, celebensis, melanops, steerii and rectunguis Coucals Centropus nigrorufus, sinensis, chlororhynchos, phasianinus (including nigricans and spilopterus) and bernsteini Coucals Centropus toulou, violaceus, goliath, bengalensis, viridis, grillii African coucals Centropus leucogaster, anselli, monachus, cupreicaudus, senegalensis and superciliosus Old World ground-cuckoos Carpococcyx and couas Coua Old World malkohas Ceuthmochares, Rhinortha, Taccocua, Zanclostomus, Rhamphococcyx, Dasylophus and Phaenicophaeus New World cuckoos Coccycua, Piaya and Coccyzus (including Saurothera and Hyetornis) Old World brood-parasitic genera Pachycoccyx, Microdynamis, Eudynamys, Urodynamis and Scythrops Glossy cuckoos in Africa and Asia, genus Chrysococcyx Glossy cuckoos in Australasia, genus Chrysococcyx Brush cuckoos Cacomantis Drongo-cuckoos Surniculus, long-tailed cuckoos Cercococcyx Crested cuckoos Clamator, hawk-cuckoos Hierococcyx Cuckoos Cuculus Eggs of brood-parasitic cuckoos and their hosts in Australia and southern Africa. Brood parasitism in cuckoos. Eggs of Village Weaver Ploceus cucullatus Mouth patterns of young cuckoos

Abbreviations AM AMNH ANSP BMNH BPBM BWYO BYU CAS CM CMNH CU CUM DM DMNH FLMNH FMNH IRSNB JFBM KMMA KU LNS LSU MCNB MCZ MHNLS MMZ MNHN MNM MSNG MSNM MSU MVZ MZB MZUS NMGL NMK NMP NMW NSA

Auckland Museum, Auckland, New Zealand American Museum of Natural History, New York Academy of Natural Sciences, Philadelphia British Museum of Natural History (Natural History Museum),Tring, UK Bernice P. Bishop Museum, Honolulu National Museums of Zimbabwe, Bulawayo Museum of Natural History, Brigham Young University, Hawaii Campus, Laie, Hawaii Chinese Academy of Sciences, Beijing Carnegie Museum, Pittsburgh Cincinnati Museum of Natural History, Cincinnati Cornell University, Ithaca Cambridge University Museum, Cambridge Durban Museum, Durban Delaware Museum of Natural History, Greenville Florida State Museum of Natural History, Gainesville Field Museum of Natural History, Chicago Institut Royal des Sciences Naturales de Belgique, Brussels James Ford Bell Museum of Natural History, University of Minnesota Koninklijk Museum voor Midden Afrika,Tervuren University of Kansas Museum of Natural History, Lawrence Library of Natural Sounds, Cornell University, Ithaca, New York Museum of Zoology, Louisiana State University, Baton Rouge Museu de Cièncias Naturals de la Ciutadella, Barcelona Museum of Comparative Zoology, Cambridge, MA Sociedad de Ciencias Naturales La Salle (Fondacion La Salle), Caracas Moscow Museum of Zoology, Moscow Muséum National d’Histoire Naturelle, Paris Malawi National Museum, Blantyre Museo Civico di Storia Naturale “Giacomo Doria”, Genoa Museo Civico di Storia Naturale, Società Italiana di Scienze Naturale, Milan Michigan State University Museum, East Lansing Museum of Vertebrate Zoology, University of California, Berkeley Museum Zoologicum Bogoriense, Cibinong nr Bogor Museo Zoologico Universitario “La Specola”, Florence National Museums and Galleries (Merseyside Museums), Liverpool National Museum of Kenya, Nairobi National Museum of the Philippines, Manila Naturhistorisches Museum Wien,Vienna British Library National Sound Archive, London

xvi Abbreviations NSM OU OSU POC QM RMNH ROM SMF SMNS SMTD SP SRSU TCWC TISTR UCLA UMMZ UNLV UNM USNM UTEP UWBM WAM WFVZ YIO YPM VW ZFMK ZMA ZMB ZMM ZMUC ZRC ZSM F I, Is M NP NSW NT P PN S

National Science Museum,Technopolis, Bangkok University of Oklahoma, Norman Borror Laboratory of Bioacoustics, Ohio State University, Columbus Phelps Ornithological Collection, Caracas,Venezuela Queensland Museum, Brisbane Nationaal Natuurhistorisch Museum, Leiden Royal Ontario Museum,Toronto Forschungsinstitut Senckenberg, Frankfurt Staatliches Museum für Naturkunde in Stuttgart, Stuttgart Staatliches Museum für Tierkunde, Dresden Sabah Parks, Kota Kinabalu Scudday Vertebrate Collection, Sul Ross State University, Alpine,Texas Texas Cooperative Wildlife Collection, College Station Thailand Institute of Scientific and Technical Research, Bangkok Dickey Collection, University of California, Los Angeles University of Michigan Museum of Zoology, Ann Arbor University of Nevada at Las Vegas, Las Vegas (MBM) University of New Mexico Museum of Southwestern Biology, Albuquerque National Museum of Natural History,Washington, D.C. University of Texas at El Paso, El Paso Burke Museum, University of Washington, Seattle Western Australian Museum, Perth Western Foundation of Vertebrate Zoology, Camarillo, California Yamashina Institute of Ornithology, Abiko, Japan Yale Peabody Museum, New Haven Vogelpark Walsrode,Walsrode, Germany Museum Alexander Koenig, Bonn Universiteit van Amsterdam Museum, Amsterdam Zoologische Museum, Humboldt-Universitat, Berlin Zoological Museum of Moscow, Moscow Universitets Zoologiske Museum, Copenhagen National Museum of Singapore (Zoological Reference Collection, National University of Singapore), Singapore Zoologische Staatssammlung München, Munich female island(s) male National Park New South Wales, Australia Northern Territory, Australia wing feather, primary (numbered from innermost outward, P1 the inner primary, P10 the outer primary) Parc National wing feather, secondary (numbered from outermost inward)

Abbreviations xvii T TA TB U WA

tail feather, rectrix (numbered from the inner pair outward,T1 the inner rectrix,T5 the outer rectrix) ambient temperature, air temperature body temperature unsexed Western Australia

This page intentionally left blank

Plan of the book

The book is a comprehensive survey of the biology of cuckoos, with descriptions of the life history and systematic status of the 141 cuckoo species. Part I gives general accounts of cuckoo biology, behavior, cooperative breeding and brood parasitism, and it develops ideas about the relationships among the species and major lineages of cuckoos, based on song, morphology and molecular genetics. In the general chapters the cuckoos are mentioned by common name or scientific name, and sometimes by both names. In part I the text, tables and graphs summarize the details of life history and comparative morphology of the cuckoos. Part II consists of species accounts of each cuckoo species, with details about plumage, size, geographic variation, behavior and breeding biology, depending on the information available.The details and references for the general chapters in Part I are in the species accounts in Part II. Recent field observations on the behavior of these birds and phylogenetic studies on cuckoo evolution, allow a synthesis of ideas about the evolution of parental behavior and adaptations for competition, cooperation and brood parasitism in these remarkable birds. The species were determined by the morphological distinctiveness of regional populations and by their songs when these were available on tape recordings. Each cuckoo species that has been studied and is well known has a characteristic song.The songs provide an objective criterion to assess the status of species: if cuckoos have the same song they are conspecific, and if the songs differ, the cuckoos are not of the same species. Female birds choose their mates on the basis of song. Females are attracted to songs of their own species and not to songs of other

species. For these reasons bird songs are a major clue to the nature and limits of biological species (Payne 1986, Alström and Ranft 2003). Another criterion used to determine species status is the fact that birds living in the same area and not interbreeding are distinct species. So song is used mainly to compare birds living in different areas, such as in continental Asia and on islands. In a few cases the results of molecular genetics were used to refine ideas about the nature and limits of cuckoo species. The scientific and common names of most birds in the text are the names in Sibley and Monroe (1990), a comprehensive recent and referenced systematic revision of birds on a worldwide scale.Where the names in that source differ from those in standard regional works, as in Japan, the regional names are also included. For cuckoos the names in Sibley and Monroe’s book are used, except where recent evidence of song or molecular genetics, or both, indicates that the species are not the ones those authors suggested. For example, in the hawk-cuckoos Hierococcyx fugax and the geographic forms nisicolor, hyperythrus and pectoralis, there now appear to be four species rather than one ‘Hodgson’s Hawkcuckoo’. In Part 1 the common and scientific names are given together at their first occurrence in the text. Subsequently the common names are used, except in the tables and chapters that discuss systematic details, where the scientific names are used again. For most cuckoo species and subspecies, 20 or more specimens were measured if the series were available in UMMZ, otherwise specimens in other museums were sampled. When possible, the sex of the specimens used was determined when the

xx Plan of the book gonads were examined, and male or female indicated on the label. In a few specimen series it became obvious that the birds had been mis-sexed. Some species are now known to differ in male and female plumage, which was not known when the collector obtained the birds, for instance with Raffles’s Malkoha Rhinortha chlorophaea (Swainson 1838, Shelley 1891). In other species such as coucals Centropus the sexes differ in size. I avoided museum series where mis-sexing of birds was apparent. In some cases the sex had been added after the field label was written. In such cases I usually did not use the specimen, although some may have been labeled appropriately from the collector’s field notes after the field label was written. The publications where each species was first described are listed in the species accounts. Citations of publications where subspecies were described are available elsewhere (Peters 1940, Payne 1997b), and for recently described forms the source is indicated in the species account. Measurements were taken on museum specimens. Wing length was measured with a stop-end mm rule and the wing flattened on the rule. Tail length was taken with a rule from the point of insertion of the central rectrices (tail feathers) on the body to the tip of the central tail feather. The tail could not be measured on some specimens of which the collector had pushed the base of the tail well into the body cavity, as far as 100 mm in the large coucals and malkohas that were collected and prepared by G. Heinrich, as described by Eck (1978). The tail feathers are numbered in the text, T1 being the innermost rectrix,T2 the next, and T5 the outermost rectrix. Bill length was measured with calipers from the skull to the tip of the bill. Tarsus length was measured from the proximal end of the tarsometatarsus to the distal margin of the most distal tarsal scute that was not divided between toes 2, 3 and 4. Specimens were measured for all species and nearly all subspecies known to vary in size, and in others that vary in plumage color or pattern. For cuckoos that I did not measure, I used the measurements given by E. Mayr, E. Stresemann, D. Rogers in Higgins (1999), where these coincided with the measurements of specimens, done by other ornithologists. Bird skins

shrink as they dry, and the wing length of museum specimens is less than that of live birds measured in the field, with a difference of about 2 mm in the Yellow-billed Cuckoo Coccyzus americanus. Measurements are given in terms of maximum, minimum, mean, and ⫾ standard deviation (sd) where sample sizes are larger (n ⫽ 6 or more).Where data are cited from another author, sd is included only when the author included this index of variation. Body weight is the main index of body size in birds. Body weight can change as much as 40% in resident cuckoos from summer to winter (Greater Roadrunner Geococcyx californianus), more than 20% above average for breeding females with a large yolking ovary and oviduct, 70% above average for birds that had large stores of fat before migration, and 40% below average after a long over-water migration (Yellow-billed Cuckoos Coccyzus americanus). Individual birds that are caught and weighed through the year even in a tropical region change their body weight by as much as 16% (Mustached Hawk-cuckoo Hierococcyx vagans).Where specimen labels noted a large amount of premigratory fat, or laying females with enlarged ovaries and oviducts and sometimes an egg in the oviduct, the body weights of these specimens were noted separately. Body weights do not vary in a statistically normal frequency distribution and for this reason sd’s are not included for weights. The wing formula captures some of the diversity in wing shape. Counting the primaries from the innermost outward, the rank of feather length from the bend of the wing to the tip of the folded wing was determined, with the longest primary listed first, followed by the next, and so on. For example, P7 ⬎ 6 ⬎5 indicates that the seventh primary (P7) is longest, then P6 and then P5. Cuckoos with the longest primary number 8 and 9 have pointed wings, and cuckoos with the longest primary number 4 and 5 have rounded wings.The details of the wing formula vary within cuckoo species, but this variation is much less than between groups of closely related species of cuckoos (Stephan 2001a). Plumage colors of specimens in different museums were compared with reference to a standard color guide (Kornerup and Wanscher 1967). Also useful were color photographs, taken with a standard

Plan of the book xxi gray reflectance card placed under the specimens. A Nikon-F3 camera was attached to a ring flash at 0.7 or 1.0 m from the specimen. Photographs taken under these conditions were compared, for birds in different museums. Photographs of birds such as the Little Bronze-cuckoo Chrysococcyx minutillus species complex show the glossy color to be the same across the visual field, except in the center of the body plumage where it appears dark. Glossy cuckoos with bronze gloss are more greenish than in their photographs, and these specimens were either compared directly with each other, or their photographs were compared with other photographs taken in the same conditions. ‘Sources’ indicates the museum collections that were used for species descriptions and, in some cases, the publications used for the appearance of nestlings. Colors of the bare skin on the face, the eye-ring, the iris, the bill and the feet were recorded from photographs and the labels on museum specimens. Where the colors of plumage and the bare parts differed between the sexes and ages, these are described. Subspecies are recognized when populations are nearly exclusive in size, color or other morphological characteristics, in different parts of their geographic range. Geographic variation in birds involves average size differences between birds in different areas, or a continuous variation between over areas. For example, Yellow-billed Cuckoos in western North America are larger on average than birds in eastern North America, but most birds are in the size range that overlaps between the eastern and western birds, and no subspecies are recognized in this case. Banders and collectors who see these birds in migration or in their winter range in South America, cannot know for certain the breeding areas where the birds originated in North America. Although early work with cuckoos involved collecting birds in the field and studying them in the museum, recent work in the field has involved conservation of their populations in nature. Wilson (1993) cites Baba Dioum:“In the end, we will conserve only what we love, we will love only what we understand . . .”. Conservationists are concerned about the birds whose numbers are low, and whose continued existence is threatened by human activity. This concern is expressed through the protection of

their habitats, to preserve the wild birds alive in natural conditions. Even though these biologists favor an agenda for conservation, we make our best estimates of the distinctiveness of bird species, on the basis of their biology rather than their conservation status. Conservationists have sometimes argued that if each population is given the special taxonomic status of a species, this will increase the legal and social protection of the population. There is, however, no evidence that systematic biologists have had a compelling impact on our own population growth, economics, migration, commercial logging or agricultural techniques. It is questionable whether biologists ought to set their scientific standards for a perceived social gain (Slobodkin 2001), such as in naming every population as a species or a subspecies, which practice becomes a political statement rather than a biological description. Even though we look for arguments and means by which to preserve wild birds and their habitats, the long-term outlook for these birds, and for tropical forests in general over the years to come, is not encouraging (Struhsaker 1997, Heaney and Regalado 1998, Oates 1999, Holmes 2000). Morphological descriptions are used to identify birds in the field or in a museum, and these descriptions are presented in the book. In earlier days a systematic revision provided a diagnosis, that is, a summary of characters that differentiate a taxon from related or similar taxa, and genera were recognized and diagnosed when an authority perceived “really trenchant external characters” (Peters 1940:45). In practice, the description of a species became the standard for the diagnosis of a genus (e.g., Cabanis and Heine 1863). Because similar body forms have evolved in different lineages of birds, as molecular genetic studies have shown (e.g., Sibley and Ahlquist 1990, Cibois et al. 1999, 2001, Sorenson and Payne 2001), the book emphasizes description rather than diagnosis. The book takes a new look at the relationships among the cuckoos, using phylogenetic analysis of discrete molecular genetic characters. The genetic characters (that is, the sequences of nucleotide bases in the DNA) determined in the study, can be accessed by computer in Genbank, the international register for comparative genetic data. These

xxii Plan of the book sequences were used to determine the groupings of cuckoo species into clades (lineages or branches of common ancestry) (Hillis et al. 1996). Molecular genetic analysis uses a large number of characters (the four nucleotide bases and their sequences at thousands of sites in the mitochondrial gene).These genetic characters provide a much larger data set than distinct morphological characters have ever allowed. Molecular genetic profiles were determined for each cuckoo species. The evolutionary relationships among cuckoos were estimated from the molecular genetic sequences and computer programs (PAUP*: Swofford 2001; POY, Gladstein and Wheeler 1996) which use parsimony, a principle of the smallest number of evolutionary changes that are necessary to derive the relationships from a common ancestor in a phylogeny. The phylogenetic estimates of relationships among the cuckoos in the molecular genetic results, were similar to the traditional sets of genera and subfamilies of cuckoos as recognized by morphologists and other museum systematists. Where they differed, the genetic phylogeny was taken as the best estimate of common ancestry, and these results are used throughout the book. I examined nearly all specimens from which the genetic samples were taken. A few re-identifications of these cuckoo specimens was a result (Payne and Sorenson 2003). The evolution of social behavior, such as cooperative breeding and brood parasitism, is a question of general significance. Phylogeny is used not only to clarify the evolutionary relationships among cuckoo species, but also to determine which behaviors evolved once or more than once, and the circumstances in the most closely related lineages that may have led to the evolution of these behaviors (Brooks and McLennan 1991, Edwards and Naeem 1993, Ligon 1993, Sorenson and Payne 2001, 2002). That is, cuckoos are a case study of evolutionary history and adaptation to behavioral ecology. Determining their phylogeny is a way to test the course of evolution of brood parasitism from a non-parasitic ancestor. These questions have been

addressed before (Poiani and Elgar 1994, Poiani 1998) using the estimate of cuckoo relationships in Sibley and Ahlquist (1990), yet that estimate was based on a genetic technique that did not use discrete characters, and it did not include all the major groups of cuckoos. The phylogenetic analysis of molecular genetic characters in the present study was designed not only to get a best estimate of relationships among the cuckoos, but also to determine processes in the evolution of social behavior and breeding biology. The arrangement of subfamilies, genera and species follows the phylogenetic convention of listing the basal taxa first, and the recent radiations or “crown” taxa last, in the order of their branching in the evolutionary tree (Wiley et al. 1991). This convention has long been used in some avian groups such as waterfowl, in contrast to recent reviews of cuckoos which first list the familiar cuckoos of Europe.Where the first convention did not apply, as in symmetric branches of the tree, other conventions were used. First, species with little or no geographic variation between populations (subspecies) were listed, before species with greater geographic variation. Second, where no subspecies were involved, the species with the more northern extent of distribution was listed first. Application of the “basal group first” norm caused the New World Cuckoos to be listed before the Old World cuckoos, and within the New World clade, the group-living crotophagine cuckoos are listed first. The order in which birds appear in the list does not necessarily mean that ancestral cuckoos were cooperative-group living birds. Within a species the geographic forms are listed from north to south and from east to west, with no necessary phylogenetic interpretation. Geographic names in the book are mainly those in the Times Atlas of 1993. In Indonesia the historical names of the Moluccas, now Maluku, and the Lesser Sunda islands, now Nusa Tenggara, have been retained, as they are widely used in current regional works on birds (White and Bruce 1986, Coates and Bishop 1997).

PART I General chapters

This page intentionally left blank

1 Introduction to the cuckoos

The parasitic breeding behavior of cuckoos has fascinated people for centuries. The brood-parasitic cuckoos lay their eggs in the nests of other kinds of birds, and never rear their own young ( Johnsgard 1997, Rothstein and Robinson 1998, Davies 2000). The natural history of the Common Cuckoo Cuculus canorus and the Great Spotted Cuckoo Clamator glandarius, the two species that live in Europe where they have been observed for many years, is well known. In Classic times,Aristotle studied the brood-parasitic behavior of Common Cuckoos (Friedmann 1964b,Aristotle 1991), and in Britain Shakespeare referred to cuckoos in his plays. The term “cuckoldry” was often used in the sense of an adulterous affair, and the implications of immorality in a later period led to a censored version of Shakespeare that eliminated the term “cuckold” (Bowdler 1861, Hamilton 1996). Many other species of brood-parasitic cuckoos live in Africa, Asia and Australia and in Central and South America where they have been watched by resident naturalists. Brood-parasitic cuckoos are diverse with 56 species in the Old World and three in the New World. Nevertheless, brood parasitism is only one of several breeding behaviors in this family of birds. A few cuckoos, the anis and the Guira Cuckoo of the New World, are sociable, with several pairs sharing a nest where they lay their eggs and care for the young; these are the cooperative breeders. Their cooperation is balanced by competition, as a female may remove the eggs of other females when she lays her own. In contrast to these birds which have gone to extremes of parental care and social behavior, most species of cuckoos live in solitary pairs and regularly build a nest and raise their own young.

Even these cuckoos occasionally lay their eggs in the nests of others, either a nest one of their own species or a nest of another kind of cuckoo. One cuckoo species is extinct, and eight cuckoo species are threatened or endangered at the global level. The distribution of cuckoos is cosmopolitan, and most species live in the Old World tropics.The variation in social behavior and parental care of cuckoos may be unmatched among the bird families of the world. The cuckoos are the most successful broodparasitic birds. They have the largest number of species, the largest number of host species that rear their young, and worldwide distribution.The cuckoos are one of the five families of birds in which brood parasitism is the only life style for one or more species—the others are two families of songbirds (the Old World finches and the New World cowbirds), the honeyguides, and the ducks. One duck is a dedicated brood parasite; indeed other waterfowl sometimes lay their eggs in the nests of other ducks of the same species or a different species. In each of these other families, an obligatory breeding style of brood parasitism has evolved only once (Sorenson and Payne 2001, 2002). Brood parasitism has originated more than once in cuckoos. Results of molecular analysis in this study indicate that the parasitic lineages of cuckoos are not all each others’ closest relatives. Because brood parasitism has evolved more than once, the cuckoos provide a replicate test of the behavioral context of the course of evolutionary changes when parental care was transformed into brood parasitism. Young cuckoos are altricial. They remain in the nest where they depend on their parents or foster

4 The Cuckoos parents to provide food, and they grow rapidly.The young of nesting cuckoos leave the nest days before they are able to fly, while the young of broodparasitic cuckoos take longer to fledge and are welldeveloped when they leave the nest. Nesting cuckoos in the New World have the shortest incubation periods and the shortest nestling periods of any birds. Even the Old World coucals have a short nestling life, unusual for such large birds.The variable period in which the cuckoos depend upon parental care is closely associated with the life style of nesting and brood parasitism. The mating systems and social organization of cuckoos range from monogamous pairs that stay together in a social bond, to the cooperatively breeding anis and Guira Cuckoo where incubation and nestling care are shared by group members that use a single communal nest to rear their young. Among these birds there is genetic evidence of polygyny and polyandry as well as social monogamy. Cuckoos have a long evolutionary history. Studies in molecular genetics have estimated that cuckoos occurred as a distinct lineage more than 60 million years ago, a lineage as ancient as that of any modern bird (Sibley and Ahlquist 1990). Fossil cuckoos have been known for only a part of this time. These fossil birds are as varied as modern cuckoos, and some appear to have been flightless. Cuckoos occur throughout the world on all habitable continents and on the island areas of southern Asia, and they are an important part of the biological diversity of tropical birds. Much of what is known about cuckoos came from birds collected in the 1800s when they were first discovered by western Europeans. A few cuckoos were described by their collector and discoverer, Stamford Raffles (1781–1826), natural historian, ethnologist, Governor of Java and founder of the trading station that became Singapore (Raffles 1817, 1830). He began his work as a clerk in London and read widely, mastered languages, cultivated friends and gained a position in Penang. Based on his interests and his rewarding discussions with the Viceroy of India, he was able to study the area’s history, culture and wildlife (Raffles 1830, Mearns and Mearns 1998). Raffles was unusual among early collectors for his access to the publishing world. Later collec-

tors visited the same areas and prepared specimens that made their way back to natural history museums, and in some regions resident naturalists were able to observe the details of their behavior and breeding biology. These collectors were employed by governments and private collectors, and most of their specimens were sent to natural history museums in Europe where the species were described and named (e.g. Blyth 1842, Gould 1845b, Cabanis and Heine 1863, Salvadori 1881, Shelley 1891). Linnaeus described more kinds of cuckoos than any other taxonomist, and half of the 22 species that he described are currently recognized as distinct kinds of birds (some others included a second sex), although the specimens from which they were described are no longer in existence. Tommaso Salvadori appears to hold the record for the number of new cuckoos described by a post-Linnaean ornithologist. Museum patrons employed bird collectors who visited Borneo, the Moluccas and New Guinea, and from specimens that made it safely back to Europe, Salvadori described 11 kinds of cuckoos which are now recognized as species and subspecies (Salvadori 1874, 1875, 1876, 1878a,b, 1879, 1881, 1889). He was based at the museum in Turin (Elter 1986,Violani et al. 1997), and his many type specimens in Turin, Genoa, Leiden and other museums rank him highest among his ornithological generation (Payne 1997b). He usually did not designate a single specimen as the holotype of a species, but rather syntypes of a type series, many of which went to other museums as “duplicates”, with some still being discovered in these collections. Salvadori worked with other families of birds as well. He worked for two years at the National Museum in London where he wrote monographs on parrots and pigeons, and he lived in Leiden where many of his type series are located (van den Hoek Ostende et al. 1997). In other museums in Germany, Cabanis and Heine (1863) described nine cuckoos which are now recognized as species and subspecies, and they described several cuckoo genera as well. A few species of cuckoos are known only from museum collections. The nests of several cuckoos have never been found, and the life histories of these and many other cuckoos are unknown. Many cuckoos live in the Old World tropics where few

Introduction to the cuckoos 5 observations of their life histories have been made in the forests, where these birds are hard to see as the nests are well hidden. Once a nest is found a predator is also likely to find it. Little collecting of anatomical and genetic materials for scientific research has been done in recent years, and the careful field studies that were the fieldmark of resident ornithologists have waned with changes in land ownership. Current research efforts have been limited by nationalistic zeal, and by the difficulty of productive research by unacclimated field biologists in the malarial tropics.The museum specimens that have been collected over the past two centuries are a valuable and unreplaceable resource and are our primary record of biological diversity.We continue to learn new details from these museum specimens, such as geographic variation and the plumages and molts of birds, and we can now retrieve genetic information from feathers or skins of these specimens. Field studies on behavior and recordings of the songs of cuckoos are much needed to understand these birds in natural conditions, and great opportunities remain to discover the life styles and the changing distribution of birds such as the cuckoos.

What are the cuckoos? Cuckoos are zygodactyl birds with the inner and outer toes directed backward and the other two toes directed forward. The bill is usually slender and slightly arched.The plumage of most cuckoos is soft and lax. The body form varies among species with their systematic relationships, body size and life style. Most arboreal cuckoos have a slender body and long tail, whereas terrestrial cuckoos are heavybodied and have long tarsi for their size (weight). The nostrils are round in many cuckoos, but slit-like in the coucals, in the crested cuckoos Clamator and Thick-billed Cuckoo Pachycoccyx audeberti, and to a lesser extent in Long-tailed Cuckoo Urodynamis taitensis, Common Koel Eudynamys scolopacea and Dwarf Koel Microdynamis parva, and slit-like in some malkohas also. Many cuckoos have long eye-lashes. The bill has no cere; the tarsi are scutellate and often unfeathered. The base of the tail has a naked and bilobed oil gland.

Cuckoos have a wing with 10 primaries and 9 to 13 secondaries (9 in several malkohas, in the large Caribbean species of Coccyzus (“Hyetornis”and “Saurothera” ), and in some squirrel cuckoos Piaya; 9 or 10 in Cuculus cuckoos; 10 in hawk-cuckoos Hierococcyx,Thick-billed Cuckoo Pachycoccyx,African Crested Cuckoo Clamator levaillantii and in some coucals; 11 in koels Microdynamis and Eudynamys, in some couas and coucals and in the Great Spotted Cuckoo Clamator glandarius); 12 in some other coucals, and 12 or 13 in the Channel-billed Cuckoo Scythrops novaehollandiae (Stephan 2001a).The alula is small in most cuckoos but large in the New World ground-cuckoos Neomorphus and Carpococcyx, and in the brood-parasitic American Striped Cuckoo Tapera naevia. In the last two cuckoos the large alulae are extended when the wings are spread in display.The wing is eutaxic.The pattern of wing molt is peculiar to cuckoos, where a wave of molt jumps over the neighboring old feathers. The odd-numbered primaries first drop and grow, followed by the evennumbered primaries. Cuckoos are capable of flight, varying from the swift and direct flight of the longwinged, long-distance migrant Common Cuckoo Cuculus canorus and Black-billed Cuckoo Coccyzus erythropthalmus of northern temperate regions, to the gliding flight of short-winged tropical forest-living malkohas and the slow awkward flight of anis, couas and coucals. The tail usually has 10 rectrices, though only 8 in anis Crotophaga and Guira Cuckoo Guira guira. The tail of some cuckoos is shorter than the wing, particularly that of the glossy cuckoos Chrysococcyx. The tail of other cuckoos is one and a half times the length of the wing, as in the arboreal squirrel cuckoos Piaya cayana and P. melanogaster, the yellowbills Ceuthmochares and several Asian malkohas, and the terrestrial ground-cuckoos Geococcyx and Neomorphus (Stephan 2001b). In North and Central America where all cuckoos have long tails, the area of the tail is larger in cuckoos than in any other family of bird of comparable body size (Hartman 1961). In many cuckoos the tail is strongly graduated, with the central feathers longest and the outer feathers short, and the steps marked by the conspicuous white tips of the tail feathers. In others the tail is rounded, square, or even forked, with the central

6 The Cuckoos feathers shorter than the outer feathers, especially in the Fork-tailed Drongo-cuckoo Surniculus dicruroides and Moluccan Drongo-cuckoo S. musschenbroeki. Cuckoos span a wide range in body size, with most species 16–70 cm in overall length.They vary in mass from 17 g with the smallest, Little Bronzecuckoo of Australasia, to 400 g with Giant Coua Coua gigas of Madagascar, and 700 g with the largest broodparasitic bird, Channel-billed Cuckoo Scythrops novaehollandiae of Australia (the next largest brood parasite is Black-headed Duck Heteronetta atricapilla at 513–565 g, Carboneras 1992). The largest nesting cuckoos are coucals in the New Guinea region, where Greater Black Coucal Centropus menbeki and Violaceous Coucal C. violaceus are more than 500 g, Goliath Coucal C. goliath are over 600 g, and Buffheaded Coucal C. milo are as large as 770 g.

The major groups of cuckoos The major groups of cuckoos include two groups with a common ancestry in the New World, two groups in the Old World, and one group that occurs in both regions. In the past, some of these groups have been called families, and others have only recently been recognized as lineages with a common evolutionary history, especially the Cuculinae, which include both New World and Old World cuckoos, and both nesting species and brood-parasitic species.

1. Crotophaginae The crotophagines are groupliving cuckoos and all cuckoos with this life style live in the New World.They have robust legs and a long tail, streaked brown plumage in the South American Guira Cuckoo Guira guira and black plumage and a deep compressed bill in the three ani species Crotophaga. Living in aggressive and noisy social groups, several pairs defend a common territory and share a joint communal nest where two or more females lay their eggs (in anis, some nests may be of single females) and the adults all rear their common young together as cooperative breeders. The nest is flat or a shallow bowl, built in a tree.The nestling mouth is pink with white marks on the palate and tongue. Anis often have a pungent odor, noticed when birds are held in the hand or when they are at their night roost. The social or adaptive

significance of the smell is unknown. The crotophagines have large and unique anal glands, but whether these produce chemical deterrents to predators or chemical signals to social partners or are accessory reproductive structures is unknown (Quay 1967).The ridge on the bill is shaped by an underlying ridge in the nasomaxillary skeleton, especially in Greater Ani Crotophaga major.Awkward in movements with the wings and tail seemingly disconnected from the body, anis are unlike the graceful arboreal cuculine cuckoos.Anis are slow and clumsy as they flutter and clamber through low bushes and weeds, where their flopping wings and waving tails flush out insects in dense vegetation; Guira Cuckoos fly like kites in the wind.

2. Neomorphinae Ground-cuckoos of the New World have short wings, long legs and a long tail. They include a dry scrub-forest terrestrial cuckoo of Central America (Lesser Ground-cuckoo Morococcyx erythropygus), lizard-catching roadrunners (Geococcyx) of the semi-arid regions of North America, and large South American ground-cuckoos Neomorphus that follow ant swarms in the tropical forests and eat the insects that escape the marauding ants. Roadrunners run across the ground at speeds of up to 30 km/hr in chase of their lizard and grasshopper prey, the birds holding their heads and tails level with the ground and swinging their tails like rudders. They often course along roads, paths and dry stream beds, and walk and run on their daily rounds for food and to patrol their territories. These ground-cuckoos can fly, though when they survey their area from an elevated perch they stay within a hop and a flap of the ground. The nesting ground-cuckoos build a nest and rear their own young nestlings which have conspicuous colored spots inside their mouths. The ground-cuckoos also include three species that are brood-parasitic (two species of Dromococcyx, both shy and secretive, and the American Striped Cuckoo Tapera naevia), and are forest dwellers.They parasitize several species of passerine hosts that breed in covered or domed nests, and occasional other hosts that breed in open nests. 3. Centropodinae Coucals are large groundforaging birds with long stout feet, a long straight

Introduction to the cuckoos 7 claw on the inner hind toe in most species (the claw is the “foot spur” of the genus Centropus), short rounded wings and a long broad tail. The plumage is black, rufous brown, or white, or a combination of these colors; the plumage is shaggy and in many species the neck and breast have spiny hackles. Coucals live in the Old World, mainly in the tropics from Africa and Asia to Australia, New Guinea and the Solomon Islands, and on many islands in Wallacea and the Papuan region. Coucals live in forest, swamp and marsh habitat, and spend most of their time on the ground, and in dense marshes they are most often seen in their slow and awkward flight. Most coucals build large globular or domed nests of grass and leaves with a side entrance. Most make deep resonant cooing calls, some species in a “water-bubbling” pattern which recalls the gurgling sound of water as it rushes from a narrow-necked bottle, and some species call in duets. Nestling coucals have an odd set of stiff natal down (trichoptiles) and as the body feathers grow these hair-like structures are attached to the tips of growing feathers, then fall off when the birds fledge; and they have conspicuous raised colored spots on the palate.Adult coucals have a pheasant-like body form with long, robust legs and a long tail, coarse plumage with stiff hackles, and a bronchial syrinx. Males of some coucals have a single testis while others have two testes; the number can vary between one and two within a species and a local population (Mayr 1937, Mayr and Rand 1937, Rand 1942a).

4. Couinae These are large colorful groundcuckoos of Madagascar and southeast Asia and have long tarsi. Ground-cuckoos Carpococcyx of tropical forests in Asia have plumages of dove-like greens and blues, bright bare skin around the face and long eye lashes, and they build flat platform nests. Couas Coua are large cuckoos in Madagascar. They have soft and lax plumage with dove-like colors of pastel pinks, purples and peach (blue in the Blue Coua Coua caerulea), a long tail and large feet, colorful bare areas of skin around the eye, and long eye-lashes. Most are terrestrial. The arboreal Blue Coua has long legs like the terrestrial couas and like them it walks, but in the trees.The Red-capped Coua Coua ruficeps is mainly terrestrial and walks on the ground,

and along branches when it is in a tree (Berger 1960). One species of coua, a snail-eating specialist on Ile Sainte Marie northeast of Madagascar, was last seen around 1835. This is the only cuckoo species known to have become extinct in the past two centuries. Nestling couas and Asian groundcuckoos lack natal down, and the palate and tongue have conspicuous patterns of raised white or blue spots that contrast with the red mouth lining and may be a signal to their parents during parental care.

5. Cuculinae These include the long-tailed arboreal nesting cuckoos of the Old World and New World.The Old World nesting birds in this group are known as the malkohas and are now grouped into six genera in Asia: Rhinortha, Taccocua, Zanclostomus, Phaenicophaeus, Dasylophus and Rhamphococcyx, and a seventh, Ceuthmochares, is found in Africa.“Malkoha” is a Sinhalese name in Sri Lanka (Ceylon) for the Red-faced Malkoha Phaenicophaeus pyrrhocephalus. It has been widely used for other species in the Oriental Region (Levaillant 1806, Raffles 1822, Jerdon 1862) and as a generic name Malcoha Schinz 1821. Malkohas have strong, unfeathered tarsi, short rounded wings, and long graduated tails tipped with bold spots and bars. Malkohas are skulking arboreal cuckoos of forests and thickets. Many have brightly colored bare skin around the eye and face, and the bill is large, often arched and brightly colored. The German term for the birds, “Buntschnabelkuckucken,” translates as the descriptive “colorfulbilled cuckoos.” The wing feathers often have wide vanes and the body feathers have stiff, shiny shafts in a color that contrasts with the rest of the feather. Malkohas build shallow nests, lay chalky white eggs and rear their own young. The malkohas are brightly-colored in plumage, with swollen and brightly colored bills, short wings and long tails. All are arboreal except for Sirkeer Malkoha Taccocua leschenaultii which is terrestrial. A cultural awareness of cuckoos as brood parasites in Europe caused some ornithologists to guess that the other cuckoos were also brood parasites, and Newton (1896) suggested that malkohas were brood-parasitic, but field ornithologists in India and the Malay Peninsula, Borneo and Sulawesi had by that time observed adult malkohas at their own nests. Little is known

8 The Cuckoos about the details of nesting such as the incubation period, nestling period and parental behavior in these birds. The brood-parasitic Old World crested cuckoos Clamator appear to be closely related to the malkohas. Several cuckoos of the New World are the ecological counterparts of malkohas and are closely related to them. Most are arboreal insect-eaters and have long tails. The coccyzine cuckoos including Coccyzus have long wings, and several species are long-distance migrants between continental North and South America, or between temperate South America where they breed and tropical South America where they winter. Others are large-bodied cuckoos that live on islands in the Caribbean and feed on lizards. The New World cuckoos build saucer-shaped nests, lay white or blue eggs, and rear their own young. The young have raised papillae inside the mouth; the color of the papillae contrasts with the palate. Cuculines also include the brood-parasitic cuckoos of the Old World, the tribe Cuculini. Many are migratory with long, narrow, pointed wings, long tails, and short legs with the tarsus feathered at the base. The tail is often graduated with the central feathers longer than the others. As noted by Aristotle, Common Cuckoos Cuculus canorus look like hawks (notably Merlin Falco

columbarius) in plumage, shape and flight; and now we know the hawk-cuckoos Hierococcyx have rounded wings like bird-hawks (Accipiter). Most brood-parasitic cuckoos are dull in plumage, and their cryptic appearance may help them to avoid detection by their hosts (Payne 1967). The parasitic glossy cuckoos Chrysococcyx of Africa and Australasia, however, are conspicuous in their colorful plumage. They are small and at first glance could pass for insect-eating songbirds, and sometimes they even feed in mixed-species flocks with songbirds. Old World brood-parasitic cuckoos lay their eggs in the nests of many passerine host species. Their eggs vary from blue and green to chocolate brown to white, spotted or unspotted, and some species of Old World brood-parasitic cuckoos have more than one kind of egg marked by different colors and patterns. The nestlings are naked when they hatch (young Shining Bronzecuckoos Chrysococcyx lucidus and Little Bronzecuckoos have sparse natal down). Inside the mouth the young lacks visual effects except for a healthy color of red, orange or yellow.The young of one brood-parasite, African Thick-billed Cuckoo, however, have a bright orange palate with contrasting pale spots, and nestling Rufous Hawk-cuckoo Heirococcyx hyperythrus have a palate of yellow and pink.

2 Distribution, habitats and conservation status

Distribution Cuckoos occur around the world in tropical and temperate regions. More species live in the Old World (n ⫽ 109) than in the New World (n ⫽ 32). The northernmost cuckoos (Common Cuckoo Cuculus canorus, Oriental Cuckoo C. optatus, Indian Cuckoo C. micropterus,Asian Lesser Cuckoo C. poliocephalus) occur in northern Europe and Russia in summer.They are migrant brood-parasites that lay in the nests of migrant Palearctic songbirds and then winter in the tropics, sometimes crossing the equator and even passing over the Indian Ocean from Asia to Africa. In the New World the two species (Yellowbilled Cuckoo, Black-billed Cuckoo Coccyzus erythropthalmus) that breed in the United States and Canada are mainly warm-weather birds that spend most of the year in the New World tropics. No cuckoo species occurs regularly and breeds in both the Old World and the New World, although vagrants from northeastern Asia (Common Cuckoo and Oriental Cuckoo) turn up in the Alaskan archipelago and northern Alaska in migration season, and vagrant Black-billed Cuckoos and Yellow-billed Cuckoos from northeastern North America show up in many years in Europe from Iceland to Italy. Cuckoos are widespread as breeding birds on islands, especially in the Indonesian archipelago and the New Guinea region. Migrant cuckoos occur on these islands in winter, mainly species from the northern hemisphere; Australian cuckoos also live on the islands during the austral winter. Cuckoos are not as likely to turn up on oceanic islands as are fruit-eating pigeons and doves, nectar-feeding white-eyes or the rails, yet migrant Oriental

Cuckoos from northern Asia occur as far from their breeding grounds as Palau in the western Pacific and Australia in the Southern Ocean. Perhaps the most widely-traveled species is Long-tailed Cuckoo Urodynamis taitensis which migrates from its breeding area in New Zealand to winter on islands in Melanesia, Micronesia and Polynesia across a wide area of the tropical Pacific. Three species (Common Cuckoo, Asian Lesser Cuckoo, Jacobin Cuckoo Clamator jacobinus) migrate from Asia to Africa for the non-breeding season, and some birds appear during migration season on islands in the Indian Ocean. Cuckoos are unusual birds in this India-Africa migration pattern. All major clades of cuckoos occur in South America, Africa, and southern Asia through to Australia.The biogeography of cuckoos suggests an ancient origin and radiation of cuckoos in the southern region of the world. Geologically this area was the ancient tectonic plate Gondwanaland. Cuckoos in the northern region are all closely related to cuckoos in the southern region and appear to have derived from the southern cuckoos. The ancient diversity of cuckoos that is suggested by the genetic differences between major groups (Sibley and Ahlquist 1990) point to an early origin of the major clades, perhaps before the Eocene and at the time of the breakup of the Gondwana plate; yet there is no independent calibration of this date from dated fossils. The North American Coccyzus (including the large lizard-cuckoos “Saurothera” and “Hyetornis” of the West Indies) is closely related to the Old World malkohas and appears to have been derived from them, perhaps following the dispersal of ancestral nesting cuculine cuckoos by way of

10 The Cuckoos South America.Within the Old World the distributions of certain cuckoo groups suggested to some workers, an initial separation traced to plate tectonic events. Marchant (1972) suggested this origin for the biogeographic separation of Chrysococcyx glossycuckoos of Australasia and Africa, and Cracraft (1988) suggested the same for the Carpococcyx groundcuckoo species in southeast Asia, Borneo and Sumatra. So far there is no evidence that the genetic distances observed between the clades and species of cuckoos correspond with the genetic distances expected from these tectonic events, based on mutation rates and the length of time since the events. The goal of biogeographic models should be to test whether the geological and the genetic times of divergence coincide (Zink et al. 2000).

Habitats Cuckoos are mainly arboreal birds of forest and woodland. Their habitats are varied but include many primary, more or less undisturbed tropical rain forests. Other cuckoos live in secondary forest and some of these have adapted to life in tree plantations, particularly that of cocoa where larger trees provide shade for the cocoa plants.The forests they inhabit range from primary evergreen forests to semi-arid scrub in Madagascar. Cuckoos often occur in the canopy and on dense vines and rattan in Asia, and the ground cuckoos are mainly forest birds. A few cuckoos live in coastal mangroves (Mangrove Cuckoo Coccyzus minor, some malkohas and coucals in southern Asia, and some broodparasitic cuckoos in Australasia). Cuckoos also live in open brushy country in Australia and the plateau country of Africa, in the southwestern United States, Mexico and the drier parts of Central America, and in southern South America. The diversity of cuckoo behavior is highest in the New World tropics. Here, many cuckoo species nest as pairs and rear their own young, others are cooperative breeders that care for the young of others in their social group, and three are brood parasites which do not rear their own young. Cuckoos are especially numerous and diverse in Amazonian Brazil. Near Manaus there are seven species of cuckoos, only one a brood parasite, one a cooperative breeder, and

others solitary birds, or ones that join mixed-feeding flocks but nest in pairs away from other birds (CohnHaft et al. 1997). In one of the largest forest reserves in the world,Tapajós National Park, with 448 species of birds, seven are cuckoos (Oren and Parker 1997); in Jaú NP, with 445 species, nine are cuckoos (Borges et al. 2001); on the Middle Rio Jiparaná, Rondônia, with 459 species, 12 are cuckoos; and in the Alta Floresta region in extreme northern Mato Grosso, with 474 species, nine are cuckoos including a nonbreeding austral migrant (Zimmer et al. 1997). In all these areas most cuckoos are of the solitary nesting species. The Atlantic coastal forests in Brazil have as many as ten cuckoos, although one, a subspecies of ground-cuckoo, may now be extinct in that region (Parker and Goerck 1997). The number of cuckoo species is highest in forests of the Old World tropics. In southern Asia at a local scale of a 2 km2 area in the Malay Peninsula, Khaso Nor Chuchi has 247 breeding bird species with 29 resident cuckoos (Round and Treesucon 1997), the Pasoh Forest Reserve has 188 species of birds and the Kerau Wildlife Reserve at Kuala Lompat has 195 species, each site within the geographic range of 21 cuckoo species (Wells 1999).The Tekam Forest Reserve has 225 species with 12 resident cuckoo species, ( Johns 1989). In the Vu Quang Nature Reserve,Vietnam, there are 220 resident bird species with 9 cuckoos, and in 14 other sites in Vietnam there are 16 cuckoo species (Eames et al. 2001). On a larger scale, the Malay Peninsula has 686 species with 22 cuckoos that are unlikely to breed there (Wells 1999) and Sabah has 339 resident species with 21 cuckoos (Sheldon et al. 2001). Island areas are less species-rich both in birds on the whole, and in cuckoos. In the Lesser Sunda Islands, 251 species are known on Flores, nine of them cuckoos (Verhoeye and Holmes 1998); and 182 species on Sumba, six being cuckoos (Linsley et al. 1998). In Central Africa, as many as 16 resident species of cuckoos live together in forests and on forest edges. Four are coucals Centropus species, one is a malkoha, Chattering Yellowbill Ceuthmochares aereus, and the others are brood parasites that lay in nests of forest passerines. In Cameroon in Lobéké FR where 305 species of birds are known there are 13 cuckoos, 10 of them in forest habitat; and Mt. Kupe has 324 resident

Distribution, habitats and conservation status 11 species with 10 cuckoos in Congo-Brazzaville Nouabalé-Ndoki has 273 species with 12 cuckoos, Odzala NP has 435 species with 17 cuckoos, and the Kouilou basin has 425 species with 16 cuckoos (Dowsett and Dowsett-Lemaire 1997, DowsettLemaire and Dowsett 2000, Bowden 2001), in the Central African Republic Dzanga-Sangha has 360 species with 11 cuckoos (Green and Carroll 1991, P. Beresford in litt., AMNH) and in Uganda Kibale Forest has 410 species with 13 cuckoos (Struhsaker 1997). In each of these Old World areas, more than half the cuckoo species are brood-parasitic. The difference in the number of cuckoo species between the New World and Old World lies mainly in the greater number of brood parasites in the Old World, and this is a result of different evolutionary histories, rather than any obvious ecological difference between the two regions. Cuckoos live not only in the warm tropics but in cool and wet conditions as well, in forests and in more open habitats. When their plumage becomes wet, the cuckoos use the sun to dry and these birds are best seen when they expose themselves high on a perch in the early morning or after a shower, spreading their wings and tail and lifting their back feathers, exposing the skin to the sun. Sunning behavior is well known in coucals and glossy parasitic cuckoos in the Old World, and in Coccyzus cuckoos, roadrunners Geococcyx californianus and anis Crotophaga and Guira in the New World. Ani plumage becomes soaked as the birds forage in wet grass and herbaceous vegetation. Guira Cuckoos Guira guira have heavily pigmented black skin between the dorsal feather tracts, and in the morning on cold days the birds droop their wings, turn their back and expose the black skin toward the sun to dry the plumage and warm the body. Anis adapt to cool weather and to poor food conditions by becoming semi-torpid; they drop their body temperature a few degrees at night, and when their body temperature drops the anis can still fly after two days of fasting (Warren 1960). A few cuckoos live in seasonally hot, arid and semi-arid habitats. These include roadrunners in the New World, a few couas Coua in Madagascar, and several brood parasitic cuckoos in Australia.

While the dry-country brood parasites escape the extreme conditions when they migrate, the nesting cuckoos typically live in the deserts as residents. Cuckoos have adapted well to human changes in habitats in some areas. In the New World tropics, anis Crotophaga have spread with forest clearings. In southeast Asia the koels Eudynamys scolopacea have spread to new areas and new host species that do well in towns and cities. In Australia the extensive plantings of gardens and flowering trees have attracted nectar-feeding birds. In southeastern Australia, Common Koels and Channel-billed Cuckoos have increased in numbers in the past ten to twenty years along with their host species, and in southwestern Western Australia Pallid Cuckoo Cacomantis pallidus are common where the large honeyeaters especially Red Wattlebird Anthochaera carunculata live in parks and gardens. Certain ground-living cuckoos including Gabon Coucal Centropus anselli, Red-billed Ground-cuckoo Carpococcyx renauldi and Red-capped Coua Coua ruficeps are often seen when they visit human areas and feed on camp refuse such as rice and noodles and on the rich insect life downstream from camp latrines.

Conservation status Although cuckoos are at some risk wherever their habitat is shared with humans, most cuckoos are not at immediate risk of extinction for at least another generation of humans. Unless otherwise indicated in the species accounts, the cuckoos are not considered threatened, when judged by the criteria in Collar et al. (1994) and BirdLife International (2001). The cuckoos most at risk are forest cuckoos. Tropical evergreen forests, the unique achievement of millions of years of nature, are being destroyed by logging for widescale economic consumption, and human populations are large and growing and are removing forest to cultivate open lands. As their forests are logged, many cuckoos are unable to adapt to any remaining scrub and secondary forest that might survive. Furthermore, forests once logged are open and vulnerable to fires and encroachment. In the past two decades, fires have burned millions of

12 The Cuckoos hectares of logged forests in Borneo and Sumatra (MacKinnon et al. 1996, Curran et al. 1999, Francis 2001). Global-imaging satellite records of change in humid forests of tropical regions between 1990 and 1997 have determined an annual deforestation rate of 1–4% in southeast Asia, 1–5% in Africa and Madagascar, and 3–4% in the upper Amazon (Achard et al. 2002). In the Old World the cuckoos at risk include the Red-faced Malkoha Phaenicophaeus pyrrhocephalus, Sumatran Ground-Cuckoo Carpococcyx viridis, Black-hooded Coucal Centropus steerii, Javan Coucal C. nigrorufus and Green-billed Coucal C. chlororhynchos. In the New World there are questions about the continued existence of Cocos Cuckoo Coccyzus ferrugineus, Rufous-breasted Cuckoo C. rufigularis and Banded Ground-cuckoo Neomorphus radiolosus (Collar et al. 1994, BirdLife International 2001). Several other species and subspecies are considered near-threatened (Collar et al. 1994), especially birds whose distributional range is small and limited to islands (Coral-billed Ground-cuckoo Carpococcyx renauldi,Verreaux’s Coua Coua verreauxi, Kai Islands race of Pheasant-coucal Centropus phasianinus spilopterus, Biak Coucal C. chalybeus, Short-toed Coucal C. rectunguis, Rufous Coucal C. unirufus and Scaled Ground-cuckoo Neomorphus geoffroyi squamiger. Although no brood-parasitic cuckoos appear on the lists of threatened and endangered birds, the loss of forests will have an impact on these birds, as well as on their host species. Most critically, the lowland and montane forests in the Malesian tropics of Malaysia, Indonesia and the Philippines are at risk, and with loss of forests these areas and the world are in danger of losing their cuckoos and the rest of their birds (Whitten et al. 1987, 1996, Collins et al. 1991, Sayer et al. 1992, Heaney and Regalado 1998, Kennedy et al. 2000, BirdLife International 2001). At least one brood-parasitic cuckoo should be added to the list of species at risk, Philippine Drongo-cuckoo Surniculus velutinus. Formerly regarded as a subspecies of a single species of drongo-cuckoo, the drongo-cuckoos are now known as four distinct species on the basis of differences in songs and in their plumage. More than 90% of forest habitat in the Philippines has been lost in recent years.

Preservation of their forest habitats is the only way to maintain the numbers and diversity of birds: necessary but perhaps not sufficient if hunting and trapping of birds continue. In addition to forest cuckoos, the cuckoos of open woodlands have seriously decreased in numbers. In Brazil, in the interior part of São Paulo State the number of species reported by earlier observers but no longer seen, is much greater than the number of species that have disappeared in the humid coastal areas, where the remaining forests are protected by law. Even there, birds are hunted and captured for the cage bird trade, and habitats of bamboo and coastal flats are developed for human housing. In the mesic and drier inland region of São Paulo, the open woodland savannas (“cerrado”) have been destroyed for large ranches, replaced by sugar cane for alcohol production, and coffee and oranges for export crops, and now the small dry forest and savanna fragments cover less than 1% of their original area. In inland regions of Brazil, Pearl-breasted Cuckoo Coccyzus euleri have little remnant habitat and their numbers are much reduced (Willis and Oniki 1992). In North America, Greater Roadrunners were controlled with a federal or state bounty on their heads as predators of quail, early in the twentieth century. In fact, these cuckoos hardly ever take quail, whereas they often take crop and household pests. Even though roadrunners have a traditional cultural status in the North American southwest (Dobie 1939), they are often shot as targets and trophies. Large ground-cuckoos undergo local extinction when their small forest patches support only a few birds, and the destruction of nearby habitats results in the loss of opportunity for birds in neighboring areas to disperse into the remnant forests. Local extinction has been documented on Barro Colorado Island, Panama, a forested hilltop that between 1912 and 1914 was isolated from the mainland forest when the Chagres River was dammed to form the Panama Canal.The area of the island is 1564 hectares, or six square miles.The site was designated a biological reserve in 1923 and has been the focus of long-term behavioral and ecological studies of birds and other tropical life (Eisenmann 1952, Willis 1979, Leigh 1999). When

Distribution, habitats and conservation status 13 the nearest populations of resident sedentary birds were lost, as forest habitats were cut around Gatun Lake, no source populations remained to repopulate the island. Rufous-Vented Ground-cuckoo Neomorphus geoffroyi was the first bird to become extinct on the island; it was last seen in 1935. The brood-parasitic Pheasant Cuckoo Dromococcyx phasianellus, was the second cuckoo lost; it was last seen on the island in 1971 as its shrub habitat was overgrown by forest (Willis and Eisenmann 1979). Of special interest and concern is Sumatran Ground-cuckoo Carpococcyx viridis, a fruit-eating bird that had not been seen by western ornithologists since 1916, but was rediscovered when one was captured in a live trap, photographed and released in 1997; another one was seen at another site in 2000 (Zetra et al. 2002). Sumatran Ground-cuckoo may persist in dipterocarp forests, where vulnerability to snaring and hunting, the widespread failure of these forests to fruit in certain years ( Janzen 1974, Whitten et al. 1987), and the loss and fragmentation of forests in response to logging and burning which restrict the dispersal of large vertebrates (Curran et al. 1999, Curran and Leighton 2000), all threaten the survival of large forest animals including the Sumatran Ground-cuckoo. The only cuckoo that is known to have become extinct as a species in historical times is Delalande’s or Snail-eating Coua Coua delalandei, which disappeared from Madagascar around 1835 during a time of deforestation. Hunting may have been involved in its extinction, along with the introduction of rats which depleted the terrestrial snails that were the coucals’ food. In regions where forest habitats have been destroyed, the species that do well in disturbed habitats have become more common and the forest species have become rare.This has led to some concern that replacement of the rare species is a result of competitive exclusion by the more common species. For example, the rare Black-hooded Coucal Centropus steerii and the more widespread Philippine Coucal C. viridis live in cut-over former forested lands in Mindanao. Elsewhere in southeast Asia and the Sunda Archipelago the uncommon forest-living Short-toed Coucal C. rectunguis, Javan Coucal C. nigrorufus and Green-billed Coucal

C. chlororhynchos are decreasing and their place taken by the widespread Greater Coucal C. sinensis. Disappearance of these rare coucals results from destruction of their habitat. It is unlikely the rare coucals would recover their numbers if the more common coucals were controlled. Conservation efforts are better applied to the protection of a diversity of habitats that are suitable for these rare coucals (BirdLife International 2001). Humans have used cuckoos for food and traditional medicine, (Payne 1997b). The cultural practice of folk medicine has declined and appears to be disappearing in rural areas, as in the use of the rare Green-billed Coucal in Sri Lanka (Wijesinghe 1999) and anis in the West Indies. Host species populations which the broodparasitic cuckoos parasitize are not a risk to cuckoos and their brood parasitism is not a conservation concern.When their hosts are thin on the ground, their nesting birds may be too few to support the cuckoos, because a population of brood parasites needs a large number of nesting hosts in order to maintain their own numbers from one generation to the next. Certain cuckoo species concentrate on only one or a few host species, and the population numbers of cuckoos are likely to be lower than the numbers of their hosts. Deteriorating habitat conditions for cuckoo hosts would drive the cuckoos into local extinction well before the hosts become threatened or endangered owing to cuckoo parasitism. In this sense the brood-parasitic cuckoos may be indicators of the environmental health of an Old World forest. Cuckoos depend on a healthy insect population for food and survival, and the brood-parasitic cuckoos depend on their hosts finding enough food to rear the cuckoo young. Environmental disturbances that lead to the reduction of hosts also lead to the loss of brood-parasitic cuckoos, and disturbances that reduce the area of wooded habitats and their insects lead to the decrease of other cuckoos as well. Cuckoos are among the first birds to disappear when a habitat is polluted by industrial and chemical waste. Their caterpillar prey accumulate toxins from chemical pollution. In North America, migratory cuckoos that flew into television transmitting towers in the early 1970s had tissues with significant amounts of chlorinated hydrocarbons. Concentrations of pesticide

14 The Cuckoos metabolites in Yellow-billed Cuckoos were higher in autumn than in spring, and were higher in autumn adults than in juveniles, so the time and context suggest that pesticides were acquired on the breeding grounds (Grocki and Johnston 1974). Finally, Greater Roadrunners in North America have disappeared from urban areas and highways as a result of human impact, as well as disturbance and predation from feral and domestic dogs and cats (Emlen 1974). Roadrunners need extensive areas of undisturbed open shrub and grassy habitats with abundant populations of reptile and insect prey, and these are destroyed with the development of agriculture, urban areas, and the fragmentation of expansive tracts of grassland and shrub lands. Few cuckoos are kept in zoos and none have bred in captivity on a large scale that could provide enough birds to reintroduce cuckoos into areas where they have been lost in the wild and where the habitat remains. Cuckoos that have been maintained in captivity where the bright bare colors of their heads and their bright plumage can be well seen by visiting people, include Red-billed Malkoha Zanclostomus javanicus and Chestnut-breasted Malkoha Phaenicophaeus curvirostris. Cuckoos are difficult to breed in captivity

(Pagel 1992), although a few cuckoos, (Great Spotted Cuckoo Clamator glandarius, Common Cuckoo, Guira Cuckoo, Greater Roadrunner and Coral-billed Ground-cuckoo) have bred at zoos, including the National Zoo in Washington, D.C. (Muller 1971) and Vogelpark Walsrode in Germany (Marcodes and Rinke 2000). Captive breeding has been useful for behavioral observations, especially in the Old World ground-cuckoos: behavior and breeding biology are better known through avicultural studies than on the field, where these birds are seldom seen and are nearly unknown (Robiller et al. 1992). Most cuckoos are shy and skulking with humans. The most easily watched in the field are the roadrunners of the southwestern United States and the social group-living Guira Cuckoos and anis in the New World tropics, where these birds are often tame and allow a close approach. Cuckoos are not agricultural pests, and the brood-parasitic cuckoos are not threats to the breeding populations of their hosts. Cuckoos are most interesting to naturalists and biologists who are fascinated with their variety of behavior, the evolution of their sociality, their mating systems and breeding styles which include monogamy, cooperative breeding and brood parasitism.

3 Behavior

Social and parental behavior Cuckoos are solitary, skulking birds, more often heard than seen. Their calls in forests and woodlands are among the most conspicuous and compelling sounds of the tropics. As a group the cuckoos are arboreal and spend their time in trees, although several couas of Madagascar, most coucals in the tropics and some cuckoos in semidesert areas are terrestrial. Many cuckoos live by themselves and call only in the breeding season, while a few kinds of cuckoos live in social groups, leaving their group perhaps only once throughout their lives, when they disperse and join another social group. Cuckoos are mainly diurnal, but both the nesting cuckoos and the brood-parasitic cuckoos call at night along with the nocturnal owls and nightjars. It is in their social and parental behaviors that cuckoos are particularly variable, with monogamous pairs that rear their own young, cooperative breeding birds that rear each others’ young, and brood-parasitic cuckoos that leave their young to be reared in the nests of other species of birds. Many cuckoos rear their own young. Some cuckoos are highly social, living together in permanent social groups where they feed and breed together. Several pairs build a nest and rear their young together in a common nest.These cooperatively breeding cuckoos, the anis and the Guira Cuckoo, live in the New World. The broodparasitic cuckoos, however, have no obvious longterm pair bond. Male and female live apart from each other and come together only to copulate. There is also no family bond between offspring

and their own parents. Females lay their eggs in the nest of another species, the host, and the hosts are foster parents to the young cuckoos. Female cuckoos also remove a host egg to make room for the nestling cuckoo which then removes the others by evicting them from the nest, pushing its own naked body under the eggs and lifting them over the rim of the nest. This behavior ensures that all the parental care of the foster parents will be delivered to the nestling cuckoo.To the host pair the nestling cuckoo means the loss of their own nesting effort. Many cuckoo hosts remove an egg from their nest if it looks unlike their own eggs, and this rejection behavior is the main line of defense against cuckoo parasitism.When the young cuckoo hatches, however, the foster parents accept the cuckoo and rear it at the cost of their own young.This host-parasite association has led to evolutionary battles which each member of the association has won in turn. First the hosts evolve a sensitivity to the cuckoo’s egg in their nest and remove it. Then the cuckoos evolve the ability to match the color and pattern of the host egg (Davies 2000). The nestling cuckoo takes the theme of sibling rivalry to its infanticidal extreme by killing its nestmates. In these broodparasites the rivals, the offspring of the foster parents, are not kin but are competitors for parental care. Cuckoos affect the breeding success of their hosts, both when the female cuckoo removes a host egg from the nest, and when the nestling cuckoo removes the rest of the hosts’ genetic interest by evicting the eggs and nestlings from the nest. Much of our interest in cuckoos and their eggs and behaviors is focused on this brood-parasitic behavior.

16 The Cuckoos

Food and feeding Cuckoos are generalist predators and mainly take large insects. The diet of many cuckoos consists of noxious, brightly colored and hairy forms of caterpillar. These include urticating caterpillars, whose hairs pierce the skin and histamines causing itching and burning, at least when they contact human skin. Cuckoos also take grasshoppers and locusts, many with distasteful “tobacco” juice; other kinds with hard legs and shields. Other cuckoo foods include large non-insect arthropods like millipedes, centipedes, spiders and phalangids, and terrestrial snails and small vertebrates such as tree-frogs.A few cuckoos concentrate on lizards. Roadrunners take many reptiles, and the lizard-cuckoos on Caribbean islands where there are no lizard hawks, feed on lizards. In addition, the brood-parasitic cuckoos take eggs and nestlings from the nests of their hosts, and the coucals take nestling birds when they find them. Old World malkohas and the arboreal New World cuckoos forage by moving through the vines and branches of tropical forests and thickets with hops, twisting the tail for balance like a squirrel, then turning and seizing an insect spotted on a leaf or axil. Most Old World brood-parasitic cuckoos feed in much the same manner, perching and peering in a characteristic posture (Figure 3.1). Cuckoos hunt by ambush as they perch motionless for many minutes and peer about from a perch, then dash or fly to a caterpillar when they sight it on the underside of a leaf or on a tree trunk, grab the insect, and return to their perch to eat it.Their feeding behavior is one of “peer and pounce.” Caterpillars have guts full of indigestible and toxic leaf products, and the cuckoos clean the caterpillars and remove these products. The cuckoo bites off the end and wipes the caterpillar back and forth on a branch until the guts are pressed out, or passes it back and forth through the bill to clean the insides, then swallows the insect. Or else it beats hairy caterpillars against a branch, removing their gut contents and toxins before eating them. These gut removal techniques do not remove the caterpillar hairs. The cuckoo swallows the hairs, which form a felted mat lining in the stomach, and the hairs are regurgitated as a pellet.

Figure 3.1. Fledgling Klaas’s Cuckoo Chrysococcyx klaas, Hans Merensky Nature Reserve,Transvaal.

North American Black-billed and Yellow-billed Cuckoos forage on the tent caterpillars Malacosoma spp. The cuckoos, along with orioles and other birds (60 species are known to feed on these insects, Witter and Kulman 1972), are attracted to wooded areas where the caterpillar tents hang on hawthorns, wild cherries and other rosaceous trees and shrubs.The caterpillars build a communal tent of silk where they retreat between foraging bouts in the tree (Fitzgerald 1995). Although their tent protects the caterpillars, many insectivorous birds pierce the tent and remove the caterpillars. Birds also feed on them as they move along marked foraging tracks. Cuckoos often nest in sites with an abundance of tent caterpillars, M. americanum in spring and M. disstria in autumn. Outbreaks of tent caterpillar in the northern states can defoliate trees each year for up to six years; in the southern states the forests can be defoliated annually for 20 to 30 years (Fitzgerald 1975). The clutch size of Blackbilled Cuckoos varies with the abundance of caterpillars, with large numbers of eggs laid when the

Behavior 17 number of caterpillars is high (Forbush 1927, Sealy 1978). The variation in tent caterpillar abundance may affect the numbers of breeding cuckoos. It is unlikely, however, that cuckoos control the populations of pest caterpillars, although over a few days a cuckoo can take hundreds of caterpillars and clean out a tree of all its caterpillars. Other large insects that appear in outbreaks are also taken, including larvae of gypsy moths Lymantria dyspar ( Jauvin 1996). Caterpillars on economically important trees are more often controlled with the application of chemical pesticides, and bacteria and viruses as biological control agents. The chemical pesticides appear in cuckoo tissues, and from the cuckoo’s point of view it is not a good idea to spray these pesticides in the woods. Coucals take large insects and small vertebrates including snakes, lizards, tree frogs, mice and rats, and small birds.These large cuckoos feed mainly on the ground, often in thick scrub and less commonly in tangles of vines and on the branches of trees. They are generally slow and clumsy in movement but move rapidly when they locate their prey. When hunting grasshoppers and lizards on the ground, they move forward with a slow stalking walk. Then when close, they change to a hop and run, and ambush or chase the prey. Their feeding behavior is one of “flush and rush.” At other times, they tear open bird nests to get the eggs and young, and rip bark from trees to get the lizards and large insects that are concealed underneath the bark. Small birds recognize coucals as predators and they mob the larger coucals. In marshes and swamps, coucals feed partly in the water, where they take frogs, crabs, and aquatic insects, and occasionally they scavenge for dead fish. In seasonally dry habitats, coucals respond to opportunity when fires burn the grass during the dry season. Like foraging Cattle Egrets Egretta ibis, White-browed Coucals Centropus superciliosus approach the smoke and feed at the edge of grass fires, where they take large insects and small mammals that attempt to escape the flames as the fire advances. Other foods include snails, which are rich in protein and in calcium. Senegal Coucals C. senegalensis take large, brightly colored, apparently aposematic bush locusts that give warning displays. They attack and

subdue the locust, then soften it with the bill, working the abdomen and thorax. They then hold the insect underfoot and remove the head and digestive tract and eat the remainder (Goodwin 2001). Anis and Guira Cuckoos are mainly ground feeders. Groove-billed Anis and Guira Cuckoos live in meadows and pastures in dry open habitats, Smooth-billed Anis in denser herbs in wetter habitats, and Greater Anis in wet forest edges. Ani species that live in habitats with larger and wetter leaves have the largest bill: The anis use the bill as a wedge to move through leaves and feed (Willis 1983b), to open the soft wet earth and grab burrowing insects, and to plow through cow dung to get insects (Gosse 1847). The name of the ani genus Crotophaga means tick-eater, a behavior that was more common in the days before chemical pesticides were used in management of livestock.An old German term for anis is “Madenfresser”, eater of ticks (Leverkühn 1894). Groove-billed Anis take grasshoppers and follow cattle much like the icterid cowbirds. Anis also take tuks from the backs of cattle. Anis bite hard-bodied insects repeatedly before they swallow or feed the insects to their young. Smooth-billed Anis sometimes feed on butterflies. In a rapid approach the bird runs to a group of butterflies, lunges, captures one, swallows it, then tries to capture and swallow another, as long as the butterflies remain: Or the bird walks slowly in a crouch towards a single butterfly, then lunges and grabs it. Adult anis are more successful at capturing a large butterfly than are the young birds. Members of an ani group forage near each other, and when they are out of sight of each other in the grass they stay in contact with loud calls. Anis feeding with a cow or horse hop near the front of the mammal and seize insects stirred up in the grass.They catch more insects per minute when they feed with cattle than when they hunt alone, especially in the dry season. Anis in the forest use ant swarms as beaters, catching roaches and other insects as they are driven from the litter by the ants (Sutton 1951, Rand 1953, Smith 1971, Skutch 1983, Willis 1983b, Burger and Gochfeld 2001). Roadrunners feed on the ground. They flush prey as the birds walk with the head low and brush

18 The Cuckoos the grass: The disturbance startles the insects, and when they jump, the roadrunner grabs them. They also flush out insects when they jump up and flap their wings slowly over the ground, then catch the insect in flight or watch it land and pick it from the vegetation.They dart into clumps of vegetation for large insects such as grasshoppers and crickets, spiders as large as tarantulas, and scorpions, and they take vertebrates such as small snakes, mice, small birds and the contents of bird nests.They run down lizards in fast pursuit. Roadrunners leap at flying insects and birds such as swifts that swoop over a dry stream, and they ambush hummingbirds at nectar sites where these visit and feed. They take young bats that fall from the ceilings of caves. In winter they take beetles and grasshoppers, finding insects around the base of rocks.They also take fruits, seeds and scattered grain when insects and reptiles are scarce. Roadrunners eat the sweet seedy fruit of the prickly pear cactus Opuntia, knocking it to the ground and tumbling it to remove the spiny cover: These fruits can make up 10% of a bird’s diet in fruiting season. Following a hand-reared free-living roadrunner through the Texas brush, Sutton (1913, 1915) recorded the food it took in a day: 263 hopping grasshoppers and 73 flying grasshoppers, 17 scorpions, 28 sowbugs, 7 caterpillars, 3 chrysalids, 14 angleworms, 39 moths, 1 butterfly, 14 centipedes, 16 spiders, 2 tarantulas, 3 walking-sticks, 3 small toads, 3 horned lizards, 14 other lizards and a mouse. When roadrunners encounter a rattlesnake, they crouch and circle it, then dart in to grab it behind the head, which they hold in their bill. Then they whip and pound the snake against the ground or stones until it no longer moves. After it is dead they pull the snake into the shade of a shrub and eat it. They also toss and batter other snakes and lizards in this way to kill their prey. Birds are killed and plucked, small mammals are killed by a blow to the base of the skull, and larger prey such as lizards, snakes and ground squirrels are held in the bill, then swung and hit repeatedly against a rock, stick or the ground. The beating kills the prey, separates and crushes the skeleton, compacts the meal and allows the bird to swallow its prey whole. Food processing may take up to 15 minutes. Roadrunners kill scorpions by biting at the tail where the poison gland

and sting are located; this procedure removes the stinger. They swallow lizards and snakes headfirst, avoiding the backwards-pointing spines and scales. They take these reptiles especially in the breeding season when there are young roadrunners to be fed: The pair sometimes cooperates in attacking a snake, circling it and alternating their attacks (Figure 3.2). In the forests of tropical America, groundcuckoos feed at ant swarms, using the ants as beaters to flushout insect prey. Ground-cuckoos are the largest birds in attendance at army-ant swarms and are regular at the swarms along with other forest birds. The cuckoo waits until an ant swarm enters its territory, then follows as the ants pass through the territory and flush out insects with their advance.The cuckoo stands on the ground or perches on a low branch, then runs along the edge of the swarm, snaps up an insect or other forest floor arthropod and runs away, active as it bounds, changes direction or spins: Occasionally it flaps over a vine or gains a higher perch, but extended flight is uncommon. One cuckoo feeds at a time at an ant swarm. When two appear near the swarm the birds either go their separate ways, or they rush at each other with rapid bill-pops, or lower the primaries and spread the tail in aggressive behavior till one bird leaves the site. At other times the ground-cuckoo wanders through the forest, and when it comes to an ant trail it follow the trail until it comes upon the ant swarm. Like a roadrunner, a hand-reared or habituated ground-cuckoo becomes tame enough for an observer to follow on its foraging rounds (Oniki and Willis 1972, Willis and Oniki 1978).

Figure 3.2. Greater Roadrunner pair killing a rattlesnake (after Meinzer 1993).

Behavior 19 A few other cuckoos form a feeding association with ant swarms, or primates and other large mammals, or mixed-species bird flocks. New World ground-cuckoos Neomorphus and Old World ground-cuckoos Carpococcyx follow foraging groups of peccaries and wild pigs, to the extent that these large ground-feeding birds in South America and in Borneo are locally known as “pig birds”. This behavior may have been more common in earlier days when large herds of pigs moved through the forests. Old World coucals sometimes feed in mixedspecies bird flocks. In Madagascar the Blue Coua Coua caerulea follow troops of lemurs as they forage on the forest floor.African yellowbills Ceuthmochares accompany other birds and squirrels, which use each other as beaters.A bird grabs the insects disturbed by the others (Brosset and Erard 1986). Bates (1930: 190) described this behavior: “In its hunt after insects it does not go alone, but follows the squirrels which live in these creepers, or joins itself to other birds, of many kinds, that seek insects, and catches those that flee from the rest; for all insect-catching birds of the trees of the forest find it to their advantage to feed in company.” Yellowbills are seen in 10% of mixed-species flocks in lowland Cameroon (Thomas 1991).Yellow-billed Malkohas Rhamphococcyx calyorhynchus follow Bay Coucals Centropus celebensis through the rattan creepers in Sulawesi, and they also follow troops of primates, Sulawesi macaque Macaca nigra and booted macaque M. ochreata. Redfaced Malkohas Phaenicophaeus pyrrhocephalus sometimes associate with hornbills and other species. Other malkohas that feed in mixed-species flocks include Raffles’s Malkoha Rhinortha chlorophaea. Black-bellied Malkoha P. diardi, Chestnut-bellied Malkoha P. sumatranus and Chestnut-breasted Malkoha P. curvirostris. New World cuculines usually feed alone but some feed in mixed-species flocks of birds. Squirrel Cuckoos Piaya cayana follow army ants in Brazil, and Squirrel Cuckoos and Blackbellied Cuckoos Piaya melanogaster in the Neotropical mainland forests and lizard-cuckoos Coccyzus longirostris in the Caribbean occur in mixed-species flocks, as do wintering Black-billed Cuckoos and Yellow-billed Cuckoos in South America. Greater Anis follow ants and squirrel monkey troops. Other anis live in flocks where they associate with cattle

and take insects when they are stirred up by the cattle, while in less disturbed habitats the anis follow native American mammals. Brood-parasitic cuckoos appear in mixed-species flocks in the non-breeding season. Several insectivorous cuckoos in New Guinea do this, as do wintering Oriental Cuckoos in Australia. Birds in mixed-species flocks benefit from the foraging activities of other birds, and they may also benefit from protection against predators (Thiollay and Jullien 1998). Fruits (especially figs, also berries, tamarinds and palm oil fruits) are the main food of the broodparasitic Dwarf Koel, Common Koel and Channelbilled Cuckoo, and fruit is important in breeding behavior in the Dwarf Koel and Common Koel when the male feeds his mate in courtship. Figs are especially important as they are high in calcium and the fruits are often present in all seasons. Koels have a wide mouth gape (⬎ 2 cm) and can eat as many as 68 intact figs before they regurgitate a pellet that contains the seeds. The young of these broodparasitic cuckoos are often fed insects, although nestling Common Koels reared by Figbirds Sphecotheres viridis get fruit as well as insects, and young Channel-billed Cuckoos Scythrops novaehollandiae are fed fruit by Pied Currawongs Strepera graculina. African Emerald Cuckoo Chrysococcyx cupreus parasitize the fruit-eating Yellow-whiskered Bulbul Andropadus latirostris, as well as insectivorous host species.The bulbuls feed insects to the nestlings during the first four days after their brood has hatched, then feed a mix of insects, regurgitated fruit pulp and whole fruits during the last week of nestling life. Fruit is also fed to young nestling Jacobin Cuckoos when they are reared by fruit-eating bulbuls (Liversidge 1971). Fruit is important to several nesting cuckoos as well. Some malkohas, couas, coucals and Old World ground-cuckoos (Carpococcyx “fruitcuckoos”) take fruit (Shanahan et al. 2001). The ground-cuckoos in Sumatra and Borneo live in dipterocarp forests which produce huge crops of fruit in some years, but in most years do not produce any fruit ( Janzen 1974, Appanah 1985). The fruiteating animals such as bearded pigs Sus barbatus that take these masting fruits have to move long distances to find food or switch to alternative foods such as figs (Whitten et al. 1987, MacKinnon et al. 1996).

20 The Cuckoos Ground-cuckoos often feed with large mammals nearby. Dipterocarp forests are heavily logged and burned and during El Niño years the trees fail to fruit (Curran et al. 1999, Curran and Leighton 2000). Dependence on the fruit of these masting trees may be responsible in part for the decline of groundcuckoos in the Greater Sundas. In Madagascar, two couas are vegetarians that take mainly fruit; couas also take the gum that exudes from acacia trees. In the New World, roadrunners, Squirrel Cuckoos and Yellow-billed Cuckoos occasionally take fruit along with a mainly insectivorous diet. Koels and other fruit-eating birds are ecologically important in the dispersal of forest fruits (Lambert 1989, 1991, Lambert and Marshall 1991, Compton et al. 1996, Kinnaird et al. 1999, Shanahan et al. 2001). Smooth-billed Anis take hot peppers Capsicum, the “Madame Jeanette” chili peppers of the Caribbean and Suriname.The capsaicin chemical in the ripe fleshy fruit is distasteful to mammals but not to birds. The function of capsaicin is to deter mammal seed predators (small rodents are fruit consumers which crush the seeds and do not allow them to germinate) and not to deter birds, which are attracted to the bright fleshy fruits and disperse the seeds in their feces. Consumption of non-pungent peppers by fruit-eating mammals results in zero germination. In contrast, consumption of both non-pungent and hot peppers by fruiteating birds results in germination rates similar to those of control seeds planted from the fruit, which proves that consumption by birds results in successful dispersal of the seeds (Tewksbury and Nabham 2001).Anis disperse gumbo-limbo fruits when they remove the fruit skin, swallow the flesh and regurgitate the seeds.

Cuckoos and their predators Larger predators such as raptors and large primates take adult cuckoos, and cuckoo nestlings and eggs are taken by a variety of predators including birds, mammals and snakes. Both young and adult broodparasitic cuckoos are eaten by mammals such as monkeys in Africa.Young roadrunners are eaten by dogs, cats and raccoons, and these mammals also catch adult roadrunners on the nest. Black-billed

Cuckoos in North America are eaten by hawks and other predatory birds (Hughes 2001), Cuculus cuckoos in the Philippines are plucked and eaten by raptors (Curio et al. 2001), anis in Central America are taken by carnivorous bats (Vehrencamp et al. 1977), and coucals in Central Africa are eaten by chimpanzees Pan troglodytes (Nishida and Uehara 1983), all observations suggesting that cuckoos are not toxic to predatory birds and mammals. In addition, rural peoples eat cuckoos for medicinal purposes as well as for food, including anis in the Caribbean, Common Koels in India and coucals in New Guinea (Walden 1869, Payne 1997b). Most cuckoos avoid predators by stealth. The arboreal New World Black-billed and Yellow-billed Cuckoos avoid predators by their watchful behavior and are easily overlooked by predators, as well as by their visually-oriented prey. Other cuckoos actively avoid predators.When they see a predator, group-living anis call and fly together to a higher perch, so their group behavior is related not only to food but also to mutual protection from predators. Adult Greater Roadrunners avoid an aerial predator by running, dodging, flashing their wings and spreading their tails, behaviors that may distract, confuse or intimidate a predator (Rand 1941b). Roadrunners often run rather than fly when a predator approaches. When I taught a field ornithology course with field trips in the Black Mesa country of western Oklahoma, we watched where the bird ran—it hid in a clump of cholla cactus.The students circled the cactus, one reached in and grabbed the roadrunner. Then we took a few photos, released the bird and watched it run again. Nestlings of many nesting cuckoos (Black-billed and Yellow-billed Cuckoos, roadrunners, coucals) void a foul black liquid from the cloaca when they are disturbed in the nest, and this may repel a predator. Nestling Guira Cuckoos of any age sometimes leap from the nest or branches and escape a predator when they land safely on a lower branch and hide; leapers sometimes die from the fall (R. Macedo, in litt.).The young cuckoos are at risk: more than half of all roadrunner nests are lost to predators (Folse 1974).Young nestling roadrunners produce a fecal sac that is removed and eaten by the parent cuckoos; older nestlings when about to

Behavior 21 leave the nest produce the foul liquid when a predator disturbs them. Brood-parasitic young cuckoos do not produce the black liquid, but like their passerine host young they encapsulate their feces and urine in a gelatinous fecal sac which is removed by the foster parent.Although nesting cuckoos take an active role in mobbing predators, the broodparasitic cuckoos do not join these mobs; they leave the defense of the young to the foster parents.

Breeding seasons In temperate latitudes the gonads of many kinds of birds develop in response to the photoperiod. Seasonal daylength changes are reliable predictors of environmental conditions such as food and temperature, which affect the ability of birds to rear their young (Murton and Westwood 1977). In the northern temperate New World, brood-parasitic Brownheaded Cowbirds Molothrus ater breed at the same time as their host species. In experiments, the cowbirds respond to seasonal changes in daylength in the same way as their hosts, and in the same way as nesting icterids in the same region (Payne 1969b, 1973b). It is likely that brood-parasitic cuckoos respond to the same environmental cues as their hosts in the initial stages of hormonal activity and gonad growth. The final stages of ovarian growth of the brood-parasitic birds depends on hearing the songs and seeing the nesting activity of their hosts (Immelmann and Immelmann 1967, Payne 1973b, 1983, Löhrl 1979). No experimental work on this has been carried out with cuckoos, but it is likely that they respond to changing conditions in much the same way as these other brood-parasitic birds do. North American Black-billed and Yellow-billed Cuckoos breed along with the earliest songbirds in April and May, and if insects are still abundant they continue to breed into August and early September. These early, long and late seasons occur notably in areas with outbreaks of tent caterpillars, which feed in huge numbers on many wild rosaceous trees and shrubs. For Greater Roadrunners, breeding varies with temperature and rainfall. In southern California they breed from late February in low elevations of the Lower Sonoran Life Zone,

and a month later in the upland deserts of the Upper Sonoran Life Zone. A pair can nest again when the male takes over the care of the fledged brood while the female lays a second set of eggs.The cooperative behavior of a pair may let them rear more than one brood when insect food is abundant during the short desert season. In southern Arizona, breeding tends to be bimodal with nesting from mid-April to mid-June and late July to midSeptember with a pause in the hot dry summer, and nesting again after the summer rains, with the breeding peaks varying with the rains of the year. In Texas they breed from March to October, and in Oklahoma from April to July or August. In the tropics, cuckoos breed in the rains; many cuckoos in anglophone regions of the world are known as “rain birds”. They begin to sing shortly before the first good rains begin, and their calls are taken as cues to local farmers to prepare their land for cultivation. Birds may begin their physiological change into breeding condition well before the rains begin (Immelmann 1971). Although their breeding seasons are indirectly timed with the season of rains, some tropical birds respond directly to changing daylength much as in northern temperate regions. A seasonal change of less than an hour, that is, the daylength change at 15° from the equator, is enough to affect the cycles (Hau et al. 1998). Outside the forest where the rains fall through fewer than six months, breeding is seasonal with changes in the vegetation and insects (Levings and Windsor 1982, Wright and Cornejo 1990). In tropical forests many birds are seasonal, nesting in all but the rainiest or driest season (Chapin 1939, Moreau 1950, Skutch 1950, Fogden 1972, Bell 1982a,b,Wikelski et al. 2000). The breeding season for brood-parasitic cuckoos corresponds with the breeding season of their host species. In tropical and subtropical regions, the cuckoos breed early in the rains when their host species begin to nest.These seasons vary within Africa. (1) In equatorial East Africa the breeding season corresponds to the local rains, where seasons differ by a few months in areas only 200 km apart (Brown and Brown 1980). These differences are associated with altitude, distance from the coast, and the presence of mountains that intercept monsoon rains from the Atlantic. East of the eastern rift (region D)

22 The Cuckoos there are two rains in a year, with long rains in April and May and shorter rains around November: January to March is a hot dry season and June to October is cooler and overcast.West of the rift and south of Lake Victoria (region C) there is only one season with rains between October and April or May, and a dry season in the middle of the year, although temperatures are cooler than in the dry season of D owing to the elevation and to the sun not being overhead at that time. In western Kenya, the Lake Victoria basin and Uganda (region B), rains come from March to November with two peaks corresponding with those east of the rift. In the dry region of northern Uganda and northern Kenya (region A) the rains come between March and October, with little rain in June and July. In coastal Kenya the rains come from April to June during the southeast monsoon. Insectivorous songbirds breed during the rains in these areas, and their brood-parasitic cuckoos breed in the same season as their hosts. Klaas’s Cuckoo Chrysococcyx klaas breed mainly in the rains as do their sunbird hosts, with most recorded from February to April in region D, December to April in region C, April in region B and September in region A. Diederik Cuckoo C. caprius breed for a longer season from March to June when most ploceids breed in region D. Finally, Red-chested Cuckoos Cuculus solitarius breed mainly from April to June in region D, and February to September in region B. (2) In West Africa, seasonal rains from the Atlantic are controlled by the inter-tropical convergence (ITCZ), which is the atmosphere’s equator.The zone moves a month or two behind the overhead movement of the sun.The vernal equinox in March is followed by rains in West Africa in May and these rains continue until October. The zone forms by movement into, or convergence of, the northeast trade winds of the northern hemisphere and the southeast tradewinds of the southern hemisphere in the equatorial belt.The inflowing air rises and clouds form in convective cells that often drift westward bringing frequent rains and thunderstorms. Dry northeasterly winds, the harmattan, blow off the Sahara desert from November to March, signaling the end of the rains in the north.The wet westerly or southwestern monsoon is more extensive and marked in summer

from May to September when thermal heating over the continent supports the northerly displacement of the equatorial trough. Rains on the coast of Ghana at Accra and Cape Coast peak in May and June, as the monsoon is intercepted along the coast west of Cape Coast. Further east in Togo the rains are affected by a low north-south mountain range (Grimes 1987, Cheke and Walsh 1996). Only 200 km inland from the coast in Ghana at Kumasi are there two rainy peaks, one in May and June and the other in September and October (Elgood et al. 1994). Further north the winds bring less rain. In coastal Togo, Lomé gets nearly 100 mm during April, while in northern Togo rains this intense arrive a month later and the growing season begins later (Cheke and Walsh 1996). Likewise northern Senegal gets less rain and the season of effective rain is from June to October (Morel and Morel 1990). In Nigeria a monthly rainfall of at least 100 mm comes a month earlier on the coast (March) than at Ibadan (April), and a month later in northern Nigeria (May) (Elgood 1994). Different breeding seasons are known for African Black Coucal Centropus grillii—in coastal Ghana from April to July, in northern Ghana in July and August.The birds also apparently migrate between north and south (Grimes 1987), much as they do in Togo. (3) Even at high latitudes the breeding seasons are strongly influenced by seasonal rains. Klaas’s Cuckoos in the Cape Region of South Africa, an area of winter rainfall at 34–35° N, breed in winter from July to September along with their sunbird hosts, and in the Eastern Cape at 33–34° N they breed mainly from October to December. Further north (c. 15° S) where rains come later the cuckoos breed in the later months, as in Zambia where they breed mainly from December to February. Great Spotted Cuckoos breed two or three months later in western Namibia, than in Natal and the Eastern Cape of South Africa, again in relation to the rains of the region. In India the breeding seasons of brood-parasitic cuckoos are similar to those of nesting cuckoos. Their breeding seasons are related to altitude in India (Baker 1934, 1942), where cuckoos and their hosts breed earlier in the plains than in the higher foothills and mountains.The rains and green growth that follow come earlier at the lower altitudes.

Behavior 23 In Africa the breeding season for the broodparasitic cuckoos varies with the ecology of the nesting cuckoos as well as with the ecology of the host species. In Malawi, nesting resident Whitebrowed Coucals breed right through the year except from July to September, while migratory African Black Coucals breed from January to May when the marshes are flooded (Benson and Benson 1977). Many insect-eating songbirds begin to breed before the first rains, and in northern Zambia like the miombo trees in Brachystegia woodland which grow and unfold their new leaves before the rains, these songbirds anticipate the rains weeks before they begin. In this pre-rain period, Black Cuckoos Cuculus clamosus and Klaas’s Cuckoos begin to sing. The nesting cuckoos tend to have a longer breeding season than the brood parasites, depending on their status as migrants or as residents (Table 3.1). In the parasite species with the largest number of records, the season is five months. In Zambia, the cuckoos with the long breeding season are resident coucals. In Zimbabwe, the brood-parasitic cuckoos with the longest season are Levaillant’s Cuckoos Clamator levaillantii, whose main host,

Table 3.1 Breeding seasons (number of months) in south-central Africa.

nesting cuckoos Centropus grillii Centropus cupreicaudus Centropus senegalensis Centropus superciliosus Ceuthmochares australis brood parasites Clamator jacobinus Clamator levaillantii Pachycoccyx audeberti Cuculus clamosus Cuculus solitarius Cuculus gularis Chrysococcyx caprius Chrysococcyx klaas Chrysococcyx cupreus

Malawi

Zambia

Zimbabwe

5 na 1 9 1

3 6 7 3 -

4 4 6 6 1

7 2 1 3 1 7 5 3

5 3 4 3 5 5 -

6 8 5 4 4 4 6 8 -

Arrow-marked Babbler Turdoides jardinei, is a cooperatively-breeding bird that nests in all months; and the partly resident Klaas’s Cuckoos whose hosts are warblers and sunbirds, some breeding in the rains and others in the dry season. The resident coucals breed over a longer period than the migratory African Black Coucals (Irwin 1981, Hustler 1997a). In Australia, the time when cuckoos breed depends on the rains. In two regions matched for latitude but with different months of rain, Horsfield’s Bronze-cuckoo Chrysococcyx basalis begins to lay in the early austral spring in July in the southeast, and in early October in the southwest. A common host in the west, Splendid Fairy-wren Malurus splendens have a longer breeding season than their cuckoos, and the latest nests escape the cuckoo parasite (Rowley et al. 1991, Rowley and Russell 1997). Further north where rains fall in austral summer, cuckoos and their hosts breed in the Kimberley Division from January to March, and from March to May and August to September in the Pilbara Region: In arid inland Australia breeding is bimodal, with rains in spring and autumn (Storr 1984a, Brooker and Brooker 1989a,b, Ambrose and Murphy 1994, Pizzey and Knight 1998, Higgins 1999).

Migration Cuckoos undertake long-distance migrations between their temperate breeding areas and their wintering grounds.With cuckoos that breed in the higher latitudes, migration is known from their alternating seasonal occurrence in the breeding region and on the winter grounds. North American Black-billed and Yellow-billed Cuckoos migrate to South America, and winter south of the equator. Another North American cuckoo is at least partly migratory: northern populations of Mangrove Cuckoos appear seasonally in Florida and the Bahamas, and are absent in winter. Migration is just as remarkable although the detail is less well known with the South American cuckoos. Austral migration occurs with other birds that breed in temperate South America while the cuckoos are nonbreeding wintering birds in the tropics (Chesser 1994, 1997). At least three cuckoos have

24 The Cuckoos this seasonal movement and tend to live in scrub woodland both in their breeding grounds and in the austral winter: Ash-colored Cuckoo and southern populations of Pearly-breasted Cuckoos Coccyzus euleri and Dark-billed Cuckoos C. melacoryphus. Ash-colored Cuckoos Coccycua cinerea breed only in the southern temperate region, but the other two cuckoos have a wide breeding range. Other cuckoos perhaps migrate from the southern extremity of their range (Greater Ani, American Striped Cuckoo: Chesser 1994), yet the odd birds that appear in the tropics in localities where they are not regular in occurrence may be local dispersers rather than austral migrants. In the absence of known morphological differences between northern and southern breeders it has not been possible to determine whether the southern cuckoos move into the northern region while northern birds are breeding, or whether both populations spend a non-breeding period in the Amazon basin. Four cuckoos in the northern temperate region of the Old World complete long-distance migrations when they move across the equator (Great Spotted Cuckoo, Common, Oriental and Asian Lesser Cuckoos). European Great Spotted Cuckoos and Common Cuckoos migrate to Africa, flying nonstop over the Mediterranean and the Sahara, a distance of more than 3000 km. Other cuckoos migrate from continental Asia to the Malay Peninsula, the Greater Sunda Islands and the Philippines (hawk-cuckoos Hierococcyx and drongocuckoos Surniculus), or over the Indian Ocean to Africa (Indian populations of Jacobin, Common and Asian Lesser Cuckoo) and appear in migration on islands in the Indian Ocean between their breeding and wintering grounds. Some cuckoos that appear on the coast in Kenya have flown over 3000 km from their nearest breeding grounds to winter in Africa. Madagascar Lesser Cuckoos Cuculus rochii migrate from their breeding grounds to live the rest of the year in continental Africa. Migration of the Common Cuckoo is known both by seasonal changes in where the birds live, and by recoveries (n ⫽ 2) of birds marked in Europe and found thousands of kilometers away in Africa in the winter. Less well known are the wintering grounds of the Common Cuckoo that breeds in eastern Asia.

Migration of brood-parasitic cuckoos that move between breeding and nonbreeding areas is timed to coincide with the season when the hosts breed, and when caterpillars emerge and become abundant with the onset of warmer weather in spring in the high latitudes, and with rains in the subtropics and tropics. Cuckoos regularly migrate within Africa, moving from their breeding grounds in South Africa to equatorial Africa (black phase Jacobin Cuckoo, Black Cuckoo). Cuckoos move with the rains in West Africa where they normally live near the humid coast in the dry season, and move northward into the savanna in the rains ( Jacobin Cuckoo, Levaillant’s Cuckoo, Black Cuckoo, African Emerald Cuckoo, Klaas’s Cuckoo, Diederik Cuckoo, African Black Coucal), while some may remain as residents of coastal areas throughout the year. In the Australo-Pacific region, certain cuckoos migrate from their southern breeding grounds in the temperate regions of Australia and New Zealand, to winter in the tropics (Horsfield’s Bronze-cuckoo, Shining Bronze-cuckoo, Channelbilled Cuckoo, Long-tailed Cuckoo). Migrating cuckoos are often very fat, and their stores of fat may persist into winter (“summer” in their nonbreeding grounds), as in a Common Cuckoo from the Palearctic measured in December in South Africa (Payne 1968, Moreau 1972). Shining Bronzecuckoo and Long-tailed Cuckoo breed in New Zealand in summer but are absent there in the winter months; and they occur on tropical Pacific islands in winter but not in summer. Few banded cuckoos in the region have been recovered to show longdistance migration, In one case, Common Koel banded in New South Wales, Australia, was recovered in New Guinea, 2950 km from its banding site (Higgins 1999). The only form of Little Bronzecuckoo known to migrate from Australia to New Guinea is Chrysococcyx m. barnardi, with a few winter records in New Guinea. Other Australian bronzecuckoos move within Australia but it is unknown whether the green-glossed C. m. minutillus migrate to New Guinea. Rufous birds that breed in Australia look like rufous breeding birds in New Guinea (C. m. poecilurus), so there is no compelling evidence of long-distance migration in these cuckoos.

Behavior 25 Migration of cuckoos that breed in tropical areas is less well known, and cuckoos seen only in one season, may be resident but inconspicuous at other times when they are silent. In Africa, Jacobin Cuckoos and Levaillant’s Cuckoos are seen seasonally in Zambia within 15° of the equator and they migrate to more equatorial regions. Klaas’s Cuckoo is conspicuous in the breeding season in southern Africa and nearly disappears in the non-breeding season. Nevertheless a few birds remain all year, but are inconspicuous during the season when they do not sing. Movements of cuckoos within Africa are not well understood, partly because several species have populations that do not differ morphologically between regions and lack distinctive subspecies, and in these birds there is no easy way to distinguish between local residents and nonbreeding visitors. The black-phase Black Cuckoos in West Africa probably come from populations further south (that is, south of the Congolese equatorial forests), if only because these black-phase birds are not known to sing or to breed in West Africa. On the other hand, there are remarkably few records of the resident forms of this species actually breeding in West Africa.These migratory forms are mainly non-forest birds. In Togo, five species of non-passerines in forest habitats show seasonal changes in their numbers, that suggest a seasonal arrival of birds from further south: three of these are cuckoos, Black Cuckoo, Red-chested Cuckoo and African Emerald Cuckoo. Among savanna cuckoos several species appear to have resident populations in the south, but these birds are present only during the rains in the northern savannas and they may undertake more local migrations. These include Diederik Cuckoo, Klaas’s Cuckoo and African Black Coucal, and the coastal swamp grasslands may also receive visiting African Black Coucals from further south (Cheke and Walsh 1996). In East Africa, Barred Long-tailed Cuckoos Cercococcyx montanus appear seasonally near the east coast where they do not breed, and these birds make local migrations. In southeast Asia, the migration of cuckoos at night has been recorded in the Malay Peninsula. On Fraser’s Hill, the records of cuckoos netted at night include more than 10 individuals each of Chestnut-

winged Cuckoo Clamator coromandus,Violet Cuckoo Chrysococcyx xanthorhynchus, (Fork-tailed) Drongocuckoo Surniculus (dicruroides) (the most abundant, with 57 records),Whistling Hawk-cuckoo Hierococcyx nisicolor and Indian Cuckoo Cuculus micropterus (Medway and Wells 1976). The seasonal appearance of small glossy-cuckoos Chrysococcyx in the Indonesian archipelago may result from the inconspicuous behavior of non-breeding birds, and there are no substantiated records of migration of the distinctive forms of Little Bronze-cuckoo C. minutillus. Brush Cuckoos Cacomantis variolosus may migrate from the central highlands to the lowlands in New Guinea: the evidence is the occasional appearance near the coast of long-billed birds, that are more often found in the mountains (Diamond 1972). Long-distance migrants sometimes appear on the wrong side of the ocean. North American Black-billed and Yellow-billed Cuckoos appear as vagrants on the North Atlantic islands and in northern Europe, and as far east as Italy and Sicily, while Palearctic Common Cuckoos turn up in northern North America in Alaska and on the Atlantic coast, and Oriental Cuckoos appear in Alaska.The dangers of migration over the ocean are illustrated by the remains of a Yellow-billed Cuckoo in the stomach of a shark (Saunders 1962). Black-billed and Yellow-billed Cuckoos migrate at night.They make an over-water flight in autumn of 2000–3000 km from their breeding grounds in North America to the West Indies, then continue onwards to South America. Or they fly as far as 4000 km from the north directly to the mainland of South America, then continue well inland to winter south of the equator. Yellow-billed Cuckoos appear through May to August on islands off the Florida coast where they do not occur as breeding birds, and their appearance suggests that spring migration continues as late as June, and the southbound return begins by July (Stevenson and Anderson 1994).These cuckoos orient by patterns of the stars: when tested in a planetarium (Marler and Hamilton 1966) they became active at night and directed their movements in the seasonally correct direction of the planetarium sky. These cuckoos appear as casualties below TV transmitting towers which attract nocturnal migrants by their lights,

26 The Cuckoos and they fly into brightly lit walls as well (Sick 1993). Cuckoos that breed in New Zealand, the Shining Bronze-cuckoo and Long-tailed Cuckoo, are nocturnal migrants that winter on remote islands in the Pacific Ocean more than a thousand kilometers from their breeding range. The large Channel-billed Cuckoos migrate during the daytime as they pass between Australia and the tropics where they winter in New Guinea and the Moluccas. Cuckoos like other nocturnal migrants can orient even without the experience of learning from their parents. When they are socially isolated and hand-reared apart from the adults, young Yellowbilled Cuckoos become active at night in the migration season and orient their restless behavior toward the stars (Marler and Hamilton 1966). A “natural experiment” demonstrated that the young brood-parasitic cuckoos have no opportunity to learn the migration time or route from the adults, as the older birds do not live with their offspring, and would have left from their northern breeding

grounds several weeks before the juveniles leave the area. Some resident cuckoos disperse after the breeding season and range far from their natal and breeding areas. Individual Greater Anis appear outside their breeding range in Brazil away from their breeding sites. Smooth-billed Anis and Groovebilled Anis occasionally show up at the end of summer in North America more than 1000 km north of their breeding grounds, usually alone and not in a group. A South American Guira Cuckoo once appeared in the Netherlands Antilles. “All three [ani] species have shown a remarkable tendency towards extralimital vagrancy, given their seemingly ineffective flight and short rounded wings. The wide distribution of Smooth-billed [Ani] in the West Indies with no phenotypic differentiation attests to its powers of dispersal” ( J.V. Remsen, in McLean 1995). Most tropical cuckoos are resident, including the ground-cuckoos of the Old World and New World, the Madagascar couas, the malkohas and many coucals.

4 Morphology

Cuckoos have a zygodactyl foot with two toes of each foot (numbers 1—the inner toe or hallux of most birds—and 4, the outermost toe) directed backward, and the other two toes (numbers 2 and 3) directed forward. Most cuckoos have a long tail. Cuckoos have 10 primaries and 10 rectrices (eight in the anis); the plumage has no aftershaft; the oil gland is naked, and in most species the nestlings have no downy feathers (Nitzsch 1867). Several other morphological features occur in cuckoos that are generally not present in other birds, and a few of these features occur only in the cuckoos.The most distinctive charactereristics are in the skeleton in details of the tarsometatarsal bone (two enclosed canals side by side in the hypotarsus, and a characteristic shape of the accessory process or sehnenhalter on trochlea IV of the distal tarsometatarsus), and in the shape of the humerus and its deltoid crest. Relationships within the cuckoos have been estimated from their anatomy, geographic distribution and breeding biology, and more recently from their molecular genetics. Earlier studies were based on morphology, primarily from museum skins. Shelley (1891) recognized one family of cuckoos with six subfamilies: typical cuckoos Cuculinae, coucals Centropodinae, malkohas Phaenicophaeinae, New World ground cuckoos Neomorphinae, New World brood parasites (the group: the birds were not yet known to be brood parasites) Diplopterinae, and anis Crotophaginae. Peters (1940) recognized the cuckoo family, also with six subfamilies (Cuculinae, Phaenicophaeinae, Crotophaginae, Neomorphinae, Couinae and Centropodinae); he included Carpococcyx with Neomorphus in Neomorphinae. Wolters (1975–1982) recognized nine families of cuckoos

(Crotophagidae, Centropodidae, Neomorphidae (including Dromococcyx), Taperidae, Coccyzidae, Clamatoridae (including Pachycoccyx), Cuculidae, Eudynamidae (including Caliechthrus, Urodynamis, Microdynamis, Rhamphomantis and Scythrops), and Phoenicophaeidae. He included Coua and Carpococcyx with Centropus in a family Centropodidae; Mason (1997a) listed characters that these groundliving birds had in common.

Comparative and adaptive features of cuckoo morphology A few general trends in morphology are related to the life style of cuckoos. In some cases we can distinguish between adaptations and features that are explained by the evolutionary relationships among lineages. Body weight was taken as the main standard of body size to allow broad comparisons across cuckoos: body weight is correlated with other aspects of size, and many life history variables are associated with body weight in birds in general, as in a sampler of behavior, breeding biology, lifespan and metabolic rates (Calder 1984). Body size was then compared with brain size in cuckoos with different breeding behavior (brood parasites, nesting pairs and cooperatively-breeding group-living birds), and also with length of the leg bones in relation to the arboreal and terrestrial feeding behavior of the cuckoos.Where body weight is not known, wing length or even tarsus length can be used as an estimate of body size, at least within a group of birds based on the same body plan such as the coucals, but where the proportions of birds differ

28 The Cuckoos among groups, body weight is the standard of comparison for body size. Species differences in brain size may be related to both phylogeny and adaptive behavioral ecology. There have been few comparative studies of birds (Pearson 1972, Mlíkovsky 1989), and those have shown few ecological correlates of brain size, except that altricial birds tend to have larger brains than precocial birds (Bennett and Harvey 1985, Iwaniuk and Nelson 2001, 2002). In cuckoos, the brain size is similar to that of other altricial birds of the same body size, although the brain is smaller than in parrots.The relationship of brain size with body size in birds varies with the systematic level of analysis, a so-called “taxon-level effect” ( Jerison 1973, Aboitiz 1996, Nealen and Ricklefs 2001). To test the variation in brain size with behavior, skeletal specimens of 76 cuckoo species were examined by George Kulesza. The orbital foramina were closed with masking tape, then the braincase was

filled through the foramen magnum. Spherical lead shots of uniform size were introduced into the braincase with the help of a flexible cylinder.The volume of shot in the braincase was compressed by shaking and tapping the skull until the shot was level with the foramen magnum. The procedure was repeated for each skull to determine repeatability (within 2%) and the two measurements were averaged. The amount of shot was determined with the use of a graduated pipette. Volumetric measurements showed that 0.17 ml of shot was equal to 1 ml or about 1 g of brain. The mean mass of a pellet (0.013 g) agreed closely with the mass µ calculated for a sphere of the measured diameter 0.136 cm (µ ⫽ ρ (4/3)πr3), where radius r ⫽ 0.068 cm, and density of lead ρ ⫽ 10.5 g (cm3)⫺1.Where several skeletal specimens of a species were available, 2–4 skulls were measured and the mean brain size was calculated. Brain size (g) was then compared with body size (g) in the corresponding sex and subspecies.

Figure 4.1. Brain size in relation to body size and behavioral ecology in cuckoos. Closed symbols indicate nesting species (N), open symbols indicate brood parasites (P).

Table 4.1 Leg length, femur, tibiotarsus and tarsometatarsus (mm), brain (g), body weight (g) and behavior (a ⫽ arboreal, t ⫽ terrestrial). Species

breeding (c, n, p)2

specimen3

41.0 43.6 38.3 36.3

t t t t

c c c c

UMMZ UMMZ UMMZ UMMZ

49.8 86.5 70.8 95.1 40.2 46.6 40.5

38.0 63.7 51.2 74.6 30.5 36.2 32.3

t t t t t t t

n n n n p p p

UMMZ 159111; 133738, 159111 UMMZ 227057; 227059, 234551 UMMZ 159112; 156461, 159112 UMMZ 200592 UMMZ 222217; 218370, 218946 MVZ 85638 LSU 101257

37.9 48.4 60.6 52.3 63.5 73.8 54.4 48.8 60.5 41.7 45.1 39.9

55.1 71.7 87.9 79.9 95.3 112.3 74.9 67.5 90.6 59.3 63.5 62.3

41.4 45.6 57.5 50.1 63.4 71.6 51.1 48.4 63.5 41.2 43.1 43.8

t t,a t t t t t t t t t t

n n n n n n n n n n n n

UMMZ 143041 USNM 226188 USNM 557171 USNM 318594 AMNH 7333 UWBM 63149 UMMZ 224761 UMMZ 207709 BMNH S/1969.1.462, USNM 562052 UMMZ 208305 UMMZ 208402 UMMZ 228027; 228025, ⫺26, ⫺27 (s)

64.4 66.3 42.4

112.3 112.4 74.4

87.4 91.7 57.9

t t t

n n n

UMMZ 219851, 236029; ⫺26 and 219043 (s) USNM 223970 USNM 432195; FMNH 356640 (s)

brain (g)

femur

tibiotarsotarsus metatarsus

142 153 105 75

1.85 2.69 1.71 1.53

37.2 41.9 38.3 34.8

58.8 64.6 56.8 53.6

62 305 178 345 52 85 48

1.33 3.59 2.77 4.39 1.25 1.11 0.86

31.4 54.5 50.0 60.8 27.9 30.3 24.3

88 – – 336 503 769 280 170 417 170 177 123

2.25 4.38 5.33 – 7.45 9.69 3.90 3.34 4.66 3.11 3.36 2.75

– – 190

5.96 – 2.63

202015; 157046, 200661, 202013, ⫺14 (s) 139988; 139989, 219553 (s) 218940; 118151, 218943 (s) 219189; 133737, 219189, 219557 (s) (s) (s) (s) (s)

Morphology 29

Crotophaginae Guira guira Crotophaga major Crotophaga ani Crotophaga sulcirostris Neomorphinae Morococcyx erythropygus Geococcyx californianus Geococcyx velox Neomorphus geoffroyi Tapera naevia Dromococcyx phasianellus Dromococcyx pavoninus Centropodinae Centropus bengalensis Centropus celebensis Centropus goliath Centropus leucogaster Centropus menbeki Centropus milo Centropus phasianinus Centropus senegalensis Centropus sinensis bubutus Centropus superciliosus Centropus toulou Centropus viridis Couinae Carpococcyx renauldi Carpococcyx radiatus Coua ruficeps

behavior (a,t)1

body (g)

Species Coua caerulea Coua cristata Coua reynaudii Coua gigas Coua serriana Cuculinae Rhinortha chlorophaea Ceuthmochares australis Clamator coromandus Clamator glandarius Clamator jacobinus Clamator levaillantii Dasylophus superciliosus Dasylophus cumingi Zanclostomas javanicus Phaenicophaeus diardi Phaenicophaeus tristis Phaenicophaeus curvirostris Rhamphococcyx calyorhynchus Coccycua minuta Piaya cayana Piaya melanogaster Coccyzus americanus Coccyzus minor Coccyzus melacoryphus Coccyzus erythropthalmus Coccyzus lansbergi Coccyzus merlini Coccyzus longirostris Coccyzus pluvialis Coccyzus rufigularis

body (g)

brain (g)

femur

235 136 140 420 298

3.48 2.38 3.47 4.75 4.15

50.0 34.7 43.5 57.8 52.5

tibiotarsotarsus metatarsus 83.8 55.4 55.5 39.3 66.4 46.3 98.1 72.0 80.1 61.0

59 66 77 124 72 124 118 174 98 61 115 189 – 40 98 102 63 64 50 51 55 154 98 176 130

1.34 1.53 1.49 1.85 1.31 1.89 2.62 3.27 2.46 – 2.39 2.98 – 1.06 2.07 1.89 1.20 1.53 1.15 0.99 1.12 2.50 1.96 2.79 2.62

28.1 31.0 32.1 35.6 31.7 36.9 40.1 46.1 37.6 34.4 40.6 45.8 49.6 25.6 42.0 38.2 29.2 29.9 27.9 26.2 28.6 42.3 37.4 47.6 48.2

41.3 44.2 45.4 49.6 44.5 51.2 57.3 67.2 51.2 47.2 59.5 65.1 67.1 36.1 61.4 53.7 39.7 43.5 39.0 37.1 41.5 61.6 51.9 68.8 –

27.1 29.9 27.8 32.5 29.0 32.5 38.7 44.0 35.6 33.5 40.5 43.2 45.7 25.4 42.4 37.4 27.4 29.7 26.3 24.6 27.0 41.9 35.1 45.1 –

specimen3

behavior (a,t)1 a a t t t

breeding (c, n, p)2 n n n n n

UMMZ 209201 UMMZ 157526; 157526, FMNH 384680 (s) UMMZ 219043; 208403, FMNH 352797 (s) FMNH 360040 UMMZ 209202; FMNH 356638, ⫺39 (s)

a a a a a a a a a a a a a a a a a a a a a a a a a

n n p p p p p n n n n n n n n n n n n n n n n n n

MCZ 342734 UMMZ 209209; 158185–86 (s) USNM 343999 UMMZ 212911; ⫺11, ⫺14 (s) UMMZ 212899; 212890, ⫺94, ⫺96 (s) UMMZ 212906 UMMZ 228068; ⫺22, ⫺23, ⫺24 (s) UMMZ 233062; ⫺63 (s) UMMZ 236439 BMNH S/1969.1.10 USNM 344002 UMMZ 236028 USNM 226196 UMMZ 139991; ⫺90, ⫺92 (s) UMMZ 222216; 133732, ⫺33 (s) UMMZ 209478; LSU 64990, 89701 (s) UMMZ 203320; 154465, 201937, 207641 (s) UMMZ 208492; 152623, 208492 (s) UMMZ 208493 UMMZ 205253; 200574, 201813 (s) LSU 114314 UMMZ 158527 USNM 554614 USNM 559182 USNM 226541

30 The Cuckoos

Table 4.1 contd.

1

40.9 36.7 58.7 19.6 16.5 16.8 17.2 20.6 18.3 23.0 25.8 20.6 19.2 19.0 22.1 19.1 33.7 29.1

57.1 51.9 80.2 29.7 25.5 24.4 25.7 28.5 25.3 31.1 36.8 28.8 28.2 27.0 30.9 27.1 46.2 39.3

34.2 34.5 44.5 20.7 17.8 16.3 18.0 18.1 15.9 18.8 22.7 18.5 18.6 18.2 20.3 16.1 28.3 23.3

25.9 26.8

36.8 35.2 34.2 43.2 32.1 31.5 36.6 34.9 37.1

21.2 19.9

32.0 22.5 22.5 37.0 36.2 28.4

27.1 19.0 17.3 21.4 19.4 20.7

a a a a a a a a a a a a a a a a a a a a a a a a a a a a a

p p p p p p p p p p p p p p p p p p p p p p p p p p p p p

UMMZ 210347 USNM 559589 KU 85378; KU 85378, MCZ 340296 (s) UMMZ 214229; 214229, ⫺30 (s) KU 85213; UMMZ 214232 (s) KU 43514; KU 41018 (s) UMMZ 214231; 214231, 231601 (s) UMMZ 212919; 217503 (s) UMMZ 158083; 158083, 209198 (s) UMMZ 218545; 209197, 218545 (s) UMMZ 206513; 214226 (s) UMMZ 223664 UMMZ 210982; 207451 (s) UMMZ 228020; 233059, ⫺60 (s) UMMZ 214227; 214227, ⫺28 (s) UMMZ 233061 UMMZ 233058 USNM 557504

FMNH 357420 UMMZ 212917; 212918, 217502 (s) UMMZ 212916 USNM 318240 UMMZ 234936 FMNH 384675 UMMZ 207052; 119430, 151115, 207051 (s) USNM 560668 UMMZ 207450

Behavior, a⫽arboreal; t⫽terrestrial. Breeding behavior, c⫽cooperative; n⫽nesting; p⫽brood-parasitic. 3 First specimen indicates leg measurements, second specimen(s) indicates brain measurements where a specimen other than the first was used, repeated number indicates that brains of all specimens were used. 2

Morphology 31

Eudynamys scolopacea 222 2.59 Urodynamis taitensis 120 2.08 Scythrops novaehollandiae 684 5.55 Chrysococcyx osculans 30 0.77 Chrysococcyx basalis 24 0.56 Chrysococcyx minutillus 18 0.58 Chrysococcyx lucidus 23 0.58 Chrysococcyx caprius 32 0.88 Chrysococcyx klaas 26 0.61 Chrysococcyx cupreus 38 0.94 Cacomantis pallidus 82 1.52 Cacomantis sonneratii 37 0.94 Cacomantis merulinus 24 0.99 Cacomantis variolosus 34 0.93 Cacomantis flabelliformis 44 1.05 Surniculus velutinus 36 0.99 Hierococcyx sparverioides 150 2.19 Hierococcyx hyperythrus 115 1.82 Hierococcyx nisicolor 75 Hierococcyx fugax 80 Hierococcyx pectoralis 76 1.62 Cuculus clamosus 85 1.54 Cuculus solitarius 74 1.59 Cuculus micropterus 119 – Cuculus poliocephalus 52 – Cuculus rochii 59 1.01 Cuculus canorus 106 1.67 Cuculus optatus 117 1.52 Cuculus saturatus 91 – Body weights are from the species accounts.

32 The Cuckoos Brood-parasitic cuckoos have smaller brains for their body size than nesting cuckoos (Table 4.1, Figure 4.1). The variation in brain size is in part explained by a taxon level effect with less scatter around a genus than between genera. Nevertheless, the trend for brood parasites to have smaller brains appears in all three clades of brood-parasitic cuckoos. Cuculus and related brood-parasitic cuckoos have smaller brains than the malkohas, Clamator have smaller brains than the nesting malkohas of the same body size, and the New World brood-parasitic Tapera and Dromococcyx have smaller brains than the nesting Lesser Ground-cuckoo Morococcyx erythropus and Lesser Roadrunner Geococcyx velox. It is unknown whether certain parts in the brain of brood-parasitic cuckoos are small and associated with breeding behavior. In the other brood-parasitic birds that have been examined, cowbirds have larger hippocampus volumes than do the non-parasitic icterids, and females cowbirds that search for host nests have larger hippocampus volumes than males that do not; total brain volume was not determined (Sherry et al. 1993, Reboreda et al. 1996). In birds, the hippocampus is associated with spatial experience and memory (Clayton et al. 1997, Clayton 1998). No other comparative studies on brains of parasitic and non-parasitic birds have been reported. The cooperatively-breeding crotophagine cuckoos do not have large brains for their body size; neither do cooperatively-breeding songbirds have larger brains than other songbirds of the same body size (Iwaniuk and Arnold 2004). In contrast, primate species with the cognitive demands of sociality in complex societies have large brains (Harvey and Krebs 1990, Dunbar 1995, Barton 1996, Kudo and Dunbar 2001). Arboreal cuckoos have smaller brains than terrestrial cuckoos Coua, Carpococcyx and Centropus, in contrast to mammals, where arboreal squirrels have larger brains than terrestrial rodents, as might be expected with the arboreal squirrels’ use of a threedimensional habitat (Eisenberg 1981). Cuckoo brains do not have unusually large olfactory lobes when compared with those of other birds, although the brains of only four cuckoo species have been examined (Bang and Wenzel 1985). The presence of large anal glands in anis (Quay 1967), and

their possible function ( perhaps as in the Hoatzin Opisthocomus hoazin or “Stink-bird” (Newton 1896) which have large olfactory lobes) suggests a direction for further study of olfactory behavior and brain structure in cuckoos. Second, the appearance of the head and bill differs among cuckoos. In the anis the head shape is determined by the bill, which is compressed laterally and expanded into a lateral shield. Bill size and shape may be important in social displays (males have a more prominent profile than females, and adults have a more prominent arc than juveniles), and bills also vary with feeding behavior, with larger bills in species that forage in wet, dense foliage. Bill size and shape vary in the malkohas where bright colors suggest their importance as social signals, and in the lizard-cuckoos where the bill is long and slender. Bills of fruit-eating cuckoos are particularly gross. The nostril shape differs among cuckoos and in early systematic treatments the nostril shape was used to characterize the genera of malkohas (Sharpe 1873).The nostril is circular in many brood-parasitic cuckoos and in the New World cuculines, and slitlike in other brood-parasitic cuckoos. The nostril shape even varies between geographic vicariants within a single species in the Chestnut-breasted Malkoha Phaenicophaeus curvirostris, where the colors of the face and brightly-colored bill with light green, red and black are in different patterns across the range of the species, and the nostril shape varies from a horizontal oval to a vertical slit, with or without a groove extending from the nostril to the cutting edge of the bill. In these birds the nostrils are as variable as the color patterns on the bill, and they may be behavioral signals within a mated pair. In other cuckoos, nostril shape is consistent across species that are closely related in the phylogenetic analyses. Colors of the bill and bare facial skin range from indistinct blackish to bright red, yellow, blue and green. The yellow banana-like bills of some Asian malkohas suggest toucans to a New World ornithologist, and the bicolor huge banana-like bill of Australian Channel-billed Cuckoo led to their original descriptions as “Anomalous Hornbill” and “Psittaceous Hornbill”.A bright blue face is striking both in real life and in photographs of Raffles’s

Morphology 33 Malkoha Rhinortha chlorophaea, Blue-faced Malkoha Phaenicophaeus viridirostris, and several couas Coua, as well as in roadrunners Geococcyx, large forest-living New World ground-cuckoos Neomorphus and Lesser Ground-cuckoo. In some cuckoos the blue contrasts with the neighboring bare skin of orange, red or purple, and in others especially couas the blue is outlined with black. In earlier work on birds, blue skin has often been attributed to the incoherent scattering of light by small particles, the bright violets, blues and greens on the bare faces of these cuckoos may be due to constructive reflection from ordered arrays of collagen arrays, as in other birds with bright blue bare skin. The red facial skin of Red-faced Malkoha is covered with rough papillae, the color disappears when the birds are dead; the source of the color is unknown. Brightly-colored bare skin occurs in many cuckoos, in both the Old World and the New World, especially in the tropical

forests. Bare skin that ranges from violet, blue and green to yellow which reflects certain colors due to constructive interference has evolved at least twice in the cuckoos, once in the Old World and once in the New World. These colors may be associated with colors of the ambient light in forests (Endler 1993, Prum and Torres 2003). Third, the length and proportion of the leg bones vary with arboreal or terrestrial behavior. The legs are longer in proportion to body size in terrestrial cuckoos than in arboreal cuckoos. This morphological adaptation across clades may not have been noted before in birds, perhaps because the mechanics of movement differ between birds and mammals, a group where functional comparisons have been made in more detail (Hildebrand 1974). Engels (1938) and Berger (1952) described the lengths of leg bones in four species of North American cuckoos, but their motive was description

Figure 4.2. Body weight vs total leg length in arboreal and terrestrial cuckoos.

34 The Cuckoos and classification, and not a comparative test of behavior over different clades. In the present study, for cuckoo species that were available as skeletons, the femur, tibiotarsus, and tarsometatarsus were measured and the percentage of the total length of these leg elements was compared with body size (weight) and with terrestrial or arboreal behavior. Across a wide range of body size (23 g to 769 g) the ratios vary with the femur 25–33% of total length, the tibiotarsus 40–44%, and the tarsometatarsus 22–31%.The distal bones do not increase in length disproportionately to the femur with increasing body size, a trend that occurs when birds are compared over a wider range of size (Calder 1984). Terrestrial cuckoos have longer legs and a longer tarsometatarsus than climbing arboreal cuckoos of the same body size (Table 4.1, Figures 4.2, 4.3). Arboreal cuckoos may have short legs to balance and keep a low center of gravity as they move along branches in the canopy. Terrestrial cuckoos with

long legs and a long tarsometatarsus move rapidly without the mechanical problem of high distal mass: angular momentum and the energy to move the legs is an exponential function of the mass of the distal elements, and speed is a function of stride with long legs. Greater Roadrunners on the desert floor for example run very fast with a long stride in pursuit of their lizard prey. The ratios of lengths of leg bones vary less in cuckoos than among other families of birds (Gatesy and Middleton 1997) and large mammals (Scott 1985). Much as in birds, in cursorial ungulates the distal leg bones are longer than the femur.Very large mammals have short distal elements of the leg and a long femur; giants support much weight on their legs, whereas cuckoos are built on a smaller scale. In contrast to the cuckoos, terrestrial squirrels are shorter-legged than arboreal squirrels; the terrestrial squirrels burrow with short legs while arboreal squirrels use their limbs to bound about on branches (Bryant 1945,

Figure 4.3. Body weight vs % tarsometatarsus: leg length in arboreal and terrestrial cuckoos.

Morphology 35 Emry and Thorington 1982, Thorington and Thorington 1989). Cuckoos with rounded wings tend to have a short tarsus or a long digit 3, and there may be other variation in toe length with life style. In all cuckoos, toe 1 is shortest, then toes 2, 4 and 3 (Stephan 2002). In the terrestrial cuckoo that is best known, the Greater Roadrunner, the pelvic muscles originate further anteriorly and laterally than in arboreal

cuckoos. These muscles may support stability and balance in a runner that places its weight alternately on one leg and then the other. Roadrunners have slender leg bones especially the tarsometatarsus, and the toes are mobile and flexible allowing the bird to vary the shape of its foot as it moves on a flat surface (Berger 1952, 1953a, Hughes 1996b). Among the couas Coua of Madagascar, the behavioral differences between species do not appear to co-vary with the dimensions of the tarsus (Milne-Edwards

Table 4.2 Comparative data on the biology of cuckoos, from the species accounts. Species Cuculinae (brood-parasitic) Clamator jacobinus, Jacobin Cuckoo Clamator levaillantii, Levaillant’s Cuckoo Clamator coromandus, Chestnut-winged Cuckoo Clamator glandarius, Great Spotted Cuckoo Pachycoccyx audeberti,Thick-billed Cuckoo Cuculus sparverioides, Large Hawk-cuckoo Cuculus varius, Common Hawk-cuckoo Cuculus vagans, Mustached Hawk-cuckoo Cuculus hyperythrus, Hodgson’s Hawk-cuckoo Cuculus nisicolor,Whistling Hawk-cuckoo Cuculus fugax, Javan Hawk-cuckoo Cuculus pectoralis, Philippine Hawk-cuckoo Cuculus solitarius, Red-chested Cuckoo Cuculus clamosus, Black Cuckoo Cuculus micropterus, Indian Cuckoo Cuculus canorus, Common Cuckoo Cuculus gularis, African Cuckoo Cuculus saturatus, Himalayan Cuckoo Cuculus optatus, Oriental cuckoo Cuculus poliocephalus, Asian Lesser Cuckoo Cuculus rochii, Madagascar Lesser Cuckoo Cuculus mechowi, Dusky Long-tailed Cuckoo Cuculus olivinus, Olive Long-tailed Cuckoo Cuculus montanus, Barred Long-tailed Cuckoo Cacomantis pallidus, Pallid Cuckoo Cacomantis leucolophus,White-crowned Cuckoo Cacomantis sonneratii, Banded Bay Cuckoo Cacomantis merulinus querulus, Plaintive Cuckoo

m wl f wl m wt f wt egg l 154 174 161 199 223 236 201 146 199 178 173 175 172 174 195 221 215 183 206 150 161 138 146 149 193 172 121 110

153 174 160 193 221 227 191 143 196 181 175 174 169 178 194 210 205 175 196 146 160 135 143 143 192 169 124 110

79 90 129 121 77 139 129 92 111 151 101 105 58 62 115 78 76 76 79 78 70 90 87 119 119 117 106 97 110 79 73 99 89 54 54 65 54 58 66 56 59 58 90 86 115 (99) 35 27 25

egg w

egg g

inc da

nestl da

27 26 27 33 24 27 31

22 21 23 24 17 19 20

7.2 6.3 7.9 10.5 3.8 5.4 6.9

12 12

18 13

12.8 13

24 28

28 24

20 16

6.1 3.4

19–20

22 26 25 23 25 21 22 22 18

18 19 19 17 18 16 16 16 14

3.9 12–14 5.2 14 5.0 12 3.6 12 4.5 3.0 3.1 3.1 2.0

17–21 20–21 21 17–20 20–23

23 21 25

16 15 16

3.3 2.6 3.5 12–14

18 20

14 14

2.0 2.2

18–20

36 The Cuckoos Table 4.2 contd. Species Cacomantis variolosus sepulcralis, Brush Cuckoo Cacomantis castaneiventris, Chestnut-breasted Cuckoo Cacomantis flabelliformis, Fan-tailed Cuckoo Chrysococcyx osculans, Black-eared Cuckoo Chrysococcyx basalis, Horsfield’s Bronze-cuckoo Chrysococcyx m. minutillus, Little Bronze-cuckoo Chrysococcyx lucidus plagosus, Shining Bronze-cuckoo Chrysococcyx ruficollis, Rufous-throated Bronze-cuckoo Chrysococcyx meyeri,White-eared Bronze-cuckoo Chrysococcyx maculatus, Asian Emerald Cuckoo Chrysococcyx xanthorhynchus,Violet Cuckoo Chrysococcyx klaas, Klaas’s Cuckoo Chrysococcyx cupreus, African Emerald Cuckoo Chrysococcyx caprius, Diederik Cuckoo Surniculus dicruroides, Fork-tailed Drongo-cuckoo Surniculus velutinus, Philippine Drongo-cuckoo Surniculus lugubris, Square-tailed Drongo-cuckoo Microdynamis parva, Dwarf Koel Eudynamys s. scolopacea, Common Koel Eudynamys s. cyanocephala, Common Koel Urodynamis taitensis, Long-tailed Cuckoo Scythrops novaehollandiae, Channel-billed Cuckoo Cuculinae (Old World), nesting Ceuthmochares australis,Whistling Yellowbill Phaenicophaeus curvirostris, Chestnut-breasted Malkoha Phaenicophaeus pyrrhocephalus, Red-faced Malkoha Phaenicophaeus diardi, Black-bellied Malkoha Phaenicophaeus sumatranus, Chestnut-bellied Malkoha Phaenicophaeus tristis, Green-billed Malkoha Phaenicophaeus viridirostris, Blue-faced Malkoha Taccocua leschenaultii, Sirkeer Malkoha Rhinortha chlorophaea, Raffles’s Malkoha Zanclostomus javanicus, Red-billed Malkoha Rhamphococcyx calyorhynchus,Yellow-billed Malkoha Dasylophus superciliosus, Rough-crested Malkoha Dasylophus cumingi, Scale-feathered Malkoha Cuculinae (New World), nesting Coccycua pumilus, Dwarf Cuckoo Coccycua cinerea, Ash-colored Cuckoo Coccyzus erythropthalmus, Black-billed Cuckoo Coccyzus americanus,Yellow-billed Cuckoo

m wl 116 110 144 118 103 94 106 95 89 109 100 96 104 117 136 117 126 105 195 217 189 354

f wl m wt f wt egg l 115 33 32 20 113 30 32 21 141 50 50 21 117 29 32 20 101 22 24 18 92 19 19 19 106 23 25 18 95 21 24 89 18 21 109 25 30 17 100 20 26 16 95 26 29 19 102 38 37 20 121 34 41 21 135 38 22.5 117 33 38 124 30 31 21 102 43 51 189 229 31 215 262 257 34 186 110 124 24 347 708 660 41

egg w 15 15 15 14 12 13 12

egg g inc da 2.5 12–13 2.6 2.6 2.2 1.4 11–13 1.8 1.4 13–14

12 12 13 15 15 17

1.4 1.3 1.8 11–12 2.5 13–14 2.6 12 3.6

16

3.0

124 169 155 130 139 163 135 158 115 144 177 151 157

124 172 157 131 139 161 131 154 116 145 180 157 159

68 64 30 152 144 39.7 36 66 57 30 114 28.5 116 114 34 84 29 219 140 36 56 62 28 98 97 29 36 115 120 174 167

23 28.4 27 24.5 23.5 26 25 26 24 23 31

102 107 135 139

102 110 139 144

36* 37 47 58

19.6 19 21 23

24.9 45 25.5 54 27 65 30

23 24 17 29

9.1 13–14 10.8 14–17 3.8 16 19.1 8.7 17.7 14.5 10.0 8.7 13.5 10.0 13.5 8.9 8.5 19.1

nestl da 17–19

18 17–19 18–21

19–21 18–20 20–22

19–28 19–28 21–29 17–24

13

5.3 12 5.1 6.6 10–11 8.8 10–11

12 8–9 8–9

Morphology 37 Table 4.2 contd. Coccyzus euleri, Pearly-breasted Cuckoo Coccyzus minor, Mangrove Cuckoo Coccyzus melacoryphus, Dark-billed Cuckoo Coccyzus lansbergi, Gray-capped Cuckoo Coccyzus merlini, Cuban Lizard-cuckoo Coccyzus vieilloti, Puerto Rican Lizard-cuckoo Coccyzus longirostris, Hispaniolan Lizard-cuckoo Coccyzus vetula, Jamaican Lizard-cuckoo Coccyzus pluvialis, Chestnut-bellied Cuckoo Coccyzus rufigularis, Rufous-breasted Cuckoo Piaya cayana, Squirrel Cuckoo Piaya melanogaster, Black-bellied Cuckoo Coccycua minuta, Little Cuckoo Couinae, nesting Carpococcyx radiatus, Bornean Ground-cuckoo Carpococcyx renauldi, Coral-billed Ground-cuckoo Coua gigas, Giant Coua Coua coquereli, Coquerel’s Coua Coua serriana, Red-breasted Coua Coua reynaudii, Red-fronted Coua Coua cursor, Running Coua Coua ruficeps, Red-capped Coua Coua cristata, Crested Coua Coua verreauxi,Verreaux’s Coua Coua caerulea, Blue Coua Centropodinae, nesting Centropus celebensis, Bay Coucal Centropus unirufus, Rufous Coucal Centropus melanops, Black-faced Coucal Centropus nigrorufus, Sunda Coucal Centropus milo, Buff-headed Coucal Centropus goliath, Goliath Coucal Centropus violaceus,Violaceous Coucal Centropus menbeki, Greater Black Coucal Centropus chalybeus, Biak Coucal Centropus ateralbus, Pied Coucal Centropus p. phasianinus, Pheasant Coucal Centropus p. spilopterus, Kai Coucal Centropus bernsteini, Lesser Black Coucal Centropus rectunguis, Short-toed Coucal Centropus steerii, Black-hooded Coucal Centropus s. sinensis, Greater Coucal

128 137 115 114 174 129 134 124 177 171 171 136 105

133 132 116 114 180 130 137 126 194 184 184 135 105

49 64 46 50 134 80 92 95 130 128 104 99 41

57 67 54

31 29 26 40 34 37 32.5 40 38 35 30 24

23 22 20 30 25 25 24 31 25 26 23 19

9.1 7.8 11–12 5.8 27.1 11.8 12.8 10.4 21.3 13.1 13.1 18 8.8 4.8

259 278 221 144 165 137 130 168 139 134 195

253 455(540) 47 278 (400) 44 214 413 420 43.5 148 33.5 168 298 139 240 163 36 133 118 34 168 190 34 146 144 145 35 131 200 236 254 37

35 34 32 25

35.6 28.1 18–19 25.7 11.6

173 156 162 208 272 263 311 225 203 202 226 228 174 164 159 203

182 162 168 201 162 211 238 224 267 769* 271 401 622 261 (500) 221 503 529 203 214 330 342 249 302 445 252 180 146 180 197 238* 164 179 163 215 255 370

155 92 105 189 100 93 40

28 23 28 26.5

15.6 9.9 14.7 13.6

28.5

16.6

31 39

26 31

11.6 21.3

43 37

34 30

27.5 18.4

41 38 36 32 37

33 29 27 25.5 30

24.7 17.7 14.5 11.5 18.4

36

28

7–9 8–13

8⫹

18–24

15

17–21

15.6 15–16

18–22

38 The Cuckoos Table 4.2 contd. Species m wl f wl m wt f wt egg l egg w egg g inc da nestl da Centropus s. andamanensis, Brown Coucal 180 183 234 35 28 15.2 Centropus toulou, Madagascar Coucal 148 168 173 180 33 26 12.3 14–16 19 Centropus grillii, African black Coucal 152 169 100 151 31 24 9.9 18–20 Centropus bengalensis, Lesser Coucal 134 159 86 148 28 24 8.9 Centropus v. viridis, Philippine Coucal 148 160 112 134 30 25 10.4 14 Centropus chlororhynchus, Green-billed Coucal 167 175 35 27 14.1 Centropus leucogaster, Black-throated Coucal 189 198 293 336 Centropus anselli, Gabon Coucal 186 195 210 Centropus monachus, Blue-headed Coucal 187 186 171 237 35 27 14.1 Centropus cupreicaudus, Coppery-tailed Coucal 214 224 272 299 38 28 16.5 17⫹ Centropus senegalensis, Senegal Coucal 160 164 169 169 34 26 12.7 17–19 18–20 Centropus superciliosus,White-browed Coucal 146 154 124 136 33 25 11.4 14–15 18–20 Crotophaginae, nesting (cooperative breeder) Crotophaga major, Greater Ani 203 201 171 152 45 38 35.9 13–14 8–10 Crotophaga ani, Smooth-billed Ani 149 149 111 92 35 27 14.1 13–14 13–17 Crotophaga sulcirostris, Groove-billed Ani 135 131 80 67 31 24 9.9 12–14 10 Guira guira, Guira Cuckoo 179 174 139 145 43 32 2.46 12–13 12–15 Neomorphinae (nesting or brood parasite) Tapera naevia, American Striped Cuckoo (brood parasite) 119 113 49.4 51.1 22 17 3.5 15–16 18 Dromococcyx phasianellus, Pheasant Cuckoo “ 166 163 85 98 25 14.5 2.9 Dromococcyx pavoninus, Pavonine Cuckoo “ 134 137 50 45 21 15 2.6 Morococcyx erythropygus, Lesser Ground-cuckoo 101 102 63 68 27 21 6.6 Geococcyx californianus, Greater Roadrunner 178 171 344 309 39 30 19.4 17–18 17–19 Geococcyx velox, Lesser Roadrunner 146 141 186 174 35 26 13.1 Neomorphus geoffroyi, Rufous-vented Ground-cuckoo 169 164 350 386 43 32 24.3 Neomorphus rufipennis, Rufous-winged Ground-cuckoo 170 164 435 328 40 31 21.3 Neomorphus pucheranii, Red-billed Ground-cuckoo 172 166 330 Data are from the species accounts: m wl ⫽ male wing (mm), f wl ⫽ female wing (mm), m wt ⫽ male weight (g), f wt ⫽ female weight (g), egg l ⫽ egg length (mm), egg w ⫽ egg width (mm), egg g ⫽ egg weight (g), inc da ⫽ incubation period (days), nestl da ⫽ nestling period (days). In Graph 1, for Carpococcyx, brain size is for Carpococcyx renauldi and body size is for C. radiatus; the ground-cuckoos are about the same in other body measurements. Figures in parens were based on single individuals of atypical body condition, as described in the species accounts, and these are not included in all graphs. * indicates unsexed birds.

and Grandidier 1879, Berger 1960); tarsi are long in the arboreal Coua caerulea and the terrestrial C. gigas and C. ruficeps, and short in the climber C. reynaudii and the terrestrial C. cursor. The main difference between couas is in the longer tendons and the site of their attachment in the long-legged forms

(Berger 1960). Fourth, wing shape in cuckoos varies with the longest primary ranging from the 9th (the most pointed wing) to the 2nd (the most rounded and broad wing).The wing length and shape vary with migratory behavior. Pointed wings are seen in the

Morphology 39

Figure 4.4. Wing length and body mass in cuckoos.

migratory Cuculus species and rounded wings in the ground-cuckoos and in coucals Centropus, where the longest primary is P7, P6 or P5 in Lesser Coucal C. bengalensis and P4, P3 or P2 in the large Greater Black Coucal C. menbeki. The migratory northern hawk-cuckoos have more pointed wings than the more tropical species (Rufous Hawkcuckoo Hierococcyx hyperythrus and Whistling Hawk-cuckoo H. nisicolor, compared with Javan Hawk-cuckoo H. fugax and Philippine Hawkcuckoo H. pectoralis). The long wings of broodparasitic cuckoos in the Old World are related to their migratory behavior and not to brood parasitism, insofar as the New World brood-parasitic cuckoos are residents and have short wings (Table 4.2, Figure 4.4). Many nesting cuckoos are shortwinged, especially resident tropical birds, and so are a few forest brood parasites especially the African long-tailed cuckoos Cercococcyx, which have a

rounded wing with the inner primaries P4-7 longer than the outer primaries P8-9. Wing shape in cuckoos has been described in detail from the lengths of the wing feathers in relation to ulna length (Stephan 2001a, Figure 4.5). In Figure 4.5, the primaries are numbered from the proximal end of the carpometacarpus, the secondaries S1 from the distal end of the ulna, S6 and the more inner secondaries from the proximal end of the ulna, and S2-5 from their corresponding points of attachment along the ulna. In another comparison, the length of “free” primaries relative to “free” secondaries captures the shape of the wing, the term indicating the length of feathers that extend beyond the wing coverts (Stephan 2001a, Figure 4.6). All cuckoos can fly and have a keeled sternum, even in the terrestrial roadrunners which more often walk and run than fly. The shape of the

Figure 4.5. Wing shape in cuckoos, with relative lengths of the primaries and secondaries (after Stephan 2001a). Lengths of the wing feathers are plotted from their positions of attachment along the ulna, which is represented by the two small circles on each wing: a, Scythrops novaehollandiae; b, Cuculus canorus; c, Eudynamys scolopacea malayana; d, Clamator levaillantii; e, Rhamphococcyx calyorhynchus; f, Piaya cayana; g, Centropus celebensis; h, Coua ruficeps; i, Tapera naevia; j, Geococcyx californianus.

Morphology 41

Figure 4.6. Wing shape in cuckoos, with extent of “free” primaries and secondaries: a, rounded wing, Geococcyx californianus; b, pointed wing, migratory Cuculus canorus (after Stephan 2001a).

sternum varies with life habits among Coua species, the more arboreal couas with a more convex and strongly keeled sternum than the terrestrial couas (Milne Edwards and Grandidier 1879). Finally, much variation in wing morphology of cuckoos can be explained by their phylogenetic associations, such as coucals being short-winged owing to being related to other short-winged coucals, and secondarily owing to their terrestrial and non-migratory behavior. Coucals have short wing bones as well (Milne-Edwards and Grandidier 1879, Berger 1952, 1953a); and the New World terrestrial roadrunners Geococcyx and to a lesser extent the anis Crotophaga have shorter wing bones relative to leg bones than the Coccyzus cuckoos (Larson 1930, Engels 1938, Berger 1952).

Sexual size dimorphism Among most cuckoos the sexes are nearly the same in size (Table 4.2, Figures 4.7, 4.8). Sexual dimorphism in body size does not vary among cuckoos as

much in relation to their breeding biology as to their systematic status. Among the coucals, females are larger than males in the monogamous species such as Pheasant Coucal, where the female is half again the mass of the male and both sexes care for the young, and in the sometimes-polyandrous African Black Coucal where the male looks after the young. The degree of sexual size dimorphism is consistent across subspecies in the coucal species that vary geographically in size (as in Greater Coucal and Philippine Coucal). The largest degree of sexual size dimorphism occurs in the smaller coucals, unlike birds such as grouse where sexual size dimorphism is greater in the larger species (Rensch 1950, Payne 1984, Abouheif and Fairbairn 1997). Among the smaller coucals the females are larger than the males, with larger coucals the sexes are the same size, and with the largest coucals the males are larger than the females. With other kinds of birds, reversed sexual size dimorphism where females are the larger sex is sometimes associated with polyandry, as in buttonquail, tinamous Tinamidae, painted snipe Rostratulidae, jacanas Jacanidae and some other shorebirds (Payne 1984, Andersson 1994, 1995). Insofar as reversed sexual size dimorphism occurs both in some monogamous birds and in polyandrous birds, the large size of female coucals does not necessarily indicate a polyandrous mating system among coucals. Other explanations of sexual size dimorphism include the ability of large females to lay more or larger eggs, or to defend the young where the small and agile males forage for the females and leave them at the nest, as with many monogamous hawks and owls. Reduced competition for food between mates among some birds is also a correlate of sexual size dimorphism, though the use of prey of different sizes may follow sexual size dimorphism that evolved for other reasons (Andersson 1994). Coucals have a long hallux claw, much as do some other terrestrial birds such as larks and pipits, while other birds have long claws on all the toes (longclaws Macronyx spp., jacanas Jacanidae, Longtoed Lapwing Vanellus crassirostris, Purple Heron Ardea purpurea), birds that live in grassland and on weak-stemmed and floating aquatic vegetation. In coucals the hallux claw ranges from 32–35% of tarsus length (in the large Buff-headed Coucal

42 The Cuckoos

Figure 4.7. Sexual size dimorphism in body size in cuckoos: body mass.

Centropus milo and Violaceous Coucal C. violaceus), to 68–76% of tarsus length (in the small marshliving African Black Coucal C. grillii and Lesser Coucal C. bengalensis). An exception is Short-toed Coucal C. rectunguis in which the hallux claw is only 23% of tarsus length. A test of whether the hallux claw is a sexually selected trait in coucals is whether the claw is proportionally larger in one sex. The diagonal line in Figure 4.9 indicates the position where male and female claws are equally long in relation to tarsus length. About the same number of coucal species have males with relatively longer claws on average than females. When claw length is compared with body size, there is no trend for species with a large size to have proportionally larger claws in females than in males. Also there is no increase in relative size of the claw of the sex with larger body size, as there is with measurements in other polygamous birds (Payne 1984).The lack of sexual

dimorphism in this prominent feature of coucals does not support the idea that the reversed sexual size dimorphism of coucals has been driven by polyandry. Sexual differences in size of cuckoos might be related to female aggressive behavior. In some hawks, falcons and owls, the female takes the lead in social dominance or territorial behavior of the pair (Pleasants and Pleasants 1989, Mueller 1990). Field observations on mating among cuckoos do not suggest female social dominance over males. In the cooperatively breeding anis, females are about the same size as males. Females compete for the position of their eggs in a communal nest and they remove each others’ eggs, yet within a group the body size is not larger in the socially dominant females at a nest (Vehrencamp et al. 1986). Large female size may also be related to egg size in anis, where large eggs provide nestlings a head start for growth in the competition for parental care in a brood.

Morphology 43

Figure 4.8. Sexual size dimorphism in body size in cuckoos: wing length.

Plumage The body plumage of ground-cuckoos, couas and malkohas is soft and lax.The plumage becomes wet in the rain, and the birds often sun themselves to dry after a shower and even on a warm morning (Forbush 1927, Cracraft 1964). The anis have smooth, dense glossy plumage, a texture that appears to be related to foraging in wet foliage. Their feathers still become soaked, and in cool wet weather the anis spread their wings in the sun’s rays to dry (Skutch 1983), like a vulture or a cormorant. Coucals that forage in wet habitats do the same. None of these plumage traits are exclusive to cuckoos, as some gruiform birds and doves have the same kind of lax plumage and soft pastel colors of purples and pinks like the plumage of couas and malkohas. Cuckoos also have bristles, specialized contour feathers with a stiff rachis, and barbs are unfused and restricted to the feather base (Stettenheim 1973).

Specialized feathers are found on the face of many nesting cuckoos. The eyelashes are small and inconspicuous in several cuckoos but are conspicuous, stiff and curved without vanes or barbs in some malkohas (Yellow-billed Malkoha Rhamphococcyx calyorhynchus, Green-billed Malkoha Phaenicophaeus tristis, Black-bellied Malkoha P. diardi, short and thin in Blue-faced Malkoha P. viridirostris, Rough-crested Malkoha Dasylophus superciliosus and Scale-feathered Malkoha D. cumingi), and they are particularly long and stout in Sirkeer Malkoha Taccocua leschenaultii which live in semi-arid sandy habitats.The eyelashes are short in Red-faced Malkoha P. pyrrhocephalus and in the African yellowbills Ceuthmochares; they are absent in the small Raffles’s Malkoha Rhinortha chlorophaea. The eyelashes are conspicuous in New World squirrel-cuckoos Piaya; they are short in Coccycua and Coccyzus. Eyelashes are also found in anis and in large terrestrial cuckoos including Coua, Carpococcyx, Centropus, Geococcyx and Neomorphus. Eyelashes are short in Clamator and are absent or

44 The Cuckoos

Figure 4.9. Size and sexual size dimorphism in hallux claw in coucals Centropus. Measurements are in the species accounts.

not noticeable in other Old World brood-parasitic cuckoos. The face and head have some feathers with long delicate projections of the rachis or shaft from the loose-webbed base of vaned feathers in the malkohas. In Sirkeer Malkoha these shafts are stiff and only a small length of the feather base has a vane. In Chestnut-bellied Malkoha P. sumatranus the feathers are recurved forward especially around the base of the bill. In Blue-faced Malkoha P. viridirostris the lower head, neck and breast feathers are forked. In these unique feathers, the bifurcation is formed by stiff terminal barbs that cling together on each side of the central feather shaft, which protrudes as a fine free black hair-like bristle. Delicate hair-like shafts occur in some other cuckoos including Great Spotted Cuckoo, although these structures are not evident in most Old World brood-parasitic cuckoos

(Cuculus, Chrysococcyx, Eudynamys, Scythrops). In the New World, the Guira Cuckoo and roadrunners have long delicate feathers around the face (Figure 4.10). The feathers on the face and head occur in an extreme form with a thickening and fusion of barbs in two Philippine malkohas. Rough-crested Malkoha Dasylophus superciliosus has superciliary crests formed by stiff reddish barbs; at the base a few more basal barbs form a loose-webbed vane on each side of the shaft. Perhaps the most bizarre head feathers are in Scale-feathered Malkoha D. cumingi whose terminal barbs are fused at the tip into a single shiny metallic flat blue-black shield that contrasts with the white vane near the base of the feather.The tips are formed of fused barbs that are densely packed with melanin, while the surfaces are thick layers of keratin with little melanin (Brush

Morphology 45

Figure 4.10. Facial feathers of cuckoos: Greater Roadrunner Geococcyx californianus and Guira Cuckoo Guira guira.

1965). The tips of the crest feathers are twisted to reveal the same iridescent color on each side of the head. Some coucals have hackles, robust elaborate feathers on the neck, with conspicuous stiff feather shafts, the shaft color often in contrast to the feather vane. In Greater Roadrunner the head has stout, bare, black terminal shafts that project outward and backward over the eyes, barbed bristles around the base of the bill, and short recurved feather shafts all around the head, while Rufousvented Ground-cuckoo has dark bare eyelash bristles above and below the eye, and semibristles in front of the eye. Many cuckoos have long tails that function as rudders in slow flight in the arboreal species, and as steering devices and balancers in the cursorial ground-living species. Cuckoo tails are also conspicuous in courtship behavior.The tails often have

contrasting colors, with broad or narrow white tips in many malkohas and New World arboreal cuckoos, and red tips on green tails in Chestnut-breasted Malkoha P. curvirostris. Raffles’s Malkohas Rhinortha chlorophaea are sexually dimorphic, the males with the tail finely barred black and gray in a washboard pattern and broadly tipped white; the females with the tail rufous and a black subterminal band and white tip. In several cuckoos (Common Cuckoo, Sirkeer Malkoha,Yellow-billed Cuckoo) the tail is spread and held above the back, exposing the color pattern on the dorsal surface in frontal display to the mate. In some cuckoos the edges of the tail feathers are “notched” with a V-shaped area of contrasting color, the apex toward the feather shaft; in others the notches are extended well towards the shaft and almost form a bar. The notches are distinctive of certain species especially of the brush cuckoos Cacomantis. Tails are most elaborate in the small ground cuckoos Dromococcyx of the New World tropics, with the long rectrices arrayed in a fan and the upper tail coverts cascading nearly the full length of the tail, each pointed covert feather with a black band and a delicate white spot at the tip. In display the bird arches the tail coverts over the body.

Natal down and naked nestlings Natal down of cuckoos is hair-like, a single keratinized shaft called a “trichoptile” (hair-feather), rather than an array of non-interlocking barbs as is the fluffy natal down of many kinds of birds.These trichoptiles are not down-like in structure but are down-like in being the first set of feathers on a young bird.They are present at hatching or within a day or two in most cuckoos, and are developmentally homologous with the natal down of other birds. The trichoptile grows from the same follicle as the juvenile contoured feather, and when the pinfeather bursts open, the hair-like natal down remains attached to the tip for a day or two in the same position on the tip of the feather as the fluffy down of songbirds. The down is often worn away by the time the nestling fledges. This down is white in many cuckoos; it differs in color between certain closely-related species (gray in

46 The Cuckoos Yellow-billed Cuckoo, white in Black-billed Cuckoo).A few nesting cuckoos have not been studied as hatchlings but have soft body feathers as feathered nestlings (Scale-feathered Malkoha); these may be loose-webbed juvenile contoured feathers rather than natal down. Malkohas, of which the nestling has been studied and described, have hair-like natal down. In young cuckoo nestlings with dark skin, such as Black-billed Cuckoo and Greater Roadrunner, the white hair-like down forms a strong pattern of contrast with the skin. In the nesting cuckoos, these feathers appear to be continuous with the sheath that covers the growing first generation of contoured feathers. Hair-like down is prominent in the coucals Centropus, where the feathers are white, stiff and sometimes as long as 20 mm and the nestlings look like spiny hedgehogs; less typical of young coucals is Lesser Black Coucal C. bernsteini which has only short white tips attached to the growing contoured feathers on the head. Natal down is absent in the nesting Old World ground-cuckoos Carpococcyx and Madagascar couas Coua, both as hatchlings and when they fledge. In adult cuckoos, fluffy down-like feathers are limited to the apteria. Natal down is absent in the brood-parasitic cuckoos, and this may be explained by the fact that the naked skin of nestlings is sensitive to the touch of host eggs which they evict, and natal down would interfere with this close contact. An exception is Shining Bronze-cuckoo, which has a few small hairlike feathers in New Zealand and New Caledonia, and few or none in southern and Western Australia where the young are naked; Little Bronze-cuckoo also has some natal down. Nestling brood-parasitic cuckoos evict the host egg from the nest; natal down in those bronze cuckoos is mainly on the head, thigh and side, rather than on the back where the egg is balanced in eviction behavior. Nestling crested cuckoos Clamator, like other Old World brood-parasitic cuckoos, lack natal down, and nestling brood-parasitic American Striped Cuckoos in the New World also hatch naked without down.

Plumage change with age Juvenile plumages are variable in a few broodparasitic cuckoos. Jacobin Cuckoo and Levaillant’s

Cuckoo juveniles occur that are dull versions of the adult plumage phases, and in East-Central Africa a rufous color phase occurs in Levaillant’s Cuckoo. Juvenile Diederik Cuckoos vary in plumage in some regions, with rufous juveniles in areas where the cuckoos parasitize bishop finches, Euplectes species. It is unknown whether the color variants of cuckoos are genetically different (diederiks have egg races in these areas, and the plumage may correspond with egg races), or the rufous plumage is a result of bishop foster parents rearing the young on seeds rather than entirely on insects as in the greenplumaged juveniles that are reared by the more insectivorous weavers Ploceus. Juvenile plumage in most cuckoos is similar to adult plumage, especially so in the nesting cuckoos. The transition from juvenile to adult appearance occurs early in life. The main difference in plumage is the shape of the rectrices (narrow and pointed in juvenile plumage, more broad and rounded or truncate in the adult) and the presence and extent of white tips on the rectrices (less in juveniles), or the presence of barred rectrices (barred in juvenile coucals, not barred in adult coucals). Juveniles are quick to assume adult head plumage. Philippine Red-crested Malkoha and Scale-feathered Malkoha with their elaborate head plumage grow these feathers within a few weeks of leaving the nest and have their showy adult head plumage by the time the tail is fully grown. A few flight feathers in wing and tail are often retained into the first breeding plumage of the coucals Centropus spp. The cuckoos with the most strongly age-graded plumage, where juvenile plumage differs from adult plumage, are certain brood-parasites. None of these cuckoos mimic the plumages of their host species. Juvenile Common Koels Eudynamys scolopacea have black plumage like their crow hosts in much of their range, yet this is characteristic only of male juveniles; female koels have a spotted brown juvenile plumage that is intermediate between that of juvenile males and adult females. Behavioral associations with the host species may nevertheless have some effect on juvenile plumage. In koels the plumage somewhat resembles that of their host species, dark like crows in India where

Morphology 47 koels parasitize crows, and streaked brown in Australia where they parasitize brown honeyeaters. The juvenile plumage of Philippine Drongocuckoo Surniculus velutinus is similar to that of a paradise flycatcher Terpsiphone cinnamomea, but the host of this cuckoo is unknown and the similarity of juvenile plumages in these birds is no more than a teasing clue to the mystery of their breeding behavior. A distinctive subadult plumage that is retained for a year after the molt from juvenile plumage occurs in the Asian hawk-cuckoos, as described for some of these birds by Mayr (1938b); previously these had been confused as a female plumage (Siebers 1930). Species with a distinct subadult plumage are Large Hawk-cuckoo Hierococcyx sparverioides, Dark Hawk-cuckoo H. bocki, Rufous Hawk-cuckoo H. hyperythrus, Whistling Hawkcuckoo H. nisicolor, Javan Hawk-cuckoo H. fugax, and Philippine Hawk-cuckoo H. pectoralis. At least one hawk-cuckoo breeds in subadult plumage, Philippine Hawk-cuckoo females laying in this plumage as well as in the definitive unstreaked adult plumage. In the Sulawesi Cuckoo Cuculus crassirostris also a distinct subadult plumage occurs and specimens with this plumage are more common in museum collections than are the juveniles, which have a white head much as that of the closely related Indian Cuckoo C. micropterus, but nothing is known of their breeding. Burchell’s Coucal Centropus superciliosus burchellii breed in subadult plumage as well as in the definitive plumage that lacks the white eye-stripe. Another coucal with subadult plumage is a subspecies of Blue-headed Coucal C. monachus heuglini in the southern Sudan, where its breeding has not been studied. In known cases, the development of sexual maturity is completed well before the development of adult plumage. Among these cuckoos the sexes are alike in plumage and both sexes have the full set of plumages. In both Centropus superciliosus and C. monachus, the other subspecies lack the extra plumage and breed in the plumage that looks like the “subadult” plumage of these two subspecies. The lack of a subadult plumage is probably related to the rapid development of sexual maturity of cuckoos in general.

Cuckoos sometimes retain a few juvenile feathers through their first year, particularly in the wing (e.g., Great Spotted Cuckoo, Diederik Cuckoo Chrysococcyx caprius and Klaas’s Cuckoo Chrysococcyx klaas).The retained juvenile feathers show that these cuckoos breed in their first year, as these birds with mixed plumage have been taken with eggs in the oviduct. Earlier studies misinterpreted the great variation in adult plumage of some cuckoos (e.g., Black Cuckoo, Bannerman 1921) to be due to the number of years it takes a bird to develop an adult plumage. The larger number of specimens now available for examination in museums shows that each adult plumage form follows a single molt from the all-black plumage of the juvenile, in which the tail tip is black and not white as in the adult. Insofar as cuckoos usually breed in their first year, there is no reason to expect a series of subadult plumages. If birds that attempt to breed in their first year are less successful in reproduction over their lifetimes than those that do not attempt to breed at that time, then a delay in sexual activity in the first year may be the more productive life history strategy, and we should see sexual inactivity in birds with a lessthan-definitive plumage. The idea of a “delayed plumage maturation” is derived from the more general concept of “delayed maturation”, which proposes that few first-year birds are competent to breed, and that early reproduction negatively affects their later success (Lack 1954, Selander 1965,Williams 1966).A logical consequence of this life history theory is that individuals with plumages unlike the older adults are sexually inactive. If this is true, we predict that these birds do not breed, or when they do breed they have a lower survival and lifetime reproduction. However in many cuckoos the birds with odd plumages are sexually active. Female cuckoos with a few juvenile feathers in their plumage have been taken with eggs in the oviduct so they were actively breeding, as described in the species accounts of Plaintive Cuckoo C. merulinus, Brush Cuckoo and black-billed Common Koel Eudynamys scolopacea melanorhyncha. There is no indication that brood-parasitic birds have a long life (Payne 1977b).They are expected to breed in their first year: if they deferred breeding in the first year, they would have a high chance of dying before the next year. In some cooperatively-living species of

48 The Cuckoos songbirds the definitive adult plumage may not occur while the young are living in a group with older, socially and sexually dominant individuals (Rowley and Russell 1997). These long-lived cooperatively breeding birds are one group where “delayed maturation” might occur, although with removal of older birds in the group, the younger birds become breeders themselves. However, in the cooperativelybreeding anis Crotophaga and Guira Cuckoo Guira guira which compete for social status at communal nests, where increased maturity might give the birds some benefit in competing with their group members, the species do not have distinct subadult plumages, and these birds do not have a long life span. A few coucals molt into a winter non-breeding plumage before they molt into the adult breeding plumage (African Black Coucal Centropus grillii, Lesser Coucal C. bengalensis, Australian Pheasant Coucal C. p. phasianinus, and some other coucals). These birds are known to molt from breeding plumage back to a nonbreeding plumage, at least in forms where specimens are available from all seasons; a second- and later-year non-breeding plumage is not known for the New Guinea races of Pheasant Coucal. In Blue-headed Coucal Centropus monachus fischeri and in White-browed Coucal C. superciliosus burchellii and C. s. fasciipygialis, a distinctive subadult plumage is retained for a year after the molt from juvenile plumage and before the molt to definitive adult plumage. In addition, Stresemann (1913a) thought Greater Coucal C. sinensis had a series of subadult plumages based on the variation in the retained juvenile feathers in birds in partial adult plumage. He also noted the absence of museum specimens of Lesser Coucal in molt from adult breeding plumage to nonbreeding plumage.The lack of molting specimens may be due to the secretive behavior of birds after the breeding season, rather than to a lack of a post-breeding molt. It would be useful to keep these birds in captivity and observe their sequence of molts and plumages. Some coucals also have a distinct adult nonbreeding plumage, at least in their first year. They molt from juvenile plumage into this alternate nonbreeding plumage and maintain the plumage for a season before they molt into an adult breeding plumage. The nonbreeding plumage has sometimes been

overlooked owing to its similarity with the juvenile plumage. Although there is no evidence of an adult nonbreeding plumage for most coucals, distinct nonbreeding plumages are known for a few species including Lesser Coucal C. bengalensis throughout its range and the Pheasant Coucals C. phasianinus nigricans, C. p. thierfeldi and C. p. spilopterus in New Guinea and the Kai Islands. The question remains whether birds with a distinct subadult or nonbreeding plumage maintain the adult breeding plumage all year into later years, or assume a distinct non-breeding plumage as in birds in molt from black to pale plumage in Pheasant Coucals of Australia, Lesser Coucal, African Black Coucal and Madagascar Coucal C. toulou. Again, it would be informative to watch the plumage change through the seasons within an individual bird species.

Sexual dimorphism in plumage A few cuckoos are sexually dimorphic in plumage color, perhaps more among the Old World parasitic cuckoos where males are conspicuous and glossy and females are cryptic and not so glossy like the African and Asian Chrysococcyx, and the Common Koel where males are black and females are brown. The eye-ring and iris color and pattern (a ring in the iris) also differ between the sexes in some cuckoos, both in glossy cuckoos where sex is apparent from eye color like the Little Bronze-cuckoo and Shining Bronze-cuckoo, and in several malkohas. Conspicuous plumage dimorphism is not known in nesting cuckoos except in Raffles’s Malkoha Rhinortha chlorophaea.

Plumage phases and color morphs Adult plumage is typically cryptically colored in the brood-parasitic cuckoos. In several brood-parasitic cuckoos more than one adult plumage phase occurs within a population and these color phases are limited to the females. In several Cuculus species the females occur in either a gray phase like the adult male or a barred rufous (or “hepatic”) color phase. The proportion of female Common Cuckoos with gray or rufous plumage varies in Europe, where rufous females are most common in Central Europe,

Morphology 49 and the proportion of color phases parallels that of Accipiter hawks and Falco falcons (Voipio 1953).The variation in plumage may be density-dependent, where a bird in the rarer plumage phase has an advantage in not being recognized by naive songbirds, while the more common form is recognized and the cuckoo is chased away from the nest before it can lay (Payne 1967). Distinct plumage color phases (gray without bars, rufous with bars) also occur in breeding female Plaintive Cuckoo and Brush Cuckoo (not in all subspecies). Plumage variation in the Pallid Cuckoo Cacomantis pallidus of Australia appears to the number of rufous feathers grown after the postjuvenile molt (not retained juvenile feathers, which are white and black) and after adult molt in later years in breeding females. Finally, female Common Koels Eudynamys scolopacea melanocephala in Sulawesi have three plumage phases, and females in each of these plumages are known to lay as they have each been collected with an egg in the oviduct (Stresemann 1940). Two species of brood-parasitic crested cuckoos Clamator, Jacobin Cuckoo and Levaillant’s Cuckoo, have color phases in both sexes: a pied phase that is white or whitish below, and a black phase where white is limited to a wing patch (and in Levaillant’s Cuckoo to the tip of the tail). In both cuckoos the black phase is limited to a small part of the distributional range of the species in Africa. Color morphs and locally distinct subspecies that differ in plumage color occur among several nesting cuckoos, particularly the coucals. In some the plumage is partly albinistic. In the Andaman Islands, Greater Coucal Centropus sinensis andamanensis has a pale body (buffy white or dusky gray) rather than a black body as in most other subspecies, and on Kangean I south of Borneo the form C. s. kangeangensis likewise has a pale body (buffy white or dusky gray). On Timor the only specimen collected of Pheasant Coucal C. phasianinus has been described as a subspecies (C. p. mui) and has a white breast, in contrast to other populations of Pheasant Coucal which have black underparts in breeding plumage and buff with paler streaks in nonbreeding plumage. Goliath Coucal C. goliath in the northern Moluccas has a whitish morph in Halmahera. Melanistic or black plumages occur in certain coucal populations on

islands, especially in Philippine Coucal C. viridis with two subspecies having black wings (C. v. mindoroensis, C. v. carpenteri) unlike the rufous wings of coucals elsewhere in the Philippines, and on Luzon an uncommon white phase occurs in a population where most birds are black and rufous. Pied Coucal C. ateralbus in the Bismarck Archipelago have four distinct color phases in the amount of white in the adult plumage; birds with intermediate plumage are also known. Senegal Coucal C. senegalensis have a dark rufous color phase “epomidis” in humid areas of coastal West Africa; other birds in these populations are white below.

Other comments on plumage The cuckoo wing is eutaxic, with eutaxy the number of secondaries equal to the number of upper greater secondary coverts (Stephan 1970). The alternative condition in birds is diastataxy, where the wing has a missing secondary S5 (or an extra secondary covert between S4 and S5). The condition varies among orders of birds and within some families such as the megapodes Megapodiidae, kingfishers Alcedinidae, rails Rallidae and waders Scolopacidae, and it varies among the birds suggested as relatives of cuckoos (diastataxic in turacos Musophagidae, parrots Psittacidae, doves Columbidae and nightjars Caprimulgidae; and eutaxic in trogons Trogonidae, woodpeckers Picidae and the perching birds Passeriformes). Because the covert varies in no apparent regular manner among orders, and it varies within some families (the kingfishers, the waders Scolopacidae), the eutaxic wing gives us no clue to the relationship of cuckoos with other orders of birds (Sibley and Ahlquist 1990). Total plumage mass of Smooth-billed Anis has been measured. Males have an average plumage mass of 4.6 g, and a feathered body of 129.1 g; plumage is 3.6% of the total body weight. Females have an average plumage of 3.9 g and a feathered body of 120.0 g, and plumage is 3.3% of total body mass (Mitsch 1974).This proportion is about average among birds of their body size.

Glossy plumage The glossy- and bronze-cuckoos Chrysococcyx have a glossy plumage, with considerable variation

50 The Cuckoos among the species in color and brilliance. Two species are named for their brilliant greens, Asian Emerald Cuckoo C. maculatus and African Emerald Cuckoo C. cupreus; other cuckoos are named for a coppery or subtle bronze copper such as the Australian Horsfield’s Bronze-cuckoo. The glossy plumage of Asian Violet Cuckoo C. xanthorhynchus is violet purple in mainland Asia and the Sundas, and blue in the Philippines subspecies. Plumage gloss is more intense in the sexually dimorphic African glossy cuckoos and less intense in the sexually monochromatic Australian glossy cuckoos, where only the colors of the bare eye-ring and iris give visual cues to the sex of a bird. There may be an evolutionary conflict in the plumage appearance of these brood-parasites, between sexual selection within the species and avoiding detection by the host (Payne 1967, Andersson 1994). Structural features of feathers affect plumage color, including the wavelength λ of reflected light, the proportion of light that is reflected and the brightness of plumage, and the consistent color of reflected light viewed from different angles. The shape and curvature of the barbules, the changing dimension of barbules along the feather shaft and the orientation of barbules all combine to maintain a constant λ over a wide range of angles of incident light (Durrer and Villiger 1970, Dyck 1987). The constancy of reflected color in plumage when seen at different angles to the incident light in the emerald cuckoos C. maculatus and C. cupreus may result from the rounded shape of the reflectors and the angle of barbules, with the proximal and distal barbules nearly at right angles to each other (Dyck 1987). The structure of body feathers of African Emerald Cuckoo has been examined by light microscopy (Dorst 1951) and scanning and transmission electron microscopy (TEM) (Durrer and Villiger 1970, Durrer 1986). Feather barbules originate on each side of the barb shaft.The barbules are broad and nearly flat along their length, with haemuli (small hooks) on the medial side of the vane, and the barbules are elongated, narrow and lack haemuli on the outer side, the feather structure giving the plumage a frosted appearance. The frosted plumage of this cuckoo contrasts with the more

glossy Diederik Cuckoo and Klaas’s Cuckoo which have only the elongated, unflattened barbules (van Someren 1925, Dorst 1951), so the appearance of texture is due to the microstructure which exposes the color-producing portions of the barbules. Internally the barbule has layers of melanin granules, rod-like, arrayed side by side with the long axis parallel to the long axis of the barbule. The convex upper surface of the barbules has 10–14 of these melanin layers and the inner surface has 5–6 layers separated by keratin, the upper and lower stacks arranged around a keratin core.The number of these layers is related to plumage brightness (Rütschke 1966); glossy cuckoo feathers are more brilliant on the outer surface which reflects light to other birds than on the inner surface which faces the bird’s body, and the Scale-feathered Malkoha has equal brilliance on both sides. Multiple layers of melanin and keratin in the feather barbule are reflectors that determine the constructive interference of light waves. Some light is reflected from the first layer, some from the second, and so on.The color observed is a summation of light waves reflected off each layer. The light reflected from each layer travels a different distance to any point in the feather, and the most reflected waves pass out of phase with one another and cancel each other out. But those waves that travel an entire wavelength in the added distance traveled, are in phase with each other. These waves sum to produce the more brilliant color of a restricted wavelength. In the plumage of other green birds that have been studied in more detail, the structural determinants of reflected color include the number of melanin granule layers (the more layers, the more total light is reflected and the more brilliant the color), and the thickness and number of layers of melanin and keratin (Dyck 1987). Insofar as reflected light waves interfere with light waves reflected from other surfaces and the interference depends on their phase relationships, the wave lengths amplified in constructive interference can be calculated from a physical model of reflected light (Durrer 1986). The model predicts that the optical thickness (actual thickness⫻ refractive index) of each layer is equal to a quarter of the wavelength λ of maximum reflectance. This is the difference in

Morphology 51 light paths (2 ⫻ sine of the angle of incidence ⫻ the distance between reflecting layers) and the refractive indices of melanin and keratin. Reflected color varies with the angle of incidence and observation of light. At the larger critical angle the melanin layers absorb nearly all light, and at an angle of 0o no light would undergo constructive interference and the reflected colors would be less intense than when light is reflected at an angle, but the bigger effect is on hue. Differences between path length of waves become larger as incident light is more direct, and the hue will vary from a shorter wavelength (more blue) to a longer wavelength (more red). In this model of constructive interference, the wavelength λ amplified in an ideal regular multilayer reflector is four times the thickness of the semitransparent layers, times the refractive index. For an African Emerald Cuckoo feather, the λ observed is close to the λ value calculated from the thickness of keratin layers in the feather (Durrer and Villiger 1970). The exact color that is most strongly amplified depends on how the keratin layers are measured in relation to the rounded melanin rods (Prum et al. 1999b). In addition to this ideal model of multiple reflectors, the arrangement of keratin and melanin may affect the reflected colors through “chirping” series of layers where the peripheral reflectors are thick and the interior layers are progressively thinner. Parker et al. (1998) found these chirping multiple reflectors in the glossy gold wings of beetles. Feather barbs of African Emerald Cuckoo have thicker peripheral layers than internal layers of keratin, and TEM figures show a 20–30% shift in keratin thickness between layers (Durrer and Villiger 1970), much as in beetle wings, and this ordered structure may affect plumage color. The layers of different thicknesses may reinforce different components of the spectrum, producing complex and broad spectrum colors, metallic hues with coppery or golden highlights. In addition to the constructive interference of colors that is produced by multiple reflecting layers, coherently scattering threedimensional nanostructures occur in feathers (Prum et al. 1998, 1999a). In other birds with glossy green feathers, the colorful green Malachite Sunbird Nectarinia famosa has

more layers of melanin and keratin in the feather than does the Black Sunbird N. amethystina, while the non-glossy Gray Sunbird N. veroxi has unordered melanin granules that are not in regular layers as in the glossy species (Farquhar et al. 1996). In African starlings there is a similar difference between the dull and modestly glossy birds, but the most glossy birds have oblong melanin granules that are filled with a refractive material, especially the Emerald Starling Coccycolius iris with multiple layers on the dorsal feather surface and a single layer on the ventral surface.The color and gloss of plumage varies with the size and shape of these granules (Craig and Hartley 1985, Durrer 1986). These structural features may determine differences in plumage color among the glossy cuckoo species. The iridescent color of plumage has been said to change with wear, as in Greater Black Coucal Centropus menbeki becoming less blue and more green (Rand 1942a). The mechanism of color reflectance described for the internal layers of melanin and keratin within the feather barb make this kind of change unlikely, and the observation of variation in color probably refers to natural variation in plumage color among birds within a population.

Molt Most cuckoos molt their plumage once a year. Exceptions are coucals which molt their body plumage twice a year, and anis which have a complete molt and an incomplete molt within a year. The bird keeps a long tail through the period of molt. Tail feathers are replaced with one of the three long rectrices T1-2-3 which are retained until another long feather on each side is completely replaced. In some coucals (best known in Pheasant Coucal Centropus phasianinus in Australia) the central rectrices T1 are replaced twice and the others once a year (Higgins 1999). Smooth-billed Anis Crotophaga ani have a complete post-juvenile body and tail molt and a partial wing molt that is similar in pattern and extent to the adult molt. In March and April the young birds undergo a partial molt, but they attain the wide adult tail feathers only after the second autumn molt (Dickey and Van Rossem

52 The Cuckoos 1938). Adult anis undergo a complete postbreeding molt and a partial prebreeding body molt, with the molt slow and lasting through much of the year, and birds sometimes breed while they are still in molt (Davis 1940a, Snow and Snow 1964, Foster 1975, Pyle 1997). Molt is generally seasonal and occurs when birds are not breeding or in migration. Molt is not a reliable guide to the seasonality of breeding in cuckoos, and some cuckoos molt while they are breeding. In addition to the anis, Lesser Ground-cuckoo Morococcyx erythropygus and Dwarf Cuckoo Coccycua pumila sometimes molt while they are in breeding condition (Foster 1975, Ralph 1975), breeding adult Yellow-billed Cuckoos sometimes begin postbreeding molt while they are still caring for their nestlings (Potter 1980), and a few brood-parasitic females in Africa begin to molt while still laying (Payne 1969a). Molt in cuckoos is a prolonged affair. Migratory cuckoos such as Common Cuckoo begin and complete their post-juvenile and post-breeding molts on their wintering grounds. In northern temperate-region cuckoos that migrate, the molt is sometimes incomplete or is delayed for months. Common Cuckoos replace their plumage in about 100 days (Seel 1984b). For the Diederik Cuckoo and Klaas’s Cuckoo, the juvenile primary molt begins at 4 months and is completed in about 80 days (Hanmer 1995). In several species including the Common Koel, Diederik Cuckoo and Klaas’s Cuckoo, juvenile molt is often interrupted with birds in their first year ending the molt with their wing primaries, secondaries and coverts a mix of juvenile and adult feathers. Most remarkable in cuckoos is the molt pattern in which the flight feathers are replaced. The wing feathers are replaced not from the innermost outward, but from separate centers within the wing, beginning with an outer primary (Stresemann and Stresemann 1961, 1966, 1969, Piechocki 1971, Stephan 1991). Molt proceeds in leaps forwards or backwards across one or more adjacent primaries, in a “transilient” pattern. As with other birds, the primaries are numbered from the innermost feather outward; the terms “ascending” and “descending” trace back to an earlier time when the feathers were

numbered from the outermost feather inward. The Old World parasitic cuckoos shed the primaries in a predictable sequence. In the simple form of transilient molt, the odd-numbered feathers are replaced first from outer to inner position, then the evennumbered feathers are replaced, again starting from the outer position. Several variations occur on this theme: in Cuculus, Cacomantis and Chrysococcyx the molt sequence is P9 - 7 - 5 - 8 - 6, with P 4 - 1 2 - 3 beginning between the old primaries P5 and P3. Long-tailed Cuckoo Urodynamis taitensis has a unique molt, first the odd-numbered primaries then the even-numbered primaries, with the molt sequence P9 - 7 - 5 - 3 - 10 - 8 - 6 - 4. Certain cuckoos have other patterns, with Common Koel Eudynamys scolopacea quite different from other parasitic cuckoos with the outer primaries in transilient ascending molt (P9 - 7 - 5 - 10 - 8 - 6) and the inner primaries in stepwise descending molt (1 - 2 - 3 - 4) (Stresemann and Stresemann 1961). Another pattern is a semi-transilient ascending molt where some adjacent feathers are molted (Scythrops,Tapera). The crested cuckoos Clamator and lizard-cuckoos Coccyzus (Saurothera) are unique as their wing molt progresses in a regular pattern to skip two feathers, with a sequence P6 - 9 - pause - 7 - 10 - pause - 8 - 5 (or - 5 - 8) (Stresemann and Stresemann 1969, Piechocki 1971). In the Great Spotted Cuckoo these feathers are of different colors in juvenile and adult plumage, and first-year birds can be identified by their primaries, the new odd-numbered adult primaries being gray-brown and the evennumbered primaries the old rufous juvenile feathers. Non-parasitic cuckoos have a variable pattern of transilient ascending molt where adjacent feathers are sometimes replaced in series (Coccyzus, Ceuthmochares, Centropus) but the same pattern of alternating juvenile and adult feathers is sometimes seen in the flight feathers of first-year birds (Pyle 1997). Molt sequences sometimes vary within a species (Stresemann and Stresemann 1961) and the number of museum specimens in active molt is limited, so it is not possible at this time to use molt patterns as characters in a phylogenetic analysis for all cuckoos. This unique transilient pattern of molt nevertheless supports the monophyly of cuckoos.

Morphology 53

Skeleton Avian anatomists and paleontologists identify birds from the details of their bones (Huxley 1867, Fürbinger 1888, Pycraft 1901, 1903, Howard 1929, Berger 1960, Olson 1985, 1990, Olson and James 1991, Feduccia 1996). Certain bones have characteristic shapes that differ among avian families and orders, and skeletons may be useful in discovering the evolutionary relationships among birds of the world. It is only since the application of cladistic reasoning or phylogenetic systematics (Hennig 1966) that ornithologists have distinguished between characters that are shared because they are primitive (inherited from a remote ancestor, and shared only because other lineages from the same ancestor have evolved different characters, and not because they are shared by a strictly monophyletic lineage), or because they are derived (synapomorphic, inherited from a unique and recent common ancestor), and that skeletal characters have been used in the logic of phylogenetic systematics (e.g. Cracraft 1974, Payne and Risley 1976, Livezey 1993, 1997, 1998, Mayr 1998a,b). Despite the accumulation of a vast amount of descriptive detail on skeletal features, systematists have only recently begun to organize the accumulated data in a phylogenetic research program to determine the evolutionary relationships among birds (Livezey and Zusi 2001). In practice, bird skeletons are identified by comparison with a reference collection where bones are matched for identification to samples without a detailed description, and the published descriptions of bones are usually limited to certain groups or to a geographic region. While these resources are helpful, they often do not include all variations found within a family, and identification is sometimes based on a process of elimination (“not as above”) (e.g. Gilbert et al. 1981). Cuckoos have a desmognathous palate, no vomer, no basipterygoid process, a holorhinal nasal septum, deep temporal fossae and short mandibular processes.The postcranial skeleton has 13 or 14 cervical vertebrae.The pectoral girdle has a large coracoid and the sternum has an internal spine.The pubis has a pectineal or preacetabular process.The leg has two bony canals arranged side by side in the hypotarsus (the crest complex on the proximal end of the

tarsometatarsus), and the foot is zygodactyl, with two toes forward and the inner and outer toe directed backward, and with an accessory process on trochlea IV of the distal end of the tarsometatarsus to support the backwards orientation of the outer toe. These characters of cuckoos have been pointed out in earlier studies as reviewed by Sibley and Ahlquist (1990), and they were observed in skeletons in the present study. These skeletal features taken one at a time are not unique to cuckoos. For example, a desmognathous palate (a palate with a medial bridge across the maxillopalatinos: Huxley 1867, Pycraft 1901) also occurs in ducks, falcons, turacos, parrots and rollers and a few other avian groups, so this palate structure has evolved more than once in birds (Huxley 1867, Pycraft 1901). Other skeletal features vary among the cuckoos, such as the pectineal process of the synsacrum, a process which is prominent only in the terrestrial cuckoos (Verheyen 1956).There has been little consideration of whether the skeletal features of cuckoos are uniquely derived (synapomorphies) within the group, or are shared with other avian groups with a common ancestry, or are independently derived in the cuckoos and in other groups (homoplasies). Characters of cuckoos that may have been independently derived in bird lineages include the zygodactyl foot, a foot structure that also occurs in parrots, owls and woodpeckers. Although these birds may not form a monophyletic lineage, they are related in one phylogenetic hypothesis (Sibley and Ahlquist 1990), and if that model is true, then zygodactyly was ancestral to the original land-bird radiation and was subsequently lost three times (in Coraciiformes; in nightjars Caprimulgiformes, some swifts and hummingbirds Apodiformes; and in the lineage leading to pigeons, rails, the wading birds Ciconiiformes and the perching birds Passeriformes). The details of the zygodactyl foot in cuckoos differ from those in other zygodactyl birds and other birds that have two toes directed forwards and two backwards (trogons, mousebirds, some swifts, osprey, parrots, owls, cuckoo-roller, piciforms and the jacamars and puffbirds) (Figure 4.11). In zygodactyl birds the outer and inner toes (digits I and IV) are directed backwards; in fact the outer toe of

54 The Cuckoos

Figure 4.11. Distal end of the tarsometatarsus of cuckoos, hoatzin, turaco and other zygodactyl birds: a, cuckoo Geococcyx californianus; b, hoatzin Opisthocomus hoazin; c, turaco Tauraco hartlaubi; d, parrot Amazona amazonica; e, owl Otus asio; f, woodpecker Colaptes auratus. ii ⫽ trochlea II; iii ⫽ trochlea III; iv ⫽ trochlea IV; Se ⫽ sehnenhalter. Each pair of figures shows the end view (above) and ventral view (below), left tarsometatarsus; scale mark is 3 mm.

climbing woodpeckers is also often directed laterally (Bock and Miller 1959). The distal end of the tarsometatarsus in these zygodactyl birds with digit IV permanently reversed (cuckoos, parrots, owls, piciforms, jacamars and puffbirds) has an accessory articulating process for digit IV, a large inflected collateral flange, the sehnenhalter or sinew holder (Steinbacher 1935, Simpson and Cracraft 1981, Olson 1983, Houde and Olson 1992), and in mousebirds the simple outer flange of trochlea IV is enlarged and birds can reverse the outer toe even

though the toe is not permanently reversed (Raikow 1985).This sehnenhalter process in zygodactyl birds is enlarged and rotated posteriomedially. In cuckoos the process is less distinct than in other zygodactyl birds. It extends posteriomedially from trochlea IV but not as far distally as trochlea IV (it extends further distally in parrots), its lateral length is less than half the width of the distal shaft of the tarsometatarsus (it is about half the width in jacamars and puffbirds (not shown), and is longer in woodpeckers), and it projects on the perpendicular

Morphology 55 only slightly from the shaft of the tarsometatarsus (it projects radially a distance more than the width of the tarsometatarsus in woodpeckers and other piciforms, in which it also extends along the shaft further than its radial projection). The structure is similar for all cuckoos though it is slightly longer in the terrestrial cuckoos such as roadrunner. In hoatzin, trochlea IV bears a posteriolateral process which is not however an extra process (not a sehnenhalter), and it is similar to the expansion of the lateral ridge of trochlea IV in the trumpeter (Psophia crepitans UMMZ 152877, Psophiidae) which has an even weaker sulcus for the trochlear IV tendon; the hoatzin is not zygodactyl and the outer three toes are directed forwards.Turacos have a semi-zygodactyl foot in which the outer toe can go forwards, it is usually directed outwards, and it can rotate backwards within 70o of the hallux (Moreau 1938); this flexibility of the outer toe does not involve a sehnenhalter (Fig. 6). In the heterodactyl trogons digits I and II are sometimes directed backwards, not digits I and IV as in the zygodactyl birds; actually trogons usually hold all four toes forwards. The arrangement of foot tendons co-varies with the tarsometatarsal trochleae in zygodactyl birds (Gadow 1892, Steinbacher 1935, Olson 1983), except that muscle attachments are similar in jacamars and puffbirds and in the typical piciform birds (Swierczewski and Raikow 1981, Raikow and Cracraft 1983). The mechanical function of this process is to articulate with the basal phalanx of digit IV in the reversed position and to reverse the direction of force of the long flexor tendons of digit IV (Scharnke 1930, Steinbacher 1935). The foot of the cuckoo is also distinctive at the proximal end of the tarsometatarsus, where tendons from the leg to the toes pass through two bony canals in the hypotarsus.The two canals are arranged side by side in cuckoos. Other zygodactyl birds have a differently structured hypotarsus. In woodpeckers two bony canals are arranged radially, owls have no bony canals, and in parrots the hypotarsus varies with a single canal in Aratinga and Ara, a single canal with a partial division into two radial canals in lories Lorius garrulus and two canals side by side with incomplete division of one of them in the cockatoo

Cacatua moluccensis. In addition a few other birds including a grebe Podiceps grisegena and some herons Botaurus (Payne and Risley 1976) have two bony canals side by side, and songbirds have a hypotarsus with two canals side by side and also have other canals arranged radially to these (Figure 4.12). Although the foot skeleton of cuckoos is distinctive, the presence of two bony canals that are arranged side by side in the hypotarsus is not unique. Cuckoos have 13 or 14 cervical vertebrae and the number has figured prominently in earlier systematic arrangements of cuckoos. Fürbinger (1888) reported 13 in Crotophaga and 14 in Cuculus, Zanclostomus javanicus and Centropus; Shufeldt (1886a,b, 1901) and Berger (1952, 1960) reported 13 in Clamator, Coccyzus (including Saurothera) and Piaya, and 14 in most other cuckoos; Hughes (2000) reported 13 only in Clamator and Coccyzus. Pycraft (1903) recognized only 11 or 12 but he also recognized 2 or 3 cervico-thoracic vertebrae; Newton (1896) and Gadow and Selenka (1893) recognized 14 or 15 cervical vertebrae in cuckoos. The number of cervical vertebrae depends on the criteria used to define them. As Beddard (1899a) noted, the last three neck vertebrae bear ribs that progressively increase in size, with the anteriormost ones being very small.The ribs articulate anteriorly with the head, and a facet on the short knob on the anteriolateral surface of the vertebral centrum, and posteriorly with the tubercle and a larger process of the neural arch. Fürbinger (1888) and Baumel (1993), distinguished cervical vertebrae from thoracic vertebrae in that they do not articulate with a complete rib, where a “complete rib” has both vertebral and sternal segments, and the sternal segment connects directly or indirectly with the sternum. Vertebrae at the root of the neck which bear moveable ribs that do not reach the sternum are transitional between cervical and thoracic vertebrae. Newton (1896) and Berger (1960) considered only vertebrae with a rib that have a direct connection with the sternum to be thoracic vertebrae, yet they found different numbers of cervical vertebrae. Pycraft (1903) was more restrictive. He considered the vertebrae behind the cervicals to be cervicothoracic vertebrae, recognized by their ventral catapophyses in addition to the median hypapophysis,

56 The Cuckoos

Figure 4.12. Proximal end of the tarsometatarsus, showing the canals in the hypotarsus: a, cuckoo Geococcyx californianus; b, hoatzin Opisthocomus hoazin; c, turaco Tauraco hartlaubi; d, songbird Corvus brachyrhynchos; e, grebe Podiceps grisegena; f, woodpecker Colaptes auratus; g, owl Otus asio. c ⫽ canals; h ⫽ hypotarsus. Each pair of figures shows the end view (above) and ventral view (below), left tarsometatarsus; scale mark is 3 mm.

the second and third cervico-thoracics having ribs and the third having a rib with an uncinate process. He also noted that the fifth or posteriormost thoracic rib did not reach the sternum.Another feature that has been used to differentiate the cervical and thoracic vertebrae is the shape of the dorsal spine, which is short in the axial plane in the cervical vertebrae and has a high longitudinal crest in the dorsal vertebrae (Newton 1896), whereas Pycraft (1903) noted that the dorsal spine of the first thoracic vertebra in Coua is barely visible and is like that of the posterior cervical vertebrae. Using Fürbinger’s definition, in skeletal specimens where the articulation of ribs from the vertebrae to the sternum is intact I recognize 13 cervical vertebrae in Surniculus and Cuculus as well as in Clamator, Zanclostomus javanicus and Carpococcyx renauldi. Inasmuch as (a) some studies did not describe the criteria to recognize the cervical vertebrae, (b) the

transitional series of these vertebrae are associated with the transitional series of the ribs, (c) the attachment of the rib to the sternum is not obvious in disarticulated specimens, and (d) the number of cervical vertebrae sometimes varies among specimens of a single species both in cuckoos and in other birds (Fürbinger 1888, Berger 1952, 1960), the definitions and homologies of avian cervical vertebrae need to be re-evaluated. The sternum varies in the depth of the keel, where the flight muscles attach. The sternum also varies in the shape of the posterior edge (notched, windowed or intact).The keel is short and shallow in terrestrial ground cuckoos such as the couas and coucals (Pycraft 1903). These cuckoos can fly. The flight muscles attach not only to the keel but also to a stout coraco-clavicular membrane that extends from the coracoid to the furcula, as in Pheasant Coucal (Feduccia 1996, p. 136).

Morphology 57

Figure 4.13. Humerus of cuckoos and other birds: a, Geococcyx californianus; b, Eudynamys scolopacea; c, Opisthocomus hoazin; d, Tauraco hartlaubi; e, Trogon citreolus; f, Colaptes auratus; g, Indicator indicator; h, Corvus brachyrhynchos. aet ⫽ anterior external tuberosity; de ⫽ deltoid crest; ect ⫽ ectepicondyle; ent ⫽ entepicondyle; head ⫽ head; rt ⫽ round tuberosity; s ⫽ shaft. Anconal view, right humerus; scale mark is 5 mm.

In cuckoos the humerus is concave along the medial edge; it has a strongly developed median crest, and it lacks the distinct round tuberosity proximal to the entepicondyle that characterizes the perching birds Passeriformes. The humerus is distinct in the shape of the deltoid crest.The crest projects in a ridge with two peaks, a minor peak located proximally at the anterior external tuberosity and a larger peak located distally on the crest with a shallow concave area between the peaks (Figure 4.13). The crest is short and does not extend far along the shaft, has a simple low ridge and often a more extensive distal projection along the shaft.The distal peak is more prominent in Eudynamys and less

prominent in the ground-cuckoos Carpococcyx and Neomorphus. In other groups of birds, the deltoid crest is a straight ridge with a single peak (songbird), or it has a single peak without the extensive ridge ( parrot, pigeon), or a single rounded ridge (mousebird, not shown in Figure 4.13), while other details in the humerus also differ between cuckoos and other birds.

Skeletons and cuckoo monophyly Seibel (1988) and Hughes (2000) described several features of cuckoo skeletons that they interpreted as unique to the cuckoos, and other features that

58 The Cuckoos characterized certain lineages, including terrestrial vs arboreal cuckoos, and the brood parasites. These features involved the shape and combination of skeletal elements, and not the presence or absence of a simple element. Hughes used the more inclusive set: 14 characters. Because these skeletal characters were used to estimate the evolutionary history of cuckoos, and the conclusions of these studies differed from conclusions of studies in molecular genetics, it is of interest to compare these characters. The skeletons of cuckoos and other avian taxa were compared, in the present study, to test whether the characters reported to support monophyly of cuckoos were in fact unique to cuckoos. One or two representatives from each major cuckoo clade were selected. The other taxa were selected from specimens in orders that were within three branches of the cuckoos in the phylogenetic estimate of Sibley and Ahlquist (1990): mousebirds Coliiformes, kingfishers Coraciiformes, puffbirds Bucconidae, buttonquail Turniciformes, woodpeckers and honeyguides Piciformes, parrots Psittaciformes, turacos Musophagiformes, owls Strigiformes, doves Columbiformes, plovers Charadriiformes, falcons Falconiformes and herons Ciconiiformes (these last three all classed as Ciconiiformes in Sibley and Ahlquist 1990) and songbirds Passeriformes. The sample was selected to include at least one taxon from all third-order branches (clades within three branching points) of the cuckoos in Sibley and Ahlquist (1990), and from each order that has been suggested elsewhere as being closely related to the cuckoos.The sample also included hoatzin, which was regarded by Sibley and Ahlquist as a cuckoo.When a bone was missing or damaged, another specimen was examined. Taxa and specimens examined are listed in Table 4.3, and the characters of cuckoo monophyly in Seibel (1988) and Hughes (2000) are described in Table 4.4. The examination of skeletons does not support the claim (Hughes 1990) that all 14 characters are unique to cuckoos (Table 4.4). Depending on the character, the structures vary among the cuckoos, or they are not unique to cuckoos, or they vary both among cuckoos and among other taxa, although a few characters are unique. Cuckoos and the other

taxa differed in (8): the occurrence and position of two hypotarsal canals side by side.A few other birds also have two hypotarsal canals arranged side by side, so this characteristic is not unique to cuckoos. The sehnenhalter or accessory trochlea IV differs between cuckoos and other zygodactyl birds, but the description of character (9) does not make this distinction. The shape of the maxillary bone is distinctive, as is the form of the humerus. The other characters are variable among cuckoos or also occur in other taxa.

Skeletons and evolutionary relationships between cuckoo species Comparison of cuckoo skeletons has led to estimates of cuckoo relationships that affect the interpretation of important questions in evolution, including the origin of brood parasitism, the adaptations of arboreal and ground cuckoos, and the interchange between or isolation of the Old and New World cuckoos (Seibel 1988, Hughes 1997d, 2000). This phylogenetic inference is based on the assumption that truth is revealed in bones, and skeletal characters are true guides to the evolutionary relationships within the cuckoos.

Single origin of arboreal and terrestrial behavior? In an analysis of the postcranial skeleton of cuckoos, Seibel (1988) concluded that ground-cuckoos of the Old World were closely related to groundcuckoos of the New World, and neither were closely related to the arboreal cuckoos in their regions. The characters that linked these groundcuckoos were features of the synsacrum and tarsometatarsus, and these may have independently derived their structure in relation to the terrestrial mode of locomotion. Seibel reported three traits as shared derived characters for the recognition of a clade of coucals, couas and the Old World and New World ground-cuckoos: the relative length of the ilioischiatic synsacrum (long in these cuckoos), and the contour and orientation of the anterior edge of the posteriodorsal iliac crest (straight and more

Morphology 59 Table 4.3 Skeletal specimens examined to test statements that cuckoos are diagnostically distinct. Cuckoos Carpococcyx renauldi Centropus senegalensis Centropus toulou Clamator glandarius Coccyzus americanus Coua caerulea Crotophaga ani Cuculus canorus Dasylophus superciliosus Eudynamys scolopacea Geococcyx californianus Neomorphus geoffroyi Zanclostomus javanicus Surniculus velutinus Other taxa Amazona amazonica (parrot) Ardea herodias (heron) Ceryle alcyon (kingfisher) Charadrius vociferus (plover) Colaptes auratus (woodpecker) Colius striatus (mousebird) Columba livia (pigeon) Corvus brachyrhynchos (songbird) Falco sparverius (falcon) Indicator indicator (honeyguide) Notharchus macrorhynchos (puffbird) Opisthocomus hoazin (hoatzin) Otus asio (owl) Tauraco hartlaubi (turaco) Trogon citreolus (trogon) Turnix suscitator (buttonquail)

laterally directed in these cuckoos). This first character is problematic, as most cuckoos had an intermediate form (Seibel 1988). Second, the couas, Old World ground-cuckoos and New World groundcuckoos were uniquely linked only by the inflection of the acrocoracoidial furcular facet of the coracoid, although the definition of character states of the facet is problematic (Seibel 1988). Third,

specimen number UMMZ 219851, 236029 UMMZ 207709 UMMZ 208402 UMMZ 212907 UMMZ 152712 UMMZ 209201, 208404 UMMZ 218940 UMMZ 151115 UMMZ 228022 UMMZ 210347 UMMZ 71909 UMMZ 200592 UMMZ 236439, 236440 UMMZ 233061 UMMZ 203942, 203944 UMMZ 107419, 224551 UMMZ 206751 UMMZ 74184, 198342, 220113 UMMZ 235110, 207776 UMMZ 234219 UMMZ 224018 UMMZ 232727 UMMZ 223964 UMMZ 212931 UMMZ 219195 LSU 62741 UMMZ 224665 UMMZ 234273 UMMZ 209256 UMMZ 205093

within this last set of cuckoos, the Old World ground-cuckoos and New World ground-cuckoos including roadrunners were linked by two tarsometatarsal traits: the size and position of a tubercle, and the position of the distal edge of the proximal foramen. Insofar as the characters are in the synsacrum and legs (or are problematic), and these birds share a common pattern of locomotion,

60 The Cuckoos Table 4.4 Skeletal characters in a statement of cuckoo monophyly (after Hughes 2000). 1. “Os quadratum. Condylus medialis, caudalis, and lateralis prominent, well-rounded, and separated by a broad notch; condylus caudalis tapered to a rounded point, deflected and extending well beyond caudal margins of quadrate in lateral aspect; processus orbitalis quadrati truncated and somewhat anvil-shaped.” (Character 21 of Hughes 2000). Comments:The description is appropriate for cuckoos, although the shape of the condyles vary considerably in detail, particularly the shape of the orbital process, and the condylus medialis is not demarcate by a distinct notch in relation to the other quadrate condyles. In the other taxa, the same description appears to match the bones in part, if not in entirety: (1a) the condyles are prominent and well rounded in the parrot, kingfisher, trogon, heron and songbird; (1b) the broad notch is present in the buttonquail, trogon, kingfisher, mousebird, hoatzin, owl, turaco and songbird, and the heron has a prominent but narrow notch; (1c) the condylus caudalis is tapered to a rounded point and deflected in the kingfisher, plover, songbird, owl and turaco; (1d) the condylus caudalis extends as in cuckoos, in the honeyguide, kingfisher, trogon, turaco, owl, plover and songbird; (1e) the processus orbitalis quadrati is truncated in the buttonquail, hoatzin, turaco, parrot, plover, falcon and songbird; and (1f ) the processus orbitalis is anvil- or cone-shaped in the hoatzin, parrot, plover, dove and falcon. Because these characters are variable within the cuckoo species, and none are unique to cuckoos, the form of the quadrate is not diagnostic in cuckoos, i.e. it does not clearly distinguish them. 2. “Os pterygoideum. Mesially deflecting spur-like processus dorsalis located caudally on facies dorsalis near, but not articulating with, os quadratum.” (Character 44 of Hughes 2000). Comments: Not all cuckoos have such a process, and in Centropus, Clamator and Eudynamys the process articulates with the quadrate. 3. “Os ectethmoidale. Large, quadrilateral in form, marginis lateralis and ventralis slightly to moderately excised.” (Character 15 of Hughes 2000). Comments:The character has three components: in cuckoos (3a) all have a quadrilateral ectethmoid; (3b) a lateral excision is apparent only in Clamator, Coccyzus, Eudynamys, Geococcyx and Neomorphus; and (3c) a ventral excision is apparent in all but Carpococcyx, Centropus, Coua and Dasylophus. In other taxa, seven have a quadrilateral ectethmoid, five have a lateral excision, and six have a ventral excision; the hoatzin is deeply excised (not slightly to moderately excised), the puffbird is similar to Centropus, although the ectethmoid is extended further laterally, the mousebird is nearly identical to Coccyzus, and the buttonquail is nearly identical to Coua. 4. “Os palatinum. Marginis medialis meeting in midline below, fully or nearly concealing, rostrum parasphenoidale; lamella caudolateralis being widest midway between processus pterygoideus and margo distalis of pars choanalis, tapering towards rostrum; processus interpalatinus extending nearly to margo caudalis of processus maxillopalatinus; in lateral aspect, cristae lateralis are deflected ventrally.” (Character 47 of Hughes 2000). Comments:The description defines the desmognathous palate and elaborates on the shape of the palatines. In cuckoos the description is appropriate, except that (4b) the parasphenoid is exposed throughout its length in Carpococcyx; and (4d) the palatines do not taper anteriorly in Clamator.The description excludes the other taxa, as most of these including hoatzin are schizognathous, although in all (4d) the palatines taper to the rostrum and (4f) the lateral crests are deflected ventrally in all but kingfisher. 5. “Os maxillare. In ventral aspect, processus palatinus bilobed, extending caudally nearly to marginis distalis of crista ventralis; margo medialis of processus palatinus not fused caudally to margo distalis of os maxillare.” (Character 48 of Hughes 2000). Comments: The description applies to cuckoos. In the other taxa, it also applies to plover, songbird and buttonquail, although the shape of the maxilla differs in other aspects from that of cuckoos. 6. “Os articulare, crista intercotylaris. Prominent, half-disk shaped with radius aligned approximately 45° to planum medianum.” (Character 54 of Hughes 2000). Comments:The description of the base of the lower mandible applies to cuckoos. In other taxa, the description does not universally apply as the articular surface is more complex (more robust in parrot) and the radius is closer to the median plane in turaco, owl, falcon and buttonquail, but the description applies well to the hoatzin, puffbird and trogon which are nearly identical to cuckoos. 7. “Tibiotarsus, incisura intercondylaris. Deep, rounded pit extending caudally at least half the length of extremitas distalis; trochlea cartilaginis tibialis poorly developed.” (Character 57 of Hughes 2000).

Morphology 61 Table 4.4 contd. Comments: This character has two components: (7a) the extent of the pit; and (7b) the development of the trochlea. The description of the distal end of the tibiotarsus applies to cuckoos. In the other taxa both 7a and 7b apply to the mousebird, turaco and owl. 8. “Tarsometatarsus, hypotarsi. Two oblong canales hypotarsi, completely enclosed in bone, positioned side by side in planum medianum both almost entirely caudal to cotylae medialis and lateralis, and corpus tarsometatarsi.” (Character 58 of Hughes 2000). Comments:This character is present in the cuckoos, while in the other taxa examined it is absent. Some have a single canal, e.g. hoatzin. However certain other birds have two enclosed hypotarsal canals, either side by side (some grebes) or in another arrangement (not side by side: woodpeckers, some herons, Payne and Risley 1976), or have two canals in the same position as the cuckoos, but have additional canals peripheral to those in a radially expanded hypotarsus (songbird) (Fig. 5). 9. “Tarsometatarsus, trochlea metatarsi quarti. Margo distalis proximal to incisura intertrochlearis medialis; caudal trochlea accessoria prominent and strongly inflected medially.” (Character 75 of Hughes 2000). Comments:This character has two components: (9a) the shape of the distal margin; and (9b) the sehnenhalter (accessory caudal trochlea). All cuckoos have the character, while in the other taxa, (9b) the sehnenhalter is inflected inwards in the owl, woodpecker and honeyguide, as in cuckoos (in owl the outer trochlea IV flange forms the base of the structure, and in woodpecker and honeyguide the inner trochlea IV flange forms this structure as in cuckoos; in parrots trochlea IV is shorter—it does not extend very far distally along the shaft—than the sehnenhalter). In hoatzin the outer trochlear flange is prominent, but less so than in cuckoos (in hoatzin, trochlea IV extends as far distally as trochlea II, the flange is about 60% as long as the metatarsal articular surface of trochlea IV, and the flange not inflected inwards (Fig. 4). 10. “Coradoiceum, facies articularis sternalis. Crista ventralis prominent, centred, rounded, and facing sternally.” (Character 99 of Hughes 2000). Comments: Gilbert et al. (1981) describe this condition as “ventral sternal facet rounded, centered and facing sternally”. Although the figures in Gilbert et al. (1981) and Hughes (2000) do not show the feature well, the coracoid has a ridge or crest on the ventral surface, perpendicular to the shaft and centered with respect to the shaft, with the curved ridge parallel to that of the margin of the sternal articulatory facet. All cuckoos have a coracoid as in this description, while in the other taxa, trogon and plover have the crest as in cuckoos, the woodpecker is similar (though the crest is more distal to the coracoid shaft), and the pigeon has a prominent ventral crest not centered but expanded on the procoracoideal side of the coracoid. The ventral crest is also present in certain other birds including herons and cormorants (Phalacrocorax auritus) although it is not always centered on the shaft in these birds (Payne and Risley 1976). It is also known in other birds including early fossil charadriiform, trogon and other Eocene fossils (Mayr 1998a, 1999, 2000). In hoatzin the coracoid is fused with the sternum and no articular surface is visible. Seibel (1988) recognized fusion of the bones and assumed that hoatzin lacked a ventral crest (his character state CO19a); Hughes (2000) did not describe hoatzin though she coded the crest as absent in hoatzin. 11. “Scapula. Facies articularis humeralis directed dorsally; facies articularis clavicularis prominent, knob-like, and directed dorsally; extremitas cranialis scapulae, delineated by acromion and tuberculum coracoideum, truncated and flattened cranially.” (Character 107 of Hughes 2000). Comments:The description fits the cuckoos, and also the other taxa, except that (11a) the facies articularis humeralis is directed ventrally in the kingfisher; (11b) the facies articularis clavicularis is not prominent in the songbird, and (11c) the facies articularis clavicularis is not directed dorsally in the plover, while in the woodpecker the caudal end of the scapula is greatly enlarged, flattened and recurved to meet the rib cage. 12. “Humerus, condylus ventralis. Rounded, not flattened or oblong, in axis proximodistalis and both plana transversialia and dorsalia.” (Character 114 of Hughes 2000). Comments: In cuckoos the description matches the specimens, while in the other taxa it matches for all except woodpecker, honeyguide and trogon. 13. “Synsacrum, extremitas caudalis synsacri. Crista dorsalis reduced and fused to form a single low ridge; marginis lateralis of processus transversus widened in axis rostrocaudalis, facies lateralis projecting cranially.” (Character 130 of Hughes 2000).

62 The Cuckoos Table 4.4 contd. Comments: In cuckoos, (13a) the crista is not prominent in Crotophaga; and (13d) there is no wide lateral margin of the transverse process in Clamator and the lateral margin does not project cranially in Neomorphus, while in the other taxa, (13a) the crista dorsalis is reduced in the kingfisher, songbird, owl, turaco and buttonquail; (13b) the crista dorsalis is unfused in all but the plover; (13c) a single low ridge is present in all but the songbird and hoatzin; (13d) there is a wide lateral margin of the transverse process in the mousebird and honeyguide; and (13e) the lateral face projects cranially in the plover and turaco. 14. “Vertebrae caudales, primus. Processes transversus tapered, not widened, paddle-shaped in ventral aspect; projecting strongly caudally.” (Character 131 of Hughes 2000). Comments: The description is appropriate for most cuckoos, although (14a) the transverse process is not tapered in Carpococcyx, Coua, Dasylophus or Eudynamys; (14b) it is not paddle-shaped in Coua; and (14c) it does not project strongly caudally in Carpococcyx, Coua or Eudynamys, while in the other taxa, (14a) the process is tapered in the kingfisher, trogon, songbird, falcon and turaco; (14b) it is paddle shaped in the woodpecker, mousebird and hoatzin; and (14c) it projects strongly caudally in the parrot, mousebird and owl.

the skeletal characters are likely to be adaptations for running on the ground, and evolved independently in these cuckoos (Payne 1997b). Only one character was found to link the arboreal cuckoos: the posterior extent of the medial angle of the posterior dorsal iliac crest, relative to the width of the vertebral centra (the main body of the vertebra).The cuckoos that share the character are the New World anis, New World broodparasitic cuckoos, New World coccyzine cuckoos and the Old World malkohas and brood-parasitic cuckoos. Seibel’s conclusion that a synsacral character is a derived synapomorphy may follow from the assumption of turaco as the outgroup (Payne 1997b). These long-legged birds have synsacra that are used in bounding locomotion, so it is questionable whether this trait is a good standard by which to estimate the evolutionary polarity of characters in cuckoos. The shape of the iliac crest in these cuckoos may be a primitive rather than a derived character. Also, it is uncertain that turacos are closely related to cuckoos.With only a single skeletal character to link these cuckoos with each other, there is no corroboration of the hypothesis that arboreal cuckoos form a single lineage.

Single origin of brood parasitism? Skeletal characters were used by Seibel (1988) and Hughes (2000) to infer a single origin of broodparasitic cuckoos.These characters were re-examined

in the present study, using some of the same specimens and others that were not available to the earlier studies. Seibel (1988) listed four characters that defined the brood-parasitic cuckoos of the Old World and New World as a single derived group, together with the New World Coccyzus cuckoos, and two additional characters that defined this group except for the parasitic genera Urodynamis and Eudynamys. Hughes (2000) used a larger set of 135 skeletal characters, including 56 cranial as well as 79 postcranial characters, and she modified several characters of Seibel (1988). In addition some taxa were different (Dasylophus cumingi in Hughes 2000, Zanclostomus javanicus in Seibel 1988). The phylogenetic estimates of the two studies were similar, with a few exceptions. In Hughes’ estimate, (1) Carpococcyx was basal to all cuckoos, (2) Coua and Centropus formed the next basal clade, and (3) the New World Morococcyx, Geococcyx and Neomorphus formed the next basal level (rather than a sister clade to Old World Carpococcyx).The differences in (1), (2) and (3) may result from rooting the outgroup next to Carpococcyx rather than between the ground cuckoos and arboreal cuckoos. Other differences in estimates were that in Hughes (2000), (4) the mostly Old World Phaenicophaeinae and Cuculinae formed separate clades, and (5) New World brood-parasitic Tapera and Dromococcyx were basal to the Old World brood-parasites with both in Cuculinae. Hughes’ estimate was similar to Seibel’s in (1) the association of New World Coccyzus with

Morphology 63 the Old World brood-parasitic cuckoos rather than with New World nesting Piaya, and (Coccyzus sub-genera) Saurothera and Hyetornis, and (2) the association of Morococcyx, Geococcyx and Neomorphus in the New World ground cuckoos. Hughes (2000) included a data matrix, and this matrix and a character analysis were presented elsewhere as an appendix (Hughes 1997d ). Both Seibel (1988) and Hughes (2000) concluded that brood parasitism evolved only once in the cuckoos. Hughes’ presumptive clade of “Cuculinae” (Old World and New World brood parasites together with the genus Coccyzus) was distinguished in Hughes (1997d, 2000) by nine skeletal characters that changed state from node #63 to node #64, and defined her “Cuculinae” parasitic cuckoos. To evaluate this estimation of a single origin of brood-parasitic cuckoos, I examined skeletons of cuckoos for the six characters that were said to define the brood-parasitic clade in Seibel (1988), and the nine characters said to define the clade in Hughes (1997d, 2000), a total of 12 characters, with some overlap between the two studies. The terminology and examples refer to the illustrations and descriptions in standard anatomical works (Howard 1929, Baumel 1993).

Cuckoo skeletons were examined to test their conformity with coding of the characters that Seibel and Hughes reported, to support their conclusion of a single origin of brood-parasitic cuckoos. Specimens used for the examination were: Clamator glandarius 212909, Cuculus canorus 2020512, 119430, Eudynamys scolopacea UMMZ 210347, Urodynamis taitensis USNM 559589, Tapera naevia UMMZ 214005, Dromococcyx pavoninus UMMZ 209207, Crotophaga major UMMZ 219553, Geococcyx californianus 227057, Morococcyx erythropygus UMMZ 133738, Ceuthmochares australis UMMZ 158186, Zanclostomus javanicus UMMZ 209478, Dasylophus superciliosus UMMZ 228024, Coccyzus americanus UMMZ 154465, Saurothera merlini UMMZ 158527, Coccycua minuta UMMZ 139991, 139990, Piaya cayana UMMZ 200591, Piaya melanogaster UMMZ 209478, Coua caerulea 209201, Coua cristata UMMZ 157526, and Centropus senegalensis UMMZ 205896. Where skeletal elements were missing, a second specimen was examined. In most cases there was no consistent or discrete difference in the skeletal trait for nesting cuckoos and brood-parasitic cuckoos, as in Table 4.5. Where the CI is less than 1.0, the character transition occurred more than once in

Table 4.5 Skeletal characters in a statement of monophyly of brood parasitism (“Cu”) in cuckoos (Hughes 1997d (node 66), 2000). Characters are numbered as in Hughes (1997d ), where the character transitions are described. Character number, transition, and description #24 (1–2). “Os quadratum, angle of proc. oticus and proc. orbitalis in lateral aspect (ordered): 0 large, 1 moderate, 2 small. CI ⫽ 0.67.” Comments: No difference in size of either orbital or otic process of quadrate between “Cu” and other cuckoos is apparent: proc. oticus and proc. orbitus in parasitic cuckoos Cuculus, Clamator and Urodynamis were as large as in nesting cuckoos of similar body size or larger (Ceuthmochares, Piaya melanogaster, Morococcyx, Centropus). Quadrate processes were larger in brood-parasitic Tapera than in the larger Morococcyx. There was a trend for larger cuckoos to have larger processes than smaller cuckoos, although this did not consistently distinguish groups, and quadrate processes varied in a gradual transitional series and not in a step function. #37 (0–1). “Os exoccipitale, shape (ordered): 0, flat or slightly raised; 1, produced into bulbous expansion raised above contour of os occipitale; 2, as in 1 but greatly expanded. CI ⫽ 1.00.” Comments:The bulbous projection from the skull base was not very large (2) in any cuckoos, and the birds in which it was relatively large included both parasitic cuckoos (Tapera, Dromococcyx, Clamator, Cuculus, Eudynamys, Urodynamis) and nesting cuckoos (Geococcyx, Zanclostomus, Dasylophus, Morococcyx, Coccyzus, “Saurothera” and both Coua). #80 (0–1). “Pelvis, relative widths of facies dorsalis (ordered): 0, narrowest point less than half the width between processes antitrochantericus; 1, approximately half; 2, more than half. CI ⫽ 0.67.” Comments: Seibel listed this as a synapomorphy for Cuculinae, less Eudynamys and Urodynamis. Hughes (1997) listed 0–1 as a synapomorphy for “Cu”, 1–2 as a synapomorphy for most “Cu” and 1–0 as a reversal from other “Cu” to Eudynamys and Urodynamis. I find all forms to be “0”, except “1” for brood-parasitic Tapera and Dromococcyx, and a nesting cuckoo Coccycua minuta.

64 The Cuckoos Table 4.5 contd. #87 (0–1). “Ilium, cross-section of margo medialis (Fig. 15) of crista iliaca dorsolateralis in axis dorsoventralis: 0, broad, forming smooth arc; 1, compressed into thin, blade-like shelf. CI ⫽ 1.00.” Comments:This character ⫽ PE10 in Seibel 1988, I find that all cuckoos are “0” except for the brood-parasitic Tapera and Clamator and the nesting Crotophaga which are “1”. #92 (0–2). (p) “Ilium, relative length of sync. ilioischiadica from crista caudalis fossa renalis to proc. marginis caudalis in axis rostrocaudalis, with reference to depth of recessus caudalis fossae: 0, intermediate between state 1 and 2; 1, sync. ilioischiadica long, recessus caudalis fossae present and deep; 2, sync. ilioischiadica short, recessus caudalis fossae shallow. CI ⫽ 0.67.” Comments: Seibel (1988) defined the character in terms of vertebral centrum width; Hughes defined it in terms of depth of the fossa renalis. Seibel found three transitions, none corresponding to those of Hughes. Seibel (1988: 64) described the character as problematic and nearly impossible to code within the Phaenicophaeinae, that is, the group related to Cu. In general aspect, birds with a longer pelvis have a longer ilioishiadic fusion H93. The character varies with the life style of cuckoos, a shorter ilioishiadic fusion in the arboreal cuculines; the structure is nearly identical in the parasitic Tapera and the nesting Morococcyx, while it is shorter in the small, terrestrial parasitic Dromococcyx. I find character 92 to have state 2 in the brood-parasitic Dromococcyx and in the nesting cuckoos Coccyzus, Coccycua minuta and Piaya melanogaster. #94 (1–2). “Sternum, cranial extent of apex carinae (ordered): 0, cranial tip of apex caudal to labrum externum of sulcus articularis coracoideus; 1, apex at same cranio-caudal level as sulcus; 2, apex cranial to sulcus. CI ⫽ 0.50.” Comments: I find no difference among cuckoos. #111 (0–1). “Humerus, extremitas proximalis humeri,“bulbous convexity” on facies cranialis just lateral to midpoint: 0, absent, 1, present. CI ⫽ 1.00.” Comments: I find no difference among cuckoos. #112 (0–2). “Humerus, position of fossa m. brachialis in cranial aspect: 0, mesial to center of corpus humeri; 1, far mesial; 2, at or lateral to meso-lateral center of corpus humeri. CI ⫽ 0.67.” Comments: Seibel found the transition from primitive cuckoos to occur in two places, one leading to the New World terrestrial cuckoos Geococcyx, Morococcyx and Neomorphus, the other to Coua. Seibel’s transitions and Hughes’s transitions occurred at different points on their phylogeny; Seibel’s transitions did not occur to define a clade “Cu”. Examination of UMMZ specimens supports Seibel’s transitions, except that the form of the fossa varies within Coua. I find all cuckoos to be similar except for Crotophaga, Geococcyx, and Coua caerulea, all with a lateral position. #124 (1–2). “Os carpi ulnare, shape of distal end: 0, blunt, expanded, or bulbous; 1, pointed, margo distalis 45⫿ to crus longum; 2, strongly pointed, approximately 30°. CI ⫽ 0.67.” Comments: Seibel defined three characters of the ulnare—his “cuneiform”—that appeared to be synapomorphies of all cuckoos; Hughes used these to define cuckoos and also recognized a new character, the shape of the distal end, a modification of Seibel’s character CU3. In parasitic cuckoos the ulnare has a more slender shape, but not in all; it is blunt in Eudynamys and especially in Tapera, whereas in nesting cuckoos it is narrow and pointed in Saurothera and somewhat narrow in Ceuthmochares and Dasylophus. The form seems to vary with behavior, blunt in terrestrial forms, and narrow and pointed in the more arboreal birds and the intercontinental migrants. In addition, Seibel (1988) reported the following characters to distinguish the clade Cuculinae (except Urodynamis, which fell outside the clade): S#57, PE5: (pectineal process, ⫽ tuberculum preacetabulare of Hughes 2000). I find no discrete variation in this process between the parasitic and non-parasitic cuckoos, although it is large in the terrestrial forms, as noted by Verheyen (1956). S#55 , UL1:0–2, ulna distal shape: Seibel had 6 states for the ulna; I find no distinct differences in ulna shape. S#52 , PE17, iliac crest shape and its anterior position relative to antitrochanter) Seibel (p. 63) found PE17 problematic and nearly impossible to code in Coccyzus, and PE17 individualistic and nearly impossible to code in Rhinortha. In Seibel (1988), PE17 character state B was coded “not as in A” with no independent description. I find the antitrochanter indistinct in cuckoos, whereas it is distinct in the goose and turkey (Howard 1929, Baumel 1993).

Morphology 65 the phylogenetic hypothesis of Hughes (2000), and the character is not unique to brood-parasitic cuckoos. The models of Seibel (1988) and Hughes (2000) of phylogeny among the cuckoo species and a single origin of brood parasitism depended on certain characters. Deconstruction of these characters reveals unquestioned assumptions and inconsistencies and leaves no reason to think that broodparasitic behavior evolved only once in the cuckoos. Also, several characters that supported their conclusion of a single origin of brood parasitism changed more than once in the skeletal phylogeny of cuckoos. The characters with a low consistency index are 24, 80, 95, and 124 (Hughes 2000). In addition, certain of these reported character states are not discrete, but rather they intergrade between extremes: these intergrading characters are 24, 37, 87, 93, 95, 110, 112, and 124 (Hughes 2000). Posso (2003) found the skulls of cuckoos to differ from the descriptions of Hughes, and his phylogenetic estimate of cuckoo relationships also differed from that of Hughes. There is no confidence in conclusions that are based on unrepeatable characters. The conclusion that brood parasitism evolved once rather than twice in the Old World cuckoos involves several of these same characters: the first skeletal estimate, based on postcranial characters (Seibel (1988). The same conclusion in Hughes (1997, 2000) involved 12 characters that changed at the node of separation of the malkohas from the Old World brood parasites. Several of these characters changed more than once in the skeletal phylogeny of cuckoos (following the numbers in Hughes (2000), these were characters 5, 33, 89, 113, 115) and had a low consistency index (CI). The other characters (4, 31, 80, 100, 108) were either not discrete, or had multiple states with questionable linear polarity, or had both these features (e.g. 66), or had a low CI and were not discrete (e.g., 24). In brief, this node is not well supported. Inasmuch as the authors did not test the relationships of cuckoos with other birds, the conclusion that Asian ground-cuckoos Carpococcyx are the basal cuckoos is a consequence of the assumption that character states in turacos are the prim-

itive character states in cuckoos. Because the assumption was not tested, and the relationship of cuckoos and turacos is not apparent in comprehensive systematic surveys of birds (Sibley and Ahlquist 1990, Livezey and Zusi 2001), the conclusions from the morphological estimates are open to question. In particular, the skeletal characters used in earlier studies do not support the conclusion of a single origin of brood parasitism in the cuckoos.

Other morphological variations among cuckoos Other morphological features that vary among cuckoos are the form of the syrinx, the arrangement of feather tracts, and the presence and absence of certain leg muscles (coded into alphabetic formulae that note the presence and or absence of these muscles). These characters are not closely associated with each other in cuckoos (Berger 1960). Anatomical studies of cuckoos should be continued, not only to add to the skeletal descriptions of Pycraft (1903), Seibel (1988) and Hughes (1997b, 2000), but also to confirm the characters and to add taxa to previous descriptions. Table 4.6 summarizes the earlier comparative morphological studies of cuckoos. Some characters in Berger (1960) were from earlier descriptions (e.g. Beddard 1885, 1898b), and some characters differed among these earlier descriptions. Characters in Table 4.6 are mainly from Berger’s Tables 4.2–4.4, with information on additional taxa from his text. The cuckoo syrinx has been described by Beddard (1895, 1898a,b, 1901, 1902) and Berger (1960). A bronchial syrinx has enlarged bronchi; a tracheobronchial syrinx has both enlarged bronchi and lower trachea, and the intrinsic muscles are attached to a bronchial semi-ring. Cuckoos with a tracheobronchial syrinx include Phaenicophaeus curvirostris (“erythrognathus”), Ceuthmochares, Scythrops, Eudynamys, Tapera, Clamator, Coccyzus, Saurothera, Piaya, Chrysococcyx, Cuculus, Hierococcyx Surniculus, Cacomantis and Pachycoccyx. Cuckoos with a bronchial syrinx Carpococcyx, Coua, Crotophaga and Guira; although Beddard (1885) indicated a tracheobronchial syrinx

66 The Cuckoos Table 4.6 Anatomical characters of cuckoos (after Nitzsch 1867, Beddard 1885, 1898a,b, 1899, Pycraft 1903, and Berger 1960). Character 1 2 3 4 5 6 7 8 9 Clamator 1 0 1 0b 1c 0 1 1 0 Cuculus 1 1 0a 0b 1c 0 0 1 0 Pachycoccyx 1 0d 0 “Surniculus” 1 0 0 0b 1c 0 1 1 0 a b Scythrops 1 1 0 0 1c 1 0 1 1 Eudynamys 1 1 0a 0b 1c 1 0 1 1 a b c Chrysococcyx 1 0 0 0 1 0 0 1 1 Ceuthmochares 1 0 0a 0b 1c 1 1 1 0 Phaenicophaeus 1 0 0a 0b 1c 1 1 1 1 Coccyzus 1 0 1 0b 0c 0 1 1 0 Saurothera 1 0 1 0b 0c 0 1 1 0 b c Piaya 1 0 1 0 0 0 1 1 0 Centropus 0 1 0 1 1 1 0 0 1 Carpococcyx 0 1 0 0 0 1 0d 0 1 Coua 0 0 0 0 0 1 1 0 1 Crotophaga 0 1 0 0 1 1 1 1 1 Guira 0 1 0 0 1 1 1 0 1 Tapera 1 0 0 0b 1 0 1 1 0 Dromococcyx 0 0 0 0 1 1 1d 0 1 Morococcyx 0 0 0 0 0 1 1 0 1 Geococcyx 0 0 0 0 0 1 1 0 1 1. syrinx: 0, bronchial; 1, tracheobronchial. 2. sternum: 0, double-notched; 1, single-notched. 3. cervical vertebrae: 0, 14; 1, 13 (aHughes found 13 only in Clamator and Coccyzus). 4. M. flexor hallicus longus inserts on hallux: 0, yes; 1, no (bBeddard 1898 reported it absent in Centropus, but Berger found it present, with a different attachment than in other cuckoos) (there was no specific description of the state in Tapera or other cuckoos, but it was inferred absent only in Centropus). 5. M. flexor metacarpi brevis: 0, absent; 1, present (csee Berger p. 77, also Berger’s Table 3). 6. M. piriformis ⫽ M. caudofemoralis, pars iliofemoralis “B”: 0, absent; 1, present. 7. M. iliacus “E”: 0, present; 1, absent (dPachycoccyx: absent in Berger’s Table 3, not mentioned p. 73; Dromococcyx: ? p. 73; (Scythrops: present in Berger’s Table 3; not listed p. 77). 8. dorsal apterium between cervical and interscapular tracts: 0, absent; 1, present. 9. ventral abdominal tracts: 0, single; 1, double.

in Guira.A few cuckoos have an intermediate syrinx, that is, a bronchial syrinx with some tracheal development: Phaenicophaeus curvirostris, Geococcyx, Morococcyx and Dromococcyx, and Centropus are reported to have either an intermediate or a bronchial syrinx (Beddard 1898b, 1902, Berger 1960). In fresh specimens of Carpococcyx renauldi (UMMZ 219043, 236029), a few details of feather tracts and leg muscles differ from those in

the descriptions of C. radiatus by Beddard (1901) and Berger (1960). I found that a dorsal median apterium was not present between the dorsal cervical tract and the spinal tract (there were scattered feathers as dense as in the interfemoral region of the spinal tract in Berger’s Fig. 3); there were two small apteria on the pelvic region of the spinal tract; and leg muscle “E” (M. iliacus) was present (Berger did not find it in a preserved specimen). In an alcoholic

Morphology 67 specimen of Rhinortha chlorophaea (USNM 223470) I found a single ventral feather tract on each side, like the single tract in Ceuthmochares aereus as described by Berger (1960), and in Zanclostomus javanicus (UMMZ 236439), Dasylophus superciliosus (UMMZ 228023, 2280244) and Piaya cayana (Beddard 1885), and not a partially double ventral feather tract as in Scythrops and Centropus as reported by Nitzsch (1867) or Phaenicophaeus curvirostris as reported by Beddard (1885). Studies of the bony palate, postcranial skeleton, leg muscles and the geographic distribution of the major groups of cuckoos show little agree-

ment between sets of data (Berger 1960, Seibel 1988, Hughes 1997d, 2000), and re-analysis revealed certain inconsistencies and questions in coding of the skeletal characters. Another estimate of cuckoo phylogeny (Hughes 1996a) used the occurrence of brood parasitism, egg color, behavioral and ecological traits. It is clear that there is no consensus about the relationships among the cuckoos from a comparison of these characters. It is unclear whether additional morphological and behavioral characters can provide a set of data that allows the same resolution as a good molecular data set.

5 A molecular genetic analysis of cuckoo phylogeny Perspective Historically, taxonomists organized the diversity of life into a system of names (or taxonomy) that reflected the general similarities and differences among different kinds of organisms.As information became available, the many species of cuckoos were recognized as belonging to a single avian family (or order) on the basis of the similarities in their behavior and morphology. Early taxonomists used a qualitative approach, in which species with similar appearance were grouped together. Systematists now agree that taxonomy should explicitly reflect the evolutionary relationships among organisms. Grouping species together in the family Cuculidae, for example, implies that they are descended from a single common ancestor that is more recent in time than the common ancestor of cuckoos and other more distantly related birds. Cuculidae also includes every species that is descended from their most recent common ancestor. In other words, each taxonomic group (e.g., order, family, subfamily, genus) is a natural or monophyletic group, and the hierarchical structure of these groups is consistent with an evolutionary tree or cladogram that depicts their historical relationships. Systematists use various kinds of information and analytical approaches to construct evolutionary trees. Overall similarity, quantified using either morphological or genetic data, can be used to establish relationships under the assumption that similarity between taxa is related to the time since common ancestry. This assumption, however, does not always hold, as evolutionary change may proceed at different rates in different lineages.

Although phenetic methods once were widely used and may be the only option for certain kinds of data (e.g., DNA-DNA “hybridization” or thermal dissociation), most systematists recognize that shared derived characters (novel morphological, behavioral, or genetic characteristics that are inherited from a common ancestor) provide the best evidence of common ancestry and in turn provide the basis for discovering the hierarchical relationships among organisms. Building a branching tree of life would be a trivial process if each character examined evolved only once and was retained in all the descendants of the ancestor in which the character first appeared, and if we could find at least one character to diagnose each group and subgroup in our tree. Similar features, however, may evolve independently in distantly related organisms and a derived character state may be lost in some descendants of the common ancestor in which it first appeared. In both situations, the organisms that share a particular character state may not represent a natural (monophyletic) group. Homoplasy, the independent origin or loss of a character state in unrelated organisms, obscures evolutionary relationships and is the essential problem that systematists confront in building trees. Systematists now use quantitative methods to select the tree or, equivalently, the hypothesis of evolutionary relationships that provides the best explanation for all the available character data. In a phylogenetic parsimony analysis, for example, the tree that requires the smallest number of evolutionary transformations to explain the distribution of character states in the terminal taxa is the preferred hypothesis of evolutionary relationships, because it

A molecular genetic analysis 69 minimizes the number of ad hoc hypotheses of homoplasy needed to account for the data. For example, if three different derived character states are shared by species A and B, and only one is shared by species A and C, we prefer the hypothesis that A is more closely related to B rather than to C.The former hypothesis requires five evolutionary transformations or steps (including one instance of homoplasy), whereas the latter requires seven steps (including three instances of homoplasy). Phylogenetic analyses can use any kind of discrete character data, including information from morphology, behavior, and molecular genetics. For a variety of reasons, systematists are relying to an ever greater degree on molecular genetic information and particularly DNA sequence data. DNA sequences offer a number of advantages, perhaps most importantly the availability of a very large number of potentially informative characters that are comparable across a broad diversity of taxa. As illustrated in the preceding section, morphological characters may be subject to a degree of ambiguity and interpretation associated with establishing the homology of characters across taxa and the scoring of characters, that may not fall neatly into discrete states. Perhaps more problematic is the potential for a suite of morphological characters associated with a particular lifestyle (e.g., arboreal versus ground-dwelling) to evolve in a correlated fashion in independent lineages, resulting in correlated homoplasy that can mislead a phylogenetic analysis. Molecular data is not immune from such problems, but it offers much less ambiguity in the initial scoring of characters (the nucleotide at any particular position in a DNA sequence must be either A, C, G, or T, the four nucleotides that comprise DNA) and they offer potentially greater character independence. Most differences in DNA sequences among closely related taxa are likely to reflect mutational changes that have little if any effect on the function of the genes involved and therefore should show little potential for correlated convergence associated with particular ecological or behavioral adaptations. Molecular data, however, is limited in the number of possible character states (four for DNA sequences), and is therefore characterized by levels of homoplasy that may be greater than that found in morphological data.As

a result, any one molecular character is likely to be inconsistent with the evolutionary history of a group of organisms, but hierarchical patterns reflecting common ancestry generally emerge when a large number of molecular characters are considered together in a single analysis. A few previous studies have used genetic data to address cuckoo relationships. These studies have used cytological, protein, and molecular sequence data and have used both phenetic and discrete character approaches to phylogenetic analysis. (1) As in other birds, chromosome numbers in cuckoos are large. The typical diploid number is 78–80, including 6 pairs of macro-chromosomes (Yamashina 1946, Ray-Chaudhuri 1967, 1973, de Lucca 1974, Waldrigues and Ferrari 1980, 1982, Waldrigues et al. 1983, Jensen 1980; other references are in Erritzoe 2000). The large number of avian chromosomes and the presence of barely visible micro-chromosomes that are difficult to count or identify, greatly complicates attempts to use chromosome numbers as evidence of systematic relationships. Nevertheless, the shapes of chromosomes may provide phylogenetic information.As an example, the karyotypic morphologies of the Common Cuckoo and the Diederik Cuckoo are more similar than either is to the Common Koel (RayChaudhuri 1973). While this result is consistent with the molecular analysis described in the next section, karyotypic studies of cuckoos are too few to support further inferences of their relationships. (2) Brush and Witt (1983) used electrophoresis of feather keratins as a genetic test of cuckoo relationships. Their similarity matrix suggested that (1) Old World and New World cuckoos are distinct groups, (2) Old World brood-parasitic cuckoos are most closely related to the malkohas, (3) this clade is the sister group of the coucals, (4) the New World Coccyzinae are most closely related to the anis, and (5) this clade is most closely related to the New World ground cuckoos (Payne 1997b). As in other studies, only a few species were compared and not all these were compared directly with each other. Nonetheless, these results are generally consistent with the analysis of cuckoo relationships presented below (except for a different placement of the coccyzine cuckoos).

70 The Cuckoos (3) In a broad survey of all birds based on thermal dissociation of species-pairs of DNA, Sibley and Ahlquist (1990) recognized six families of cuckoos: Cuculidae, including the Old World parasitic cuckoos, Old World malkohas, couas and Old World ground-cuckoos Carpococcyx; the coucals Centropodidae; the American cuckoos Coccyzidae; the anis Crotophagidae; the hoatzin Opisthocomidae; and the New World groundcuckoos Neomorphidae. Recognition of these six cuckoo families was based on an arbitrary criterion for delineating family level taxa, using genetic distances derived from thermal dissociation data.Their study was limited in not including any malkohas, couas, Old World ground-cuckoos, or any crested cuckoos Clamator, or New World brood parasitic cuckoos; and although their systematic scheme (Sibley and Ahlquist, Fig. 360) shows a coucal, no coucal appears in their thermal dissociation curves (Figs. 58, 76–84) or in their pooled-species analysis of cuckoos (Fig. 333). Their analysis (Fig. 333) shows Coccyzus (along with Piaya and Saurothera) and Cuculus (with other Old World brood parasites) as sister groups, and Geococcyx and Crotophaga (with Guira) as sister groups, results that are consistent with our analyses (see below). Other aspects of Sibley and Ahlquist’s phylogeny are inconsistent with our analyses and appear to have been based on limited data (inclusion of hoatzin in Cuculiformes and placement of coucals Centropus as the sister group to Cuculidae), or no data at all (inclusion of malkohas, couas, and Carpococcyx in Cuculidae). (4) In an analysis of partial sequences of the mitochondrial cytochrome b and ND2 genes from 15 cuckoo species, Aragón et al. (1999) found that brood-parasitic cuckoos occurred in three separate clades: (1) Old World brood-parasitic cuckoos (Cuculus, Cercococcyx, Cacomantis, Chrysococcyx), (2) malkohas (including an Old World nesting cuckoo Dasylophus superciliosus, New World Piaya cayana and the Old World brood parasites Clamator jacobinus and C. glandarius), and (3) New World anis and ground-cuckoos (Guira, Geococcyx, Neomorphus, along with the brood-parasites Tapera and Dromococcyx). Following Hughes (1996a), the authors suggested that nesting behavior in some cuckoos may have been derived from parasitic behavior,

although this conclusion does not follow from a parsimonious reconstruction of reproductive behavior on their phylogeny. Limited sampling of both molecular characters (851 base pairs per taxon) and cuckoo taxa, and the inclusion of a single outgroup species may all have contributed to somewhat ambiguous results in this study. (5) Johnson et al. (2000) analyzed mitochondrial DNA sequences for 26 cuckoos, the hoatzin, and two turacos, with a focus on the couas. Their results suggested that the Old World brood parasites are most closely related to the New World “Coccyzinae” (Coccyzus and Piaya) and that these two clades are in turn sister to the couas. Coucals Centropus were at the base of the tree within a clade also including the New World cuckoos Geococcyx, Guira, and Crotophaga.This study was also based on a limited set of molecular characters (951 base pairs per taxon) and did not include a number of key taxa, such as the Old World malkohas, Carpococcyx, and the brood-parasitic Tapera, Dromococcyx, and Clamator. In spite of some differences, the above studies show several results in common, and suggest the potential for molecular sequence data to provide a robust analysis of cuckoo systematics. All the above studies included a limited sample of taxa and therefore generate inferences primarily about relationships among the major groups of cuckoos. By necessity, a complete phylogenetic hypothesis that resolves relationships at the genus and species level requires the sampling of character information for all species. In addition, the two recent DNA sequencing studies on cuckoos have included a relatively small number of molecular characters per taxon, limiting the power of these analyses to resolve certain aspects of the phylogeny. In the present work, we have obtained a larger set of DNA sequence data for a complete sample of extant cuckoo species, to provide a robust and comprehensive analysis of the evolutionary relationships among cuckoos. As in many other avian systematics studies, our analysis is based on DNA sequences from the mitochondrial genome. The mitochondrial DNA (mtDNA) is a small genome, comprising about 17,000 base pairs of double-stranded DNA in a

A molecular genetic analysis 71 closed loop. It resides within the cell’s mitochondria and is independent of the chromosomal DNA in the nucleus.The mtDNA includes a small set of genes that is associated with cellular respiration and mitochondrial protein synthesis.The same complement of 13 protein-coding genes, 2 ribosomal RNA (rRNA) genes and 22 transfer RNA (tRNA) genes is shared by nearly all metazoa. A number of characteristics make mtDNA well suited for phylogenetic analyses of relatively closely related taxa. Because it is maternally inherited and effectively haploid, mtDNA has a smaller effective population size than nuclear genes, which results in a greater likelihood that the mitochondrial “gene tree” will correspond to the history of speciation events in the group under study (Avise 1994, Moore 1995, Klein and Takahata 2002). In addition, mtDNA exhibits a relatively rapid rate of evolutionary change, such that it provides ample information about relationships among closely related species. On the downside, this high rate of evolution results in a relatively high level of homoplasy in comparisons among more distantly related taxa. Particularly important for our study, mtDNA also has a high copy number per cell in avian tissues other than blood.This makes it feasible to amplify and sequence from samples with only small amounts of degraded DNA. Our goal of a comprehensive analysis of cuckoo relationships and the fact that many kinds of cuckoos have not been collected recently required that we obtain samples for many taxa from museum specimens that were collected up to 100 or more years ago. We find that mtDNA sequences can be obtained readily from most specimens less than 50 years old, and, by amplifying smaller segments of DNA, from specimens ranging up to 100 years or older (Payne and Sorenson, 2003).

Methods We extracted DNA from muscle tissue obtained from museum tissue collections or from the base of a feather plucked from live birds in the field or in captivity. For analyses of older museum skins, we extracted DNA from the base of a single feather or a small piece of skin cut from the foot. DNA extractions from museum skins were carried out in

a dedicated laboratory from which PCR products and extracts from fresh tissues were excluded. As in previous studies (Sorenson et al. 1999, Sorenson and Payne 2001), we amplified mitochondrial gene regions in overlapping fragments using the polymerase chain reaction (PCR) and sequenced the products using fluorescently labeled di-deoxy nucleotides and an automated DNA sequencer. For this study, we sequenced the mitochondrial genes for NADH dehydrogenase subunit 2 (ND2) and the small subunit ribosomal RNA (12S) along with portions of the transfer-RNA genes flanking both ends of both genes. Our choice of genes to sequence was intended to generate a large number of phylogenetically informative characters while also maximizing the potential independence of the data.We therefore chose two genes evolving under different functional constraints, such that possible systematic biases (Swofford et al. 1996) would be less likely to affect both genes in the same way. The mitochondrial cytochrome b gene has been sequenced more often than any other gene in avian molecular systematics studies to date (Sorenson et al. 1999), but the utility of this gene has been questioned (Meyer 1994, Weins and Hollingsworth 2000), and recent analyses suggest that other, more variable genes, provide greater phylogenetic signal (Yang 1998, Klicka et al. 2000).Among avian taxa for which complete mitochondrial genomes have been sequenced (Desjardins and Morais 1990, Harlid et al. 1997, Mindell et al. 1999), all protein-coding genes show similarly high levels of variation in third-codon positions, but vary greatly in amino acid conservation. Our choice of ND2 was based on the availability of robust avian-specific primers for this gene (Sorenson et al. 1999), and the expectation that it would provide greater phylogenetic information, particularly in the first and second positions, than other mitochondrial protein-coding genes (ND2 is the third most variable gene in amino acid sequence, after ATPase 8 and ND6, both of which are much shorter in length; Saccone et al. 1999). An important advantage of the 12S gene for this study is the presence of many conserved sequence blocks associated with the secondary structure of the rRNA molecule. This allowed us to design primers for both DNA strands spaced

72 The Cuckoos at 200–300 base intervals and to amplify the gene in small fragments when working with degraded DNA samples from older museum specimens. Obtaining ND2 sequences from older samples was more difficult because primers designed to amplify smaller fragments had to be tailored for different groups of cuckoos. Table 8 lists the cuckoo samples included in our genetic analysis of cuckoo phylogeny. Our objective was to include samples from all cuckoo species, and we obtained at least some sequence data for 202 individual cuckoos representing all but one of the 141 species recognized here. Sequences of both ND2 and 12S are complete for 143 samples, representing 119 species. As in the analysis of skeletal characters, we also include in our analysis several outgroup taxa (not in Table 8): mousebirds Coliiformes (2 species), parrots Psittaciformes (2 species), turacos Musophagiformes (2 species), the hoatzin Opisthocomiformes, doves Columbiformes (2 species), songbirds Passeriformes (3 species), a duck (Anseriformes), chicken (Galliformes), and 2 ratites (Struthioniformes). The entire data set comprises 399,155 base pairs of sequence information. For samples with complete data, the 12S and ND2 sequences (along with portions of the flanking tRNA genes), respectively, include 1020– 1042 and 1085–1098 base pairs of sequence data per taxon. In a preliminary alignment of these sequences, 737 positions of the 12S region and 1074 positions of the ND2 region (including the entire coding sequence) align unambiguously and without gaps in any taxa. Given that a significant portion of the 12S data set comprises sequence regions that are variable in length among taxa, and given that much of the informative variation in this gene occurs in these regions (Sorenson and Payne 2001), sequence alignment was a very important issue in our analysis.Typically, molecular systematists align sequences as the first step in a phylogenetic analysis. Where sequences vary in length among taxa, gap characters (“-”) are introduced as necessary so that nucleotides that are homologous across taxa are included in the same columns of the data matrix. For a set of closely related taxa such as the cuckoos, it is relatively easy to identify and align conserved portions

of the sequence, either manually or with the aid of a computerized alignment algorithm (e.g., Clustal W; Thompson et al. 1994). Inevitably, however, variable regions remain for which the alignment of sequences is ambiguous, and these regions are typically excluded from phylogenetic analyses. Exclusion of this data makes sense if the arrangement of nucleotides into columns is essentially arbitrary, such that there is little confidence that the nucleotides in a column share positional homology. Unfortunately, these regions of uncertain alignment may contain a great deal of potential information about phylogenetic relationships (Giribet and Wheeler 1999, Sorenson and Payne 2001). We think the best solution to this problem is the optimization alignment approach described by Wheeler (1996) and implemented in the computer program POY (Gladstein and Wheeler 1996). In an optimization alignment tree search, sequences are input into the analysis without a priori alignment. In effect, the alignment is reconsidered (optimized) for each of the different tree topologies considered by the tree search algorithm, to find the alignment/topology combination that provides the most parsimonious explanation of the data. This approach leads to far more parsimonious solutions than the usual two-step process of creating a static alignment and then searching for the best or shortest tree for that particular alignment. It allows the information present in variable “gap regions” to be used in a rigorous and unbiased way. Based on preliminary analyses, we excluded 45 individual samples with sequences that were identical or nearly identical to another individual of the same species, leaving a set of 171 individuals (157 cuckoos and 14 outgroup taxa) for our principal analyses. Our analysis of this data set in POY closely followed the approach used by Sorenson and Payne (2001). For each of two weighting schemes, we completed 50 replicate tree searches, each with a different random addition of sequences to the analysis. In the first set, all changes (transitions, transversions, insertions and deletions) were given equal weight. In the second analysis, transitions, (changes between the two purines, A and G; or between the two pyrimidines, C and T), were given a weight of 1, whereas transversions (changes between purines

Table 5.1 List of specimens used for DNA sequencing. The museum specimen number is provided for genetic samples that are associated with a voucher specimen. Note that one or more additional numbers identifying the collector, preparator, and/or tissue collection are often associated with these specimens. Tissue types included T: samples from frozen tissue collections (generally muscle); M: a feather and/or skin sample from museum skin specimens; and F: feather samples for which no voucher specimen is available (e.g., samples taken from live birds in captivity or birds that were captured and released in the field). Sequence data collected to date from the mitochondrial small subunit ribosomal RNA (12S) and NADH dehydrogenase subunit 2 (ND2) genes are indicated for each sample (complete sequence: √; partial sequence: 0.5, 0.75). Locality

Guira guira Guira guira Crotophaga major Crotophaga ani Crotophaga sulcirostris Tapera naevia Dromococcyx phasianellus Dromococcyx pavoninus Morococcyx erythropygus Geococcyx californianus Geococcyx velox Neomorphus geoffroyi squamiger Neomorphus geoffroyi salvini Neomorphus radiolosus Neomorphus rufipennis Neomorphus pucheranii Centropus milo Centropus ateralbus Centropus menbeki Centropus menbeki Centropus chalybeus Centropus unirufus Centropus chlororhynchos Centropus chlororhynchos Centropus celebensis celebensis Centropus celebensis celebensis

Bolivia, Santa Cruz Dept. captive Ecuador Puerto Rico, Cabo Rojo Nicaragua Bolivia, Santa Cruz Dept. Peru, Ucayali Dept. Peru, Ucayali Dept. Nicaragua Arizona, Pima Co. Nicaragua Brazil, Amazonas, Parintins, rt bank R. Amazon Panama, Darién Prov. Colombia, Cauca Dept. Guyana Peru, S. Rio Amazonas Solomon Islands, New Georgia Bismark Archipelago, New Ireland Indonesia,West Irian, Koesik Indonesia,West Irian, Lake Sentani Biak Philippines, Luzon Sri Lanka Sri Lanka Sulawesi Sulawesi

Museum Specimen No.

Tissue Type LSU 125706 (skeleton) T UMMZ 236027 T ANSP 83032 T LSU 150033 T UNLV 4308 T LSU 150544 T LSU 156873 (partial skeleton) T LSU 156024 T UWBM 69016 T UMMZ, feathers only F UWBM 68990 T AMNH 278613 M LSU 108160 T YPM 54524 M ANSP (skin in Guyana museum) T LSU 114631 T UWBM 63013 T ZMUC Noona 17.04.1962 M RMNH 4268 M YPM 74879 M ANSP 132995 M DMNH 68328 M MCZ 184954 M YPM 42613 M MCZ 270146 M RMNH 61684 M

Sequences 12S ND2 √ √1 √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ 0.5 √ √ √ √ 0.5 √ √ √

A molecular genetic analysis 73

Species

Species

Locality

Museum Specimen No.

Centropus rectunguis Centropus melanops Centropus steerii Centropus nigrorufus Centropus s. sinensis Centropus s. andamanensis Centropus toulou Centropus phasianinus nigricans Centropus phasianinus spilopterus Centropus phasianinus phasianinus Centropus bernsteini Centropus goliath Centropus goliath Centropus violaceus Centropus grillii Centropus viridis viridis Centropus bengalensis lignator Centropus bengalensis javanensis Centropus senegalensis “epomidis” Centropus senegalensis Centropus leucogaster leucogaster Centropus leucogaster efulensis Centropus anselli Centropus monachus Centropus cupreicaudus Centropus superciliosus burchellii Centropus superciliosus loandae Centropus superciliosus sokotrae Carpococcyx viridis

Malay Peninsula Philippines, Mindanao Philippines, Mindoro Java India, Haridwar Andaman Is Madagascar New Guinea, Morobe Prov. Kei Islands,Toeal Australia, QD, Shoalwater Bay Papua New Guinea, Maiwara Halmahera Halmahera Bismark Archipelago, New Ireland Malawi, Mchinji Dist. Philippines, Negros Taiwan Philippines, Negros Nigeria, Lagos Gambia Liberia Cameroon Angola, Cuanza Norte Prov. Angola, Malanje Prov. Zimbabwe South Africa, KwaZulu/Natal Namibia, Caprivi Strip Socotra Island Sumatra

AMNH 628149 CMNH 35734 YPM 34670 USNM 219367 UMMZ, feather only BMNH 87.12.2.1445 FMNH 384682 MVZ 149111 AMNH 628276 QM 29481 YPM 89146 MCZ 277981 USNM 571638 ZMUC Noona 17.04.1962 NMM 1988.1.34 CMNH 37080 MVZ 140273 CMNH 37079 BMNH 1951.7.8 UMMZ 235198 BMNH 1977.20.218 BMNH 1951.34.240 YPM 50356 YPM 50358 UMMZ, tissue, photo UWBM 62986 UMMZ 202732 USNM 518004 RMNH 04/EB

Tissue Type M T M M F M T M M M M M M M M T M T M T M M M M T T M M M

Sequences 12S ND2 √ 0.75 √ √ √ √ √ 0.5 √ √ 0.25 √ √ √ √ 0.75 √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ 0.75 √ √ √ √ √ √ √ √ 0.5 √ √ √ √ √ √ √ 0.75 0.15 √

74 The Cuckoos

Table 5.1 contd.

captive captive Sabah Madagascar Madagascar Madagascar Madagascar Madagascar Madagascar Madagascar Madagascar Madagascar Madagascar Thailand Sumatra Sumatra Central African Republic South Africa India, New Delhi Java, Jakarta market captive Malaysia, Sarawak Sri Lanka, Uva captive Sumatra India,Tamil Nadu India, Assam Thailand Philippines, Luzon Philippines, Luzon Philippines, Luzon Sulawesi Sulawesi

UMMZ 236029 UMMZ 219043 BPBM 10588 FMNH 352802 FMNH 384680 FMNH 356640 (skeleton) AMNH 628959 YPM 69417 FMNH 352799 FMNH 352797 FMNH 384681 FMNH 356638 (skeleton) MCZ 84298 OU 15103 MCZ 177724 YPM 42608 AMNH 10824 (alcoholic) DM 36943 MSU 4063 UMMZ, feather only UMMZ 236439 YPM 51093 FMNH 278258 UMMZ 236628 ANSP 139384 FMNH 229803 UMMZ 142974 OU 15102 CMNH 636502 UMMZ 233062 UMMZ 228024 RMNH 61661 MCZ 270138

T M M T T T M M T T T T M M M M T M M F T M M T M M M M T M M M M

√ —2 √ √ √ √ √ √ √ √ √ √ 0.25 √ √ √ √ √ √ √ √ √ √ √ 0.75 √ √ √ √ —2 √ √ 0.25

√ √ √ √ √ 0.75 √ √ √ √ √ √ 0.5 √ √ √ √ √ 0.75 0.5 √ 0.25 0.5 0.75 √ √ √ √

A molecular genetic analysis 75

Carpococcyx renauldi Carpococcyx renauldi Carpococcyx radiatus Coua caerulea Coua cristata Coua ruficeps olivaceiceps Coua verreauxi Coua coquereli Coua cursor Coua reynaudii Coua gigas Coua serriana Coua delalandei Rhinortha chlorophaea Rhinortha chlorophaea Rhinortha chlorophaea Ceuthmochares aereus Ceuthmochares australis Taccocua leschenaultii Zanclostomus javanicus Zanclostomus javanicus pallidus Phaenicophaeus sumatranus Phaenicophaeus pyrrhocephalus Phaenicophaeus curvirostris curvirostris Phaenicophaeus diardi Phaenicophaeus viridirostris Phaenicophaeus tristis “tristis” Phaenicophaeus tristis “longicaudatus” Dasylophus cumingi Dasylophus cumingi Dasylophus superciliosus Rhamphococcyx calyorhynchus Rhamphococcyx calyorhynchus

Species

Locality

Clamator coromandus Clamator glandarius Clamator jacobinus serratus Clamator jacobinus pica Clamator levaillantii Coccycua minuta Coccycua pumila Coccycua cinerea Coccycua cinerea Piaya melanogaster Piaya cayana Coccyzus melacoryphus Coccyzus minor Coccyzus ferrugineus Coccyzus euleri Coccyzus americanus Coccyzus erythropthalmus Coccyzus lansbergi Coccyzus pluvialis Coccyzus pluvialis Coccyzus rufigularis Coccyzus vieilloti Coccyzus merlini Coccyzus longirostris Coccyzus vetula Pachycoccyx audeberti Microdynamis parva grisescens Eudynamys scolopacea alberti Eudynamys scolopacea melanorhynca

Java, Jakarta market Gambia South Africa India, Haridwar Zimbabwe Bolivia, La Paz Dept. Colombia Bolivia, Beni Dept. Argentina Guyana Bolivia, Santa Cruz Dept. Ecuador Venezuela Cocos Island Ecuador Michigan Michigan Ecuador Jamaica Jamaica Dominican Republic Puerto Rico, San German Cuba, La Guira NP Dominican Republic Jamaica Central African Republic Indonesia,West Irian Solomon Islands Sulawesi

Museum Specimen No.

Tissue Type UMMZ, feather only F UMMZ, feather only F UNLV MBM 7489 T UMMZ, feather only F UMMZ, feather, photo F LSU 101256 (skeleton) T CU 35069 M LSU 123537 T USNM 614640 (wing, skeleton) T USNM 621716 (skeleton) T LSU (skin in Bolivia) T ANSP 187026 T MHNLS 10097 T CM 123814 M ANSP 185132 T UMMZ 225060 T UMMZ 236139 T ANSP 186069 T MVZ 149905 M UF 26233 M AMNH 164128 M LSU (skin in Puerto Rico) T ANSP 186545 T UMMZ, feather only F AMNH 155237 M AMNH 831777 T UMMZ 114159 M UWBM 58734 T MCZ 39733 M

Sequences 12S ND2 √ √ √ √ √ √ 2 — √ √ √ √ √ √ √ √ √ 2 — √ √ √ √ √ √ √ √ √ 0.75 0.75 √ √ √ √ √ √ √ √ √ √ √ √ 0.5 √ √ √ √ √ √ √ √ √ √ √ √ √ √ √

76 The Cuckoos

Table 5.1 contd.

India, Haridwar Australia, NSW Australia, QD New Zealand, Hamilton Australia, QD Thailand, Lampang Prov. Thailand,Tak Prov. Sumatra, Padang South Africa Uganda Central African Republic Kenya, Kericho Ghana Liberia Indonesia,West Irian Australia, NSW Australia, QD, Cape York Peninsula Australia, QD, Gowrie Ck/Herbert R junction Australia, QD, Cockatoo Ck Tanimbar New Caledonia Australia, NSW Indonesia,West Irian Papua New Guinea, Baiyer River Papua New Guinea,Wau Australia, NSW New Guinea Java, Jakarta market Java, Jakarta market Nepal Indonesia, Bacan Philippines, Mindanao Philippines, Mindanao

UMMZ, feather only CU 39830 QM 29927 AM 98.55 QM 29923 YPM 39592 YPM 68155 UWBM 67472 UWBM 53071 FMNH 364198 AMNH 24783 (skeleton) NMGL 1986.2.107 LSUMZ 168425 AMNH 827449 YPM 75000 UWBM 57616 ANWC 28619 QM 28434 QM 27712 RMNH P4503 UMMZ 221890 UWBM 57460 YPM 74870 YPM 89142 YPM 89143 UWBM 57432 MVZ 149109 UMMZ feather, photo UMMZ feather, photo FMNH 275817 YPM 74832 FMNH 357425 FMNH 357424

F M M M M M M T T T T M T T M T T M M M M T M M M T M F F M M T T

√ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √

√

√ √ √ √ √ √ √ √ √ √ 0.25 √ √ √ √ √ √

√ √ √ √

√ √

A molecular genetic analysis 77

Eudynamys scolopacea scolopacea Eudynamys scolopacea cyanocephala Eudynamys scolopacea cyanocephala Urodynamis taitensis Scythrops novaehollandiae Chrysococcyx maculatus Chrysococcyx xanthorhynchus Chrysococcyx xanthorhynchus Chrysococcyx caprius Chrysococcyx caprius3 Chrysococcyx klaas Chrysococcyx cupreus Chrysococcyx cupreus Chrysococcyx flavigularis Chrysococcyx megarhynchus Chrysococcyx basalis Chrysococcyx osculans Chrysococcyx minutillus poecilurus Chrysococcyx minutillus barnardi Chrysococcyx minutillus crassirostris Chrysococcyx lucidus layardi Chrysococcyx lucidus plagosus Chrysococcyx ruficollis Chrysococcyx meyeri Chrysococcyx meyeri Cacomantis pallidus Cacomantis leucolophus Cacomantis sonneratii Cacomantis merulinus lanceolatus Cacomantis passerinus Cacomantis variolosus “heinrichi” Cacomantis variolosus sepulcralis4 Cacomantis variolosus sepulcralis

Species

Locality

Museum Specimen No.

Cacomantis variolosus sepulcralis Cacomantis variolosus variolosus Cacomantis castaneiventris Cacomantis flabelliformis Cercococcyx mechowi Cercococcyx mechowi Cercococcyx olivinus Cercococcyx montanus Surniculus dicruroides Surniculus dicruroides Surniculus velutinus velutinus Surniculus velutinus velutinus Surniculus musschenbroeki Surniculus lugubris Surniculus lugubris Hierococcyx vagans Hierococcyx varius Hierococcyx varius Hierococcyx sparverioides Hierococcyx hyperythrus Hierococcyx pectoralis Hierococcyx pectoralis Hierococcyx nisicolor Hierococcyx nisicolor Hierococcyx nisicolor Hierococcyx fugax Cuculus clamosus clamosus Cuculus clamosus gabonensis5 Cuculus clamosus gabonensis6

Borneo, Sabah Australia, QD, Brisbane Australia, QD, Claudie River 3rd Crossing Australia, QD, Maleny Central African Republic Uganda Angola, Cuanza Norte Prov. Uganda Nepal Bhutan Philippines, Mindanao Philippines, Mindanao Halmahera Sabah Sabah Thailand India, Assam Oman Nepal Japan, Nagano Pref. Philippines, Mindanao Philippines, Mindanao India, Darjeeling Dist. Singapore Singapore Singapore South Africa Nigeria, Ibadan Nigeria, Ibadan

Sabah Parks 17065 QM 26747 QM 27099 QM 30248 AMNH 831292 UMMZ 221537 YPM 50330 FMNH 355263 FMNH 221725 USNM 519749 FMNH 286173 YPM 61680 USNM 571628 WFVZ 37595 WFVZ 37594 USNM 534592 UMMZ 142758 BMNH 1989.5.1 AMNH 831293 UMMZ, skin lost in Japan FMNH 357420 CMNH 35841 MCZ 297472 UWBM 67458 UWBM 67545 UWBM 67529 UWBM 52912 CU 29762 CU 29759

Tissue Type T M M M T M M T M M M M M M M M M M T T T T M T T T T M M

Sequences 12S ND2 —2 √ √ √ √ √ √ √ √ √ —2 0.5 √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ 0.75 √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ 0.5 0.5

78 The Cuckoos

Table 5.1 contd.

Cameroon South Africa Zambia Japan Vietnam,Tonkin Sulawesi Singapore Madagascar South Africa,Transvaal Botwswana Japan Philippines, Luzon Philippines, Panay Philippines, Panay Java, Jakarta market Nepal Philippines, Luzon China, Xinjiang UK Japan Russia Russia

FMNH 188943 UWBM 52799 UMMZ, feather only UMMZ 234916 AMNH 833647 AMNH 298824 UWBM 64917 FMNH 384675 UMMZ 216724 USNM 527286 UMMZ 234743 CMNH 36498 CMNH 36760 NMP (tissue: CMNH 4334) UMMZ feather only AMNH 831292 FMNH 6025 (voucher missing) CAS 98001 UMMZ, ex Univ. East Anglia UMMZ tissue only UWBM 46350 UWBM 49796

M T F T T M T T M M T T T T F T T M T T T T

√ √ —2 √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √ √

0.5 √ √ √ √ √ √ √ √ √ √ √ √

captive captive Paraguay captive Gambia Gambia Florida Japan Cameroon Gambia

UMMZ UMMZ UMMZ UMMZ UMMZ UMMZ UMMZ UMMZ UMMZ UMMZ

T T T T F F T T T F

√ √ √ √ √ √ √ √ √ √

√ √ √ √ √ √ √ √ √ √

234219 233646 227492 232536 (A573) (A794) 225650 234875 232516 (A525)

√ √ √ √ √ √ √

A molecular genetic analysis 79

Cuculus clamosus gabonensis Cuculus solitarius Cuculus solitarius Cuculus poliocephalus Cuculus poliocephalus Cuculus crassirostris Cuculus micropterus Cuculus rochii Cuculus gularis Cuculus gularis Cuculus optatus Cuculus optatus Cuculus optatus Cuculus optatus Cuculus saturatus or c. optatus 7 Cuculus saturatus Cuculus saturatus 8 Cuculus canorus subtelephonus Cuculus canorus canorus Cuculus canorus canorus Cuculus canorus canorus Cuculus canorus canorus Outgroup Taxa Colius striatus Urocolius macrourus Nandayus nenday Neophema elegans Musophaga violacea Crinifer piscator Columba leucocephala Treron sieboldii Vidua chalybeata Lanius senator

Species

Locality

Museum Specimen No.

Sequences Tissue Type 12S ND2 Smithornis sharpei Nigeria UMMZ 233887 T √ √ Sayornis phoebe Louisiana UMMZ 226790 T √ √ Opisthocomus hoazin Bolivia LSUMZ 131920 T √ √ Opisthocomus hoazin Peru LSUMZ 156874 (skeleton) T √ √ 1 The ND2 sequence for this sample may be a nuclear copy but it is closely related to the sequence from the other Guira sample. 2 The 12S gene was not sequenced for these samples because their ND2 sequences were identical or nearly so to another sample from the same species. 3 This specimen (tissue FMNH 2051) was misidentified as C. klaas in Johnson et al. (2000). 4 This specimen (tissue FMNH 6635) was misidentified as C. merulinus in Johnson et al. (2000). 5 Breast black, as in four of five specimens from this locality. 6 Breast rufous, as in one of five specimens from this locality, which is intermediate between the forest zone and Guinea woodlands. 7 The tail feather of this specimen was identified to one of those species. 8 This specimen was identified on the basis of its DNA sequence because we were unable to locate the voucher specimen.The bird was identified as H. (“C.”) fugax by the collector, and as H. (“C.”) vagans in Johnson et al. (2000). Its DNA sequence is similar to other C. saturatus and C. canorus. Both C. canorus and H. vagans can be excluded, however, because neither occurs in the Philippines (Dickinson et al. 1991) and the DNA sequence suggests that it is neither H. vagans or H. pectoralis (formerly considered a form of H. fugax). Identification to C. saturatus is based on a wing measurement of 181 mm.

80 The Cuckoos

Table 5.1 contd.

A molecular genetic analysis 81 and pyrimidines) and “indels” were given a weight of 2. On average, transitions occur much more frequently than other changes (Wakeley 1996) and therefore may be less reliable as evidence of historical relationships. This is not to suggest that a weighted analysis will necessarily produce a tree that is a better estimate of phylogeny. Rather, our interest in weighting the data was to explore the stability of the results with regard to a change in assumptions. In general, results for cuckoos from the two different weighting schemes were very similar. Because of the complexity of the alignment problem, POY uses heuristic methods to estimate tree length (i.e., the number of evolutionary changes or steps required to explain the data), and therefore the tree length score assigned to any tree, particularly in a large analysis such as this, includes a small but unknown over-estimation. This makes it difficult to conclude that one particular tree (or trees) is the most parsimonious. We therefore considered not only the tree(s) with the lowest estimated score, but also trees within five steps of the best score. In the unweighted analysis, 11 trees with 16119 to 16124 steps were found in 9 of the 50 tree search replicates. In the weighted analysis, 10 trees with 20419 to 20424 steps were found in 7 of the 50 replicates.We used these 21 “best” trees to construct a majority rule consensus tree in which only those groups found in a majority of the 21 trees are shown.We use this consensus tree from the 171-taxon analysis as our primary hypothesis of cuckoo relationships in the discussion and figures that follow, but also point out alternative relationships found in one or more of these 21 best trees.We also conducted 100 replicate tree searches under each weighting scheme for five subgroups within Cuculidae, comprising respectively, (1) Crotophaginae plus Neomorphinae, (2) Centropodinae, (3) Couinae, (4) Phaenicophaeini, and (5) Cuculini.Analyses of these smaller sets of taxa allowed us to more thoroughly explore tree space within each subgroup with less ambiguity in tree scores. In general, however, results of these analyses were highly congruent with the overall analysis. Calculation of bootstrap values (Felsenstein 1985), which measure the relative support for different branches in the tree, was based on a standard static alignment and excluded 138 characters in

regions of particularly ambiguous alignment. In bootstrapping, a large number of replicate data sets are generated by randomly sampling characters from the original data set and then determining the proportion of replicate data sets that support each node in the tree.

Relationships, evolution and biogeographic histories The molecular data provides strong support for the monophyly of cuckoos and for the other avian orders in our analysis, respectively, but it provides only limited support for higher-level (inter-ordinal) relationships. None of our analyses indicated the inclusion of hoatzin within Cuculiformes, as suggested by Sibley and Ahlquist (1990), or a sisterrelationship between cuckoos and hoatzin.We also found no evidence for a close relationship between turacos and cuckoos or between turacos and hoatzin, relationships that have been previously suggested (Hughes and Baker 1999). Cuckoos were monophyletic in all our analyses (Tree 1), a result consistent with a historical view of cuckoo systematics, insofar as one or more species in all the major groups of cuckoos were originally described as a Cuculus: Linnaeus in his 1766 edition of Systema Naturae listed 22 species in his genus Cuculus. We have Cuculus canorus as the “typical” cuckoo from an historical and Eurocentric point of view, and many Old World brood parasites and nesting malkohas such as Raffles’s Malkoha that were originally described as Cuculus. In addition, Blue Coua and Crested Coua were described as Cuculus caeruleus and C. cristatus by Linnaeus, Javan Coucal as Cuculus nigrorufus by Cuvier, Pheasant Coucal as Cuculus phasianinus by Latham, Madagascar Coucal as Cuculus Toulou by P. L. S. Müller, Lesser Coucal as Cuculus bengalensis by Gmelin, Philippine Coucal as Cuculus viridis by Scopoli, Senegal Coucal Cuculus senegalensis by Linnaeus, Guira Cuckoo as Cuculus Guira by Gmelin, Lesser Roadrunner as Cuculus velox by A. Wagler, and American Striped Cuckoo as Cuculus naevius by Linnaeus. Within the cuckoo, species molecular analysis provides a remarkably good resolution of evolutionary

82 The Cuckoos

88 100

99

93 89 71 100 59 90

99

87 97

99 63 90

89

8 52

87

80 90 52

70

20 100 54 90

64 61 61 100

50 92 69

Crotophaginae

Neomorphinae Centropodinae Couinae

Phaenicophaeni

100

Guira Crotophaga Tapera Dromococcyx Morococcyx Geococcyx Neomorphus Centropus Carpococcyx Coua Rhinortha Ceuthmochares Taccocua Zanclostomus Phaenicophaeus Dasylophus Rhamphococcyx Clamator Coccycua Piaya Coccyzus Pachycoccyx Microdynamis Eudynamys Urodynamis Scythrops Chrysococcyx Cacomantis Surniculus Cercococcyx Hierococcyx Cuculus

Cuculinae

Cuculini

100 96

Figure 5.1. Genus-level phylogeny of cuckoos (Cuculidae) based on mitochondrial DNA sequences. Obligate broodparasites are underlined.The tree shown is a majority rule consensus of the 21 most parsimonious and nearly most parsimonious trees found in optimization alignment analyses using two different weighting schemes (see text).The analysis included 157 cuckoo sequences, plus 14 outgroup taxa (not shown). Genus-level relationships were consistent among trees, except for six branches (shown as dashed lines) that were found in most, but not all trees.The proportion of the 21 “best” trees including each of these clades is indicated below each branch (underline, italics). Alternative arrangements of cuckoo subfamilies and genera within Cuculinae were found in a small proportion of the best trees and these are described in the text. Bootstrap percentages based on 500 replicate data sets are shown above each node. Monophyly of all genera recognized here was strongly supported.

relationships and strong support for most groups or clades in the tree. A phylogenetic hypothesis for cuckoo genera is shown in Figure 5.1.The basal division in this tree separates a clade of cuckoos found only in the New World, from a more species-rich clade that includes all the Old World cuckoos plus a few New World genera. We recognize five subfamilies as representing the main phylogenetic lineages among cuckoos. These are Crotophaginae and Neomorphinae in the New World, Centropodinae and Couinae in the Old World, and Cuculinae which has its origin and greatest diversity in the Old World, but includes a clade of New World cuckoos represented by Coccycua, Piaya, and Coccyzus (Figure 5.1). Some of the trees found in our optimization

alignment analyses included alternative arrangements of the five cuckoo subfamilies. In 62% of the 21 “best” trees, subfamily relationships were as shown in Figure 5.1. In 29% of the best trees, Couinae was the sister group to Cuculinae (this result was obtained only in the weighted analysis); and in 9% of trees, the Centropodinae/Couinae clade was the sister group of all other cuckoos (this result was found only in the equal weights analysis). Because the topology shown in Figure 5.1 was found most often and under both weighting schemes, we used it as our working hypothesis of cuckoo relationships. Sequence data from additional mitochondrial genes or nuclear genes, and more thorough analyses, are needed to further test this hypothesis and provide

A molecular genetic analysis 83 better resolution of some aspects of cuckoo phylogeny. The linear systematic treatment of cuckoo genera and species in this book is based on Tree 1 and on the species-level phylogenies of cuckoo subgroups. By convention, the descendant branch in each bifurcation with fewer taxa is considered “basal”, and these taxa are listed first in the linear sequence. Monophyly was used as a minimal criterion for recognition of taxa at each systematic level above the species. In principle, each node or branching point in the tree defines two clades which can be named in a hierarchical system, but only subfamilies and genera are consistently named in this review. Each genus recognized was monophyletic in all of our phylogenetic analyses, and this criterion emphasizes the common ancestry of the species within each genus, rather than the morphological distinctiveness of a set of similar species. In general, the genera we recognize are consistent with morphological and behavioral traits, with the relative degree of genetic divergence between lineages, and with traditional classifications of the cuckoos. In the following trees, we consider the specieslevel relationships among cuckoos and describe in greater detail the results of our phylogenetic analyses. The trees are drawn as phylograms, in which branch lengths are proportional to the minimum number of changes in DNA sequence that must have occurred in each lineage. The total branch length separating any two terminal taxa provides a rough measure of the genetic distance between them and in turn an indication of the time since they shared a common ancestor, although reconstruction of branch lengths using parsimony may substantially underestimate the number of mutation events that occurred in each lineage. In the trees for each cuckoo subfamily or tribe, we show the majority rule consensus tree from the overall analysis, but describe any alternative arrangements of taxa that were found either in the 21 best trees from the overall analysis or in the tree(s) with the best score in our analyses of cuckoo subgroups. Owing in part to the benefits of dense taxon sampling (Zwickl and Hillis 2002) and perhaps also to the spacing of cuckoo diversification events in history, the mitochondrial data provide a

remarkably good resolution of almost all aspects of cuckoo phylogeny above the species level, and strong support for most clades in the phylogeny. In addition, results were generally stable in relation to changing the weights applied to different kinds of character change (i.e., transitions versus transversions). Some of the remaining uncertainties in the cuckoo tree involve taxa with incomplete sequence data,whereas others will require data from additional genes to resolve. Genetic material for taxa with incomplete data came from older museum specimens, from which DNA sequences were more difficult to obtain. The basal cuckoo lineage includes two groups that occur only in the New World (Figure 5.2). Crotophaginae includes the cooperatively breeding Guira Cuckoo Guira guira and the anis Crotophaga in which birds in social groups lay in a common nest from which females remove the eggs and sometimes the nestlings of their group mates. Guira is the basal taxon in this set and Crotophaga ani and C. sulcirostris are sister species, as they are in a phylogenetic analysis of cranial skeletons of the Crotophaginae (Posso and Donatelli 2001). The other lineage, Neomorphinae, includes the roadrunners Geococcyx and ground-cuckoos Morococcyx and Neomorphus, which are monogamous and rear their own young. Phylogenetic relationships among the large ground-cuckoos Neomorphus in the molecular analysis are much the same as suggested by Haffer (1977) on the basis of geography and plumage pattern. Neomorphinae also includes the brood-parasitic cuckoos Tapera and Dromococcyx. The placement of these parasitic birds within this clade is strongly supported by the genetic data and indicates independent origins of obligate brood parasitism in New World and Old World cuckoos. In coucals Centropodinae, the molecular results indicate several lineages (Figure 5.3).A basal lineage includes Centropus menbeki and C. chalybeus in New Guinea and Biak, C. ateralbus in the Bismarck Archipelago, and C. milo in the Solomon Islands. Close relationships among these species are consistent with the known geological history of the New Guinea – Melanesia region. Although the details of this history are particularly complex (Pigram and Davies 1987), the Melanesian arc with the

84 The Cuckoos 100 100 99 100 86 100

50 changes

Guira guira (2) Crotophaga major Crotophaga ani 100 Crotophaga sulcirostris Tapera naevia Dromococcyx phasianellus 100 Dromococcyx pavoninus Morococcyx erythropygus Geococcyx californianus 100 Geococcyx velox 99 Neomorphus rufipennis 87 Neomorphus pucheranii 100 Neomorphus radiolosus Neomorphus g. geoffroyi 97 98 Neomorphus g. squamiger

Figure 5.2. Phylogeny of the New World cuckoos (Crotophaginae and Neomorphinae).This single “best” tree was found in all optimization alignment analyses. In this, and in the following figures, the trees are drawn as phylograms in which branch lengths are proportional to the number of character state changes on each branch, as inferred by parsimony reconstruction using ACCTRAN optimization in PAUP* (Swofford 2002). Bootstrap values are shown above each node.Two samples of G. guira had nearly identical sequences of the 12S gene, but somewhat divergent sequences for the ND2 gene. We suspect that a nuclear copy of the ND2 gene was amplified and sequenced for one of the two samples.

Bismarcks and Solomon Islands came into contact with New Guinea around 10 to 5 million years ago (Lee and Lawver 1995, Michaux 1998, Polhemus and Polhemus 1998), suggesting the maximum age for a common ancestor of Melanesian coucals, and the period when coucals could move from New Guinea into this region. Coucals are absent on Buka and Bougainville but occur on each side of this area, with two species in the Bismarck Archipelago, and one further south in the Solomons. This distribution gap may reflect an ancient extinction or a more recent historical extinction. In the Pleistocene when sea levels were lower, the land mass of Bougainville and most of the Solomons region was larger, presumably allowing coucals to move from the Bismarck Archipelago to the Solomons by way of Bougainville. Bougainville and most island groups in the Solomons were separated from each other by rising sea levels only over the past 10,000 years. Melanesian people have maintained a fully developed and widespread cultural complex in the region for the past 3,500 years (Wickler and Spriggs 1988, Spriggs 1997, Merriwether et al. 1999), and for at least 29,000 years people have occupied the region where cultural remains have been found at the northern end of the

Solomons on Buka Island just north of Bougainville. Indigenous mammals have disappeared on Buka over the past few millenia, after the human introduction of marsupial possums (Phalanger orientalis), rats (Rattus elegans, R. praetor), pigs and dogs (Flannery and Wickler 1990, Flannery and Roberts 1999), all which may have been predators of ground-cuckoos. Buff-headed Coucals Centropus milo occur on other Solomon Islands that never were connected to the large island “Greater Bougainville” (“Greater Bukida”) during low sea level times of the Pleistocene (Diamond et al. 1976, Spriggs 1997, Mayr and Diamond 2001). Circumstantial evidence therefore points to the absence of large coucals as a result of colonization, and growth of human population and associated animals, and yet the two known sites with avian fossils on St. Matthais and New Ireland have not revealed any species that are known only from fossils and do not still live in the area (Steadman and Kirch 1998, Steadman et al. 1999). The extinction of birds on the islands of northern Melanesia has not been as extensive as on smaller islands that are more remote from continental sources, islands such as Hawaii, Easter Island and Henderson Island (Steadman and Olson 1985, Steadman 1995, Mayr and Diamond 2001).

A molecular genetic analysis 85 Centropus milo C. ateralbus 99 C. menbeki (2) C. chalybeus C. unirufus C. chlororhynchos (2) 100 C. melanops 95 33 91 C. steerii C. rectunguis C. c. celebensis (2) 19 C. anselli 100 C. l. leucogaster C. l. efulensis 95 82 C. senegalensis C. senegalensis “epomidis” C. monachus 71 C. cupreicaudus 72 C. superciliosus sokotrae 79 99 C. superciliosus burchellii C. superciliosus loandae C. nigrorufus 100 C. s. sinensis C. s. andamanensis 40 C. toulou 87 C. goliath (2) C. grillii 100 100 C. viridis C. bengalensis lignator 98 C. bengalensis javanensis 32 C. violaceus C. bernsteini 67 C. p. nigricans 95 C. p. phasianinus 50 changes C. p. spilopterus 100

97

100

Figure 5.3. Phylogeny of Centropodinae (Centropus).The tree shown is the majority rule consensus from the overall analysis. Relationships within this genus were stable among analyses, with two exceptions. First, the positions of C. senegalensis and the anselli/leucogaster clade were reversed in weighted analyses. Second, C. goliath was sister to either C. toulou or to the violaceus/bernsteini/phasianinus clade. Bootstrap values are shown above each node.Two samples of C. menbeki, C. chlororhynchos, C. celebensis, and C. goliath, respectively, had nearly identical sequences.

A second coucal lineage consists of species that occur from Africa to southern Asia, the Philippines, the Greater Sundas, New Guinea and Australia. Within this group, the Philippine Centropus unirufus and the Sri Lanka C. chlororhynchos are basal, followed by a clade comprising Short-toed Coucal C. rectunguis from southeast Asia, and two additional shorttoed species from the Philippines, C. melanops and C. steerii (Figure 5.3). Next is C. celebensis from Sulawesi. These coucals have relatively short hallux claws, largely independent of their body size or habitat. Another clade comprises the African coucals Centropus leucogaster, C. anselli, C. senegalensis, C. monachus, C. cupreicaudus and C. superciliosus.

Sister relationships between the large forest coucals C. leucogaster and C. anselli and between C. cupreicaudus and C. superciliosus are strongly supported. The latter result differs from earlier views based on plumage, that regarded C. cupreicaudus, and C. monachus as a single superspecies (Snow 1978), but is consistent with a more recent proposal by Irwin (1985). C. senegalensis (including the plumage phase “epomidis”) was sister to a clade comprising C. monachus, C. cupreicaudus and C. superciliosus in the equal weights analysis, but was the basal lineage in this African clade in the weighted analysis. The broadly distributed Greater Coucal Centropus sinensis and the Javan Coucal C. nigrorufus comprise

86 The Cuckoos another well-supported clade. Finally, we find a geographically and morphologically diverse clade including the Madagascar Coucal C. toulou; two island species with large body size, C. goliath and C. violaceus; a clade comprising the African Black Coucal C. grillii, the Philippine C. viridis, and the widely distributed C. bengalensis, a small bird but with the longest hallux claw of any coucal; and a well-supported group comprising Pheasant Coucal C. phasianinus and Lesser Black Coucal C. bernsteini of New Guinea. Alternative placements of C. goliath were sister to C. toulou or sister to a clade comprising C. violaceus, C. bernsteini, and C. phasianinus, the latter seemingly more consistent with morphology and biogeography. Historically the coucals within a geographic region have been grouped by systematists in terms of plumage and size. Salvadori (1881) recognized three genera in Sulawesi and the Moluccas to New Guinea: (1) Nesocentor (Cabanis and Heine, 1863) (the large species goliath, milo, violaceus, menbeki, chalybeus and ateralbus), (2) Polophilus (Leach, 1814) (including the forms spilopterus, nigricans and bernsteini) and (3) Centrococcyx ( javanensis ⫽ bengalensis, rectunguis), the black-and-brown coucals.Then after a period when several genera were recognized, the coucals were again collapsed into a single genus Centropus (Hartert 1891, Shelley 1891). Mason et al. (1984) and Mason (1997b) retained a single genus but recognized three groups in Australasia based on juvenile plumage and habitat. These three groups corresponded in part with Salvadori’s scheme, although they were not based on phylogenetic analysis and they did not include all coucal species of the Australasian region. The genetic results lead to very different conclusions about the relationships among coucals and indicate that features of body size and plumage color have evolved repeatedly. For example, the large black coucals Centropus goliath and C. phasianinus spilopterus are in the same lineage as most of the smaller black-andbrown coucals, some of which have a seasonal change of plumage (C. toulou of Madagascar, C. grillii of Africa, C. bengalensis of Asia) while others do not (C. viridis of Asia, C. goliath). This result is consistent with the similarity in size and dull black plumage between C. goliath and Pheasant Coucal C. p. spi-

lopterus of the Kai Islands; these large coucals are also similar in the pale patches on their wing coverts. Large body size in C. goliath and C. p. spilopterus, however, reflects a general pattern of body size evolution in island birds, much as in the large subspecies of Lesser Coucal on islands (C. bengalensis sarasinorum on the Lesser Sunda Islands, C. b. medius on the Moluccas), of C. p. mui on Timor, and of Coccyzus lizard-cuckoos in the West Indies. Large body size does not necessarily indicate phylogenetic relationship, insofar as the large C. violaceus and C. milo of the Bismarck Archipelago and Solomon Islands are not closely related, nor are C. goliath of the northern Moluccas and C. menbeki of New Guinea. Couinae includes as sister groups the Madagascar couas Coua and the southeast Asian ground-cuckoos Carpococcyx (Figure 5.4).The two genera are similar in plumage, nests, and the naked skin and brightly patterned mouths of nestlings. Our results for species-level relationships among couas are consistent with those from a recent molecular study ( Johnson et al. 2000); our analysis includes four additional species. The two coua species living in open country, Verreaux’s Coua C. verreauxi and Crested Coua C. cristata, are closely related to each other.The extinct Snail-eating Coua Coua delalandei is most closely related to the Giant Coua C. gigas or Red-breasted Coua C. serriana, which it resembles in size and in dark rufous plumage. Placement of C. delalandei was uncertain because to date we have obtained only 334 base pairs of the 12S sequence from the ~140 year old specimen that we sampled. Within the ground-cuckoos Carpococcyx, relationships are not well-resolved because we obtained only 174 base pairs of the 12S sequence from an old specimen of Sumatran Ground-cuckoo C. viridis. The data nonetheless suggests that C. viridis is distinct as a species in that its 12S sequence differs from that of C. radiatus and C. renauldi by 5% and 9%, respectively. On the basis of genetic distance, C. viridis and the Bornean Ground-cuckoo C. radiatus may be more closely related to each other than either is to the Coral-billed Ground-cuckoo C. renauldi. The common origin of Madagascar couas and Asian ground-cuckoos may date from a time when these land areas were in contact or were closer together than at present. To understand the

A molecular genetic analysis 87 Carpococcyx viridis 100

Carpococcyx radiatus 65

Carpococcyx renauldi (2)

100

Coua cristata 100

Coua caerulea

96

Coua verreauxi

100

Coua ruficeps olivaceiceps 98

Coua reynaudii 99

100

Coua coquereli Coua cursor

47

Coua gigas 98 54 50 changes

Coua delalandei Coua serriana

Figure 5.4. Phylogeny of Couinae (Carpococcyx and Coua), with bootstrap values shown above each node. Relationships within this group are strongly supported with the exception of Coua reynaudii which is sister to the C. gigas/delalandei/ serriana clade in one of the 21 “best” trees. In addition, placements of Carpococcyx viridis and Coua delalandei are not well-supported because limited data was available for these taxa (149 and 334 base pairs of sequence data, respectively).The short terminal branches for these two species, and the seemingly close relationship between C. delalandei and C. serriana are artifacts of missing data. C. delalandei is only slightly more similar to C. Serriana than to C. gigas. Two samples of C. renauldi had nearly identical sequences.

distribution of land birds it is necessary to know the history of land and sea, and the distance between the land areas above sea level (Hall 1998).The geological and climatic history of the Indian Ocean region and the distribution of other terrestrial vertebrates suggest that historically more land was exposed and the sea itself was less of a barrier to dispersal and colonization of land animals than it is now. Madagascar and the drifting Indian subcontinent separated c. 130 mya (Krause et al. 1999), and the Indian subcontinent accreted with Eurasia c. 65–55.5 mya with a suture complete by 49 mya (Beck et al. 1995, Gower et al. 2002). The granitic Seychelles in the Indian Ocean between India, Madagascar and Africa are of continental origin and separated from Africa around 75 mya; the Indian Ocean has been in place between India and Madagascar for at least 50 million years and is very deep (Hall 1998). The Seychelles are emergents of a much larger land mass, the Seychelles Microcontinent, a now largely submerged, continental fragment in the Indian Ocean, about 55,000 km2

in area. This land would have been above sea level during periods of glaciation in the Quaternary, when the sea level was as much as 140 m below the current sea level. These lowstands have occurred at regular intervals over the past 500,000 years (Colonna et al. 1996, Rohling et al. 1998, Raxworthy et al. 2002). Although this period is too recent to account for the movement of large terrestrial cuckoos between southeast Asia and Madagascar and the subsequent speciation of couas and ground-cuckoos, the possibility of dynamic changes on ocean mounts and sea levels suggests an earlier but post-Gondwana opportunity for dispersal between the land masses. In the late Oligocene there were emergent ridges above sea level around the Mascarene Plateau and these ridges only subsided slowly thereafter. Phylogenetic resolution of the genetic relationships among other terrestrial vertebrates also supports a history of postGondwana dispersal between Madagascar and Southeast Asia (Raxworthy et al. 2002, Shapiro et al. 2002).

88 The Cuckoos The large subfamily Cuculinae includes more than half of all cuckoo species and occurs in both the Old and New World. The phylogeny indicates an origin in the Old World followed by a single invasion of the New World (Figure 5.1).The New World taxa comprise a single derived clade (Coccycua, Piaya, and Coccyzus), whereas the basal lineages within the group (e.g., Raffles’s Malkoha Rhinortha chlorophaea, The yellowbills Ceuthmochares) and the sister clade to Cuculinae (Couinae plus Centropodinae) are all Old World birds. Cuculinae also includes both nesting cuckoos and brood parasitic cuckoos. One of the most significant results of the mitochondrial genetic analysis is that brood-parasitism evolved in this clade on at least two separate occasions, once in the lineage leading to Clamator and once in the common ancestor of all other Old World brood parasites, tribe Cuculini (Figure 5.1). Three tribes are recognized: Rhinorthini, Phaenicophaeini (Figure 5.5) and Cuculini (Figure 5.6). A surprising result of the molecular analysis was the position of Raffles’s Malkoha Rhinortha chlorophaea, basal to the all other Cuculinae, including both nesting and brood-parasitic cuckoos. Rhinortha is highly divergent genetically and its placement as the sister group to all other taxa in Cuculinae is strongly supported. This result was obtained in separate analyses of the two different mitochondrial gene regions sequenced, and it was confirmed by sequencing three specimens of this species, all of which had nearly identical sequences. Rhinortha is the only extant member of a divergent clade recognized here as a tribe, Rhinorthini. The tribe Phaenicophaeini includes all other nesting species of Old World and New World Cuculinae as well as the brood-parasitic Clamator (Figure 5.5). This arrangement suggests that Clamator and the New World nesting cuckoos (Coccycua, Piaya and Coccyzus) were derived from malkoha-like ancestors. The large malkoha genus “Phaenicophaeus” in the broad sense of Delacour and Mayr (1945) and Schlegel (1862) is not a monophyletic group but represents a paraphyletic assemblage that includes the basal lineages in this tribe. The African yellowbill genus Ceuthmochares (in African tradition called a “coucal” by Layard 1874, Sclater 1930, Bannerman 1933, Roberts 1940 and Maclean 1993), a malkoha in its morphology,

chattered vocalizations and nesting behavior, is the sister group to all other species in Phaenicophaeini. The other malkohas (Taccocua, Zanclostomus, Phaenicophaeus, Dasylophus and Rhamphococcyx) may or may not form a monophyletic group: a clade comprising all malkoha species in these five genera was found in 2 of the 10 best trees from the weighted analysis on 171 taxa but not in analyses of Phaenicophaeini taxa only.The relative positions of Rhampococcyx, Dasylophus, and Clamator, in particular, were variable among analyses. Malkohas are diverse in morphology and plumage ornaments, and for many years these birds were classified in several separate genera, based on differences in the feathering on the head and the shape of the bill and nostril (Sharpe 1873, Shelley 1891). This view changed when Delacour and Mayr (1945) combined all the Asian malkohas and the African Ceuthmochares (separate genera that they were “unable to admit”) into a single genus Phaenicophaeus, with the assertion that “in general characters and habits the various Malcohas are so closely related that it seems more logical to consider them all as species of a single genus (Phoenicophaeus—1815).”The authors described no characters, even though several skeletal and other morphological differences among these cuckoos have been described by others (Sharpe 1873, Pycraft 1903, Seibel 1988, Hughes 2000). Delacour and Mayr apparently based their opinion on similarities they saw in skins. The genetic results (Figure 5.5) suggest the recognition of five Asian genera in addition to Rhinortha and the African genus Ceuthmochares. In Figure 5.5, six malkoha species are included in the genus Phaenicophaeus. Chestnut-breasted Malkoha P. curvirostris (previously included in Rhamphococcyx) is the sister group to a clade of more closely related species, Chestnut-bellied Malkoha P. sumatranus, Redfaced Malkoha P. pyrrhocephalus, Blue-faced Malkoha P. viridirostris, Black-bellied Malkoha P. diardi, and Green-billed Malkoha P. tristis (all but pyrrhocephalus formerly included in Rhopodytes).These birds have a large area of brightly colored bare skin around the eyes and sides of the face, an arched or deep bill, and many have contrasting colors on the bill. Relationships within Phaenicophaeus were generally stable among analyses, yet the result was not strongly

A molecular genetic analysis 89 Ceuthmochares aereus Ceuthmochares australis Taccocua leschenaultii 99 Zanclostomus javanicus (Java) Zanclostomus javanicus (captive) 80 Phaenicophaeus curvirostris 91 Phaenicophaeus sumatranus 93 Phaenicophaeus pyrrhocephalus 48 Phaenicophaeus viridirostris 70 Phaenicophaeus d. diardi 69 99 Phaenicophaeus tristis "tristis" Phaenicophaeus t. "longicaudatus" Rhamphococcyx calyorhynchus (2) 52 Dasylophus superciliosus 99 Dasylophus cumingi (2) Clamator coromandus 100 63 Clamator glandarius 95 Clamator levaillantii 100 Clamator jacobinus pica Clamator jacobinus serratus 8 Coccycua minuta 96 100 Coccycua pumila Coccycua cinerea (2) 100 87 Piaya cayana Piaya melanogaster Coccyzus melacoryphus 70 100 100 Coccyzus americanus Coccyzus euleri 97 100 Coccyzus minor 100 Coccyzus ferrugineus Coccyzus erythropthalmus 84 Coccyzus lansbergi 100 Coccyzus pluvialis (2) 36 95 Coccyzus rufigularis Coccyzus vetula 89 Coccyzus vieilloti Coccyzus merlini 58 100 50 changes Coccyzus longirostris 100

87

Figure 5.5. Phylogeny of the Old World “malkohas” and allies (tribe Phaenicophaeini).The tree shown is the majority rule consensus from the overall analysis, and is identical to the weighted analysis for Phaenicophaeini only.The position of Rhinortha chlorophaea is indicated in Figure 5.1. Alternative arrangements for Rhamphococcyx, Dasylophus, and Clamater are described in the text. Bootstrap values are shown above each node.Two samples of R. calyorhynchus, D. cumingi, C. cinerea, and C. pluvialis, respectively, had nearly identical sequences.

supported inasmuch as complete sequence data were obtained only for P. curvirostris. Two species, Sirkeer Malkoha Taccocua leschenaultii and Red-billed Malkoha Zanclostomus javanicus, are basal to Phaenicophaeus and are recognized as separate genera; they differ from other malkohas in the feathered face and straight bill with a single color. Yellow-billed Malkoha Rhamphococcyx calyorhynchus of Sulawesi and the crested malkohas Dasylophus of the Philippines generally did not group with other malkoha genera in our analyses, but instead branch off singly or together either before or after Clamator in most trees. In other trees, Rhamphococcyx and Dasylophus form a clade sister to Clamator. Although its placement is uncertain,

results for Rhamphococcyx are nonetheless surprising. Most systematic arrangements that recognize several genera of malkohas treat R. calyorhynchus and Chestnut-breasted Malkoha P. curvirostris as sister species in a single genus Rhamphococcyx. These two are morphologically similar and their geographic ranges are complementary. Whereas P. curvirostris is well known as a nesting cuckoo, only one nest of R. calyorhynchus has been described (Meyer 1879) and a fledgling was seen to be fed repeatedly by another species (Rozendaal and Dekker 1989). The fledgling malkoha was not necessarily reared in the nest of that species, insofar as songbirds can be attracted to a begging fledgling of another species and feed it repeatedly, especially when their own

90 The Cuckoos 20

Pachycoccyx audeberti Microdynamis parva Eudynamys s. scolopacea 100 Eudynamys s. melanorhyncha Eudynamys s. alberti Eudynamys s. cyanocephala (2) Urodynamis taitensis Scythrops novaehollandiae 100 Chysococcyx maculatus Chysococcyx xanthorhynchus (2) 76 Chysococcyx caprius (2) 100 Chysococcyx klaas 44 Chysococcyx flavigularis 100 Chysococcyx cupreus Chysococcyx cupreus Chysococcyx megarhynchus 85 Chysococcyx basalis 48 Chysococcyx osculans 62 18 Chysococcyx ruficollis Chysococcyx lucidus layardi Chysococcyx lucidus plagosus Chysococcyx meyeri (2) 14 Chysococcyx minutillus crassirostris 67 Chysococcyx minutillus poecilurus Chysococcyx minutillus barnardi 83 Cacomantis pallidus Cacomantis leucolophus 94 100 Cacomantis castaneiventris Cacomantis flabelliformis 73 Cacomantis sonneratii 67 Cacomantis merulinus 100 Cacomantis passerinus 100 Cacomantis variolosus “heinrichi” Cacomantis variolosus sepulcralis (3) Cacomantis variolosus variolosus Cercococcyx mechowii (2) 100 100 Cercococcyx olivinus 50 Cercococcyx montanus Surniculus dicruroides (2) 100 Surniculus velutinus (2) 95 Surniculus lugubris (2) Surniculus musschenbroeki 87 Hierococcyx vagans 96 Hierococcyx bocki 100 Hierococcyx sparverioides 96 Hierococcyx varius (2) 98 Hierococcyx hyperythrus 100 Hierococcyx pectoralis (2) 69 Hierococcyx fugax 100 Hierococcyx nisicolor (3) Cuculus c. clamosus Cuculus c. gabonensis (3) 93 Cuculus solitarius (2) 100 Cuculus poliocephalus (2) 88 Cuculus crassirostris Cuculus micropterus 82 Cuculus rochii 63 Cuculus gularis (2) 96 Cuculus saturatus (2) 100 Cuculus canorus Cuculus optatus (4) Cuculus saturatus 98 Cuculus canorus (5)

100

64 54

99 61

61

100

92

50 changes

Figure 5.6. Phylogeny of the Old World parasitic cuckoos (tribe Cuculini). Relationships within this group were identical in all analyses except for variations within Chrysococcyx, which are discussed in the text. A number following the taxon name indicates that multiple individuals were sequenced, and in general these individuals had identical or nearly identical sequences. Each of twelve C. canorus and C. saturatus samples had unique but similar sequences (four of which are shown here). In addition, the sequences from these two species did not form monophyletic groups (see chapter 6 species). Bootstrap values are shown above each node.

brood has recently failed and the birds are in the hormonal condition for parental care (Welty and Baptista 1988, Payne and Payne 2002). Nonetheless, this observation and the lack of information on the

nests of Dasylophus species are intriguing, particularly if Rhamphococcyx and Dasylophus form the sister group to Clamator. The reproductive biology of Rhamphococcyx and Dasylophus need to be

A molecular genetic analysis 91 investigated to determine if they are facultative or perhaps even obligate brood parasites. Perhaps just as important, the malkohas in these two genera are all east of mainland Asia and the sunda Shelf, in an area never connected to the Asian mainland (Lee and Lawver 1995, Hall, 1996), in the area originally described as the zoogeographic region east of Wallace’s line (Huxley 1868, Mayr 1994). The arboreal New World cuckoos Coccycua, Piaya and Coccyzus are closely related to the Old World malkohas and form a monophyletic group within the tribe Phaenicophaeini (Figure 5.5).This is in contrast to the account of Payne (1997b) which was based on an analysis of feather keratins (Brush and Witt 1983) suggesting a close relationship between Coccyzus and New World anis. Berger (1952, 1960) also questioned the idea that the coccyzine cuckoos were closely related to the malkohas, insofar as the two groups had a different pelvic muscle formula (AXYAm, vs ABXYAm). Sibley and Ahlquist (1990) grouped malkohas, Old World ground-cuckoos and couas with the Old World “Cuculidae” and apart from their New World family “Coccyzidae;” in fact none of these groups were sampled in their study.The mitochondrial evidence for including the arboreal New World cuckoos within Phaenicophaeini is strong and we refer to this clade as the coccyzine cuckoos. Within the coccyzine cuckoos the basal lineage is Coccycua. Little Cuckoo Coccycua minuta was called a Coccyzus by Vieillot in 1817 and a Piaya by Gray in 1846 and Shelley in 1891, and was described as a distinct genus Coccycua by Lesson in 1830. In the phylogeny (Figure 5.5) Coccycua includes two other species, C. pumila and C. cinerea, that were formerly listed as species of Coccyzus.Although C. minuta is like the Squirrel Cuckoos Piaya cayana and P. melanogaster in plumage color and pattern and in the long graduated tail, minuta is similar to Dwarf Cuckoo Coccycua pumila in the short, curved bill and rufous plumage (the rufous is restricted in pumila to the throat and breast), and pumila in turn is like Ash-colored Cuckoo C. cinerea in the shape of the bill and the rounded (not graduated) tail. C. cinerea was described as Coccyzus by Vieillot and as Piaya by Gray (Cabanis and Heine 1862). Although Dwarf Cuckoo C. pumila was more similar to C. cinerea than to other Coccyzus cuckoos in mtDNA cytochrome b sequence, Hughes (1997d ) compared only the cuckoos listed by Peters (1940) in

the genus Coccyzus, and did not include Coccycua minuta.When all these cuckoos are considered together, the molecular genetic data shows that the closest relatives of C. minuta are C. pumila and C. cinerea. The species pumila and cinerea were treated as the genus Micrococcyx by Ridgway (1912, 1916) based on their more rounded wing than that of other small New World cuckoos, and the genera Coccycua and Micrococcyx were considered to be separate based on a more rounded wing in minuta than in pumila and cinerea (Ridgway 1916). Shelley (1891) listed Coccyzus with Cuculus because they had long straight wings, and Piaya with Phaenicophaeus and other Old World malkohas because they had short curved wings. In Shelley’s view, minuta was a Piaya because it had short curved wings, while pumila was a Coccyzus because it had long straight wings.The molecular genetic results indicate that wing shape is not the most useful guide to the evolutionary relationships among these New World cuckoos. The large Squirrel Cuckoo Piaya cayana and Black-bellied Cuckoo P. melanogaster are each others’ closest relatives and form the next clade within the coccyzine cuckoos. The other New World coccyzine cuckoos are recognized here in a single genus Coccyzus. Most species are small and have long pointed wings. South American Pearly-breasted Cuckoo C. euleri and North American Yellow-billed Cuckoo C. americanus are sister species which are in turn sister to the closely related Mangrove Cuckoo C. minor and Cocos Cuckoo C. ferrugineus.The large cuckoos of the West Indies that have generally been referred to as Hyetornis, and the lizard-cuckoos Saurothera, appear to be derived from smaller cuckoos within the genus Coccyzus.The genus Coccyzus is monophyletic only if these large island cuckoos are included, and this is the basis for combining all these birds into a single genus. The genetic, morphological and distributional data suggest that these large cuckoos of the West Indies were derived from the continental Coccyzus cuckoos, perhaps birds like South American Gray-capped Cuckoo C. lansbergi, which some of them resemble in plumage color and pattern.The large island cuckoos are also similar to continental Coccyzus in the sound of their throaty and guttural rattling calls.The cuckoos of the West Indies evolved large body size and rounded wings (P8 is one of the shorter primaries)

92 The Cuckoos on the islands, a trend that other kinds of birds have undergone as well (e.g., Livezey 1993, 1998). Evolution of large body size of island cuckoos is also indicated in Mangrove Cuckoo C. minor, which is larger than its close relatives C. americanus and C. euleri, and in Cocos Cuckoo C. ferrugineus, an isolated island cuckoo that is larger than its allospecies, C. minor. Within the West Indies, the Hyetornis and Saurothera species groups apparently evolved by repeated invasions across the Greater Antilles, where two islands (Hispaniola and Jamaica) each have two species, one from each group. In the phylogeny of Clamator, C. coromandus rather than C. glandarius is basal to the other species. This result is inconsistent with a division of these cuckoos into two genera, Clamator and “Oxylophus” ( jacobinus and levaillantii ) (Wolters 1975–1982, Sibley and Monroe 1990).Those studies apparently considered the Asian species C. glandarius to be more similar to C. coromandus than to the two black-and-white species, but they did not consider discrete characters or the evolutionary branching sequence among these cuckoos. Perhaps the most significant result of the molecular phylogeny is the implied double origin of brood parasitism in the Old World cuckoos, a result also found by Aragón et al. (1999). The crested cuckoos Clamator are more closely related to the malkohas and coccyzine cuckoos in Phaenicophaeini than to other Old World brood parasites, indicating that these crested cuckoos evolved obligate brood parasitism independently of the other Old World brood parasites.This conclusion is supported by several anatomical and behavioral traits. (1) Brain size relative to body weight in Clamator is like that of the nesting cuckoos, and unlike the small brains of the other Old World brood-parasitic cuckoos. (2) Both Clamator and other malkoha genera have elongate rather than raised circular nostrils, and both have the same formula of thigh muscles, AEXYAm; in Phaenicophaeus pyrrhocephalus and Ceuthmochares an additional muscle is present (Berger 1960).The same groups have a double-notched sternum, unlike most other Old World brood parasites in Cuculini; however, several other anatomical characters that vary among cuckoo genera do not sort out along these same lines (Berger 1960, Seibel 1988, Hughes 1997d, 2000). (3) Egg size in relation to body weight is larger

in Clamator and the malkohas than in the other Old World brood-parasitic cuckoos, which lay eggs that are relatively small for their body size. (4) Clamator have a plumage crest, like the closely related Dasylophus malkohas and unlike the other Old World brood-parasitic cuckoos. (5) Clamator and the lizardcuckoos Coccyzus (Saurothera) share an unusual molt pattern in which wing molt progresses in a regular pattern that skips two feathers, with a sequence P6 9 - pause - 7 - 10 - pause - 8 - 5 (or - 5 - 8). (6) The calls and songs of Clamator are harsh repeated sounds, like the calls of malkohas but unlike the whistled songs of most Old World brood-parasitic cuckoos. (7) Clamator resemble the nesting Old World and New World malkohas in having a pair bond; the male assists in brood parasitism by accompanying the female to host nests. (8) Nestling Clamator like nestling malkohas do not evict eggs from the nest, as do other Old World brood parasites. (9) The begging posture of Clamator nestlings is unlike that of nestling brood-parasitic Cuculus and Chrysococcyx, which crouch in the nest and hold their head low over the back and do not flutter their wings, and do so only inconspicuously after they fledge. Young Great Spotted Cuckoos Clamator glandarius stretch their neck upwards in high-intensity begging, much as do nesting Black-billed Cuckoos Coccyzus erythropthalmus, Chestnut-bellied Malkohas Phaenicophaeus sumatranus, Red-capped Couas Coua ruficeps, and Coquerel’s Couas Coua coquereli (Appert 1970, 1980, Ong Kiem Sian, Figure 5.7). At other times they crouch in the nest (von Frisch and von Frisch 1967, M. Soler, in litt.). Begging postures of young Jacobin Cuckoos Clamator jacobinus also are upright with the neck stretched upwards at an early age (Liversidge 1971), and fledged Levaillant’s Cuckoos C. levaillantii conspicuously flutter their wings (Vincent 1934), unlike Cuculus and Chrysococcyx cuckoos. However, when nestling Rufous Hawk-cuckoos Hierococcyx hyperythrus beg, they raise the wings in display to their foster parents. (10) Young nesting cuckoos grow rapidly and fledge early, whereas brood-parasitic cuckoos grow slowly and let the parents provide prolonged parental care in the nest. Nestling growth in young C. jacobinus is slower than in nesting cuckoos such as Black-billed Cuckoo and Yellow-billed Cuckoo, but is faster than in Cuculus of the same adult body size. Nestling growth in Clamator jacobinus, for example, is

A molecular genetic analysis 93

Figure 5.7. Begging behavior of nestling cuckoos: a, Groove-billed Anis, Costa Rica; b, Black-billed Cuckoo, Michigan; c, Horsfield’s Bronze-cuckoo,Western Australia.

more rapid than in Cuculus solitarius, with Clamator reaching half of fledging weight (60 g) by day 7 rather than day 9. Like nesting cukoos, C. jacobinus and C. levaillantii also have shorter nestling periods (11–14 days) than other broodparasitic cuckoos of the same body size. Comparison with the Old World malkohas is not possible inasmuch as the details of their breeding biology are unknown. Old World brood-parasitic cuckoos, except for the crested cuckoos Clamator, share a common ancestor and are each others’ closest relatives in most of the best trees from our overall analyses.These Old World cuckoos are recognized as a tribe Cuculini (Figure 5.6). A sister relationship between the Thick-billed Cuckoo Pachycoccyx audeberti and the

koels (Eudynamys and Microdynamis) was consistently found in our analayses, but was weakly supported by the measure of bootstrap support.This clade was the sister group of all other Cuculini in most trees. In 2 of the 21 best trees from the overall analysis, however, the clade comprising Pachycoccyx, Microdynamis, and Eudynamys is basal to both Phaenicophaeini and other Cuculini, branching off just after Rhinortha, a result that would imply a fourth origin of brood parasitism in cuckoos.Additional data are needed to further test the monophyly of Cuculini. Channel-billed Cuckoo Scythrops novaehollandiae and Long-tailed Cuckoo Urodynamis taitensis of Australasia are each other’s closest relatives and form the next most basal lineage within Cuculini.

94 The Cuckoos The glossy cuckoos Chrysococcyx are monophyletic (Figure 5.6). The African and Australasian species form two well-supported clades and the Asian C. maculatus and C. xanthorhynchus appear to be more closely related to the African glossy cuckoos than to the Australian and New Guinea bronze-cuckoos. In some analyses, however, the two Asian species are the basal lineage within the genus. Long-billed Cuckoo C. megarhynchus, once recognized as a genus Rhamphomantis by itself, is a member of the Australian and New Guinea glossy cuckoo lineage within Chrysococcyx. The placements of C. basalis and C. oscualans were variable among analyses, with C. basalis sister to C. megarhynchus and/or C. osculans sister to the clade comprising C. meyeri and C. minutillus in some trees. Incomplete data for C. megarhynchus and C. ruficollis likely contribute to this uncertainty. Other brood-parasitic cuckoos with gray and brown plumage and barred underparts form a monophyletic lineage (Figure 5.6).This group includes the brush cuckoos Cacomantis, a clade comprising the African long-tailed cuckoos Cercococcyx and Asian drongo-cuckoos Surniculus, and a clade comprising the hawk-cuckoos Hierococcyx and the canonical cuckoos Cuculus. Realtionships within this entire clade were generally well-supported and no variations in tree topology were found in our analyses.The Hawk-cuckoos Hierococcyx are the sister group of Cuculus cuckoos, but not all birds that have been called “hawk-cuckoos” are each others’ closest relatives. Sulawesi Cuckoo Cuculus crassirostris, which was known as a “hawk-cuckoo”, is not closely related to either the H. fugax or H. sparverioides species complexes, but is more closely related to the more typical (non-hawk) Cuculus cuckoos such as C. micropterus and C. canorus. C. crassirostris was originally described as a Hierococcyx and it was long called a “hawkcuckoo” because it has broad, rounded wings and a broadly barred tail (Meyer 1879, Shelley 1891, Baker 1927, Stresemann 1940, Wolters 1975–1982), but its calls are mellow whistles like the calls of Common Cuckoo C. canorus, and its plumage below is barred not streaked; it is better known as Sulawesi Cuckoo. Wolters (1975–1982) listed Indian Cuckoo C. micropterus in his genus Hierococcyx based on its broadly barred tail, but C. micropterus is also in the Cuculus lineage, both in the molecular genetics’ results

and in its rhythmic whistled song. Drongo-cuckoos Surniculus and White-crowned Cuckoo Cacomantis (Caliechthrus) leucolophus were once thought to be related to koels Eudynamys (Peters 1940), if only because the adult plumage is black (as in adult male Common Koel). The genetic results indicate that both drongo-cuckoos and White-crowned Cuckoo are more closely related to Hierococcyx and Cuculus than to the koels. Drongo-cuckoos and Whitecrowned Cuckoo resemble Cuculus and Cacomantis in the step-graded whistles of their song (especially Fork-tailed Drongo-cuckoo S. dicruroides and Pallid Cuckoo Cacomantis pallidus), and a bubbling crescendo call (as in Black Cuckoo Cuculus clamosus and in hawk-cuckoos of the Hierococcyx fugax complex). The mitochondrial sequence data presented above provides a remarkably good resolution of the evolutionary relationships among cuckoo subfamilies, genera and species, and a framework for future analyses of cuckoo systematics and population biology. Nonetheless, our results should be considered preliminary, in that additional data and more thorough analyses of the currently available data are needed. Sequence data from the nuclear genome needs to be collected to test our conclusions and resolve some of the uncertainties noted above. Issues in need of further analysis include the relationships among cuckoo subfamilies; monophyly of Cuculini and the position of Pachycoccyx; and the relationships among Rhamphococcyx, Dasylophus, Clamator and other malkohas. At the species level, more data are needed for Carpococcyx viridis, Coua delalandei, some of the coucals Centropus, various malkohas Phaenicophaeus, and Chrysococcyx megarhynchus.A number of clades that were consistently found in our analyses were weakly supported by the measure of bootstrap value (Figures 5.1–5.6), and it is these clades that may be most likely to change as additional data become available. Finally, there are many opportunities for population level studies exploring geographic variation and species limits, particularly within the Old World brood parasites (Cuculini). The lack of reciprocal monophyly in the mitochondrial DNA of C. canorus, C. optatus and C. saturatus, for example, needs to be examined with a broad geographic sample of populations of these two species.

6 Species

What is a species? In his book On the origin of species, Darwin (1859, 1875) recognized a species as a morphologically distinct class of individuals. He avoided a formal definition of the term, and he even referred to “the vain search for the undiscovered and undiscoverable essence of the term species”. Earlier, when Linnaeus (1758) wrote Systema naturae and established systematic biology, he described species as morphologically distinct kinds.A more current idea recognizes a biological species as a class of individuals that interbreed with each other, at least when they live in the same population (Mayr 1963, Klein and Takahata 2002). In most kinds of birds it is clear whether individuals are members of the same species, where sex for sex and age for age the birds look alike within a population and among geographic populations. However, some populations live in different regions and the birds can differ somewhat in plumage but otherwise appear to be the same kind of bird.This geographical separation of populations can be a challenge to the biological species concept (Cracraft 1983). However, most biologists find reasonable and rewarding the notion of species being composed of populations with restricted gene flow between these and other populations (Avise 1994). In birds, if pairs that feed their young at the nest are mated, exclusively from birds that look different, they can be considered to be behaving as a distinct biological species. However, the brood-parasitic cuckoos do not live together in pairs near the nest as they have no nest. Males and females do not generally associate together and are not seen as pairs

except briefly when they copulate, or more often when the male feeds the female before they mate. In most brood-parasitic cuckoos it is impossible to observe mated pairs, and only when lucky we see the social associations that lead to mating. Only in the few brood-parasitic cuckoos where male and female visit a nest together (as do the crested cuckoos, especially Jacobin Cuckoo Clamator jacobinus, where the pair visits the nest of a host), do mated pairs live together. In southern Natal, where two color phases occur together, many pairs are assortative, the black-phase birds mated with black-phase birds and the white-bellied birds mated with white-bellied birds. This suggests some degree of like mating with like. Nevertheless, the songs of black-phase and white-phase birds in this region are the same. Also, further south in the Eastern Cape where white-phase birds are outnumbered by black-phase birds, all white-phase birds observed are mated with black-phase birds. The similarity of songs and the common mating between the two color phases indicates that these Jacobin Cuckoos are members of a single species. Perhaps most useful in identifying species of cuckoos is their song. Of course the birds do not sing to us, they sing to other birds, and they use song to attract a mate. Among cuckoos, the birds that share their songs and morphology are recognized as biological species. In the field we can test whether birds respond to playback of recorded song equally well to foreign songs and to local songs, and if they do so respond, this equal responsiveness indicates that they are the same species (Payne 1986, Alström and Ranft 2003). In addition to their appearance and song, another criterion of species distinctiveness

96 The Cuckoos is whether two forms live together without interbreeding: the test of sympatry. If they live together and do not interbreed, then the populations are distinct species (Mayr 1963). Although certain birds occasionally hybridize even when they are not each others’ closest rela-tives (e.g., orioles Icterus: Freeman and Zink 1995, Omland et al. 1999), interspecific hybrids are unknown in the cuckoos. Genetic information is the best source of data to estimate the phylogenetic relationships among species. But species pairs of birds differ from each other in degrees of genetic divergence. That is, we have no molecular genetic criteria that define a species or tell us which allopatric populations that differ in appearance, are distinct enough to recognize as species. In the molecular genetic comparisons (Figures 5.5, 5.6), some species pairs differ in genetic distance by several times that of other species pairs (perhaps some species in one lineage became extinct, so the lineages did not diverge at the same time and not all species pairs at present are equally old), and some cuckoos are more variable within a species (e.g. Zanclostomus javanicus, Chrysococcyx cupreus) than in some cases between species.The genetic distance between the species Cacomantis merulinus and C. passerinus is less than between populations within some species such as Chrysococcyx cupreus, but the songs of the first two cuckoos are distinct, whereas the songs and appearance of birds in populations of C. cupreus are all similar and they are regarded as a single species. Genetic information can tell us which kinds of birds are each others’ closest relatives and share a common ancestor. Genetic monophyly is a condition necessary to define a clade, but monophyly is not necessary or sufficient to define a species. Generally we expect members of a species to form a monophyletic class, but then also we expect the same for members of a clade at any higher systematic level such as a genus. In addition, gene monophyly is not a necessary criterion of a species, insofar as gene trees do not always coincide with species trees. In some cases, two species live together and do not interbreed, yet they do not have mutually exclusive sets of genes. Genes can be shared between recently diverged species when the genes in their ancestral population have not yet become differentially

extinct in each daughter species. In consequence, the evolutionary divergence of gene trees can lag behind that of species trees (Avise 1994, Klein and Takahata 2002).A case of this was found in the cuckoos, where Oriental Cuckoos Cuculus optatus. Himalayan Cuckoos C. saturatus, and common Cuckoos C. canorus do not consistently differ in mtDNA gene sequences (Figure 5.6). The overlap of genetic sequence between these two species may result from recent speciation and incomplete lineage sorting or unilateral extinction of genetic morphs that were present in their common ancestor. Lack of reciprocal genetic monophyly might also result from hybridization, but we know of no case of hybridization in the cuckoos (as mentioned in the species account for Yellow-billed Cuckoo). In these Cueulus species the birds differ morphologically in their songs, so they are distinct species. Except in this one case, mtDNA monophyly was the rule among cuckoo species that were compared in more than one individual. When populations differ consistently in morphology and in song, they are distinct species: together these are sufficient criteria to recognize species. On the other hand, there is no general standard measurement of genetic difference to say that populations are one species or more than one species. For this reason, morphological and biological criteria are more general criteria to use in recognizing a species, than are degrees of genetic differences between populations. With this rationale it was possible to use two criteria to recognize species as distinct. First, if two or more kinds of birds live together and do not interbreed, then they are considered distinct species. Second, if populations have the same calls and songs, they are considered one species, and if they have different calls, they are different species.

Songs and species Birds use songs in choice of mate, and males use songs in behavior towards competitors, chasing away other birds with the same song and excluding them from their territories, while ignoring birds with different songs. The response of males in attacking a tape recording of songs of their own species while ignoring songs of other kinds of birds is often used as a test of whether birds are distinct

Species 97 enough to be considered different species (Payne 1986). Females are more difficult to test because females often do not respond to song playback in the field, but when females do respond, they are as selective in response as the males (Searcy 1992). Female cuckoos have not been tested for their choice of song, and our use of song to indicate species identity is based on what is known about birds that do respond to song in this way, as well as on the response of male cuckoos to song.The reasoning is that females use song to choose a mate and the specificity of male song determines the set of potential mates. Among cuckoos it is reasonable to use song in the same way as the birds themselves, in recognizing a species when they hear the song. The songs and calls are remarkably similar through the range of a cuckoo species (Chappuis 1974, 2000).The Common Cuckoo Cuculus canorus “cúck-oo” is a standard through the Palearctic from Britain to China and Japan (allowing for songs out of tune when the breeding season ends). Songs of the Oriental Cuckoo C. optatus are also the same across a continuous expanse in northern Asia. In Africa the songs of the African Cuckoo sound the same from Gambia and Nigeria through Zambia and South Africa. Similarly, songs of Diederik Cuckoo Chrysococcyx caprius, Klaas’s Cuckoo C. klaas and African Emerald Cuckoo C. cupreus sound the same across the range of each species. From Western Australia to New Zealand, Shining Bronze-cuckoos C. lucidus are alike in their songs. The songs also sound the same to the cuckoos. Laura and I compared the response of cuckoos in Western Australia to the playback of cuckoo songs recorded in New Zealand, to songs of cuckoos in their own Australian population, and to songs of another species of cuckoo.The cuckoos did not respond to songs of Horsfield’s Bronze-cuckoo C. basalis, but they responded by approaching, calling and flying over the speaker to songs of Shining Bronzecuckoo from New Zealand as quickly as they did to songs of their own population. Although New Zealand cuckoos and Western Australian cuckoos differ in plumage with females in New Zealand looking like males in Australia, the birds are the same species on the basis of performance and response to their common songs.

The similarity in song across a wide geographic range of these cuckoos suggests that cuckoo song is innate and not learned. A behavior is said to be “innate” when a bird develops the normal behavior whether or not it has any contact with another individual of its own species (Lorenz 1935). Indirectly this term is sometimes taken to indicate a “genetic” basis. Mayr (1942) argued,“There is also no question that the song of parasitic cuckoos must have a strictly genetic basis, for the young cuckoo is obviously more exposed to the song of his foster father than to that of his own”. However, the term “innate” refers to the social conditions of development, and not to genetics. There is some evidence for innate song development in cuckoos. A young Channel-billed Cuckoo was reared in captivity and developed normal adult calls (Goddard and Marchant 1983). Species songs are nearly the same across a wide range of nesting cuckoos as well, as they are in Yellowbilled Cuckoos Coccyzus americanus across North America. Call development may be innate in Greater Roadrunner Geococcyx californianus, as hand-reared birds that do not hear their own species have adult calls like the calls in wild adults (Whitson 1971), and a hand-reared Smooth-billed Ani Crotophaga ani developed normal calls like those of wild birds (Quinn and Startek-Foote 2000). Because their songs are similar across a species range of cuckoos, we use song as a guide to the species of the less well-known cuckoos to determine the biological limits of species. If songs are the same in geographically separated populations, then the birds are likely to be conspecific; and if songs are different (allowing for song repertoires), then the populations are likely to be two or more species. If the birds differ both in morphology and in song, then the birds are distinct species.This guideline has led to the discovery and recognition of several new species of birds by their songs in recent years (Alström and Ranft 2003). Using their song as a guide, African Cuckoo Cuculus gularis is a species distinct from Common Cuckoo C. canorus.African birds sing with an accent on the second note of “cuckóo,” and this note rises rather than drops in pitch from the first note. Palearctic birds sing with an accent on the first, higherpitched note; the familiar “cúck-oo.” In playback

98 The Cuckoos tests in the field, males were attracted to their own kind of song, but not to the song of the other species (Payne 1986). In addition, the birds differ in morphology—in adult and juvenile plumage, in bill color (yellow in African Cuckoo, black in Common Cuckoo) and in bill width, wide at the base in African Cuckoo and narrow in Common Cuckoo (Payne 1977a).These cuckoos were indifferent to the songs of the other kind of cuckoo, and their songs are different.We use this difference in their songs as a standard to recognize other species of cuckoos. Song recordings have identified more species of cuckoos in the Old World than was done before their songs were studied. In a few cases, their distinctive songs were identified only in the past few years. (1) The number of hawk-cuckoo species recognized in the Hierococcyx fugax complex has increased from one (Salvadori 1874, Shelley 1891, Peters 1940, King and Dickinson 1975) to two (Payne 1997b) or three (King 2002), or even four.These hawk-cuckoos were described as separate species in the early 19th century, the latest one by Gould (1856) at a time when populations with any morphological differences were described and published as separate species.These hawk-cuckoos were then most dramatically combined into a single species by Salvadori (1874), who included not only nisicolor, hyperythrus and pectoralis with H. fugax but also sparverioides and varius within this species. He did not include H. vagans or its synonym H. nanus in his review, which he wrote before he worked at Leiden (where the holotype of C. vagans is kept) and at the British Museum (where Hume’s type series of H. nanus is kept) (Shelley 1891, Warren 1966,Violani et al. 1997). Shelley (1891) recognized H. varius, H. sparverioides and H. bocki as distinct species, but he combined the forms nisicolor, hyperythrus and pectoralis and even H. vagans with H. fugax.The larger forms were considered subspecies of H. fugax by Mayr (1938a), on the basis of being geographic counterparts. Only after their songs were recorded and found to co-vary consistently with the plumage forms were the four hawk-cuckoos Hierococcyx fugax, H. nisicolor, H. hyperythrus and H. pectoralis again recognized as distinct species. Songs of Philippine Hawk-cuckoo Hierococcyx pectoralis (“wheet wheet wheet . . .”) are distinct from songs of Javan Hawk-cuckoo H. fugax (a shrill “gee-

whizz, gee-whizz”) (Payne 1997b). Songs of Rufous Hawk-cuckoo H. hyperythrus in northern China, Russia and Japan are even more distinct, a loud harsh call “wee wee-pit” (Figure 6.1). The songs in Japan are so different from songs in the Himalayas that I suspected a recording problem and excluded the Japanese songs in an earlier review of cuckoo species (Payne 1997b), but P. Alström and B. King have recorded the same songs in northeastern China. Songs of Whistling Hawk-cuckoo H. nisicolor in the Himalayas are similar to songs of H. fugax in the Malay Peninsula and the Malay Archipelago, but there are differences. Southern birds H. fugax have lower-pitched notes than northern birds H. nisicolor, with the first note loudest between 3 and 4kHz in H. fugax and the first note loudest between 4 and 5kHz in H. nisicolor. In addition, in southern birds H. fugax the first note in the song rises in pitch through its length; northern birds H. nisicolor hold the first note on one pitch for most of its length before it rises. In the recordings available, six H. fugax in southern Thailand and the Malay Peninsula (NSA 15709, 32612, 34934, 41317; Scharringa 1999) have songs lower in pitch and with the first note sliding upward in pitch, compared with northern birds H. nisicolor (Sichuan, P.Alström; Guangdong, B. King; northern India, P. Singh; Bhutan, S. Connop 1995; northern Thailand, P. Round; Figure 6.1).Across the species complex, northern birds hold the first note on one pitch the longest time and Philippine birds hold it the shortest time. In addition to these songs, the four species of hawk-cuckoos give a long call that rises and then falls in pitch. Northern birds differ in wing shape as well as in song from southern birds. The outer primaries are longer than the inner primaries in the migratory cuckoos in northeast Asia (H. hyperythrus) and the Himalayas (H. nisicolor); the outer primaries are shorter and the wing is more rounded in the cuckoos of the Malay Peninsula and Greater Sunda Islands (H. fugax) and the Philippines (H. pectoralis).The species accounts describe other morphological differences. (2) The small southern Dark Hawk-cuckoo H. bocki and the large northern Large Hawk-cuckoo Hierococcyx sparverioides have different songs, perhaps more different from each other than either is from Common Hawk-cuckoo H. varius (Figure 6.2).

Species 99

Figure 6.1. Hierococcyx fugax complex: a–c, Rufous Hawk-cuckoo H. hyperythrus (a, Nagano, Japan, LNS 08259; b, Japan, LNS 08260; c, Heilongjiang, China, P.Alström); d–f,Whistling Hawk-cuckoo H. nisicolor (d, Sichuan, P.Alström; e, Bhutan, Connop 1995; f, Khao Yai NP,Thailand, P. Round); g–i, Javan Hawk-cuckoo H. fugax (g, Krabi,Thailand, Scharringa 1999; h, Ampang, Malay Peninsula, NSA 94048; i, Ampang, NSA 15709); j–k, Philippine Hawk-cuckoo H. pectoralis (j, Mt Katanglad, Mindanao, Philippines, NSA 44887; k, Baralatan, Mindanao, Philippines, NSA 34307).

Figure 6.2. Larger hawk-cuckoos: a, Common Hawk-cuckoo Hierococcyx varius (N Pakistan, P. Alström); b, Large Hawk-cuckoo H. sparverioides (Uttar Pradesh, India, P. Alström); c, Bock’s Hawk-cuckoo H. bocki (Malay Peninsula, White 1984).

Insofar as the two differ with no overlap in size, adult plumage color and juvenile plumage as well as in voice, they appear to be distinct species, as Shelley (1891) concluded from their morphological differences. The molecular phylogeny indicates that H. bocki is basal to these other two hawk-cuckoos, and that bocki and sparverioides are not a monophyletic set and so are not conspecific. (3) The northern and southern Asian drongocuckoos formerly were thought to comprise a single species, Surniculus lugubris, as in Peters (1940) and

Payne (1997b). Nevertheless, songs of Asian birds and Philippine birds were recognized as different by several field observers, and their plumage differs as well. The juvenile plumage of the Philippine Drongocuckoos S. velutinus is black with white spots in the Asian birds, uniformly rufous in southern Philippine S. velutinus (although black with white spots in the northern Philippines) and, the adult plumage has a glossy mantle and forked or square tail in Asia, and a velvet mantle and square tail in the Philippines. Songs of drongo-cuckoos differ in the number of notes, in

100 The Cuckoos change in pitch within the notes, and between notes in a song. Songs in southern Asia are usually of six notes sung in arpeggio,“one, two, three, four, five, six” rising smoothly up the scale (Legge 1880, King and Dickinson 1975). Songs in the Philippines have 8 to 10 notes, given more rapidly in a series, and the notes are closer to one pitch (Figure 6.3). Moluccan Drongo-cuckoo S. musschenbroeki in the northern Moluccas and Sulawesi differ from Philippine birds in song with a long series of notes: each note drops in pitch in the Moluccas, whereas each note rises in Philippine birds. Finally, songs of many drongo-cuckoos in mainland Asia from the Himalayan region, eastern Assam (Dig Boi) and the Chin Hills of Burma to northern Vietnam and China do not progress smoothly up the scale. Rather, the second note is no higher than the first, especially at the beginning of the note (sometimes the end of the second note is higher than the end of the first). These birds are large and have a deeply forked tail and are Fork-tailed Drongocuckoos S. dicruroides. In contrast, drongo-cuckoos of western Assam, central Thailand, the Malay Peninsula, Sri Lanka, Sumatra, Borneo and Palawan have songs

that rise uniformly through the series of notes, and the individual notes are more mellow in tone, less strident and piping, than notes of Fork-tailed Drongo-cuckoos.These more lowland and southern birds are Square-tailed Drongo-cuckoos S. lugubris. The occurrence of both kinds of songs in Assam suggests sympatry of two species of drongo-cuckoos. And in Sri Lanka the presence of both square-tailed and fork-tailed birds during the breeding season suggests two species; only the songs of S. lugubris appear to have been recorded there. Birds were not recorded as they sang in the field and then captured or collected to confirm a match of song and morphology, but in the information available on songs and morphology (tail shape), it appears that there are two species of drongo-cuckoos in India and southeast Asia.The two drongo-cuckoos overlap in their breeding areas.The extent of geographic overlap and morphological differentiation of these two cuckoos that are best identified by their songs remains to be worked out. (4) Indian Gray-bellied Cuckoo and rustybellied races of Plaintive Cuckoo (Cacomantis

Figure 6.3. Drongo-cuckoos: a–f, Fork-tailed Drongo-cuckoo Surniculus dicruroides (a, Himachal Pradesh, P. Singh; b, Kathmandu Valley, Nepal, NSA 12449; c, Dig Boi oilfields, Assam, NSA 100506; d, Mt Victoria, Chin Hills, Burma, P. Alström; e, Hainan, P. Alström; f, Ho Ke Go, Vietnam, NSA 46844); g–m, Square-tailed Drongo-cuckoo S. lugubris (g, Kaziranga,Assam, NSA 15022; h, Sri Lanka, Deepal Warakagoda; i, Huai Kha Khaeng,Thailand; j,Ampang, Malay Peninsula, NSA 31616; k, Sumatra, NSA 06806; l, Sabah, NSA 36035; m, Palawan, NSA 17837); n–o, Philippine Drongo-cuckoo S. velutinus (n, Mindanao, NSA 65939; o, Luzon, NSA 16435); p–r, Moluccan Drongo-cuckoo S. musschenbroeki (p, Sulawesi, NSA 68079; q, Sulawesi, LNS 32990; r, Halmahera, NSA 32969).

Species 101 passerinus and C. merulinus) in southeast Asia, Malaysia, Indonesia and the Philippines differ in song. Indian birds C. passerinus give a “piteer” call for their rising song, while the southeast Asian birds C. merulinus give “eat more froueet”; and the two cuckoos also differ in the cadence call (Figure 6.4). Insofar as their songs and morphology both differ, these two sets of populations are now considered distinct species, in contrast to reports that were made before good samples of songs were compared (Payne 1997b). (5) Three small Cuculus that are similar in adult plumage now are recognized as distinct species because they differ in juvenile plumage, in song, and (at least in the two species that were examined) in molecular genetics that indicate each is more closely related to a different cuckoo species than to each other.The three previously were recognized as a single species, “Lesser Cuckoo” C. poliocephalus (Peters 1940).The characteristic plumages of adults and juveniles are described in the species accounts, and the phylogenetic relationships are illustrated in chapter 5. The songs are illustrated for the three species, Madagascar Lesser Cuckoo C. rochii, Asian Lesser Cuckoo C. poliocephalus and Sunda Lesser Cuckoo C. lepidus in Dowsett and Dowsett-Lemaire (1980),Wells (1982), Becking (1988) and King (2005), and the song differences between all three species first were pointed out by Dowsett and Dowsett-Lemaire (1993). In other cuckoo species, the songs are the same between populations that are geographic counterparts. This similarity in song indicates that the birds would respond behaviorally to each other. Several sets of geographically replacing forms of brood-parasitic cuckoos in the Old World have nearly identical songs.

(1) Indian Cuckoos Cuculus micropterus in northern and southern Asia have nearly identical songs except for a slightly lower pitch in the northern birds, which are often larger but overlap in size with the southern birds. (2) Common Koels Eudynamys scolopacea throughout Australasia have similar songs,The koels are sometimes split by systematists into two or more species (“E. melanorhyncha”, “E. orientalis” and “E. cyanocephala”) by bill color and female plumage. Among these cuckoos each bird has a repertoire with more than one call (“ko-el”,“ke-woo” or “who are you” and “kwow kwow-kwow-kwow” in Asia (E. s. scolopacea); “kuOw” or “whooo” and “oeoeoeoeoeoe” in Sulawesi (melanorhyncha); “cooee” or “kooeel” and “wurroo” in Australia (cyanocephala). Apparently these are two sets of calls given in different contexts and each have their counterparts in each region. More field recordings are needed to test whether the corresponding calls are the same. (3) Northern and southern populations of Brush Cuckoos Cacomantis variolosus (combining the graybreasted and the rusty-breasted or Indonesian cuckoos “C. sepulcralis” ) have a similar rising song (“where’s the tea?”) and cadence song (“fear-fear”). The songs are not identical in northern and southern birds, and the evidence for the birds being conspecific involves the morphological intergradation of populations that live in the Moluccas and New Guinea.The rising song is longer and lower in pitch in the larger southern birds, yet when the birds are excited they shift into a high pitch that is nearly identical in Australia with the rising song of birds in the Malay Peninsula and Sulawesi (Figure 6.5).The

Figure 6.4. Plaintive cuckoos Cacomantis passerinus and C. merulinus: a–d, Gray-bellied Cuckoo C. passerinus, (a, b, “kateer” (a, Himachal Pradesh, Pratap Singh; b, Kaziranga, Assam, NSA 14045); c, d, “pee-pipee-pee-pipee” (c, Thekkady, Kerala, NSA 15016; d, Manas, Assam; NSA 15017)); e–h, Plaintive Cuckoo C. merulinus: (e, f, “eat more froueet” (e, Mt Victoria, Chin Hills, Burma, Per Alström; f, Sulawesi, Smith 1993a); g, h, cadence song (g, Mt Victoria, Burma, Per Alström; h, Mindanao, Philippines, NSA 31714)).

102 The Cuckoos descending “fear” calls may differ between areas, and additional recordings are needed to compare the behavior of northern and southern birds and test whether they are conspecific. (4) In the Little Bronze-cuckoo Chrysococcyx minutillus geographic complex, two or more species are sometimes recognized (Parker 1981). C. minutillus occur in many geographic forms from southeast Asia through the Malay Archipelago and the Philippines to New Guinea and northern and eastern Australia.The songs of C. m. minutillus, C. m. poecilurus, C. m. crassirostris and other geographic populations appear to be identical (Figure 6.6). The plumages of C. m. minutillus and C. m. poecilurus intergrade where their ranges come together in northern Australia. Because their trilled songs are similar throughout their overall geographic range, the green-backed forms (minutillus and others) and rusty-backed forms ( poecilurus and others) appear to be a single species, and the morphologically distinct pied form crassirostris has the same trilled song. None of the forms are known to occur together in their breeding range, and song as well as morphologically intermediate individual birds are sufficient criteria to consider these birds as a single biological species.

(5) African Black Cuckoos Cuculus clamosus differ remarkably in appearance across their range in Africa. Birds breeding in southern Africa are nearly completely black in plumage, and birds in the forests of central Africa are rufous on the throat and are barred and appear more or less white on the underparts. These birds have the same “papa’s boy” slurred whistles for their song, and birds in intermediate plumages occur where the extreme color forms come into geographic contact, so these cuckoos are considered a single species both on the basis of their song and on the basis of the morphologically intermediate individuals, even though they are difficult to spot while mating and there is no observational evidence of whether mated cuckoos tend to be similar to each other in appearance. Song similarity as a standard of species identity also applies to nesting cuckoos.African Black Coucal Centropus grillii have been considered conspecific with Lesser Coucal C. bengalensis of Asia and Madagascar Coucal C. toulou of Madagascar, but their voices differ as do their size and plumage, so they are now regarded as distinct species. The C. grillii song is “kop-kop” repeated at 2-sec intervals, sometimes in a long series, each note rising and

Figure 6.5. Brush Cuckoos Cacomantis variolosus: Northern populations, a–b, “where’s the tea” (a, Gombak, Malay Peninsula, NSA 32539; b,Tegah, Sulawesi, NSA 32750); c–d,“fear-fear” (c, Lore Lindu NP, Sulawesi, NSA 13137; d, Mt Katanglad, Mindanao, NSA 41264); southern populations, e–g,“where’s the tea” (e, f, FitzRoy River,Western Australia, Buckingham and Jackson 1990; g, Mt. Hagen, New Guinea, NSA 38418); h–i, “fear-fear” (h, Morobe, New Guinea, NSA29701; i, Lamington NP, Queensland, NSA14259). Note: time scale 0–4s correct for a, b, e–g; time scale to read 0.8s for c, d, h, i.

Species 103

Figure 6.6. Little Bronze-cuckoo Chrysococcyx minutillus: a–g, descending trill (a, jungei, Sulawesi, Ben King; b, crassirostris, Tanimbar, NSA 46764; c, poecilurus, Timor, NSA 72300; d, poecilurus, Timor, Ben King; e, minutillus, Kimberley Range, WA, Graeme Chapman; f, poecilurus, Cardwell, QD, Buckingham and Jackson 1990; g, barnardi, Brisbane, QD, Barry Morgan); h–n, whistled song (h, peninsularis, Malay Peninsula, Scharringa 1999; i, Sulawesi, Ben King; j, Flores, NSA 72301; k, Kimberley Range,WA, Graeme Chapman; l, Darwin, NT, Buckingham and Jackson 1990; m, Cardwell, QD, Buckingham and Jackson 1990; n, Brisbane, QD, Barry Morgan).

falling in pitch within the range 0.8–1.0kHz, and the notes are repeated 6 per sec for a couple of seconds with no change in pitch. C. toulou gives a muffled series of 5–10 hoots “toulou toulou toulou . . .” decreasing in volume, short notes given c. 6 per sec at 0.6kHz, and it also gives water-bottle call duets, one bird in a “toulou” series, the mate joining with a higher-pitched series at 14 notes per sec dropping in pitch and slowing to 8–10 notes per sec; while the Asian Lesser Coucal has paired notes lower in pitch, and not mellow and modulated as in the African Black Coucal. In Figure 5.3, these three coucals are not each others’ closest relatives, insofar as the species complex includes the Philippine Coucal C. viridis which is sympatric with C. bengalensis in the Philippines. The differences in song are consistent

with morphology and results of molecular genetics which show that these coucals are distinct species. In other cases, sympatric, non-interbreeding coucal species may sound alike.The pitch of the call differs between the sexes; the larger female has a lower-pitched call, noticed in duetting pairs. Taking into account the pitch differences between sexes and the variation in calls given by a single bird, the geographically remote populations of Greater Coucal Centropus sinensis in southern Asia and the Andaman Islands appear to be the same species, although these birds differ in size and sometimes in plumage.The sex of the bird was not noted in field recordings, and measurements of the pitch of calls in the coucal species accounts may not be comparable where a male was recorded for one species and a female for

104 The Cuckoos another species. Chappuis (2000) described interspecific song duets between Senegal Coucal Centropus senegalensis and Blue-headed Coucal C. monachus, and song duets between Black-throated Coucal C. leucogaster and C. monachus, and Brosset and Erard (1996) described duets between Gabon Coucal C. anselli and C. monachus. Also the calls of C. leucogaster in Nigeria and C. anselli in Congo are similar to each other and to calls of C. monachus, and coucals sometimes react to the playback of recorded calls of other coucal species much as they do to calls of their own species (Dowsett and Dowsett-Lemaire 1993). Song of yellowbill malkohas Ceuthmochares vary together with the plumages of malkohas across Africa and this covariation indicates there are two species,

rather than one species. Across West Africa from Guinea-Bissau to Ivory Coast, Chattering Yellowbill C. aereus have a monotonic chatter that quickens to a rattle; from Nigeria and Cameroon eastward to western Kenya, they have a similar quickening rattle but the song often begins with a high pitch an octave above the underlying basic pitch and it drops more than an octave to the terminal rattle. In contrast,Whistling Yellowbill C. australis from coastal Kenya to South Africa have a song with a set of long whistles interposed between the introductory notes and the terminal rattle (Figure 6.7). In southeast Asia, Chestnut-breasted Malkohas Phaenicophaeus curvirostris present a problem of species distinctiveness. The plumage pattern, bill

Figure 6.7. Ceuthmochares African yellowbills. a-f, Chattering Yellowbill C. aereus (a, Bijagos Archipelago, GuineaBissau, C. R. Barlow; b, N’Douci, Ivory Coast, Chappuis 2000; c, Mt Bengoué, Gabon, Chappius 1974; d, Obale, west of Lake Kivu, NSA 80609; e, Nyungwe Forest, Rwanda, NSA 30638 cf. Dowsett-Lemaire 1990; f, Kakamega forest,W Kenya, NSA 48766); g-i, Whistling Yellowbill C. australis (g, Sokoke forest, E. Kenya, C. R. Barlow; h, Mrima hill, SE Kenya, NSA 18022; i, Kruger NP South Africa, Gillard 1987).

Species 105 pattern, and the shape of the nostril and groove on the bill of these malkohas differ among islands across the western Sumatran islands and Greater Sunda Islands to Palawan. It is unknown whether songs differ between these malkoha populations. Questions about the species of cuckoos remain to be investigated. Fieldwork in India and Southeast Asia is needed to test whether both kinds of song occur in extensive sympatry in the drongo-cuckoos Surniculus dicruroides and S. lugubris. In Australasia, questions of song and species distinctiveness need to be studied in detail in the Koels Eudynamys scolopacea, in the hawkcuckoos Hierococcyx sparverioides, H. bocki and H. varius, in Brush Cuckoos Cacomantis variolosus and in Little Bronze-cuckoos Chrysococcyx minutillus. Finally, additional field studies are needed on several Asian Cuculus cuckoos.The songs of three species, Oriental Cuckoo Cuculus optatus, Himalayan Cuckoo C. saturatus and Sunda Lesser Cuckoo C. lepidus differ from each other; their plumages differ only in size and color of the underparts (and juvenile of C. optatus differ in plumage from the other two). Here they are recognized as three distinct species, as in King (2005).

Speciation New species originate when their common ancestral populations are isolated over time and evolve independently, with genetic differences accumulating to the degree that the populations become reproductively independent (Mayr 1942, 1963). Geological changes that divide birds into geographically isolated populations (vicariants), and dispersal of individuals from a region into new areas where the birds breed separately from their source populations, are two processes that can isolate these populations (Zink 1996, Klicka and Zink 1997, Zink et al. 2000). On the other hand, speciation in the brood-parasitic cuckoos might involve another process, where speciation of the cuckoos is closely linked to the speciation of their hosts. This section develops the idea that allopatric speciation accounts for the differentiation of both nesting cuckoo species and of the brood-parasitic cuckoos.

Geographic speciation model The idea of recent speciation occurring in vicariant populations that live in isolated areas leads to a pre-

diction of geographic complementarity, much as in Mayr (1963).The extent to which regional geological changes have led to speciation is tested by finding common biogeographic patterns across different lineages (Cracraft 1985, 1988).Vicariance biogeography is based on a congruence of historical branching sequences among clades. The reasoning is that regions became geographically disjunct at the time of unique geological events that shaped the face of the earth’s landscape, and the isolated populations diverged into species (Haffer 1978, 1992, 1997, Cracraft 1982, Cracraft and Prum 1988, Marks et al. 2002). The effect of dispersal on geographic isolation is clear on islands that were never connected to continental areas, and closely related species on islands and mainland were never in contact. These birds must have dispersed from mainland to island. In many cuckoos the geographical distribution of closely related species is complementary, where the species breed only in neighboring regions where the other does not occur. Cuckoo species with distributions consistent with this model include brood-parasitic hawk-cuckoos Hierococcyx hyperythrus, H. nisicolor, H. fugax and H. pectoralis in Asia, the drongo-cuckoos Surniculus in southern Asia, the plaintive cuckoos Cacomantis merulinus and C. passerinus in southern Asia, and the long-tailed cuckoos Cercococcyx olivinus and C. montanus in Africa. Other sets of closely-related species that live in geographically separate areas include the nesting cuckoos such as lizard-cuckoos Coccyzus (“Saurothera” and “Hyetornis” ) in the West Indies, the ground-cuckoos Carpococcyx in southern Asia, ground-cuckoos Neomorphus in the New World tropics, and coucals Centropus in the New Guinea region. In South America, the ground-cuckoos Neomorphus live in regions that are now separated from each other by large rivers. Haffer (1977) proposed that ground-cuckoos vary in morphology in the way predicted by a series of climate changes with dry periods that separated the formerly continuous continental forests, with the forest patches persisting in refugia that became separated by large expanses of savanna habitat. First a speciation event separated the ancestors of radiolosus and geoffroyi from the ancestors of pucheranii and rufipennis, then each of these groups speciated once more. Ground-cuckoos could not cross from one forest refuge to another over broad

106 The Cuckoos rivers or across broad areas of savanna. The phylogeny of Neomorphus is consistent with this model. N. geoffroyi has subspecies both north and south of the upper Amazon in Brazil; in the lower Amazon ground-cuckoos are not known north of the river in Pará and Amapá; they occur south of the river. N. pucheranii occurs on both banks of the upper Amazon, with different subspecies on each bank as far towards the headwaters as they have been sampled. N. radiolosus is separated from other groundcuckoos by the Andes, a distribution that suggests the cuckoo was isolated from N. geoffroyi with the rise of the northern Andes. N. rufipennis occurs in the northern drainage basin east of the Orinoco and west of the Courantine rivers, and its restricted distribution suggests the Courantine is a barrier to dispersal into Suriname; the species also lives on the upper Rio Branco in the northern Amazon basin. Dispersal in a continental area is likely to lead to gene flow and a breakdown of local differentiation. Wallace (1853) observed that certain mammal species replace one another on opposite banks of the large rivers in Amazonia. A river of a kilometer or more in width may prevent dispersal and genetic interchange of forest birds with limited abilities of flight.At the same time, the continuous forests along a river may allow dispersal and gene flow upstream where the river narrows in the headwaters and is no longer a barrier to dispersal, and the birds can disperse over or around the river and downstream again on the opposite side. Wallace noted, “Toward their sources, rivers do not form a boundry between distinct species.” In time, populations may become separated when the river changes course, and populations on each side of the river become isolated. The idea is that populations differentiate into species when barriers arise at the periphery of their range, but these barriers are effective only where they prevent dispersal, and the isolation of populations is not caused by dispersal. Dispersal would lead to speciation only where occasional movements of individuals leads to the establishment of new geographic isolates across regions that more often bar dispersal. Changes in climate during the Pleistocene, the rise of the Andes, and changes in river courses in the western Amazon with a breakdown of distinction between flooded várzea forest and non-flooded tierre firme forest repeated over time by means of channel

migration and sedimentation of the meandering rivers (Räsänen et al. 1987) may all have affected these Amazonian birds.The changes might isolate populations at one time, and at other times they might allow them to disperse, interbreed and exchange genes again. In Neomorphus there is biogeographic support for a vicariance model of speciation. Nevertheless it is unclear whether the geographic isolation of populations in separate regions of relict forests when these were separated during a Pleistocene dry period was involved in the initial separation and speciation, or whether the forest remained intact over a long period and the initial isolation of the populations occurred when a new broad river divided a formerly continuous forest. Because more than one geological process can account for the present-day distribution of the ground-cuckoo species, neither process is excluded, and the ecogeographical processes that were responsible for the speciation of these ground-cuckoos remain uncertain. Recent history witnesses rivers as barriers to dispersal in the short term, if not in speciation. Ground-cuckoos on Barro Colorado Island disappeared a few decades after the island was isolated from mainland populations. Birds fly, and they disperse, and as a result, dispersal is important in the distribution of many species. Many cuckoos are strong fliers and move thousands of kilometers each year. In the New World, the closely related Yellow-billed Cuckoo, Pearl-breasted Cuckoo and especially the insular Mangrove Cuckoo Coccyzus species likely trace their geographic distribution in the breeding season to dispersals that are independent of major geological change, insofar as all three cuckoos are seasonally migratory. In the Old World, the hawk-cuckoos Hierococcyx hyperythrus and H. nisicolor migrate from their northern breeding areas into tropical regions where other hawk-cuckoos H. fugax live year-round. And yearround resident malkoha species live on small islands well offshore as well as on the mainland of Asia and the Greater Sunda Islands. The coucals Centropus are of interest for models of allopatric speciation. These birds live in a region of active geological history where both geological processes of landscape change and the biological process of dispersal have been involved in isolation and speciation. Movements of the Australian and

Species 107 Philippine-Caroline continental plates present a complex background to a vicariance model of speciation. In the Tertiary, Australia and southern New Guinea were on a shallow sea plate along with the Aru Islands. Halmahera and the other islands of the northern Moluccas were not part of this block. The Banda Sea arc including the Kai Islands as its eastern bend had a separate origin from the New Guinea plate; these western islands were never connected with Australia and New Guinea (Hall 1996, 1998, Milson et al. 1996). In the genetic phylogeny, Goliath Coucal Centropus goliath of the northern Moluccas is not closely related to Greater Black Coucal C. menbeki of New Guinea, as proposed by White and Bruce (1986). Rather, C. goliath is most closely related to Pheasant Coucal C. phasianinus, which occurs in Australia, New Guinea, the Kai Islands (C. p. spilopterus) and Timor (C. p. mui). The geological history of land masses suggests that dispersal as well as a vicariance isolation of populations by land movement led to speciation in these coucals. The same argument applies to the question of whether C. goliath is closely related to either species. In both cases a dispersal would be required to get the birds from New Guinea to the northern Moluccas. We expect that episodic dispersal has led to the establishment of populations in new areas, where they diverged into species that differed from their source population. Even short-winged coucals disperse, as one (African Black Coucal Centropus grillii ) is a seasonal migrant.There is historical evidence of success in dispersal on a broad geographic scale. Krakatau Island, between Sumatra and Java, blew up in a devastating volcanic explosion in 1883 and all land birds and land plants were exterminated (Dammerman 1992, Thornton 1996). Lesser Coucals C. bengalensis appeared there by 1908 and nested soon afterwards; Greater Coucals C. sinensis were common by 1919. Greater Coucals occur on offshore islands in the Indian Ocean, and they have historically spread into northern India. Common Koel Eudynamys scolopacea have colonized offshore islands near New Guinea after the islands lost all their animals in volcanic explosions about 300 years ago. In India the resident Sirkeer Malkoha Taccocua leschenaultii spread through dry scrubland as they dispersed along irrigation canals in the past century. In the New World, Groove-billed Ani Crotophaga sulcirostris established themselves in

Florida after a hurricane in 1926, while other populations have come, gone and come again in dispersal with loss in storms and recolonization from other islands in the West Indies, and in Central America the anis enlarged their distributional range when forests were cleared and the birds dispersed from one open habitat to another.

Host-parasite cospeciation model The close association between a brood parasite and its host species might provide an opportunity for reciprocal evolution, or coevolution. The cuckoo evolves mimicry of its own egg to the host egg, and the host might evolve to recognize a cuckoo egg and reject it from the nest. Or the parasite-host association might lead to speciation of cuckoo populations matched with the speciation of their host populations, in parallel speciation or cospeciation. The brood-parasitic cuckoos as a group are not hostspecific, as far as we know. Only a few cuckoo species are restricted to a single kind of host. In many cuckoos the hosts are not well known, and further fieldwork may reveal a broader range of their host species. Perhaps the most host-specific cuckoo is Little Bronze-cuckoo Chrysococcyx minutillus which uses flyeaters of the genus Gerygone. These cuckoos have not undergone speciation in parallel with their hosts. The problem of host specificity and speciation in these birds involves determining not only the songs and genetic histories of cuckoos but also the species limits in Gerygone. Sulphur-bellied Flyeater G. sulphurea in the Malay Peninsula, Java and Flores comprise a single species with no morphologically distinct geographic populations or subspecies, whereas the cuckoos differ geographically in these regions. Among the cuckoos, the bronze C. m. poecilurus parasitizes Large-billed Flyeater G. magnirostris in New Guinea and northern Queensland, and the greenish C. m. minutillus in northeastern Queensland are in more open habitats where White-throated Flyeater G. olivacea may be the host species.The same cuckoos also parasitize sunbirds in New Guinea and Australia. Also, Shining Bronze-cuckoo parasitize two species of Gerygone in New Zealand and the Chatham Islands. These populations show a lack of cospeciation in the cuckoos and their songbird hosts.

108 The Cuckoos In a second case,Thick-billed Cuckoos Pachycoccyx audeberti in Africa use host helmet-shrikes Prionops, although they must use other hosts in Madagascar where helmet-shrikes do not occur.The host species are more closely related to other songbirds (insofar as we can compare genetic distances across different genes, Cibois et al. 1999, 2001,Yamagishi et al. 2001, Sefc et al. 2002), than Thick-billed Cuckoos are to other cuckoos. Furthermore, the subspecies of Thick-billed Cuckoos within Africa do not coincide geographically with species of their helmet-shrike hosts (Fry et al. 2000). In a third case, African Cuckoos Cuculus gularis in most parts of Africa use Fork-tailed Drongos Dicrurus adsimilis although in west Africa they also use Yellow-billed Shrikes Corvinella corvina, which being local as hosts may be secondarily-acquired hosts. It appears that brood-parasitic cuckoos have switched from time to time from old hosts to a new host species. In these cuckoos, the alternate host species are not closely related to the primary host species. First, in the Indopacific region, Common Koels have switched hosts from crows Corvus spp. to parasitize mynahs Acridotheres spp. in southern Thailand and in the Malay Peninsula within the past century, as one host became scarce and the other became more numerous with changes caused by human alteration of the habitat.The variation within time suggests a number of host switches in the past in this species, which parasitizes mynahs elsewhere in Indonesia and large honeyeaters in Australia. Great Spotted Cuckoos have begun to parasitize certain host species in recent years.The cuckoos now parasitize House Crows Corvus splendens in Israel, with records beginning in 1997 (Yosef 1997, 2002). The cuckoo and the host have only recently come into contact. House Crows spread from India to Egypt and the east coast of Africa in the first half of the twentieth century, and they have bred in the Middle East only since the 1970s (Shirihai 1996).The cuckoos and House Crow were not in sympatry in earlier years and had no earlier opportunity to interact.These cuckoos parasitize other species of crows in the Middle East, Egypt and around the Mediterranean, so colonizing the new host did not involve a great change in cuckoo behavior. Also, Great Spotted Cuckoos parasitize Common Mynah

Acridotheres tristis near Estcourt and Colenso, Natal, South Africa (Symons 1962, Sharp 1976). Mynahs in South Africa were introduced from Asia in 1888 to 1900 (Maclean 1993). The cuckoos in Weenen County regularly parasitize open-nesting crows and hole-nesting starlings (West et al. 1964), and mynahs nest in the same kind of holes in buildings and trees as the native starlings. Certain Asian cuckoos parasitize two or more closely related species of hosts even though these cuckoos are not all each others’ closest relatives. In this case the five cuckoos Cuculus poliocephalus, C. canorus, C. optatus, C. lepidus and C. saturatus each parasitize small warblers of the genus Phylloscopus, and these cuckoos sometimes parasitize the same host species. These cuckoo-host associations indicate that these hosts have been colonized more than once by more than one kind of cuckoo (Becking 1981, Higuchi and Sato 1984). The same cuckoo species also parasitize other species of hosts that are not closely related to Phylloscopus warblers.These associations of brood-parasitic cuckoos and hosts tell a history of independent colonization by cuckoo species of their host species, rather than speciation of cuckoos that went along with the speciation of their hosts. In all these associations, the brood-parasitic cuckoos and their hosts have speciated independently. There is no evidence that cuckoos and their hosts have cospeciated, or that phylogenetic relationships among the hosts give clues to the relationships among the cuckoos. In their independent histories of speciation, the parasitic cuckoos are similar to the African finches. Among the finches, the brood parasites and their hosts speciated along completely independent lines where one tree of life did not shadow the other. Genetic differences between closely related viduid species are less than the genetic distances between the corresponding host species even though both lineages have the same generation times. The genetic phylogenies point towards an evolutionary history of successful switches of a line of parasites from an old host to a new host (Klein and Payne 1998, Sorenson and Payne 2001, 2002, Payne et al. 2002). In summary, geographic isolation has led to speciation in cuckoos, much as it has in other birds (Mayr 1963). Still, it is curious that the cuckoo species which are genetically most similar to each other are brood parasitic, particularly the Cuculus cuckoos.

7 Fossil and comparative evidence of cuckoo relationships Fossil cuckoos In the early Tertiary (Houde and Olson 1988, Mayr 1998a,b), the oldest fossil birds that might be cuckoos are known in the Paleocene and Eocene, the time when many avian families first appeared in the fossil record. None of these ancient fossils are known for certain to be cuckoos (Feduccia 1996, G. Mayr, in litt.). An Eocene fossil Foro panarium Olson 1992 from Wyoming is a nearly complete specimen that has some cuckoo-like characters such as a pectineal process of the synsacrum. Foro is said to represent a primitive group of land birds (Olson 1992). Cuckoos are not unique in having a pectineal process, as it also occurs in other terrestrial birds, and Foro may not be a cuckoo. Fossils from the early Tertiary have been identified as cuckoos from a tarsometatarsus or humerus. Marsh (1872) described a small fossil, Uintornis lucaris, from the Eocene of Wyoming as a woodpecker, on the basis of the distal end of a tarsometatarsus. The fossil was illustrated and described in detail by Shufeldt (1915), who thought it was not zygodactyl and it was neither woodpecker nor cuckoo. Brodkorb (1971) listed it as a cuckoo; others listed it in a fossil family Primobucconidae (Feduccia and Martin 1976, Olson 1985). Fossil tarsometatarsi from the London Clay deposits were the material used to describe two other small birds, Parvicuculus minor and Procuculus minutus (Harrison and Walker 1977). Olson and Feduccia (1979) noted that cuckoos have a posteromedial twist of the outer trochlea (number IV), and this twist was absent in the

London Clay specimens. Harrison (1982) attributed the lack of a distinctive trochlea IV to a “generalized” cuculiform condition and cited the lack of the feature in turacos and hoatzin as evidence, but the assumption that turacos and hoatzin are cuculiform birds is questionable. Harrison (1982) described several features of “cuculiform” tarsi, and proposed a new family “Parvicuculidae” to accommodate the London Clay fossils with the view that they were not cuckoos in the modern sense. Baird and Vickers-Rich (1997) described a fossil tarsometatarsus from the Late Paleocene of Brazil as a cuckoo, Eutreptodactylus itaboraiensis. In this fossil, trochlea IV is large, rounded in shape, lacks a sulcus and lacks a distinct sehnenhalter, and trochlea IV is rotated only slightly from the axis formed by the distal projections of trochleae II and III, whereas it is rotated more than 60° in modern cuckoos and other zygodactyl birds, so it is questionable that the fossil tarsometatarsus is of a zygodactyl bird. Finally, Chandler (1999) described Eocuculus cherpinae from Colorado (Chandler 1999), complete except for a skull. The distal part of the tarsometatarsus may have a large trochlea IV with an accessory articulating process or sehnenhalter. This process is like that of cuckoos insofar as the process in cuckoos is weakly developed, yet the bone of Eocuculus is not clearly zygodactyl, and the similarity with cuckoos is the lack of a diagnostic sehnenhalter such as occurs in the other modern groups of zygodactyl birds. In these Tertiary fossils the identification of a tarsometatarsus as a cuckoo was “cuculiform by elimination, and more similar to Cuculidae than Musophagidae or Zygodactylidae” (Harrison and

110 The Cuckoos Walker 1977). One diagnostic definition of a cuckoo metatarsus is “elongate, narrow corpus tarsometatarsus, lenticular-shaped trochlea metatarsi IV lacking any sulcus” (Baird and Vickers-Rich 1997). But some cuckoos (Scythrops) have a broad tarsometatarsus. The other half of the definition, like the descriptions of Harrison (1982), reflects the lack of distinctive features of a cuculid tarsometatarsus. The other skeletal element used to identify early Tertiary fossils as cuckoos is the humerus. Dynamopterus velox Milne-Edwards 1892 in the EoOligocene chalks of southern France was described from a humerus but with no comparison with modern cuckoos other than in size. The next candidate for oldest cuckoo is Neococcyx mccorquodalei Weigel 1963 from the Lower Oligocene Cypress Hills in Saskatchewan, from a humerus that differed from modern cuckoos in having the entepicondyle reduced, in having the ectepicondyle ridge less angular, and in other features of shape. The humerus was about the size of the humerus of Coccyzus; the internal condyle was smaller and the olecranal fossa was more shallow (Weigel 1963, Martin and Mengel 1984). Because these authors did not describe differences in the humerus of cuckoos and other birds, the identification of fossil Dynamopterus and Neococcyx as cuckoos is not well supported. Fossil cuckoos are known from the middle to late Tertiary and the Quaternary in the Northern Hemisphere, and in the late Quaternary in the Southern Hemisphere. The fossils establish that cuckoos were present and were similar in morphology to modern cuckoos. A fossil cuckoo Cursoricoccyx geraldinae Martin and Mengel 1984 from the early Miocene c. 20 mya (million years ago) of Colorado was smaller than the modern Lesser Roadrunner Geococcyx velox, and had a less strongly developed coracoid, and tarsometatarsal elements in specialization for cursorial life; the carpometacarpus was longer than in modern terrestrial cuckoos. Fossil roadrunners, perhaps the same species as Greater Roadrunner G. californianus, and a larger one known as Conkling’s Roadrunner G. conklingi (the two may be conspecific), are known from the Pleistocene. They per-

sisted into the Holocene to less than 6000 years ago in arid regions of the southwestern United States and Mexico (Larson 1930, Howard 1931, Miller 1943). G. conklingi was larger and may have been adapted to inland temperatures and a continental climate (Harris and Crews 1983). In Australia a fossil coucal Centropus colossus, known from a humerus, was larger than the presentday Australian Pheasant Coucal C. phasianinus. The fossil coucal lived within the past 40,000 years (Baird 1985a,b, Rich and Baird 1986). In Madagascar, a large fossil coua Coua primavea is known from the Quaternary, and a larger subfossil coua C. berthae is known from recent cave surface deposits (Milne-Edwards and Grandidier 1895, Goodman and Ravoavy 1993, Burney et al. 1997), along with several extinct or locally extinct birds and mammals. C. primavea had a gracile tarsometatarsus of 84 mm, longer and more slender than that of C. delalandei and C. gigas; C. berthae had a broad pelvis and a robust tarsometatarsus of 93 mm, longer and more robust than any other coua. C. berthae may have been flightless.These fossil couas lived 1000 years ago and perhaps as recently as 600 years ago. Remains of other cuckoos in the upper layers of the deposits include the extant Madagascar Lesser Cuckoo Cuculus rochii, Madagascar Coucal Centropus toulou, and other couas including Giant Coua Coua gigas.The associated human artifacts appear to be more recent than the extinct fossil couas (Burney et al. 1997). Finally, on St. Helena Island a small bird Nannococcyx psix was described from an unmineralized partial humerus, identified as a cuckoo and dated as later than 1502, when the island in the South Atlantic Ocean began its ecological degradation with the impact of man and the loss of its forests (Olson 1975, 1990, Rowlands et al. 1998). Recent and prehistoric remains of extant cuckoo species in the Old World include Great Spotted Cuckoo in Israel and Common Cuckoo in Europe, and in the New World the Yellow-billed, Mangrove and Black-billed Cuckoos Coccyzus, two squirrel cuckoos Piaya, and a lizard-cuckoo Coccyzus (“Saurothera”) in the West Indies, two anis Crotophaga in the West Indies and Yucatan, American Striped Cuckoo Tapera naevia in Brazil,

Fossil and comparative evidence of cuckoo relationships 111 and roadrunners in the North American southwest (Brodkorb 1971, Olson and Hilgartner 1982, Morgan 1994). Sites in Melanesia include remains about 8000 years old of Brush Cuckoo Cacomantis variolosus and coucals Centropus thought to be extant species (Steadman et al. 1999); and Longtailed Cuckoo Urodynamis taitensis is known as a subfossil on Henderson I in the Pitcairn Islands from prehistoric Polynesian occupation c. 400– 1000 years ago (Weisler 1995, Wragg 1995), and from Tonga c. 400–2800 years ago (Steadman et al. 2002). In summary, the recent fossil cuckoos show a greater diversity of couas than now live in Madagascar.The fossils do not suggest a great loss of birds caused by human immigrations such as happened in island avifaunas of the Pacific (Olson and James 1982, 1991, James and Olson 1991, Collar et al. 1994, Steadman 1995). The fossils also show extant cuckoos in the same areas where they now occur in the Old and New Worlds.

Cuckoos and the clocks of evolutionary time Estimates based on molecular genetics, of the time of separation of the major groups of birds tend to be older than estimates from the fossil record (Sibley and Ahlquist 1990, Feduccia 1996, García-Moreno and Mindell 2000, van Tuinen and Hedges 2001). The former suggest divergence of modern orders of birds in the Cretaceous, whereas the earliest fossils of the same orders are known from the Tertiary. The first estimate of an early separation of the major groups of cuckoos was that of Sibley and Ahlquist (1990) at 69 million years ago.Their estimate was based on their values for molecular thermal dissociation of Old World and New World cuckoos, compared with the value of thermal dissociation between Ostrich Struthio camelus and Greater Rhea Rhea americana, and a time calibrated from the separation of continental plates by the Atlantic Ocean. Another molecular estimate of the time of separation of Old World and New World cuckoos is 52 million years ago, with a calibration based on a fossil anseriform Presbyornis and a divergence of Ostrich from other ratites at 80 million

years (van Tuinen and Hedges 2001). Other estimates of the time of separation of Africa and South America are 100–120 million years (Hay et al. 1999), somewhat earlier than the estimate used by Sibley and Ahlquist. A more recent possible touchstone is the collision of tectonic plates of the Bismarck Archipelago and New Guinea about 10–15 million years ago (Lee and Lawver 1995, Michaux 1998, Polhemus and Polhemus 1998); coucals in these areas are each others’ closest relatives and may have diverged from a common ancestor sometime after this tectonic event. The basal phylogenetic split of Old World and New World cuckoos (Figure 5.1) suggests an evolutionary history of cuckoos dating to before the separation of Gondwana, or sometime earlier than 60 million years ago, as in other Gondwana clades (Cracraft 2001). Genetic differences between Old World and New World cuckoos also suggest a divergence around 60 million years ago (van Tuinen and Hedges 2001). The estimate of age of 53 million years for the oldest fossil cuckoo (Feduccia 1996) was based on an undescribed bone of a bird of uncertain relationships. Thus, biogeography and molecular estimates point to an origin of cuckoos somewhat earlier than the known fossil record.The genetic and the fossil estimates agree that cuckoos have been a distinct lineage for tens of millions of years. The most secure conclusion is that cuckoos were around after the Cretaceous and diverged during the Tertiary.

What birds are related to the cuckoos? From a historical perspective, the ideas about the relationships of cuckoos are seen in the names given when cuckoos were originally described as new species and were thought to be members of other families of birds. Channel-billed Cuckoo Scythrops novaehollandiae was once known as “Anomalous Hornbill” and “Psittaceous Hornbill”,Asian Emerald Cuckoo Chrysococcyx maculatus was described as a trogon Trogon maculatus, and Pallid Cuckoo was described as a dove Columba pallida. Conversely, Greater Honeyguide Indicator indicator was described by Sparrman as Cuculus indicator, and Green Turaco

112 The Cuckoos Tauraco persa was described by Linnaeus as Cuculus persa. In addition, several birds of other families were named at the species level as cuckoo-like: an owl, Asian Barred Owlet Noctua (Glaucidium) cuculoides and a hawk, African Baza or Cuckoo-falcon Aviceda cuculoides (Sibley and Monroe 1990). Common names reflect the similarity in appearance of cuckoos to other groups, as Cuckoo-doves Macropygia, Cuckoo Rollers Leptosomas discolor and cuckooshrikes Campephagidae in the Passeriformes. Other cultures likewise have remarked on the resemblance of cuckoos to owls and hawks, as in the common names of these birds in the Moluccas (Taylor 1990). Traditional systematic views of avian relationships are based on morphology and suggest that ratites and tinamous are basal to all other birds (Huxley 1867, 1868, Fürbinger 1888). Recent molecular systematics studies support (a) this conclusion, (b) that waterfowl and Galliformes are each other’s closest relatives, and (c) that these are basal to the remaining groups of birds (Groth and Barrowclough 1999, van Tuinen et al. 2000, Johnson 2001, Sorenson et al. 2003b), but they do not resolve the relationship of the cuckoos with these other birds. One recurring idea in avian systematics is that cuckoos are related to the Neotropical Hoatzin Opisthocomus hoazin (Walters 2003). In their anatomical work, Huxley ( 1867, 1868) and Fürbinger ( 1888) found hoatzin most similar to the Galliformes. Sibley and Ahlquist ( 1972) cited Huxley as listing hoatzin “adjacent to the Coccygomorphae”; in fact Huxley concluded that hoatzin is related to galliform birds and doves, and he did not list the cuckoos and hoatzin next to each other. In an anatomical survey of avian leg muscles, the cuckoos and hoatzin were said to be each others’ closest relatives (McKitrick 1991), but the survey included only those cuckoos with formula leg muscles like those of hoatzin and it excluded cuckoos with other leg muscles (e.g. Berger 1952, 1960). In a comparison of egg-white protein electromorphs, hoatzin was more like crotophagine cuckoos than were other cuckoos, and Sibley and Ahlquist (1972, 1973) concluded that hoatzin was a crotophagine cuckoo (however, the results could not be repeated: Brush 1979); and in pairwise tests of molecular thermal association (DNA-DNA

“hybridization”), hoatzin was unlike the cuckoos, and except for Galliformes no other birds were included in the comparison (Sibley and Ahlquist 1990: Fig. 79). More recent, sequence data show that cuckoos and hoatzin are not related (Sorenson et al. 2003b). Cuckoos share several morphological features with turacos, and the cuckoos and turacos are sometimes listed together in a large single order Cuculiformes. This convention was most strongly established by Fürbinger (1888), the first avian anatomist to present a comprehensive phylogenetic hypothesis based on discrete characters. Cuckoos (Sorenson et al. 2003b) and turacos shared more morphological characters (“Merkmale”) with each other than either family did with any other group of birds. Fürbinger’s Tables xli and xlii listed 46 morphological characters (“Merkmale”) including the skeleton, muscles and feather tracts in the families and orders of birds, and features of birds, based on his own anatomical studies. A total of 32 out of these 46 characters were shared by cuckoos and turacos; though a few characters were inconsistent within one family, many characters were also shared with many other avian taxa, and none of the 32 characters were shared exclusively between cuckoos and turacos.The most nearly similar traits shared nearly exclusively by the two groups were not discrete characters, but were the relative lengths of the wing bones and the leg bones (Fürbinger’s characters 20, 21 and 23). Cuckoos are also similar in body form and plumage to doves Columbidae, and their leg muscles and tendons are like those of some parrots Psittacidae (Raikow 1985). In some features of the skull, Hoatzin is similar to cuckoos, in others to turacos, and it is not similar to the curassows Cracidae (Marceliano 1996).The palate of hoatzin is schizognathous, like the palate of Galliformes and some Gruiformes such as the Neotropical trumpeters Psophiidae, and unlike the desmognathous palate of turacos and cuckoos (Huxley 1867, Sibley 1955, Sibley and Ahlquist 1990, LSU, UMMZ). Another idea is that cuckoos are related to both turacos and hoatzin. One molecular study found hoatzin more closely related to turacos than to cuckoos, and these three groups formed a clade

Fossil and comparative evidence of cuckoo relationships 113 (Hughes and Baker 1999). However, only a few other avian orders were used in that comparison, and more extensive comparisions failed to support those two conclusions (Sorenson et al., 2003b). Finally, a preliminary comprehensive analysis of skeletal characters located the hoatzin, turacos, and cuckoos each at different regions of the tree of avian relationships, with cuckoos most closely related to certain coraciiform birds, hoatzin to gruiforms, and turacos to a wide assemblage of orders

including mousebirds Coliiformes, rails Gruiformes, woodpeckers Piciformes and passerines Passeriformes (Livezey and Zusi 2001). Sibley and Ahlquist (1990) concluded that “DNA comparisons show that the cuckoos have no close living relatives.” Mitochondrial and nuclear gene sequences also indicate no close relatives (Sorenson et al. 2003). These genetic studies suggest that we know no single order of birds that is most closely related to cuckoos.

8 Breeding biology and life histories

The breeding behavior of cuckoos ranges from solitary pairs, to social groups that breed cooperatively, to brood parasitism. Cuckoos are altricial, and the young hatch and remain in the nest in a nearly helpless state, unable to feed themselves, and are completely dependent on their parents or their foster parents to provide care and to feed them. The adult females vary from providing parental care to their young in their nest, to the absence of both nest-building behavior and parental care in the brood-parasitic cuckoos. The adult males take on the life style of their females, and some males even provide more parental care than their mates. Among nesting cuckoos, many are secretive and solitary and live alone or in breeding pairs. In contrast, the communally nesting anis and Guira Cuckoos are conspicuous and social birds, living in the same social group both in the breeding season and also when they are not breeding.

Breeding displays Breeding displays are known for only a few cuckoos. A few common themes are seen when the male displays in courting the female. The tail is raised towards the female, and the male offers a bit of food to the female he courts.The display behavior of the Greater Roadrunner of North America, a nesting species, and the Common Cuckoo of the Palearctic, a brood-parasitic species, both of which have been closely observed in the field can be taken to represent that of cuckoos in general. In Greater Roadrunners the male and female spend time foraging together and calling back and forth before they show a sexual interest in each

other. Early in courtship, one bird runs after the other on the ground, often for several hours; both birds stop and rest between chases.The chasing bird runs and lunges at the forward bird with its wings and tail raised and fanned, while both birds give “clack” calls. The male gives a “coo” call from an elevated perch. One bird carries a stick in its bill and presents it to the mate; both sexes pick up sticks and pass them to the mate or drop the stick at the mate’s feet. The behavior may provide the mate with an assessment of its partner before it commits to a nest. Before they copulate the roadrunners exhibit with a “prance display.” The male approaches the female on the ground in a short burst of speed, holds an insect or snake or lizard in his bill, and wags his tail back and forth. The male runs to and from the mate with its wings and tail lifted, then he lowers his wings and brings them inward with a “pop.” He holds his tail over his body, then gradually lowers his tail, exposes the postorbital bare skin, erects his crest, and sleeks the contour feathers. The display involves four or five cycles of lifting the wings: it lasts more than two minutes. In a “tail-wag display” the male wags his tail from side to side while he bows, then slowly lifts his head, as he faces the female and gives a “whirr” call. After this he jumps into the air and leaps over the female or mounts her from the rear. He holds the food in his bill and presents it to the female as they copulate. Then he dismounts, they walk away from each other, they flick the tail, and the female eats the offering or feeds it to her young if copulation occurs after the young have hatched: roadrunners copulate socially, and not only for fertilization (Whitson 1975).

Breeding biology and life histories 115 The displays of males to other males near their territory borders differ from displays directed towards the mate, but the displays have elements in common. Males call across the territory border, give “coo” calls and bow the head in display of the crest and the brightly colored bare skin behind the eye. When one male intrudes, the resident male approaches with head lowered, the bare skin fully exposed (particularly the orange skin, Folse 1974), the tail held upright and wagging from side to side. The intruder exposes his head colors, and the birds move from side to side until the intruder backs down.The resident approaches in short bursts, running and dodging, popping the wings.The resident female may join in defense as she follows the male to the border and makes “clack” sounds until the intruder is gone. In Common Cuckoos, courtship begins when the female gives a bubbling call and a male approaches, or the female approaches a calling male. The male then bobs the head or bows the body, the wings are open and drooped, and the tail is raised and fanned; in the fan posture the male may rotate the body slowly and sway from side to side.The female looks on; only rarely does the male offer her a caterpillar. In mating, the male silently follows the female and may give a piece of vegetation to her, the female opens her wings slightly, moves her tail to one side, and calls just before mating. The male mounts, then droops his wings slightly and moves his tail slowly from side to side. Copulation occurs after the female has laid an egg, and this fertilizes the next egg, which is laid 1-1/2 days later (Wiley 1981, Cramp 1985). The female areas of laying activity do not match the male territories. A male may have more than one female in his area, and a female’s area may overlap that of more than one male (Dröscher 1990, Nakamura and Miyazawa 1997). In some areas where cuckoo density is high, females are territorial and exclude other females from their laying area, even when the other females’ eggs mimic the eggs of another host species. The male sometimes aids his mate, the female in his territory, in expelling an intruding female; at other times he mates with the intruder (Riddiford 1986). Cuckoos of several species spread and raise their tails in sexual display. Secretive as cuckoos are,

glimpses of this behavior have been seen in Old World parasitic cuckoos, in New World Coccyzus cuckoos, in American Striped Cuckoo, in the anis and in Guira Cuckoo. The bold patterns of white spots on the tail are shown in breeding displays in many other Old World malkohas and New World coccyzines. Courtship feeding has been observed in broodparasitic cuckoos (Figure 8.1) including Clamator ( Jacobin Cuckoo, Levaillant’s Cuckoo, Great Spotted Cuckoo), several glossy cuckoos Chrysococcyx (Blackeared Cuckoo, Horsfield’s Bronze-cuckoo, Little Bronze-cuckoo, Shining Bronze-cuckoo, Diederik Cuckoo, Klaas’s Cuckoo, African Emerald Cuckoo), and the fruit-eating koels (Dwarf Koel, Common Koel) and Channel-billed Cuckoo where the male presents the female with fruit, and in Thickbilled Cuckoo. In these birds, courtship feeding is common. Courtship feeding is often seen in several Cacomantis cuckoos (Pallid Cuckoo, Plaintive Cuckoo, Brush Cuckoo, Fan-tailed Cuckoo), and it is seen occasionally in Cuculus cuckoos (Black Cuckoo, Red-chested Cuckoo and Common Cuckoo).

Figure 8.1. Courtship feeding by Jacobin Cuckoos Clamator jacobinus, Lochinvar National Park, Zambia.

116 The Cuckoos In courtship feeding the male captures and holds a caterpillar, then calls and attracts a female to which he gives the prey. In the brood-parasitic Diederik Cuckoo, the male has a perch on his territory where he calls with a caterpillar in his bill, and here he attracts a series of females.When he has a caterpillar he gives a call, and the female gives another call when she attracts the male, the female “caterpillar call” of “deah-deah-deah . . .” which signals courtship-feeding. The male spreads his wings and tail in display as he presents the caterpillar to her; the female repeats her own call at this time, then he mounts her as she takes the food. Females that take the caterpillar are often in the act of ovulation of the yolk into the oviduct, before the germ cell and yolk are surrounded by albumen and egg membranes (Payne 1973a), and this is the time when copulation may effect fertilization. In Diederik Cuckoos in the Eastern Cape Province of South Africa I observed courtship repeatedly, many times in an hour, day after day, as a male attracted his breeding females to his call-site. Female cuckoos may gain a significant amount of their food to form their eggs during the period of courtship feeding. Courtship feeding occurs in nesting cuckoos as well as in the brood-parasites. Courtship feeding has been seen in African yellowbills, but has not been reported in the Asian malkohas. In the New World the behavior has been seen in Dwarf Cuckoo, Yellow-billed Cuckoo, Black-billed Cuckoo, Mangrove Cuckoo, Dark-billed Cuckoo, Jamaican Lizard-cuckoo and Squirrel Cuckoo. In other cuckoos, courtship feeding has been described in couas (Red-fronted Coua, Blue Coua), in coucals (Violaceous Coucal, Pheasant Coucal, Greater Coucal, Madagascar Coucal, African Black Coucal, Lesser Coucal, Green-billed Coucal (although courtship feeding has not been seen in copulation, the male feeds the female on the nest) and White-browed Coucal), in cooperatively-breeding crotophagines (Smooth-billed Ani, Groove-billed Ani), and in the New World ground-cuckoos (Greater Roadrunner). These birds are among the more readily observed cuckoos, though courtship feeding may be wides-

pread among other species of nesting cuckoos as well.

Songs and calls Cuckoo species have unique songs and distinctive calls. Some cuckoos have melodious whistles that recall the musical song of songbirds and humans. In other cuckoos the calls are not whistled but are used in much the same social context and can be referred to as “songs” that are loud, distinctive, and used in the same behavior contexts as the songs of songbirds, in announcing their territories and attracting a mate. Cuckoo songs range from clear whistles to unmusical noises and mechanicalsounding vocalizations. The glossy cuckoos Chrysococcyx and the large Cuculus brood-parasitic cuckoos of the Old World have clear whistles that are musical to listening humans, Great Spotted Cuckoos give harsh screams; malkohas give chatters and rattles, the coucals give dove-like coos, the roadrunners give coos and rattles, the anis give hisses and glissandos, and the New World groundcuckoos make guttural sounds. Cuckoos are distinctive in their songs, especially in the Old World where several species occur together. Many cuckoos call in a position like calling doves, bringing the head forward and down until it nearly touches the breast (roadrunners, coucals, Old World groundcuckoos). Cuckoos sing to establish and maintain their territories and attract their mates.The brood-parasitic cuckoos sing repeatedly all day and even at night during the breeding season, and they sing in territorial behavior and courtship. In the nesting cuckoos, a Squirrel Cuckoo announces his territory by calling as many as 96 notes a minute for several hours through the day (Sick 1993). Females also call, as New World Yellow-billed Cuckoos do with a short version of the male call, and cuckoos often call back and forth with the inmates while on the nest, the females lacking the terminal notes. Jacobin Cuckoo females give a long call “kleeuw, kleeuw, kleeuw” like the male, but only the male follows the series with a faster series of short rising notes “kwik-kwik-kwik”, a difference in behavior that

Breeding biology and life histories 117 was useful to know when females were collected during an ovary study of the otherwise similar males and females (Payne 1973a). Females in Old World brood-parasitic Cuculus cuckoos give a bubbling call unlike the calls of the males, and females do not have loud persistent songs. Some cuckoos such as the Old World ground-cuckoos and New World ground-cuckoos such as the roadrunners signal with bill-pops, clicks and chatters, particularly in aggressive contexts. Group defense of territory against other groups in group-living anis and Guira Cuckoos involves harsh vocal calls. Squirrel Cuckoos have been said to imitate and mimic the sounds of other kinds of birds (Sick 1993). However, vocal mimicry has not been recorded in the field, and no cuckoos are known to learn their songs and calls. Vocal repertoires of more than a single song are best known in the Common Roadrunner. These birds have several distinct calls, and they are used in different situations (Whitson 1971, 1975). (1) The most commonly heard call is a “coo.” The call is a low-pitched hoot like that of a dove or an owl, a downward slurred note, given in a series of about five notes, “coo-coo-coo-coo-cooo,” by a male as he perches on a post or other elevated site. As the male calls, he lowers his head with the crest erect and the bare skin behind the eye exposed, and moves the bill away from the body with each note. (2) A short “coo” is given by the female in a series of two notes, and is not as loud, audible to about 30 m; this is also given by the male. (3) A single “coo” is a soft note heard only up to 2 m away and is given by both sexes, with a flick-bow display in courtship. (4) A “bark” is a rapid series of short notes that sound like coyote yelps and this loud call carries for up to 300 m. It is given by the female, often with the crest raised and the bare skin exposed. (5) “Growl-coo” is 3–4 notes of low frequency, given by both sexes. As the bird calls its throat bulges, it fluffs its feathers, raises its crest and usually conceals bare skin. (6) A “whirr” is a soft, low pitched sound where a “whirr” regularly comes between a series of “put-put” calls heard up to 2 m away.The male gives the call during a tail-wag display in courtship. (7) A “Whine” is a single low-

pitched call heard up to about 5 m, and is given by both sexes. The bird shakes its head from side to side as it moves the head downward. In addition to these vocal calls, roadrunners rattle their bills together in a “clack” while they produce a vocal whine in the syrinx, and the male makes a “pop” by bringing his wings together inward towards the body. In pair formation the “coo” and “bark” are usually given when birds forage apart, and the softer “growl,”“coo” and “whine” when they forage near each other. The birds become quiet while they are rearing the nestlings and more vocal when they have fledglings. Of these calls, the “coo” series is the one like the long-distance territorial song in songbirds. The call is given loudly by the male, from a conspicuous perch, and the male may call for a couple hours beginning at sunrise; the male often remains on the same perch, but he sometimes calls from more than one perch site, and males in neighboring territories alternate calls during the early morning. In the breeding season, many cuckoos call both in the day and at night, both the nesting species such as the Black-billed Cuckoo in North America and several brood-parasitic cuckoos in the Old World. In tropical regions when other birds are quiet, the cuckoos often call persistently in the hot middle of the day, like doves. They also call when rain approaches or has just passed, and for this behavior the cuckoos are known as “rain birds.” “Brain-fever” cuckoos are named for their persistent calls, as in Pallid Cuckoo in Australia, Black Cuckoo in Africa, and Plaintive Cuckoo, Large Hawk-cuckoo and Common Hawk-cuckoo in Asia.The name “Brain-fever” was used for cuckoos in India and Burma as they called all night; in Uganda an administrator was petitioned to send his collectors to shoot these birds as they disturbed the residents’ sleep ( Jackson 1938). Some cuckoos including Asian hawk-cuckoos (Large Hawkcuckoo, Common Hawk-cuckoo) even scream “Brain-fever!” Night sounds are also given by New World nesting cuckoos such as the Black- and Yellow-billed Cuckoos. In Finland an Oriental Cuckoo called day and night. When its song was recorded and played in the field at midnight, the

118 The Cuckoos cuckoo rapidly flew to the playback and was captured in a mist net in conditions so dark that its plumage color could not be seen (Lindholm 2001). These cuckoo songs appear to be as active in advertising a territory at night as in the daytime. The origins of many scientific names of birds have been traced by Jobling (1991). Several cuckoo genera were named for their calls, Cuculus itself for the onomatopoeic “cuckoo” call of Common Cuckoo C. canorus, and descriptions and interpretations of the calls (Clamator, the shouting cuckoos; Cacomantis (Gr. kakos, evil, ill-boding; mantis, prophet) refers to the “rain bird” role in folklore, as these birds are said to predict ill fortune and bad weather; its synonym “Penthoceryx” (Gr. penthos, misfortune, grief; kerux, a herald) indicates the same; “Caliechthrus” (Gr. kaleo, to call; ekhthros, odious) the loud monotonous calls, which according to tradition foretell bad weather and ill fortune; Coccyzus derives from Gr. kokkuzo, to cry “Cuckoo”; Hyetornis (Gr. huetos, rain; ornis, bird) refers to the calls of the local rain-bird; Coua is a Malagasy name for these birds that repeat the calls transliterated as “coua”; and Guira derives from güirá , a Paraguayan Indian name for a bird named after its calls.

Mating systems Most nesting cuckoos are socially monogamous: a male and a female live together during the breeding season. Both sexes rear the young.The cooperatively breeding anis appear to maintain distinct social pairs of a male and a female. In a few cuckoos there have been molecular genetic studies to test whether the members of a social pair are monogamous or mate outside the pair bond as well. Mating systems of the brood-parasitic cuckoos are not well known: male and female show no strong pair bond. A male Diederik Cuckoo regularly calls in a bush and attracts a series of females; he feeds each with caterpillars and courts them one by one over a day or several days (Payne 1973a), so there appears to be no special pair bond. In one case a male Diederik Cuckoo had two females, one that spent her time in a colony of Fire-crowned Bishops Euplectes hordeaceus, the other in a colony of Black-headed Weaver Ploceus melanocephalus

(Verheyen 1953). Males occasionally feed a fledged young cuckoo and this behavior may be misdirected courtship feeding. Adult brood-parasitic cuckoos do not feed the young while the nestling cuckoos are still in the hosts’ nest. Exceptions to the lack of a pair bond in brood-parasitic cuckoos are seen in cases when male and female act together, as when the male draws the host off the nest while the female slips onto the nest to lay her egg, in crested Jacobin Cuckoo, Common Koel and Channelbilled Cuckoo. In Common Cuckoos in Britain and Japan, the mating system was exposed by comparing nuclear microsatellite genetic profiles of adults and nestlings. In both areas, cuckoo mothers were identified by comparing their profiles with those of cuckoo chicks in the nests of the host species. Cuckoo mothers were also identified by comparing mitochondrial genes of adult females and chicks, as these genes are transmitted and inherited in a matrilineal pattern. The young of each female had the same father, and the results show that the female was monogamous. One male was polygamous and was father of the offspring of two females. The results indicate that each female prefers to lay in the nests of the host species that fostered her, but she does not preferentially mate with a male that was reared by the same foster species ( Jones et al. 1997, Marchetti et al. 1998, Gibbs et al. 2000). Cuckoos do not learn or mimic the songs of their foster species, and they do not have a vocal signal that guides a female to mate with a male that was reared by her own foster species. In Great Spotted Cuckoos a female sometimes lays more than one egg in a crow’s nest, and more than one female cuckoo sometimes lays in a nest. Microsatellite markers reveal some cases of monogamy and other cases of a cuckoo mating with more than one partner. No mitochondrial markers were included in the study, and it is unknown whether these polygamous cases were a female mating with more than one male, a male with more than one female, or both (Martínez et al. 1998a,b). In contrast to the brood-parasitic cuckoos, males and females in the brood-parasitic finches Vidua have a common genetic interest in finding a mate that was reared by the same foster species. Each

Breeding biology and life histories 119 species of indigobird Vidua mimics the mouth pattern of its host species, and the nestling indigobird has a mouth color and pattern like that of the host nestlings with which it is normally reared in the nest (Payne 1973, Payne and Payne 1994). Both male and female form a behaviorally imprinted association with their foster species. Males mimic the songs of their foster species, which they learn from their foster parents. Females are attracted to the songs of male indigobirds that mimic their own foster species. Moreover, the females reared by an alternate foster species parasitize their own foster species and not the nests of their normal host species where the nestlings would have the mouth pattern of the females’ own offspring, even when nests of the normal host species are available. The inclination of both male and female indigobirds to a common foster species indicates a common genetic interest in their offspring, as both sexes contribute nuclear genes that determine mouth mimicry in their young, and both benefit when their mate has the same developmental experience and genetic traits for nestling mimicry and the eggs are laid in the nest of the same species that had reared the adults (Payne et al. 1998, 2000, Sorenson et al. 2003a). Most breeding cuckoos are socially monogamous.The partners of a pair are often together, and both male and female incubate the clutch and rear the brood. Molecular genetic studies have been carried out with cooperatively breeding New World anis and with Guira Cuckoos. Early results in Guira Cuckoos suggest that nestmates in a communal nest are more likely to be half-sibs rather than full sibs, indicating that a parent had mated with more than one partner. The genetic profiles of nestmates in one nest, revealed that a female mated with more than one male and a male mated with more than one female (Quinn et al. 1994). Fieldwork continues in social behavior and molecular genetics of families in these cuckoos. In the South American Dwarf Cuckoo Coccycua pumila and Dark-billed Cuckoo Coccyzus melacoryphus occasionally two females lay in the same nest. The birds may be polygynous with two females mated to the same male, or they may be facultative brood parasites (Ralph 1975, Sick 1993). In North

America, Black-billed and Yellow-billed Cuckoos occasionally lay in the nests of other species such as songbirds and doves, and more often they lay in each other’s nests (Darwin 1875, Bendire 1895, Nickell 1954a, Wiens 1965, Nolan and Thompson 1975, Peck and James 1983, Fleischer et al. 1985). In these cases it is unlikely that interspecific promiscuous sex was involved. Molecular genetic studies and behavioral observations of marked cuckoos are needed to test whether the cuckoo eggs and nestlings in these nests are related at the half-sib level, and whether one male is the mate of both females. Perhaps a female lays in the nest of another pair when her own nest has been disturbed. In the nests where these cuckoo eggs appear, the nesting species sometimes rears the young cuckoo (Darwin 1875, Nickell 1954b); more often the outcome is unknown. Most coucals that have been watched in the field are monogamous, living in pairs (Baker 1934, Frauca 1967, Dhindsa and Toor 1981, Taplin and Beurteaux 1992, Natarajan 1997, Hustler 1998). Males often carry out the larger share of parental care, with both sexes incubating and caring for the young. Behaviors that suggest a strong social pair bond in coucals includes courtship feeding by the male and song duets by the pair. These behaviors are more common in monogamous birds with a strong pair bond than in birds with other mating systems (Lack 1968, Payne 1971, Kunkel 1974, Farabaugh 1982). On the basis of the sex roles in parental care, some coucals are suspected to be polyandrous. In polyandrous birds a female mates with more than one male and in these birds often the male takes a more active and conspicuous role than the female in incubation and care of the young. Polyandry has been seen once in the African Black Coucal, when a female had three males and mated with each male in turn, while each male held a separate territory within her large domain.The males built the nests, the female laid the eggs, then each male incubated the clutch in a nest and brooded and fed the young. The female took no active role in parental care. Only two females were observed, and only one had more than one mate (Vernon 1971). Reversed sexual size dimorphism with females larger than males coincides with clades or lineages

120 The Cuckoos of coucals, rather than with widespread reversal of sex roles or polyandry as suggested by Ligon (1993, 1999) and Andersson (1995). As in other birds, the female coucals have only one ovary and oviduct. Male coucals often have only one large testis (Rand 1933, 1936, Chapin 1939, Centropus milo in UWBM) and when two testes are present they vary in whether the left or right testis is larger even within a species (Mayr 1937, Mayr and Rand 1937, Rand 1942a), while the testes are extremely asymmetric in some other cuckoos such as the cooperatively breeding anis (McNeil 1968). Explanations of the origin and adaptation to polyandry in coucals in terms of their asymmetric testes (Ligon 1997, 1999) are not well supported, insofar as field studies show that coucals are generally not polyandrous.The idea that nocturnal incubation by the male in cuckoos led to increased male parental care and to relief of the female from nesting duties in cuckoos, suggested by Vehrencamp (1982a, 1985) and Ligon (1993, 1999), could be explored by determining the incubation roles of males in other kinds of polyandrous birds. Does staggered asynchronous hatching make it likely that a single parent can care for a brood? Again, there is no evidence that asynchronous hatching which is common in other cuckoos is associated with polyandrous mating. African Black Coucals are seasonally migratory in parts of their range, and Andersson (1995) suggested that migratory behavior may have led to polyandry, yet other

migratory cuckoos are not polyandrous.Additional fieldwork is called for on breeding behavior of these cuckoos.

Nests and nestbuilding Nest structure with most cuckoos is simple. The cuckoos build an open platform nest, using sticks as the main nest material, and they line the nest with finer sticks (Figure 8.2).These nests are often flimsy and the cup that holds the eggs is shallow. Coucals build a covered nest, sometimes called a globular or domed or thatched nest. These nests are usually built of grass or other soft vegetation, or sticks thatched together with the stems of grass in which the nest is concealed. The sides are built upward and form a roof. The nest has an entrance on one side, and in some the entrance has a covered tunnel that leads between the nest chamber and the surrounding vegetation. A few cuckoo species have a variable nest structure, sometimes open and sometimes covered. Pheasant Coucal Centropus phasianinus, Greater Coucal C. sinensis and Madagascar Coucal C. toulou sometimes build an open nest rather than a covered nest, and only an open nest has been described for Bay Coucal C. celebensis in a group which otherwise have covered nests. Also, White-browed Coucal C. superciliosus builds a bower-like nest when it is built in dense overhanging vegetation. Senegal Coucal C. senegalensis usually build a covered nest.

Figure 8.2. Black-billed Cuckoo and nest, Ann Arbor, Michigan: a, adult on nest; b, nest with 3 eggs.

Breeding biology and life histories 121 However, one pair laid in an open nest that they built in a dense spiny citrus tree where the canopy overlaid the unroofed nest when the eggs were laid—we saw this at Bakary Sambou, near Brikama, The Gambia, in early October 1999. In some cuckoos that usually build an open nest, the yellowbills Ceuthmochares and Blue Coua Coua caerulea sometimes build a covered nest.Timing may determine when the nest is completed and the eggs are laid, with some building continuing after the first eggs are laid, as in the open nest of the coucals. Or the nest structure may vary with the nesting substrate, as in the coucals. Some cuckoos continue to bring nesting material to the nest, especially green leaves, after the nest has been built, during incubation and after the young have hatched.This behavior has been seen in coucals, anis, Guira Cuckoo, roadrunners and the New World ground-cuckoos. It is unknown whether the leaves have an antibiotic effect that keeps bacteria from invading the eggs or young, or a strong odor that covers the odor of the brood and avoids detection by a predator.

Nesting pairs Most cuckoo species that build their own nests and rear their young appear to be monogamous, the birds appearing in pairs and two birds caring for the eggs and young in the nest.The nesting cuckoo that has been observed in most detail, Greater Roadrunner, shows behavior that appears to be typical of a monogamous bird. The male brings food—mice, small birds, snakes and lizards—to his mate before mating and during copulation. The pair copulates at the place where they will build the nest.They continue to copulate after their clutch is complete, and copulation appears to have a social function in bonding the pair as well as a fertilizing function when the female is ovulating. The nest is usually built in isolated thickets of small trees such as mesquite and bushes, in an area near open ground or short grass where the birds display and forage. The nest site is 1–3 m above the ground, occasionally higher than this, in the crotch of a tree or resting along a horizontal branch, near the center of the bush or tree. Nests are built mainly by the

female. The male brings building material in the form of thorny twigs and branches to the nest site, and gives them to the female when she gives a “whine” call. She incorporates them into the nest, laying the larger sticks around the edge and smaller sticks near the center of the nest. The pair may begin several nests and desert a site after building for a few minutes (Whitson 1975). In the site selected for the active nest, she builds the platform of sticks, then adds a lining of finer nest material such as leaves, grass, mesquite pods, snakeskin, and dry flakes of cattle and horse manure as well as feathers.The nest is a flattened cup, about 30 cm in outer diameter, 5–10 cm in inside depth and 15–20 cm in outside depth (Sutton 1940, Hughes 1996b). Some pairs continue to build during incubation and the nest may grow in height as the nestlings hatch and grow and require a larger and more protected living space (Calder 1976). Roadrunners lay a clutch of smooth white eggs, usually 3–6 (Meinzer 1993). Occasional sets of 9–13 eggs may be mixed clutches in which more than one female laid in the same nest (Sutton 1940), although the number of eggs a single female lays can vary. The clutch is often larger after good summer rains when food is abundant.The eggs are usually laid on alternate days but also at irregular intervals (1 to 4 days or as long as 9 days), and the birds begin to incubate before the complete clutch has been laid.The eggs are covered continuously, as the incubating parent remains on the nest until the other parent arrives to relieve it. Incubation that effectively transfers heat begins only after the parent has covered the eggs, and as some eggs are laid after incubation has begun, the eggs hatch asynchronously on a staggered schedule over several days. Roadrunner eggs hatch in about 18 days.The young in a brood are of different sizes with age differences as great as 7 days between the largest and smallest siblings.When the young bird hatches, the parent carries away the eggshell, then she breaks and eats it (Calder 1967a, Muller 1971, Smith 1981). The last-laid eggs are often abandoned after the first-hatched young have left the nest (Folse 1974). The last young may remain in the nest up to a week after the other nestlings have fledged. Success of the brood varies with feeding conditions,

122 The Cuckoos where parents are able to rear all their young when food is abundant. The young remain in their nest while both parents bring food. Both parents brood their young, nearly continuously for the first four days, and at cool times later.The parents bring intact food items such as lizards and snakes and the only treatment they give is to beat the vertebrate prey into a flexible form, and for grasshoppers they remove the hind legs (Meinzer 1993). The adult eats the nestlings’ fecal sac.Adults eat their own young nestlings when the young are unresponsive or weak, or feed these young to the larger and stronger nestlings when food is short. Nestlings may evict their younger siblings from the nest (Hughes 1996b), a behavior like that in the brood-parasitic cuckoos. Nestling roadrunners give a loud vocal “churr” and they rattle the bill when disturbed and excrete a blackish foul-smelling liquid reminiscent of disturbed snakes. When the young have grown, the parents give soft “coo” calls to the young to leave the nest (Calder 1967).The young fledge in 17–19 days, or as late as 25 days, or earlier if they are disturbed in the nest. Roadrunner young at fledging are only 50% of average adult weight, and young have been observed foraging with their parents when still only 70% of adult size. The adults lead their young far from the nest a few days after they fledge. Fledged young follow their parents to feeding areas, quivering their wings as they beg for food (Meinzer 1993) until they can feed themselves, with independence gradually developing about 30–40 days after they leave the nest (Whitson 1975). Roadrunner nests are spotted by watching and following an adult as it perches or runs over the ground carrying a snake, lizard or other large item of prey in its bill (Figure 8.3).The bird makes a beeline for the nest shrub, then hops up to the nest. Laura and I watched a roadrunner in southern Arizona. It crossed the road with a lizard in its bill, ran to the base of a mesquite, then ran up the sloping trunk and branches directly to a nest about 5 m above the ground, and fed the nearly-grown young in the nest. The female roadrunner sometimes lays a second clutch while her fledglings are being fed by the

Figure 8.3. Greater Roadrunner Geococcyx californianus with food for its young, Joshua Tree National Monument, California.

male, and this biparental care allows the pair to hatch and raise two broods. In Texas one pair had three nests in succession, the first in April, the second in June and the third in August, and all three broods fledged.Another pair began a second nest in July while the pair were still feeding the fledglings from the earlier nest. In some renestings the same nest is used, in others a new nest is built (Meinzer 1993). Parental behavior has been studied using radio-telemetry. The male incubates at night, and the pair takes turns during the day with the female taking two long spells on the nest. Males that incubate at night maintain a body temperature of about 2°C lower than in the daytime. Females and

Breeding biology and life histories 123 non-incubating males drop their body temperature 5°–8°C at night and save about 36% of their energy requirements during this period. Breeding males are more robust with fat deposits and heavier body mass than nonbreeding males, and breeding female roadrunners recover their prebreeding body mass while their males work the night shift (Vehrencamp 1982a). Roadrunners may be typical of nesting cuckoos in their nesting, although it is hard to know and most cuckoos have not been observed in detail. In the features that may be widespread among nesting cuckoos, both the male and the female incubate, feed the young in the nest, and tend the fledglings. In Black-billed Cuckoos a pair alternate their parental care during incubation with each member on the nest for about two hours. In most coucals both parents care for the young, and in some the males may take a greater part than females. Also, courtship feeding is widespread in nesting cuckoos; Black-billed Cuckoos and coucals pass food from one mate to the other, a behavior that provides the female with extra food to develop her eggs. In nesting cuckoos, courtship feeding may give the female a cue to assess the male as a provider of parental care before she decides to mate with him. Couas are often seen in pairs. The birds build a shallow nest of sticks, and both parents rear the young. Courtship feeding has been seen in two coua species, Coua reynaudii and C. caerulea (Goodman et al. 1997). The breeding biology of Australian Pheasant Coucals and other coucals differs from the biology of many altricial tropical land birds.The coucals lay large clutches, have a short nesting cycle, the nestlings grow rapidly, and the juveniles mature early. Some other tropical cuckoos also have these traits. Many other tropical bird families have small clutches, the nestlings grow slowly and the young mature slowly as juveniles. Coucal nesting success is high (Taplin and Beurteaux 1992). The adults add green leaves to the nest before laying and continue to add leaves through incubation. Nestling coucals have curious hair-like natal down formed by a long keratinized sheath of the growing contour feather. Nestlings also have prominent marks and papillae on the palate and sometimes these have contrasting

colors.The young have a prominent egg-tooth as in other cuckoos, and they emit a foul-smelling liquid excreta from the cloaca when nestlings are disturbed in the nest (Frauca 1967).

Eggs Cuckoo eggs are remarkable in the brood-parasites, as they often match the color and pattern of the eggs of their hosts (Davies 2000). In contrast, most nesting cuckoos have plain white eggs.The general features of egg size, shape, and thickness are not unusual in cuckoos as a group, but the eggs of brood-parasitic cuckoos are special in these features, which appear to be adaptations to their breeding behavior. Some features of egg size and shape in cuckoos, however, appear to be more directly accounted for by the evolutionary relationships of these birds than by the special demands of their breeding behavior.

Egg size and shape The eggs of nesting cuckoos are similar in size to that of other altricial birds of their body size, but the eggs of brood-parasitic cuckoos tend to be small (Darwin 1875, Payne 1974;Table 4.3, Figure 8.4). The small eggs of brood-parasitic cuckoos are similar in size to the host eggs. Small eggs may allow a female parasitic cuckoo to lay more eggs in a season, for reasons of economy, and the eggs may go undetected in the nest if they resemble the host eggs in size as well as in color and pattern.Also, small eggs may allow the female to find and lay her eggs in a larger number of nests insofar as small host species are more abundant than larger hosts the size of the cuckoo (Payne 1974). However, at least in Britain, small hosts are not parasitized proportionately more often than larger hosts (Soler et al. 1999). Common Cuckoo egg size does not vary with host egg size (Latter 1902, 1905). The exceptions were the large cuckoo eggs in nests of the large Corn Bunting Emberiza calandra, but this host accounted for only 27 of 11,870 cuckoo eggs, or 0.002 percent of the sample (Moksnes and Røskaft 1995).The trend for small eggs is unique to cuckoos among all broodparasitic birds. The other brood-parasites (the

124 The Cuckoos

Figure 8.4. Egg size and female body size in cuckoos. Egg mass (g) is estimated from the equation, g = 1.08 ⫻ 0.512 ⫻ 1 ⫻ d2, where 1.08 is egg density (g/mm3), 0.512 is a geometric factor, l is egg length (mm) and d is egg width (mm) (Payne 1977c).

honeyguides, the cowbirds and the African viduid finches) do not have small eggs for their body size. The cuckoos are several times larger in body size than most of their host species, whereas the other brood-parasitic birds are not as disproportionately large (Payne 1977b,c, 1989). In nesting cuckoos, the eggs in coucals are smaller than in couas, malkohas and large New World roadrunners and ground-cuckoos.The communal-nesting anis and Guira Cuckoos have large eggs for their body size. In addition, the eggs of these crotophagine cuckoos are remarkably variable in size within a local population, perhaps owing to differences in physical condition of laying females within a social group, perhaps with nestlings hatched from large eggs having a size advantage within a crowded communal brood. Nevertheless, in Guira Cuckoos the egg characteristics of size and markings vary as much within a female as between

females and there is no suggestion that females can recognize each other’s eggs (Cariello et al., 2002, 2004). Cuckoos in comparison with other birds have rounded eggs, although less round than in parrots, owls, trogons and coraciiforms (Rahn and Paganelli (1988). Among the cuckoo species, large eggs tend to be rounder than small eggs (Figure 8.5). The trend is the reverse of the trend in other orders of birds in the same size range. Brood-parasitic cuckoos lay elongated eggs, and nesting cuckoos lay round eggs. And the largest brood-parasite eggs are more elongate than the eggs of nesting cuckoos of the same egg size, suggesting that parasitic cuckoos tend to have more elongate eggs than nesting cuckoos, independently of their size. Eggs of the brood-parasitic crested cuckoos Clamator are more rounded than those of other brood-parasitic cuckoos of the same size. Brooker

Breeding biology and life histories 125

Figure 8.5. Egg shape in cuckoos. The shape (elongation) is the ratio of length to width; large eggs tend to be less elongate (more round), and the eggs of brood parasites tend to be more elongate (less round) than the eggs of nesting cuckoos of the same body size.

126 The Cuckoos and Brooker (1991) suggested that the round shape is an adaptation to avoid damage when more than one cuckoo egg is laid in a host nest. Great Spotted Cuckoo Clamator glandarius lays while perched on the edge of a nest, and the host eggs in nests that have been parasitized are sometimes cracked. In experiments, when eggs are dropped from this height, on impact they crack the host eggs in the nest (Soler et al. 1996). A few other brood-parasitic cuckoos also lay more than one egg in a nest, particularly Common Koel Eudynamys scolopacea and Channel-billed Cuckoo Scythrops novaehollandiae, but these cuckoos do not lay rounded eggs. In addition, the egg shape in Clamator is about the same as in the malkohas, birds of the same body size that do not lay communally in a single nest. Finally, in the crotophagine anis, cooperative breeding birds in which several females lay competitively within a common nest, the eggs are not round, except in the largest species, Greater Ani, in which body size is large and egg shape is in line with the trend for large birds to lay rounded eggs. Because the prediction of rounded eggs with multiple nest parasitism does not account for egg shape in these cuckoos, it is more likely that eggs of brood-parasitic Clamator are round due to their ancestral relationship to the nesting malkohas.

Eggshell thickness, strength and ultrastructure Cuckoos as a group do not have unusually dense eggshells. In a comparison of egg structure in all avian orders, the only birds that were found to have unusually thick eggshells and eggs with high shell mass for their body size were the galliforms (Rahn and Paganelli 1989). Eggs of parasitic cuckoos may have a hard and thick shell that resists cracking when the female lays her egg from a perch above the nest.The host often attempts to remove the cuckoo egg from its nest, and the eggs of parasitic cuckoos might be structurally formed to avoid being rejected. Some hosts, mainly the larger species, remove the eggs by grasping around the eggs with their bills, while the smaller hosts with smaller bills can remove the eggs by puncturing the shells, grasping the edge and lift-

ing the eggs from the nest. A structural adaptation by which the eggs of brood-parasitic cuckoos might avoid rejection is a thick or dense eggshell that resists puncture by the hosts’ bill. Another is a round shape, which would be more difficult for the host to grasp and to puncture. In an experimental study of eggs of brood-parasitic cowbirds, which have a thicker eggshell and rounder shape than icterid species that rear their own young, the cowbird eggs that were thicker-shelled and more rounded had a greater resistance to being punctured than eggs that were thin and more elongate. In the cowbird eggs, eggshell thickness and egg shape contributed equally to puncture resistance and this structure appears adapted to their brood parasitism (Picman 1989). To test whether brood-parasitic cuckoos have eggshells that are thick and dense compared with those of nesting cuckoos, the data on eggshell thickness and mass in Schönwetter (1964) were compared with egg size in Table 4.3. Eggshell thickness and weight were closely associated; eggshell weight is used in the comparison of egg size and eggshell mass (Figure 8.6). The eggshell is not thicker or denser in brood-parasitic cuckoos than in the nesting cuckoos, when egg size or body size is taken into account.Within the region of similar egg size and body size, the parasitic cuckoos do not have a more robust eggshell. Picman and Pribil (1997) found that Cuculus have denser eggshells than Clamator, and this is not the result predicted if eggshell density were related to multiple nest parasitism, nor is the difference apparent in the data when egg size and body size are taken into account. The eggshell of Cuculus eggs is thicker than that of their nesting hosts, and the hatching cuckoos began to peck earlier in relation to hatching and made more pecks at the shell from inside the egg than did young Great Reed Warblers Acrocephalus arundinaceus (Honza et al. 2001). It remains to be tested whether the eggs of the brood-parasitic cuckoos and nesting cuckoos hatch in the same manner. The eggshell structure in brood-parasitic cuckoos differs from that in nesting cuckoos (Mikhailov 1997). Cuckoo eggshells are like those of coraciiform and piciform birds, and differ from

Breeding biology and life histories 127

Figure 8.6. Egg size and eggshell weight in nesting cuckoos and brood-parasitic cuckoos.

other orders of birds in having two outer layers in distinct and sharply separated zones (SqZ ⫹ EZ). An external zone is present in all cuckoo eggs and is absent in passerine eggs. In nesting anis, Guira Cuckoo and roadrunners there is a thick carbonate microglobular outer layer (15–40 µm) that appears chalky on the surface (Schmidt 1964, Board and Perrot 1979). The egg is covered with vaterite, a form of calcium carbonate that forms small globes. This material is resistant to crushing (Tyler 1969) and it may be an adaptation to the competition over eggs that occurs in the communal nest of anis where one female drops her eggs onto the eggs of another female. In eggs of Centropus and Coccyzus, the cuticle is thin (3–10 µm) with microglobular and granular inclusions, and the egg looks dull and non-crystalline, again with a bloom. Eggshells of some brood-parasitic cuckoos have similar globeshaped particles on the surface (Becking 1975a), and this layer may be an adaptation for parasitism in

birds that drop their eggs into a nest. The eggs of most brood-parasitic cuckoos are glossy and lack the thick layer of vaterite that characterizes ani eggs. The eggshells of Old World parasitic cuckoos have a thick outer layer that differs from eggshells of nesting cuckoos and are like those of some other birds such as the Passeriformes (Mikhailov 1997). Although these features of the eggshell of parasitic cuckoos may contribute to resistance to damage by the hosts, no functional design is obvious in the microstructure of the eggshell.

Incubation behavior, incubation period and temperature tolerance In most cuckoos both sexes incubate the eggs, transferring heat from the body of the adult to the eggs in the clutch. Heat is transferred through a brood patch, a bare area on the ventral side of the adult.The brood patch becomes vascular and thick

128 The Cuckoos in breeding Yellow-billed Cuckoos and presumably in other nesting cuckoos. Both sexes incubate and both are said to have a brood patch (Pyle 1997). The brood patch has not been examined in nesting cuckoos particularly the male, and some observations may have mistaken the ventral apterium which all cuckoos have for the specialized shortterm brood patch which occurs only when birds are nesting. There is no brood patch in broodparasitic cuckoos, though some have been reported (e.g. Thompson 1966). In fact no brood-parasitic birds of any kind are known to have a brood patch. The brood-parasitic cuckoos have a shorter incubation period than their hosts (Payne 1977b, Davies 2000). The incubation period is short in many nesting cuckoos of similar body size as the brood-parasitic cuckoos as well. For this reason it is uncertain whether the incubation period has been selected to be short in the brood parasites. It is likely to be so, if only because other brood parasitic

birds such as cowbirds Icteridae and parasitic finches Viduidae also have shorter incubation periods than their host species and have shorter periods than that of related species of nesting icterids and nesting estrildid finches (Payne 1977b). The egg of a parasitic cuckoo is held for a day in the oviduct of the female before she lays.This head start in development of the embryo contributes to shortening of the incubation period of these brood parasites (retention of the egg is unknown in other broodparasitic birds, Payne 1965, 1973a, 1977c, 1989). When a female cuckoo lays an egg, the embryo is developed to a 24-hour stage where the primitive streak can readily be seen on the yolk surface within the egg (Perrins 1967, Vernon 1970a, Liversidge 1971, Payne 1973a). Davidson (1886) appears to have been the first to note that a cuckoo egg laid in a parasitized nest is more developed than the host eggs. Retaining the egg in the oviduct for a day is also useful when a female cuckoo finds that

Figure 8.7. Egg size (g) and incubation period (days) in cuckoos.

Breeding biology and life histories 129 a target nest is lost to a predator, as she then has time to find another nest. Brood-parasitic cuckoos have short incubation periods, with 12 days in the smaller species and 15 in the larger species (Figure 8.7), a general trend in birds with about the same body size (Calder 1984). Nesting cuckoos also have short incubation periods when compared with other nesting altricial birds. There is no clear difference in incubation period between brood-parasitic and nesting cuckoos, when the longer time for incubation of larger eggs is taken into account.The brood-parasitic Clamator have an incubation period about the same as the small Cuculinae (Coccyzus species). Most of the variation in incubation period in cuckoo species is due to the size of the egg and not to whether the cuckoo is a brood parasite. In addition to a short incubation period, the developing cuckoo embryo is more tolerant of cooling than the embryo of most birds. In Australia,

an egg of Horsfield’s Bronze-cuckoo was taken from a nest along with eggs of the host Malurus sp. fairywren.The cuckoo embryo was still alive in the egg 24 hours later, while the fairy-wren embryos were dead (Serventy and Whittell 1976). Tolerance of developing cuckoos to cool temperatures continues into the nestling period, as discussed below.

Development of the young Cuckoos are altricial birds. Nesting cuckoos remain in the nest while they grow.They depend on their parents or foster parents to provide food, and they grow rapidly. Nestling arboreal cuckoo young (Little Cuckoo,Yellow-billed Cuckoo, Black-billed Cuckoo, Squirrel Cuckoo, Smooth-billed Ani, Groove-billed Ani) grow rapidly and leave the nest in as little as 10 days, and can climb about when they are disturbed and fledge earlier if at risk from predation. The larger ground cuckoos including

Figure 8.8. Egg size (g) and nestling period (days) in cuckoos.

130 The Cuckoos Coral-billed Ground-cuckoo, Pheasant Coucal, White-browed Coucal and Greater Roadrunner fledge in 17–20 days, a longer time than in the smaller nesting cuckoos. The nestling period of brood-parasitic cuckoos is longer than that of nesting cuckoos (Figure 8.8). Nestlings of the small brood-parasitic cuckoos (Chrysococcyx basalis, C. lucidus, C. caprius, C. klaas, C. cupreus,Tapera naevia) take 18–20 days to fledge, Common Cuckoo take 18–20 days, Common Koel take 20–28 days, Great Spotted Cuckoo take 22–26 days, and Channel-billed Cuckoos take as long as 24 days to fledge. The nestling period of small brood-parasitic cuckoos is twice as long as the nesting cuckoos of the same body size, and they often monopolize the parental care of their foster parents for a few more weeks after they leave the nest. In the brood-parasitic Jacobin Cuckoo and Levaillant’s Cuckoo the nestling period is shorter than in Great Spotted Cuckoo or other parasitic cuckoos. The nestling period in these Clamator is

the same as that observed in the nesting cuckoos, notably the New World malkohas (breeding information is less well known for the Old World malkohas). The short nestling period in Clamator may be explained by their choice of hosts and the advantage to fledge together with the host young in the mixed-species brood. Nestling Clamator often grow up together in the nest with their foster young, rather than evicting the host eggs and young like most other brood-parasitic cuckoos do. The babbler host young of Jacobin and Levaillant’s cuckoos fledge early from the nest (Gaston 1976, 1977, Hustler 1977b), and in contrast the crow host young of Great Spotted Cuckoo remain in the nest for more than three weeks (Cramp et al. 1994). In these cuckoos it appears that the nestling period is adjusted within the brood to match the nestling period of their host young, and by fledging together in the brood the young cuckoos are likely to be fed and receive parental care after the foster parent leads its brood away from the nest.

Table 8.1 Size at fledging in cuckoos in relation to parental behavior. Species Chrysococcyx basalis Chrysococcyx caprius Chrysococcyx lucidus Clamator glandarius Clamator jacobinus Cuculus canorus Cuculus clamosus Cuculus solitarius Eudynamys scolopacea Pachycoccyx audeberti Centropus phasianinus Centropus senegalensis Centropus sinensis Coccyzus americanus Coccyzus erythropthalmus Crotophaga ani Dasylophus superciliosus Geococcyx californianus a Fledging weight/adult weight. n, nesting cuckoos, reared by their own parents. p, brood parasite.

Weight (g)a 20.9/23 (91%) 35/38 (92%) 22/25 (88%) 110/134 (82%) 56/80 (70%) 90/112 (80%) 62/90 (69%) 60/76 (79%) 125/225 (56%) 83/110 (75%) 123/302 (41%) 125/169 (74%) 160/310 (52%) 38/60 (63%) 23.8/50 (48%) 100 (35%) 60/117 (51%) 150/320 (47%)

Breeding behavior (p, n) p p p p p p p p p p n n n n n n n n

Breeding biology and life histories 131 Development of young cuckoos can also be compared in terms of their size at fledging. Growth rates in altricial birds vary with their altriciality and precociality (with higher growth rates in altricial birds which complete more of their development in the nest) and tend to be lower for birds of larger adult body size (Ricklefs 1973, Starck and Ricklefs 1998). Nestling brood-parasitic cuckoos continue to grow until they nearly reach adult body size, then they leave the nest. Nesting cuckoos such as roadrunners and anis fledge before they are fully grown. In Table 9, the fledging weights, adult weights and breeding behaviors are extracted from the cuckoo species accounts.Young of the brood-parasitic cuckoos stay in the nest until they are well grown, with most species waiting until they are at least 75% of adult body weight. The nesting cuckoos for which there is data fledge when they are only half grown. Other cuckoos have not been studied. In some species disturbance of the nest to measure the birds can cause the nestlings to fledge prematurely; as in

young Guira Cuckoos that are more than half the adult size when they normally fledge, but the young leave early when the nest is disturbed (R. Macedo, in litt.). Nesting cuckoos are unusual among altricial birds in leaving the nest long before they are able to fly. This is the pattern in most nesting cuckoos, including birds for which no weights at leaving the nest are available as in Piaya cayana and Crotophaga sulcirostris. The brood-parasitic cuckoos are unusual among the cuckoos in remaining in the nest until they are nearly grown. Their foster parents continue to care for the young parasitic cuckoos long after they leave the nest, well beyond the period of post-fledging care in the nesting cuckoos. Even in the small broodparasitic Shining Bronze-cuckoo, the foster Western Thornbills continue to feed their fledged cuckoo for as long as 42 days after fledging (Payne and Payne 1998a), well past the time when thornbill young would be independent and the parents could nest again (Nicholls et al. 2000).

9 Cooperative breeding

Cooperative breeding birds live in social groups. In the group some birds, the “helpers”, help rear the young of other individuals, the breeders. Helpers may provide relief to the breeders by taking part in the care of a brood, and the helpers may increase the number of brood young that survive. This behavior is curious. The helpers, by not breeding themselves, may not be behaving in their own best interest. This altruistic behavior would under most conditions be selected against, if only because other individuals would leave more offspring and the helpers themselves would fail to reproduce (Stacey and Koenig 1990, Ligon 1999). Nevertheless the New World anis and Guira Cuckoo are among the most social of all cooperative breeding birds. Two ideas explain how cooperative breeding is a successful behavior. First, cooperative breeding may relate to the benefits of living in a social group. Group-living birds can find more food and defend their brood against a predator (Payne et al. 1985, 1988, Rowley and Russell 1997). Group-living birds can hold an area against other social groups, as when the resident group is larger than the competing group, and the helpers not only gain resources in the present but also take over this extra land themselves in the future (Woolfenden 1975, Cockburn 1998, Kokko et al. 2000). Another way the helpers can aid their own survival and future breeding success is in gaining experience in parental care (Komdeur 1996). Finally, a “helper” may also be a breeder at the same time, as when a female lays in the nest of the older breeding female (Brown 1987, Cockburn 1998). These complex societies are potential arenas for reproductive conflict: when more than one adult female is in a

group, there is a potential opportunity for all to breed and for a socially dominant female to coerce others to care for her own offspring (Reeve et al. 1998, Johnstone et al. 1999). Second, the helpers are often closely related to other birds in their group. They may be the older offspring from an earlier brood of the parents or they may be otherwise closely related, such as brothers helping their sisters. The more closely related the helpers are to the breeders and to the young they care for, the more likely they are to share the same genes. By providing help to relatives the helpers are indirectly increasing the success of their common genes in the next generation (Hamilton 1964). This is Hamilton’s theory of inclusive fitness. In contrast to early predictions from the theory of inclusive fitness and the evolution of behavior that favors relatives, ideas that were developed in studies of social insects, recent studies of social insects in which several females breed in a nest show a low degree of genetic relatedness among the breeders (Hamilton 1972, 1996, Queller et al. 1988, Strassmann et al. 1991, 1992). Cooperative breeding in certain birds is associated with genetic relatedness and kin discrimination that is based on social experience between siblings reared in the same social group (Emlen and Wrege 1988, 1989, Stacey and Koenig 1990, Emlen 1997). Cooperative breeding also occurs in birds where helpers are not closely related to the brood or parents of the brood; and in a social group where not all birds are closely related, the helpers are as likely to care for young to which they are not closely related as to young to which they are (Payne et al. 1985, Dunn et al. 1995, Dunn and Cockburn 1996,

Cooperative breeding 133 Magrath and Whittingham 1997, Wright et al. 1999, Finn and Hughes 2001). Cooperative behavior involves three behaviors of the “helpers.” First, the birds remain in a social group rather than dispersing to their own territories. Second, they do not breed themselves. Third, they care for the dependent young of other individuals (Wiley and Rabenold 1984, Mulder and Langmore 1993, Green et al. 1995, Emlen 1995, 1997). These behaviors are not rules that all cooperatively-living birds follow, and the cuckoos are creative in the way they breed cooperatively. In the cooperatively breeding cuckoos, the anis and Guira Cuckoo, more than one female often lays in the nest (Vehrencamp 1977, 1978, Loflin 1983, Vehrencamp et al. 1986, 1988, Macedo 1992, 1994, Macedo and Bianchi 1997a,b). Field studies are now in progress to determine the social interactions and reproductive success of individual females in the cooperative-breeding cuckoos and may give answers to the question of whether group living provides opportunities for hopeful breeders, or control over the lives of the younger females by socially dominant females in the group, or both kinds of reproductive success. 1. Guira Cuckoos live in open habitats in South America.They often lay in the nests of other birds. Because their eggs appeared in the nests of other birds, not only passerines but also such unlikely birds as caracaras, there was doubt that these cuckoos built their own nests, although nesting behavior had been reported in the field (e.g. Euler 1867). When the guiras were seen to build their own nests in captivity in the Copenhagen Zoo, the doubters conceded that the birds were not obligate nest parasites (Leverkühn 1894). Guiras sometimes nest in the same bush with Smooth-billed Anis C. ani (Euler 1867). Azara (1809: 23) reported that both species often lay in the same nest: “J’ai vu plusieurs de ce nids dans lesquels étaient des oeufs des deux espèces”.Azara’s observations have often been interpreted as two species sharing parental care at a common nest (e.g. Sclater and Hudson 1889, Sick 1993) but no one else appears to have seen the cuckoos do this. Although Guira Cuckoos breed together in a communal nest, there is much conflict and com-

petition among the breeding adults. A breeding group has as many as ten females and several females lay in a nest where the common clutch has as many as 20 eggs. Adults perch on the nest, remove an egg from the nest and drop the egg on the ground, and they also fly off with the egg and bury other eggs in the nest. In Brazil, 32% of all eggs laid were lost or disappeared, and 76% of these eggs were found on the ground below the nest, tossed out by other adult guiras during laying and incubation. Other eggs are ejected shortly before hatching, and chicks are sometimes removed and killed by infanticidal adults (Macedo 1992, 1994, Quinn et al. 1994, Macedo and Bianchi 1997). In spite of the adults’ conflict over breeding and interference with each other’s nesting success, the adults also “cooperate” in incubating the clutch, feeding the brood and guarding them from predators. Usually all young in a nest hatch on the same day. At several nests all the adults attended the nest and cared for the brood, and the young were fed equitably except in the larger broods, where some young were underfed. Nests of the larger breeding groups did not fledge more young, and the number of young fledged per adult did not differ among the small, medium and large breeding groups. The probability of fledging at least one young did not vary with the number of adults. In groups with more adults, more eggs were laid but so were the number of eggs that were lost. After the brood hatches, an adult remains in attendance in the nest tree while other adults forage away from the nest. When a threat approaches the nest, the attendant gives loud staccato calls, and the other group members converge to the nest in defense (Macedo 1992, 1994). The genetic markers of young Guira Cuckoos were compared with the markers of the adults. In this study, chicks in a nest with two or more breeding pairs were often the offspring of more than one pair of adults. The chicks were usually half-siblings but not full-siblings, indicating that the adults were not sexually monogamous. Two or more females accounted for more than half the young in a communal nest. Although a single female was not responsible for most of the young in a nest, a dominant female may leave more than her share of

134 The Cuckoos young. Adults in a breeding group were sometimes related and sometimes not related to each other, and there appeared to be no strong kinship among members of a group (Quinn et al. 1994). The eggs of a female are as variable as the eggs of different individual females in their size and appearance as determined by egg proteins (Cariello et al. 2002, in press). Guira Cuckoos are difficult to study: they nest high in a tree and the birds are hard to capture and mark as individuals. Most inferences about social relationships within a group has come from molecular genetic studies rather than from observation of marked birds. 2. Groove-billed Anis breed in groups or pairs in open habitats with scattered trees. Most groups have two or three pairs. Within the social group each female has a mate with which she spends time in a social bond. Females lay eggs in a single communal nest. In the breeding season all members of a group use the territory; in the dry season the birds move into the areas of other groups, or abandon their territories and move to the edge of a forest. Anis in social groups with a larger number of adult birds have a higher survival rate in the breeding season, but not a higher average number of young fledged; the larger groups produce more fledglings although not in proportion to the number of females in the group. The eggs of early-laying females are tossed from the nest by the later-laying female, who may be socially dominant in the breeding group, in observations where eggs were identified by size and shape (rather than by molecular genetics, not yet determined in this species) and where social interactions between females were seen at the nest (Vehrencamp 1977,Vehrencamp et al. 1986). Males brood at night, females in the daytime.Although the males that mate with dominant females may leave more young (their mates remove the eggs and young of other females), these males are more likely to be taken by a predator. One risk is the nocturnal carnivorous bat Vampyrum spectrum which find anis on the nest and in roosting sites and kills and eats them (Vehrencamp et al. 1977). During the daytime, brooding females on the nest are occasionally fed by another ani, apparently her mate. Juvenile anis when reared in an earlier brood sometimes feed the young in the second brood (Skutch 1959, 1983).

After anis fledge and are independent, the young birds usually disappear from their group and move out of their natal area. Nevertheless a few males remain as nonbreeders or become breeders in their natal group, when another bird disappears or when their group size enlarges. Most anis do not breed in the group where they were born, and when they do it is only males that stay and breed in their natal group (Bowen et al. 1989, Koford et al. 1990). 3. Smooth-billed Anis breed in communal nests where as many as 10 adults attend to a nest and only one adult broods at a time. Eggs are sometimes layered in the nest with the lower layers covered with leaves (Euler 1867). Breeding females compete for the position of eggs in the nest.The females that lay later cover the eggs of the early-laying females with green leaves. Egg burial occurs in nests where two or more females lay. As a result, many eggs that are laid early do not hatch while eggs of the laterlaying females have a higher hatch success. The young from early eggs have a head start in development and are larger in a mixed-age brood, and young of the last-laying females are more likely to starve. In some nests, the last-laying female and her mate desert the nest, and their young are reared by the earlier-laying pairs (Loflin 1983). Females stay on the nest and are fed by their mates (Skutch 1983, Köster 1971). This cooperative behavior between members of a pair may allow the female to remain on the nest and guard her eggs from her rivals.The fledged young often remain in the group for a season and help care for later broods of their natal group as sentinels, and occasionally feed them. Groups with many anis tend to have larger territories, perhaps because the extra birds can exclude other anis from the group territory (Quinn and Startek-Foote 2000). In addition to communal nesting in their social group, Smooth-billed Anis sometimes lay in nests of other ani social groups.A female may lay in another group’s nest when her own nest is lost due to predation or to bad weather. These intruding females and their mates are attacked by resident breeding anis. If the intruders succeed in laying, their eggs are often buried by the resident group if they are laid early, but are incubated if they are laid when the last resident female has begun to lay (Loflin 1983).

Cooperative breeding 135 And anis sometimes nest as single pairs with one female laying in the nest, as they regularly do in Florida (Loflin 1983). All cooperatively breeding cuckoos not only nest together, they are social through the year, remaining in groups when they feed and roost. They perch pressed in contact against each other, and they roost piled in layers on the backs of their fellows.When they are in contact side by side, they preen each other (Sclater and Hudson 1889, Wetmore 1926, Barbour 1943, Wetmore 1968, Skutch 1983, Figure 9.1). Allopreening in cuckoos is known only in these social species, so it appears to be a social adaptation for living in groups.These birds have a high rate of infestation by parasitic insects, and allopreening perhaps is a social adaptation to the shared mallophaga or feather lice that are transmitted among members of the social group, much as allopreening is in Old World primates. Mallophaga are also common in their nests, and in Guira Cuckoos where a group may have as many as five nests in a season (R. Macedo, in litt.), most groups move to a different tree for the second nest.This change of nest site may decrease the nest infestation by the feather lice and other insects or nest detection by predators (Abrahamovich and Cicchino 1985). Cooperative breeding in these

cuckoos is of concern not only because of conflicts of interest between adult females in the group, but also because of the functional overlap of cooperative breeding and brood parasitism. In both the cooperative and brood-parasitic breeding styles, one bird cares for the young of another to which it is not closely related. One question in all these cooperative-breeding cuckoos is their relatedness, where birds in a social group may not be close kin. In Groove-billed Anis, birds often disperse from their natal area, and as a result the breeders may not be close relatives (Vehrencamp et al. 1986), and Guira Cuckoos in a group are not closely related, at least in a preliminary genetic study (Quinn et al. 1994). In Smoothbilled Anis, genetic markers provide evidence of complex relationships, with suspected cases of polygamy, extra-pair fertilization, intraspecific brood parasitism, or all these behaviors (Quinn and Startek-Foote 2000). Living in a group may benefit the individual birds in the group, and helping behavior may be a social mechanism that allows these birds to remain in the group and be tolerated by the other group members. A second question for future studies is the reproductive success of each female in a group. Early field studies estimated the contributions of each

Figure 9.1. Social behavior in group-living Guira Cuckoos Guira guira (after Fandiño Mariño 1986).

136 The Cuckoos female cuckoo by size and shape of the eggs (Vehrencamp 1976, Loflin 1983,Vehrencamp et al. 1986), and not by the more reliable molecular markers of identification of eggs laid and matched to individual females (Cariello et al. 2002). A third question is whether one female controls the reproductive success of other females in a group, or whether each female has the incentive to succeed within a changing social system (Reeve et al. 1998, Johnstone et al. 1999). In Guira Cuckoos the sequence in which females lay within a communal nest can change from nest to nest, and even the last-laying females sometimes lose eggs, so there is no obvious long-term social rank in relation to laying sequence (Macedo et al. 2004). On a broader scale, these cooperatively-breeding cuckoo species are each other’s closest relatives. Even though they live in different habitats ranging from the swamp forests of Greater Ani to the scrubbier habitat of Groove-billed Ani and Smooth-billed Ani

and the savannas of Guira Cuckoo, the crotophagine cuckoos share features of social behavior, nestbuilding, competition and cooperation among females, and infanticide. These cooperative cuckoos are an evolutionary parallel with the Australian fairy-wrens, a clade in which all 13 species are cooperative breeders and live in habitats that range from scrub to rainforest to desert. In fairy-wrens a sufficient phylogenetic explanation of their behavior is that their common ancestor was a cooperative breeder (Russell 1989, Edwards and Naeem 1993, Rowley and Russell 1997). This explanation is more likely than the idea that cooperative breeding is somehow caused by a common habitat across all species with this behavior, as there is no common ecological feature that is associated with the behavior. The same perspective applies to the crotophagine anis and Guira Cuckoo. In one sense, the most likely explanation of why each is a cooperative breeder is that their common ancestor had this behavior.

10 Brood parasitism

Brood parasitism is the most fascinating feature of cuckoo biology. Cuckoo parasites use other species to rear their young, and how the behavior evolved as a reproductive strategy demands an explanation (Darwin 1859, Hamilton 1972). Brood-parasitic cuckoos lay their eggs in the nests of other species, their hosts, and the nesting pair that rears the cuckoos are the foster parents.These birds incubate the cuckoo eggs along with their own, and when the cuckoo egg hatches they feed the young cuckoo. The foster parents forfeit their own reproductive success when the nestling cuckoo removes the hosts’ eggs or nestlings. Some cuckoo eggs mimic the background color and spot pattern of the eggs of their host, and the eggs of a single species of cuckoo may vary among females, each matching the eggs of a different species of host. Host egg mimicry in cuckoo eggs has been known for a long time. In Europe, the eggs of Common Cuckoo Cuculus canorus often match the eggs of the host nest where they lay, as noted in 1850 (Kunz 1850), described in detail by Baldamus (1853, 1892), and explained as a result of selection by the host, which removes from the nest any eggs unlike its own, by A. Newton (1896). In fact, Newton cited naturalists’ knowledge of cuckoo egg mimicry as early as the second century CE. Later biologists have reported new observations and developed ideas about the evolutionary process of natural selection in brood parasitism (Rey 1892, Swynnerton 1916, Rensch 1924, Makatsch 1937, 1955, Baker 1942, Brooke and Davies 1988, Davies et al. 1989a, Øien et al. 1995, Rothstein and Robertson 1998, Davies 2000). The Common Koel in India was the first bird described as a brood parasite. Four thousand year

old Sanskrit literature referred to these birds as “anya-vapa”; “anya” indicating “other” and “vapa” indicating “reared” or “brought up”; the term “anya-vapa” means the koels were “reared by others”. In a later Vedic drama, the Sakuntala by Abhijnanashakuntalam, dated about 375 AD, is the passage, “Koels manage to have all their young reared by other birds before they move into the air.” The same drama refers to the “other” as House Crow Corvus splendens, called “para-bhrit,” or “one who rears others” (Friedmann 1964b). Later the Moghul Emperor, the naturalist Jahangir (1569–1627) wrote: “A strange thing about the Koel is that it does not bring up its young from the egg, but finding the nest of the crow unguarded at the time of laying, it breaks the crow’s eggs with its beak and throws them out, and lays its own in the place of them and flies off. The crow thinking the eggs its own, hatches the young and brings them up. I have myself seen this strange affair at Allahabad”. Jahangir also described the Jacobin Cuckoo as a brood parasite: “I have seen in Kashmir that the papiha [the Jacobin] lays its eggs in the nest of the ghaughai [a babbler] and the ghaughai brings up its young” (Ali 1927). Brood parasitism in Common Cuckoos in Europe has been common knowledge since the classical period of the Greeks and was described in the natural histories of Aristotle. The evicting behavior of nestling cuckoos and other broodparasitic behaviors of these birds in Britain was described in detail more than 200 years ago ( Jenner 1788). In addition to these accurate observations, there have been misunderstandings concerning these cuckoos. “Cuckoldry” derives its meaning

138 The Cuckoos from a misunderstanding of cuckoo parasitism. Centuries ago it was thought that the female cuckoo mated with the male of the foster species, and this would “explain” both the color and the small size of the cuckoo’s eggs (Stresemann 1934, Rothschild and Clay 1952). It was common folk knowledge that the cuckoo “changeling” in the nest— substituted for the real offspring of the pair—is not the offspring of the pair that gives the parental care. In the contemporary sense of cuckoldry, rearing the young of another bird is what happens when a female mates with one male and her social partner rears a youngster that is genetically not his own. References to cuckoo changelings in Shakespeare’s plays were concerned with parents rearing young other than their own offspring, in an age when civil wars were fought and England was divided over kinship claims; the question of parentage and royal succession was a matter of life and death (Payne 1997b).This concept of cuckoldry has been central to the development of ideas in the late twentieth century about the evolution of parental care (Hamilton 1990). In the New World, brood parasitism in cuckoos was first reported about a hundred years ago (Friedmann 1964b), a century later than brood parasitism was described in the cowbirds. Hartert and Venturi (1909) mentioned an American Striped Cuckoo Tapera naevia parasitizing a spinetail Synallaxis nest in Argentina, but they did not imply that this was news to them, so brood parasitism may have been documented earlier. Azara (1809) mentioned nothing about parasitism of the New World cuckoos, and apparently the earliest record is from Suriname (then, Dutch Guiana) when the Penard brothers collected eggs of this cuckoo in the nests of other birds from 1898 to 1904 (Penard and Penard 1908). The next revelation of broodparasitic cuckoos in the New World was an observation by von Ihering (1914). He found a strange egg in the nest of a Black-backed Water-Tyrant Fluvicola albiventer in 1913, and guessed that it may be the as yet unknown egg of Pheasant Cuckoo Dromococcyx phasianellus.A female cuckoo taken the next year had in her oviduct an egg that matched the egg in the tyrant nest, clearing up the identity of these eggs (Naumburg 1930).

Identification of cuckoo eggs is a problem for field ornithologists as well as for the nesting hosts. Skead (1951) recognized the problem in South Africa,“When attributing cuckoo eggs to a particular species in an area where there are several species of cuckoos one is faced with the inevitable question, ‘How do I know they belong to that species?’ Unless laying is seen the only way to answer the question is to allow the egg and its contents to mature. But predatory Nature so often intervenes that one’s efforts are usually frustrated. I feel as sure as it is possible to be without absolute proof that the eggs of the Black-crested Cuckoo are pure white, shiny and nearly round. All those found in bulbul nests have been thus and have contrasted strikingly with the eggs of their hosts both in size and colour.” In India where several cuckoo species occur and two or more may overlap in their use of hosts, it is uncertain which eggs go with which species of cuckoo. The eggs identified by Baker (1906, 1907, 1908, 1934, 1942) and other egg collectors may not have been correctly identified. Even though Baker described eggs taken in the oviduct of females, the cuckoo itself was rarely saved as a museum skin, so doubts remain. Also, eggs that were found in host nests did not hatch, and the eggs collected were labeled years after they were taken from the nest (Harrison 1969, Becking 1981). A cuckoo egg can be identified by measuring and photographing the egg, then letting it hatch and develop into a feathered young cuckoo; or by sampling the eggshell or the nestling, with a feather or a drop of blood to compare its genetic sequence with that of a known cuckoo, as we have done with brood-parasitic finches (Payne et al. 2000, 2002). Identification of nestling and juvenile cuckoos has been a problem as well. Some errors, as in juveniles of Plaintive Cuckoo being mistaken as juveniles of Violet Cuckoo (Baker 1906, 1927,Wells 1999), can be traced back to earlier plumage descriptions, in this case to Hume (1875). Only recently have cuckoos been handreared from the egg or nestling stages to determine the juvenile plumage (Higuchi and Sato 1984). Fieldwork is needed to determine the basic life history details of many brood-parasitic cuckoos. The mysteries of brood parasitism in the cuckoos have led to evocative book titles such as

Brood parasitism 139 Cuckoo problems (Baker 1942), The Cuckoo’s secret (Chance 1922) and The truth about the Cuckoo (Chance 1940). We now phrase these mysteries as biological questions, and ask how birds without a family life of their own know what species they are? And, how they can find a nest where their own young will be reared? These questions lead us to consider how birds recognize their own young, and why the hosts rear young that are not their own. Do the brood parasites have a real effect on the breeding success of their hosts? Do the hosts distinguish their own young and the young of the cuckoo? Why do the hosts care for these foundlings? How have cuckoos evolved eggs that mimic the color and pattern of the eggshell of their host species, and how do their hosts respond to this challenge? Finally, how did the behavior of leaving their young in the nests of other birds evolve in the first place?

Behavior of the female cuckoo How females find a nest and lay in it During the breeding season female Common Cuckoos often perch on high branches and watch small songbirds as these make trip after trip with nest material to their nest sites. Nests of a host species are more likely to be parasitized when they are near regular cuckoo perches than when far from a cuckoo perch (Øien et al. 1996). Female cuckoos also search through the vegetation where songbirds are likely to nest. Cuckoos respond to calls of the hosts and may find a nest as the host gives alarm calls that increase in number and loudness when the cuckoo gets closer to the nest (Seppä 1969). We too find songbird nests in the same way by keeping track of alarm calls of the nesting birds (Payne and Payne 1998b). A female cuckoo visits the nest when the host is not there. She often takes an egg of the host in her bill, or she perches on the nest rim and swallows egg after egg as she takes them from the nest. She then sits on the nest and lays her egg directly into the nest. Egg-laying is rapid, 10 seconds or less; then the female flies away. With the best of timing she lays when a host egg has been laid but before

incubation has begun (Gurney 1897, Chance 1922, Molnár 1944, Gärtner 1981a,b, 1982, Wyllie 1981, Yoshino 1988, 1999). Edgar Chance first filmed Common Cuckoos in the nests of Meadow Pipits Anthus pratensis at Pound Green Commons in Worcestershire, England, in 1918–1921. By knowing where the pipits were nesting and by disrupting a few nests causing pipits to nest again at a time convenient for himself and the cuckoo, Chance was able to anticipate where the female cuckoo would visit and lay, in late afternoon on every other day, and he was there with his camera before the female visited the nest (Chance 1922). In earlier years it was thought that a female cuckoo laid her egg elsewhere and carried it to the nest in her bill, but reports of a cuckoo carrying an egg in the bill likely were of a female as she removed a host egg from the nest. In Britain, a committee was appointed to solve this cuckoo problem (Snow 1992).There was no final committee report, nor was one necessary: Chance’s films documented that cuckoos lay directly into the nest. Adapting current field methods in their observations, a Norwegian research group set up videocameras to make continuous recordings at nests of Reed Warbler Acrocephalus scirpaceus every day through the laying period.Videocameras were often in the right place at the right time: Common Cuckoos were recorded as they laid in the monitored nests (Moksnes et al. 2000). Cuckoo behavior here was similar to the behavior in England, although the female did not avoid visiting a nest when the host was nearby, and she often laid in the presence of the host. Cuckoos took host eggs from nests that were never parasitized as well as nests where the cuckoo laid. Parasitized warblers were more likely to remove the cuckoo egg from their nest when they had seen the cuckoo at their nest. In video recordings of Horsfield’s Bronzecuckoos Chrysococcyx basalis in Australia, a female visits the host nest alone and sometimes lays in a nest that she cannot enter. She glides from a perch to the nest site and lays while perched on the rim of the nest, or perched above it. In nests that are enclosed or in a hole the female pushes her abdomen into the nest cavity and lays directly into the nest (Brooker et al. 1988).

140 The Cuckoos In a few cuckoo species the female lays in the host nest with the help of her mate. In Jacobin Cuckoo Clamator jacobinus in South Africa the male often distracts the host from the nest, the host chases the male cuckoo, and the female slips into the unguarded nest and lays her egg (Liversidge 1971). In Common Koel Eudynamys scolopacea the male may distract the crow and allow the female to lay in the nest, though she usually visits the nest alone (Lamba 1969, 1976). Channel-billed Cuckoo Scythrops novaehollandiae pairs sometimes approach the nesting host together; the male distracts the host while the female lays in the unattended nest (Crouther 1980, Larkins 1994b).

Why cuckoos parasitize certain species Several biological and ecological factors affect the suitability of bird species as hosts (Sealy et al. 2002). These include whether a potential host species is likely to accept the cuckoo egg, incubate it, care for the young and meet its nutritional needs. Insectivorous songbirds with short nestling periods provide the best parental care because these hosts bring food for the nestlings at a high rate. In addition, the number of host pairs in the local population may affect the choice of a host, as a common host provides more nests for a cuckoo to parasitize. In a comparative study of songbird species available to Common Cuckoos in Europe, Soler et al. (1999b) determined that nests of the more abundant species were more likely to be parasitized. Other factors that affect the rate of parasitism were the length of nestling period, the time of breeding season, and the nest structure and location (open nesting species are more likely to be parasitized than cavity nesting species). Friedmann (1967) suggested that competition among cuckoo species for parental care may have led the brood-parasitic cuckoos to focus on different hosts. Although many cuckoos are host specialists, in some regions two or more cuckoo species overlap in their choice of hosts. A nest of Western Thornbill that we found on Gooseberry Hill in Western Australia had two cuckoo eggs, one an unspotted copper egg of Shining Bronze-cuckoo

and the other a spotted whitish egg of Horsfield’s Bronze-cuckoo (Payne and Payne 1998a). Splendid Fairy-wrens are used by Shining Bronze-cuckoo and Horsfield’s Bronze-cuckoo in South Australia; Jungle Babblers Turdoides striatus are used by Jacobin Cuckoo and Common Hawk-cuckoo in India (sometimes both kinds of eggs appear in the same nest: Baker 1906); some warblers and sunbirds are used by two glossy cuckoos Chrysococcyx species in Africa, and so on (Baker 1942, Payne and Payne 1967, Becking 1981, Brooker and Brooker 1989a, Brooker 1992).There is no evidence that a cuckoo species uses more host species in the absence of another cuckoo species than when it is the only brood-parasitic species, as would be expected from the competition hypothesis. Also two species of brood-parasitic cuckoos do not respond to each other’s songs as would be expected if they were strongly competitive. The host specificity of cuckoos is better explained by the success in parasitism of each host species by a cuckoo species, survival of cuckoo eggs and nestlings, imprinting, and in some species the coevolutionary match of cuckoo eggs and host eggs, than by interactions between cuckoo species.

Does a female imprint on her foster species? Imprinting is a behavioral process in which the young bird forms a social attachment to an object to which it directs its adult behavior (chooses a mate, or seeks and parasitizes a nest of a socially familiar kind of bird), an attachment that occurs during a critical period in early life, is irreversible, and is generalized to other individuals of the species (Lorenz 1935, Irwin and Price 1999, ten Cate and Vos 1999, Payne et al. 2000). Maintenance of the egg morphs of females in the egg-mimetic cuckoos may involve imprinting. Female Common Cuckoos in Britain and Europe lay eggs that differ from the eggs of other females, and often each female’s eggs match the eggs of a host songbird (Baldamus 1892, Rey 1892). Each female normally lays in the nests of a single kind of host (Chance 1922, 1940, Marchetti et al. 1998). How does a female decide which host to parasitize?

Brood parasitism 141 Imprinting to its own foster parent’s species might be the behavior that maintains a tradition of host specificity in brood parasitism from one generation to the next. A young cuckoo could imprint to her foster species, so that her adult behavior in host choice would be directed towards the species that had reared her. If she formed an association with this host species, the cuckoo may return as an adult to the same habitat and find nests by recognizing the host species and matching a memory of calls of her own foster parents to the calls of nesting birds, then lay in those nests. According to the “gens” theory, a population of Common Cuckoos consists of distinct biological “races” or “gentes” (Newton 1896) that are closely associated with their foster species and mimic the egg color and pattern of this host species. The female cuckoos then seek and choose the nest of this species to lay their eggs. A female cuckoo is host specific, by this reasoning, because she was behaviorally imprinted to her foster parents. Her own experience as a chick directs a female cuckoo to search out the species that reared her and parasitize that same host species. By using a host like the one that reared her, the female chooses a host species that has reared a cuckoo, and she can lay an egg that matched the egg from which she hatched and that has a high probability of acceptance by the same host species. There has been little research with cuckoos to test the idea of imprinting. Löhrl (1979) kept wildreared Common Cuckoos in aviaries. A female reared by Pied (White) Wagtail Motacilla alba saw wagtails outside the aviary, and although she could not see the wagtail nest, the cuckoo laid an egg in the aviary two days before the wagtail laid. Other studies have tested cuckoos reared in different conditions, where the cuckoos had alternate foster species or had alternate habitats. First, Brooke and Davies (1991) cross-fostered nestling cuckoos from Reed Warbler nests into nests of alternate foster-parents, including European Robin Erithacus rubecula, which reared the young cuckoo to independence. Other young were reared by moving the nest, cuckoo and adult warblers into an aviary where the warblers reared the cuckoo. Next year the captive cuckoos were tested to see if they

perched near their own foster species. When they had a choice of Dunnock Prunella modularis, robin or warbler, the cuckoos showed no preference for perching near their foster species, even when one, a female cuckoo, was implanted with estradiol. Second,Teuschl et al. (1998) removed young cuckoos from nests of Reed Warbler and Marsh Warbler A. palustris, hand-reared them with different visual effects (“habitats”) in small cages, then kept the cuckoos in an aviary. Over the next two years the cuckoos were tested in a cage where each could perch near the six visual “habitats”. One cuckoo perched near its natal “habitat”, but in a second test she did not. Neither study tested whether a cuckoo imprints to her foster species, then lays her eggs in the nests of the same host species. The imprinting model is a reasonable explanation for the host-specific preference of female cuckoos. However, the difficulty of breeding cuckoos in captivity, and the dispersal of cuckoos from their natal site to first breeding site in the field, have prevented a convincing test of the imprinting model. The imprinting model has been tested in another broodparasitic bird, Village Indigobird Vidua chalybeata. The indigobirds are host-specific brood parasites, and their nestlings mimic the nestling colors of their normal host species. Indigobirds were foster-reared in aviaries, either by their normal host species, the Redbilled Firefinch Lagonosticta senegala, or by an experimental foster species, the Bengalese Finch Lonchura striata. Captive-reared female indigobirds were tested as adults for host choice. In the aviaries where both finch species were nesting, the females reared by firefinches laid in the nests of firefinches, and the females reared by Bengalese Finches laid in the nests of Bengalese Finches. Females caught in the wild showed the same behaviors as captive-bred females reared by firefinches.The results showed imprinting in which young indigobirds focus their attention on their foster parents, rather than an innate bias for the normal host species, or a nonspecific choice of a host. In these imprinting experiments the females recognized and laid in the nest of the species that reared them, even when the foster species was not the normal host species (Payne et al. 2000). Behavioral conservatism may explain why certain host species are parasitized in one region and not in

142 The Cuckoos others. But even in a population where all females imprint to their normal host species, a female may occasionally lay in the nest of a novel host, as she does when a nest she has been observing is destroyed before the cuckoo has laid; she may then lay in the nearest available nest of another species, which may rear the young cuckoo (Rose 1982). In southern Germany where Common Cuckoos parasitize other host species, 3000 nests of Blackcap Sylvia atricapilla were checked before one was found with a cuckoo; the Blackcap pair reared the young cuckoo (Berthold et al. 1995). In another species, Great Spotted Cuckoo in Spain parasitize Magpie Pica pica and Carrion Crow Corvus corone.The cuckoo eggs in the nests of these two hosts look alike. Some female cuckoos lay eggs (identified by genetic markers that matched the adult females) in nests of both species of host, Magpies as the common host, and Carrion Crows as alternative host when no Magpie nests are available (Martínez et al. 1998a). The same cuckoo species has colonized new host species when these species have been introduced into the normal range of the cuckoos, the House Crow in Israel and the Indian Mynah in South Africa.

Egg color and pattern, and the genetics of mimicry Common Cuckoo eggs often resemble the color and pattern of their hosts’ eggs.The eggs of this cuckoo vary in color and pattern within a population. Some eggs are similar to eggs of one host species and others are similar to those of another host species (Baldamus 1853, 1892, Rey 1892, Oates 1903, Makatsch 1937, 1971, Yoshino 1999). Many hosts recognize the eggs of a cuckoo as different from their own, and these hosts remove or “reject” the cuckoo egg from their nest if it does not resemble their own eggs. The function of cuckoo egg matching is thought to increase the change of parental care by the nesting host. Cuckoo eggs that mimic the color and pattern of the host eggs are more likely to be accepted by the host and incubated along with the host’s own eggs. In field experiments, the mimetic eggs are less likely to be rejected from the nest than are eggs that differ from the host eggs (Swynnerton 1916, 1918, Davies and Brooke 1988, Davies 2000).

Fine tuning of cuckoo egg to host egg by natural selection may lead to specialization, with eggs that match one species of host.There may be a coevolutionary “arms race”, where the host is selected for an ability to discriminate among eggs, and the cuckoo is selected to lay an egg that mimics the hosts’ eggs (Baker 1907, 1942, Jourdain 1925, Livesey 1936, Southern 1954, Davies 2000). In Britain many female Common Cuckoos lay distinct eggs that closely resemble the eggs of the most common host species (Davies and Brooke 1989). Mimicry of the cuckoo eggs to the host eggs is not perfect, and the cuckoo eggs are still distinguishable as cuckoo eggs by color and pattern as well as by size and shape (Cramp 1985: plate 96). In Central Europe the cuckoo eggs laid in the nests of four host species of Acrocephalus warblers all look alike and do not match the differences between the eggs of these host species (Edvardsen et al. 2001), and in Japan the same is true for cuckoo eggs laid in the nests of different host species (Nakamura et al. 1998). In surveys of museum egg collections in Europe only 25–50% of cuckoo eggs are a close match to the host eggs (Harrison 1968, Perrin de Brichambaut 1993). Several authors have noted a difficulty in distinguishing cuckoo eggs from host eggs, and the low match in museum collections may result in part from a bias of collectors having passed over the clutches they did not realize had a cuckoo egg when the egg was a good mimic. Although eggs of Common Cuckoos do not always mimic the color and pattern of the host eggs, the eggs that match have a greater chance of being accepted by the nesting host (Davies 2000). When models of cuckoo eggs were tested in the host nests, certain hosts were more likely to accept the model egg if it was like the cuckoo egg normal for their nest. In nesting hosts that are parasitized by mimetic cuckoo eggs (Meadow Pipit, Pied Wagtail, Reed Warbler), the model eggs were either accepted or rejected (the eggs were removed from the nest, or the nest was deserted). Meadow Pipit and Reed Warbler accepted the model eggs that matched their own cuckoo egg morph, but rejected the other eggs. Pied Wagtail rejected the model eggs half the time when these matched their own eggs and more than half when they did not match, but

Brood parasitism 143 the difference was not significant.The fourth host, European Robin, is parasitized by cuckoo eggs, like the cuckoo eggs in nests of Pied Wagtails (much like the robin eggs); while the fifth host, Dunnock, is parasitized by spotted whitish cuckoo eggs, although Dunnock eggs are unspotted blue. These last two host species accepted model eggs whether or not they were unlike their own eggs (Davies and Brooke 1989a).The tests demonstrated the importance of egg mimicry in acceptance of cuckoo eggs by certain species of nesting hosts. The tests also demonstrated the discriminating ability of the hosts. Avoidance of predation has been proposed to explain the function of egg mimicry.Wallace (1889) suggested that cuckoo eggs in the nests of birds with matching eggs would be less conspicuous to a predator. Egg predation could involve other cuckoos; as when a female cuckoo lays in a nest, a second female would be less likely to remove the rival first cuckoo’s egg if the first cuckoo egg matched the host eggs (Brooke and Davies 1988, Brooker et al. 1989b, Brooker et al. 1990).There is no evidence of a higher rate of predation in clutches where a cuckoo egg was unlike the hosts’ own eggs. In field experiments there was a tendency for second cuckoos to remove a model cuckoo egg when it did not match the host eggs, but the tendency was not strong enough to be statistically significant. In addition, few nests have second cuckoo eggs, so the risk of egg predation by a second cuckoo is low and selection to match to escape egg predation by a second cuckoo would not be as strong as discrimination by the host (Davies and Brooke 1988). Female and male genetic contribution to the egg color and pattern of cuckoo eggs once puzzled ornithologists. Current genetic research has demystified this puzzle. A female bird produces the color and pattern on the egg while the eggshell is in her oviduct; the male that fertilizes the egg has no effect on the appearance of the eggshell (Brooke 1989). Certain genes are transmitted to the offspring only by females and not by males. One set is on the sex chromosomes, a pair that differ in size and shape in one sex but match in the other sex. In most birds, including the cuckoos, females are heteromorphic for sex chromosomes (females are WZ, males are

ZZ), in contrast to most mammals in which males are the heteromorphic sex (males XY, females XX) (Punnett 1933, Ray-Chaudhuri 1967, 1973, Waldriguez and Ferrari 1979, 1980, 1982,Waldriguez et al. 1983, García-Moreno and Mindell 2000, Graves and Shetty 2001). In birds, genes that are transmitted from mother to offspring (and not from father to offspring) occur on the W chromosome (Graves and Shetty 2001). Genes that are transmitted in this way also occur on mitochondria, outside the nucleus (Avise 1994). Genes that are transmitted only along female lineages behave much like the female brood parasites do when they lay their eggs in nests of the same kind of hosts from one cuckoo generation to the next (Sorenson and Payne 2002). In mammals the X chromosome carries about 10% of the nuclear genome and has an estimated 3000 genes; the short Y has fewer, some with counterparts on the X and some unique to the Y and expressed only in males (Graves and Shetty 2001, Klein and Takahata 2002). In birds, the Z comprises 7% of the nuclear genome; the W has much less. In the only bird (chicken) that has been gene mapped, the DNA of the terminal PAR (pseudoautosomal) region of the W and Z are the only parts that recombine. Genes on W and Z differ in sequence detail and do not recombine in meiosis (Fridolffson et al. 1998, Ellegren 2000, Graves and Shetty 2001), and so the genes on the nonrecombining region of W are transmitted only from mother to daughter. In principle, genes coding for traits that are expressed only in one sex should disproportionately have their addresses on these sex chromosomes (Roldan and Gomendio 1999). It is unknown whether W genes affect in birds the color end pattern of eggshell pigments.The mitochondrial genes are unlikely to code differences in eggshell color and pattern. The coding functions of mitochondrial genes are well known: mitochondrial gene content consists of 2 ribosomal RNA genes, 22 transfer RNA genes, and 13 protein genes that code for subunits of enzymes that function in electron transport and ATP synthesis (Moritz et al. 1987, Desjardins and Morais 1990, Mindell et al. 1999). In birds the eggshell pigments are porphyrins rather than melanins, and apart from domestic fowl there is no information about the

144 The Cuckoos developmental genetics of these porphyrin pigments (Lang and Wells 1987, Miksik et al. 1996). Do mother and daughter birds lay the same kind of egg? There are no observations in cuckoos (Hamilton 1972), but there are in another bird. Great Tit Parus major daughters have eggshell patterns like mother and maternal grandmother, but not like paternal grandmother (Gosler et al. 2000). The observations are consistent with the idea that eggshell pattern is inherited along a maternal lineage in these birds. To test whether Common Cuckoos reared by different hosts differ genetically, and whether cuckoos reared by the same host share the same genetic profile, Gibbs et al. (2000) compared mitochondrial genes of cuckoos in the nests of their main host species. In Britain, nestling cuckoos were sampled in the nests of Reed Warbler, Meadow Pipit and Dunnock. Each variant gene was found mainly in nestling cuckoos with a single species of host. These genes differed between localities, and cuckoos in the nests of one host species (Reed Warbler) in different localities shared some gene variants. Because mitochondrial genes are inherited along the maternal line, the variation among cuckoos in the nests of a host species indicates that female cuckoos have switched hosts from time to time. It also indicates that some races of the cuckoos have more than one origin, perhaps from times when separate cuckoo lineages colonized the same host species: multiple colonizations can account for the different genes in cuckoos in the nests of the same host species. In Japan, nestling cuckoos were sampled in the nests of Great Reed Warbler and Azure-winged Magpie Cyanopica cyanea. Cuckoos in magpie nests shared the same gene variant with each other. Cuckoos in warbler nests had four variants including the one in cuckoos in magpie nests. Warblers are parasitized through much of their range, whereas magpies are parasitized locally: the genes suggest that magpie-cuckoos were descendants of a single female. In contrast to the mitochondrial genes, the nuclear microsatellites did not show significant differences between cuckoos in nests of different host species in Britain or in Japan (Gibbs et al. 2000). The genetic results are consis-

tent with the field observations that female Common Cuckoos focus on a particular host species and males do not, and females do not tend to mate with males with the same genetic variants (Marchetti et al. 1998). A few other species of brood-parasitic cuckoos have variable colors and markings on their eggs (Plate 17). The Australian cuckoos including Horsfield’s Bronze-cuckoo, Shining Bronze-cuckoo and Black-eared Cuckoo have eggs that are similar to some of their host species, but only the Brush Cuckoo has polymorphic eggs. In southern Asia and Africa, the Jacobin Cuckoo eggs are of a similar blue color to eggs of their host species, but in southern Africa the eggs are unmarked white, unlike any of their host species. Egg color and pattern in the Black Cuckoo appears to be similar across its range in Africa, and these birds parasitize a species group of hosts: bush-shrikes Laniarius that all have similar eggs, usually bluish green with indistinct purplish or brownish spots and blotches across the different species (Fry et al. 2000, Plate 18d). Red-chested Cuckoo eggs vary geographically, unmarked dark brown in southern and East Africa where they parasitize robin-chats Cossypha with brown eggs, and pale blue, unmarked or with fine spots, in other parts of East Africa where they parasitize scrubrobins Cercotrichas; they do not always mimic the scrub-robin eggs (Plate 18e,f). The Chrysococcyx glossy cuckoos have polymorphic eggs, best known in Diederik Cuckoos with eggs of several colors and patterns. Female diederiks lay in the nests of a host with an egg color and pattern like their own (Hunter 1961, Payne 1967, Reed 1968, Jensen and Vernon 1970). Red Bishops Euplectes orix which lay unspotted blue eggs reject model cuckoo eggs that have dense spots or a darker background color than their own eggs (Lawes and Kirkman 1996).Village Weavers Ploceus cucullatus are host species that lay a variety of egg colors and patterns in a single colony (Plate 19). The weavers discriminate against eggs that differ from their own eggs, removing them from the nest, and the proportion of rejections depended on the dissimilarity in color and pattern of the other eggs to their own eggs (Lahti and Lahti 2002).

Brood parasitism 145

How does a cuckoo know its own species? The young brood-parasitic cuckoos do not grow up with their own parents, and they have no early opportunity to learn their own species from other cuckoos.When they are adults, the males apparently develop their song by themselves without having to hear it from an adult. Female mate choice behavior may develop normally whether or not they are exposed to that of others of their own species. Male cuckoos do not imitate the songs or other behaviors of their foster parents, and in consequence the female has no information of whether males were reared by the same host species that reared her.As far as is known, a female innately recognizes a conspecific male as a potential mate. Then on the basis of his song rate and other behaviors that indicate his physical condition and ability to defend an area, and on his attraction to the female and the interaction between the pair, she decides to mate with him. Although they are not reared by their own family, the social behavior of certain brood-parasitic cuckoos might be affected by other cuckoos, either their nestmates or older cuckoos. Great Spotted Cuckoos often lay more than one egg in a host nest (Friedmann 1964) and the young cuckoos are often reared together in the nest. Soler and Soler (1999) placed young cuckoos into the nests of Magpies in an area where the young had no social contact with adult cuckoos.When it was reared alone in a nest, a fledgling did not join other fledged young cuckoos; but when it was reared together with another cuckoo, it associated with other fledged young cuckoos. An influence of early social life on the mate choice of adult cuckoos is unknown. Because Great Spotted Cuckoos are sometimes reared without another cuckoo in the nest, these cuckoos can select a conspecific mate whether or not they experience early life with another cuckoo. In most species of brood-parasitic cuckoos the young grows up alone in the nest, and it evicts any eggs and young of another cuckoo or the host, or the nestling cuckoo kills the host young with the hooks on its bill. As a consequence, we think that adult cuckoos mate with conspecific adults

whether or not they interact with other cuckoos when they are young.That is, the choice of a conspecific mate appears to be innate in the cuckoos.

Behavior of the nestling cuckoo Eviction behavior The nestling in many species of brood-parasitic cuckoos actively removes the host eggs from the nest. After a female cuckoo removes a host egg from the nest, the infanticidal nestling cuckoo evicts the remaining contents of the nest.The cuckoo evicts the host eggs within a few hours of hatching, and if it fails then it succeeds in a couple days when it is larger and strong enough to lift and push the eggs from the nest.The cuckoo braces each foot against the bottom of the nest and braces its head by pushing the bill into the nest, forming a stable tripod. It balances the egg on its back and raises the hooked wings to each side to hold the egg in place. This behavior is aided by the flat or hollow shape of the nestling cuckoo’s back, and by the enlarged first digit which holds the egg in place on the nestling’s back.Then the nestling moves backwards and thrusts its rump upward and back until the egg falls free of the nest (Payne and Payne 1998a, Figure 10.1).

Figure 10.1. Eviction behavior of nestling Horsfield’s Bronze-cuckoo in a nest of Splendid Fairy-wren, Gooseberry Hill,WA.

146 The Cuckoos Fatal competition with host nestmates is won by the nestling cuckoo when it hatches first and the nesting parents do not interfere with the mayhem in their nest. When the cuckoo nestling removes the hosts’ eggs, the foster parents do not rear their own young.They give all their parental care to the cuckoo and rear it as their own. The cuckoo takes all the food that would have been delivered to the host brood. Eviction by the nestling is known in about half the species of brood-parasitic cuckoos. The behavior has been watched in most detail in the Common Cuckoo (Wyllie 1981, Khayutin et al. 1982, Malchevsky 1987, Yoshino 1999, Davies 2000) and it has been seen in several species of Cuculus, Cacomantis and Chrysococcyx and in Thickbilled Cuckoo Pachycoccyx audeberti. Evicting behavior is sensitive to the temperature in the nest. Unlike other altricial birds, the young cuckoos become more active at cool temperatures. Nestling cuckoos can evict at cool body temperatures, conditions when the host nestlings are motionless in the nest. In northern Russia, nestling Common Cuckoos are active and evict the host eggs when the nestlings are cool. “Temperature changes greatly affect the activity of the Cuckoo, which is most active and vigorous at the relatively low temperature of 15°[C], when the young of the host may be rigid with cold” (Dement’ev and Gladkov 1966). Promptov and Loukina (1940) found that the nestling cuckoo evicts better at lower temperatures, when its skin is cold.They suggested that this behavior is adaptive: the cuckoo evicts the host young when the host female is absent, the nest is cold and host chicks are inactive. Nestling cuckoos, like passerines, are inactive below 15°C, and are otherwise most active in begging when TB is 35–37°C, their temperature when they are brooded (Malchevsky 1987). Nestling Common Cuckoos begin to develop a high internal temperature about the age when they no longer evict, with temperature regulation beginning 4–5 days after hatching (Hund and Prinzinger 1980). In temperate Western Australia Chrysococcyx nestling cuckoos evict at low temperatures. Nestling Horsfield’s Bronze-cuckoo and Shining Bronze-cuckoo are active and evict at cool envi-

ronmental TA and body TB temperatures over a range of TB from 34°C to 20°C. Nestlings evict persistently when their TB drops 2–3°C and the nestlings are touched on the back, and they maintain this eviction push-up posture while the TB is above 20°C. When they are no longer touched on the back, they move around the nest with increased “searching” activity as they cool, and they resume the eviction posture when touched again on the back. Small nestlings (1.9 to 3.0 g) cool rapidly, TB dropping from 5.7° to 11.7°C below the initial TB of 30–32°C in a few minutes. Nestling cuckoos stop the eviction behavior when the TB drops to 20°C or cooler. They do not shiver, and they cool as rapidly as a host nestling. Nestling fairy-wrens of the same age (0–5 days) are cool at TA between 22° and 30°C and are motionless until they are warmed to 30°C (Figure 10.2).The nestling cuckoo remains inactive while it is brooded and kept warm, and when its foster parent leaves the nest for its first feed early morning, the cuckoo becomes active and pushes itself around the nest until it comes in contact with the eggs in the nest (Payne and Payne 1998a).This remarkable active behavior of the altricial parasitic cuckoo nestlings at cool temperatures that cannot maintain their own body temperature has no known counterpart in other altricial birds. New World brood-parasitic cuckoos also eliminate the competition for parental care. Nestling

Figure 10.2. Movements of young Horsfield’s Bronzecuckoo Chrysococcyx basalis in a nest of Splendid Fairywren Malurus splendens, Gooseberry Hill, WA. The nesting fairy-wren is immobile.

Brood parasitism 147 nestling cuckoo does not evict. In Spain, young Great Spotted Cuckoos in the nests of magpies grow at the expense of the host young, whereas young cuckoos in the nests of the larger crows often grow up together with their foster nestmates which do not have a lower success in these parasitized nests.

Begging behavior and begging calls

Figure 10.3. Hooks on bill of a five-day-old nestling American Striped Cuckoo Tapera naevia (after Morton and Farabaugh 1979).

American Striped Cuckoos Tapera naevia are siblicidal.They have a sharp hook on the bill tip and use the hook to kill their nestmates (Figure 10.3, Morton and Farabaugh 1979) much like a broodparasitic nestling honeyguide uses its bill hook to kill the nestlings of its foster parents (Friedmann 1955, Payne 1998).The other two brood-parasitic cuckoos in the New World, Pheasant Cuckoo and Pavonine Cuckoo, are less well known, though insofar as they are reared in an ovenbird nest that is more enclosed than the nest of the Australian fairy-wrens (a closed nest from which a young bronze-cuckoo can evict), the eviction of the egg or nestmates by these nestling neomorphine cuckoos would be difficult. Nestlings of other brood-parasitic cuckoos grow up together with their host young. In part, their tolerance of nestmates may be forced by the larger size of the host.The host-tolerant cuckoos include crested cuckoos Clamator spp. ( Jacobin Cuckoo, Levaillant’s Cuckoo, Great Spotted Cuckoo, Chestnutwinged Cuckoo), Common Koel (in India, where they parasitize crows), Channel-billed Cuckoo and Long-tailed Cuckoo. These tolerant cuckoos are sometimes reared alone, however, as they grow rapidly and out-compete the host young for food. In Australia where there is only one cuckoo egg in a nest and the host is smaller than the cuckoo, young Common Koels sometimes evict the host eggs. In India the young are often reared with another young Koel in the nest, the host is larger and the

The young parasitic cuckoo crouches in the nest, tosses the head back and opens the mouth. It calls loudly when the foster parents approach the nest, and calls more loudly and rapidly when a parent places food into its open mouth (Payne and Payne 1998a). It continues to beg with open gape after it is fed. The brood-parasitic nestlings differ from the nesting cuckoos in the mouth, which is unmarked in the parasites but often has species- or genus-specific colored patches in the nesting species (Plate 20). Nestling brood-parasitic cuckoos deposit their excreta in a gelatinous wrap as in passerine birds and the package of feces is carried away by the foster parents. After they fledge, the young cuckoo actively associates with its foster parents, flying towards them, calling, and begging with fluffed plumage, slightly drooped wings and open mouth (Figure 10.4). The nestling behavior is similar to the behavior of young songbirds and is remarkably different from that of the young of nesting cuckoos. In Yellow-billed and Black-billed Cuckoos, the nestling stands upright in the nest, stretches its neck upward and flaps its wings as early as its first day after hatching, and the young excrete a foul-smelling liquid (Hughes 1999a, UMMZ photos, Figure. 5.7). Young brood-parasitic cuckoos are loud and persistent in begging for parental care, both in species that are reared together with the host young and in species where the young cuckoo kills its nestmates and is the only nestling in the brood.The calls are so loud and given so often that the begging cuckoo makes as much noise as an entire brood of its host species (Davies et al. 1998, Davies 2000). Young cuckoos call when the foster parents are out of sight and also when they are nearby. Unattended fledged cuckoos use short calls to maintain social contact with the foraging foster parent. When the foster

148 The Cuckoos

Figure 10.4. Begging behavior of young Common Cuckoo Cuculus canorus, fed by foster parent Vinousthroated Parrotbill Paradoxornis webbianus: a, in nest; b, after fledging (after Yoon 2000).

parent approaches, the young cuckoo increases the rate and duration of its calls (Payne and Payne 1998a).The young of other brood-parasitic birds, the cowbirds Icteridae and brood-parasitic finches Viduidae, are also conspicuous in loud and persistent begging calls that not only attract their foster parents but also put them at risk of attracting a predator. Begging calls of most cuckoos differ from begging calls of their hosts, and the young cuckoos are readily identified in the field as cuckoos by their calls. Species-specific differences in mimicry of begging calls are unknown in the parasitic cowbirds (Gochfeld 1979, Woodward 1983, Broughton et al. 1987). Although some young brood-parasitic Vidua finches match the begging calls of their host, others do not (Payne and Payne 2002), and matching their

hosts’ begging calls is not common in nestling brood-parasitic birds. In a few cuckoos, the young give different begging calls when reared by different host species. A handreared young American Striped Cuckoo had a call like the begging call of the young of its host species, and it differed from the call of a cuckoo in the nest of another host species (Haverschmidt 1961, Morton and Farabaugh 1979). Fledged Levaillant’s Cuckoos are said to give calls like the begging calls of their babbler hosts ( Jubb 1952, Mundy 1973, Steyn 1973, Vernon 1982, Hustler 1997b, Barry 1998). Great Spotted Cuckoo young in nests of crows Corvus spp. and in nests of Magpies in Spain have different begging calls (Arias de Reyna and Hidalgo 1982, Redondo and Arias de Reyna 1988, Redondo 1993), and so do cuckoo young in nests of crows and nests of starlings in Africa (Mundy and Cook 1974). Begging calls of young Channel-billed Cuckoos in Australia are said to be similar to begging calls of their hosts. It is unknown whether female cuckoos are species-specific in host selection and have call races, or the nestling cuckoos learn their calls. Begging calls of Shining Bronze-cuckoos in New Zealand reared by Grey Gerygone Gerygone igata are identical to those of cuckoos in Western Australia reared by Yellow-rumped Thornbill and Western Thornbill (Payne and Payne 1998a). The host species differ from each other in their begging calls, so there is no evidence in this cuckoo for mimicry of begging calls as suggested elsewhere (McLean and Waas 1987). In another glossy cuckoo, Diederik Cuckoo in Africa, the young in nests of hosts Southern Masked Weaver Ploceus velatus, Red Bishop Euplectes orix and Cape Sparrow Passer melanurus are said to differ in begging calls (Reed 1968), and begging-call races may coincide with mimetic egg races in this cuckoo; recordings are needed to compare these begging calls. In Horsfield’s Bronze-cuckoo, the begging calls of the young differ when the nestlings are reared by different host species; and further, the begging calls of each are similar to the begging calls of the host young. On a hillside in Western Australia the begging calls of young cuckoos reared by Splendid Fairy-wrens at first are a “peep”, then change to “zeba” and “sit” calls and a “twe” note by late nestling

Brood parasitism 149 life; after fledging the young cuckoos combine “twe” calls into a “reel” like a young fairy-wren. Begging calls of cuckoos reared by Western Thornbills are a “whine” like the calls of a young thornbill.The begging calls of young reared by an occasional host, the irruptive Scarlet Robins Petroica multicolor which appear after grass fires, are like the calls of young cuckoos reared by thornbills and are not like the begging calls of young robins (Payne and Payne 1998a). Nestling cuckoos in the nests of fairy-wrens and thornbills had distinctive calls by the time they were 10 days old, and by 14 days the calls were the same as in the fledgling cuckoos.The rising buzz of young thornbills was not matched exactly by the cuckoos, nor was the descending buzzy note of robins, and some calls (“ze-ba,” “sit”) of cuckoos in the care of fairy-wrens did not match the calls of the host’s own young. Nevertheless, the begging calls of cuckoo and host species were similar, especially in cuckoos reared by fairy-wrens. Where the host-specific begging calls have been recorded in southeastern Australia (Buckingham and Jackson 1990), young Horsfield’s Bronze-cuckoos reared by Superb Fairy-wren Malurus cyaneus sound like young fairy-wrens and like the cuckoos reared by Splendid Fairy-wrens in WA, and young cuckoos reared by Brown Thornbill Acanthiza pusilla sound like young Western Thornbills and like the

cuckoos reared by thornbills in WA (Figure 10.5). Nestling Horsfield’s Bronze-cuckoos in nests of Superb Fairy-wren have a begging call like the missing young of the host as early as day 2 after hatching. The begging calls are thought to attract the parental care of the foster parents. In experiments, lone chicks of Superb Fairy-wren and of Horsfield’s Bronzecuckoo are less likely to be deserted than lone chicks of Shining Bronze-cuckoo, which have a different begging call (Langmore et al. 2003). Because nestling bronze-cuckoos evict the host eggs from the nest by day 2 after hatching, the young cuckoos have no opportunity to learn the begging calls of their host species from the foster parents’ nestlings. A possible explanation of the begging calls is that Horsfield’s Bronze-cuckoos have begging-call races, where the young of females that lay in the nests of fairy-wrens differ genetically from the young of females that parasitize thornbills. The irruptive host robins rear cuckoo young with begging calls like the calls of thornbill-reared cuckoos.The observation is consistent with the idea of a begging-call race of cuckoos that regularly use resident fairywrens and that opportunistically use other host species (Payne and Payne 1998a). It is not consistent with the idea that young cuckoos learn and copy the calls of their foster species, if only because the robinreared cuckoos did not call like the young robins,

Figure 10.5. Begging calls of young Horsfield’s Bronze-cuckoos Chrysococcyx basalis: a–c,Western Australia (a, cuckoo reared by Splendid Fairy-wren Malurus splendens; b, cuckoo reared by Western Thornbill Acanthiza inornata; c, cuckoo reared by Scarlet Robin Petroica multicolor); d–e, eastern Australia (d, cuckoo reared by Superb Fairy-wren M. cyaneus, NSW; e, cuckoo reared by Brown Thornbill A. pusilla,Victoria (d, e from Buckingham and Jackson 1991)); f–h, young hosts in Western Australia (f, Splendid Fairy-wren; g,Western Thornbill; h, Scarlet Robin).

150 The Cuckoos and because nestling cuckoos never hear the host young, as the cuckoo chick evicts the host eggs before they hatch. Bronze-cuckoos in some parts of Australia have mitochondrial genetic differences between populations, and so do cuckoos within a local population (Joseph et al. 2002). It will be of interest to test whether genetic differences are found between nestling cuckoos that are reared by different host species, and to transfer cuckoo eggs between the nests of different hosts and record the begging behavior of young cuckoos reared by these hosts. The matched begging calls of cuckoo chicks to calls of the host may occur only in certain glossyand bronze- cuckoos Chrysococcyx, insofar as Common Cuckoos in Europe do not differ in their begging calls when their host species differ (Butchart et al. 2003).

Ecology of host-parasite associations Unlike the young of nesting birds, the young brood parasites are not related to the adults that feed them. A young cuckoo can gain extra care by eliminating the host nestlings. It is not related to the foster parents, its demand for unlimited parental care does not put a genetic sibling at risk, and it loses no kin by overworking the foster parents; but if a foster parent dies, the young cuckoo starves (Gochfeld 1979, McLean and Griffin 1991, Briskie et al. 1994, Dearborn 1998, 1999, Payne et al. 1998, 2001, Payne and Payne 1998b, Kilner and Davies 1999, Davies 2000).

Effects of a brood parasite on its host Brood parasites reduce the nesting success of their hosts (Rothstein 1975b, Payne 1977b, 1997a). In the relatively benign brood-parasitic finches and cowbirds, the removal of a host egg by the female parasite accounts for most of the reduction in nesting success of the host (Morel 1973, Payne 1997a, Payne and Payne 1998b). In addition, brood-parasitic nestlings compete with host nestlings for parental care.The tactics of young parasites to control parental care for their own benefit are a counterpart of the sibling competition of certain nonparasitic nesting

birds.Within a brood of nesting birds, asynchronous hatching and not-so-subtle siblicide may be advantageous to the rival that wins the struggle with its siblings (Nathan et al. 2001).The young brood parasites carry the theme of sibling rivalry to its extreme. Most parasitic cuckoos have a more rapid incubation than their hosts and are larger at hatching. The head-start in development gives the young parasite an advantage in winning parental care (Payne 1977b). Second, some cuckoos are larger, grow more rapidly and take more food than their hosts, as in the Great Spotted Cuckoos. Third, brood-parasitic cuckoos are the only nestlings that regularly and actively push and remove the host eggs and nestlings out of the nest. Brood-parasitic New World cuckoos are equally deadly to their nestmates, killing the host young with their hooked bills. Although the female cuckoo causes some loss of host success when she removes an egg from the host clutch, the later eviction and biting behavior by the nestling cuckoo causes the total loss of the host brood. The frequency of brood parasitism affects the breeding success of the host population and the evolution of host defense against brood parasitism (Davies et al. 1996). The proportion of nests parasitized by Common Cuckoos is less than it was in earlier decades in Britain (Brooke and Davies 1987), yet in some areas of Europe the intensity of cuckoo parasitism remains high. There is often a great deal of regional and local variation in parasitism. Songbirds parasitized in one area are not parasitized in another area, even though cuckoos live in both areas (Makatsch 1937, 1955). In southern England where the regional rate of parasitism of Reed Warblers is about 5%, the local rate of cuckoo parasitism of warbler nests ranges from 0 to 16% (Lindholm 1999). In Central Europe where the rate of cuckoo parasitism per nest averages 8.3% for Reed Warblers and 6.3% for Marsh Warblers, from 3% to 28% of the warbler nests in local populations are parasitized. In Central Europe the most frequently parasitized host is the Great Reed Warbler, and in the marshes of Hungary nearly 50% of warbler nests were parasitized in the early 1940s (Molnár 1944).The rate of parasitism has remained high for decades, and surveys in the late 1990s showed a higher rate of parasitism (57%) with half

Brood parasitism 151 the parasitized nests having 2 or more cuckoo eggs (Moskát and Honza 2002). In Africa and Australia a high proportion of host nests are parasitized at least locally by several cuckoo species, though the proportion is much less than 50% (Payne and Payne 1967, Brooker and Brooker 1989a, Payne 1997a). The effect of parasitism on a host population can be calculated from field data. The proportion of nests that are parasitized, multiplied by the difference in success in parasitized and unparasitized nests (the number of host young that fledge, or the proportion of nests that fledge at least one host young), allows for not all unparasitized nests succeeding and not all losses of parasitized nests being due to parasitism (Payne 1977b, 1997a). Egg predation by cuckoos also reduces breeding success of the hosts, and in some populations more nests have eggs taken by a cuckoo than nests parasitized by the cuckoo (Schulze-Hagen 1992). Costs of Common Cuckoo parasitism to Reed Warblers were calculated from the breeding success of host pairs with their own young (Øien et al. 1998). In a population in the Czech Republic, 16.1% of 1108 warbler nests were parasitized; in this population the cuckoo eggs are not good mimics of Reed Warbler eggs. In parasitized nests the nesting pairs that removed the cuckoo egg had a success of 0.29 for each warbler egg, and pairs that accepted the egg had a success of only 0.03, yet the warblers removed the cuckoo egg from only 8% of the nests. Some birds deserted and nested again; 30% of parasitized nests and 1.8% of unparasitized nests were deserted.These observations suggest that warblers which accept a cuckoo egg in the nest do not behave in their own best interests. Success in the number of warbler young fledged from unparasitized nests was 1.95, and the number fledged from parasitized nests was only 0.11. When the effect of cuckoo parasitism on production of young in the host population is calculated as % parasitized nests x% difference in success of parasitized and unparasitized nests (the mean number of eggs in unparasitized nests ⫽ 4.22), the success of unparasitized nests was 1.95/4.22 ⫽ 0.462, and the success of parasitized nests was 0.11/4.22 ⫽ 0.026), the difference is 0.436, or 44%.That is, cuckoo parasitism was responsible for nearly half the loss in breeding success in the population. In nests where cuckoos

may have removed a host egg, the number of host eggs in unparasitized nests (3.88) and number in the parasitized nests (2.12) allows an assessment of costs that are accounted to the female cuckoo and to the nestling cuckoo. In this case, 1.76 eggs were lost to the cuckoo female, and the difference (1.95 young fledged in unparasitized versus 0.11 young in parasitized nests) was 1.84 young to the cuckoo nestling. That is, nearly half the cost of being parasitized was the result of egg removal by the cuckoo female, and more than half the cost was due to eviction of host eggs by the cuckoo nestling. When the cuckoo female removes a host egg or two from the nest, the host cannot recover the loss by removing the cuckoo egg; nevertheless a host could avoid the cost of nestling eviction by a cuckoo chick if the host removes the cuckoo egg. If the host were uncertain which egg is the cuckoo’s, the host could gain only 0.03%, the proportion of nests that would otherwise fail. For comparison, the cost of removing a cuckoo egg is more than 2%, the proportion of times it damages or removes its own egg when it attempts to remove a cuckoo egg (Davies et al. 1996).A host should remove a cuckoo egg when the cuckoo egg differs from its own egg, but not when it is uncertain which egg is the alien egg, as when the cuckoo egg is a good mimic. In fact, female warblers that lay variable eggs themselves are less likely to remove a cuckoo egg than females whose own eggs look alike. Perhaps the variability in their own eggs leads to uncertainty and inaction against the parasite egg (Stokke et al. 1999). The most complete records of the effects of cuckoo parasitism in the southern hemisphere are for hosts of the glossy cuckoos Chrysococcyx spp. (1) In New Zealand, Grey Gerygones are parasitized by Shining Bronze-cuckoos. Nests not parasitized had a 33% success in fledging at least one young, and nests with a cuckoo egg had only 1.9% success in fledging a young of the host. In this population 55% of 40 nests were parasitized (Gill 1983a).The impact of cuckoos on decreasing the production of host young was (0.55 ⫻ (32.9 ⫺ 1.9)/32.9) ⫽ 17% (Payne 1997b). (2) In Western Australia, Splendid Fairywrens are parasitized by Horsfield’s Bronze-cuckoos (Rowley et al. 1991).The incidence of brood parasitism was 20%. In some parasitized nests the cuckoo

152 The Cuckoos egg was laid before the host had laid and the fairywrens then deserted, and in others the cuckoo egg failed to hatch or survive, but parasitized nests failed completely when the cuckoo evicted the host eggs. Nests that were not parasitized had a mean success of 1.70 fledged fairy-wrens, and nests that were parasitized had 0.11 fledged fairy-wrens (Brooker and Brooker 1996). Because mean clutch size in unparasitized nests is 2.91, when the female cuckoo removes an egg and mean clutch size in parasitized nests is 1.89 host eggs, the cost of parasitism is due mainly to evicting behavior by the nestling cuckoo. The overall impact of cuckoos was to decrease the success of fairy-wrens by 19%. (3) In South Africa, Southern Masked Weavers are parasitized by Diederik Cuckoo. Nests not parasitized fledged on average 1.48 young weavers, and nests with a cuckoo egg fledged only 0.16 young weavers. In a few parasitized nests the cuckoo egg failed to hatch and the young weavers survived; in both kinds of nests some broods failed owing to storms and predators. In this population 43% of 76 nests were parasitized (Hunter 1961). For the host of Diederik Cuckoo, the impact of the cuckoo was to decrease the production of host young by 38%. A female cuckoo can make an impact on the breeding success of a local host population. In Britain, Chance (1922) found that a female Common Cuckoo that he knew on sight laid as many as 25 eggs in a season in nests of its host Meadow Pipit. Similar large numbers of eggs were laid by a female cuckoo in host nests in northeastern France (Blaise 1965). In southern Africa, female cuckoos lay 12–24 eggs in a season. Seasonal egg laying was determined by the proportion of females with an egg in the oviduct, the number of recently ovulated follicles in the ovaries, and the length of the local breeding season.A female Diederik Cuckoo lays 16–21 eggs in a season and a female Jacobin Cuckoo lays 19–25 eggs. Other species that have been examined (Great Spotted Cuckoo, Levaillant’s Cuckoo, Black Cuckoo, Redchested Cuckoo, Klaas’s Cuckoo, African Emerald Cuckoo) have similar rates of laying. Small cuckoos lay on successive days, while large cuckoos and even the small glossy cuckoos sometimes lay on one day, carry an egg in the oviduct for a day then lay on the following day, and do this in a series (or “clutch”) of 3–5 eggs in a set with a few days between the series.

Between sets the ovary has only small ovarian follicles, then it yolks up another series, staggered in size with the largest developing follicle about twice the size of the next one and so on. A female cuckoo can lay no more than one egg in a day. Like other birds the cuckoos have only one ovary and one oviduct, and their reproductive anatomy enforces the “egg-aday” limit in birds, even in chickens that are bred to lay many eggs. Ovaries of brood-parasitic cuckoos were compared with those of breeding Yellow-billed Cuckoos taken at nests where they incubated or had small nestlings, to calibrate how the ovarian follicles of cuckoos change from the time the egg is ovulated and laid, to the time the bird was on the nest (Payne 1973a, 1974).

Behavioral response by the nesting host to the cuckoo Nesting hosts respond to an adult cuckoo near their nest by mobbing, flying at it, striking it and calling loudly (Smith and Hosking 1955, Payne et al. 1985, Macdonald 1990, Duckworth 1991, 1997, Yoshino 1999, Hanmer 2001). Mobbing may drive a cuckoo from the nest, and a host may avoid parasitism of its nest if the cuckoo leaves before she lays. But if a host mobs when a cuckoo is not near the nest, the conspicuous behavior may help the cuckoo to locate the nest, so a host often limits its mobbing behavior to the nest site after the cuckoo has already located the nest. Active defense near the nest can drive away the cuckoo, and strong defense can kill the cuckoo. A Smooth-billed Ani (an occasional within-species brood parasite) was killed by other anis as she attempted to lay in an ani nest (Loflin 1982). A Common Cuckoo with an egg in the oviduct ready to lay was killed by a Great Reed Warbler near a warbler nest; the female was found freshly dead, floating in the water by the nest just after the warblers were excited around the nest ( Janisch 1954).A Common Cuckoo near a nest of Bull-headed Shrike Lanius bucephalus was photographed aftger being killed and plucked by the shrike (Yoshino 1999). And a Diederik Cuckoo was killed when a male Southern Masked Weaver defended its nest site, harried the cuckoo to the ground, pecked it on the head and continued to attack it on the ground;“Mr Hoffman then held the

Brood parasitism 153 cuckoo up whereupon the weaver attacked the dead cuckoo in his hand” (Lorber 1985). Most small nesting songbirds in Europe reject the eggs of other birds. Rejection of a cuckoo egg is the main line of defense against brood parasitism. In field experiments, when an egg unlike their own (a “cuckoo” egg) was placed into the nest, eight species always or almost always (⬎ 80%) rejected the unlike egg, 16 species often (20%–80%) rejected the egg, and ten species seldom or never (0%– ⬍ 20%) rejected it. In the ten accepters, five were hole-nesting songbirds such as tits and wrens that are not parasitized in natural conditions (Øien et al. 1995).Variation in acceptance and rejection of an egg is also affected by the time when the egg is laid in the nest. The nesting females more often reject when a cuckoo egg appears while the hosts are laying, than when an egg appears after they have begun to incubate (Moksnes et al. 1994, Palomino et al. 1998). The egg colors and patterns of cuckoos have evolved together with the responses of the intolerant host species. Cuckoos of the mimetic egg morphs later colonized other host species that are not as discriminating and accept a cuckoo egg that does not closely match their own. In Europe, several songbirds are parasitized by Common Cuckoos with blue eggs, but of these hosts, Redstart Phoenicurus phoenicurus (blue eggs) and Whinchat Saxicola rubetra (pale blue eggs, sometimes speckled rusty) are the most discriminating, and the nondiscriminating hosts such as Wheatear Oenanthe oenanthe (white eggs spotted brown) and Pied Flycatcher Ficedula hypoleuca (pale blue eggs, rarely with fine speckling) may have been parasitized later by the cuckoo lineages (Moksnes et al. 1995, Gibbs et al. 2000). The risk of removal or damage to their own egg may explain why some hosts accept a cuckoo egg in their nest. When a cuckoo egg is laid late or when it is sterile, there is no benefit in removing it, and the same hosts that remove a cuckoo egg when it is laid along with their own clutch will accept a cuckoo egg when it is laid well after incubation has begun. But when the hosts removes an egg, they risk removing or (perhaps more importantly) damaging their own egg (Marchetti 1992, Soler et al. 1996, Røskaft et al. 2002).

In contrast to the small hosts of Common Cuckoos in Europe which reject a cuckoo egg, small songbirds in North America generally accept the egg of the brood-parasitic Brown-headed Cowbird (Rothstein 1975a, 1990).The behavior of the nestling parasites in a parasitized nest accounts for this difference in the behavior of the nesting hosts. A young cuckoo kills the host eggs or nestlings; a young cowbird often grows up together with the host nestlings. In cuckoo hosts the main cost of being parasitized is paid after the cuckoo hatches, whereas in cowbird hosts the cost is paid when the female cowbird removes a host egg (Moksnes et al. 1991, 1994, Payne 1998, Øien et al. 1998, Payne and Payne 1998b).The greater impact of a nestling cuckoo on the nesting success of the host explains why cuckoo hosts go to great lengths to remove a cuckoo egg, even puncturing the eggshell and risking their own eggs. A young cuckoo looks different from the hosts’ own young, and it is amazing that the foster-parents rear these strange young in their nest. Observations of songbird hosts and of other brood-parasitic birds give only partial answers to the question of why the foster parents rear young that are not their own. Brood-parasitic finches Vidua have nestlings that mimic the nestlings of their foster parents’ own chicks with which they grow up together in the nest.The visual resemblance of nestling Vidua to chicks of the foster species may allow the parasite to be accepted and provided with parental care (Payne 1997a, 1998). Except for Thick-billed Cuckoos which have mouth spots and whose host bush-shrikes Prionops have mouth spots (the details are undescribed), no such visual mimicry is known in nestling cuckoos.Within a species, parents do not discriminate in caring for young that are genetically their own and young that were fathered by another bird or were laid in their nest by a conspecific female (Westneat and Sherman 1993, Westneat and Sargent 1996, Whittingham and Dunn 1998, Payne et al. 2001). Magpies are more likely to accept an experimentally added chick of the Great Spotted Cuckoo if their nest already has a nestling cuckoo, than if it is not parasitized (Soler et al. 1995a). However, the hosts of some brood parasites rear the young of other species even in a mixed-species brood where the nestlings differ in appearance and in begging calls (Payne et al. 2001).

11 The evolution of brood parasitism in cuckoos

Equilibrium and coevolution Although a cuckoo nestling destroys the host eggs and nestlings, the host parents often do not discriminate against a cuckoo egg and they incubate it with their own eggs.The hosts may lack the genetic variation that would allow evolutionary selection to favor action against cuckoos (Rothstein 1990). Or, the hosts and their parasitic cuckoos may be in a state of equilibrium, where the gain of rejection is balanced by the cost of rejection of the cuckoo egg (Davies and Brooke 1988, Davies et al. 1996). The costs and benefits of brood parasitism and response by the hosts to the parasite can be described as a mathematical model, which assumes genetic variation in both the parasite and host and tracks the coevolutionary outcome of these costs and benefits. These two hypotheses, the lack of genetic variation in hosts and the equilibrium in costs and benefits in response to cuckoo parasitism, differ in their ability to predict the variation in behavior among host birds (Payne 1997a). Genetic differences that cause behavioral differences or are heritable along family lineages have not been demonstrated in the field or in the genetics’ laboratory. Models that explore the effects of population size and the intensity of brood parasitism assume genetic differences between hosts that defend and hosts that accept the cuckoo egg, with “rejecter genes” in the host and “egg color genes” in the cuckoo, then track the success of the assumed gene under different ecological conditions (Kelly 1987, Takasu et al. 1993, Takasu 1998). Although there may have been genetic differences in the past, the status of this

model is one of indirect reasoning. In contrast, the ecological factors that determine the alternative behavior have been identified, and the costs and benefits of alternative behaviors to birds with a cuckoo egg in their nest have been measured in the field. In consequence we can describe the benefits to individual hosts that use these behaviors, in terms of the number of their young that survive the period of parental care.When the frequency of parasitism and the decrease in fitness of parasitized nests are high, we expect selection to favor hosts that remove the cuckoo egg. This ecological and genetic interaction is combined in the coevolution model. A balance or dynamic equilibrium in the ecological and genetic interaction between host and brood parasite accounts for the adaptations of host and cuckoo.The evolution of response of a host species to cuckoo parasitism depends on the hosts’ frequency of being parasitized, the costs of being parasitized and the costs of defense (Lotem et al. 1995, Davies et al. 1996). Individual hosts vary in their response to a cuckoo egg, and a host that removes an egg from its nest at one time may accept it another time. Whether a host rejects or accepts the cuckoo egg depends on variable conditions such as the host’s age and experience, the stage of nest, and the habitat, as will be seen. (1) The cost of removal may be greater than the benefit gained by removing the egg. Costs occur if the nesting bird fails to discriminate between its own egg and a cuckoo egg, and removes its own egg in error; or if it damages its own egg when it attempts to remove a cuckoo egg (Røskaft et al. 1993, Lotem et al. 1995, Davies et al. 1996). (2) The numbers of cuckoos in a nesting area

The evolution of brood parasitism in cuckoos 155 may affect whether a host removes the egg (Davies et al. 1996). Reed Warblers have a higher rate of rejection of model cuckoo eggs in parasitized than in unparasitized populations (Lindholm and Thomas 2000). Over a 12-year period from the 1980s to the late 1990s in one population, the rate of cuckoo parasitism declined from 20% to about 5%, along with a decline in the number of cuckoos. In experiments with model eggs in warbler nests, there was a decline in host rejection of nonmimetic eggs, from 75% in 1985–1986 to 25% in 1997, and a decline within a season, with the host less likely to reject when cuckoos disappear (Brooke et al. 1998, Lindholm 2000). The decline in the rate of rejection was too rapid to be owing to genetic change, and it may have been in response to the changed environmental conditions such as the number of cuckoos in the breeding area. (3) In field experiments, nesting birds were likely to reject the egg when an adult Common Cuckoo dummy was near the nest, but to accept it when there was no cuckoo dummy (Davies and Brooke 1988, Moksnes and Røskaft 1989, Moksnes et al. 1990, 1993a, Lerkelund et al. 1993, Alvarez 1996, Davies et al. 1996). In tests with another host and parasite, Magpies were no more likely to reject a model egg when a hand-reared Great Spotted Cuckoo perched near the nest than when the nest had no cuckoo nearby (Soler et al. 2000).This host is more likely to accept a cuckoo egg than are hosts of the Common Cuckoo, because nestling Great Spotted Cuckoos do not evict the host eggs. (4) The stage of breeding affects rejection, as birds are more likely to reject a cuckoo egg during laying than during incubation, when a late cuckoo egg would not hatch before the host eggs hatch (Moksnes et al. 1990, 1993a). (5) The number of eggs in the nest may affect whether a host deserts the nest as happens when the cuckoo takes more than one egg, or when a predator reduces the clutch size (Øien et al. 1998). (6) Experience with her own eggs may affect whether a nesting host rejects an egg in a later nest. In experiments that introduced dummy eggs into the nests of Great Reed Warbler, the yearling females were more likely to accept an egg than were the older females (Lotem et al. 1992). (7) Experience in rearing a cuckoo may determine the

host’s behavior in a later nesting. Magpies that have reared a young Great Spotted Cuckoo are more likely to accept another cuckoo if it is experimentally placed into the nest, compared to adults that have not reared a cuckoo. When Magpie and cuckoo nestlings are reared together, the older they are the more the Magpie parents discriminate between them. When a fledged cuckoo is placed near their nest, they feed their own young in preference to the young cuckoo, and they tolerate alien Magpie nestlings more often than cuckoo nestlings (Soler et al. 1995a). Although the host young often survive and leave the nest, the young cuckoo takes its share of parental care and food and affects the growth and development of the young Magpies (Soler et al. 1996). Cuckoo fledglings are loud and persist in begging, they attract other birds and these newcomers may feed the cuckoos: as many as four species have fed a fledgling cuckoo (Kikkawa and Dwyer 1962, McBride 1984,Yoshino 1999. In summary, the response of hosts to a cuckoo egg varies with the risk of being parasitized, the cost of accepting the cuckoo egg, and the gain to be realized by rejecting the cuckoo egg. A conditional response explains why some hosts accept the cuckoo egg at one time, but reject it at another time (Lotem et al. 1992, Lotem and Nakamura 1998, Soler et al. 2000). A conditional response can be tested with predictions in experiments and in observations of long-term changes in the behavior of a bird (Davies et al. 1996). Coevolution is the genetic evolution of adaptations between interacting species. This idea describes this interaction over evolutionary time between brood parasites and their hosts, especially where the interaction is reciprocal with each responding to the adaptations of the other. Brood parasites that lay eggs with the color and pattern of the host are less likely to have their eggs rejected, and the evolution of similar eggs is a response to selection by their hosts (Baker 1923, 1942; Payne 1977a; Brooke and Davies 1988; Davies and Brooke 1989a,b; Rothstein 1990). The hosts use both color and spot differences between a cuckoo egg (or a model egg) and their own egg; and the more similar the alien egg is to their own, the more likely it is to be accepted by the host, as in

156 The Cuckoos Village Weavers in West Africa (Lahti and Lahti 2002). Parasitized species are more variable in egg color and pattern than are unparasitized species, and the different levels of variation suggest that egg color and pattern evolve in response to cuckoos (Øien et al. 1995, Soler and Møller 1996). The critical test is to determine whether the egg color and pattern change through time and in what direction they change. The behavior of nesting birds may have evolved towards the mismatched eggs in their nest in response to cuckoos, but their behavior also has other explanations. Some birds may reject in response to an individual bird’s experience and history of intraspecific parasitism (Brown and Brown 1989, Westneat and Sherman 1993).The variability in egg color between females in some colonial weavers may have evolved to allow them to distinguish their own eggs from those of other female weavers, rather than from eggs of cuckoos, which parasitize the weavers infrequently (Freeman 1988, Jackson 1998). More often, many birds remove an egg in their nest if it is damaged; and the removal of an egg that looks different from others in the clutch may increase the success of the remaining eggs (Rothstein 1990). In the case of antagonistic coevolution, when the mimic gains an advantage in resembling another species, its model, the model may be at a disadvantage.The model species may be selected to escape this mimicry, in a process of chase-away selection (Fisher 1930, Gavrilets and Hastings 1998). Fisher (1930) noted that a model might change its signals to escape its mimic. On the other hand, if a genetic variant in the model species produces a different egg color and pattern from the egg-mimetic cuckoo, then the model might be challenged to recognize its own eggs, because in changing its characters it might not identify its own eggs; and the disadvantage might be greater than the advantage gained in being different from the mimic (Nur 1980).Turner ( 1984: 154) argued that “The advantage of being a mimic…is considerably greater than the disadvantage of being a model”, and he questioned whether a model would diverge from its mimic (Turner 1995). In at least some cuckoo hosts, a female recognizes her own eggs in the nest, by learning their appearance when she lays

her first eggs, accepting them and rejecting eggs of different appearance that may appear late in the nestr (Lotem et al. 1995). This development of behavior provides a mechanism to escape the challenge of egg mimicry and to coevolve along with the egg-mimetic cuckoo. Both a coevolution model and an equilibrium model predict that nesting birds with a long history of being parasitized are more likely to reject a cuckoo egg than birds in an unparasitized population. Indeed, Meadow Pipits are more likely to accept nonmimetic model eggs in Iceland where cuckoos do not now occur, than in Britain (Davies and Brooke 1989a). Pipits also accept such eggs in Norway, where they were derived from a population that was parasitized but where the probability of being parasitized is low (Moksnes et al. 1993). Both a coevolution model and an equilibrium model also predict that the variation in color and pattern of eggs among clutches of the host species is greater in parasitized birds than in unparasitized birds. The eggs of European songbird hosts are in fact more variable than are the eggs of unparasitized songbirds (Moksnes and Røskaft 1995). At the same time, the variation among eggs within a clutch is lower in the parasitized host species that remove a cuckoo egg (Øien et al. 1995, Soler and Møller 1996). It appears there has been selection within a host species to evolve eggs they can recognize, and to reject the cuckoo egg. The coevolution model and the equilibrium model lead to the same predictions, and they are not mutually exclusive. One concerns an evolutionary time scale.The second can be examined directly and tested experimentally on a behavioral time scale. In practice, the hypothesis of lack of genetic variation is a null hypothesis used by default when other hypotheses fail to show a difference in the behavior of individuals within a population.The null hypothesis has been applied in comparisons between species, rather than within a species at the level where natural selection is effective (Rothstein 1975a, 1990).

Have cuckoos and hosts changed their behavior in historical time? Brood parasitism in cuckoos sets a wide stage to consider interactions between species, and different

The evolution of brood parasitism in cuckoos 157 acts are in progress on this stage. A change in the behavior of hosts and parasites over the years might indicate natural selection of one species (the parasite or the host), and an interaction between parasite and hosts; cuckoos have changed their host species in some regions (Baker 1942, Nakamura 1990). In Burma near Mogok, Maymyo and Taunggyi, Common Cuckoos changed their hostchoice behavior during the early twentieth century (Osmaston 1916, Livesey 1933, 1933, 1936, 1938a, Baker 1940, 1942). Hosts that were once common became scarce as the habitat changed from forest to open country, and the most common cuckoo egg changed at the same time. In the early twentieth century (1900–1915) the common host was the Grey Bushchat Saxicola ferrea, which laid blue eggs. In the 1930s the common host was the Stonechat Saxicola caprata, which laid pink eggs with reddish spots. In the early years cuckoos parasitized the common Grey Bushchat, and cuckoo eggs matched the blue bushchat eggs. In later years, with increased cultivation of rice, gardens and village vegetation that replaced the scrub forest, stonechats replaced bushchats, and the most common cuckoo egg morphs in the region were pink with reddish spots, like the stonechat eggs (Baker 1940, 1942). Egg colors are genetically determined, and a change in the frequency of color morphs of these cuckoo eggs indicates a change in the genetic makeup within the cuckoo population through natural selection. In this case the change in cuckoo eggs occurred when habitat change favored the Stonechat and the egg race of the cuckoo that mimicked the new common Stonechat host. In Britain, nest records of songbirds compiled over 40 years for the six main hosts of the Common Cuckoo include a remarkable sample of 73,750 nests, with 1,016 nests parasitized by cuckoos (Brooke and Davies 1987). During this period there were decreases in the proportion of parasitized nests in the Dunnock, European Robin and Pied Wagtail. For the Reed Warbler there was a small increase in the proportion of nests parasitized. No change appeared in the accuracy of cuckoo egg mimicry of Reed Warbler eggs, and no cuckoo eggs appeared that mimicked Dunnock eggs. In this population, the change in the cuckoo’s use of host species was

not accompanied by a change in cuckoo egg mimicry or in host behavior (Brooke and Davies 1987, 1998; Davies and Brooke 1988, 1989a,b). In three case histories a change in behavior has been reported in the host, the cuckoo, or both host and cuckoo.These cases have been considered natural experiments in evolution (Davies 2000). Individual experiences of the host or the cuckoo can also explain the observed changes. (1) Village Weavers in Africa are parasitized by Diederik Cuckoos, and in field experiments in West Africa these weavers remove from their nest any eggs unlike their own (Lahti and Lahti 2002). In the West Indies, where weavers were introduced into Hispaniola 200 years ago, there are no broodparasitic cuckoos. In a field experiment in Hispaniola the nesting weavers did not discriminate against eggs unlike their own (Cruz and Wiley 1989); in a subsequent test the weavers did remove eggs unlike their own (Robert and Sorci 1999).The lack of rejection in the early test suggested that when there were no cuckoos to parasitize the weavers, rejection was not selected, perhaps because rejection was costly as the birds removed their own eggs. However that study did not carry out a control test in Africa to determine whether the lack of response was due to the experimental method. Hispaniolan weavers all have blue eggs; in contrast, within a local population in Africa, egg color and pattern is highly variable (Collias and Collias 1959, Maclean 1993, Lahti and Lahti 2002).The low range of variation in the model eggs, which were mainly the blue eggs of local weavers in Hispaniola, may explain the results of the early test, while in the subsequent test the model eggs were more variable and the results were similar to the tests in West Africa. (2) Magpies in Mediterranean Europe are parasitized by the Great Spotted Cuckoo. Magpies sometimes reject a model cuckoo egg from their nest, but in northern Europe where the cuckoo does not occur, the Magpies do not reject the model egg. The difference suggests that Magpies have changed their behavior in response to the risk of cuckoo parasitism, either where they are parasitized or even where they are not parasitized. Great Spotted Cuckoos have been studied most intensively by a research group at the University of

158 The Cuckoos Granada. Soler and Møller (1990) found a difference in Magpie behavior in three sites, two in Spain and one in Sweden. Magpies were more discriminating in Spain at Santa Fe, less so 60 km from there at Guadix, and not at all in Sweden where the cuckoos do not occur. The research group suggested that the longer coexistence of host and cuckoo explained the difference between Santa Fe and Guadix (cuckoos were not reported at Guadix before 1962).The results were repeated (Soler et al. 1994b) and the authors interpreted the difference between sites as “micro-evolution”.This interpretation has been questioned. The sites were only a day’s walk apart on a flat road, and their report of differences in the time when cuckoos and Magpies lived together were based on conversations with hunters.There was no direct observation of change in Magpie behavior or cuckoo parasitism over time within a population (Zuñiga and Redondo 1992a). The difference in host behavior between sites could also be explained by the numbers of cuckoos present at each site. In accord with the authors’ interpretation, however, when a tame cuckoo was perched near the nest, there was no increase in rejection rates (Soler et al. 2000). (3) The third case involves a reported change in both host and cuckoo. In Nagano Prefecture in Japan, the Common Cuckoo first appeared in 1960 as a parasite of the Azure-winged Magpie Cyanopica cyana. Cuckoo parasitism increased in the 1960s and 1970s when the altitudinal ranges of cuckoo and magpie expanded and the birds came into contact. Cuckoo parasitism increased from 0 to 30%, then it decreased, and at about this time the magpies rejected cuckoo eggs (Yamagishi and Fujioka 1986, Nakamura 1990). Observations suggested that cuckoos had parasitized magpie nests for a different length of time in three localities: 20 years at (a) Azumino, 15 years at (b) Nagano, and 10 years at (c) Nobeyama. When host magpies were tested with model cuckoo eggs, in (a) and (b) they rejected eggs at about the same rate. Habitat, breeding density and behavior differed among the cuckoo populations, and the comparison does not necessarily reflect a change in parasitism through time, although the field workers interpreted the results in this way (Nakamura et al. 1998).

Cuckoo eggs were compared to see whether cuckoos changed genetically to mimic the host species’ eggs. Cuckoo eggs were sampled in the localities where magpies were tested for their response to model cuckoo eggs. Eight cuckoo egg morphs were defined by lines, dots and blotches, with five egg morphs in magpie nests (Nakamura et al. 1998). There were no statistically significant differences between populations when all egg morphs were considered together, and at Nagano the three host species all had the same most common cuckoo egg morph in their nests.The authors concluded that the different proportions of cuckoo egg morph “A” between (a) Nagano and (c) Nobeyama resulted from the longer history of cuckoo parasitism at Nagano. Magpie eggs have dots and blotches but no lines, unlike the cuckoo eggs in their nests. Because there was no precise mimicry in cuckoo eggs, no shift in the makeup of eggs in the region through time, no overall significant difference in the frequency of cuckoo egg morphs between localities, and no higher proportion of mimetic cuckoo egg morphs in magpie nests in the locality with the longest known history of cuckoo parasitism, there is no compelling evidence to support the conclusion that cuckoos changed by natural selection within two decades to match the eggs of a new host species. Magpies banded at Nagano dispersed within their breeding area and were not seen outside the area (Hosono 1969). Local studies underestimate dispersal (Payne 1990), and it is likely that magpies born outside the study area moved into the population and cuckoos moved as well. Azure-winged Magpies are parasitized elsewhere, in China by the Indian Cuckoo (Shaw 1938, Hoffman 1950) and in Spain by the Great Spotted Cuckoo (Arias de Reyna 1998). The magpie response to cuckoo parasitism may have evolved in other times and places. Azurewinged Magpies are known from fossils dated to 44,000 years ago in the Iberian peninsula (Cooper 2000), and magpies have had a long time to respond to cuckoos.Any changes in behavior of cuckoos and magpies in Japan may not have involved natural selection, insofar as a switch in host acceptance of a cuckoo egg might involve their own experience rather than genetic change within a population.

The evolution of brood parasitism in cuckoos 159 In summary, none of the three cases demonstrate a change in the host, in response to the onset and progress of brood parasitism through time, and none show a change in the eggs or the parasitic behavior of the cuckoos more clearly than the observations in Burma in the early 1900s.

Adaptation and the evolution of brood parasitism The parasitic cuckoos are “obligate” brood parasites. Within a species, all females lay their eggs in the nest of another species of bird and never nest themselves; they are committed to this behavior and are constrained by the lack of any nestbuilding behavior or parental care, to depend on other species to incubate their eggs and to brood and rear their young. The origin and evolution of this brood-parasitic life style has two sets of explanations. One involves the adaptive ecology of parental care and dispersion of the eggs to avoid predation, and a second compares life styles among birds and the phylogenetic relationships of brood-parasitic birds to nesting birds (Hamilton and Orians 1965, Payne 1997a).

Adaptations and behavioral ecology Nesting birds are attentive parents, incubating their eggs, then bringing food to their young until they are independent, and they show the same behavior to the parasitic cuckoos (Lack 1954, Clutton-Brock 1991). In contrast the brood parasites depend on other species to hatch and care for their young. Studies on parental care of breeding birds define the limits of rearing their own young and the consequences of laying in the nests of other birds. First, caring parents may be limited in the amount of food they can find and provide to their brood (Lack 1954, 1968). Second, birds that lay additional eggs in the nests of other birds that provide parental care can leave more offspring in a season than a bird that lays only in its own nest (Darwin 1859). Interspecific brood parasitism in the cuckoos may have originated from within-species parasitism that produces two advantages: overcoming the limits of

parental care, and avoiding predation and minimizing the risk of failure.

Risk of predation Predation at the nest is a major limit to success in breeding birds. Birds that lay all their eggs in one nest are at risk of losing the entire brood. Laying eggs in the nests of other birds can spread the risk and increase the chance that at least some young survive. This diversification of parental effort is a case of “spreading the risk” or “bet hedging” (Seger and Brockman 1987, Brown and Brown 1996) and in particular it is a case of “risk aversion” by spatially dispersing the reproductive effort (Slatkin 1974)— much as some birds may be risk averse in nesting repeatedly, each time with a small clutch.When there is a high risk of nest predation (the chance of losing a nest to a predator is as high as 0.9 in some regions, Ricklefs 1969), dispersing the risk across several nests may increase reproductive success. The risk of nest predation may have led brood-parasitic birds to put their eggs up for adoption in other nests, and not “all their eggs in one basket” (Payne 1977b,c). Consider a female that either lays 4 eggs in one nest or lays them in several nests. Females that lay all the eggs in one nest have a success of 0 or 4, with the rate of nest predation determining the mean success. Females that lay each egg in a different nest have a reproductive success (RS) of 0, 1, 2, 3 or 4, and the proportion that have 0 success will be less than in females that lay all the eggs in one nest.The probability that all eggs will survive is the last term in expansion of a binomial (p ⫹ q)n ⫽ 1, where p is the probability of predation, q is the probability of not being taken by a predator, and n is the number of nests in which eggs are laid. Insofar as predators usually take all the eggs and young in a nest, the probability that at least one egg will survive is the same as the survival of at least one nest with the egg, 1 ⫺ pn. If predation risk p is high, then the chance that at least one egg of a female that lays her eggs in several nests will survive is higher than that of a female that lays all her eggs in one nest. In an analytic solution based on field data of risk, then p ⫽ 0.8 and q ⫽ 0.2, the chance that at least 1 egg in 4 survives, 1 ⫺ p4, is (1⫺0.41), or 0.59, higher

160 The Cuckoos than when all eggs are in the same nest q, or 0.20. The proportion of females with success of 4 (n, when all eggs survive) is less when females lay each egg in a separate nest, qn ⫽ (0.2)4, or 0.0016, than when females lay all their eggs in one nest, q1 ⫽ 0.2. Within a population the RS of females laying “all eggs in one nest” will vary, high in some females and low in others. In a large population the average success of females with all eggs in one nest is the same as that of females that spread the risk across nests. Although the likelihood of an individual female leaving any offspring is higher when she lays her eggs in the nests of other birds, her expected fitness is lower than in a female that has a successful nest with a full set of her own eggs. Laying eggs in the nest of another bird may spread the risk against nest predation, but eggs in the nest of another species are at risk of discrimination and rejection by the nesting birds. If the probability of rejection (i.e., infanticide) is high, then the better strategy may be to not subject all the offspring to adoption. Alternatives are to lay from 1 to 4 eggs in the bird’s own nest and the remaining eggs in other nests, in mixed laying strategies. If p is the probability of predation and r is the probability of rejection, then the expected number of surviving offspring in each strategy would be the expanded form of the equation n(1⫺p⫺(1⫺p)r)^n ⫹((n⫺1)(1⫺p⫺(1⫺p)r)^(n⫺1))(p⫹(1⫺p)r) ⫹ (( n ⫺ 2 ) ( 1 ⫺ p ⫺ ( 1 ⫺ p ) r ) ^ ( n ⫺ 2 )) ⫻(p⫹(1⫺p)r)^2 ⫹ (( n ⫺ 3 ) ( 1 ⫺ p ⫺ ( 1 ⫺ p ) r ) ^ ( n ⫺ 3 )) ⫻(p⫹(1⫺p)r)^3 . . . (I thank Charles Rosa for this formulation). This model refers to the probability of rejection, given no predation; that is, r is a conditional probability. In addition, the value of r is less when the bird lays all her eggs in her own nest than when she lays in the nests of other females. For example, with a risk of predation of 0.8 and a risk of rejection of 0.7, scattering all the eggs would be only marginally more successful than a mixed strategy of laying

some eggs in the nests of other birds and rearing the other eggs herself. If r ⬎ p, then the expected number of survivors in the second term of the equation could drop below the first term, “all eggs in one basket”. The dispersion strategy of one egg in each nest occurs at the value of r that makes the above expression ⬍ n(1 ⫺ p). In addition to risk of rejection, there is a cost in searching for nests, especially if the female also cares for a nest of her own. The trade-off between predation and rejection is somewhat like an investment problem where putting some resources out of one’s direct control reduces the chance of total loss, but also has a risk not present in an undiversified folio.The model of risk aversion is sensitive to both r and p. Consistent with the theory of risk management, when values were determined in cuckoos and their hosts, the observed values of egg rejection r were less than the values of nest predation p (Russell and Rowley 1993, Davies et al. 1996, Stokke et al. 1996, Rowley and Russell 1997). Regional variation in nest predation and the number of brood-parasitic species allow a partial test of the idea of brood parasitism as a strategy to minimize risk of loss. Predation risk on nests is high in the regions where brood parasitism is common. In the northern temperate region, fewer than half the bird nests are taken by a predator, whereas in tropical forest regions nearly 90% of nests are taken by a predator (Snow 1962, Ricklefs 1969, Willis 1973, Barnard and Markus 1990, Masuda and Ramanampamonjy 1996). Brood parasites are more diverse in tropical regions where the risk of nest predation is high. In northern temperate regions, only one species (Brown-headed Cowbird) is a brood parasite in most of North America; and whereas seven cuckoos (Hierococcyx hyperythrus, H. nisicolor, Cuculus micropterus, C. poliocephalus, C. canorus, C. optatus, C. saturatus) are brood parasites in the Old World, no more than four occur in a local site. In temperate Argentina there are three cowbirds and three cuckoos, and nearly the same number of brood parasites occur in the New World tropics with four cowbirds and three cuckoos. In tropical Africa there are 20 brood-parasitic finches, 14 brood-parasitic cuckoos, and 15 brood-parasitic

The evolution of brood parasitism in cuckoos 161 honeyguides (Sibley and Monroe 1990, Payne 1996). In tropical Asia the number of brood-parasitic cuckoos within a local region is as high as 13 along with a parasitic honeyguide (Malay Peninsula:Wells 1999). Although there are more species of birds in the tropics than in temperate regions, the number and diversity of brood parasites in the Old World tropics may be disproportionate to the total number of bird species in that region.

Cost of reproduction Even though the risk of nest predation may affect the evolution of brood parasitism, the extra parental care given to their young is the most conspicuous feature of reproductive behavior in altricial birds. Obligate brood parasitism is more common in altricial birds where it has evolved independently in several different families, but only once in the precocial birds where there is no extended parental care after the young has hatched (Sorenson and Payne 2001, 2002).The dependence of the altricial young through a long period is a major reason that cuckoos have evolved brood parasitism to exploit the parental care of other birds, because if the number of young produced is limited by parental care that the parents can provide, then altricial birds can lay more eggs and produce more young when they have at least some of their young reared by other birds (Payne 1965, 1977b, Lyon and Eadie 1991). The parental care provided by another species allows a female brood parasite to lay more eggs and leave more chicks than the bird could provide with its own parental care, much as Darwin (1859) suggested for a facultative brood parasite that lays some eggs in the nests of other birds, and rears other eggs the chicks of which it feeds in its own nest.This reasoning implies that reproduction is not constrained by a set number of eggs, as it is in the risk model. Rather, there is a trade-off between reproductive costs due to forming the eggs and costs due to parental care, so with no parental care, the brood-parasitic birds are expected to lay more eggs.When the number of eggs is greater than the difference between the expected survival of eggs in

a bird’s own nest and the survival of eggs in other nests, the brood parasite can beat the odds of predation as well as the costs of reproduction. In a test of this idea, brood-parasitic birds including the cuckoos lay more eggs than their nesting relatives (Payne 1974, 1977b,c). Further, when siblicidal behaviors that increase fitness by gaining parental care are selected in a facultative brood parasite, a mixed-strategy of facultative brood parasitism is expected to be less successful than a strategy of obligate brood parasitism. Once a lineage laid in the nests of other birds, aggressive behavior of these nestlings could favor the lineage that attacked or removed the nestmates (Roche and Glanz 1998). Survival of chicks in a nest is lower when more than one such supercompetitor cuckoo is in the nest (Robert and Sorci 2001). That is, obligate brood parasitism can be selected rapidly once the young are siblicidal, as they are in many obligate brood-parasitic cuckoos. The evolution of brood parasitism involves the number of eggs laid in pure and mixed laying strategies, the success of young in a nest with no other cuckoo and with other competitive cuckoos, and the costs of parental care in estimating optimal reproductive strategies.

Phylogeny and behavior models of brood parasitism Three models of breeding behavior Three life styles may have been involved in the initial evolution of brood parasitism. Each reproductive style could give an advantage to an opportunistic female, when her young gained extra-parental care as well as care by the breeding female (Payne 1977a, 1998). First, brood parasitism may have evolved from occasional laying in the nests of others of her own species. Second, brood parasitism may have involved frequent laying in the nest of cooperative breeding conspecifics such as in the communally nesting anis. Third, brood parasitism may have involved birds taking over the nest of other birds, first to rear their own young and then to leave their young to the care of

162 The Cuckoos the nesting bird. Each of these behaviors have been observed in bird populations, including some species that are related to the brood-parasitic cuckoos. To infer the likely historical origin of brood parasitism in a lineage of the cuckoos, these three summary models of behavior were compared among birds. Then the phylogenetic relationships of the brood parasites and their nesting relatives were compared, to determine which of these breeding behaviors occur in the lineages of the nesting birds that are most closely related to the brood-parasitic birds.

Intraspecific brood parasitism Obligate brood parasitism may have evolved through a mixed or conditional reproductive strategy (Payne 1977b,c) of birds to rear their own young and to lay an egg in the nests of neighbors. If these females laid more eggs, and these eggs were hatched and reared by neighbors while the errant females also reared their own brood, then they would achieve a greater reproductive success than females that laid only in their own nest and reared their own young, and the behavior would be selected. Black-billed Cuckoos and Yellow-billed Cuckoos usually rear their own young, but sometimes they lay eggs in other nests of their own species or another cuckoo’s nest (Fleischer et al. 1985). Darwin (1859) was aware of these nesting cuckoos’ behavior. He reasoned, “if the old bird profited by this occasional habit”, then laying eggs in a neighbor’s nest could lead to more offspring and to the evolution of brood parasitism. Facultative brood parasitism in nesting cuckoos suggests an adaptive precursor to obligate brood parasitism. Nestling Black-billed and Yellow-billed cuckoos have the same pattern of spots on the palate and tongue, displayed when they beg from the parents (Plate 20a,b). These patterns are likely to be traits of their common ancestor, rather than one species being a nestling mimic of the other. Facultative brood parasitism is common only in the precocial waterfowl. This behavior in the precocial birds argues against the importance of facultative brood parasitism as a necessary precursor for the evolution of obligate brood parasitism in altricial birds such as the cuckoos. In waterfowl, this

facultative parasitism involves the visual stimulus of an early nesting species such as Canvasback Aythya valisineria for the later nesting birds such as Redhead A. americana (Sorenson 1991, 1993, 1997), or nests available in which to lay when their own nests fail in Snow Goose Chen caerulescens that nest close to neighbors (Cooke 1987), or birds that lived together in their natal year (Andersson and Åhlund 2000). In some cases a female may specialize in laying in the nest of another. But in most field observations, a female’s decision to lay in the nest of another is related to her own condition or her recent nesting failure. A female whose nest is taken by a predator is likely to lay a few eggs in the nest of a neighbor, and a young female that arrives early when other ducks are nesting is likely to lay some eggs in the nest of another duck before she lays in her own nest.These females would leave more offspring on average than females that laid only in their own nest (Sorenson 1991, Åhlund and Andersson 2001), and it would be of interest to compare the seasonal success of the individual birds. The same idea applies to altricial birds. Facultative brood parasitism is known in some altricial species, especially birds nesting in dense colonies such as the Cliff Swallow Hirundo pyrrhonota (Brown 1984, Brown and Brown 1988, 1989, 1996, 1998) and Ploceus weaver finches (Dhindsa 1983, 1990, Jackson 1998).

Cooperative breeding Interspecific brood parasitism may originate with cooperative breeding.The cooperatively nesting anis and Guira Cuckoo live in social groups where several pairs share a nest. One female may be socially dominant over the others in an ani group, at least within a nesting cycle, and she tosses some eggs of other females from the nest before she lays. She and her mate incubate the clutch, while other pairs continue to incubate, brood and feed the young, some of them perhaps their own (Vehrencamp et al. 1986, Koford et al. 1990, Macedo 1992, 1994, Quinn and Startek-Foote 2000). Guira Cuckoos have laid in the nest of Smooth-billed Anis (Azara 1809). Parallels in the breeding of these crotophagine cuckoos with the breeding of female paper wasps

The evolution of brood parasitism in cuckoos 163 Polistes which fight over a nest, some winning and others abandoning, suggested to Hamilton (1964) the source of brood parasitism in cuckoos, as in Guira Cuckoos which remove the eggs of other females in their nest, and the altruistic behavior of guiras and anis in rearing any young in the nest. In this model, shared nesting provides a context for a female to lay in the nest of a conspecific. An unresolved problem is to shift the parental care to another species while retaining an advantage to the individual that is a parasite.

Nest takeover A bird may take over the nest of another species, then lay its eggs and care for its own young. In mixed-species colonies of weavers in Africa, Chestnut Sparrows Passer eminibey usurp other birds’ nests, though at times they build their own nests.The sparrows fight with weavers for access to fresh weaver nests, then lay their eggs and rear their own young (Betts 1966, Payne 1969c). In Kenya I watched the sparrows display persistently at fresh green nests that were built by other species, including Grey-capped Social-weaver Pseudonigrita arnaudi at Olorgesailie in 1967 and 1976, Goldenbacked Weaver Ploceus jacksoni and Black-headed Weaver at Kisumu in 1976, and Cardinal Quelea Quelea cardinalis at Kacheliba in 1976, and even at fresh, dry stick nests of White-billed Buffalo-weaver Bubalornis albirostris at Kacheliba in 1976.The sparrows displayed on the nests, attracted females, copulated on the nests, and entered the nests, the males first then the females. Both males and females brought a few nest materials into the nest and lined the nest. The nestbuilding birds chased away the sparrows, but the sparrows were persistent and they returned and displayed until the weavers deserted. Sparrows took over the nests of the other species; they were unsuccessful with the much larger buffalo-weavers. In other mixed-species finch colonies, some estrildid finches (Cut-throat Finch Amadina fasciata, African Silverbill Euodice cantans and Indian Silverbill E. malabarica) do the same, appropriating active nests, taking over old nests of weavers, or building their own nests (van Someren and van Someren 1945, Goodwin 1982, Sorenson

and Payne 2001). If a sparrow or finch were to leave its own egg when the nestbuilder is laying, rather than driving off the nestbuilder and appropriating the nest, then the weavers might rear the sparrow or finch young and these young would gain extra parental care from the weavers. The situation calls first for a mixed-species interaction (perhaps laying in other nests of their own species or sharing a nest between females, as in the silverbills), then a change in the behavior of the candidate parasitic bird to leave its eggs in the nest for the nestbuilding pair to incubate and rear, rather than driving away its potential host (Payne 1977b).

Phylogeny and the lineages of brood-parasitic birds In the larger view, it is likely that nesting evolved once in modern birds then was lost independently within a few lineages, three times in the cuckoos and four times in other brood-parasitic birds (once each in the waterfowl Anatidae, the honeyguides Indicatoridae, the parasitic finches Viduidae, and the parasitic cowbirds Icteridae, Payne 1977b, Sorenson and Payne 2001, 2002), and was lost in birds that breed in unusual ecological contexts such as penguins Spheniscidae that incubate their eggs on their feet (Williams 1995) and megapodes Megapodiidae that bury their eggs in the ground and leave their eggs to hatch with decomposing organic, solar or geothermal sources of heat, into precocial waifs without parental care ( Jones et al. 1995). The phylogenies uncovered in the mitochondrial genetic analyses of the cuckoos give an estimate of the number of times that cuckoos have evolved brood parasitism and the behaviors of the closest relatives of each of these parasitic lineages. Information on the phylogeny of brood parasites and their nesting relatives allows an inference of the breeding behavior of their most recent common ancestor. Using this line of reasoning, a variation on the theme of nest takeover was proposed for the origin of brood parasitism in cowbirds (Friedmann 1929). The Bay-winged Cowbird “Molothrus” badius takes over the nests of other birds, then lays its eggs and rears its own young in the

164 The Cuckoos nest (Hudson 1874, Sclater and Hudson 1888). The Bay-winged Cowbird in turn is parasitized by an obligate brood parasite, the Shining Cowbird Molothrus bonariensis. Current estimates of the evolution of icterids suggest that the nesting cowbird is not closely related to the brood-parasitic cowbirds: Bay-wing is now recognized in another genus, Oreopsar badius (Johnson and Lanyon 1999, Lanyon and Omland 1999). On the other hand, the birds most closely related to the brood-parasitic finches Viduidae are the waxbills Estrildidae. Brood parasitism evolved only once in the finches, including the cuckoo-finch Anomalospiza imberbis as well as the Vidua whydahs and indigobirds. Many species of estrildids appropriate the old nest of other nesting birds rather than building their own nest, but only a few are known to lay their eggs in the active nest of other species and none are cooperative breeders. The phylogeny of the Old World finches suggests that nest appropriation is the most likely behavioral origin of brood parasitism in the viduid finches (Sorenson and Payne 2001).The brood-parasitic honeyguides Indicatoridae are most closely related either to woodpeckers Picidae or to barbets Capitonidae (Short and Horne 2001). There have been no genetic sequence data published to resolve the relationship among these three groups. If the honeyguides are more closely related to the barbets, as suggested by Short and Horne (2001), then the nest appropriation behavior of the barbets in taking over old holes of other hole-nesting birds may have been involved in the evolution of brood parasitism in the honeyguides. Because most cuckoos rear their own young and the basal branches in all clades involving brood parasites are birds that build a nest and rear their own young, there is good reason to use phylogeny to estimate the breeding behavior of nesting ancestors in each clade of brood-parasitic cuckoos, and to infer that the nesting behavior of the ancestor was a behavior that facilitated the evolution of brood parasitism. In the cuckoos, brood parasitism has originated independently in the Old World and the New World. New World brood-parasitic cuckoos are more closely related to the terrestrial New World ground cuckoos than to the communal and cooperatively breeding anis and guira, where several

females lay in the same nest and care for the common brood. Old World brood-parasitic cuckoos are closely related to the nesting Old World cuckoos. The independent origin of brood parasitism in Clamator, and its cladistic relationship to the malkohas were unexpected. The basal bird in the Cuculinae is the Raffles’s Malkoha and as far as known it makes its own nest and rears its own young. The closest relatives of Old World brood parasites such as Cuculus are socially monogamous and territorial nesting birds. The molecular phylogeny gives no support to the idea that brood parasites were ancestral to nesting cuckoos, as Hughes (1997a,d) and Aragón et al. (1999) proposed for the nesting Black-billed Cuckoos. Traits that are associated with brood parasitism in the cuckoos include courtship feeding, mimicry in color and pattern of the host egg, and the siblicidal behavior of nestling cuckoos in evicting the host eggs. Courtship feeding that occurs in several lineages of Old World brood-parasitic cuckoos may be retained from courtship feeding in the nesting cuckoos, as this behavior is widespread in nesting cuckoos. Mimicry of the host eggs has evolved at least twice. The New World brood-parasitic Tapera have variable egg colors, and Dromococcyx have spotted eggs, unlike the eggs of nesting cuckoos and somewhat like the eggs of the cuckoos’ hosts. The Old World cuckoos evolved mimicry at least twice, once in Clamator (Clamator jacobinus in part of its range has white eggs; most Clamator eggs are blue like their hosts’ eggs, and unlike the white eggs of the nesting cuckoos to which they are most closely related), and at least one other time with the colored and spotted eggs in the main lineage of brood-parasitic Cuculini). Siblicide with the sharp hooked bill evolved in the New World Tapera. Finally, egg eviction by nestling cuckoos appears to have evolved twice, once in the Thick-billed Cuckoo Pachycoccyx and another time in the Old World Cuckoo lineage before the Cuculus and Chrysococcyx lineages diverged from each other.The four crested cuckoos Clamator, Common Koel, Channel-billed Cuckoo and Long-tailed Cuckoo generally do not evict, whereas Pachycoccyx and the Chrysococcyx and Cuculus lineages of brood-parasitic cuckoos evict host eggs from the nest.

The evolution of brood parasitism in cuckoos 165 In the New World, the birds that are most closely related to Tapera and Dromococcyx are the roadrunners and ground cuckoos.The sister group of these New World cuckoos are the anis and Guira Cuckoo, all cooperative breeders. The phylogeny of New World brood-parasitic cuckoos can be interpreted in two ways: if the most recent common ancestors of the neomorphine cuckoos are taken as the referent, then the second hypothesis is supported and cooperative breeding may have been the starting point in the evolution of brood parasitism in the New World cuckoos. If the sister group of the New World broodparasitic cuckoos is taken as the referent, then the first model (facultative brood parasitism) and the third model (nest takeover) are both supported: in Greater Roadrunner, two females may lay in a single nest and a female may lay in the nest of other kinds of birds. The model of cooperative breeding cannot be rejected from the molecular phylogeny. Cooperative breeding evolved only once in the cuckoos (the crotophagines), so there is no evidence in the tree as a whole to support the notion that cooperative breeding occurred in a common ancestor of the neomorphines and crotophagines, nor that the neomorphines have traits of the cooperative breeders. In both Old World and New World cuckoos, the third model of behavior, the usurpation or claiming of old nests or active nests of other species cannot be rejected, as both the roadrunners Geococcyx and

the cuculine Coccyzus and Coccycua cuckoos from time to time take over the nest of other birds. Nevertheless, the first hypothesis of nesting behavior, that of facultative brood parasitism, appears likely to have been an adaptive precursor of obligate brood parasitism in the cuckoos. Both the New World coccyzine cuckoos (including Black-billed, Yellow-billed, Dwarf and Dark-billed Cuckoos) sometimes lay their eggs in the nests of other conspecifics or in the nests of other kinds of birds, as described in the species accounts. In Dwarf Cuckoo and Dark-billed Cuckoo two females may lay in the same nest, but whether these females mated with the same male or intruded on a nesting pair is unknown. Close relatives of the crested cuckoos Clamator include two Philippine malkohas whose nests are unknown. In Sulawesi a young Yellowbilled Malkoha was fed by another species, a sunbird, and this behavior suggests possible occasional laying and rearing of malkohas in a sunbird nest. What is available from the estimate of conditions that led to brood parasitism are observations of occasional brood parasitism in the groups most closely related to the brood-parasitic cuckoos. The phylogenetic relationships of brood-parasitic cuckoos and nesting cuckoos show that Darwin (1859) had the right idea. So did Friedmann (1933), who first argued that brood parasitism evolved independently in the Old World and New World cuckoos.

This page intentionally left blank

PART II Species accounts

This page intentionally left blank

Crotophaginae Genus Guira Lesson Guira Lesson, 1830, Traité d’Ornithologie, livre 2, p. 149. Large cuckoos of the New World with streaked brown plumage, a crest, and eight rectrices (vs 10 in other cuckoos, except Crotophaga

which also has eight).Type, by monotypy and tautonymy, Cuculus Guira Gmelin. Guira is an application of güirá , a Paraguayan Indian term for bird. One species.

Guira Cuckoo Guira guira (Gmelin, 1788) Cuculus Guira Gmelin, 1788, Systema Naturae 1, pt. 1, p. 414. (Brazil) Other names: Anú branco, White Ani, Piririgüa (Azara). Monotypic.

Description ADULT: Sexes alike, appearance shaggy, head ochre with dark shaft streaks, crest orange-rufous, upper back brown, feathers streaked with a white shaft and light gray edges.Wing coverts dark brown with white edges, wing dark brown with inner vane of wing feathers rufous, outer vane brown with a pale gray edge, giving a frosted appearance.The lower back is white, the rump is buff.Tail blackish with a yellowish base and a broad white band at the tips of T2 to T4, T1 dark brown with a yellowish base: Below pale ochre to whitish, throat and breast streaked brown, under wing whitish, darker on the secondaries; bare facial skin yellow to pea green; iris yellow to orange; bill orange to yellow; legs bluish gray. JUVENILE: Similar to adult, except that the tail has a narrower terminal white band (white of T2 is about 18 mm, white of T3 is 22–28 mm, white of T4 is 20 mm; vs 30, 35–38, and 38–40 mm in the adult); iris light gray; bill black and white, or gray. NESTLING: Skin black and face pale orange, above with long yellowish-white hair-like down, dark

brown on the tips of the primaries and whitish on the tips of rectrices; underparts with short white hair-like down; bill light orange with black ridge along culmen and mandibular rami; iris dark brown; feet black; mouth lining bright pink with raised papilla pads, with a white lateral bar just behind the tip and a white spot behind that, a large circle (open anteriolaterally) on each side of the palate, a ridge of small white papillae at the posterior end of the anterior palate groove, and a fringed white bar perpendicular to the choana pars caudalis; the tongue pink with a black band and white tip (UMMZ photos). SOURCES: AMNH, BMNH, CM, FMNH, MVZ, ROM, UMMZ, USNM, ZMA, ZSM.

Measurements and weights Wing, M (n ⫽ 10) 170–187 (178.5 ⫾ 6.1), F (n ⫽ 8) 161–180 (174.3 ⫾ 5.9); tail, M 214–235 (223.2 ⫾ 6.6), F 198–243 (227.5 ⫾ 13.2); bill, M 27–32 (29.2 ⫾ 1.4), F 36–40 (27.88 ⫾ 1.5); tarsus, M 36–40 (38.0 ⫾ 1.6), F 36–40 (38.0 ⫾ 1.8) (UMMZ). Weight, M (n ⫽ 19) 128.6–168.2 (138.7), F (n ⫽ 8) 113.0–168.6 (144.8) (AMNH, USNM). Wing formula, P7 ⫽ 6 ⬎ 5 ⬎ 8 ⬎ 4 ⬎ 3 ⬎ 9 ⫽ 2 ⬎ 1 ⬎ 10.

Field characters Overall length 36 cm. A long-tailed, brown bird with a shaggy crest, streaked above with a rusty

170 Guira Cuckoo Guira guira crown, streaked below on the breast, the tail with a white base and white tip when seen from below; the wings rufous in flight; the birds often in flocks. Tail jerks up and down, sideways and diagonally.

Voice Noisy, with a great variety of notes, including plaintive whistles “pio..pio..pio..pr..prr..prrr”, yodels, descending phrases “kee-ey, kee-ey, kee-ey, kee-orr, keeorr, cure curecure”; flight call a quiet “yew yew yew yew . . .”; also guttural calls, a high gargled trill, and a “creep” (Azara 1809, Friedmann 1927, Eisentraut 1935, Davis 1940b, Fandiño Mariño 1986, 1989, Hardy et al. 1990, Macedo 1992, Sick 1993).

Range and status South America in Bolivia, Brazil (except in forested regions), Paraguay, Uruguay and Argentina (Hudson 1920, Laubmann 1939, Short 1975, Hilty and Brown 1986, Remsen et al. 1986, Remsen and Traylor 1989, Fjeldså and Krabbe 1990, Narosky and di Giacomo 1993, Sick 1993, Novaes and Cunha Lima 1998, de la Peña and Rumboll 1998, Magalhães 1999). An emaciated vagrant was found in Curaçao (ZMA 28750,Voous 1957); the cuckoo is unknown on mainland northern South America. Resident through their range; in southern part of range more numerous in summer than in winter (Cuello and Gerzenstein 1962). Guira have increased their distributional range in the last 100 to 200 years, appearing in open lands when the forests

have been cleared (Azara 1809, Sclater and Hudson 1889, Sick 1993).

Habitat and general habits Second-growth scrub, drier tree and scrub savanna and scrub woodlands, pampas, pastures, fields, coastal dunes; sea level to over 1200 m (Stotz et al. 1996). They live in drier areas than Smooth-billed Anis Crotophaga ani. They feed on the ground and roost in trees. Common in non-forested areas such as Mato Grosso and the Amazonian savannas, and deforested areas near towns and pastures, they disappear when an area becomes forested, and they are absent in forest areas of Amazonia (Sick 1993). Gregarious, group-living, the members of a group breed communally (Davis 1940b, 1942, Reichholf 1974, Salvador 1981, Cavalcanti et al. 1991, Macedo 1992, 1994, Souza 1995, Stotz et al. 1996). Live and feed in flocks; mean size 6 to 8, with 20 birds in the largest flocks; the members of a pair preen each other (Davis 1940b, Durrell 1956). Birds huddle together on cold days and perch together in a tree at night. On some nights the birds of a flock roost together or in sub-groups. The cuckoos have heavily pigmented black skin between the dorsal feather tracts (UMMZ 236027). On cool days, the wings drop to the side with feathers raised, the black skin exposed to the sun (Sclater and Hudson 1889, Durrell 1956, Storer 1989). Night after night the birds roost in the same trees, huddling together, facing outward, with as many as 14 birds in a clump. This group roosting behavior may help the birds save energy; nevertheless some birds die during cold nights (Friedmann 1927, Davis 1940b, Belton 1984, Gallardo 1984). In spring some groups split into pairs to breed (Sclater and Hudson 1889, Sick 1993). Groups defend territories from other social groups, and some nesting pairs hold breeding territories within these areas and defend these from other pairs in their group (Davis 1940b, 1942, Macedo 1992, Macedo and Bianchi 1997a). While feeding, the birds move from a perch to the ground, or they stalk across the ground in a group; they walk, jump and run after prey which when captured, they swallow whole (Martins and Donatelli 2001).

Guira Cuckoo Guira guira 171

Food Insects, mainly grasshoppers; also cicadas, flying termites, frogs, small reptiles; eggs and nestlings of small birds such as Pied Water-tyrant Fluvicola pica and Fork-tailed Flycatcher Tyrannus savanna, (Azara 1809, Sclater and Hudson 1889, Wetmore 1926, Friedmann 1927, Belton 1984, Gallardo 1984, Mason 1985, de la Peña 1995, Martins and Donatelli 2001). They also take frogs, mice and rats in captivity (Pagel 1992, R. Macedo, in litt.). Nestlings are fed invertebrates and lizards. Most items brought to the young in the nest are ⬍ 6 cm; though some are ⬎ 12 cm (Melo and Macedo 1997).

Displays and breeding behavior The birds live in pairs and in groups throughout the year. In agonistic behavior the guiras raise and open their wings, erect their head feathers in a crest and parade up and down in front of each other with a specific call. One bird preens another, the preening bird with its head feathers sleeked and the tail feathers overlapped into a narrow length, and the head and body held lower than those of its social partner; the bird receiving the preening in a more erect posture, fluffing its head feathers, erecting its crest and closing its eyes, and spreading its tail (Fandiño Mariño 1986, Figure 9.1). Pairing behavior is simple and takes place during the everyday social life of the group without obvious sexual display. Copulation has not been seen and is apparently infrequent. Paired birds carry a leaf in the bill and take the material to a nest site, then bring in more sticks and leaves. Birds are vocal both when nesting and when in nonbreeding groups (Davis 1940b, Gallardo 1984, Macedo 1992; R. Macedo, in litt.). The birds do not appear to behave aggressively in the nest (R. Macedo, in litt.).

Breeding and life cycle In tropical Brazil the cuckoos breed in the dry months May–August, near Brasilia mainly during the rains from August to November (also December to March: Melo and Macedo 1997), near Rio de

Janeiro from August to November, and in Rio Grande do Sul from November to February (Snethlage 1928, Belton 1984). In Uruguay they breed in November and December (Narosky and Yzurieta 1987), and in Argentina from October to February (de la Peña 1995).The nest is a large open platform of sticks and twigs, 15–40 cm in diameter and 14–20 cm high, placed high in a thorny tree or a cactus. Guiras often use and renovate an old nest from the previous season, and add green leaves to line the nest (Friedmann 1927, de la Peña 1995). They sometimes take over old unused nests of other species of birds such as mockingbirds Mimus and add nest material to these (Daguerre 1924). Sometimes guiras lay in nests of other birds, even those of the caracara Milvago chimango and lapwing Vanellus chilensis (Serie 1923a,b, Daguerre 1924, Eisentraut 1935, Sick 1993, Jenny 1997). Eggs are blue to greenish blue and are covered with whitish chalky splotches and streaks raised in relief. Egg size is remarkably variable within a population, 37.2 to 50.8 mm in length, width of 2.78 to 3.86 mm, a mean of 42.7 ⫻ 31.9 mm (n ⫽ 560 eggs); egg weight is 1.67 to 3.22 g (mean 2.46 g, n ⫽ 451 eggs) (R. Macedo, in litt.). Eggs laid by a single female are as different in size and appearance as eggs laid by different females. A few nests have only one breeding pair, whereas in most groups several females lay in the same nest, the number of eggs being larger in the larger groups. Clutch size averages 8–10, ranging from 4 to 30 eggs in a nest. During the period of laying, it is common to find 2, 3 or even 4 eggs laid on the same day (Cariello et al. 2002, 2004).The incubation period is 12–13 days (Macedo and Bianchi 1997a), for 15 days, (de la Peña 1995). The young usually hatch on the same day, or as much as four days apart. The nestling period varies with nesting conditions: the young leave the nest by day 5–6 when disturbed, whereas in undisturbed nests they fledge at day 12–15.The young beg with loud calls and flap their wings when an adult approaches with food. Nestlings that fall from the nest climb up the tree back to the nest, using their spread wings to grasp the trunk, even in an Auracaria where the trunk is covered with spiky leaves. Young guiras

172 Greater Ani Crotophaga major are malodorous in the nest, perhaps due to their carnivorous diet; they leave the nest when more than half grown and are fully adult in size within a month (R. Macedo, in litt.). Parental care continues for three weeks after fledging. Nesting success: 26% of eggs and 55% of hatchlings survive to fledge. Eggs and nestlings are removed or killed by certain adults in the breeding group, and this is the greatest source of loss of eggs and chicks in the nest (Gallardo 1984, Macedo 1992, 1994, Quinn

et al. 1994, Macedo and Bianchi 1997b, Macedo and Melo 1999). The nests are occasionally parasitized by Screaming Cowbirds Molothrus bonariensis (de la Peña 1995). Across one to three nesting attempts within a year, breeding groups lay on average 13 eggs in a season (Cariello et al., in press). Guira Cuckoos have bred in captivity, and hand-reared birds socialize with humans and become inquisitive pets (Sclater and Hudson 1889, Leverkühn 1894, Pagel 1992).

Genus Crotophaga Linnaeus, 1758 Crotophaga Linnaeus, 1758, Systema Naturae (ed. 10) 1, 105. Large cuckoos of the New World with black plumage, a deep and laterally compressed bill, no crest, and eight rectrices (vs 10 in other cuckoos, except Guira which also has eight).Type, by monotypy, Crotophaga ani Linnaeus 1758. All species are

cooperative breeders, either usually or part of the time. The name Crotophaga means “tick eater” (Gr. kroton, tick; phagos to eat) and some are called “tick birds” as they associate with cattle and eat ticks, or did before the period of chemical pesticides.Three species.

Greater Ani Crotophaga major Gmelin, 1788 Crotophaga major Gmelin, 1788, Systema Naturae ed. 10, 1, p. 363. (Cayenne) Monotypic.

NESTLING: Naked, skin black, corners of mouth yellow. The growing black feathers retain a tip of the natal sheath, which is lost at fledging.

Description

SOURCES: AMNH, ANSP, BMNH, CM, FMNH, MCZ, MVZ, ROM, UMMZ, USNM.

ADULT: Sexes alike, plumage glossy black, the narrow nape feathers and the broad back and breast feathers with a dull blue gloss and edged with a bright glossy bronze-green; wing black glossed blue, tail long, broad, black glossed purple, rectrices broad at the tip; iris greenish white; bill black, compressed with a high arched ridge on upper mandible, with two parallel grooves, one along base of ridge and the other from nostril towards tip of bill; feet black. JUVENILE: Wing glossed blue, tail glossed purple, the rectrices narrower at the tip than in the adult, the bill compressed but without the elevated ridge; iris dark brown.

Measurements and weights Wing, M (n ⫽ 10) 184–217 (202.6 ⫾ 8.7), F (n ⫽ 8) 192–211 (201.0 ⫾ 6.7); tail, M 252–282 (271.8 ⫾ 9.1), F 248–276 (262.6 ⫾ 10.2); bill, M 46–48 (47.2 ⫾ 0.9), F 43–46 (45.1 ⫾ 1.5); bill depth at nostrils, M 22.1–24.7 (23.3 ⫾ 0.8), F 21.0–24.4 (22.1 ⫾ 1.1); tarsus, M 42–44 (42.6 ⫾ 1.0), F 38–41 (39.4 ⫾ 0.9) (UMMZ). Weight, M (n ⫽ 11) 139.5–259 (171.2), F (n ⫽ 3) 145.1–156 (151.6) (AMNH,ANSP, FMNH, MVZ, USNM, UMMZ). Wing formula, P7 ⬎ 6 ⬎ 8 ⫽ 5 ⬎ 9 ⬎ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 10.

Greater Ani Crotophaga major 173

Field characters Overall length 46 cm. Long-tailed black bird with a crested and compressed bill and pale greenish to white eye (dark in other anis); larger than other anis and a stronger flier. Ridge on bill is highest well anterior to nostrils, and the bill is tapered (rounded in Smooth-billed Ani C. ani). Juveniles with dark eye are identified by their association with adults.

Voice Noisy, a guttural to bubbling, melodious “korokoro” or “toodle-doodle-doodle” given by several birds in a group when they fly or gather on a branch. Flock call, a loud croak “kqua”; alarm call, a harsh raspy note; other calls variously described as growls, croaks, grates, hisses and whirrs “shhhrrrrrrrr” (Davis 1941, Wetmore 1968, Hilty and Brown 1986, Hardy et al. 1990, Sick 1993). In the Orinoco region of Venezuela the sound of bubbling, boiling water gives the birds their local name, hervidor, derived from the Spanish verb hervir, to boil (Cherrie 1916).

1984, Cuello and Gerzenstein 1960, Hayes et al. 1994, de la Peña and Rumboll 1998, Hinkelmann and Fiebig 2001). Seen in São Paulo state, Brazil, in the 1950s, but not in the 1970s or later; locally extinct (Magalhães 1999). Fairly common in much of the range; uncommon south of Tropic of Capricorn, only two records in 100 years in Buenos Aires Province, Argentina (Narosky and di Giacomo 1993).

Range and status

Habitat and general habits

South America, from E Panamá (on the Caribbean slope from western Colón eastward; on the Pacific slope from the Canal area eastward), Colombia, Venezuela, Trinidad, and the Guianas southward through tropical South America east of the Andes to E Ecuador and E Peru, E Bolivia, Paraguay, Brazil, Uruguay and N Argentina. Common along the upper Amazon River in Brazil (Gyldenstolpe 1945a), occasional elsewhere in Amazonia with a record of a lone bird suggesting dispersal (CohnHaft et al. 1997). A population occurs in SW Ecuador at S Los Ríos, Guayas, east of Guayaquil and coastal El Oro (Phelps and Phelps 1958, Wetmore 1968, Davis 1993, Sick 1993,AOU 1998, Ridgely and Greenfield 2001). In addition to the birds in this range, two birds were taken in 1960 at Río Tamesí,Tamaulipas, Mexico, 1200 km north of the nearest known wild population in Panamá (Olson 1978). Resident in most of their range, or with local movements, breeding birds seasonally migratory in Paraguay south to Argentina (Belton

Flooded and riverine habitats, tropical evergreen forests near water, river-edge forests by oxbow lakes, gallery forests, in dense vegetation, thickets and trees along rivers, lakes, swamps and mangroves, wet forest edges, lake edges, grassy edges, thick vine growth and shrubbery, bamboo, marshes, often in vegetation overhanging water. They move into clearings and around houses, sometimes tame around humans; not far into forests. In tropical lowlands they occur mainly below 500 m, casually much higher in the savannas in postbreeding movements or migration, to 2600 m in the eastern Andes of Colombia, and to 2550 m in Cochabamba, Bolivia (Cherrie 1916, Wetmore 1968, Short 1975, Remsen et al. 1986, Remsen and Traylor 1989, Fjeldså and Krabbe 1990,Tostain et al. 1992, Haverschmidt and Mees 1994, Stotz et al. 1996, 1997). Like other anis, the body has a pungent odor. Gregarious, feed in flocks, occasionally follow army ants and squirrel monkey groups which flushout insects. Anis are arboreal and

174 Smooth-billed Ani Crotophaga ani terrestrial ground-gleaning insectivores (Willis 1983, Fjeldså and Krabbe 1990, Sick 1993). Groupliving, two or more pairs live and breed communally in loose colonies.A group defends its territory against other groups of anis (Belcher and Smooker 1936, Davis 1941, Robinson 1997).

Food Insects, mainly orthoptera (grasshoppers, roaches), caterpillars (Brassolidae), beetles, occasionally odonates; other arthropods, spiders, small vertebrates including fish; fruit, berries and euphorbia seeds (Wetmore 1968, Hilty and Brown 1986, Terborgh et al. 1990, Haverschmidt and Mees 1994, Lau et al. 1998).

Breeding In Trinidad they breed from August to November (Belcher and Smooker 1936), in Guyana from May to December (Young 1929), in Suriname from April to September (Haverschmidt and Mees 1994), in French Guiana in April (Tostain et al. 1992) and in Brazil (Pará) in May (Stone 1929). The nest is a bulky mass of sticks and vines, lined with green leaves, built on branches hanging 3–5 m over water. Eggs are greenish blue with a white chalky surface, subspherical or oval, 45 ⫻ 38 mm (Belcher and

Smooker 1936); egg weight 2.4 g (Hellebrekers 1942). Clutch size in Panamá is 2–3 (Wetmore 1968), in Venezuela the average clutch is 7.6 (Lau et al. 1998), in Brazil 3 (Stone 1929) and in Argentina 4–5 (de la Peña 1986). Clutches with more (e.g. 10) eggs in a nest may be eggs of more than one female. Broods with as many as seven young are successful and all young fledge from the nest (Willis and Eisenmann 1979). The incubation period is about 13–14 days.Adults defend nests only after the eggs have hatched; then the ani group successfully drives away Snail Kites Rostrhamus sociabilis and Black-collared Hawks Busarellus nigricollis. Nestlings when disturbed by day 5 or older are able to jump from the nest into the water below, and swim actively on the water surface for several meter to shore, where they run on the ground and conceal themselves in vegetation, then climb back into their nests (Lau et al. 1998). Nestlings also expel large amounts of dense cloacal fluid which has a strong odor and may deter a predator. Nestlings when undisturbed fledge 8–10 days after hatching and remain near the nest for a few more days. Breeding success is greater in nests in isolated sites than in continuous riparian vegetation (av. 4.8 fledged young, vs 1.3 young), due to more eggs hatching in the isolated nests (Lau et al. 1998).

Smooth-billed Ani Crotophaga ani Linnaeus, 1758 Crotophaga Ani Linnaeus, 1758, Systema Naturae (ed. 10), 1, p. 105. [ Jamaica] Other names: Black Witch, Tick Bird, Savanna Blackbird Monotypic.

Description ADULT: Sexes alike, plumage glossy black, head and nape feathers with bronze edges contrasting with the glossy bluish feathers on the back, the feather streaking wide and the tips of the feathers pointed, giving a curved appearance; the nape feathers stiff, wing black with a slight violet gloss, tail long, black, the rectrices broad and truncate; skin around eye black, iris brown to black; bill

arched high and laterally compressed into a thin keel, swollen at base, the arch more pronounced in male than female with the arch half the length of the bill; black, often with pale patches due to frequent scaling, some with a few shallow creases, the lower mandible with an angle on the lower edge; legs black. JUVENILE: Plumage dull black, rectrices more narrow than those of adult and the tips tapered; bill with shallow keel. NESTLING: Naked, skin black; palate pink with white raised patches, one in front and one on each palate fold, row of white marks behind tongue and

Smooth-billed Ani Crotophaga ani 175 U-shaped mark under the tongue; tip of tongue is black, edge of mouth whitish, egg-tooth is present; sheathed pin-feathers grow in a few days (Skutch 1966, Sick 1993, Quinn and Startek-Foote 2000).

calls, with certain calls given in distinct social contexts (Davis 1940a).

SOURCES: AMNH, BMNH, CM, FMNH, MCZ, MVZ, ROM, UMMZ, USNM.

South Florida, Bahama Islands, West Indies, Caribbean islands; mainland and islands of Quintana Roo in Mexico, Honduras, Nicaragua, Costa Rica, Panamá, Trinidad, Tobago, South America on the west coast south to Ecuador and Peru, and east of the Andes from Colombia, Venezuela, the Guianas and Brazil through the eastern lowlands to northern Uruguay and the Chaco region of Paraguay to northern Argentina (Misiones, Cordoba, Buenos Aires) and the Galapagos Is (Ridgway 1916, Blake 1956, Cuello and Gerzenstein 1962,Wetmore 1968, Short 1975, Remsen et al. 1986, Buden 1987, Benito-Espinal 1990, Fjeldså and Krabbe 1990, Narosky and di Giacomo 1993, Sick 1993, Biancucci 1995, Howell and Webb 1995, CohnHaft et al. 1997, Novaes and Cunha Lima 1998, Raffaele et al. 1998). In Ecuador and Peru they occur both east and west of the Andes, and are replaced in drier areas in the west by the Groovebilled Ani C. sulcirostris (Fjeldså and Krabbe 1990). Resident or locally nomadic. Local groups join together in flocks of 15–35 ⫹ (a record 65) birds in winter, while other groups remain in their breeding territories in all seasons (Loflin 1983). Anis are one of the first birds to colonize newly cleared areas. In Florida, anis became established and bred after a hurricane in 1926, then spread north in the 1960s and bred on the east and west coasts north to Tampa Bay (Merritt 1951, Fisk 1979, Stevenson and Anderson 1994). In the 1970s they became less common and now are local in the Florida Keys and north to West Palm Beach on the east coast and Collier County on the Gulf coast. After the breeding season, some birds disperse out of their breeding range and appear northward and west along the coasts and occasionally inland, to New Jersey, Pennsylvania and Ohio, but records of these are rare (McLean 1995, McLean et al. 1995, Mlodinow and Karlson 1999). Anis appear on the Dry Tortugas, southwest of Florida, apparently arrived from Cuba. In Central America the anis expanded their range during the 20th century.They

Measurements and weights Paraguay: Wing, M (n ⫽ 12) 143–159 (141.8 ⫾ 29.9), F (n ⫽ 12) 136–156 (145.9 ⫾ 6.0); tail, M 172–191 (180.8 ⫾ 5.8), F 160–188 (173.1 ⫾ 8.7); bill, M 30–33 (31.8 ⫾ 1.1), F 28–31 (29.6 ⫾ 1.00); bill depth at nostril, M 20.9–23.4 (21.9 ⫾ 0.9), F 19.2–21.4 (20.4 ⫾ 0.7); bill smaller in juveniles; tarsus, M 32–38 (34.2 ⫾ 1.8), F 32–35 (33.5 ⫾ 0.9) (UMMZ). Weight, Panamá: M (n ⫽ 11) 81.9–133.1 (110.5), F (n ⫽ 11) 77.8–115.6 (92.6) (UMMZ). Wing formula, P7 ⬎ 6 ⬎ 8 ⬎ 5 ⬎ 9 ⬎ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 10.

Field characters Overall length 35 cm. Black birds in flocks; large arched bill, the hump or arch extending above the forehead and in front of a notch between forehead and base of bill, the bill without grooves (sometimes shallow grooves on basal half of bill), the lower mandible gonys with an angle; the head and nape with a bronze gloss, contrasting with the darker green gloss of the back; tail relatively shorter than in the smaller Groove-billed Ani C. sulcirostris. The differences in bill shape of the two species are not apparent in juveniles. Feed on the ground, head and tail held high, outline concave above.

Voice Call, a whining whistle “ani”, “ah-nee”, “ooooeee” or “coo-reek keyoreek”, rising in pitch and given in flight, a querulous “que-lick”; also chatters, mewing with a thin descending “teeew”, growls and clucks (Gosse 1847, Gundlach 1874, Bendire 1895, Hilty and Brown 1986, Stiles and Skutch 1989, Downer and Sutton 1990, Hardy et al. 1990, Sick 1993, de la Peña and Rumboll 1998, Reynard and Sutton 2000).Anis have as many as 13 different

Range and status

176 Smooth-billed Ani Crotophaga ani

first appeared in southern Costa Rica in 1931; by 1975 they were in Guanacaste and replaced Groove-billed Anis in the southern half of the Pacific slope, and they expanded their range in Panamá from the 1940s to 1960s with the human activity of clearing the forests (Wetmore 1968). In the West Indies anis are common in the Bahamas, Greater Antilles, Virgin and Cayman Islands and Providencia; in the Lesser Antilles common on Dominica, St.Vincent and Granada; uncommon on Martinique and Guadeloupe; and rare or absent on other islands. Numbers fluctuate on small islands where birds are absent for a time, then recolonize and disappear again (Terborgh et al. 1978, Raffaele et al. 1998). One population survived a hurricane in 1989 with no decrease in numbers, when anis could feed on weakened migratory prey (Wauer 1989). Recent settlements in the Galapagos may be due to introduction by farmers (Harris 1973, 1982, Rosenberg et al. 1990). Common in much of their range.

Habitat and general habits Second-growth scrub, river island scrub, humid areas, clearings in forests, scrub and scattered trees in pastures, woodland, thickets, mangroves, lake margins, grassy edges, roadsides, marshes, thick vine growth, Cecropia and dying trees, and marshes, most common where they occur in outskirts of towns and old cultivation. In Puerto Rico they live in disturbed dry habitats where the vegetation is dominated by exotic grasses (buffel grass Centris ciliaris, guinea grass Panicum maximum) with scattered trees, often mesquite.They feed in fields of grass and use

their high crowned bill to separate the wet, dense vegetation as they forage (Willis 1983b, Quinn and Startek-Foote 2000). They live mainly in lowlands, from sea level to 500 m, occasionally higher (Stotz et al. 1996), to 2400 m north of Orinoco in Venezuela, to 2000 m or more in Colombia, 2400 m in southern Ecuador, and breed in the temparate side valleys of the Marañón Valley in northern Peru (Fjeldså and Krabbe 1990). In South America, their heart and lungs are longer, and body weight greater when they are above 2000 m in the Andes, than in the lowlands (Köster 1976). Anis live in small, noisy flocks of 2–11, occasionally of as many as 30 birds, each bird a few meters from the others (Loflin 1983) as they walk or hop on the ground and on branches, and rise into the air and fly in pursuit of insects. On the ground they feed with cattle, moving among the feet and head of the cattle. They take ticks from the legs of cattle and they sometimes perch on cattle (Gosse 1847,Wetmore 1916, 1968). Anis follow cattle, horses and indigenous mammals (tapu Dasypus sabanicola, de Visscher and Moratorio 1984); they follow a plow, and also army ants (Willis 1983b). They capture insects that are flushed out; and they catch other small animals that flee from grass fires. A bird finds food at twice the rate when it follows cattle, than when it feeds alone (Smith 1971). The flight is straight and rapid with shallow wingbeats and a short glide. Large groups of anis have larger territories than small groups. In Puerto Rico, a 3-adult group had territory of 6.2 ha, a group of 5 adults had 8.7 ha, and a group with 6 adults (increasing to 9) had 9.6 ha (Quinn and Startek-Foote 2000). Once established, a group can hold its territory against challenges from other groups, sometimes even when the resident group size changes (Davis 1940a). Anis often spread their wings and tail to sun after they wet their plumage while climbing through wet places and dense herbs; they appear lean and shabby (Bendire 1895).They allopreen, touch each other, expose their skin to the sun while they spread their wings and fluff their feathers. They huddle together during the rain, and at night they roost in marshes, huddling to save heat; at night their body temperatures become 8°C cooler than

Smooth-billed Ani Crotophaga ani 177 in the daytime; and they bask in the sun, with wings spread, to warm their bodies in the morning before they become active (Bent 1940, Merritt 1951,Warren 1960,Wetmore 1968).

Food Insects, mainly grasshoppers; also mantids, other orthoptera including mole cricket nymphs, cockroaches, beetles including fireflies and weevil rootborers which are pests of sugar cane and citrus orchards, squash bugs and assassin bugs, dragonflies, moths and caterpillars (Brassolidae, Noctuidae, Geometridae), butterflies (Pieridae), hymenoptera (euglossine bees), but few ants and spiders (Gosse 1847, Gundlach 1874, Wetmore 1916, Danforth 1925, Friedmann 1927, Stockton de Dod 1981, de Visscher and Moratorio 1984, Sick 1993, Haverschmidt and Mees 1994, Rosenberg et al. 1990, Magalhães 1999, Burger and Gochfeld 2001). Recent observations show that few ticks are taken (Wetmore 1916, 1968, Rand 1953), but in the 1800s before chemical pesticides were used on livestock, ticks were important in their diet (Gosse 1847). Anis plow into soft earth and cow dung to get insects, and they chase flying insects.They take tree snails, Anolis lizards, small snakes, frogs, mice, nestling birds and eggs (anis are mobbed by other birds: Wetmore 1916, 1927, Danforth 1925, Dubs 1982, Loflin 1983, de Visscher and Moratorio 1984); and they take small birds from mist-nets (Gill and Stokes 1971). In the dry season they take fruits, including snake-withe berries, fiddle-wood Cytharaxylon and wild grape Cissus sicyoides (Gosse 1847, Quinn and Startek-Foote 2000), Madame Jeanette hot peppers (Haverschmidt and Mees 1994) and seeds (Davis 1940a, Downer and Sutton 1990, Sick 1993).They eat the fruits of the gumbolimbo Bursera simaruba, removing the skin, swallowing the inner part and regurgitating the seed (Trainer and Will 1984).

Displays and breeding behavior Breeding adults are often socially monogamous; rarely polygynous, a male mating with two females, or polyandrous, a female copulating with two males when the earlier mate disappears. Anis copulate

both in the nest tree and in other trees. In mating, the male simply mounts the female. In copulating, the female remains stationary and raises her tail, then pulls her wings together in front while the male flaps his wings. Copulation lasts as long as 42 seconds. A pair copulates with either sex on top, males mounting females as well as females mounting males. Copulation occurs throughout the breeding cycle but is most common in the nestbuilding period. In courtship feeding, the male presents the female with a grasshopper or lizard, copulates and then she swallows the food, or the female accepts the food without copulating. Besides feeding the female during her fertile period, the male feeds the female while she incubates (Köster 1971, Loflin 1983, Quinn and Startek-Foote 2000). Social hierarchies are apparent in the nest, where one bird displaces another by pushing her bill underneath the sitting bird, and this has the effect of delaying the onset of incubation of the clutch until the last female has laid her eggs (J. S. Quinn, in litt.).

Breeding and life style In Florida they nest after the rains begin in May through October. Where rains are not seasonal in the West Indies, Suriname and French Guiana, the anis may breed all year round. In more seasonal areas of the Neotropics they breed during the rains, in SW Puerto Rico starting in September after the rains have begun, in Cuba from April to October, in Central America during the rains after the herbs have grown and grasshoppers are abundant, and in southern Brazil in January and February (Gundlach 1874, Wetmore 1916, Danforth 1925, Belcher and Smooker 1936, Hellebrekers 1942, Loflin 1983, Belton 1984, Tostain et al. 1992, Haverschmidt and Mees 1994, Raffaele et al. 1998). On Grand Cayman Island, where the wet season occurs from May to November, they breed throughout the year (Bradley 2000). Gregarious, group living with several pairs breeding together in a single nest.The nest is built and attended by a single pair or by 2–4 pairs with several females laying in a common nest. As many as nine adults may take part in building the nest. In Florida, anis nest either in pairs or in groups,

178 Groove-billed Ani Crotophaga sulcirostris whereas in other tropical areas, anis are communal and competitive (Gosse 1847, Eisentraut 1935, Bent 1940, Davis 1940a, 1942, Skutch 1966, Köster 1971, Reichholf 1974, Loflin 1983, Stiles and Skutch 1989, Sick 1993, Stevenson and Anderson 1994, Quinn and Startek-Foote 2000). The nest is built in a thorny tree, shrub or thicket, a large, bulky shallow mass of interlaced sticks. It is 17–26 cm in outside diameter, 13–19 cm inside, the inside depth being 5–9 cm (Quinn and Startek-Foote 2000). Green leaves are added each day before laying, during laying and during incubation. Eggs are greenish blue covered by a white chalky surface, 35 ⫻ 26 mm in size; weight 1.15 g (Hellebrekers 1942), and are quite variable within a population (34.5 to 37.2 ⫻ 26.0–28.6 mm; weight 10.8 to 13.0 g) (Loflin 1983). Eggs are laid at 2-day intervals. Clutch size averages 9.7, varying with the number of females at the nest, where nests with one laying female average 4.3 (range, 3–7), nests with two females 9.5 eggs, nests with 3 females 14.8 eggs, and with 4 females 21.3 eggs; as many as 36 eggs are laid in a nest. In nests where more than one female lays, the female covers with leaves the eggs of the female that laid earlier, before she lays her own.When the early females attempt to raise these eggs to the top layer, the later-laying females chase the early females from the nest and continue to bury the early eggs. As a result, often only the top layer of eggs hatches. Occasionally a female tosses eggs of an earlier female from the nest. In communal nests,

37% of eggs are lost when buried in the nest lining by competing females. Other sources of loss are infertility, predators and storms. Incubation begins soon after the first egg in single pair nests, but later in multiple-female nests. Competitive behavior occasionally leads to one adult killing another. Both sexes incubate. The incubation period is 13–14 days. Nestlings are likely to starve if latehatched—these may be the young of the last-laying female.The last-laying pair sometimes abandons the nest, and their young are reared by the early-laying pairs that rear their own brood as well.The nestling begs by stretching its head and neck and flapping its wings ( J. S. Quinn, in litt.). Nestling feathers begin to emerge from sheaths in five days.The young can grasp with their feet and crawl when they are disturbed; they scramble out of the nest and cling to branches as early as day 4 ( J. S. Quinn). Nestlings normally fledge at 10–17 days. Both sexes care for the young, as do juveniles that remain in their natal territory for as long as a year. As many as nine anis bring food to the nestlings (Davis 1940a), and both the adults and nestlings defecate over the side of the nest (Quinn and Startek-Foote 2000). Breeding adults give less care to broods with helpers, which give care to the brood. Group size has no effect on feeding rates or growth of the young. The young fledge at 35 g, less than half the weight of adults, and they complete most of their growth after they leave the nest. Most anis disperse from their natal group at 8 to 12 months of age (Oniki and Ricklefs 1981, Loflin 1982, 1983).

Groove-billed Ani Crotophaga sulcirostris Swainson, 1827 Crotophaga sulcirostris Swainson, 1827, Philosophical Magazine (n.s.) 1, p. 440. (Temascaltepec, Mexico) Monotypic.

Description ADULT: Sexes alike, plumage glossy black with a blue gloss due to the iridescent edge of the feathers on the head, nape and back, the feathers narrow and straight with the tips pointed, the skin under

the feathers also black, wing black, tail long, black; skin around eye black, iris brown to black; bill arched and laterally compressed, no hump at the base, the upper ridge continuous with crown, upper mandible with parallel grooves along the bill, extending to near the tip in older adults, the lower mandible nearly straight in profile; feet black. JUVENILE: Plumage dull black, soft and loosewebbed; cheeks bare at fledging, feathers develop

Groove-billed Ani Crotophaga sulcirostris 179 later, tail feathers narrower and more pointed (not truncate) than in adult; bill smaller than adult, without conspicuous grooves, but grooves are present.

hollow “pep, pep . . .” (Skutch 1959, Wetmore 1968, Vehrencamp et al. 1986, Stiles and Skutch 1989, Hardy et al. 1990).

NESTLING: Naked, skin black, no hair-like down when they hatch, but covered with dark downy contour feathers when they fledge; bill with grooves in the older nestlings; palate bright pink with white raised patches, one on each side of the palate fold or choana, white edge on underside of tongue and white line to tip of under tongue, tip of tongue black, edge of mouth whitish.

Range and status

SOURCES: AMNH, BMNH, CM, FMNH, MCZ, MVZ, RMNH, ROM, UMMZ, USNM.

Measurements and weights Texas: Wing, M (n ⫽ 12) 129–138 (134.7 ⫾ 3.5), F (n ⫽ 10) 127–135 (131.2 ⫾ 2.7); tail, M 168–184 (177.0 ⫾ 6.9), F 166–177 (171.9 ⫾ 4.0); bill, M 26–30 (29.0 ⫾ 1.3), F 26–29 (27.5 ⫾ 0.8); bill depth at nostrils, M 17.3–19.8 (18.9 ⫾ 0.9), F 16.6–19.1 (17.5 ⫾ 0.9), smaller in juveniles; tarsus, M 31–35 (33.2 ⫾ 1.4), F 31–34 (33.1 ⫾ 1.1) (UMMZ). Weight, Mexico and Panamá, M (n ⫽ 14) 61.1– 94.5 (79.6), F (n ⫽ 22) 55.5–79.8 (66.6) (UMMZ); Costa Rica, M (79), F (70) (Vehrencamp 1978). Wing formula, P6 ⬎ 7 ⬎ 5 ⬎ 8 ⬎ 9 ⫽ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 10.

Field characters Overall length 32 cm. Black birds in loose flocks; bill large and with grooves, arch not extending above the forehead in profile and the lower mandible rather straight; the blue gloss of head and nape uniform with the gloss of the back; the glossy feathers narrower than in Smooth-billed Ani C. ani, and the tail long, often pointed upwards when the birds are on the ground.

Voice Squeaky and chattering high-pitch sounds, “tee-ho tee-ho” in series, the first note slurring up, given in excitement or alarm; also a faster series, a whining, querulous “whee-o”, a long series of rapid whistled “kiw” slurred down at the end, a grating “krr krr . . .”, and a “conk” or sharp,

Gulf of Mexico from central and southern Texas and (once breeding) Louisiana, Mexico (southern Sonora south along both the Pacific and Caribbean slopes and near-coastal islands including Cozumel I and Ambergris Cay off eastern Yucatán Peninsula), south through Caribbean and Pacific Central America and western South America along the coastal lowlands to Ecuador, Peru, northern Chile and NW Argentina. In the southern Caribbean region they occur in northern South America from Colombia (mainly N Córdoba east to Guarija, south in drier parts of Magdalena Valley to N Huila) to Venezuela and Guyana, and on the Netherlands Antilles and Trinidad (Phelps and Phelps 1958, Snyder 1966, Meyer de Schauensee and Phelps 1978, Voous 1983, Araya Mödinger and Mille Holman 1986, de la Peña 1986, Hilty and Brown 1986, Stiles and Skutch 1989, AOU 1998, Ridgely and Greenfield 2001). Nonbreeding movements occur when birds disperse northward into continental North America, as far as panhandle and peninsular Florida, the southwestern states, California, the southern Great Plains and the Great Lakes region to southern Ontario and Manitoba, and an ani is accidental in the Revillagigedo Islands (Socorro) ( Johnson 1967, Sutton 1967, Oberholser 1974, AOU 1998, Kent 1988, Russell and Monson 1998, Mlodinow and Karlson 1999). Although anis are resident in most of their range, after the breeding season some appear as far as 1000 km from their nearest breeding areas. In central United States they appear in autumn from late September to November in Florida and Alabama they appear from September through April and May, and in the southwestern states they appear in all seasons, mainly September to November. In the nonbreeding season they live in flocks and roost in groups of as many as 30–40 birds. Many birds remain on their breeding territories and live in the same groups throughout the year, particularly in the better habitats where insects occur throughout the year; while birds in

180 Groove-billed Ani Crotophaga sulcirostris

xeric habitats such as pasturelands move to riparian areas. A few birds occur in the northern part of their range in winter in northern Mexico, the Gulf Coast of Texas and further east (Root 1988, Bowen 2002). Groove-billed Anis are common in much of their range, but are replaced by Smooth-billed Anis C. ani in the Pacific areas of Central and South America and in the West Indies.Territories of social groups occupy an area of 1–10 ha; birds in marshy areas have larger breeding groups and smaller territories than birds in pastures. Anis formerly lived on the cape and islands of southern Baja California (Grinnell 1928); the last breeding record was in 1896, and since about 1910 only one has been seen there, a vagrant (Howell and Webb 1995, AOU 1998). A subspecies Crotophaga sulcirostris pallidula described by Bangs and Penard (1921) as paler was recognized for birds on the west coast of Mexico in Sinaloa, Colima and Tehuantepec (Bangs 1930), but these birds are not distinct. Even the anis taken at Cape Baja in 1887 do not differ consistently from anis in other regions: some old specimens are brownish due to post-mortem change (van Rossem 1938, MVZ). On the other hand, historically the range has extended to other areas, with birds colonizing Curaçao then Bonaire and Aruba within the past century (Voous 1957).

to tropical scrub, less often in humid regions. Anis occur mainly in lowlands, and they range mainly from sea level to 500 m, occasionally in Ecuador and Peru to 2700 m (Lowery and Dalquest 1951, Wetmore 1968, Hilty and Brown 1986, Fjeldså and Krabbe 1990, Stiles and Skutch 1989, Stotz et al. 1996, AOU 1998, Ridgely and Greenfield 2001). Gregarious, group-living, weak fliers.They feed in groups, walking on dry ground, often in pasture with cattle or other livestock, where they catch insects more rapidly when in groups than when hunting alone (Dickey and Van Rossem 1938, Rand 1953, Skutch 1959, 1983, Smith 1971, Stiles and Skutch 1989). They take ticks from the legs and undersides of cattle, or did so in the days before the widespread use of cattle dips with pesticides. They dig in dung for grubs, dung-beetles and other insects (Voous 1957, 1983).They follow army ants into the forest and capture insects that are flushed out by the ants (Bent 1940, Willis 1983). Territory size of a group is large, 3–11 ha. The local density of birds is higher in marsh than in pasture (1.9 vs 0.4 birds / ha), even though the density of insect prey is about the same in these habitats. As many as four pairs live in a group and lay in a common nest. The number of birds in a group does not affect their territory size or the number of young that survive and fledge in their common nest. However, the number of birds that help rear the young may reduce the costs of parental care to the breeding adults (Vehrencamp 1978, Bowen et al. 1989).

Food Insects, mainly grasshoppers, also cockroaches, grubs, cicadas, flies, wasps, ants, termites; ticks, spiders, small vertebrates including Anolis lizards, and fruits, seeds and berries. Most foods are larger than 4 mm (Bent 1940, Dickey and Van Rossem 1938, Voous 1957, Wetmore 1968, Oberholser 1974, Vehrencamp 1978). The same foods are fed to the young anis in the nest (Bowen 2002).

Habitat and general habits

Displays and breeding behavior

Tropical lowlands in scrub, thickets, pastures, fields, marshes, cleared and disturbed areas in the range of original vegetation from tropical evergreen forest

A contact species like the other anis, the birds roost and huddle together in all kinds of weather. They preen each other: one bird stretches its neck

Groove-billed Ani Crotophaga sulcirostris 181 upward and the other bills and nibbles at the feathers, then the birds switch roles as the preener and the preened. In the early morning they spread their wings and sunbathe, a behavior that accounts for their name “zopilotillo,” a diminutive of “zolpilote” or vulture (Skutch 1959,Wetmore 1968). Agonistic displays are aggressive in nature: (1) Broadside threat, where the threatening bird lowers its head, presents its large bill in profile, drops its near wing and raises its far wing to enlarge its body profile, sometimes erecting its plumage as well, and gives a “conk” call. Males threaten another male that approaches their mate; females threaten other females when they interact at a nest; and adults threaten juveniles of the same sex during the dry season between breeding seasons when the young disperse from their natal territory. The response of the threatened bird is to sleek its plumage, lower its head, and turn or move away, or to solicit preening from the other bird. (2) Aerial chase, which is directed towards potential immigrants and territory invaders from outside the social group, and to rival males within the social group. (3) Supplanting behavior, which occurs near a nest and controls the access of other females to the nest. Anis in a group do not live in a strict dominance hierarchy, and a female’s status changes when alone and when her mate is nearby (Vehrencamp et al. 1986). Anis join a social group in different ways depending on the number of applicants to the group and the number of birds already in the resident social group. (1) A single bird attempts to join a group; it gives a call given only in that context, sometimes persisting in calling for days, while it is threatened and chased away by the dominant resident birds. When a bird repeatedly fails to enter a group, it moves to another group and tries again. A resident group admits a new bird when they have lost a member of the same sex. (2) A pair joins a group with less apparent control over the pair by the resident group; the new pair sometimes enters by laying in the nest of the resident pair, then remains and takes part in rearing the young. (3) A resident pair splits and each bird admits an immigrant of the opposite sex as its new mate (Vehrencamp et al. 1986). Anis breed as early as their yearling year (Vehrencamp 1978).

A nestbuilding pair is socially monogamous.The male and female spend time together, perch in contact and allopreen, call back and forth near the nest, and call over a long distance when one mate leaves the other at the nest. The pair remains together from year to year as long as both birds are in the social group (Vehrencamp et al. 1986). In colormarked birds, nearly all copulations took place between social mates. The few extra-pair copulations were between nesting attempts or were at the end of laying and were unlikely to lead to fertilization. During the breeding season and shortly before the eggs are laid, the male mounts the female. Males develop small testes while they incubate. Reversed mounting with the female on the male sometimes follows the loss of a nest. The female also mounts the male outside the breeding period; during this dry period all the matings observed of individually color-marked anis have been female-on-male (Bowen et al. 1991,Vehrencamp 1982b). In courtship feeding, one bird approaches the mate with an insect in its bill, and it holds its breast up and out. The bird approached lowers its body into a crouch. As the first mounts, the second spreads its wings; the bird on top flutters its wings and wags its tail; then the upper bird’s tail dips and cloacal contact takes place. Females take food offered by the male, but males do not take food when the courtship is initiated by the female (Bowen et al. 1991).

Breeding In Texas from May to August with young in the nest as late as September (Oberholser 1974), in the Cape region of Baja California in August and September (Brewster 1902), in Oaxaca from May to July (Rowley 1984, Binford 1989), in the Caribbean lowlands during the rains from June to early September; in Curaçao the same birds have apparently bred four times in succession from December to September (Voous 1983); in Costa Rica the birds breed in the rainy season of June to November, later in years when the rains are late (Vehrencamp et al. 1986, 1988, Stiles and Skutch 1989), in Colombia they breed in October and January (Hilty and Brown 1986), in Venezuela they breed in July and September to November (Thomas 1979).

182 Groove-billed Ani Crotophaga sulcirostris The nest is built in a marsh, thicket, shrub or tree, often a spiny citrus tree, 3–5 m above ground. It is a large, open bulky platform, measuring about 25 mm across and 10 mm deep, with an inner cup about 10 mm across and 6 mm deep, built of sticks, roots and thorns, and bits of vine and herbs, lined with green leaves which they continue to add during laying and incubation. Occasionally anis take over an old nest of another bird, such as the Boat-tailed Grackle Quiscalus mexicanus, and add a few sticks and leaves (Skutch 1959). A nest is built and by solitary pairs or by 2–4 cooperating pairs. Among these communal breeders, several females lay in a common nest, each female attended by her mate. Eggs are greenish blue covered with a white chalky surface, easily scratched; 31 ⫻ 24 mm (Bendire 1895, Wetmore 1968); 11 g (an egg is c. 17% of female body weight,Vehrencamp 1978). Eggs are laid at any time of day, often in the early afternoon. Clutch: a female lays 3–4 eggs at intervals of 2–3 days. More than one female may lay in a nest, which may have as many as 18 eggs. A female that has not yet begun to lay often tosses the eggs of other females out of the nest.When she has begun to lay she does not remove the other eggs in the nest, suggesting that females do not recognize their own eggs in a nest where more than one female has laid; rather, they change their behavior and avoid removing their own eggs.When a female from outside the social group lays, the group often abandons the nest.The incubation period is 12–14 days. An adult incubates for 30 min to 1 h at a time during the day; both male and female take about the same time. In solitary pairs the male incubates and later broods the young through the night. In communal nesting groups, one male incubates and broods the young at night. He is generally the oldest male in the group, the largest male, and the mate of the last-laying female, which differs from other females in the group in size (longer wing) and age (older) (Skutch 1959, Vehrencamp 1978, Vehrencamp et al. 1986). Nestlings develop rapidly and their eyes open two days after hatching. They are brooded for the first week; then their feathers break through the sheaths and the young look like feathered birds.

Both parents bring food to the brood. Together parents and helpers bring food at a rate of about six feeds per nestling in an hour.Two adults sometimes divide food between themselves at the nest, then each feeds the young. The nestlings beg in an upright posture when they are hungry, with neck stretched upward and wings stretched out and fluttered at the side (B. S. Bowen, in litt.). Nestling anis remain in the nest for 8–10 days, but in asynchronous broods the later-hatched birds are smaller and sometimes die of starvation. Parents remove nestling feces when the nestlings are young, and after the nestlings are two or three days old the young birds squirt their excrement over the side of the nest, which becomes marked with the droppings. Nestling feathers begin to emerge from sheaths in five days, when the young can grasp with their feet and crawl when they are disturbed (B. S. Bowen). Wing length develops independently of the number of young in the brood, or the number of adults in the breeding group (Vehrencamp 1978). After the young leave the nest, they can make short flights from branch to branch by day 11, but often spend 10 to 15 days near the ground. They normally fly by about day 17–20 and are tended by their parents until they are about 6 weeks of age.Young anis remain in their natal area for several months, and when they are about 10 weeks old they sometimes aid in incubation and feeding a later brood in their natal group (Miller 1932, Dickey and Van Rossem 1938, Skutch 1959, Vehrencamp 1977, 1978, 1983, Koford et al. 1986, 1990, Vehrencamp et al. 1986, 1988, Bowen et al. 1989, Stiles and Skutch 1989, Bowen 2002). Young anis that were marked in their natal site and survived until the end of the breeding season often remained in their natal area. Early in the next breeding season, 48 of 196 independent young were still in their natal area; then 25 dispersed within the study area and bred 1–3 territories away from their natal group, while another 15 bred in the group where they were born. Males are more likely than females to remain and breed in their natal group, and more males than females that breed in a group were born there. When a bird leaves the group, it disperses alone rather than with a nestmate (Bowen et al. 1989, Koford et al. 1990).

American Striped Cuckoo Tapera naevia 183 Survival of adults is 66–72% from one year to the next, although some local loss of adults may be due to dispersal rather than to mortality. Most known deaths occur during the breeding season when adults

are at risk on the nest.Anis are taken through the year at night roosts by their predators, Trimorphodon lyre snakes and Vampyrum carnivorous bats (Vehrencamp et al. 1977,Vehrencamp 1978, Bowen et al. 1989).

Neomorphinae Genus Tapera Thunberg, 1819 Tapera Thunberg, 1819, Handlingar Kungl Vetenskaps och vitterheetssamhället i Göteborg , 3, p. 1. New World Cuckoos with a streaked brown plumage, a crest, a long and broad alula, and a long tail with 10 narrow rectrices. Type, by monotypy, Tapera brasiliensis

Thunberg ⫽ Cuculus naevius Linnaeus (Peters 1940). Tapera is a Tupí (Brazilian) name matim tapirera for a cuckoo whose cries represent the voices of ghosts. One species.

American Striped Cuckoo Tapera naevia (Linnaeus, 1776) Cuculus naevius Linnaeus, 1766, Systema Naturae (ed. 12) 1, p. 170. Based on “Le Coucou tacheté de Cayenne” Brisson, Ornithologie 4: 127, pl. 9, fig. 1. (in Cayania ⫽ Cayenne) Other common names: Striped Cuckoo, Cuatro Alas, Quatro Asas,Tres Pesos, Four Winged Cuckoo Monotypic.

Description ADULT: Sexes alike, upperparts brown streaked buff and black, head with a striped blackish and rufous shaggy crest; the alula large and black, wing brown to rufous, white stripe across base of primaries seen in flight; upper tail coverts light rufous becoming dark brown with the loss of the pale feather edges; tail long and graduated; in fresh plumage the brown central vane with light rufous edges gives a sandy appearance but in worn plumage the loose-webbed edges are gone and the tail is dark brown; face with a white eyebrow and black whisker line; underparts whitish with narrow black malar streak, throat and breast with distinct black streaks, belly white; bare skin around

eye yellow, iris brown to greenish; bill brown to orange-brown or yellowish below; feet gray brown. Underparts fade with the season; fresh plumage in autumn is ochraceous, plumage in spring less so, and by midsummer the breast and flanks are a dirty gray. JUVENILE: Crown black with buff spots (not streaks), upperparts rufous brown streaked buff and black with buff spots, alula black with buff terminal spots, upper wing coverts brown with buff spots, rump and upper tail coverts brown with buff spots, tail brown with rufous edge and tips, underparts buff with fine black bars on throat and breast, belly white with small dark spots (not streaks). NESTLING: At hatching naked, skin pinkish; gape orange-yellow, bill with a sharp curved tip; within a week the skin turns darkish violet (Haverschmidt 1961). SOURCES: AMNH, BMNH, CM, FMNH, MCZ, MVZ, ROM, UMMZ, USNM.

184 American Striped Cuckoo Tapera naevia

Geographic variation Regional populations have been described as subspecies on the basis of size. Nevertheless, measurements of museum specimens show a considerable overlap in size between these regions. Tapera naevia excellens (Sclater, 1858) was recognized in Mexico and Central America to E Panamá, described as smaller in the bill and wing than birds in northern South America, but the size of birds in those regions overlap considerably and the form is not recognized here. A third form, Tapera naevia chochi (Vieillot, 1817), was recognized in southern Brazil south from Mato Grosso and São Paulo and in Paraguay and Argentina, and was said to be larger and different in plumage color (Bangs and Penard 1918, Cory 1919, Pinto 1947), but size and plumage color of birds in this region overlap extensively with those of birds in northern South America (Wetmore 1926, Hellmayr 1929, Traylor 1958, Sick 1993, and measurements) and the form is not recognized here.

Measurements and weights Mexico to Costa Rica: Wing, M (n ⫽ 13) 116–123 (118.8 ⫾ 3.2), F (n ⫽ 8) 107–121 (113.4 ⫾ 4.0); tail, M 171–190 (178.5 ⫾ 8.9), F 152–194 (169.9 ⫾ 13.1); bill, M 19.5–21.7 (20.4 ⫾ 0.6), F 17.1–20.5 (19.1 ⫾ 1.1); tarsus, M 30.8–35.9 (33.4 ⫾ 1.7), F 30.0–34.1 (31.8 ⫾ 1.5) (AMNH, UMMZ); Panamá: Wing, M (n ⫽ 10) 108.1–117.5 (112.4), F (n ⫽ 10) 104.0–112.3 (108.2); tail, M 148.0–165.0 (157.7), F 140.0–162.0 (146.2); bill, M 21.0–23.2 (21.8), F 20.2–22.6 (20.8); tarsus, M 32.3–36.1 (34.2), F 30.5–33.0 (32.0) (Wetmore 1968); Colombia, Venezuela, the Guineas, N Peru, Ecuador and Brazil to River Amazon: Wing, M (n ⫽ 13) 103–125 (110.3 ⫾ 6.4), F (n ⫽ 9) 101–109 (105.1 ⫾ 3.2); tail, M 143–169 (154.8 ⫾ 9.5), F 136–152 (148.8 ⫾ 7.8); bill, M 16.4–19.7 (18.1 ⫾ 0.9), F 15.3–18.7 (17.0 ⫾ 1.1); tarsus, M 25.5–34.4 (30.7 ⫾ 2.3), F 26.9–32.0 (29.5 ⫾ 1.9) (AMNH, UMMZ); Paraguay: Wing, M (n ⫽ 11) 110–118 (113.9 ⫾ 3.3), F (n ⫽ 4) 106–110 (108.3 ⫾ 1.7); tail, M 153– 173 (163.4 ⫾ 6.2), F 142–146 (144.0 ⫾ 2.3); bill, M 17.7–19.8 (18.7 ⫾ 1.0), F 15.2–17.4 (16.5 ⫾ 1.0);

tarsus, M 26.0–33.0 (30.3 ⫾ 2.3), F 26.7–30.9 (28.8 ⫾ 1.7) (AMNH, UMMZ). Weight, M (n ⫽ 8) 43–56.5 (49.4), F (n ⫽ 7) 41–59 (51.1) (AMNH, MVZ, UMMZ); Bolivia, M (n ⫽ 2) 40.0–47.7 (Schmidt et al. 1997). Wing formula, P7 ⬎ 6 ⬎ 8 ⫽ 5 ⬎ 4 ⬎ 9 ⫽ 3 ⬎ 2 ⬎ 1 ⬎ 10.

Field characters Overall length 26–29 cm. Long-tailed cuckoo with streaks, brown back striped black and a blackstreaked rufous crest. In flight the wing is white across the base of the primaries.When perched the bird is erect, raises and lowers its crest and swings its tail from side to side.

Voice The most common song is a mellow whistle, the second note half a tone higher than the first, often repeated,“sem-fim” or “fee-fee”, repeated or sometimes with a third note, each note at nearly 3 kHz and lasting 0.5–0.6 sec, and the song lasting nearly 1.5 sec. A second song has an ascending series of 4–7 short whistled notes,“tres pesos pide” or “peepee-pee-PEEdee”. The song rises through the first five notes from 2.6 to 3.0 kHz, the last note “dee” drops to 2.7 kHz; the song lasts about 2 sec. A third song is a soft variant of the second, with fewer notes (2–4), and other soft sounds are given. Songs in Central America, Colombia, Venezuela and southern South America are similar (Friedmann 1927, Slud 1964, Belton 1984, Hilty and Brown 1986, Stiles and Skutch 1989, Hardy et al. 1990, de la Peña and Rumboll 1998, Smith and Smith 2000). Regional differences in song as reported in Brazil (Sick 1953b, 1993) may be behavior variants as are known within an area and in a single bird (Hardy et al. 1990, Haverschmidt and Mees 1994, Smith and Smith 2000). The young cuckoo gives a short whistle like that of its host young (Morton and Farabaugh 1979).

Range and status Southern Mexico (Veracruz, Oaxaca, Tabasco, Chiapas and Quintana Roo), south through Central America along the Caribbean and Pacific slopes to eastern Panamá (AOU 1998), and in

American Striped Cuckoo Tapera naevia 185

South America from Colombia,Venezuela and the Guianas (Guyana, Suriname and French Guinea), western Ecuador (Esmeraldas south through El Oro and W Loja), northern Peru, and east of the Andes in Brazil (lower Rio Madera, Goais, Bahia, Mato Grosso and São Paulo) south to Bolivia, Paraguay, Uruguay and northern Argentina (Tucumán, Buenos Aires); and on Margarita and Trinidad (Chubb 1916, Wetmore 1926, Hellmayr 1929, Pinto 1938, Lowery and Dalquest 1951, Cuello and Gerzenstein 1962, Short 1975, 1976, Meyer de Schauensee and Phelps 1978, Hilty and Brown 1986, Remsen and Traylor 1989, ffrench 1991, Tostain et al. 1992, Narosky and di Giacomo 1993, Sick 1993, Haverschmidt and Mees 1994, Novaes and Cunha Lima 1998). Resident; in southern part of range their numbers vary with the season (Cuello and Gerzenstein 1962, de la Peña and Rumboll 1998, Magalhães 1999). The species range has expanded in the past 50 years as birds moved into cleared and scrub lands in response to deforestation in Central America and in Brazil (Mitchell 1957, Belton 1984, Stiles and Skutch 1989, Skutch 1999) and have appeared at high altitudes and into southeastern Ecuador in cleared habitat (Ridgely and Greenfield 2001). Common to uncommon, shy and solitary.

Habitat and general habits Shrubby clearings, scrub, low seasonally wet grassland, overgrown pastures, mangroves, river island scrub, open country with scattered trees, thickets and bushes, clearings in forested areas, and brush at

edge of forests.They are more common in tropical and subtropical regions than in arid regions. Birds occur from sea level in dense scrub on coastal plains to 800 and 1500 m, occasionally to 2300 and 2500 m in Venezuela and Ecuador (Hilty and Brown 1986, Fjeldså and Krabbe 1990, Tostain et al. 1992, Haverschmidt and Mees 1994, Howell and Webb 1995, Stotz et al. 1996, Ridgely and Greenfield 2001). Shy and skulking most of the time, they are conspicuous when they sing on post or wire, raising the crest and spreading the alula. They sand bathe and dust bathe on the ground. Solitary in feeding, they forage in vegetation and hop on the ground, sway the body side to side, fan the large alula, perhaps to flush prey; then move in a sudden rush after prey after flashing the alula. They stand on the ground or rest on a low perch, raising and lowering the long crown, stretching and closing the wings one at a time, extending and waving the alula, and shifting the body from side to side. They also flash the alula when disturbed or frightened, and in social interactions when they raise and lower the crest (Stiles and Skutch 1989, Haverschmidt and Mees 1994, Sick 1999).

Food Insects, notably grasshoppers, caterpillars, dragonflies, cockroaches and beetles, and spiders and snails (Pelzeln 1871, Friedmann 1927, Dickey and Van Rossem 1938, Belton 1984, Haverschmidt and Mees 1994, Skutch 1999).Young, like other parasitic cuckoos, take what the foster parents bring: insects including crickets, roaches, grasshoppers and caterpillars, and some are fed fruit (Loetscher 1952).

Displays and breeding behavior Territorial birds give loud “fee-fee” songs in long series, while birds that countersing with a neighbor or that respond to the playback of either “fee-fee” or “tres pesos pide” give the “tres pesos pide” song. Birds that appear excited in approaching a playback song raise and lower the crest, hunch the body, fan the tail, and hold the wings out and down with the dark alulae extended, and sway from side to side (Smith and Smith 2000).The conspicuous behavior of erecting the long black alulae and flipping these back and forth across the paler breast gives the

186 American Striped Cuckoo Tapera naevia bird its local name of “quatro asas” or “four wings” (Skutch 1999).

Breeding and life cycle In Oaxaca, southern Mexico, in June (Binford 1989), in Panamá the birds sing from January to June (Wetmore 1968), in Colombia at S Bolívar they breed in April (Hilty and Brown 1986), in Venezuela in September (Thomas 1979). In Suriname they breed nearly all year (Hellebrekers 1942, Haverschmidt and Mees 1994). In Trinidad they sing or breed nearly all year, mainly from June to August (Belcher and Smooker 1936, ffrench 1991) and in French Guiana early in the dry season (Tostain et al. 1992). Birds in Argentina breed from October to February, mainly in November and December (Friedmann 1927, 1933, Morgensen 1927, de la Peña 1995). Brood-parasitic. Hosts of the cuckoo are small birds with covered or domed nests, especially the furnariid ovenbirds. Twenty species, mainly suboscines, are known to be parasitized: Chotoy Spinetail Schoeniophylax phryganophila, Buff-browed (Azara’s) Spinetail Synallaxis azarae, Sooty-fronted Spinetail S. frontalis, Chicli Spinetail S. spixi, Palebreasted Spinetail S. albescens, Plain-crowned Spinetail S. guajanensis, Stripe-breasted Spinetail S. cinnamomea, Rufous-breasted Spinetail S. erythrothorax, Yellow-fronted Spinetail Certhiaxis cinnamomea, Short-billed Castanero Asthenes baeri, Red-eyed Thornbird Phacellodromus erythrophthalmus, Plain Thornbird P. rufifrons, Greater Thornbird P. ruber, Buff-fronted Foliage-gleaners Philydor rufus, Whiteheaded Marsh-Tyrant Arundinicola leucocephala, flycatchers Myiozetetes spp., Plain Wren Thryothorus modestus, Rufous-and-white Wren T. rufalbus, and Black-striped Sparrow Arremonops conirostris (Friedmann 1927, 1933, Belcher and Smooker 1936, Loetscher 1952, Wetmore 1968, Kiff and Williams 1978, Salvador 1982, Sick 1993, Haverschmidt and Mees 1994, de la Peña 1995, Skutch 1999). The female cuckoo enters the bulky nest of mud

or interlocked sticks through the narrow entrance tube, and it also enters nests of Buff-fronted Foliagegleaners in burrows in the ground or a cavity in a tree (Wetmore 1968, Skutch 1999). The female enters at dawn the nest to lay, when nesting parents are foraging.When female cuckoos open the ovenbird nest near the incubation chamber, the nesting birds repair the damage. The eggs are white or bluish white to bluish green in Central America (Skutch 1999), pale blue in Colombia (Carriker, in Hilty and Brown 1998), greenish, bluish white, or greenish-blue in Suriname and Brazil (Haverschmidt 1961, Haverschmidt and Mees 1994, Dubs 1992, Sick 1993), white in Argentina (Friedmann 1927), 21 ⫻ 16 mm (Schönwetter 1964), including a record that was earlier identified as Pavonine Cuckoo Dromococcyx pavoninus which is unknown in Suriname: the eggs were white as in T. naevia (Pelzeln and Pelzeln 1910, Hellebrekers 1942). Usually one, sometimes two cuckoo eggs are laid in a host nest. Incubation period is 15–16 days (host Synallaxis incubation is 17–18 days). The host young do not survive, and although eviction has not been seen, within 24 hours of hatching, the nestling cuckoo kills the host young with its sharp bill and the foster parent removes its dead young (Morgensen 1927, Friedmann 1927, 1933, Belcher and Smooker 1936, Sick 1953a, 1981, 1993, Haverschmidt 1955, 1961,Wetmore 1968, Morton and Farabaugh 1979, Salvador 1982, Stiles and Skutch 1989, de la Peña 1993, 1995, Skutch 1999).The nestling feathers out by day 10, fledges in 16–18 days but the young bird is flightless for another week while it remains with its foster parents. The young continues to accept food from its (human) foster parent to the age of 36 days (Morton and Farabaugh 1979). Young at fledging excrete a foul-smelling liquid; behavior of nestlings was not noted. Frequency of brood parasitism: in Suriname 14 of 21 nests of Yellowthroated Spinetail had a cuckoo egg (Haverschmidt 1961).

Pheasant Cuckoo Dromococcyx phasianellus 187

Genus Dromococcyx Wied, 1832 Dromococcyx Wied, 1832, Beitrage zur Naturgeschichte von Brasilien, 4(1), p. 351. New World cuckoos with brown plumage, a crest, a slender, straight bill, broad rectrices, and upper tail coverts that extend to the end of the tail.The breast muscles and sternum are

large, and the leg muscles are weak.Type, by monotypy, Macropus phasianellus Spix 1824. The genus name refers to the terrestrial running behavior (Gr. dromos, running; kokkus, cuckoo).Two species.

Pheasant Cuckoo Dromococcyx phasianellus (Spix, 1824) Macropus phasianellus Spix, 1824, Avium Species Novae, quas in itinere per Brasilium annias 1817–20 / collegit et descripsit . . . , 1, p. 53, pl. 42. (Tonantins, Brazil) Other common names: Peixe-frito-verdadeiro (Common Fried Fish),Yasi-yatére Grande Monotypic.

back and wing coverts brown with round buff spots on tips, secondaries with buff tips, tail feathers more pointed than adult and with no subterminal black band or white tips, face with a buff eye streak that extends to nape, the throat and breast rich buff with a black border on sides, the belly and under tail coverts white and unmarked; eye-ring dusky green, iris dark gray or grayish brown, feet gray.

Description ADULT: Sexes alike, the crown chestnut with a short chestnut crest, back blackish brown, the upper back and wing coverts with white margins, alula large, wing primaries and secondaries dark brown with white tip, rump blackish, upper tail coverts nearly as long as the tail, graduated in shape, dark brown, each covert ending with a subterminal black band and a white spot, the tail long and graduated, tail feathers dark brown with a subterminal black band and white tip of the outer feathers T2 to T5; face with a white eye-streak which extends to nape, cheek blackish brown; the chin whitish, the breast buffy white with fine blackish spots and streaks formed by dark rachis and dark terminal barbs, belly and under tail coverts white and unmarked, under wing coverts white, base of primaries white and a white bar across center of middle primaries; bare skin around eye yellow-green to blue-green, bare lores bluish green to dull greenish gray, iris brown to yellow, bill blackish above, below gray, feet grayish brown. JUVENILE: Crown black with fine buff spots in a short black crest, upperparts sooty brown, the upper

NESTLING: Undescribed. SOURCES: AMNH, ANSP, BMNH, CM, DMNH, FMNH, LSU, MCZ, MVZ, NMW, ROM, SMF, SMT, UMMZ, USNM, ZSM.

Subspecies None. Regional texts on birds in Mexico and Central America often recognize a subspecies D. p. rufigularis Lawrence, 1867. The form was described from a bird in juvenile plumage, and the type (AMNH 44460) was compared only with birds in adult plumage. Neither juveniles nor adults differ in size or in plumage between geographic regions.

Measurements and weights Wing, M (n ⫽ 22) 154–176 (166.2 ⫾ 5.9), F (n ⫽ 18) 153–173 (162.8 ⫾ 5.6); tail, M 173–242 (209.4 ⫾ 17.0), F 177–248 (203.1 ⫾ 19.0); bill, M 17.7–25.2 (20.4 ⫾ 2.0), F 17.5–23.6 (20.0 ⫾ 1.6); tarsus, M 27.0–35.7 (31.5 ⫾ 2.8), F 27.3–36.9 (30.9 ⫾ 3.2) (AMNH, ANSP, UMMZ). Weight, M (n ⫽ 4) 78–98 (85.0), F, laying (n ⫽ 1) 98.1 (AMNH, ANSP, UMMZ).

188 Pheasant Cuckoo Dromococcyx phasianellus Wing formula, P7 ⬎ 8 ⫽ 6 ⬎ 5 ⬎ 9 ⬎ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⫽ 10.

Field characters Overall length 36 cm. Long-tailed cuckoo of the forest floor, with a small head and thin neck, a long fan-shaped tail with white spots, a chestnut crest and a white line through the eye, white with dark streaks below. Juvenile has a blackish crown with fine spots, buff scallops on upperparts, and rich buff on the throat and breast. The tail feathers are broader than in the Pavonine Cuckoo (D. phasianellus, adult and juvenile width 28–34; D. pavoninus, width 20–24), and the tail tip is less marked with white.

Voice Clear whistles, “whee whee wheerr-rr”, the first notes a series of 2–4 successively higher pitched whistles rising from 1.8 to 2.0 kHz, the last note a tremulous trill, the song lasting about 1.5 sec. Other calls include a series of rattling, clucking or growling “grrr” notes (Dickey and Van Rossem 1938, Wetmore 1968,Willis and Eisenmann 1979, Belton 1984, Hilty and Brown 1986, Stiles and Skutch 1989, Hardy et al. 1990, Howell and Webb 1995, de la Peña and Rumboll 1998). The bird approaches when it hears a good imitation of its whistle calls.

Range and status Southern Mexico (Guerrero, Puebla, Oaxaca, Veracruz, Chiapas and the Yucatan Peninsula) through the Caribbean and Pacific slopes of Central America (mainly on Caribbean slope, in Belize, Guatemala, northern Honduras and Nicaragua, and eastern El Salvador and adjacent southeastern Honduras) and Panamá, and in South America east of the Andes from Colombia (middle Magdalena Valley, eastern base of Andes, in northeast in Arauca and Vichada (Río Meta) and in the extreme southeast in Amazona (Leticia), Venezuela (Zulia, Carabobo, Aragua, Guárico, Sucre, NE Bolívar), Brazil (Roraima, Para (Aramanaí), Amazonas (Rio Madeira and Borba), Maranhão (Tabocas), Piauí, Bahia (Barra, Bomfin), Goiás (Araguay), Minas Gerais, Mato Grosso (including Engenho do

Gama, Descalvado), Mato Grosso do Sul, Paraná, Rio de Janeiro, São Paulo and Rio Grande do Sul), Amazonian Ecuador (E Sucumbíos, E Napo, E Pastaza, E Morona Santiago) and eastern Peru (Ucayali, Rio Apurimac in Cuzco), to N Bolivia (Rio Beni, Rio Chapare), Paraguay (scarce in Chaco, present in Paraguarí, Ñeembucú and Cordillera) and NE Argentina (Misiones) (Pelzeln 1871, Ridgway 1916, Hellmayr 1929, Pinto 1938, Meyer de Schauensee and Phelps 1978, Hilty and Brown 1986, Contreras 1993, Hayes 1995, Howell and Webb 1995, Parker and Goerck 1997, AOU 1998, Cappeer et al. 2001, Ridgely and Greenfield 2001). In Ecuador it is seen near Kapawi Lodge along the Río Pastaza near the Peruvian border. Unknown in the Guianas. Resident. Widespread, uncommon to rare and local, secretive, solitary.The species was not seen after 1971 on Barro Colorado I, Panamá, where it was one of the first birds to disappear when the island lost its cuckoos and the possibility of immigrants due to deforestation on the nearby mainland (Willis and Eisenmann 1979).

Habitat and general habits Humid understory of young woodland, tropical lowland evergreen forest, flooded tropical evergreen forest, tropical deciduous forest, thickets and undergrowth, low-altitude cloud forest, forest borders and secondary woodland, lake margins with grassy edges, thick growth of vines, Cecropia and dying trees; mainly lowlands, occurring from sea level to 700 m in Ecuador), 1200 m in Panamá and 1600 m in Colombia (Wetmore 1968, Terborgh et al. 1984,

Pavonine Cuckoo Dromococcyx pavoninus 189 Hilty and Brown 1986, Sick 1993, Stotz et al. 1996, 1997, de la Peña and Rumboll 1998, Ridgely and Greenfield 2001). Solitary, secretive and skulking, they occur on the ground or near the ground, less often in the crowns of tall trees. Although the cuckoos are terrestrial, walking and wagging the tail, the tail and sternum are large and well muscled and the leg muscles are weak (Dickey and Van Rossem 1938, Wetmore 1968). The foraging bird bobs the body and makes rattling noises, vibrating the wing feathers and bill while it fans the tail downward and brushes the litter, then the bird pauses, rushes forward flicking its wings, and pecks in the litter (Sieving 1990).

Food Insects, including large grasshoppers and cicadas, beetles; small vertebrates including lizards (anoles, gekkos) and nestling birds (Pelzeln 1871, Wetmore 1968, Stiles and Skutch 1989).

Displays and breeding behavior Territorial, a singing adult raises head and crest, speckled breast feathers puffed out, alula and outer

primaries partly extended to show white spots, with upper tail coverts strongly arched. Two birds walk in parallel in display within 1 m of each other (Sieving 1990).

Breeding In Oaxaca, Mexico, from April to June (Binford 1989), in Bahia and Mato Grosso, Brazil, in November and December (Naumburg 1930, Wetmore 1968), and calling daily at dawn in Matogrosense, Paraguay, in September (Capper et al. 2001). Brood-parasitic. Hosts are small birds, mainly tyrannid flycatchers, that lay in either open or closed nests. In Mexico they use Yellow-olive Flycatcher Tolmomyias sulphurescens (Wilson 1992), in Brazil and Argentina the Eye-ringed Flatbill Rhynchocyclus brevirostris, Myiozetetes sp. and Pied Water-Tyrant Fluvicola pica and Barred Antshrike Thamnophilus doliatus (Sick 1993). Eggs are dull white or pale buff, with a wreath of rufous spots on the large end (Naumburg 1930, Sick 1953a,b, 1993, Friedmann 1964b, de la Peña 1986), 25 ⫻ 14.5 mm (oviduct egg, Naumburg 1930). Incubation and nestling periods are unknown.

Pavonine Cuckoo Dromococcyx pavoninus Pelzeln, 1870 Dromococcyx pavoninus Pelzeln, 1870, Ornithologie Brasiliens, 3, p. 270. (Araguay, Engenho do Gama and Arimani, Brazil). Other common names: Peacock Cuckoo, Yasiyatére Chico Monotypic.

Description ADULT: Sexes alike, crown and face rufous, short rufous crest, back dark brown with feathers edged in white, alula large, wing coverts edged whitish, wing dark gray brown, rump blackish, elongated and narrowly graduated black upper tail coverts with a small white spot at the tip, coverts cascading nearly the length of the tail, tail long and graduated, brown with broad white tip, the tail from below

gray with a subterminal black band and white tip, face with ear patch dark rufous and a deep buff eye streak extending to the nape; the throat and breast unspotted rufous, belly and under tail coverts white, under wing coverts white; orbital skin greenish gray with an inner ring yellow and black, iris brown, bill black above and gray below, feet gray to gray brown. JUVENILE: Crown dark brown, back dark brown without white edges of feathers, wing coverts and wing much as in the adult, upper tail coverts and tail long without terminal markings of white, cheek gray brown, eye streak light buff, chin white, throat gray, breast brown gray, belly whitish. NESTLING: Undescribed.

190 Pavonine Cuckoo Dromococcyx pavoninus SOURCES: AMNH, ANSP, BMNH, CM, DMNH, LSU, NMW, SMF, UMMZ, USNM, ZSM. A subspecies D. p. perijanus Aveledo H. and Ginés 1950 from the Upper Río Negro, Zulia,Venezuela, was described as dark on the crown. The holotype (Phelps collection no. 1172, in AMNH) is a juvenile with a few spotted upper tail feathers, the crown like the dark crown of juveniles from other areas.

Measurements and weights Wing, M (n ⫽ 13) 129–145 (134.4 ⫾ 4.7), F (n ⫽ 5) 126–142 (133.8 ⫾ 3.4); tail, M 132–174 (150.4 ⫾ 13.0), F 135–170 (152.8); bill, M 18.3–22.2 (19.6 ⫾ 1.4), F 18.4–19.7 (19.1); tarsus, M 26.6–33.2 (28.2 ⫾ 1.2), F 26.2–26.9 (26.5) (AMNH, ANSP, BMNH, CM, DMNH, LSU, USNM). Weight, M (n ⫽ 1) 50, F (n ⫽ ?) 45.2; U (n ⫽ 2) 48–49 (AMNH, Sick 1993, Magalhães 1999). Wing formula, P7 ⫽ 6 ⬎ 5 ⬎ 8 ⬎ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 9 ⬎ 10.

Field characters Overall length 28 cm. Long-tailed cuckoo of the forest floor, small head and thin neck, barred blackish back, long graduated tail with white spots, rufous crest and buff line through the eye, throat and breast unspotted rufous. Juvenile head gray-brown, eye streak white and back unmarked. Tail feathers narrower than in Pheasant Cuckoo (D. pavoninus, feather width 20–24; D. phasianellus, feather width 28–33).

Voice Song a high whistle “fee fee, feefee,” the first note lowest, second and fourth notes a half tone higher. Song is higher pitched (2.4–2.6 kHz), lasts about 2 sec, and lacks terminal tremolo that occurs in song of Pheasant Cuckoo (Hardy et al. 1990).They call before dawn (Boesman 1998).

Range and status South America east of the Andes from Colombia (Tierra Nueva, Sierra Negra SE of Fonseca 4000’: USNM 368717), Venezuela (subtropical western slope of the eastern Andes in NW Zulia, Aragua, S Bolívar, S Amazonas), Guyana (mainly near coastal

rivers: Abary River, Kartabo, Bartica, Georgetown, Annai River, Iwokrama Reverve on Essequibo River, Moraballi Creek, Ituribisi River, Supenaam River, Bonasika River; inland at Mt Roraima), French Guinea (Saut Tamanoir on Fleuve Mana), E Ecuador (Río Suno in W Napo, observed on the slopes of Cordillera del Cóndor), E Peru (Ucayali), Bolivia (La Paz, Santa Cruz, Cochabamba), Brazil (Amazonas (Manaus, Rio Branco, Rio Madeira), Maranhão, Pará (Obidos, Upper Rocaua), Goiás, Minas Gerais, Mato Grosso (Chapada), São Paulo (Botucatú, Itapura, Iguape, Albuquerque Lins), Paraná and Rio de Janeiro, Paraguay (mainly in the east) and NE Argentina (Misiones). Resident. Solitary, uncommon, their distribution is discontinuous; there is one record in Colombia and none in Suriname (Pelzeln 1871, Chubb 1916, Pinto 1938, Snyder 1966, Short 1975, Meyer de Schauensee and Phelps 1978, Remsen and Ridgely 1980, Dubs 1982, Remsen and Traylor 1983, Hilty and Brown 1986, Sick 1993, Haverschmidt and Mees 1994, Hayes 1995, Stotz et al. 1996, Cohn-Haft et al. 1997, Boesman 1998, de la Peña and Rumboll 1998, Kirwan and Sharpe 1999, Capper et al. 2001, Ridgely and Greenfield 2001, Salaman et al. 2001).

Habitat and general habits Understory of humid lowland tropical evergreen forest, montane evergreen forest, tangled thickets with dense secondary woodland, bamboo, heavy brush, transitional forest, seasonally inundated forest, occur with abundant Heliconia and Ficus, usually near water. Live in lowlands to 1600 m, occasionally to

Lesser Ground-cuckoo Morococcyx erythropygus 191 1900 m in Venezuela, in Ecuador c. 400 m (Terborgh et al. 1984, Hilty and Brown 1986, Haverschmidt and Mees 1994, de la Peña and Rumboll 1998, Ridgely and Greenfield 2001). Solitary. Forage on the ground and in forest understory.

Food Insects, mainly orthoptera, caught on the ground (Chubb 1916, LSU).

Breeding In Guyana a female cuckoo had a large ovum (7 mm) in July (ANSP 187796). Brood-parasitic. Hosts are several species of suboscine passerines with closed or bag-shaped nests (Tyrannidae:

Ochre-faced Tody-Flycatcher Todirostrum plumbeiceps, Eared Pygmy-tyrant Myiornis auricularis, todytyrants Hemitriccus spp; Thamnophilidae: Plain Antvireo Dysithamnus mentalis). It is unknown how the cuckoo gets its eggs into a small covered nest it cannot enter, such as the nest of a 6 g todyflycatcher: perhaps the cuckoo lays while it holds onto the nest outside the nest entrance (Sick 1993). Eggs are rose white with purplish spots, 21 ⫻ 15 mm (Giai 1949, Neunteufel 1951, Sick 1953a,b, 1993, Schönwetter 1964, de la Peña 1986, Haverschmidt and Mees 1994). Incubation and nestling periods are unknown. The host young disappear after the cuckoo hatches, a circumstance pointing to infanticide by the nestling cuckoo.

Genus Morococcyx Sclater, 1862 Morococcyx Sclater, 1862, Catalogue of Collection of American Birds, p. 322. Cuckoo with a short, decurved bill and long legs. Type, by monotypy, Coccyzus

erythropyga Lesson 1842.The name refers to the harlequin pattern of the head plumage (Gr. moros, clown, silly, foolish; kokkux, cuckoo). One species.

Lesser Ground-cuckoo Morococcyx erythropygus (Lesson, 1842) Coccyzus erythropyga Lesson, 1842, Revue et Magasin de Zoologie Pure et Appliquee (Paris) 5, p. 210. (San-Carlos, Centre Amérique ⫽ La Unión, El Salvador) Other common names: Rufous-rumped Cuckoo Polytypic. Two subspecies. Morococcyx erythropygus erythropygus (Lesson 1842); Morococcyx erythropygus mexicanus Ridgway 1915.

Description ADULT: Sexes alike, crown brown streaked with black and tipped buff, back grayish brown, wing brown with bronzy gloss, lower back and rump black with rufous tips, loose-webbed upper tail coverts rufous, long brown tail,T1 unmarked,T2 to T5 with subterminal black bar extending onto lateral vane and with buff tip, face with a buffy white superciliary streak which broadens to pale gray or

buffy gray over the ear; underparts, throat and breast rufous buff, lower belly and under tail coverts dark cinnamon; bare skin above and behind the eye blue (darker below the eye) bordered with black feathers, eye-ring and bare skin in front of eye yellow, iris brown, bill decurved, black above and yellow below, long legs yellow-brown. JUVENILE: Duller above, feathers scaled with pale buff edges and tips from crown to rump, rectrices more pointed (less rounded) than in adult, T1 unmarked brown, T2 to T5 brown without a subterminal black bar and with a small buff tip; underparts rufous smudged dusky; bare skin around eye gray, iris brown, bill above brown below lighter. NESTLING: Undescribed.

192 Lesser Ground-cuckoo Morococcyx erythropygus SOURCES: AMNH, ANSP, CM, FMNH, MCZ, MVZ, ROM, UMMZ, USNM.

Subspecies and geographic variation Morococcyx erythropygus mexicanus Ridgway, 1915; larger and paler, crown uniformly colored or indistinctly streaked, black tail marks indistinct; Pacific slope of tropical western Mexico from S Sinaloa south to Guerrero and Isthmus of Tehuantepec in Oaxaca and to Arriaga in SW Chiapas; Morococcyx erythropygus erythropygus (Lesson, 1842); smaller and darker, crown darker than the back and streaked with dark central shafts, black tail marks conspicuous when seen from below, tail tips buff or white; S Mexico in Chiapas east of the Isthmus of Tehuantepec, and Pacific Central America south to NW Costa Rica, and the arid interior valleys on Caribbean slope of Guatemala and Honduras. Plumage varies in color of the rufous underparts, the darker birds in El Salvador, paler eastward to Costa Rica; the gradient between rufous birds on the east and paler birds on the west side of the Isthmus of Tehuantepec is abrupt.

Measurements and weights M. e. mexicanus: wing, M (n ⫽ 9) 98–107 (101.3 ⫾ 2.7), F (n = 8) 96–105 (100.6 ⫾ 3.0); tail, M 123– 148 (135.9 ⫾ 8.2), F 125–143 (136.1 ⫾ 6.7); bill, M 21–24 (23.9 ⫾ 2.9), F 22–26 (23.7 ⫾ 1.6); tarsus, M 30–38 (34.6 ⫾ 2.8), F 30.5–39 (34.8 ⫾ 3.4) (AMNH, FMNH, UMMZ; M. e. erythropus: Guatemala to Costa Rica, wing, M (n ⫽ 22) 93–105 (98.7 ⫾ 3.6), F (n ⫽ 15) 90– 102 (96.9 ⫾ 3.8); tail, M 126–140 (130.2 ⫾ 6.7), F 120–142 (128.7 ⫾ 7.4); bill, M 19.4–25.2 (22.2 ⫾ 1.6), F 20–23.5 (21.8 ⫾ 1.0); tarsus, M 30.2–36.4 (33.6 ⫾ 1.8), F 31.4–36.0 (33.5 ⫾ 1.6) (FMNH, UMMZ). Weight, M. e. mexicanus: M (n ⫽ 5) 58–70.5 (63.1), F (n ⫽ 2) 62.9–86.8 (68.0); M. e. erythropus: M (n ⫽ 10) 53.1–66.2 (61.1), F (n ⫽ 7) 56–76 (65.3) (AMNH, MVZ, LSU, UMMZ). Wing formula, P7 ⫽ 6 ⬎ 8 ⫽ 5 ⬎ 4 ⬎ 3 ⬎ 2 ⫹ 1 ⬎ 9 ⬎ 10.

Field characters Overall length 25 cm. Greyish brown cuckoo with rufous buff underparts, blue skin around the eye bordered with black, and a pale mark over the ear.

Centropus rectunguis Wing, M (n ⫽ 4) 156–170 (165.5), F (n ⫽ 2) 166– 186 (176); tail, M 192–204 (198), F 194–238 (216); bill, M 35–37 (36), F 34–42 (37); tarsus, M 42–46 (44.3), F 45–52 (48.5); hallux claw, M 8–12 (10.0), F 9–19 (14) (AMNH, BMNH, ZRC).

Voice A loud mellow curlew-like whistle “teeeee,” flat at first then rising in the second half of the call and ending with a drop, the whistle mainly around 1.4 kHz and rising to 2 kHz, and lasting 0.6 sec. Variations include a clear trilled whistled series with notes lower and more widely spaced, the series introduced by 2–3 clear ascending whistles. Second call type is a soft burry whistle “whirrr” of four rising notes, the second two notes louder and higher-pitched than the first and last notes, the phrase lasting 0.3 sec. Female clacks the mandibles when disturbed (Stiles and Skutch 1989, Hardy et al. 1990, Howell and Webb 1995).

Range and status Southern Mexico and Central America. On the Pacific slope of western Mexico they occur from southern Sinaloa south to interior Mexico in the Balsas River region in Michoacan and Guerrero,

Greater Roadrunner Geococcyx californianus 193 and along the coast in Oaxaca and Chiapas, and in western Central America from Guatemala, El Salvador and Nicaragua to northwestern Costa Rica (Río Grande de Tárcoles).They also occur in the arid interior valleys on the Caribbean slope of Guatemala (Motagua) and Honduras (Quimistán, Sula, Comayagua, Aguán) where the population range is continuous with birds of the Pacific slope (Ridgway 1916, Hellmayr 1929, Howell and Webb 1995, AOU 1998). Resident. Fairly common, and locally common in Guanacaste, Costa Rica.

furtive and skulking, yet inquisitive and not shy (Stiles and Skutch 1989). Usually alone or in pairs, they behave like miniature roadrunners in terrestrial behavior, but walk slowly and deliberately, and freeze in position when alarmed (Howell and Webb 1995).

Food Insects, mainly grasshoppers (Dickey and Van Rossem 1938).

Breeding and life cycle Habitat and general habits Arid lowland scrub, tropical deciduous forest edge in semi-arid scrub and woodland, thorny thickets on arid slopes, agave plantations; sea level to 1500 m, locally to 1800 m in Guatemala (Stotz et al. 1996). Terrestrial, the birds forage in understory and near the edge of cultivation, and walk or hop on bare ground and among leaves on the ground with head and leg movements like a chicken, pecking food from the ground (Rowley 1984). They leap into foliage and walk along branches much like a dove (Berger 1960).They are

In Oaxaca, Mexico, they breed in the rainy season, from May to November (Rowley 1984, Binford 1989, UMMZ), in Costa Rica from February to May (Stiles and Skutch 1989). Nest is a shallow bowl of sticks and leaf petioles and rachises, lined with dead leaves, built on the ground. Eggs are chalky or smooth white, 27 ⫻ 21 mm, clutch 2 (1–3) (Skutch 1966, Rowley 1984, Howell and Webb 1995). Both sexes incubate, each taking its turn for 2–4 hours at a time before the mate takes over (Skutch 1966, 1983). Incubation and nestling periods are unknown.

Genus Geococcyx Wagler, 1831 Geococcyx Wagler, 1831, Isis von Oken, col. 524. Large, slender terrestrial cuckoos with streaked plumage, slender build, a crest, long legs and a long tail, living in arid and semi-arid regions of North and Central

America. Type, by monotypy, Geococcyx variegata Wagler ⫽ Saurothera californiana Lesson (Peters 1940). The name refers to the terrestrial habits of the bird (Gr. geo, earth; kokkux, the cuckoo).Two species.

Greater Roadrunner Geococcyx californianus (Lesson, 1829) Saurothera Californiana (Lesson, 1829), Oeuvres Complètes de Buffon, 6, p. 420. (Californie ⫽ San Diego, California). Monotypic.

Description ADULT: Sexes alike, above the crown and nape black with whitish spots and streaks, shaggy blueblack crest, streaked brown with bronze gloss,

upper back blackish brown with buff edges of the feathers forming whitish streaks, wing coverts blackish brown with buff and white, wing dark brown with two white stripes across the primaries, tip of outer vane of primaries edged white, lower back and rump brownish gray, rectrices broad and truncate, tail black with white outer edge on rectrices T1 to T5 and with broad white oval tips on T3 to T5; underparts whitish with brown streaks

194 Greater Roadrunner Geococcyx californianus and a black shaft, chin whitish, flanks whitish to whitish gray, under tail coverts light gray; bare skin behind eye blue and orange (behind eye males are usually white, females blue), the colorful skin normally concealed by feathers is exposed in display, eye-ring blue, iris brown with yellow ring around the pupil, bill blackish, tarsi brown to gray. JUVENILE: Plumage without gloss, feather tips notably the outer primary coverts and the rectrices more narrow and tapered than in adults, irregular edge of white patch on rectrices T3 to T5; postorbital skin as in adult female, white patches on tips of outer primary coverts have inverted U shape (less concave in adult); inner iris with a gray ring around pupil, outer iris brown. NESTLING: Nearly naked, skin black (throat whitish) with short (1–7 mm) whitish hair-like down; gape flange pink, mouth red with upper palate white, a raised area of white papillae on each side of palate and on the rear edge of the tongue, and tongue with a black tip, iris dark grayish brown without yellow ring; legs and feet are black at hatching and gray by 11 days. SOURCES: AMNH, CM, CMNH, FMNH, LSU, MCZ, MVZ, ROM, UMMZ, USNM.

Measurements and weights Arizona, New Mexico, Oklahoma and Texas south to Mexico (Sonora and Tamaulipas to Michoacan): Wing, M (n ⫽ 16) 163–190 (178.3 ⫾ 8.3), F (n ⫽ 12) 158–186 (170.7 ⫾ 9.0); tail, M 256–302 (283.6 ⫾ 14.3), F 260–308 (275.5 ⫾ 14.4); bill, M 43–53 (48.1 ⫾ 3.2), F 41–51 (45.5 ⫾ 3.2); tarsus, M 56–66 (61.4 ⫾ 2.7), F 56–64 (59.5 ⫾ 2.8) (UMMZ); California, Arizona and New Mexico: wing, M (n ⫽ 21) (185.5 ⫾ 18.4), F (n ⫽ 11) (171.6 ⫾ 10.0); tail, M (301.6 ⫾ 18.3), F (282.1 ⫾ 17.0); bill, M (36.7 ⫾ 2.3), F (35.4 ⫾ 1.2); tarsus, M (61.3 ⫾ 1.8), F (59.8 ⫾ 1.6) (Hughes 1996b). Weight, California: M (n ⫽ 8) 275–430 (344.0), F (n ⫽ 6) 264–352 (308.5) (MVZ); Texas, M (n ⫽ 6) 187–333 (270.0) (UMMZ); Oklahoma: early winter (n ⫽ 5) 412–588 (465), midwinter (n ⫽ 3) 300–325 (314.7) (Geluso 1969, 1970a, Sutton 1973).

Wing formula, P7 ⫽ 6 ⫽ 5 ⬎ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10. Oberholser (1974) noted that birds from east and central Texas south through eastern Mexico to Veracruz are smaller on average, and described a subspecies G. c. dromicus from Brownsville, Texas. These birds overlap in size with birds elsewhere in the species range (Browning 1978, Hughes 1996b).

History The first recorded description of the roadrunner was by a Franciscan priest in California in 1790, as reported by Coues (1900). “The Churcha is a kind of pheasant which has a long bill, dark plumage, a handsome tail and four feet. It has these latter facing outward in such fashion that when it runs it leaves the track of two feet going forward and two going backward.” Apparently “feet” replaced “toes” in a translation of this account.

Field characters Overall length 56 cm. Large terrestrial cuckoo with long legs and long tail, cocked up or hanging or held straight behind as bird runs over the ground. Greater Roadrunner differs from Lesser Roadrunner G. velox in having a breast more uniformly streaked with brown (not boldly streaked black), a lack of buff on the sides, a gray (not chestnut) back and rump, and pale under tail coverts.The colors of the bare skin behind the eye are exposed when the crest is raised. (Plate 2), at other times the full colors are not seen (Figure 4.10)

Voice Mournful dove-like “coo” notes given in long series “co-coo-coo-coo-cooooooo” at low frequency, slow and resonant, the notes dropping in pitch, the first notes about 0.6 kHz and the last notes 0.5 kHz, in a series the first note short and the later notes longer, the song lasting 3 or 4 sec, heard to about 250 m and given by the male; sometimes gives a short coo or paired coo notes in a soft two-note series “cooocooo.” Other calls are growl-coos, “whir” calls, whines, and bill clacks and rattles “trrrt”. Another sound is produced by the wings pulled inward towards the body in a series of four to eight loud

Greater Roadrunner Geococcyx californianus 195 counties of California with increasing agricultural and urban development and human disturbance (Ridgway 1916, Grinnell and Miller 1944, Emlen 1974, Garrett and Dunn 1981, Tweit and Tweit 1986, McCaskie et al. 1988, Hughes 1996b). Roadrunner population numbers showed no significant change over most of the range of the species in the period 1966–1993 (Price et al. 1995).

Habitat and general habits

“pop” sounds. Females and males often engage in mutual calling.The female gives a bark, and the male responds with a “growl-coo”, often while she is on the nest during egg-laying and incubation.The male also gives a “growl-coo” in response to his female’s “bark” while they are foraging apart from each other (Whitson 1971, 1975, 1983, Hardy et al. 1990, Hughes 1996b, K. Groschupf ).

Range and status North America, in southwestern United States from the Central Valley of California, extreme S Nevada, southern Utah, Kansas and Oklahoma south to the Gulf coast of Texas and east to W and NE Louisiana, Arkansas and extreme S Missouri, and in Mexico in Baja California, Sonora, Sinaloa, Durango, Zacatecas, Jalisco, eastern Michoacán, Puebla, Veracruz and northern Tamaulipas (Ridgway 1916, Grinnell and Miller 1944, Oberholser 1974, Howell and Webb 1995, Hughes 1996b, AOU 1998). Resident. Common to fairly common, roadrunners are most numerous in SE California, southern Arizona (the Chihuahuan desert) and Texas west of Pecos River and south of Edwards Plateau (Price et al. 1995). Their range expanded north and eastwards through the twentieth century in California, Kansas and Oklahoma and across the Ozark Mts into the Mississippi River basin (James and Neal 1986, Hughes 1996b). Land clearance, overgrazing and the invasion of shrubs may have caused eastern range expansions. Their range has decreased in regions where their native habitat was developed for agricultural and residential use. Roadrunners have disappeared in the Central Valley and southern

Semi-arid and arid lowland scrub, desert scrub, arid montane scrub, widely dispersed in dry open country with scattered brush, mesquite, palo verde, creosote, cholla cactus, prickly-pear cactus, upper oak and piñon pine scrub, in yucca and shortgrass plains of north Texas panhandle, also in tall pines and magnolias in east Texas, clearings in farms and dry scrubby woods ( James and Neal 1986, Stotz et al. 1996). They occur mainly in arid and semi-arid regions that have at least 140 clear sunny days. The northern extent of the range is limited by prolonged snow cover (Beal 1981). In Mexico, roadrunners occur from the lowlands to as high as 2900 m in the Cerro San Andres, Michoacán (UMMZ). Greater Roadrunners, cuckoos of arid scrub and desert, are the best known of any cuckoos for their morphological and physiological adaptations to their habitat. They have dorsal skin with large black patches of melanin, a high body weight, a salt-secreting nasal gland and an ability to reabsorb water through the cloaca (Calder and Bentley 1967, Ohmart et al. 1970a,b, Ohmart 1972, Dunson et al. 1976). They also have an ability to cool the body by evaporating water through the respiratory system and the skin (Lasiewski et al. 1971), to lower the body temperature at night, and to use solar radiation by basking in the early morning, exposing the unfeathered and pigmented black dorsal skin which absorbs the sun’s radiant heat (Ohmart and Lasiewski 1971, Whitson 1983). In cool times the birds sunbathe in early morning by spreading the back feathers, drooping the wings and exposing the black dorsal skin to the sun. Sunning behavior reduces the energy cost to raise its body temperature from the night-time level; the savings can be as much as 60% (Ohmart 1989). Roadrunners have fat deposits under their skin in winter, and some

196 Greater Roadrunner Geococcyx californianus males also have fat deposits under the skin when they breed (Vehrencamp 1982a). Metabolic rate is about the same as that of other birds of the same body size. At air temperatures above 36°C their metabolic rate increases and at 44.3°C it is 31% above the standard metabolic rate. Roadrunners maintain their basal metabolic rate at air temperatures as low as 9°C when they bask in the sun, but without solar radiation their metabolism begins to rise if air temperature falls below 27°C, and metabolic rate increases with lower temperature below this level (Calder and SchmidtNielsen 1967). During daytime conditions of extreme heat the adult roadrunners crouch, spread the wings and lift the upper body feathers, allowing air to flow between the layers without exposing the down feathers to the sun (Meinzer 1993).They also remove heat when they flutter their gular region, pant, and extend the wings to expose lightly feathered areas under the wing (Calder 1968a), and their heat exchange in arteries and veins cools the brain (Kilgore et al. 1976). By controlling their behavior the roadrunners reduce their food demands. Nestling roadrunners warm themselves by moving into a patch of sunlight in the nest; the skin (except the chin and vent) is black and absorbs solar radiation (Ohmart and Lasiewski 1971), and they cool themselves by fluttering their gular skin and aligning their bodies to the shade. Water lost in gular flutter and cutaneous evaporation is balanced by salt-secretion of the nasal glands and by the parents regurgitation of a clear liquid (water?) to the nestlings. Adults conserve water by reabsorption through the lining of the rectum, ceca and cloaca. Roadrunners excrete salt through the nasal glands in front of the eye; the glands are more prominent in nestlings than in adults. Roadrunners can maintain their body weight with water from their diet of reptiles and rodents, but they drink when water is available. Breeding adults eat their nestlings’ urine and feces, which are enclosed in a fecal sac.The nestlings’ excreta contains more water than adults’ excreta, and adults meet part of their daily water requirements through parental care (Ohmart 1973). In addition, adult roadrunners rest inactive in the shade during

the hot mid-day, and hunt in the early morning and late afternoon when the sun is low and their insect and lizard food is active. Resident, occur alone or in pairs. Behavior has been watched by following wild or tame handreared roadrunners on their daily rounds in the field (Sutton 1915, 1922, Marshall 1957, Whitson 1971, 1983). Roadrunners run in the open or under cover on bare ground, as they cover a large foraging area quickly and secretively. They run faster than a human, as fast as 30 km / h (Kavanaugh and Ramos 1970, Ohmart 1989).They feed in the cool hours of the morning and late afternoon, and rest in shade in mid-day heat (Calder 1968a). Occasionally they fly to a tall perch to sing or to a nest, and they ascend trees almost vertically by gaining momentum from running down another branch. In a tree they climb and scramble where the vegetation allows, moving by leaps and short flights, run along the horizontal branches, and glide down slopes just over the scrub. At night they roost high in vegetation. Terrestrial, they run after prey, leap at flying insects, ambush sparrows, wait at nectar feeders and take insects and hummingbirds. Roadrunners toss and batter lizards and snakes against a stone, and they attack scorpions at their tail (Sutton 1913, Calder 1968c, Mayhew 1971, Beal and Gillam 1979).A pair sometimes cooperate in attacking a snake (Whitson 1983, Meinzer 1993). Territory size averages 0.5 km2. Population density on rocky slopes in southern California is 0.65 birds / km2 and is lower in other desert habitats ( Weathers 1983); in coastal south Texas it is estimated at 2.5–3.1 birds / km 2 (Folse and Arnold 1976, 1978). Adult annual survival is at least 60% (Folse 1974). Roadrunners breed in their first year (Smith 1981).There are few banding recoveries.

Food Opportunistic carnivore: insects (grasshoppers, crickets, cicadas, caterpillars, beetles), scorpions, centipedes, spiders, tarantulas, toads, lizards, small snakes including rattlesnakes, small birds and eggs, mice, roadside carrion, young ground squirrels,

Greater Roadrunner Geococcyx californianus 197 young rabbits and young bats fallen from cave ceiling or colliding with the cave wall (Bryant 1916, Gorsuch 1932, Sutton, in Bent 1940, Herreid 1960, Zimmerman 1970, Shetlar 1971, Oberholser 1974, Bleich 1975, Binford 1979, Laycock 1985, Hughes 1996b). Uses bill to turn over dried mud crust on fields, and feed on the exposed crickets ( Jaeger 1947). Fruits and seeds are taken as well (prickly pear cactus Opuntia, sumac Rhus integrifolia) (Bendire 1895, Hughes 1996b). In winter, feeds on insects and grain (Geluso 1970b, Sutton 1973).

Displays and breeding behavior A pair remains on its territory all year and from year to year—one pair remained in the same site over five years. Sometimes a pair move together to a new territory and renest, while other birds move independently of the mate (Folse 1974). The pair spends time foraging together, calling back and forth, before the birds show sexual interest in each other. In courtship, one bird runs after the other on the ground, often for hours; both birds stop and rest between these chases.The chaser runs and lunges at the other bird with its wings and tail raised and fanned, while both birds give “clack” calls.The male gives “coo” calls from an elevated perch. Both sexes pick up sticks and pass them to the mate or drop them at the mate’s feet. Before they copulate the roadrunners display with a “prance display.” The male approaches the female in short, quick burst of speed, holds food in his bill, and wags his tail back and forth. He runs towards and away from the mate with his wings and tail lifted, then lowers the wings and brings them inward with a “pop.” In display he holds his tail over his body and then gradually lowers the tail, exposes the postorbital bare skin, erects the crest, and sleeks the contour feathers, and lifts the wings; the behavior lasts more than two minutes. In a “tail-wag display” the male wags the tail from side to side, while he bows then slowly lifts his head, faces the female and gives a “whirr” call. He usually holds food or plant material in his bill. After this he jumps into the air and leaps over the female or mounts her from behind. In courtship feeding, the male gives mice, small birds, snakes and lizards to his mate. He holds the food in his bill and presents it to the

female as they copulate. Then he dismounts, the pair walk away from each other, flicking the tail, and the female eats the offered food, or she feeds it to her young if copulation occurs after the young have hatched (Rand 1941b, Calder 1967a,Whitson 1971, 1975, Hughes 1996b). Roadrunners have bred successfully in captivity (Muller 1971,Whitson 1971, 1975, Smith 1981).

Breeding and life cycle In southern California in low elevations of the Lower Sonoran Life Zone the birds begin nesting in late February, and in the upland deserts of the Upper Sonoran Life Zone they nest a month later (Miller and Stebbins 1964). In southern Arizona, a few nest in late March, more nest from mid-April to mid-June, and birds nest again from late July to midSeptember with a pause in the hot dry summer. Nesting after the summer rains varies with the rains of the year (Ohmart 1973, 1989). In Texas they breed from March to October (Van Tyne and Sutton 1937, Whitson 1971, Oberholser 1974, Hughes 1996b), in Oklahoma from April to July or August (Sutton 1967, Baumgartner and Baumgartner 1992), in Sonora from May to July (Russell and Monson 1998). The nest is built 1–3 m above ground in a bush, low tree, thicket, or cactus clump, an open bulky flat platform of sticks, about 30 cm wide, often placed in shade, lined with leaves, snakeskins, cattle dung and mesquite seed pod debris. Birds continue to add material to the nest during incubation and after hatching. Occasionally they use an old nest or even an active nest of other species. Eggs are white, the inner layer of eggshell dull white, outer layer glossy sometimes with a yellowish film, 39 ⫻ 30 mm (Bent 1940). The eggs are laid at intervals of 2 days, the timing is variable. Clutch is 3–6, with the larger sizes after summer rains, and a nest may have as many as 12 eggs, apparently laid by two females (Bendire 1895, Pemberton 1925, Johnston 1964). Incubation period is 17–18 days (Bendire 1895, Sutton, in Bent 1940). Incubation starts when the second egg is laid, and early incubation before all eggs are laid results in asynchronous hatching and age differences of the nestlings in a brood. Both parents incubate, the male at night, both male and female in the daytime,

198 Lesser Roadrunner Geococcyx velox the female in one bout in the morning and one in the afternoon (Vehrencamp 1982). Both parents care for the young. Nestlings give a loud vocal “churr,” rattle the bill when disturbed and excrete a blackish foulsmelling liquid. Parents eat their nestlings’ fecal sacs and gain water this way (Calder 1968b).The parents feed the young nestlings with insects, and older nestlings are fed with lizards and small snakes while the adults feed themselves on insects. The weaker nestlings are eaten by the parent or are fed to stronger nestlings. Nestlings are 14 g at hatching, 50% of fledging weight by day 7, and fledge at 150 g, less than half the weight of the adults (Ohmart 1973, Folse 1974, Smith 1981).The young fledge in 17–19 days (Ohmart 1973). Natal down is lost on the day of fledging. The red and white mouth colors persist until day 50–55 when the mouth becomes marked with black, by day 80–85 the gape is black; the bare postorbital skin is orange

by 14 days (bluish color behind eye develops later). By 60 days the size and appearance of the young bird are similar to those of the adult (Muller 1971, Meinzer 1993). They forage actively with their parents when they are 70% of adult size, they follow their parents to feeding areas, and they catch their own food within 2–3 weeks (Sutton 1940). A parent sometimes flops a wing at its side and appears injured, a behavior that might draw a predator from the nest (Pemberton 1916).A pair of roadrunners in California renested a month after their first nesting, when the male took over care of the fledglings and the female laid a second set of eggs (Woods 1960). Nesting success: in Texas 73% of nests were taken by a predator, 22.5% of eggs survived to fledge (Folse and Arnold 1978), and a pair that nested twice could produce on average 3.5 chicks in a season. In Oklahoma and New Mexico, 66% of eggs laid hatched, 87% of nestlings fledged, and overall nest success was 72% (Maxon, in Hughes 1996b).

Lesser Roadrunner Geococcyx velox (Wagler, 1836) Cuculus velox A. Wagler, 1836, Gelehrte Anzeicher, München , 3, col. 96. (Mexico ⫽ outskirts of Mexico City) Monotypic.

Description ADULT: Sexes alike, crest and long tail, above the crown and nape black with white spots, short black crest with white spots, back chestnut brown with narrow white streaks, wing coverts chestnut brown with narrow white streaks, wing blackish with two white stripes across the primaries and the primaries with white tips, lower back and rump chestnut brown, tail long, blackish, with white tips on the outer rectrices T3–5 and white outline on the inner rectrices, underparts whitish buff, chin and upper throat white, sides of foreneck and breast whitish buff boldly streaked blackish, middle of neck and breast unstreaked, flanks and belly buff, under tail coverts sooty gray; bare skin behind eye blue to purplish, extending back to red patch on the neck, iris yellow to brown with ring of silvery white

around the pupil, bill above brown, darker on culmen, lower bill gray, tarsi gray to gray green. JUVENILE: Above buff tips to feathers, unstreaked, breast with dark spots not streaks, tail feathers narrower than in adult, bare skin around eye blue as in adult. NESTLING: Undescribed. SOURCES: AMNH, CM, CMNH, FMNH, MCZ, MVZ, ROM, UMMZ, USNM.

Measurements and weights Wing, M (n ⫽ 11) 133–154 (145.9 ⫾ 6.1), F (n ⫽ 9) 132–156 (142.1 ⫾ 8.5); tail, M 253–288 (271.8 ⫾ 11.6), F 247–280 (262.2 ⫾ 13.9); bill, M 34–40 (37.5 ⫾ 1.9), F 33–39 (36.9 ⫾ 2.3); tarsus, M 46–52 (48.7 ⫾ 1.8), F 43–50 (46.9 ⫾ 2.1) (UMMZ); Weight, M (n ⫽ 5) 174–203 (186.0), F (n ⫽ 4) 162.8–192 (174.0) (MVZ, UMMZ, USNM). Wing formula, P7 ⫽ 6 ⫽ 5 ⬎ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10.

Lesser Roadrunner Geococcyx velox 199

History and variation The birds of Mexico were first collected by Europeans shortly after Mexican independence in 1821. One of the first collections was made by Fernando Deppe, a gardener and naturalist at Berlin University who left for Mexico in 1824.A roadrunner thought to be this species appears as a manuscript name “Geococcyx viaticus” in his specimen price list which was prepared in 1830 by his brother Wilhelm Deppe. F. Deppe sent specimens of these birds with a nest and two eggs to the Zoological Museum at Berlin, where Lichtenstein adopted the names of Deppe in his own collection register but did not publish or describe them (Stresemann 1954). In the Munich Museum, Wagler found another specimen taken by Deppe, and he described the species velox from Mexico and distinguished it from the G. californianus. Plumages vary with the season. Birds in fresh plumage in autumn are often rufous buff below, and the underparts become white by spring. Subspecies have been described (Moore 1934, Carriker and Meyer de Schauensee 1935, Peters 1940) based on color of the underparts, the pattern of white spots on the head, the pattern of white and black on the tip of the tail, and the shape of the tip of the bill. Because these traits also vary within a population in the tail, and with season in the tip of the bill and in color of the underparts, the following named subspecies are not recognized here: Geococcyx velox affinis Hartlaub 1844, arid subtropical zone of El Salvador and western Guatemala; G. v. longisignum Moore, 1934, Honduras and northern Nicaragua; G. v. melanchima Moore 1934, western Mexico from southern Sonora to Isthmus of Tehuantepec; G. c. pallidus Carriker and Meyer de Schauensee 1935, semi-arid lowlands of eastern Guatemala and Yucatan.

Field characters Overall length 48 cm. Large terrestrial cuckoo with long legs and long tail, cocked up or hanging or held straight behind as bird runs through the scrub. Lesser Roadrunner differs from Greater Roadrunner G. californianus in the bold streaks on side of throat and breast, the middle of the throat and breast unmarked, the rich buff color of the

sides and flanks, the chestnut not gray back and rump, and the dark under tail coverts.

Voice Call a series of soft “coo” notes, repeated on a descending scale, low-pitched around 0.5 kHz and often breaking into a lower voice at 0.3 kHz, each note in the series lower in pitch and voice breaking earlier in the note. Notes are about 0.5 sec long; the song has 5–7 notes in a series and lasts 5–6 sec.Also has a dove-like single “coo” at a higher pitch, about 0.6 kHz (Hardy et al. 1990, Howell and Webb 1995).

Range and status Western Mexico from southern Sonora to Oaxaca and Chiapas and on the central plateau from Nayarit and Jalisco south to México, Morelos, Puebla and Veracruz (local), and an isolated population occurs in Yucatán and extreme N Campeche. In Central America they occur from Guatemala and Honduras to El Salvador and central Nicaragua (Howell and Webb 1995, AOU 1998; UMMZ). Not threatened, although their habitat at lower elevations has been taken over by humans. Resident.

Habitat and general habits Arid lowland scrub, arid montane scrub and pine forest at higher altitudes; dry open country with scattered brush, thorn scrub, grassy and lightly wooded hillsides, farmland, habitat similar to that of Greater Roadrunner. In El Salvador the birds occur on upper slopes and in brushy gullies, above timberline on Volcán de Conchagua and Volcán de San Miguel and

200 Banded Ground-cuckoo Neomorphus radiolosus on the summit of Volcán de Santa Ana, there and on Volcán de San Salvador they also occur on the lower slopes below cultivation (Dickey and Van Rossem 1938). Sea level to 2800 m (Stotz et al. 1996, AOU 1998).

Food Insects, mainly grasshoppers and caterpillars (Dickey and Van Rossem 1938).

Breeding and life cycle In Mexico they nest in Veracruz in April, in Sonora in June and July, in Oaxaca from May to July, in Chiapas in April (Short 1974, Rowley 1984,

Binford 1989, Russell and Monson 1998; UMMZ 109131 from Chiapas); in Guatemala in April and May (Owen 1861), in El Salvador in August (Miller 1932, Dickey and Van Rossem 1938).The nest is a bulky stick platform, with a cup about 15 cm in diameter, built above ground in a bushy tree, a thorn bush or the center of an Opuntia cactus thicket. Eggs are white, 35 ⫻ 26 mm (Schönwetter 1964). Clutch 2, less often 3 or 4 (Owen 1861, Miller 1932, Dickey and Van Rossem 1938, Rowley 1984, Howell and Webb 1995).The incubation and nestling periods are unknown. Both male and female attend the nest (Dickey and Van Rossem 1938).

Genus Neomorphus Gloger, 1827 Neomorphus Gloger, 1827, in Froriep’s Notizen aus dem Gebiete der Natur-und Heilkunde, 16, col. 278. Large terrestrial cuckoos of New World tropical forests.Type, by original designation and monotypy, Coccyzus geoffroyi Temminck 1820. Spectacular birds

with bushy crests, striking plumage patterns and very long tails. The name Neomorphus refers to a new or different kind. The nostrils are broadly operculate and the opening slit-like, unlike in other cuckoos. Four species.

Banded Ground-cuckoo Neomorphus radiolosus Sclater and Salvin, 1878 Neomorphus radiolosus Sclater and Salvin, 1878. Proceedings of the Zoological Society of London 1878, p. 439, pl. 27 (Intac, Ecuador). Monotypic.

white, under tail coverts blackish; bare skin around eye blue, iris dark brown, bill deep, black to dusky above and paler below and near the tip, feet bluish gray.

Description

JUVENILE: Head dull black, back black and underparts black with rusty bars; bare skin around the face blue and gray (López-Lanús et al. 1999).

ADULT: Sexes alike, large terrestrial forest cuckoo, plumage mainly black, forehead glossy black with buffy white bars, crest glossy black, nape black, upper back black with narrow, buffy white bars, wing coverts dark purplish brown, inner primaries and secondaries deep purplish red with blackish inner vanes, outer primaries black, lower back brown with fine black bars, rump brownish black, tail long and black, central feathers glossed green to purple and outer rectrices glossed purplish; underparts blackish with scaly feathers barred buffy

NESTLING: Undescribed. SOURCES: AMNH, BMNH, FMNH.

Measurements Wing, M (n ⫽ 2) 165–167 (166 ), F (n ⫽ 5) 162–172 (167.0 ⫾ 4.1), U (holotype) 177; tail, M 236–242 (239), F 236–254 (245.4 ⫾ 7.0), U 256; bill, M

Banded Ground-cuckoo Neomorphus radiolosus 201 45–46 (45.5), F 42–46 (44.8 ⫾ 1.6); tarsus, M 71–74 (72.5), F 68–74 (69.8 ⫾ 3.4) (AMNH, BMNH, FMNH). Wing formula, P5 ⱖ 4 ⱖ 3 ⬎ 2 ⬎ 1 ⬎ 6 ⬎ 7 ⬎ 8 ⬎ 9 ⬎ 10; P1 to P5 are nearly equal in length.

Field characters Overall length 50 cm. Large terrestrial cuckoo of forests in its range, crest and hind neck glossy black, forehead barred white, underparts black with buffy white scallops, the black breast band indistinct and obscured by the blackish underparts, wings purplish red and black.

Voice Bill-snaps are given around ant swarms. Call is a deep-pitched “oooo” like that of a cow or a dove, repeated in an irregular pattern, with notes at 400 Hz.

Range and status South America in foothills of southwestern Colombia (lower Río San Juan southward) and northwestern Ecuador. In addition to museum records, they have been seen in Colombia (Valle, Cauca, Nariño, Río Ñambi valley) and Ecuador (Esmeraldas, Imbabura, Pichincha, Mache-Chindul hills). There were only three records in Colombia since 1956 until contact was made in with the birds

1997 in the Río Ñambi valley north of Junín, and only one in Ecuador since 1936 until the birds were observed in 1996–1998 at Estación Biología Jatun Sacha Bilsa in the Mache-Chindul hills, Esmeraldas, in northwest Ecuador (Hornbuckle 1997, Ridgely and Greenfield 2001). In Nariño they were reported to be fairly common in the 1990s. Resident. Rare and local, little known (Hilty and Brown 1986) and threatened due to widespread forest destruction (Collar et al. 1994).

Habitat and general habits Tropical lowland evergreen forest, in secondary forest patch within primary forest, in foothills and lower slopes on Pacific slope of the west Andes from 450 to 1525 m, with most records from 700 to 1200 m; in recent years in Ecuador recorded only below 500 m (Hilty and Brown 1986, Ridgely and Greenfield 2001). Individual birds forage on the ground and in the understory, often in mixed-species flocks at army ant swarms; the foraging flocks were mainly of Ocellated Antbird Phaenostictus mcleannani, also Plain-brown Woodcreeper Dendrocincla fuliginosa, Immaculate Antbird Myrmeciza immaculata, Bicolored Antbird Gymnopithys leucaspis and other ant-following birds. It perches on the ground or fallen logs, scours live leaves and stems of understory, and examines tree trunk bases from the ground. Moves rapidly on the ground at ant swarms, sprints in bursts and stops abruptly, runs forward to capture food, and when it catches prey items, it runs in a zig-zag pattern. It follows peccaries, and in Ecuador and Colombia it is known locally as “guide of the wild pigs” (AMNH 475939, 475942) or “seinero” (“running with the peccaries”); in Colombia it is the “correlona” (“fast runner”). The bare blue skin around the eye expands and contracts, perhaps in display (LópezLanús et al. 1999).

Breeding Unknown.

202 Rufous-winged Ground-cuckoo Neomorphus rufipennis

Rufous-winged Ground-cuckoo Neomorphus rufipennis (G. R. Gray, 1849) Cultrides rufipennis G. R. Gray, 1849. Proceedings of the Zoological Society of London, 1849, p. 63. [lower Orinoco River,Venezuela]. Monotypic.

Description ADULT: Sexes alike, forehead, crown and crest black glossed purplish, head and neck purplish blue black, back and wing coverts dark bronzy green, wing black primaries, secondaries dark brownish red, lower back to upper tail coverts dull olive, tail black with central feathers metallic purple, other tail feathers with a greenish gloss; underparts mostly unmarked, throat ashy white to gray, lower throat feathers gray with black borders forming a “V”, breast black, lower breast and belly dull gray brown to slate gray, under tail coverts slate gray; bare skin on face red to red-violet, skin on nape cobalt blue covered by black feathers, eye-lashes above and below the eye, iris brown, bill black with the tip greenish to greenish white, feet greenish gray to gray-blue. JUVENILE: Crown black, back and wing coverts brown-slate and loose-webbed, primaries black, secondaries dark brownish red as in adult, lower back and rump black, tail black, glossy purple and green, rectrices narrower than in the adult; underparts slate, feathers loose-webbed, skin black; facial skin dull red, skin behind ear cerulean blue covered by black feathers, eye-lashes above and below the eye, iris brown, bill black and not deep as in the adult. NESTLING: Hatchling, undescribed. A just-fledged young has black skin below and a few dark brown hair-like feathers attached to tips of the crown feathers. SOURCES: AMNH, ANSP, FMNH.

Measurements and weights Wing, M (n ⫽ 6) 164–176 (169.5 ⫾ 4.2), F (n ⫽ 6) 152–174 (164.3 ⫾ 8.3); tail, M 262–290 (275.7 ⫾

12.9), F 266–300 (288.5 ⫾ 14.5); bill, M 39–42 (40.7 ⫾ 1.4), F 38–44 (41.3 ⫾ 2.0); tarsus, M 63–71 (67.4 ⫾ 3.4), F 68–71 (69.8 ⫾ 1.6) (AMNH, ANSP, FMNH). Weight, M (n ⫽ 2) 350–520 (435); F (n ⫽ 2) 315–340 (327.5), and an emaciated fledged young F 144 (ANSP). Wing formula, P4 ⫽ 3 ⫽ 2 ⬎ 5 ⫽ 1 ⬎ 6 ⬎ 7 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 50 cm. Large ground cuckoo with bare red skin around the eye, above dark bronzy green, head, neck and breast glossy blackish and belly dull grayish brown.The red eye skin (red and blue in N. pucheranii) and breast pattern distinguish this species from other ground cuckoos.

Voice Single hoot “whoú”, a low call at 0.8–0.9 kHz and lasting 0.3 sec; also loud single snaps with the bill (Haffer 1977, Hardy et al. 1990).

Range and status South America in Amazonian Colombia (Chiribiquete: Stiles et al. 1995), southern Venezuela (Bolívar, northern and central Amazonas),

Red-billed Ground-cuckoo Neomorphus pucheranii 203 Guyana (Siparuni, Ituribisi, Supenaam and Mazaruni Rivers, Iwokrama Reserve near Kurupukari, Kamakusa, Annai and Bartica) and northern Brazil (Roraima near upper Rio Branco) (Chubb 1916, Snyder 1966, Haffer 1977, Meyer de Schauensee and Phelps 1978, Hilty and Brown 1986, ANSP). Resident. Rare and not well known.

Habitat and general habits Tropical lowland evergreen forest, seasonally flooded forest, also forest highlands in interior foothill zone, lowlands to ⬎1100 ? m. Solitary, wary, restless, terrestrial, runs fast on the ground, perches in the middle stratum of the forest canopy. Adults are seen on ant mounds (ANSP) and near bands of peccaries (Meyer de Schauensee and Phelps 1978).

Food Insects, including grasshoppers, crickets, spiders (Pelzeln 1871, ANSP).

Breeding An emaciated just- fledged short-tailed young was taken on the Siparuni River in Guyana on 17 March 1997, and a female had large ova 6 ⫻ 5 mm on 23 Sept 1997 (ANSP 187615, 188270).The nest is unknown. Eggs are yellowish white, 40 ⫻ 31 mm (Schönwetter 1964). Incubation and nestling periods are unknown.The fledgling is accompanied by the parents who care for it, as the just-fledged young was taken on the same ant mound where an adult male and female were collected a few days earlier.

Red-billed Ground-cuckoo Neomorphus pucheranii (Deville, 1851) Cultrides Pucheranii Deville, 1851, Revue et Magasin de Zoologie Pure et Appliquee (2), 3, p. 211. [Rio Yaguas, Peru] Polytypic.Two subspecies. Neomorphus pucheranii pucheranii (Deville, 1851); Neomorphus pucheranii lepidophanes Todd, 1925.

Description ADULT: Sexes alike, crest black glossed purple, back olive brown, rump brown, tail dark chestnut with central feathers T1 above olive glossed green, tail below appears black; underparts, throat unmarked gray, breast gray with blackish scales at tips of feathers, a black band on the breast, the band wider on flanks (15 mm) than in center (7–8 mm), belly buffy to light gray, flanks and under tail coverts dark brownish gray, wing purplish chestnut, outer primaries black glossed violet blue; face with bare red skin bordered behind with bright blue, iris brown to purplish red, bill red or orange with a yellow tip, feet dark gray. JUVENILE: Above brown without green gloss, crest blackish, wing purplish chestnut; underparts,

blackish throat and breast, belly brown, plumage loose-webbed, tail feathers narrower than in adult; facial skin bare, bill black, feet slate gray to black. NESTLING: Buffy hair-like natal down attached to tips of growing contour feathers on the head, fluffy plumage unmarked dark brown; bill black (BMNH 88.8.23.60). SOURCES: AMNH, BMNH, CM, LSU, ROM, UMMZ, USNM.

Subspecies Neomorphus pucheranii pucheranii (Deville, 1851); forehead light brown, breast feathers with narrow inconspicuous scaly tips, breast and belly buffy to light gray; Amazonian Peru and E Brazil north of River Amazon; Neomorphus pucheranii lepidophanes Todd, 1925; forehead lacks the distinct brown contrast with the black crown, breast feathers with conspicuous scaly tips, breast and belly light tan or clay color; Amazonian Peru and E Brazil south of River Amazon.

204 Rufous-vented Ground-cuckoo Neomorphus geoffroyi

Measurements and weights Neomorphus p. pucheranii: Wing, M (n ⫽ 6) 166–178 (171.7 ⫾ 4.8), F (n ⫽ 6) 160–168 (165.8 ⫾ 3.0); tail, M 266–285 (273.8 ⫾ 7.3), F 262–273 (267.0 ⫾ 3.6); bill, M 48–56 (51.9 ⫾ 3.2), F 44–55 (50.7 ⫾ 3.6); tarsus, M 65–68.5 (66.9 ⫾ 1.4), F 60.5–67.2 (65.9 ⫾ 1.1) (AMNH, CM, LSU, UMMZ); Neomorphus p. lepidophanes: Wing, M (n ⫽ 6) 162–182 (170.2 ⫾ 7.4), F (n ⫽ 5) 162–183 (169.8 ⫾ 8.0) (AMNH, CM, FMNH); Weight, M (n ⫽ 1) 330 (LSU). Wing formula, P5 ⫽ 4 ⫽ 3 ⫽ 2 ⬎ 1 ⬎ 6 ⬎ 7 ⬎ 8 ⬎ 9 ⬎ 10; P1 to P6 nearly equal in length.

Field characters Overall length 50 cm. Large ground cuckoo, above dark bronzy green, head, neck and breast gray and belly dull gray, and red bare skin around eye and red bill.The red face differs from N. geoffroyi and N. radiolosus, and the red bill differs from N. rufipennis.

Voice The bird makes a guttural, roaring hum like a curassow, and it snaps the bill (Gyldenstolpe 1945a, Hilty and Brown 1986).

Range and status South America in Amazonian Peru (Río Curaray and Río Napo) and Brazil (Amazonia, from right bank of Rio Negro south to left bank of the Rio Purus) (Gyldenstolpe 1945a, Haffer 1977, Hilty and Brown 1986, Sick 1993). Sight observations have been reported in Amazonas, Colombia (Cuadros 1991), and in Amazonian Ecuador (Zábalo north of Rio Aguarico in E Sucumbíos,Yuturi in E Napo), the last

in near-sympatry with Rufous-vented Groundcuckoo N. geoffroyi (López-Lanús 1999, Ridgely and Greenfield 2001). Resident. Uncommon.

Habitat and general habits Tropical lowland evergreen forest, lowlands of upper Amazonia; lowlands to 700 m.Walks and runs on the ground, leaps from ground. It regularly forages at army ant swarms, and follows peccaries and tamarin monkeys Saguinus (Gyldenstolpe 1945a, Willis and Oniki 1978, Willis 1982, Siegel et al. 1989, López-Lanús 1999).

Food Insects, also plant matter (LSU).

Breeding In Peru a nestling was taken in February (BMNH). The nest and eggs are unknown except for a report of two eggs brooded by a pair (Castelnaux 1855, in Sick 1993).

Rufous-vented Ground-cuckoo Neomorphus geoffroyi (Temminck, 1820) Coccyzus geoffroyi, Temminck, 1820, Nouveau recueil de planches coloriées d’oiseaux, livre 2, pl. 7. [Pará, Brazil] Polytypic. Seven subspecies. Neomorphus geoffroyi geoffroyi (Temminck, 1820); Neomorphus geoffroyi salvini Sclater, 1866; Neomorphus geoffroyi aequatorialis Chapman, 1923; Neomorphus geoffroyi

squamiger Todd, 1925; Neomorphus geoffroyi dulcis Snethlage, 1927; Neomorphus geoffroyi maximiliani Pinto, 1962.

Description ADULT: Sexes alike, crest feathers black with a blue gloss, back and wing coverts bronzy brown, wing

Rufous-vented Ground-cuckoo Neomorphus geoffroyi 205 glossed green, outer primaries black with a purplish gloss, rump and upper tail coverts rufous brown, tail glossy black, forehead and underparts buff to cinnamon brown, breast feathers with dusky semicircular bands, breast with a narrow black band, lower breast buff, belly and under tail coverts rufous; bare facial skin blue, iris dark brown or with an inner ring of yellow and an outer ring of red, bill large and yellow-green to pale green, tarsi gray. JUVENILE: Plumage loose and fluffy, darker and duller than adult, head and back blackish, wing and tail black with greenish gloss, underparts blackish with rufous gray on belly; bare skin around eye gray with a bright blue spot behind the eye, iris dark brown to grayish yellow, bill black, tarsi bluish gray. NESTLING: Hatchling, undescribed. A just-fledged young bird in fluffy blackish plumage had a mouth lining of rose pink color, with small white papillae in the maxillo-palatine area. SOURCES: AMNH, BMNH, CM, FMNH, LSU, MCZ, MVZ, UCLA, UMMZ, USNM.

Subspecies Neomorphus geoffroyi salvini (Sclater, 1866); forehead and crown unbarred rufous brown, upperparts bronzy brown, breast feathers with broad pale tips, breast band distinct; Nicaragua (Caribbean slope), Costa Rica (mainly Caribbean slope, also Pacific drainage in Guanacaste), Panamá and the Pacific coast of Colombia; Neomorphus geoffroyi aequatorialis Chapman, 1923; breast feathers distinctly barred gray and buff; Ecuador, Peru, SE Colombia; Neomorphus geoffroyi geoffroyi (Temminck, 1820); forehead and crown brown barred with black feather edges giving a scaly appearance, upperparts olive green, breast with dusky bars formed with a dark brown subterminal bar on each buff feather, a lower breast band present; Brazil south of the Amazon, Pará; Neomorphus geoffroyi squamiger Todd, 1925; no black breast band; Brazil south of Amazon, lower Rio Tapajóz area;

Neomorphus geoffroyi australis Carriker, 1935; breast nearly unmarked in center and dusky bars on the side; Bolivia and Peru; Neomorphus geoffroyi maximiliani Pinto, 1962; upperparts dark bronze green, breast marks distinct; SE Brazil: Bahia; Neomorphus geoffroyi dulcis Snethlage, 1927; upperparts dark blue to green, breast marks distinct; SE Brazil from Espirito Santo to Rio de Janeiro, the range now a single locality in the Rio Doce basin. Seven subspecies are currently recognized in two groups, the first with the forehead and crown unbarred, the second with the forehead and crown barred. Genetic interchange between these populations in the not too remote past is suggested by intermediate populations between most of the extremes that are characterized in subspecies descriptions. Dispersal and genetic interchange between regions that are not separated by major rivers is suggested within Rufous-vented Ground Cuckoo N. geoffroyi, as (1) the color of the barred forehead and crown intergrade with salvini in NW Colombia having less rufous than salvini in Nicaragua (birds in Colombia are intermediate in color between salvini and aequatorialis), (2) the breast markings of the subspecies geoffroyi, squamiger and australis intergrade clinally, although locally in southern Amazonia and extreme northern Mato Grosso (Alta Floresta) the plumage is not well known, (3) the back color of geoffroyi and australis intergrade, and a Brazilian bird (AMNH 430469) taken in Rio Tocantins has a breast band that is nearly broken in the center and is intermediate between these two forms, and (4) the back colors of geoffroyi, maximiliani and dulcis vary clinally from green to blue (Griscom and Greenway 1941, Gyldenstolpe 1945a, Haffer 1977, Zimmer et al. 1997; AMNH, BMNH, LSU). N. g. squamiger is sometimes considered a separate species (Haffer 1977). However, the genetic distance between squamiger and N. geoffroyi salvini from Panama is much less than that between those birds and the other ground-cuckoos, whereas the genetic distances between the four species represented in Figure 5.3 indicate very nearly identical rates of genetic change within the Neomorphus groundcuckoos. Taken together, these genetic distances

206 Rufous-vented Ground-cuckoo Neomorphus geoffroyi indicate that squamiger is not a species distinct from N. geoffroyi.Also, examination of museum specimens shows intergradation across their geographic range between N. g. salvini and N. g. aequatorialis, and between N. g. geoffroyi, N. g. australis and N. g. squamiger. Early observers failed to find a nest of any Neomorphus species, and for a time there was speculation that these ground-cuckoos were broodparasitic. Sick (1949) discovered the parents and young together and established that these cuckoos cared for their offspring, and the adults have been seen with their nest.

bare colored skin on the face and a flat frontal erectile crest. Rufous-vented Ground-cuckoos are blackish or dark brown to rufous above, scaly brown on the breast, have a black breast band (lacking in squamiger) and rufous belly, and bare skin on the face gray to blue.

Measurements and weights

Range and status

Neomorphus geoffroyi salvini: Wing, M (n ⫽ 6) 165–175 (169.3), F (n ⫽ 11) 160–169 (163.6); tail, M 244–262 (254.8), F 237–266 (250.0); bill, M 43.6–49.5 (46.4), F 45.7–51.1 (47.9); tarsus, M 69.2–72.4 (71.6), F 69.5–73.1 (71.5) (Wetmore 1968); Nicaragua to Panamá, wing, M (n ⫽ 11) 167–184 (172.4 ⫾ 5.1), F (n ⫽ 15) 161–194 (170.1 ⫾ 7.9); tail, M 250–281 (263.3 ⫾ 10.7), F 247–278 (264.3 ⫾ 10.0); bill, M 40.0–52.3 (46.5 ⫾ 4.5), F 41–51.5 (43.3 ⫾ 3.0); tarsus, M 61–71 (67.6 ⫾ 3.2), F 63–71 (67.0 ⫾ 2.9) (AMNH, LSU, UMMZ, USNM); N. g. aequatorialis: M (n ⫽ 8) 162–188 (170.4 ⫾ 8.7), F (n ⫽ 2) 162–170 (166) (AMNH, FMNH); N. g. squamiger: M (n ⫽ 3) 163–171 (166), F (n ⫽ 2) 160–170 (165) (AMNH, CM); N. g. dulcis: M (n ⫽ 2) 172–173 (172.5), F (n ⫽ 1) 163, U (n ⫽ 2) wing 170–177 (173.2) (AMNH, BMNH). Weight, N. g. salvini: M (n ⫽ 1) 350, F (n ⫽ 2) 375–400 (387.5) (LSU); N. g. aequatorialis, M (n ⫽ 1) 340 (LSU); N. g. australis: Bolivia, M (n ⫽ 1) 370 (LSU); N. g. dulcis: M (n ⫽ 3) 339–355 (349), F (n ⫽ 1) 349 (Sick 1949, Haffer 1977); N. g. squamiger: M (n ⫽ 1) 340 (Graves and Zusi 1990). Wing formula, P6 ⬎ 5 ⬎ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 7 ⬎ 8 ⬎ 9 ⬎ 10.

Voice Loud bill-clapping, a soft dove-like moaning “oooooo-oóp”, a low “woof ” at army ant raids, and a loud “kchak!” in feeding competition (Slud 1964, Haffer 1977, Hilty and Brown 1986, Stiles and Skutch 1989, Hardy et al. 1990, Sick 1993).

Central and South America, from Nicaragua south through Costa Rica and Panamá to South America from western Colombia (Panamá border south on Pacific coast to Baudó Mts and east along N base of Andes in Córdoba to upper Río Nechí, and east of Andes in W Caquetá and W Putumayo), eastern Ecuador (close to base of Andes and from south of Río Napo, at Limoncocha, La Selva, Yuturi, Jatun Sacha, SE of Pompeya, Santiago in MoronaSantiago), Peru (except in eastern Amazonian region, where N. pucheranii occurs), NW Bolivia and Brazil (widespread except north of Rio Negro and Manaus in Rio Branco region where N. rufipennis occurs) (Howell 1957, Pinto 1962, Haffer 1977, Hilty and Brown 1986, Wetmore 1988, Remsen and Traylor 1989, Stiles and Skutch 1989, Dubs 1992, Sick 1993, Novaes and Cunha Lima 1998, López-Lanus 1999, Ridgely and Greenfield

?

Field characters Overall length 50 cm. Large terrestrial cuckoo of Neotropical forests, ground-cuckoos are fowl-sized, long-legged, long-tailed birds with a decurved bill,

Rufous-vented Ground-cuckoo Neomorphus geoffroyi 207 2001). Resident. Rare and local, solitary. They live in large blocks of natural forest, and like other ground-cuckoos they appear unable to cross the large rivers, which limit dispersal to movements along a river bank. May occur together with Redbilled Ground-cuckoo N. pucheranii near Yuturi in eastern Ecuador. This ground-cuckoo was one of the first birds to disappear from the protected but isolated forest at Barro Colorado, Panamá, where last seen in 1935 (Willis 1974, Eisenmann and Willis 1979). Population density, 0.25 pairs / 100 ha in floodplain forest in Amazonian Peru (Terborgh et al. 1990). Ground cuckoos in southern Brazil are disappearing due to widescale deforestation (Sick 1993). Because of its restricted range, N. g. squamiger are considered near-threatened (Collar et al. 1994). N. g. dulcis were formerly widespread in southeastern Brazil (Snethlage 1927, Pinto 1952) but have not been reported at the site where last observed, Rio Doce State Park, since before the 1980s, and these birds may be extinct (Knox and Walters 1994, Machado et al. 1998). Population density is sparse, at Cocha Cashu, Peru, estimated 0.25 territories in 100 ha of mature forest (Terborgh et al. 1990, Robinson and Terborgh 1997).

Habitat and general habits Tropical lowland forests, high ground forest and river bluff forest, and forest with seasonal flooding, in Ficus and Heliconia, canebrakes and thickets; in lowlands and foothills to 1000 m. Usually seen in riverine or slope forests in treefalls and vines, in dense undergrowth in primary forest and immature forest.Terrestrial, agile, can run, and can flutter to an elevated perch to lookout and to roost, but they are not seen in sustained flight. One or a pair (sometimes more, perhaps a family group) are seen at a time as they feed on forest floor in open undergrowth in mature forest. Feeding birds look into heaps of branches, armadillo holes and empty terrestrial termite mounds.Ant-following insectivores,

they forage at swarms of the army ants Eciton along with other ant-following birds, bills snapping at the edges of ant raids, tossing aside the leaf litter. They also follow peccaries. They leap over logs, dash about on the ground, flick aside leaves and peck at insects in the litter, and capture flushed out arthropods as the cuckoos take short runs along the ground (Sick 1949, 1953b, 1993, Howell 1957, Willis 1974, 1982, Terborgh et al. 1984, Stiles and Skutch 1989, Stotz et al. 1997). Crest erects when birds are excited.

Food Large insects, roaches, beetles; also scorpions, centipedes, spiders, small frogs, lizards, small birds, and occasional seeds and fruit (Pelzeln 1871, Sick 1949, Wetmore 1968, LSU).

Breeding In Central America breeds in the northern summer or wet season, in Nicaragua there was a fledged young in July, in Costa Rica there were fledged young in July and December. In northern Colombia a bird had an oviduct egg in April; in Mato Grosso, Brazil, they breed in September; in SE Brazil they breed in the southern summer (Sick 1949, Willis 1974, 1982, Roth 1981, Stiles and Skutch 1989; AMNH, BMNH, UMMZ, USNM). The nest is a broad shallow bowl of sticks, 25 cm across with a 12-cm cup, built 2–3 m above the ground in a shrub in dense swampy vegetation and lined with green leaves added through incubation. Eggs are yellowish white, 43 ⫻ 32 mm (oviduct egg, Wetmore 1968) to 40 ⫻ 32 mm (Roth 1981), clutch 1. The incubation and nestling periods are unknown.When disturbed, the incubating bird slips off the side of the nest and walks away. The first observation that ground-cuckoos rear their own young was made in 1941, when a pair was seen to attend their fledged young in Brazil (Sick 1949); another just-fledged young was taken with an adult male in Panamá (LSU).

208 Buff-headed Coucal Centropus milo

Centropodinae Genus Centropus Illiger, 1811 Centropus Illiger, 1811, Prodromus systematis mammalium et avium, p. 205. Large, heavy-bodied ground cuckoos of the Old World.Type, by subsequent designation, Cuculus aegyptius Gmelin (G. R. Gray, List Gen. Birds, 1840, p. 56). Large cuckoos with a stout bill, short rounded wings and long, broad tails. The genus name refers to the foot (Gr. kentron, a spike or spur, and pous, the foot), describing the long, straight hallux claw of most coucals; the German term for coucals “Sporenkuckucke” also refers to the spurred

foot. The broad wings cover the back; the exposed upper back (or mantle) is often the only region not covered by the wing; the back and rump are often covered by short blackish feathers that are concealed by the closed wing, in smaller species much as in the large broad-winged ground cuckoos (Carpococcyx, Neomorphus). The common name “coucal” may derive from “coucou” and “alouette”, the latter indicating the straight hallux claw as in a lark (“alouette”) (Cuvier, in Newton 1896). 26 species.

Buff-headed Coucal Centropus milo Gould, 1856 Centropus Milo Gould, 1856, Proceedings of the Zoological Society of London, 1856, p. 96, 136. (Guadalcanar, Solomon Islands) Polytypic. Two subspecies. Centropus milo milo Gould, 1856; Centropus milo albidiventris Rothschild, 1904.

Description ADULT: Sexes alike; buffy white head, upper back, throat and upper breast, lower back to rump and upper tail coverts black, wing black, tail long, graduated and black; belly either buffy white or black; iris red to red-brown, bill black, skin at base of bill dark gray, feet blue gray. JUVENILE: Upperparts and underparts brown mottled and barred with black, wing and long tail barred black and rufous; iris gray to brown, bill brown, lower mandible pale horn, whitish below, feet bluish gray. NESTLING: Undescribed. SOURCES: AMNH, BMNH, FMNH, MVZ, UMMZ, UWBM.

Subspecies Centropus milo albidiventris Rothschild, 1904; breast, belly and thighs whitish; Solomon Islands (Vella Lavella, Bagga, Ganongo, Simbo, Gizo, Kulambangra, Kohinggo, New Georgia,Vangungu, Gatukai, Rendova,Tetipari); Centropus milo milo Gould, 1856; breast, belly and thighs black; Solomon Islands (Florida, Guadalcanal).

Measurements and weights Wing, M (n ⫽ 9) 260–283 (270.6 ⫾ 7.0), F (n ⫽ 9) 245–290 (266.8 ⫾ 16.5); tail, M 325–410 (358.0 ⫾ 25.0), F 312–372 (346.2 ⫾ 21.5); bill, M 52–63 (58.4 ⫾ 2.5), F 56–73 (63.9 ⫾ 5.8); tarsus, M 60–77 (65.6 ⫾ 6.4), F 67–74 (70.4 ⫾ 3.0); hallux claw, M 19–23.5 (21.3 ⫾ 1.8), F 17–23.5 (21.2 ⫾ 2.6) (AMNH, UMMZ, UWBM). Weight, M (n ⫽ 5) 742–790 (769.4) (UWBM). Wing formula, P 4 ⬎ 5 ⫽ 3 ⬎ 2 ⬎ 6 ⬎ 1 ⬎ 7 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 60–68 cm. Large coucal with upperparts, wing and tail black, with a buffy white

Buff-headed Coucal Centropus milo 209 head, upper back, throat and breast (and belly in some island regions). Juvenile is brown with black markings.

Voice A “sawing” or “snoring” noise, or a roar like a lion, “na-ow”’; members of a pair alternate calls “urrrh,” “-uh,” and a solitary bird gives a guttural “kkkkow . . . kkkk . . . kk . . . kk,” a throaty bark, and a deep churring “argh-a-argh,” and mellow grunts. Juveniles have a stuttering cough “koko, kokokkokoko . . .” (Sibley 1951, Cain and Galbraith 1956, Doughty et al. 1999, Diamond 2002, C. E. Filardi recording).

Range and status Solomon Islands (Vella Lavella. Gizo, Kuiambangra, New Georgia, Rendova, Guadalcanal and Florida island groups; absent on Choiseul, Ysabel, Malaita and San Cristobol island groups). Common resident (Hartert 1925b, Galbraith and Galbraith 1962, Blaber 1990, Sibley and Monroe 1990, Doughty et al. 1999, Mayr and Diamond 2001). Coucals apparently have had little success in dispersal between islands. In the New Georgia island group, coucals occurs on the nine largest islands, and on the 11th to 13th next largest islands, and are absent on smaller islands. In the Florida group they are

present on the largest island but not on the second largest, even though the islands are separated by only 150 to 1000 m of water. These birds are easy to detect because they roar loudly and they are well known by local people (Diamond 2002). The absence of large coucals on Buka and Bougainville may be the result of extinction after the Solomons were colonized from the Bismarcks by an ancestral coucal. extinction that might be owing to the human population with their introduced possums, rats, pigs and dogs. Or the absence may be due to another factor leading to extinction of coucals on the northern Solomons.

Habitat and general habits Lowland, primary and secondary forest and thickets, in undergrowth along stream in Casuarina hill forest, forested karst topography on coastal and raised limestone, hill and mist forest, Campernosperman plantations, and to 500 m on Kulambangra they occur around human habitation in villages and gardens, and secondary vegetation along the edge of airstrips. They live on the ground in the forest and outside it near the forest edge.They feed on the ground as they walk; and they are seen in trees and glide from tree to tree. They are mobbed by monarch Steel-blue Flycatcher Myiagra ferrocyanea, suggesting they take small birds and eggs and young in the nests (Sibley 1951, Blaber 1990, Doughty et al. 1999, C. Filardi, pers.comm.).

Food Insects, including stick insects, grasshoppers, beetles, pupae; giant centipedes (Sibley 1951, Blaber 1990, Doughty et al. 1999).

Breeding Males have enlarged testes (one per bird, or one with a vestigial second testis) in February, May and October (Sibley 1951, UWBM).The nest and eggs are unknown.

210 Pied Coucal Centropus ateralbus

Pied Coucal Centropus ateralbus Lesson, 1826 Centropus ateralbus Lesson, 1826, Bulletin Universel des Sciences et de l’Industrie, 8, sect. 2, p. 113. (New Ireland) Other common names:White-necked Coucal. Monotypic.

Description ADULT: Sexes alike, plumage black and white, and a long black tail. Plumage is variable in the extent of white. In the most common plumage phases, the cap and face are black, wing coverts black with a white patch, wing glossy purplish blue, neck and upper breast white, belly and under tail coverts black, and others are (1) all white with black forehead and lores, (2) all white on head and upper back, (3) black only on crown, (4) all black with white in wing patch, while other plumage variations have the white replaced by brownish white or pearly gray; iris red, bill black, legs slaty blue to black. Subadult or immature, blackish with white feathers on neck and wing. JUVENILE: Anterior half of body streaked and mottled with buff or rufous, rachi of hackles pale strawwhite, barbs lacking near tip, underparts plumage loose-webbed with barbs lacking; iris brown to gray, bill black, legs blue-gray. NESTLING: Undescribed. SOURCES: AMNH, BMNH, FMNH, RMNH, ZMUC.

Measurements and weights Wing, M (n ⫽ 7) 192–217 (201.6 ⫾ 8.8), F (n ⫽ 7) 210–227 (214.1 ⫾ 10.4); tail, M 252–270 (263.0 ⫾ 8.8), F 195–227 (293.6 ⫾ 13.1); bill, M 39.5–44.0 (42.5 ⫾ 1.6), F 38.0–43.5 (41.2 ⫾ 1.9); tarsus, M 44.5–52.0 (49.4 ⫾ 2.9), F 45–54 (48.8 ⫾ 3.2); hallux claw, M 16–23 (19.3 ⫾ 2.3), F 18–28 (22.4 ⫾ 3.6) (AMNH, ZMUC).The hallux claw is large in the chick: 14 mm in a nestling with a tarsus of 37 mm, and 15 mm in one with a tarsus of 40 mm. Weight, M (n ⫽ 1) 330, F (n ⫽ 1) 342, U (n ⫽ 1) 345 (AMNH, USNM).

Wing formula, P6 ⬎ 7 ⫽ 5 ⬎ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 44–48 cm.A large coucal with black and white plumage, the only coucal like this in its range.

Voice Song a duet of two birds sounding like two hollow metal drums, “soo-hoo,” the duet lasting as long as 17 sec, the call a rising pitch, two birds drifting in and out of phase with each other. Alarm, a short “chit,”“krek” or “chunk”, sometimes repeated rapidly, also gives a nasal call “k-k-k-naaah!” (Coates 1985, 2001, Beehler et al. 1986). Dahl (1899) likened the call to the hollow toot of a village night watchman, and heard the call day and night.

Range and status Bismarck Archipelago (Umboi, New Britain, Lolobau, New Ireland, Dyaul); absent on the St. Matthais group and the western Bismarcks (Hartert 1925c, Gilliard and LeCroy 1967a, Orenstein 1976, Coates 1985, 2001, Mayr and Diamond 2001). Resident. The species was first seen by western eyes when Lesson, zoologist on the Coquille expedition, visited the Bismarck Archipelago in 1823; and it was one of the first bird species recorded by the first European explorers to the region. Lesson described

Greater Black Coucal Centropus menbeki 211 the bird as the “coucal atralbin” (Mayr and Diamond 2001).

Habitat and general habits Forest, forest edge, secondary forest, vegetation on sea shore, coconut plantations and gardens.They mainly frequent shrubbery and regrowth. Occur in lowlands near the coast and in hills, locally to 1200 m in New Britain and 1400 m in New Ireland, more common in New Britain than in New Ireland.They climb in coconut trees, walking on the midribs of hanging fronds, gleaning in and around the coconuts and in the center of a tree, and in the undergrowth in overgrown coconut plantations (LeCroy and Peckover 1983).They forage from the ground climbing to the tops of shrubs and into high trees, gliding down to a low level again, active and clumsy in movements. They occur in pairs and in family groups numbering up to four birds (Hartert 1926, Coates 2001).

Food Insects, including longicorn beetles Cerambycidae and large stick insects Phasmatodea, lizards, probably other small animals (Dahl 1899, Coates 1985, 2001).

Breeding Feathered nestlings and recently fledged young are known from November to February and a justhatched nestling was seen in February (Dahl 1899). Feathered nestlings have also been taken in May (ZMUC, Noona Expedition, 2.1.1982.4) and in July or August and (Rothschild and Hartert 1914). Nest is a hollow or chamber, built of dry stems and leaves of reed-grasses, sometimes with two holes front and back, lined with green vegetation. Eggs are whitish, 41 ⫻ 34 mm, clutch 2 or 3 (Meyer 1933, 1936, Schönwetter 1964). The incubation and nestling periods are unknown.

Greater Black Coucal Centropus menbeki Lesson and Garnot, 1828 Centropus Menbeki Lesson and Garnot, 1828, Voyage autour du monde sur la corvette la ‘Coquille,’ Atlas, pl. 33; Zoologie, 1, livre 13, 1829, p. 600. (New Guinea (Dorey, ⫽ Manokwari)) Other common names: Greater Coucal, Menbek’s Coucal, Black Jungle Coucal. Polytypic. Two subspecies. Centropus menbeki menbeki Lesson and Garnot, 1828; Centropus menbeki aruensis (Salvadori, 1878).

JUVENILE: Upper-parts and underparts dull blackish, some neck and upper back feathers with strawcolored rachi, wing black sometimes with faint rufous bars, tail black with indistinct narrow rufous bars, the feathers 40–50 mm broad at mid-length and nearly pointed at the tip; iris light gray (AMNH 425917, 425920, 766255).

Description

SOURCES: AMNH, ANSP, BMNH, FMNH, MCZ, MSNG, MVZ, MZB, RMNH, ROM, SMTD, USNM, ZFMK.

ADULT: Sexes alike, huge size, upperparts and underparts glossy black, stiff black hackles on forepart of body, wing black, tail long, graduated, glossy black, the feathers 60–70 mm broad at midlength and wide at the tip; iris red, bill whitish with black base, legs black. SUBADULT: (or NON-BREEDING): Upperparts dull black, stiff black hackles, tail black with distinct narrow rufous bars on all but the tip, the feathers 50–60 mm broad at mid-length and rounded at the tip (AMNH 425915).

NESTLING: Undescribed.

Subspecies Centropus menbeki menbeki Lesson and Garnot, 1828; plumage glossed blue to blue-green; New Guinea and the western Papuan islands,Yapen Is; Centropus menbeki aruensis (Salvadori, 1878); plumage glossed purple; Aru Islands. History: An earlier account and illustration by Martin in 1785, “The Green Bird of Paradise” had a red iris, glossy black plumage with scale-like body

212 Greater Black Coucal Centropus menbeki feathers, a long tail, zygodactyl feet and a short hallux claw, but had a black bill (Stonor 1937). Geographic variation: Birds of Yapen (Japen, Jobi) are more blue (less green) than certain birds of New Guinea, and were described as a subspecies C. m. jobiensis Stresemann and Paludan 1932, but birds with the same blue plumage color occur on the Sepik River (Rothschild et al. 1932), the Fly River area (AMNH, Rothschild and Hartert 1907) and on Misool (ANSP, Mayr and Meyer de Schauensee 1940c).

Measurements and weights C. m. menbeki, New Guinea: Wing, M (n ⫽ 13) 210–240 (224.5 ⫾ 9.1), F (n ⫽ 10) 209–240 (220.8 ⫾ 8.3); tail, M 344–411 (367.9 ⫾ 23.0), F 326–391 (355.2 ⫾ 17.5); bill, M 46–54 (49.1 ⫾ 3.5), F 44–55 (50.3 ⫾ 3.8); tarsus, M 55–64 (59.0 ⫾ 3.2), F 57–61 (59.6 ⫾ 1.4); hallux claw, M 16–23 (20.3 ⫾ 2.9), F 18–21 (19.6 ⫾ 1.1) (AMNH, FMNH, MSNG, RMNH);Yapen:Wing, M (n ⫽ 3) 225–235 (230.7), F (n ⫽ 2); 224–225 (224.5); tail, M 350–360 (355), F 333–360 (346.56) (AMNH, FMNH, MSNG); C. m. aruensis, Aru Islands: Wing, M (n ⫽ 2) 234–238 (236), F (n ⫽ 1) 260; tail, M 328–378 (353), F 416 (AMNH, SMTD). Weight, New Guinea: M (n ⫽ 6) 430–575 (493), F (n ⫽ 2) 505–553 (529) (AMNH, FMNH, MZB, RMNH, Hartert 1930, Ripley 1964);Yapen: F (n ⫽ 1) 450 (FMNH). Wing formula, P4 ⫽ 3 ⫽ 2 ⬎ 1 ⱖ 5 ⬎ 6 ⬎ 7 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 67 cm. A large black coucal of the forests of New Guinea and the Aru Islands.

Voice Low-pitched, resonant booming hoots staccato, single note “oodle,” also pairs of “hoo hoo,” or four hollow “hoo” notes, and a series that descends in pitch “Uh-oo-oo-oo-oo-oo-oh”, a grunt or croak followed by a dry rattle.The call is like “an un-oiled door being shut” (Gyldenstolpe 1955). Calls are deeper and delivered more slowly than in Pheasant Coucal C. phasianinus in New Guinea and more complex and longer than in Lesser Black Coucal C. bernsteini. Pairs call in duet. Coucals sometimes

call after sunset and at night (Rand 1942a, Diamond 1972, Coates 1985, Beehler et al. 1986).

Range and status New Guinea, western Papuan islands (Batanta, Salawati, Misool), Yapen and Numfor islands, and Aru Islands (Salvadori 1881, Rothschild and Hartert 1907, Hartert 1930, Rothschild et al. 1932, Hartert et al. 1936, Maxwell 1938, Mayr and Meyer de Schauensee 1940b, Mees 1982, Gregory 1995, Coates 1985, Sibley and Monroe 1990). Resident. Widely distributed, uncommon. Density, c. 1 bird / 10 ha (Bell 1982a).

Habitat and general habits Forest, forest edge, in low undergrowth, shrub and lower middle stories, and in swampy grasslands; from sea level to 850 m (Rand 1942a,b), while in the eastern highlands of New Guinea, they have been seen at Karimui at 1275 m, well above their usual altitudinal range (Diamond 1972). Feed on the ground, also hop up tree trunks and vines while they switch the tail from side to side. Terrestrial, they seldom fly (Ogilvie-Grant 1915, Stein 1936, Diamond 1972, Coates 1985, Beehler et al. 1986).

Food Small vertebrates (snakes, frogs, small birds), arthropods, large insects (grasshoppers, cicadas, caterpillars, beetles) taken on the ground (Rothschild et al. 1932, Rand 1942a,b).

Breeding Near the Idenburg River in northern West Papua a female had an oviduct egg in April, fledged young

Biak Coucal Centropus chalybeus 213 were seen in the middle Sepik River area in January and near the Fly River in October, and a female was in breeding condition in September (Rand 1942a, b, Gilliard and LeCroy 1966). Nest is a large mass of

leaves, in pandanus, built in the wet season.The egg, known from the oviduct of a breeding female, is white, oval, 37 ⫻ 30.3 mm (Rand 1942b). Clutch size, incubation and nestling periods are unknown.

Biak Coucal Centropus chalybeus (Salvadori, 1875) Nesocentor chalybeus Salvadori, 1875, Annali del Museo Civico di Storia Naturale di Genova, 7, p. 915. (Misori ⫽ Biak) Monotypic.

Description ADULT: Sexes alike, plumage black, upperparts with a dull purple gloss, foreparts with spiny feathers, wing black, long black tail, underparts black to dark brown on belly; iris yellow, bill black, legs and feet black. JUVENILE: Plumage unbarred, dark and duller than in adult, some birds are washed rufous.

Voice NESTLING: Undescribed. SOURCES: AMNH, ANSP, MSNG, MZB, SMTD, ZMB.

Measurements

Loud, hollow notes, a series of upslurred “hoot” notes, sometimes slightly accelerated, the notes descending or going up and down in waves, also a harsh rasp, and a repeated “bup”; a noisy bird (Beehler et al. 1986).

Wing, M (n ⫽ 5) 189–214 (203.0), F (n ⫽ 6) 189–220 (203.0 ⫾ 9.7); tail, M 260–285 (274.2), F 253–315 (280.3 ⫾ 23.0); bill, M 44–50 (47.2), F 40–49.3 (45.0 ⫾ 3.1); tarsus, M 47–54 (50.4), F 50–57 (53.6 ⫾ 2.4); hallux claw, M 15.0–16.2 (15.6), F 15.2–16.2 (15.6) (AMNH,ANSP, MSNG, SMTD, ZMB). Wing formula, P5 ⬎ 4 ⬎ 3 ⬎ 2 ⫽ 6 ⬎ 1 ⬎ 7 ⬎ 8 ⬎ 9 ⬎ 10 (P1 is half the length of P5).

Range and status

Field characters

Habitat and general habits

Overall length 44–46 cm. Small coucal with black plumage, a relatively short tail and a yellow iris.This is the only coucal on Biak Island. It is smaller and has a straighter bill than the Greater Black Coucal C. menbeki of New Guinea, and a stockier build and yellow iris unlike the Lesser Black Coucal C. bernsteini of New Guinea.

Primarily lowland forest, also thick second growth, on ground, in vines and in trees.Terrestrial, it hops and feeds on the ground. It is awkward in flight, like other coucals.

Biak Island, in Geelvink Bay north of Irian Jaya (Salvadori 1875, Mayr and Meyer de Schauensee 1940a, Rand and Gilliard 1967, Beehler et al. 1986). Resident. Hunting and habitat destruction have reduced the numbers in southern Biak, and the main population is in the forests of Supiori (Bishop 1982); conservation status near-threatened (Collar et al. 1994).

Breeding Unknown.

214 Rufous Coucal Centropus unirufus

Rufous Coucal Centropus unirufus (Cabanis and Heine, 1863) Pyrrhocentor unirufus Cabanis and Heine, 1863, Museum Heineanum, 4(1), p. 118. (Philippines [Luzon]) Monotypic.

Description ADULT: Sexes alike, plumage rufous chestnut, head slightly darker, wing and tail uniform rufous chestnut; bare skin around eye yellow, iris light brown, bill green with yellow tip, feet black. JUVENILE: Like adult, rectrices narrow; bill black. NESTLING: Yellowish-white hair-like down on head. SOURCES: AMNH, CM, FMNH, USNM.

Measurements and weights Wing, M (n ⫽ 11) 145–162 (155.7 ⫾ 5.6), F (n ⫽ 8) 155–165 (161.5 ⫾ 3.3); tail, M 213–242 (228.2 ⫾ 1.9), F 220–262 (237.6 ⫾ 12.8); bill, M 32–36 (33.8 ⫾ 1.1), F 33–38 (35.3 ⫾ 1.6); tarsus, M 36–45 (40.2 ⫾ 3.3), F 39–45 (41.8 ⫾ 2.5); hallux claw, M 16–20 (18.2 ⫾ 1.5), F 17–19 (17.9 ⫾ 0.6) (FMNH). Weight, M (n ⫽ 6) 145.7–187.7 (168.3), F (n ⫽ 8) 146–227.9 (201.4) (FMNH, USNM, Goodman and Gonzales 1990). Wing formula, P5 ⫽ 4 ⬎ 3 ⬎ 6 ⬎ 2 ⬎ 1 ⬎ 7 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 38–42 cm. Coucal with uniform light brown to rufous plumage, pale iris and green bill.

Voice Call, a snapping staccato of two to five pipsqueak notes,“squip-whip” or “squip-whip-whip . . .”, high in pitch at 4–5 kHz as the birds move through the undergrowth; also squeaks and short metallic trills, and a shrill, plaintive “kaow” when a bird is separated from its mate (Gilliard 1950, Scharringa 1999, Kennedy et al. 2000).

Range and status N Philippines (Luzon, Polillo, Catanduanes). Resident. Uncommon and local (Stresemann 1939, Gilliard 1950, Gonzales 1983, Dickinson et al. 1991). Because of its restricted range and the loss of forest habitat in the Philippines, the species is considered near-threatened (BirdLife International 2001).

Habitat and general habits Hill forest in matted undergrowth, tangled lowland forest with bamboo, usually on or near the ground, also in trees. They live in the understory in dense undergrowth and are associated with climbing vines and bamboo thickets. Occur both in primary forest and in selectively logged forest (Poulsen 1995). On Mt Isarog they were seen between 550 and 760 m in 1961, but they were not seen there in 1988 (Goodman and Gonzales 1990). The behavior as it moves about is more like a malkoha than a typical coucal. Seen singly, in pairs and in groups of up to 10 to 12 birds, noisy and walking away from disturbance; once seen following a mixedspecies flock of birds (Poulsen 1995, Kennedy et al. 2000).

Food Unknown.

Green-billed Coucal Centropus chlororhynchos 215

Breeding A female had an egg ready to lay on 14 April, and a juvenile with a short wing and tail and with hair-

like down on the head in late November is from an October nesting (FMNH). The nest and eggs are unknown.

Green-billed Coucal Centropus chlororhynchos Blyth, 1849 Centropus chlororhynchos Blyth, 1849, Journal of the Asiatic Society of Bengal 18, pt. 2, p. 805. (Ceylon) Note: Often spelled Centropus chlororhynchus, the original description is Centropus chlororhynchos. Other common name: Ceylon Coucal, Sri Lanka Green-billed Coucal Monotypic.

Field characters

Description

Voice

ADULT: Sexes alike, upperparts and underparts dull black, neck and tail glossed purple, the back, wing coverts and wing chestnut in no great contrast to the upperparts, tail black, wing lining black; iris red to reddish brown, bill ivory to pale green, legs and feet black.

Song is a resonant, sonorous, low-pitched two or three notes “hoop-poop-poop,” the last note lower in pitch, the song 0.3–0.35 kHz, given at 2 notes per sec, the calls often given as a duet; also a short double syllable “hu, hu,” or “hoooooEE” and a single cough “chewkk” (Hoffmann 1989a,b, Grimmett et al. 1999,Wijesinghe 1999).

JUVENILE: Plumage like adult’s, but wing coverts indistinctly barred black; iris gray, bill dark greenish with base and culmen gray, legs and feet dusky gray. NESTLING: Skin black, white hair-like down on head and neck, bill with an ivory-colored stripe on bill from lores to tip of upper mandible, lower mandible black (Wijesinghe 1999). SOURCES: ANSP, BMNH, MCZ,YPM, ZMB.

Measurements Wing, M (n ⫽ 7) 158–173 (167.3 ⫾ 4.9), F (n ⫽ 7) 168–184 (174.6 ⫾ 7.7); tail, M 222–252 (234.3 ⫾ 8.8), F 225–254 (239.3 ⫾ 10.7); bill, M 36–41 (38.9 ⫾ 1.7), F 38.9–47.0 (42.0 ⫾ 2.9); tarsus, M 40–46 (43.3 ⫾ 2.1), F 42–47 (42.9 ⫾ 2.6), hallux claw, M 15–18 (17.1 ⫾ 1.1), F 17–21 (18.1 ⫾ 1.5) (BMNH). Wing formula, P6 ⫽ 5 ⫽ 4 ⬎ 3 ⬎ 2 ⬎ 7 ⬎ 8 ⬎ 1 ⬎ 9 ⬎ 10.

Overall length 43–46 cm. Dull blackish head, body and tail, and dark rufous wings. On Sri Lanka, distinguished from the larger Greater Coucal C. sinensis by pale green rather than black bill, by its dark wings, and by its shorter deep call lacking the water-bottle phrase; occurs in wet forest habitat.

Range and status Sri Lanka. It occurs from the center (Dambulla to Ambepussa and Kandy) to the southwest. Observations since 1980 are from Amanawala Ampane, Kitulgala, Htaramune and Dehiowita to Morapitiya Forest Reserve, Neluktiya Mukalana,

216 Black-faced Coucal Centropus melanops Ingiya Forest Reserve, Peak Wilderness Area, Gilimale forest, Sinharaja Forest Reserve, and Nakiyadeniya and other sites in this region (BirdLife International 2001). Resident. Local forests are separated by extensive areas of cultivation and coffee plantations, habitats that may discourage dispersal between forest sites.The total population is perhaps a few thousand individuals. The most securely protected forest site that supports these coucals is on the lower edge of Peak Wilderness Sanctuary in Sinharaja National Heritage Wilderness Area, Ratnapura District. Globally threatened, endangered by loss and fragmentation of forest habitat as the coucals’ home is converted into agricultural land and the remaining forests are cut and cleared of undergrowth (BirdLife International 2001).

Food Omnivorous, they take beetles, grasshoppers, termites, spiders; snails, ammonites, worms; frogs, small lizards including skinks, and snakes as long as 60 cm, and they also feed on plant material (Legge 1880, Hoffmann 1989a,b,Wijesinghe 1999).

Breeding behavior Courtship behavior has not been described, nor has courtship feeding, although the male brings food to the nest while the female incubates, and the male is said to remove her fecal sacs from the nest. The adult bill color changes from whitish to green through the nesting period (Wijesinghe 1999).

Breeding Habitat and general habits Humid high evergreen forest with dense undergrowth, dense scrub with bamboo and rattan cane rushes in disturbed areas, tangled thickets, wooded river banks, occur in wet zone forests west, southwest and south of the central mountain massif; low country to 760 m (Legge 1880, Lewis 1898, Wait 1925, BirdLife International 2001). The coucal is closely associated with the bamboo Ochlandra stridula, which grows in wet places, and the Sinhala name for this bamboo is bata aetikukula, or “coucal bamboo” (Wijesinghe 1999). Occasionally appears in patches of abandoned slash-and-burn agriculture close to forest and in other disturbed areas with tangled vegetation (BirdLife International 2001). The coucals feed on the ground and in trees and creepers. In the morning they perch in the open and spread their wings to the sun (Wijesinghe 1999).

Prolonged season with nesting records in January, April to July, September and November. Nest is a covered mass of twigs, roots and grass, with a side entrance, lined with green or dried leaves especially bamboo. The nest is built in thorn canes, bamboo or in a tree, near the ground to as high as 3.6 m above ground in a tree, especially in Wendlandia bicuspidata (Rubiaceae). Both sexes build the nest, which measures 60 c m ⫻ 45 cm. Eggs are chalky white, 35 ⫻ 27 mm, the clutch is 2(3). Incubation period 17 days or shorter; nestling period 12 days or longer (hatching day was not determined, Wijesinghe 1999). Both parents feed the young. The young fledgling flutters the wings and tail when it is fed. It returns to roost in the nest for the first two days after it leaves the nest (Wait 1925, Baker 1934, Wijesinghe 1999, BirdLife International 2001).

Black-faced Coucal Centropus melanops Lesson, 1830 Centropus melanops Lesson, 1830, Traité d’Ornithologie, livre 2, p. 137. (“Java” [⫽ Mindanao]) Monotypic.

Description ADULT: Sexes alike, center of crown buffy white, neck darker pinkish buff, upper back rufous, wing

rufous with dark tips to primaries and secondaries, lower back and upper tail coverts black, tail black, face whitish with a black mask from forehead around eye to base of lower mandible; underparts, chin and throat buffy white, breast buff, lower breast and belly to under tail coverts black; iris red, bill black, feet black.

Black-faced Coucal Centropus melanops 217 JUVENILE: Like adult, rectrices narrow. NESTLING: Undescribed. SOURCES: AMNH, CM, CMNH, FMNH, MVZ, RMNH, ROM, UMMZ, ZMUC.

Geographic variation A subspecies C. m. banken Hachisuka 1934 was described from Bohol, Leyte and Samar, with more black on the face, but these do not differ consistently between birds on these islands and birds elsewhere in the Philippines (Rand and Rabor 1960, Parkes 1971; AMNH, FMNH).

Measurements and weights Wing, Mindanao: M (n ⫽ 8) 152–167 (162.0 ⫾ 4.1), F (n ⫽ 5) 158–172 (162.0 ⫾ 5.4); tail, M 222–252 (237.7 ⫾ 10.6), F 215–270 (243.8 ⫾ 20.2); bill, M 30–37 (34.6 ⫾ 2.9), F 33–36 (34.1 7 ⫾ 1.2); tarsus, M 34–47 (42.0 ⫾ 5.5), F 39–44 (41.3 ⫾ 1.7), hallux claw, M 15.5–18 (16.7 ⫾ 0.9), F 14–17 (16.0 ⫾ 1.2) (CMNH, ZMUC); Weight, M (n ⫽ 10) 197.5–237.3 (211.1), F (n ⫽ 12) 214–265.5 (237.9) (Rand and Rabor 1960; AMNH, CMNH, USNM). Wing formula, P6 ⫽ 5 ⫽ 4 ⬎ 7 ⬎ 3 ⬎ 8 ⬎ 2 ⬎ 1 ⬎ 9 ⬎ 10.

Field characters Overall length 42–48 cm. Coucal with buff foreparts and a black face with whitish crown streak running forward to the bill, rufous back and wings and a black tail.

Voice Call is a distinctive loud booming “wooop wooop wooop wooop wooop”, the first note longer and lasting 0.5 sec, then a pause of a second, the

other notes becoming softer, the series lasting about 5 sec. Other calls are a single “wooop” and a descending “boop boop boop boop” (Kennedy et al. 2000).

Range and status Philippines (Basilan, Mindanao, Nipa, Dinagat, Sirgao, Bohol, Leyte, Samar). Resident. Common in Mindanao 60 years ago, its forests are nearly gone (Hachisuka 1934, Dickinson et al. 1991).

Habitat and general habits Forests, in treetops, second growth; lowlands to 1200 m.

Food Unknown.

Breeding April (oviduct egg).The nest is undescribed. Eggs are white, fine texture, not glossy, 31 ⫻ 26 mm (Rand and Rabor 1960) or 37 ⫻ 29 mm (Schönwetter 1964). Incubation and nestling periods are unknown.

218 Black-hooded Coucal Centropus steerii

Black-hooded Coucal Centropus steerii Bourns and Worcester, 1894 Centropus steerii Bourns and Worcester, 1894, Occasional Papers of the Minnesota Academy of Natural Sciences, 1, p. 14. (Mindoro) Other common names: Steere’s Coucal. Monotypic.

Voice

Description

Range and status

ADULT: Sexes alike, head black glossed blue green, hackles on head and throat black, neck, back and wing sooty brownish black, tail glossy greenish black, chin and throat black, breast and belly dark brown; iris red to brown, bill black, legs and feet black.

Philippines (Mindoro) (Dickinson et al. 1991). Resident. Rare and near extinction as a result of loss of their lowland primary forest. Where forest habitat has been degraded, these coucals are replaced by the Philippine Coucal C. viridis. Blackhooded Coucals are known from recent observations in four sites: Puerto Galera, MUFRC Experimental Forest (the only inland site), Malpalon (where rough ground has given the forest respite from clearing and cultivation) and Siburan (where the Sablayan Penal Colony has provided restricted human access into the forest, although prisoners now disturb the forest undergrowth collecting rattan and logging for furniture production; and refugees who were resettled to Mindoro from their homes in Luzon in 1991–1993 during a period of volcanic activity on Mt Pinatubo have had an impact on the forest as well). The conservation status is critical, as this coucal is rare and lives in very small fragmented forest sites, which are few and mostly not effectively protected. It is the next candidate cuckoo for global

JUVENILE: Plumage above brown, soft-webbed and lacking hackles, tail dull black, chin and throat brownish gray; bill pale and depth less than in adult. NESTLING: Undescribed. SOURCES: AMNH, ANSP, BMNH, CM, FMNH, SMF, YPM.

Measurements and weights Wing, M (n ⫽ 8) 152–171 (159.4 ⫾ 5.7), F (n ⫽ 9) 157–170 (164.4 ⫾ 5.0); tail, M 217–253 (238.6 ⫾ 11.5), F 232–277 (251.3 ⫾ 16.6); bill, M 35–40 (36.9 ⫾ 1.9), F 35–41 (37.1 ⫾ 1.9); tarsus, M 35–45 (39.6 ⫾ 3.4), F 38–45 (43.2 ⫾ 3.7); hallux claw, M 16–17 (16.5 ⫾ 0.5), F 16–18 (16.8 ⫾ 1.2) (AMNH, BMNH,YPM, SMF). Weight, M (n ⫽ 4) 160–200 (179), F (n ⫽ 7) 190–238 (163.4) (BMNH, YPM, Ripley and Rabor 1958. Wing formula, P5 ⫽ 4 ⬎ 3 ⬎ 7 ⫽ 6 ⬎ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 46 cm. An all dark coucal with head black, back and wing fuscous, and a glossy greenish black tail. Larger and stockier than the black form of Philippine Coucal C. viridis which also lives on Mindoro; plumage sooty brown not glossy black.

Resonant, deep booming notes, “hoot hoot hoot hoot . . .”, 5 to 9 or more notes in a series, the first note louder and longer, the final notes descending slightly and becoming softer (Kennedy et al. 2000).

Short-toed Coucal Centropus rectunguis 219 extinction (Dutson et al. 1992, Brooks et al. 1995, BirdLife International 2001).

Habitat and general habits Primary old growth dipterocarp forest in lowlands, and in mountains to 800 m in transition dipterocarp-mid-mountain forest. The birds live in primary forest on rocky ridges with small forest openings with bamboo and creepers, in tangled thickets of vines and rattan, and streamsides within primary forest, all minimally disturbed forest habitats (Bourns and Worcester 1894, McGregor 1905,

Kennedy et al. 2000, BirdLife International 2001). Secretive, they move slowly through dense vines and foliage, in understory and forest canopy; they run along tree branches and spend time in thick tangles of forest vines and creepers.

Food Unknown, save for a specimen noted to have “larger insects” (BMNH).

Breeding Unknown.

Short-toed Coucal Centropus rectunguis Strickland, 1847 Centropus rectunguis, Strickland, 1847, Proceedings of the Zoological Society of London, 1846, p. 104. (Malacca)

Description ADULT: Sexes alike, above especially the nape glossy purplish-blue black, shafts of hackles black, wing chestnut, tail black, underparts black, under wing coverts black; iris red, bill black, legs and feet black. Hallux claw short and straight, not long as in most coucals. No eclipse plumage. JUVENILE: Head above chestnut rufous with brown shaft streaks, back rufous with indistinct blackish bars, wing coverts chestnut with black bars, flight feathers unbarred chestnut with no dark on the edges or tips, tail black with fine white bars; chin to belly dark brown barred with dull white, V-shaped bars on each feather of throat and breast and fine wavy bars on the belly (some birds are nearly black below, others are brown or buff ), under wing coverts black; iris gray, bill blackish horn to brown, paler below, legs and feet blackish. NESTLING: Undescribed. SOURCES: AMNH, BMNH, ZRC.

Measurements and weights Wing, M (n ⫽ 4) 156–170 (165.5), F (n ⫽ 2) 166–186; tail, M 192–204 (198), F 194–238 (216);

bill, M 35–37 (36), F 34–42 (37); tarsus, M 42–46 (44.3), F 45–52 (48.5); hallux claw, M 8–12 (10.0), F 9–19 (14) (AMNH, BMNH, ZRC). Weight, F? (a large unsexed bird) (n ⫽ 1) 237.5 (Wells 1999). Wing formula, P6 ⫽ 5 ⫽ 4 ⫽ 3 ⬎ 2 ⬎ 1 ⬎ 7 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 38–43 cm. Black plumage with uniformly rufous wing and black tail, wing and tail shorter than Greater Coucal C. sinensis, and hallux claw short and nearly straight.

Voice Call, deep resonant booming notes on a descending scale, “whu, huup-huup-huup-huup” at a pitch of 0.3 kHz, given 2 notes per sec, each note lasting 0.25 sec; and a rapid series of deep rising notes heard at dusk. A snorting squirrel-like “hut, hut, hut” and an explosive scold “jeézaw” may also be from this coucal (Smythies 1981, Scharringa 1999, Wells 1999, Robson 2000a, Sheldon et al. 2001).

Range and status Malay Peninsula, Sumatra and Borneo. Exact distribution is not well known and many published records are questionable, especially reports in upland habitats where observations may be of the lookalike Larger Coucal Centropus sinensis (BirdLife

220 Short-toed Coucal Centropus rectunguis

Habitat and general habits

International 2001). Many museum specimens registered as C. rectunguis in fact are C. sinensis (e.g., specimens at MZB and RMNH). In peninsular Thailand C. rectunguis are reported at Bala, Khlong Hala (Robson 2000b), and in peninsular Malaysia in 18 localities including Belum Forest Reserve, Taman Negara NP, Tumboh FR, Tekam FR, Dusun Tua, Subang FR, Pasoh FR, Sungai Kinchi, EndauRompin Conservation Area and Rengam FR (Wells 1999, BirdLife International 2001). In Sumatra they are reported at Barat in Padang highlands and southern Barisan Range, Jambi in Ketalo and Tembesi Rivers, and Selatan in Barisan Range (Salvadori 1879, Büttikofer 1887, van Marle and Voous 1988); but most specimen records are old and are unsupported by date, and the upland records are questionable because these coucals are lowland birds (BirdLife International 2001). In Borneo most records are in lowland Sarawak and Brunei, they also occur in Sabah (where they occur in lowlands from sea level to 400 m, as at Danum Valley Conservation Area) and a few lowland sites in Kalimantan (Kutai NP, Sungai Sebangau, Gunung Palung NP) (BirdLife International 2001, Sheldon et al. 2001).They do not occur on nearshore islands. Resident.The coucals are fairly common at low densities in tall lowland forest south of 5o N latitude in the Malay Peninsula (Wells 1999). Their conservation status is considered vulnerable, due to their rapidly disappearing lowland mature forest habitat (BirdLife International 2001).

Ground and understory of lowland closed canopy forest (broad-leafed alluvial forest, peat swamp forest and riverine forest). Occur in lowlands, and sometimes in nearby hills to 600 m. In the Malay Peninsula they are found in mature forest, with a few records from recently logged forest mainly in the lowlands below slopes where logging is less complete. In Sumatra they live in undisturbed and selectively logged primary forest and forest edges. In Sabah (Gunung Mulu) and Kalimantan (Gunung Palung) they occur in lowland and alluvial forest (not swamp or upland forest) (Wells 1976, Laman et al. 1996). Sight and sound records in open habitats in scrub, padi fields and tall grass perhaps apply to Greater Coucal Centropus sinensis (Johns 1989, Holmes 1995, Sheldon et al. 2001). Resident. Solitary and terrestrial, Short-tailed Coucals live on the ground and under-stratum of lowland forest.

Food Few records are known, the coucals eat insects and frogs and take birds from mist nets (Wells 1999, BirdLife International 2001).

Breeding In the Malay Peninsula the coucals call mainly from November to March and also in May and August, there is a nest record in September, and dependent fledglings with an adult were seen in May (Chasen 1939, Medway and Wells 1976, Wells 1999). In Sumatra, a nestling was found in March (van Marle and Voous 1988). The nest is an untidy covered structure of leaves, some still attached to twigs, the nest lined with teased palm-frond pith, built in fronds of a stemless Salacca palm, 2 m above ground in mature forest (Wells 1999). Eggs are white, 37 ⫻ 30 mm (Schönwetter 1964). Clutch size, incubation and nestling periods are unknown. Both parents attend to the nest (Wells 1999).

Bay Coucal Centropus celebensis 221

Bay Coucal Centropus celebensis Quoy and Gaimard, 1830 Centropus celebensis Quoy and Gaimard 1830, Observations zoologiques faites à board ‘de l’Astrolabe , 1, 1, p. 230; Atlas, Oiseaux, pl. 20. (Manado, Celebes) Polytypic. Two subspecies. Centropus celebensis celebensis Quoy and Gaimard, 1830; Centropus celebensis rufescens (Meyer and Wiglesworth, 1896). Other names: Pyrrhocentor celebensis (Quoy and Gaimard 1830).

(34.6 ⫾ 1.9), F 33–41 (38.1 ⫾ 3.0); tarsus, M 36–44 (40.6 ⫾ 2.2), F 38–50 (41.6 ⫾ 3.9), hallux claw, M 16–20 (17.6 ⫾ 1.3), F 17–21 (18.4 ⫾ 1.3), the hallux and hallux claw are short and curved, one M (SMF 28232) had a straight hallux claw (AMNH, BMNH, SMF). Wing formula, P7 ⬎ 6 ⬎ 5 ⬎ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10.

Description

Field characters

ADULT: Sexes alike, head and upper back buffy gray, shaft streaks buff, lower back and rump rufous chestnut, wing and tail rufous chestnut, breast and belly gray to rufous chestnut, under tail coverts rufous chestnut; iris red, bill black, legs and feet black.

Overall length 44–50 cm. Coucal with unmarked light gray to brown plumage, wings and tail dark rufous, bill black.

JUVENILE: Like adult, throat and breast paler, rectrices narrow; iris light gray, lower mandible and tip of upper mandible pale. NESTLING: Undescribed. SOURCES: BMNH, MSNG, MZB, ROM, SMF, UMMZ, ZMUC, ZSM.

Subspecies Centropus celebensis celebensis Quoy and Gaimard, 1830; crown and upper back buffy olive gray, breast dark gray; N Sulawesi; Centropus celebensis rufescens (Meyer and Wiglesworth, 1896); crown buffy brown to rufous, breast dark rufous; C, E, S and SE Sulawesi, Labuan Blanda, Muna and Butung. White and Bruce (1986) considered this coucal to form a superspecies with Rufous Coucal Centropus unirufus, but its song is more like that of Black-faced Coucal C. melanops. C. celebensis, C. unirufus and C. melanops lack the long straight hallux claw of most other coucals.

Measurements Wing, M (n ⫽ 10) 164–184 (173.2 ⫾ 7.1), F (n ⫽ 8) 172–187 (181.6 ⫾ 5.9); tail, M 250–302 (267.1 ⫾ 14.3), F 282–333 (296.0 ⫾ 22.7); bill, M 31–38

Voice Calls include (1) a series of “woop” notes; (2) a series of deep “hoo” notes that accelerate; then the mate joins and the pair duets together, one decreasing the pitch then slowing the phrases and raising the pitch again, the “water bottle” song (Meyer 1879, Holmes and Phillipps 1996). Solitary birds give a repeated “wheeze” (Watling 1983, NSA; Smith 1993a).

Range and status Sulawesi and neighboring islands of Labuan Blanda, Muna and Butung. Resident (Riley 1924, Stresemann 1924, van Bemmel and Voous 1951,White and Bruce 1986).These birds occasionally fly across small treefall gaps in the forest (Coates and Bishop 1997), although larger gaps may be barriers to local dispersal. Sparsely distributed, locally common, they are

222 Gabon Coucal Centropus anselli difficult to observe in dense vegetation.They live in Tangkoko Nature Reserve to 600 m, in Gurung Rurukan to 800 m, in Tentolo-Matinan Mts to 600 m, in Lore Lindu NP to 1000 m, in the Mekonga Mts to 550 m and on the Lompobattang Massif to 1100 m (Coates and Bishop 1997); they are locally common in Dumoga-Bone NP (Rozendaal and Dekker 1989). Rattan, the key habitat factor to the coucals’ survival, is rapidly being stripped from the coucals’ forests.

climbing through lianas and thick foliage (Watling 1983). On Sulawesi they occur in more forested habitat than the Lesser Coucal C. bengalensis (Rozendaal and Dekker 1989).They sometimes forage in small groups (Watling 1983), and they often associate with Yellow-billed Malkoha Rhamphococcyx calyorhynchus and troops of macaques Macaca nigra (Heinrich, in Stresemann 1940,Wardill et al. 1998).

Habitat and general habits

Fruit, including nutmeg (Meyer 1879); large insects, orthoptera, locusts, beetles; and spiders (Stresemann 1940).

Lowland and lower montane rain forest, forest edge and dense regrowth, in lianas and rattan thickets, ranges into secondary lowland and hill forest, open woodland and scrub, also mangrove forest (Rozendaal and Dekker 1989, Coates and Bishop 1997). Mainly in lowland forest, occur from sea level to 1100 m. Arboreal, they live in the undergrowth and lower forest canopy. In behavior this coucal is more like a malkoha than a typically terrestrial coucal,

Food

Breeding A female had an enlarged ovary in February (AMNH 298842). Nest is built of brushwood, flat and open not covered or domed, above ground on trees in forests (Meyer 1879). The eggs, incubation and nestling periods are unknown.

Gabon Coucal Centropus anselli Sharpe, 1874 Centropus anselli Sharpe, 1874, Proceedings of the Zoological Society of London 1874, p. 204, plate 33, fig. 1. (Danger River, Gabon) Monotypic.

Description ADULT: Sexes alike, plumage dark, above head and upper back black glossed purplish, lower back dark brown, wing rufous with darker tips, inner secondaries dark brown, lower back and rump buffy white barred black, upper tail coverts shaggy and long, barred black and buff, tail black with a bronze gloss, T1 narrowly barred buff, the outer tail feathers T2 to T5 unmarked black, hackle shafts black above to ochre on chin to breast, face and side of neck black to well below the eye, chin to belly and under tail coverts rufous buff, under wing coverts rufous buff; iris red to reddish brown, some females with a whitish outer ring, bill black, feet black. Nonbreeding plumage, restricted to the first year, above the head and nape to upper back dark brown with

pale inconspicuous shaft streaks, back unbarred dark brown, face dark brown with irregular buff spots, throat, breast and belly rufous buff with narrow dark brown bars, wing barred as in juvenile. Birds molt into this barred plumage from the juvenile plumage. JUVENILE: Head brownish black with fine pale shaft streaks, upper back black with black shaft streaks, lower back dark brown, wing coverts and wing rufous with dark brown bars, inner secondaries dark brown, rump and upper tail coverts barred buffy white and dark gray, tail black with buff bars, throat and upper breast rufous buff with fine black bars and shafts have buff and black bars, belly unbarred buff, under tail coverts rufous buff with indistinct blackish bars, under wing coverts rufous buff;iris gray brown,bill dark gray above and pale gray below, feet slate to bluish gray. NESTLING: Undescribed. SOURCES: AMNH, BMNH, FMNH, UMMZ, ZFMK.

Gabon Coucal Centropus anselli 223

Measurements and weights Wing, M (n ⫽ 11) 178–197 (186.0 ⫾ 7.0), F (n ⫽ 14) 180–209 (194.8 ⫾ 11.8); tail, M 242–302 (266.2 ⫾ 24.5), F 280–380 (302.2 ⫾ 22.7); bill, M 30–37 (33.1 ⫾ 1.9), F 31–38 (34.8 ⫾ 2.2); tarsus, M 40–48 (44.2 ⫾ 2.2), F 45–52 (48.9 ⫾ 2.3); hallux claw, M 17–18 (17.2 ⫾ 0.5), F 16–20 (17.4 ⫾ 1.3) (AMNH, FMNH). Weight, M (n ⫽ 1) 210 (FMNH). Wing formula, P5 ⫽ 4 ⫽ 3 ⫽ 2 ⬎ 1 ⬎ 6 ⬎ 7 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 46–58 cm.A coucal of central African forest, with crown and sides of head black, back dark brown and underparts rufous buff. Juvenile is very similar to juvenile Black-throated Coucal Centropus leucogaster neumanni, the head more brown and the feathers with fine streaks not spots.

Voice Fast call is a trill, more rapid and more modulated than that of other large African coucals; the note starts at 0.5 kHz then jumps to 0.7 kHz and slides down to 0.4 kHz, repeated in a trill on one pitch at 7–8 notes per sec, slowing to 5 notes per sec near the end (Chappuis 2000). Another call is a slower series of loud, deep notes at 0.6 kHz, the note 0.2–0.3 sec long and repeated 3 in a second, a melancholy bass “ouh ouh ouh ouh . . .”, like the call of the Black-throated Coucal C. leucogaster. Calls are delivered with head lowered and throat puffed out, and pairs call in duet, one bird giving the trill and the other the longer notes (Heinrich 1958, Chappuis 1974, 2000, Irwin 1988, Dowsett and Dowsett-Lemaire 1993, Christy and Clarke 1994).

1997), and an isolated population occurs in northwestern Angola (Dean 2000). Resident. Frequent to common.Territories are a few hectares in area, in dense vegetation where the birds are hard to observe. Population density is c. 3–5 pairs / 100 ha in Gabon, where territories are exclusive to Blueheaded Coucal C. monachus (Brosset and Erard 1986). The range complements that of Blackthroated Coucal C. leucogaster.

Habitat and general habits Primary forest, undergrowth in swampy forest, forest second growth, forest edge along rivers, old cultivation, grassy swamps. Terrestrial, active, they feed mainly on or near ground and they scavenge around camps and villages (Bannerman 1933, Chapin 1939, Irwin 1988, Louette 1986).

Food Omnivorous, taking insects (grasshoppers, katydids, beetles), mollusks, frogs, small snakes, small birds (Bates 1930, Chapin 1939, Mackworth-Praed and Grant 1970, Brosset and Erard 1986, Irwin 1988, AMNH).

Range and status Central Africa from southern Cameroon (Ebolowa, Sangamélima, Abong Mbang, Bitye on River Dja) (Good 1952, Louette 1981a; AMNH, BMNH, CM, FMNH, UMMZ), southwestern Central African Republic (Dzanga-Ndoki NP) (Green and Carroll 1991), Gabon, Congo (north to Nouabele-Ndoki) and central Zaire (Chapin 1939, Snow 1978, Dowsett and Dowsett-Lemaire

Breeding Laying begins with the rains or in the little dry season in the wettest forest region. In NE Gabon, coucals lay in December and February (Brosset and Erard 1986), young are seen in Uele in November (Chapin 1939) and a female had an egg nearly ready to lay in Angola in March (Heinrich 1958; FMNH 224380).The nest and eggs are unknown.

224 Black-throated Coucal Centropus leucogaster

Black-throated Coucal Centropus leucogaster (Leach, 1814) Polophilus leucogaster Leach, 1814, Zoological Miscellany, 1, p. 117, pl. 52. [“New Holland” ⫽ Gold Coast Colony] Other common names: Great Coucal Polytypic. Three subspecies. Centropus leucogaster leucogaster (Leach, 1814); Centropus leucogaster efulensis Sharpe, 1904; Centropus leucogaster neumanni (Alexander 1908).

Centropus leucogaster efulensis Sharpe, 1904; black plumage glossed blue to green, inner secondaries dark olive brown; SW Cameroon; Centropus leucogaster neumanni (Alexander 1908); plumage as in efulensis, smaller; NE Zaire and extreme W Uganda. The small eastern form neumanni has also been considered a distinct species, more closely related to C. anselli than to leucogaster (Louette 1986).

Description ADULT: Sexes alike, above head and upper back black, lower back, rump and upper tail coverts black barred buff, feathers broad, scallop-shaped (not lanceolate as in the other African coucals), wing and wing coverts rufous chestnut, flight feathers darker at tip, tail black glossed blue, often barred buff, underparts black with the lower breast and belly white with the feathers tipped buff, flanks and under tail coverts rufous buff; iris red, bill black, feet black to blue gray. JUVENILE: Head and upper back blackish, feathers spotted with buff base and short buff shaft streaks, shafts black with a subterminal bar of yellowishwhite rump and upper tail coverts black with fine buff bars, wing rufous with dark brown bars, tail black with narrow buff bars, more extensively barred than in adult, throat and upper breast black with small buff spots, lower breast to belly and under tail coverts unbarred whitish to rufous buff; iris gray or brown, bill blackish above, lower mandible horn. NESTLING: Upperparts with dense white hair-like down; tongue has a black U-shaped mark (AMNH, Chapin 1939). SOURCES: AMNH, BMNH, CM, FMNH.

Subspecies Centropus leucogaster leucogaster (Leach, 1814); black plumage glossed violet blue, inner secondaries dark rufous; West Africa from southern Senegal and Guinea-Bissau to Ivory Coast, Ghana and SE Nigeria;

Measurements and weights C. l. leucogaster: Wing, M (n ⫽ 8) 182–207 (190.3 ⫾ 7.4), F (n ⫽ 8) 187–213 (198.4 ⫾ 8.6); tail, M 244–320 (276.5 ⫾ 23.0), F 262–310 (306.0 ⫾ 20.0); bill, M 31–46 (35.1 ⫾ 5.0), F 34–41 (37.1 ⫾ 2.6); tarsus, M 39–46 (45.1 ⫾ 3.6), F 46–54 (51.2), hallux claw, M 15–22 (17.8 ⫾ 2.4), F 16–22 (19.2 ⫾ 2.3) (AMNH, BMNH, FMNH, RMNH); C. l. efulensis: Wing, M (n ⫽ 14) 180–204 (188.4 ⫾ 6.8), F (n ⫽ 6) 197–223 (207.3 ⫾ 10.7), tail, M 254–291 (274.3 ⫾ 10.7), F 240–336 (288.8 ⫾ 33.3) (AMNH, BMNH, FMNH); C. l. neumanni: Wing, M (n ⫽ 3) 174–182 (178.7), F (n ⫽ 1) 176; tail, M 241–268 (252), F 250 (AMNH, BMNH); wing, U (n ⫽ 36) 165–198.5 (179.9) (Louette 1986). Weight, C. l. efulensis: M (n ⫽ 1) 270 (Eisentraut 1963); C. l. neumanni: M (n ⫽ 1) 293; F (n ⫽ 2) 327–346 (336) (BMNH). Wing formula, P6 ⬎ 5 ⫽ 7 ⬎ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 46–58 cm. The largest African coucal, rufous chestnut above with a black head, mantle and throat and a white belly (underparts rich buff in Gabon Coucal C. anselli).

Voice Series of 8–20 deep “hoo” notes given 4 per sec at 0.4 kHz, held on one pitch, a falling and rising song with the series slowing in tempo and falling in

Black-throated Coucal Centropus leucogaster 225 pitch; calls at night (Bates 1930, Chapin 1939, Chappuis 1974, 2000, Irwin 1988).

Range and status West and Central Africa from southern Senegal (Casamance) and Sierra Leone to Togo, in southern Nigeria and Cameroon, and a disjunct population neumanni in NE Zaire and western Uganda. Resident. Fairly common in forests of lower Casamance, locally common in Liberia, uncommon in Ghana, rare in Togo, and common in the rivers area of southern Nigeria (Chapin 1939, Lamarche 1980, Louette 1986, Irwin 1988, Morel and Morel 1990, Dowsett and Dowsett-Lemaire 1993, Elgood et al. 1994, Rodwell 1996, Barlow et al. 1997, Gatter 1997; BMNH, CM, FMNH). In southern Cameroon it occurs in Korup NP, Kumba, Mt Kupé and Mt Cameroon (Buea, Victoria) and from Yaounde to the lower Sanaga River (Edea), inland at Lolodorf, and south to Kribi and Efulen, while Gabon Coucal C. anselli occurs further south and east (Good 1952, Louette 1981a,Thomas 1991, Bowden 2001); the two closely approach each other with C. leucogaster at Change, 20 miles NW of Ebolowa, and C. anselli at Ebolowa (Good 1952, FMNH). A report of C. leucogaster in the Mts Mandigues in Mali (Lamarche 1980) is questioned (R. J. Dowsett, in Lachenaud 2003); a report in Niger near Niamey (Debout et al. 2000) was rejected (Demey et al. 2001), the habitat there is not forest and the species has never been seen in Parc W (Crisler et al. 2003; C. Jameson, pers. comm.); and a provisional report of C. leucogaster in Gabon (Bannerman 1933) was rejected (Dowsett and Dowsett-Lemaire 1993).

C. l. neumanni occurs in Zaire north of the equator on the Mbomou River, Uele River and the right bank of the Congo River and in western Uganda (Bannerman 1919, Chapin 1939, Snow 1978, Louette 1986, Rossouw and Sacchi 1998).

Habitat and general habits Forest. Dense undergrowth in forest edge, forest remnants, dense gallery forest, secondary forest, thickets, dense grass, tall reeds and sedges; lowland forest zone, dense and swampy areas, along logging roads and Raphia swamps. In dense gallery forest and thick second growth more than in primary forest. In Cameroon they call on farms above Nyasoso bordering the Mt Kupé forests (Bowden 2001). In Uganda they live in Semliki NP (Rossouw and Sacchi 1998). Mainly in lowlands, to 1000 m in Liberia in montane ravines.Terrestrial, they feed on or near the ground (Chapin 1939, Eisentraut 1973, Irwin 1988, Gatter 1997).

Food Insects, including caterpillars, beetles, grasshoppers, mantids, bugs; also spiders, terrestrial snails and frogs (Bannerman 1933, Chapin 1939, Irwin 1988, Gatter 1997).

Breeding In Liberia from November to February with young attended to by parents in March, independent in April; birds call from August to November and are quiet in April; in Ghana a laying bird was caught at Kumasi in August; in Cameroon they breed from June to November, in Zaire (Uele and Ituri Districts) from March to December (Bannerman 1919, Chapin 1939, Bannerman 1951, Serle 1954, Grimes 1987, Irwin 1988, Gatter 1997). The nest is a large ball of leaves and grass, lined with green leaves, placed in a bush up to 30 cm above ground, in forest or in tall grass (Chapin 1939). Eggs are white, elliptical, of unknown size (the record in Mackworth-Praed and Grant 1970 is questioned, M. P. Walters). Clutch is 2. Incubation and nestling periods are unknown.The nest is tended by male (and female?) (Chapin 1939).

226 Senegal Coucal Centropus senegalensis

Senegal Coucal Centropus senegalensis (Linnaeus, 1766) Cuculus senegalensis Linnaeus, 1766, Systema Naturae ed. 12, 1, p. 169. (Senegal) Polytypic. Three subspecies. Centropus senegalensis senegalensis (Linnaeus, 1766); Centropus senegalensis aegyptius (Gmelin, 1788); Centropus senegalensis flecki Reichenow, 1893. Other common names: Rufous-bellied Coucal.

Description ADULT: Sexes alike (female slightly larger), crown and neck black with gloss, feather shafts stiff and shiny black, upper back and wing rusty brown, rump unbarred black, tail blackish with green gloss; underparts white, feather shafts of breast stiff and yellowish, flanks finely barred blackish, under tail coverts buff, under wing coverts light rufous; iris red, bill black, feet gray.A dark rufous colored plumage phase “epomidis” occurs in humid parts of West Africa (Liberia, Ivory Coast, Ghana, Nigeria; Serle 1950a, Elgood 1955, 1973, Louette 1981b, Grimes 1987, Elgood et al. 1994, Gatter 1997, AMNH, BMNH, RMNH) with the crown, face and throat dark brown to black, belly, flanks and under tail coverts all dark rufous. In Nigeria about half the birds are this of plumage form near the coast, with the proportion in rufous plumage decreasing away from the coast, and none further than 200 km from the coast. JUVENILE: Crown dark brown or gray, back rusty brown barred dark brown, wing and wing coverts rufous barred dark brown, primaries unbarred except for dark at tips, secondaries barred dark brown, upper tail coverts long, blackish barred buff, tail blackish with faint buff bars at tip, throat and breast buffy white with prominent shafts straw to brown, belly buffy white without the glossy shafts, flanks buffy barred blackish, under tail coverts dark buff, under wing coverts light rufous, iris olive brown to yellow brown. NESTLING: At hatching, skin black above, dark pink below, with long white hair-like down; bill pale pink, gape pink, tongue red, iris gray, legs gray.

SOURCES: AMNH, BMNH, BWYO, CM, FMNH, MVZ, ROM, UMMZ, USNM, ZMUC.

Subspecies Centropus senegalensis senegalensis (Linnaeus, 1766); above, head black glossed green, back rufous; Senegal and The Gambia to Nigeria, Chad, Somalia, Uganda and W Kenya, south to Zaire; Centropus senegalensis aegyptius (Gmelin, 1788); sootier, crown dull brownish black, back brown not rufous, larger; the Faiyum and Nile of Egypt south to 28°N; Centropus senegalensis flecki (Reichenow, 1893); top of head and neck black with blue gloss, larger than West African birds; southern Africa from Zimbabwe and Botswana to Zambia, southern Zaire, Malawi and SW Tanzania.

Geographic variation Most striking in the rufous-bellied form “epomidis”, once described as a distinct species Centropus epomidis (Bonaparte 1850). This plumage form of C. s. senegalensis is common in humid coastal west Africa, especially in Nigeria where it occurs within 200 km of the coast, and it has been observed to 500 km inland in Ivory Coast (Brunel and Thiollay 1969, Lachenaud 2003). A plumage form with black throat and upper breast and a white belly has been seen in Ivory Coast near Abidjan where both white- and rufous-breasted color phases are common (Lachenaud 2003); this may be the coucal observed near Niamey, Niger, that was identified as another species (Debout et al. 2000). Birds near Lake Chad are sometimes pale, with crown grayish brown and back pale rufous brown, and were once known as a distinct form C. s. tschadensis Reichenow 1915.

Measurements and weights C. s. senegalensis: Wing, M (n ⫽ 9) 150–176.5 (159.7 ⫾ 7.5), F (n ⫽ 10) 155–185 (164.3 ⫾ 10.0); tail, M 179–195 (188.9 ⫾ 11.1), F 172–224 (195.7 ⫾ 16.0); bill, M 28.6–31.8 (29.6 ⫾ 1.7),

Senegal Coucal Centropus senegalensis 227 F 27–33.4 (30.5 ⫾ 2.0); tarsus, M 36.1–40 (37.7 ⫾ 1.7), F 36.5–42 (38.3 ⫾ 1.6); hallux claw, M 16.4–23.4 (19.1 ⫾ 3.3), F 16.4–21.0 (19.6 ⫾ 2.0) (FMNH, UMMZ, ZMUC); form “epomidis”: Wing, M (n ⫽ 3) 161–163 (162), F (n ⫽ 5) 154–166 (160.4 ⫾ 4.5) (BMNH, RMNH); C. s. aegyptius: Wing M (n ⫽ 16) 170–183 (175), F (n ⫽ 13) 177–187 (183) (Irwin 1988); C. s. flecki: Wing, M (n ⫽ 32), 163–176 (169), F (n ⫽ 32) 170–186 (176) (Irwin 1988). Weight, M (n ⫽ 4) 160–178 (169), F (n ⫽ 5) 163–180 (169) (Irwin 1988). Wing formula, P6 ⫽ 5 ⫽ 4 ⬎ 7 ⬎ 3 ⬎ 2 ⫽ 8 ⬎ 1 ⬎ 9 ⬎ 10.

Field characters Overall length 36–42 cm. Medium-sized coucal, bill more slender than in the larger Coppery-tailed C. cupreicaudus and Blue-headed Coucal C. monachus. Upper tail coverts black in adult, not barred as in adult Burchell’s Coucal C. superciliosus burchellii.

Kordofan) (Lynes 1925, Chapin 1939, Lamarche 1980, Nikolaus 1987, Snow 1978, Irwin 1988). They occur on offshore islands in the Bijagos Archipelago, Guinea-Bissau (Naurois 1969) and formerly on islands in Lake Chad (Bannerman 1951). In Egypt early in the 20th century it was in the Faiyum, Cairo and the Rosetta Nile; now it is in the Nile Delta and Lower Valley (Goodman and Meininger 1989). Common.

Voice

Habitat and general habits

Song a series of deep “coo” notes, the first two coos often shorter than the others, given at a pitch of about 0.5 kHz, the notes vary in length from 0.06 to 0.12 sec, repeated 3–4 per sec, “ouh ouh ouh ouh ouh ouh ouh ouh ouh ouh ouh ouh ouh”, sometimes dropping in pitch near the end of the series; pairs often duet. Other calls are a short single “ouh”, and a sharp series of “guk, guk” notes at 10 per sec given in alarm or excitement. Calls of dark morph “epomidis” are identical to calls of white-bellied coucals in West Africa (Chappuis 1974, 2000, Irwin 1988, Stjernstedt 1993, Christy and Clarke 1994).

Scrub and thickets, coarse grass, edge of reedbeds, dense riverine bush, sugarcane plantations and other farmlands, towns and gardens, palm groves; less closely associated with wet areas than other African coucals (Irwin 1988, Lewis and Pomeroy 1989, Maclean 1993, Christy and Clarke 1994, Cheke and Walsh 1996, Zimmerman et al. 1996). In southwestern Tanzania near Mbeya, and in Zambia and Zimbabwe along the upper Zambezi River from Kazungula to Katombora, this coucal lives along borders of a dry forest belt, in areas where White-browed Coucal C. superciliosus lives in dense low scrub and cane on flats and shore vegetation of small rivers and occasionally appears in low acacias (Ripley and Heinrich 1966b; RBP). In southern Africa they live in disturbed habitats, including sugarcane fields and gardens (Rowan 1983).They feed mainly on the ground, taking insects from dung of cattle, buffalo and elephants and more dispersed sources. They move in a slow stalking walk then hop and run, and they forage on escaping insects along the edge of grass fire. Flight is low and clumsy as the bird takes a few wingbeats and a glide, then

Range and status Africa south of the Sahara and to Egypt along the Nile. In West Africa they occur from Senegal to Nigeria and Cameroon, in the east from S Sudan and Ethiopia to Lake Victoria and through Central Africa from Uganda to Zaire and N Angola, and in southern central Africa from Zambia and Kenya south to Zimbabwe, the Okavango, Tanzania and S Mozambique. Resident in most of its range, locally migratory in the arid regions (Mali and west

228 Blue-headed Coucal Centropus monachus a crash landing. They sunbathe, perched with back to the sun, back feathers raised, wings drooped and tail spread wide (Hamling 1937, Rowan 1983).

Food Insects, mainly grasshoppers, caterpillars, termites, beetles and bugs; also frogs, small rodents, small snakes, lizards, birds, bird eggs and nestlings; snails (Chapin 1939, Cramp 1985, Irwin 1988).They take distasteful bush locusts Phymantous viridipes, a brightly colored insect with a waxy appearance that shows off its colors and emits a pungent foul liquid, an insect avoided even by ants (Goodwin 2001). Take live birds including Red-billed Queleas Quelea quelea (Ewbank 1985) and scavenge dead fish (Steyn 1970).

Displays and breeding behavior Territorial, the pair duets and advertises its area of c. 6 ha.

Breeding and life cycle During the rains while grass is pliant enough to build the nest and high enough to provide support and concealment. In Egypt they breed from March to August (Goodman and Meininger 1989), in northern Senegal from May to October (Morel and Morel 1990), in The Gambia from July to November (Gore 1990), in Liberia March to April (Gatter 1997), in Mali on the Bani River in August (Bannerman 1951), in northern Ghana July to August, in

southern Ghana at Accra and Cape Coast from March to June (Greig-Smith 1977, Grimes 1987), in Togo they care for fledged young in July (Cheke and Walsh 1996), in Nigeria they breed from March to August (Elgood et al. 1994), in Uele from September to November (Chapin 1939), in Ethiopia from April to August (Urban and Brown 1971), in Kenya from March to May (Brown and Britton 1980), in Malawi from November to May (Benson and Benson 1977), and in Zimbabwe and Botswana from October to March (Irwin 1981, Skinner 1996, Vernon et al. 1997). Birds call in duet nearly all year. The nest is a ball of coarse dry grass, lined with green leaves, the base of the nest more substantial than the dome, to 4 m above ground in bush.A nest I found in The Gambia in 1999 was open, without a cover built by the bird: the nest was built just below the crown of a citrus tree that provided a natural cover of dense spiny branches and leaves. Eggs are white, 34 ⫻ 26 mm, laid in a clutch of 2–5 (Granvik 1923, Bannerman 1933, 1951, Chapin 1939, Irwin 1988, Tarboton 2001). The incubation period is 17–19 days. Nestlings hatch asynchronously, and both parents give parental care. The nestlings are 24 g at day 3, half fledging weight by 8 days, and their wing feathers and ventral contour feathers burst their sheaths by 12 days. The young fledge in 18–20 days at a weight of 125–145 g, well before they can fly; their legs are well developed and they move about in dense vegetation (Steyn 1972).

Blue-headed Coucal Centropus monachus Rüppell, 1837 Centropus monachus Rüppell,1837,Neue Wirbelthiere zu der Fauna von Abyssinien gehorig, entdeckt und beschrieben, Vögel, p. 57, pl. 21, f. 2. (Kulla, northern Ethiopia) Polytypic. Three subspecies. Centropus monachus monachus Rüppell, 1837; Centropus monachus fischeri, Reichenow 1887; Centropus monachus occidentalis Neumann, 1908.

Description ADULT: Sexes alike, above, forehead and crown to upper back black with blue gloss, shaft of the hackles black, lower back and wing reddish brown, rump unbarred black, upper tail coverts either

barred or unbarred, tail black with greenish or bronze gloss, throat to belly and under tail coverts whitish to pale buff, darker buff on flanks; iris dark red, bill black, feet black. JUVENILE: Crown and neck dull blackish with narrow buff shaft streaks, back barred rufous and dark brown, wing rufous barred blackish, rump and upper tail coverts blackish with narrow buff bars, tail black narrowly barred buff, the central rectrices T1 either unbarred or barred near the tip, chin and throat buffy white, breast darker pinkish buff with small black spots and buff shaft streaks, belly buffy white.

Blue-headed Coucal Centropus monachus 229 NESTLING: At hatching, skin is black with long white hair-like down. SOURCES: AMNH, BMNH, CM, FMNH, MAC, KMMA, ROM, ZFMK, ZMB.

Subspecies Centropus monachus occidentalis Neumann, 1908; back and wing rufous brown, larger than C. m. fischeri; West Africa from Liberia and Ivory Coast to Cameroon, Gabon, N Angola and Zaire; Centropus monachus fischeri Reichenow, 1887; adult back and wing dark chestnut, juvenile darker than other populations; Upper Nile basin in S Sudan, Uganda,W Kenya (Lake Victoria basin) and NW Tanzania, Burundi and Rwanda; Centropus monachus monachus Rüppell, 1837; back and wing rufous, brighter than other forms; Eritrea and Ethiopia to C Kenya.

Geographic variation Additional subspecies have been described for Centropus monachus and the plumage variation with age is peculiar to certain forms. The form heuglini Neumann, 1911, with the head dull violet-blue black, in the large permanent swamps of the upper Nile in Sudan, is regarded as synonymous with C. m. fischeri and may be an immature C. m. fischeri (Friedmann 1930). Specimens in ZFMK in the series from southern Sudan taken by A. Koenig and his associates from 1910 and 1913 (including part of the type series of heuglini (van den Elzen and Rheinwald 1984) suggest a distinct subadult plumage; one specimen (CI.3.b´.␥) has a dull black crown (without pale shaft streaks), dark back, one long barred juvenile rectrix and the other (shorter) rectrices dull black, and the rufous wing feathers are incompletely barred, so the bird is not a juvenile and appears to be subadult. The holotype as recognized by van den Elzen and Rheinwald, and described by Neumann and illustrated in color (Koenig 1911) is in Stuttgart, SMNS 5662. Others in the Koenig series are the same in size and in adult plumage, and have a bright purple-blue gloss on the crown and nape.These birds are smaller than C. m. occidentalis in Cameroon, Gabon and Sudan.A form C. m. verheyeni Louette 1986, from Mabwe,

Upemba, in S Zaire was described as smaller, but birds in this locality overlap in size with populations of C. m. fischeri, with which this form may be associated. More work is needed on these coucals to determine their plumage sequence and their relationships to each other.

Measurements and weights C. m. occidentalis, Cameroon, Gabon and Zaire: Wing, M (n ⫽ 8) 180–191 (186.8 ⫾ 4.5), F (n ⫽ 6) 176–196 (185.5 ⫾ 7.1); tail, M 200–228 (216.0 ⫾ 8.4), F 210–224 (217.2 ⫾ 6.2); bill, M 30–34 (31.8 ⫾ 1.2), F 32–35 (33.0 ⫾ 1.1); tarsus, M 43–47 (45.3 ⫾ 2.0), F 43–48 (44.5 ⫾ 1.9), hallux claw, M 18–22 (20.5 ⫾ 2.0), F 21–28 (23.2 ⫾ 2.6) (AMNH); C. m. fischeri (“heuglini” ), southern Sudan: Wing, M (n ⫽ 10) 159–171 (166.5 ⫾ 4.7), F (n ⫽ 5) 167–177 (170.8 ⫾ 4.1); tail, M 181–223 (197.8 ⫾ 11.5), F 188–230 (290.8 ⫾ 15.6) (ZFMK); C. m. monachus, Ethiopia and C Kenya, wing, M (n⫽ 3) 188–220 (201.7), F (n⫽ 14) 182–226 (198.6⫾ 12.3); tail, M 218–237 (225.7), F 208–253 (229.3⫾ 13.3) (AMNH, FMNH). Weight, M (n ⫽ 5) 163–177 (171), F (n ⫽ 3) 206–283 (237) (BMNH). Wing formula, P6 ⫽ 7 ⱖ 5 ⬎ 4 ⬎ 3 ⬎ 8 ⬎ 2 ⬎ 1 ⬎ 9 ⬎ 10.

Field characters Overall length 45–52 cm. Coucal with large bill, crown and upper back black, back and wings rufous-brown, rump dark, unbarred in adult (barred in the larger Coppery-tailed Coucal C. cupreicaudus), tail long with deep bronze sheen, below white in the adult, bill large and black. Differs from White-browed Coucal C. s. superciliosus in unstreaked head. The immature has a rich buff breast and is more broadly barred on the wing than in the Senegal Coucal C. senegalensis; immature C. cupreicaudus is not barred on the wing.

Voice Deep and resonant, a slow “hoo, hoo, hoo-wu-wuwu-wu-wu hoo hoo, hu” series, the first two notes short and given with a pause before the rest of the series, the first at 0.48 kHz and the later notes

230 Blue-headed Coucal Centropus monachus slightly lower (0.43 kHz) in pitch, the later notes 0.10–0.12 sec long given at a constant rate of 4 per sec. Pairs often duet. Other calls include a monkeylike bark and a raucous cackle (Bates 1930, Chappuis 1974, 2000, Irwin 1988, Christy and Clarke 1994, Zimmerman et al. 1996).

Range and status Africa from Liberia and Ivory Coast to Cameroon, Gabon, southern Sudan, Ethiopia, western Kenya, Uganda, and east and southern Zaire (Friedmann 1930, 1966, Chapin 1939, van Someren 1949, Ripley and Heinrich 1966a, Traylor 1960, Urban and Brown 1971, Snow 1978, Louette 1986, Nikolaus 1987, Irwin 1988, Morel and Morel 1990, Gatter 1997). Resident. Common in village areas in NE Gabon (Christy and Clarke 1994) and E Zaire (Kizungu 2000) and in savannas of N and E Zaire, and in western Uganda around Lake Victoria and Lake Kyoga (Lewis and Pomeroy 1989, Rossouw and Sacchi 1998).

Habitat and general habits Swamps, high grass, marshes, papyrus and river banks, forest edge and second growth, mesic savannas near water or forest, dense cover, sugar cane fields, old manioc fields, edges of villages and along trails (Bannerman 1933, Chapin 1939, Thiollay

1985, Brosset and Erard 1986, Grimes 1987, Zimmerman et al. 1996). Occur in lowlands to 1000 m in lower montane regions in western Cameroon (Eisentraut 1973, Bowden 2001), to 2000 m and 2500 m in Kivu volcanoes (Chapin 1939) and to 2000 m in Kenya where they are within the 1000⫹ mm rainfall region (Lewis and Pomeroy 1989), in wetter habitats than C. senegalensis. Creep about on ground and through dense vegetation.

Food Generalist carnivore; takes insects, mainly grasshoppers and beetles, also snails, frogs, lizards and snakes, small birds (scrub warbler Bradypterus and reed warbler Acrocephalus, swallowed whole), nestling birds, bird eggs, mice and rats (Bates 1930, Chapin 1939, Verheyen 1953, Irwin 1988). One coucal killed a four-foot long cobra by attacking its head (Bates 1909). In captivity, they take slugs (Royston 1981).

Breeding In Togo they breed in June and July (Cheke and Walsh 1996), in Nigeria from March to June (Elgood et al. 1994), in Cameroon in January and from April to September (Bates 1909, Serle 1950b, Mackworth-Praed and Grant 1970, Irwin 1988), in Gabon from August to March (Christy and Clarke 1994), around Lake Victoria in September and from February to June (Brown and Britton 1980), in Zaire in Uele District from May to November (Chapin 1939,AMNH) and in Irangi in December (Kizungu 2000). The nest is an oval mass of dry grass and sedges (or sticks and dry leaves), lined with green leaves, with a side entrance, concealed in bushes or tall grass or a dense tree, placed about 0.2 to 3 m above ground. The eggs are white, with a light gloss, 34.5 ⫻ 27 mm, clutch 3–4 (Bates 1909, 1930, Bannerman 1933, Chapin 1939, Brosset and Erard 1986, Kizungu 2000). The male attends the nest when it has eggs (AMNH 159145, 627794). The incubation and nestling periods are unknown.

Coppery-tailed Coucal Centropus cupreicaudus 231

Coppery-tailed Coucal Centropus cupreicaudus Reichenow, 1896 Centropus cupreicaudus Reichenow, 1896, Ornithologische Monatsberichte, 4, p. 53. (Okawangoland and Angola ⫽ S Angola] Polytypic. Two subspecies, Centropus cupreicaudus cupreicaudus Reichenow 1896; Centropus cupreicaudus songweensis Benson 1948.

Description ADULT: Sexes alike, above the forehead to upper back black with violet gloss, back dark brown, wing and wing coverts reddish brown, rump and upper tail coverts black, often finely barred, tail blackish brown with coppery gloss; underparts creamy white; iris red, bill black, deep in shape, feet black. JUVENILE: Head dull black with black shaft streaks, back dark brown, wing reddish brown barred dark brown, rump and upper tail coverts blackish, tail blackish with fine buff tip, chin to belly whitish, breast has buff shaft streaks and small blackish spots. NESTLING: At hatching the skin is black, chin and abdomen tan, egg tooth white; long hair-like white down on head and spinal tract, shorter gray hairlike down on the wing and legs. SOURCES: AMNH, BWYO, BMNH, FMNH, UMMZ.

Subspecies Centropus cupreicaudus cupreicaudus Reichenow 1896; larger and paler; Angola to Zambia and Zimbabwe; Centropus cupreicaudus songweensis Benson 1948; smaller and darker; southern Tanzania and northern Malawi.

(41.1 ⫾ 1.5), F 41.4–43.0 (42.4 ⫾ 0.7); tarsus, M 47.0–57.6 (51.2 ⫾ 3.8), F 44.0–53.2 (49.1 ⫾ 2.7); hallux claw, M 21.8–30.4 (27.5 ⫾ 3.0), F 26.2–30.0 (27.2 ⫾ 2.3) (FMNH); C. c. songweensis: Wing, MF 195–203 (Irwin 1988). Weight, M (n ⫽ 5) 250–293 (272), F (n ⫽ 5) 245–342 (299) (Irwin 1988); M (n ⫽ 7) 203–228 (216.0), F (n ⫽ 6) 218–233 (223.0) (FMNH). Wing formula, P7 ⬎ 6 ⬎ 5 ⬎ 8 ⬎ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 9 ⬎ 10.

Field characters Overall length 42–50 cm. Large coucal with large bill, crown and upper back black, back dark rufousbrown and wings rufous, rump dark brown usually with indistinct buff bars (unbarred in the smaller Blue-headed Coucal C. monachus), tail long and deep copper; underparts white. Head and neck with violet (not blue) gloss. Juvenile differs from juvenile C. monachus in the lack of barring above and in whitish not buff breast. Wing with dark trailing edge (not dark in C. senegalensis). Where it occurs with Burchell’s Coucal C. s. burchellii, Coppery-tailed Coucal has a larger size and deeper voice; where it occurs with the smaller Whitebrowed Coucal C. s. superciliosus, it has an unstreaked head.

Voice Calls a series of low “coo” notes, at 0.4 kHz either on one pitch or descending in pitch, the note 0.12–0.20 sec long given 3–4 per sec, similar to Blue-headed Coucal C. monachus; contact notes a low “cou . . . cou . . .”. The pair often duets (Chappuis 1974, 2000, Stjernstedt 1993).

Measurements and weights C. c. cupreicaudus; Zambia and Botswana: Wing, M (n ⫽ 7) 202–224 (214.0 ⫾ 8.9), F (n ⫽ 7) 216–236 (224.4 ⫾ 7.6); tail, M 220–263 (245.1 ⫾ 17.9), F 243–273 (254.2 ⫾ 10.7); bill, M 38.8–42.5

Range and status South-Central Africa from Angola, southern Zaire, Zambia, Okavango, Boteti, Linyati and Chobe rivers in Botswana and E Caprivi, upper Zambezi

232 Coppery-tailed Coucal Centropus cupreicaudus

Food Large insects, mainly grasshoppers; also snails, crabs, fish, frogs (including painted reed frog Hyperolius viridiflavis and grass frog Ptychadena sp.), snakes, lizards, small birds (Ploceus weavers, Coturnix quail taken in scavenging), small rodents; also green grass and waterweed. Scavenge for dead fish, and tear open bird nests to take the nestlings (Vincent 1946, Rowan 1983, Irwin 1988).

Breeding River east to Victoria Falls in Zimbabwe, and there are a few records in Malawi and Tanzania. Sparsely distributed but locally common, as in marshes along streams and rivers near upper Zambezi. This species and Blue-headed Coucal C. monachus are allopatric through most of their composite range (Traylor 1960, Snow 1978, Irwin 1988, Louette 1986). Resident. Local movements occur with burning or heavy grazing of wetlands. Birds disappear and reappear when water and wetlands return, perhaps persisting in riparian thicket refuges (Vernon et al. 1997).

Habitat and general habits Wetlands in marshes, swamps, papyrus Cyperus papyrus, reedbeds Phragmites, river bank vegetation, thickets, long rank grass, floodplains and adjacent bush, dambos in miombo woodland. Occurs in lowlands, to 1250 m in Angola. They walk on the ground, feeding on insects, the birds alone or in pairs. Breeding adults search mainly near the nest in long-awned water grass Echinochloa stagnina on floodplain; they also forage up to 1 km from the nest, and occasionally they enter woodland adjacent to the floodplain (Hustler 1997a). A few hundred meters from water and marsh, this species and White-browed Coucal C. superciliosus are replaced by Senegal Coucal C. senegalensis.

In Zaire December to February (Vincent 1946, Irwin 1988), in Zambia during the rains in September and November to February (Benson et al. 1971), in Zimbabwe during the rains of January to April (Hustler et al. 1996, Hustler 1997a). Nest is a coarsely built sphere of fresh grass, coarse straws or twigs, lined with leaves, an entrance on the side, sometimes with a runway of flattened grass to the entrance, built low in reedbeds within 0.5 m of water level, on a square meters few of high ground overgrown with reeds or in a tuft of grass above water. Eggs are white, with little gloss, 38 ⫻ 28 mm, clutch 2–4, sometimes laid before the nest is complete. Incubation period is unknown; it probably begins when the first egg is laid, as the young hatch asynchronously and are of different sizes in a brood (Keith and Vernon 1969). Nestlings are fed locusts and frogs. Both parents provide care to the brood, sometimes regurgitating water for the young to drink. The nestling period is at least 17 days. The young leave the nest when well feathered but before they can fly, and after fledging they are cared for by the adults for at least 40 days. The young beg by wing fluttering and incessant “chuck” calls. The young develop the blue sheen on the nape plumage when they still have a short tail, dark eye and broad barring on some flight feathers. Nest predators include monitor lizards Varanus sp. and water mongoose Atilax paludinosus (Benson and Pitman 1964, Irwin 1988, Maclean 1993, Hustler 1997a).

White-browed Coucal Centropus superciliosus 233

White-browed Coucal Centropus superciliosus Hemprich and Ehrenberg, 1833 Centropus superciliosus Hemprich and Ehrenberg, 1833, Symbolae Physicae, 2, Aves, vol. 1, un-numbered p. 35, footnote 3. (Arabia and Ethiopia) Other common names: Burchell’s Coucal, Lark-heeled Cuckoo. Polytypic. Five subspecies. Centropus superciliosus superciliosus Hemprich and Ehrenberg, 1833; Centropus superciliosus burchellii Swainson, 1838; Centropus superciliosus fasciipygialis Reichenow, 1898; Centropus superciliosus loandae Grant, 1915; Centropus superciliosus sokotrae Grant, 1915.

Description ADULT: Sexes alike, upperparts blackish brown, head with black shaft streaks on hackles, upper back blackish streaked whitish, back rufous brown, rump and upper tail coverts finely barred black and buff, wing rufous chestnut, long and broad tail black glossed green with a thin white band on tip of each tail feather in fresh plumage, face with broad white superciliary streak (lacking in terminal adult plumage of C. s. fasciipygialis and C. s. burchellii); underparts whitish, breast with yellow strawcolored hackle shaft streaks, flanks and under tail coverts buff finely barred black, under wing coverts rufous buff; iris red, bill black, feet bluish gray to black. JUVENILE: Head and neck streaked buff and brown, back barred rufous and blackish, rump and upper tail coverts blackish barred buff, wing rufous chestnut with blackish bars on the tips, tail black with buff bars, face brown with yellow shaft streaks and an indistinct whitish superciliary streak; underparts whitish with buff yellow shaft streaks on throat and breast, belly whitish, flanks and legs buffy gray barred gray; iris brown to red (by the age when the tail is grown). NESTLING: At hatching skin black with long white hair-like down, like a spiny hedgehog. At fledging the young retains the hair-like down on the forehead, crown and neck, and the tail is very short.

SOURCES: AMNH, BMNH, BWYO, CM, FMNH, MSNG, MVZ, ROM, UMMZ, USNM, UWBM, ZFMK, ZMUC.

History and subspecies Centropus superciliosus superciliosus Hemprich and Ehrenberg, 1833; paler on crown and back, smaller than other geographic forms, flanks and under tail coverts barred, only center of belly unmarked white; juvenile more buffy (paler); Ethiopia, W Somalia, E Sudan, Kenya, NE Uganda and NE Tanzania; Centropus superciliosus sokotrae Grant, 1915; paler and grayer, pale shaft to dorsal hackles, not all individuals are distinct from Ethiopian superciliosus; southern Arabian Peninsula, E Somalia and Socotra I; Centropus superciliosus loandae Grant, 1915; larger and plumage darker on crown and back than superciliosus, juvenile crown black with rufous streaks; East Africa from Uganda, SW Kenya, Tanzania, SE Zaire, N Malawi, Zambia and Angola to N Zimbabwe and Botswana; Centropus superciliosus fasciipygialis Reichenow, 1898; crown, head and nape black, no superciliary streak, plumage similar to C. s. burchellii but lacks an intermediate age plumage and size smaller; eastern Zimbabwe, Tanzania, Mafia Island and Mozambique south to Beira; Centropus superciliosus burchellii Swainson, 1838; crown, head and nape black, no superciliary streak, black shaft to dorsal hackles, no apparent streaks on back, underparts white with no melanin in breast hackles; also an intermediate age plumage with pale buff superciliary streak, crown brown, neck blackish with straw-colored streaks; juvenile much more rufous than in other geographic forms, crown blackish with dark rufous spots and shafts, cheeks black, no light line over the eye, throat to breast, flanks and under tail coverts light rufous (not buff ) with indistinct blackish bars on the flanks, center of belly whitish; southern Malawi, southern Zambia, southern Zimbabwe, eastern Botswana and South Africa.

234 White-browed Coucal Centropus superciliosus The geographic complex of populations is considered a single biological species and the southern birds are White-browed Coucals without a white brow. Although the unbrowed forms might be recognized as a species C. burchellii (as in Clancey 1989), birds in intermediate plumage with a partial brow, apparent hybrids between white-browed and unbrowed forms, occur in the regions where these forms approach each other in Tanzania, Zambia, Mozambique and northern Malawi.The difference in plumage suggests that adults of northern and western regions have a juvenile-like plumage with white superciliary streak and brown crown and hind neck, while adults in the south have a glossy black head and neck. The terminal adult plumage of southern birds C. s. burchellii appears to be an extra plumage that is grown after the streaked adult plumage; birds breed in both streaked and unstreaked plumage (Lawson 1962, Snow 1978). In NW Zimbabwe along the upper Zambezi River near Kazungula, in 1997 and 1998 I saw mixed pairs mate and breed, one bird with a partial white eye streak and its mate without a streak. Birds in eastern Tanzania, C. s. fasciipygialis (described as a species fasciipygialis by Reichenow 1898, and considered a form of Centropus senegalensis by Friedmann and Loveridge 1937) have a blackish head without a superciliary streak (as in burchellii), and are small (as in loandae); their plumage varies along the coast with a trace of an eye streak in some birds at Mikindani, with different amounts of buff on the belly, and with a dark submarginal edge to the breast feathers giving a streaked breast (as in nominate superciliosus) that increases in the intensity of striping in the Saga Hills and Same District (ZMUC). Nominate superciliosus and C. s. loandae in East Africa and Zaire differ only slightly in plumage (Chapin 1939, Louette 1986). Because birds vary in plumage within a local population and intergrade with regionally neighboring forms, interbreed where they come together, and have the same behavior, they are considered conspecific.

Measurements and weights C. s. superciliosus; Kenya: Wing, M (n ⫽ 6) 135–152 (145.8 ⫾ 7.5), F (n ⫽ 6) 148–158 (154.2 ⫾ 4.2); tail, M 176–200 (189.3 ⫾ 8.5), F 185–207

(197.3 ⫾ 7.5); bill, M 27.7–29.3 (28.8 ⫾ 0.6), F 28.6–33.3 (31.3 ⫾ 1.8); tarsus, M 28.9–38 (33.3 ⫾ 3.7), F 33.7–38.1 (36.0 ⫾ 1.5); hallux claw, M 15–18.5 (16.9 ⫾ 1.8), F 14.8–21 (18.4 ⫾ 2.0) (FMNH); C. s. sokotrae: Wing, M (n ⫽ 6) 149–160 (155.2 ⫾ 4.0), F (n ⫽ 2) 165–168 (166.5) (MSNG, USNM); C. s. loandae: Wing, M (n ⫽ 6) 157–168 (161.5 ⫾ 3.7), F (n ⫽ 10) 160–172 (167.4 ⫾ 8.2) (FMNH); C. s. fasciipygialis: Wing, M (n ⫽ 6) 149–165 (156.4 ⫾ 6.0), F (n ⫽ 4) 156–166 (161.5 ⫾ 4.4) (ZMUC, UMMZ); C. s. burchellii: Wing, M (n ⫽ 11) 158–179 (168.1 ⫾ 6.3), F (n ⫽ 9) 166–182 (172.6 ⫾ 5.7) (AMNH, BMNH, FMNH, USNM). Weight, C. s. superciliosus: M (n ⫽ 6) 110–140 (124), F (n ⫽ 7) 125–170 (136) (Irwin 1988); C. s. loandae: M (n ⫽ 14) 142–213 (160), F (n ⫽ 10) 153–212 (180) (Irwin 1988); C. s. burchellii: F (n ⫽ 1) 159 (UWBM). Wing formula, P7 ⱖ 6 ⱖ 5 ⬎ 4 ⬎ 3 ⬎ 8 ⬎ 2 ⬎ 1 ⬎ 9 ⬎ 10.

Field characters Overall length 36–42 cm. Coucal with rufous back and wings and a long tail, whitish below, in marsh and on the ground.White-browed coucals in most of their range have a whitish superciliary streak in adult plumage, in contrast to all other coucals. Burchell’s Coucal C. s. burchellii are similar to Senegal Coucal C. senegalensis except they have a finely barred rump (black rump in C. senegalensis).

Voice First song type is a rapid series of bubbling “coo” notes repeated about 8 per sec, each note at 0.4–0.5 kHz, the song lasting two or three seconds. The first note is longer than the other notes in the series. The series often descends in pitch and increases in tempo, like water gurgling from a narrow-necked bottle. Another bubbling song type is a slower more deliberate series of notes beginning at 0.5 kHz and dropping to 0.4 kHz then rising to 0.5 kHz, the first notes at 6 per sec, the low middle at 9 per sec and the end at 4 per sec, slowing then rising at the end.A single bubble call is given to the

White-browed Coucal Centropus superciliosus 235 mate. Members of a pair call in a duet; one begins as the other completes its call and they often overlap. Excitement or threat call is a rapid chatter of sharp notes given 6 or 7 per sec. Alarm and excitement calls are a hiss and a single low “chuck”. The songs and calls are the same in C. s. burchellii in South Africa, C. s. superciliosus in Kenya and C. s. loandae in Zambia (North 1958, Stannard 1966, Chappuis 1974, 2000, Gillard 1987, Gibbon 1991, Stjernstedt 1993, Christy and Clarke 1994, Skead 1995, Stevenson and Fanshawe 2002, RBP).

Range and status Africa and the Arabian Peninsula (Ogilvie-Grant and Forbes 1903, Ripley and Bond 1966, Cornwallis and Porter 1982, Porter et al. 1987, Kirwan et al. 1996, Stevenson and Fanshawe 2002). They occur in the southern Arabian Peninsula in Yemen south of 17°N from the eastern Tihamah and Jabal Bura east to the western Hadhramaut around 48°E (Meinertzhagen 1954, Brooks et al. 1987), on Socotra Island (Ripley and Bond 1966), and widespread in Africa from Eritrea and Ethiopia, Sudan along the Nile, Zaire, and East Africa to Angola and South Africa. Resident: Locally they sometimes appear with the rains (Lewis and Pomeroy 1989). Common in much of the range (Chapin 1939, van Someren 1956).

Habitat and general habits Riverside, dense bush, scrub, acacia savanna and thickets, reeds and papyrus, moist vegetation, tall rank grass (Pitman 1928), marshes, shrubby once-

cultivated old fields, along watercourses and lush scrub and cultivation in arid country; mostly lowland, also montane areas, in foothills to 1000 m in Yemen ( Jennings 1981, Brooks et al. 1987) and in bracken briar as high as 2800 m in East Africa (van Someren 1956), and cultivated irrigated habitats in arid country. In regions where Senegal Coucal C. senegalensis also occur, C. superciliosus live in more marshy habitat. Solitary, retiring and skulking in behavior, often hidden, seen when they perch on a bush and spread wings and tail to dry and to sun. Flight is weak and horizontal, flapping then gliding, then crashing into the vegetation.

Food Insects, mainly grasshoppers, crickets and locusts, beetles (Curculionidae), also ants and other hymenoptera; spiders, land and garden snails, crabs, scorpions, lizards, snakes, frogs, mice, and small birds including nestlings and eggs (Ripley and Bond 1966, Johnson 1968, Maclean 1983, Rowan 1983, Frere 1984, Irwin 1988, Skead 1995). Small prey are swallowed whole; larger prey are broken into pieces or pulped by pecking, and snails are beaten against a stone. They also take fruit including loquats and buffalo thorn Ziziphus mucronata (Rowan 1983).

Displays and breeding behavior Pairs are monogamous and territorial. Male feeds the female a large insect at copulation (Brooke et al. 1990).

Breeding and life cycle Breeding season, during the rains. In Ethiopia they breed from March to June (mainly in April and May) (Friedmann 1930, Urban and Brown 1980), in southern Somalia fledglings appear in October (Ash and Miskell 1998), in Uganda in S Ankole they breed in October (Pitman 1928), in Kenya they breed in all months, mainly in the wet seasons as in Ngong where they breed from April to July and August (van Someren 1956, Brown and Britton 1980, Zimmerman et al. 1996), in Zanzibar and Pemba there are two distinct breeding seasons, the April to July and the November to January monsoons (Pakenham

236 Javan Coucal Centropus nigrorufus 1979), in Tanzania they breed from January to March (Schuster 1926), in Zambia from December to February (Benson et al. 1971), in Malawi from October to March (Benson and Benson 1977), in South Africa from September to February (Maclean 1993, Skead 1995), and in the SW Cape from August to January (Rowan 1983, Skinner 1996,Vernon et al. 1997). The nest is a large bulky untidy dome with a side entrance, built of grass and twigs, sometimes grass alone, without a distinct cup, the inner surface lined with leaves. It is built low in reeds or bush, from ground level to 10 m high. Nests built of grass are covered with the grass that is twisted around the top, or are built in a bower with several live twigs in the supporting shrub bent and woven over the nest, or in a bush, hedge or thorny tree (Pitman 1928,Tarboton 2001). Eggs are dull white, becoming stained yellow after a few days in the nest, 34.6 ⫻ 22.8 mm (C. s. superciliosus, Sudan, Egypt), 34.5 ⫻ 26 mm (C. s. burchellii, (Irwin 1988)), 32.9 ⫻ 24.7 mm (C. s. loandae, Uganda, (Pitman 1928)). Clutch size is 3–5(6) (Irwin 1988), in

Uganda 2–3 (once 4) (Pitman 1928), in Tanzania 3–4 (Schuster 1926), in Natal 2–5 (4 most common) (Dean 1971), in Eastern Cape 4 (Skead 1995). Eggs are laid at intervals of 2–3 days (Skead 1995). Incubation is mainly by male, beginning with the first egg or later.The incubation period is 14–15 days. Nestlings hatch asynchronously, a day apart. Both sexes, mainly the male, feed the young. The nest accumulates layers of feces (the earliest feces are in capsule form), cast quill-sheaths of growing nestlings, and remains of eggshells on the bottom layer; fresh green leaves are layered over this detritus (Skead 1995). Nestlings hiss and emit a foul black cloacal liquid when disturbed.They fledge at 18–20 days (or as early as 14 days if disturbed), and the older young stay near the nest until all are ready to leave the nest. Fledgling coucals are barely able to fly. They creep about quietly under cover until parents arrive with food (van Someren 1956, Johnson 1968, Irwin 1988, Skead 1995).Young are said to be flown to safety, the parent holding a chick between the parent’s thighs, saving it from an advancing fire (Bannerman 1933).

Javan Coucal Centropus nigrorufus (Cuvier, 1817) Cuculus nigrorufus Cuvier, 1817, Regnè Animale 1, 1817 (1816), p. 426. ( Java) Other common names, Sunda Coucal. Monotypic.

Description ADULT: Sexes alike, female slightly larger, upperparts glossy black with stiff hackles, upper back with a purplish gloss, wing coverts rufous with black ends on inner vane, inner wing coverts black, wing dark rufous with blackish inner vanes and black tips, inner secondaries black, tail long and black, below black glossed purplish; iris red, bill black, feet black. JUVENILE: Plumage like adult. NESTLING: Undescribed. SOURCES: MZB, RMNH, USNM.

Measurements Wing, M (n ⫽ 9) 195–218 (one “M” 238 not included) (207.4 ⫾ 6.4), F (n ⫽ 4) 210–227 (220.8 ); tail, M 182–242 (221.1 ⫾ 20.7), F 242–262 (250.3 ⫾ 10.4); bill, M 37–41 (39.0 ⫾ 1.5), F 40–41 (40.7 ⫾ 0.6); tarsus, M 57–60 (58.0 ⫾ 1.7), F 57–70 (62.3 ⫾ 6.8); hallux claw, M 26–42 (29.5 ⫾ 6.2), F 25–40 (32.0 ⫾ 8.2) (RMNH, USNM). Wing formula, P7 ⫽ 6 ⫽ 5 ⫽ 4 ⬎ 3 ⬎ 8 ⬎ 2 ⬎ 1 ⬎ 9 ⬎ 10.

Field characters Overall length 46 cm. Coucal with black upperparts glossed purple and stiff hackles, wings rufous with black edges and black areas on the coverts, and a long tail. Differs from Greater Coucal C. sinensis in black back (rufous in C. sinensis), purplish gloss, darker rufous of wings and black inner vanes of the wing flight feathers and coverts.

Javan Coucal Centropus nigrorufus 237

Voice Not recorded or described, said to be similar to C. sinensis (BirdLife International 2001).

Range and status Java. Resident in flat coastal swamps and wetlands: Muara Angke and Muara Gembong near Jakarta, Cangkring and Muara Cimanuk near Indramayu, and Ujung Pangkah and Sidoarjo on the Brantas river delta near Surabaya. Earlier records include Ujung Kulon NP on the tip of Western Java, and Anakan Lagoon near Cilacap on the south coast. They have disappeared in most sites where they were known earlier, with the loss of their habitat to development and urbanization and with trapping. Their numbers are low and their conservation status is vulnerable (Andrew 1990, MacKinnon and Phillips 1993, BirdLife International 2001).

Habitat and general habits Mangrove and other wooded tidal swamps in estuaries, associated with swamp fern Acrostichium, thickets and tall grasses Saccharum and Imperata in coastal lowlands, and Nypa palm swamps in brackish water behind mangroves; they may also occur in inland grass swamps. In coastal wetlands they occur in fringing swamps with pioneer mangroves Avicennia.They are not forest birds and are replaced

in dense stands of mature mangroves Rhizophora and Bruguiera by Greater Coucals Centropus sinensis. Where coastal habitat has been converted to fish and shrimp ponds and agricultural lands, they are replaced by Lesser Coucal C. bengalensis and Greater Coucal C. sinensis. In the wet season Javan Coucals have been seen in Imperata grassfields and swamps and along partly flooded forest edge, and they were netted in a sugarcane plantation near Surabaya (Ir Darjono, 2004). In the dry season they forage as they walk near puddles and on dry marshland and in high grass meadows (Bartels 1915–1930 diaries and notebooks on Javan birds, in BirdLife International 2001).

Food Omnivorous, they take many insects, including grasshoppers, beetles, large moths, hairy and smooth caterpillars, cicadas, large bugs and dragonflies; also snails, slugs, centipedes, bird and insect eggs, frogs, tree snakes and rodents (Bartels 1915–1930 and Sody 1953, in BirdLife International 1989). In one study the coucals took frogs Rana, geckos Hemidactylus, water snakes, rat Rattus dragonflies, beetles, grasshoppers and a cicada; in total prey numbers 40% amphibians and reptiles, 2% mammals and 34% insects, other items were not identified (Arifin 1997 in BirdLife International 2001).

Breeding Eggs seen in March and June; in a recent study the nesting period was January to March (Hellebrekers and Hoogerwerf 1967, Arifin 1997 in BirdLife International 2001). One nest was in an Acrostichium fern in a clearing along a river, a loosely-built structure of fresh and dry fern leaves, with a nest lining of fresh fern and grass leaves; other nests were 3–6 m above the surface in mangroves (Bartels 1915–1930, in BirdLife International 2001). Eggs are white, 39 ⫻ 31 mm, weight 1.8 g, clutch 1–3 (Bartels, in Hellebrekers and Hoogerwerf 1967) to 3–5 (BirdLife International 2001). Incubation and nestling periods are unknown.

238 Greater Coucal Centropus sinensis

Greater Coucal Centropus sinensis (Stephens 1815) Polophilus sinensis Stephens 1815, in Shaw’s General Zoology 9, pt. 1, p. 51. (“China” ⫽ Ning Po) Other common names: Crow-Pheasant, Common Coucal, Hume’s Crow-Pheasant, Larkheeled Cuckoo, Brown Coucal (andamanensis). Polytypic. Seven subspecies. Centropus sinensis sinensis (Stephens, 1815); Centropus sinensis bubutus Horsfield, 1821; Centropus sinensis andamanensis “Tyeleri” Beavan, 1867; Centropus sinensis intermedius Hume, 1873; Centropus sinensis kangeangensis Vorderman, 1893; Centropus sinensis anonymus Stresemann, 1913; Centropus sinensis parroti Stresemann, 1913.

Description ADULT: Sexes alike, female larger; plumage black on the head, mantle and underparts, head glossed blue to purplish, shaft streaks above are black, wing chestnut, tail black glossed greenish, long and broad, under wing coverts black (some birds have thin white bars); iris brown to red, bill black, feet black. There is no seasonal alternation of plumages by molt. The plumage of the head, mantle and breast varies in gloss, with some birds glossy blueor purple-black and others dull sooty black. The dull birds sometimes have remnant juvenile plumage; however, some birds are in molt from the barred juvenile plumage into glossy plumage on the head and breast. JUVENILE: Head dull black with brown spots, nape black with whitish bars, back black with rufous bars, upper tail coverts black with buff bars, wing coverts, inner secondaries, and tips of outer flight feathers rufous barred and tipped blackish, tail black with narrow buff bars, underparts from throat to under tail coverts blackish barred dull white with yellowish shaft streaks; iris gray to brown, lower mandible partly pale. Birds molt into an unbarred black body plumage while retaining a few juvenile barred flight feathers (Stresemann 1913a, 1939, Whistler and Kinnear 1934, UMMZ).

NESTLING: At hatching, skin black with long white hair-like down, the down hanging forward in a fringe over the eyes and bill, center of belly pinkish; upper mandible black with pink edges and an egg tooth, gape yellow, iris brown, feet brownish gray (Shelford 1900, Inglis 1903, Roberts 1991, Natarajan 1997). SOURCES: AMNH, BMNH, CM, FMNH, RMNH, ROM, SMTD, UMMZ, USNM, ZSM.

Subspecies Centropus sinensis sinensis (Stephens, 1815); as above; Pakistan, India from Punjab and Kashmir east through Himalayas and Gangetic Plain to Assam, Nepal, the Bhutan foothills and southern China (Guangxi, Zhejiang, Fujian); Centropus sinensis parroti Stresemann, 1913; upper back black, wing of juvenile without bars; peninsular India from Bombay, Madhya Pradesh and Orissa south to Kerala; and Sri Lanka; Centropus sinensis intermedius Hume, 1873; smaller; Bangladesh, India (west Cachar), Burma north to Chin Hills, China (Yunnan, Hainan), Thailand, Indochina and northern Malay Peninsula; Centropus sinensis bubutus Horsfield, 1821; wing paler rufous, larger; southern Malay Peninsula, Sumatra, Nias, Mentawai Islands, Java, Bali, Borneo, western Philippines (Balabac, Cagayan Sulu and Palawan); Centropus sinensis anonymus Stresemann, 1913; shorter wings, wing darker brown than bubutus; southwestern Philippines (Basilan, Sulu Islands); Centropus sinensis andamanensis Beavan, 1867; two plumage phases, one pale with head and body buffy white, wing brown, tail whitish above and rufous below; the other dark with head and back brown, rump darker, wing brownish purple with tip darker brown, tail bronzy purple, belly dusky, wing lining gray buff, iris red to yellow; Andaman Is and neighboring islands of Burma (Table, Great Coco and Little Coco Is);

Greater Coucal Centropus sinensis 239 Centropus sinensis kangeangensisVorderman, 1893; two plumage phases, one pale with head and body buffy white, wing rufous and tail rufous gray; the other dark with head, back, tail and throat brownish gray and the breast mottled gray, larger than andamanensis; Kangean Is. The mainland subspecies are not very distinct. In the Bombay Natural History Society collection, C. s. intermedius are not consistently smaller, sex for sex, than C. s. sinensis (Biswas 1960). In the Malay Peninsula around 5–6°N, C. s. intermedius intergrade with bubutus (Wells 1999). Sumatra birds do not differ from Java birds C. s. bubutus ( Junge 1948). On the other hand, C. s. parroti is distinct from C. s. sinensis. The island forms C. s. andamanensis and C. s. kangeangensis, each sometimes called a distinct species, have adults in a pale plumage phase and a dark plumage phase, the dark phase buff to brown rather than black as in adult C. sinensis on the mainland and larger islands.

Measurements and weights Centropus sinensis sinensis; Assam (Palasbari, Garo Hills, Khasi Hills, Naga Hills): Wing, M (n ⫽ 9) 197–208 (203.0 ⫾ 3.9), F (n ⫽ 10) 202–228 (214.6 ⫾ 9.1); tail, M 230–256 (245.8 ⫾ 9.1), F 241–277 (262.9 ⫾ 11.3); bill, M 31–34 (32.8 ⫾ 1.1), F 34–38 (36.1 ⫾ 1.6); tarsus, M 53–58 (54.9 ⫾ 2.2), F 56–64 (60.1 ⫾ 2.7); hallux claw, M 22–30 (25.6 ⫾ 2.4), F 27–32 (28.8 ⫾ 1.8) (UMMZ); Centropus sinensis parroti; Madhya Pradesh: Wing, M (n ⫽ 5) 176–197 (186.0), F (n ⫽ 5) 194–237 (210.4); tail, M 248–273 (266.3), F 258–290 (280.0) (UMMZ); Deccan to southern Indian peninsula and Sri Lanka: Wing, M (n ⫽ 23) 177–196 (187.6 ⫾ 5.1), F (n ⫽ 22) 196–212 (201.6 ⫾ 4.4) (Stresemann 1913a); Centropus sinensis intermedius; Thailand:Wing, M (n ⫽ 6) 184–212 (199.5 ⫾ 9.5), F (n ⫽ 6) 190–226 (207.0 ⫾ 13.0); tail, M 242–288 (263.3 ⫾ 16.5), F 250–290 (274.0 ⫾ 14.9); bill, M 30–36 (32.2 ⫾ 2.3), F 33–39 (35.5 ⫾ 2.6); tarsus, M 53–60 (57.2 ⫾ 2.4), F 55–60 (58.0 ⫾ 2.5); hallux claw, M 16–26 (23.0 ⫾ 3.6), F 17–28 (24.0 ⫾ 3.8) (ZMUC);

Centropus sinensis bubutus: Wing, M (n ⫽ 20) 202–221 (211.8 ⫾ 5.2), F (n ⫽ 18) 211–239 (227.4 ⫾ 7.3) (Stresemann 1913a); Centropus sinensis anonymus: Wing, M (n ⫽ 3) 182–191 (186.3), F (n ⫽ 1) 199 (Stresemann 1913a); Centropus sinensis andamanensis: Wing, M (n ⫽ 13) 167–200 (180.0 ⫾ 8.9), F (n ⫽ 13) 177– 190 (182.9 ⫾ 4.9); tail, M 235–260 (242.6 ⫾ 7.1), F 226–253 (242.9 ⫾ 9.4); bill, M 30–36 (33.0 ⫾ 2.0), F 30–36 (31.8 ⫾ 1.8); tarsus, M 41–50 (46.0 ⫾ 3.0), F 44–53 (49.0 ⫾ 3.6); hallux claw, M 20–24 (21.5 ⫾ 1.4), F 20–24 (23.1 ⫾ 2.2) (AMNH, BMNH); Centropus sinensis kangeangensis: Wing, M (n ⫽ 4) 192–205 (200.8 ⫾ 6.0), F (n ⫽ 2) 211–222 (216.5) (Stresemann 1913a); the hallux claw is long and straight (c. 18 mm) as in other C. sinensis. Weight, Nepal and India: M (n ⫽ 1) 255, F (n ⫽ 1) 370 (USNM, ZSM); India, C. s. parroti: M (n ⫽ 2) 208–218 (213) (Saha and Dasgupta 1992); Kangean Is, C. s. kangeangensis: M (n ⫽ 2) 254–270 (262), F (n ⫽ 4) 275–380 (305.8) (ZMB). Wing formula, P5 ⫽ 4 ⬎ 6 ⫽ 3 ⬎ 7 ⫽ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 47–52 cm.A large black coucal with chestnut wings and a long, broad, black tail (except in island forms). Island coucals (Andamans, Kangean) are polymorphic, the body either buff or gray-brown, with chestnut wings. Tail is proportionately longer than in Short-toed Coucal C. rectunguis. Differs from the smaller Lesser Coucal C. bengalensis in its black not pale shaft streaks above and the uniform rufous chestnut of the wings. Differs from Green-billed Coucal C. chlororhynchos in Sri Lanka by its black bill, paler wings, higher-pitched and longer calls and its jungle habitat. Differs from the smaller Philippine Coucal C. viridis by the black shaft streaks both above and below. Juvenile differs from C. bengalensis on the head and back, which are barred (streaked in C. bengalensis), and on the underparts, dark gray with narrow whitish bars (pale and streaked in C. bengalensis).

240 Greater Coucal Centropus sinensis

Voice Call, deep hooting notes, “hoop, hoop, hoop, hoop”, 3–4 notes a second, slow, low and mellow, the notes at 0.3 kHz given about 3 per sec, the note lasting 0.1 sec.The full song runs down and up the scale, faster in the middle of the series, followed by more “hoop” notes. Both sexes give the call, and a pair often duets with their calls overlapping in time, the female at a lower pitch than the male. Taking into account the differences in pitch of males and females, the lower pitch of the larger forms, and the variation in completeness of the repeated calls and songs, the geographic forms of these coucals appear to have nearly the same calls and are similar on mainland Asia and in the Andaman Is, although southern Indian birds appear to differ from northern Indian birds (P. C. Rasmussen). Other calls, a “tok, tok,” a rapid double-note rattle “lotok, lotok . . .”, a harsh scold “skaaah”, a soft “meeaow” by the female, and an alarm or threat “k’wisss”. Females and juveniles call “tch-truu, tchtruu” when they see the male carrying food (Smythies 1960, King and Dickinson 1975, Lekagul and Round 1991, Roberts 1991, Natarajan 1997, Scharringa 1999, Wells 1999, Kennedy et al. 2000, Robson 2000a, Sheldon et al. 2001, Supari 2003).A common name of the bird in Burma is “Bok” after the sound of the call (Smythies 1940), and an interpretation of the call “bu-bu-bu” in Borneo is the source of the local name and the subspecies name bubutus (Smythies 1999).

Range and status Indian subcontinent from Pakistan (Punjab and Indus plains and into the perennially irrigated tracts of the NW Frontier District), India from Jammu along the base of the Himalayas to Assam, south to Bangladesh and Bihar; from the Gangetic plains south from Orissa throughout the peninsula; southern China (south of Yangtse and the coastal plains in Yunnan, S Zhejiang, Fujian, Guangdong, E Gwangxi and Hainan), mainland and island southeast Asia; Sri Lanka, Nicobar Is, Sumatra (including Nias and Siberut), Borneo (including N Natuna Is), Java, Bali (including Kangean I) and the southwest Philippines (Sulu Islands, Cagayan Sulu, Basilan Is,

Balabac, Palawan) (Riley 1938, Smythies 1981, van Marle and Voous 1988, Sibley and Monroe 1990, Cheng 1991, Dickinson et al. 1991, Roberts 1991, MacKinnon and Phillipps 1993, Grimmett et al. 1999, Robson 2000a, Thomas and Poole 2003). Resident. Common. On the continent their range has extended in recent years into northern India (Smetacek 1974). They live on many offshore islands. They appeared on Krakatau I and were more common than C. bengalensis on Sebesy I in 1919, 36 years after the volcanic activity of 1883 exterminated all land birds on these islands (Dammerman 1992, Thornton 1996). The form andamanensis on Andaman Is and neighboring islands of Burma (Table, Great Coco, Little Coco Islands) where it is the only coucal (Stresemann 1939, Ali and Ripley 1969, Ripley and Beehler 1989), is considered near-threatened because of restricted distribution (Collar et al. 1994). The birds are common within their range (Tikader 1984, Davidar et al. 1996).

Habitat and general habits Forest edge and disturbed habitats, secondary forest, logged-out country, overgrown plantations, banks of large rivers, tall grassland, thickets, bamboo, reedbeds, grasslands and scrub near cultivation, cocoa, gardens, paddyfields, cover near swamps, streams and lakes, and in villages with shade trees. Widespread, except in dense primary forest. A bird of the lowlands in open grass and scrub country, in India they are mainly in the plains; in the hills they are not often seen above 900

Greater Coucal Centropus sinensis 241 to 1200 m, rarely to 2100 m; in Sri Lanka they have extended their range into the highest hill country, in Malay Peninsula they are below 700 m, in Sumatra below 700–800 m, in Borneo in the lowlands and in grassy uplands on the Kelabit plateau, and in Sabah from sea level to 750 m (Jerdon 1862, Hume and Oates 1890, Baker 1934, Ali and Whistler 1937, van Marle and Voous 1988,Thewlis et al. 1996, Robson et al. 1998, Grimmett et al. 1999, Smythies 1999, Wells 1999, Sheldon et al. 2001). In the Andaman Is, they are in forest, sugarcane and padi plantation and mangrove swamp (Ali and Ripley 1969). Greater Coucals are less common than Lesser Coucals where their ranges overlap. Terrestrial, they stay under cover. They stalk, walk, hop and run in pursuit of prey; they walk into, through and out of thickets looking for food, creep through shrubs, prowl on muddy river banks near the water edge, sometimes flash wings forward to flush their prey. They get into trees where they clamber on inclined branches, ascend into canopy, hop from branch to branch, and glide down to undergrowth. Flight is slow and labored; the bird alternates flaps and glides (Ali 1953, Smythies 1986). Feeding techniques include probing in holes of tree trunks and bark, gliding after insects that fall to the ground while birds feed in trees, jumping and catching low-flying winged termites in the air, chasing grasshoppers and lizards on the ground, and hopping and flying while chasing snakes on the ground (Natarajan 1993). They roost in reedbeds (Roberts 1991). Sun-bathe in morning and after rain, in singles or in pairs. The young coucals disperse from their parents’ territory a few months after they fledge. Territory size of nesting pair, 0.9 to 7.2 ha (mean 3.8 ha) in southern India (Natarajan 1997).

Food Opportunistic predator, takes insects (caterpillars, roaches, grasshoppers, katydids, beetles, larvae of rhinoceros beetles and other grubs, dragonflies, ants); centipedes, millipedes, scorpions, spiders, crabs, large garden snails (Achatina snails are smashed, soft parts then extracted), slugs, earth-

worms, small mammals (mice, hedgehogs), snakes, lizards, frogs and toads, bird eggs and nestlings; fruits (including of oil palm) and seeds (Legge 1880, Ali and Whistler 1937, Ali and Ripley 1969, Smythies 1981, Natarajan 1993, Sody 1989, Roberts 1991, Wells 1999). The coucals search under dense herbage for snails, and they scavenge on dead carcasses. Around villages in Tamil Nadu, snails Helix vittata are the main food (Natarajan 1993).

Displays and breeding behavior Monogamous, the birds occur in pairs. Male struts and chases female on the ground, she advances with her tail depressed and her wings drooped and quivered, then gives a harsh call, and the pair copulates on the ground or in a tree (Ali and Ripley 1969).The nest is built mainly by the male but the female also builds; the pair takes three to eight or more days to complete the nest (Dhindsa and Toor 1981, Roberts 1991, Natarajan 1997). The male brings food to the female before mating, and he feeds her during copulation or eats it himself after they mate.

Breeding In the rains; in northern India they nest mainly from June to September, in peninsular India from November to May (Hume and Oates 1890, Baker 1934, Ali and Whistler 1937, Ali and Ripley 1969, Gaston 1981, Zacharias and Gaston 1983, Natarajan 1997), in the Andamans mainly at the end of the hot season and beginning of the rains, May to July, and in February and April (Hume 1874, Hume and Oates 1890, Baker 1934), in Sri Lanka nearly all year with a peak in March and April (Henry 1971), in southeast Asia from January to August (Robson 2000a), in Burma from April to August (Oates 1877, Hume and Oates 1890, Baker 1934) and a nest fledged in October (Stresemann and Heinrich 1939), in Thailand from May to September and a half-grown young was taken in February (Riley 1938), in the Malay Peninsula eggs occur from January to May (Wells 1999), in Java from January to April, June and October to December (Hoogerwerf 1949, Hellebrekers and Hoogerwerf

242 Goliath Coucal Centropus goliath 1967), in Sabah nests with eggs are seen in September and March and nests with young in November, December, February and March (Sheldon et al. 2001). The nest is a large globular dome of twigs and leaves, coarse grass or reeds, palm leaves or coconut frond fibers or the ribbon-like saw-toothed leaves of screw-pine Pandanus, and it is sometimes strengthened with mud. It is lined with leaves and grass, and it has a side entrance. Some nests have the sides thick, the back thin, and the bird sits with its head towards the back of the nest and the long tail hanging out the side entrance (Herbert 1924). Less often the nest is open, built of green leaves and lined with dry grass (Legge 1880, Macdonald 1906).The nest is built under cover of creepers and vines, concealed in thick bush, fern tangles and pandanus crowns, as high as 6 m above ground or low in a thorny tree, or on the ground in dense clumps of grass or in rice fields (Baker 1934, Ali and Ripley 1969, Natarajan 1997, Wells 1999, Sheldon et al. 2001). Eggs are chalky white, with a yellowish glaze when laid; this wears away leaving a white chalky shell, often stained by nest dirt (Hume and Oates 1890); size 36 ⫻ 28 mm and weight 14.8 g in India (Baker 1934), 34 ⫻ 29 mm in Sri Lanka (Henry 1971), 35 ⫻ 28 in the

Andaman Is (Baker 1934), 38 ⫻ 30 mm in Java (Hoogerwerf 1964), 33 ⫻ 28 in Sabah (Sheldon et al. 2001); smaller eggs of 29 ⫻ 24 mm in Borneo (Gibson-Hill 1949c) appear to be of the smaller Lesser Coucal C. bengalensis. Clutch 1–4 (mean, 2.7) in India, 3(4) in Pakistan, 2–5 in southeast Asia, 2–3 in Sri Lanka, 3–4 (once, 5) in Thailand, 2 in the Malay Peninsula, 3 in Sabah and 2–3(4) in the Andamans (Herbert 1924, Baker 1934, Riley 1938, Ali and Ripley 1969, Roberts 1991, Wells 1999, Robson 2000a, Sheldon et al. 2001). Eggs are laid before the nest is complete. Incubation period is 15–16 days. Both sexes incubate the clutch and feed the brood. During the first four days the adults provide nestlings with regurgitated food, mainly snails. Parents visit the nest 2–5 times an hour depending on the size and number of nestlings. Both parents remove fecal sacs. Nestlings are half the fledging weight by 8 days, reach fledging weight by 16 days, and fledge at 160 g in 18–22 days, then accompany and beg from their parents for another 2 months. In a population in India, 77% of the eggs hatched and 67% fledged. Nests with eggs were sometimes deserted, and many nests were robbed by Jungle Crow Corvus macrorhynchos (Natarajan 1997).

Goliath Coucal Centropus goliath (Bonaparte 1850) Centropus goliath Bonaparte 1850, “Forsten” Conspectus Systematis Ornithologiae, 1, p. 108. (Halmahera) Other common names: Giant Coucal. Monotypic.

Description ADULT: Sexes alike, upperparts and underparts black, wing coverts black with a large white or whitish patch, wing glossed bluish black, tail long, graduated, broad and black; iris dark brown, bill black, feet black. On Halmahera there is a whitish phase with buff head, and some birds have pied

plumage or have irregular white marks elsewhere in the plumage. JUVENILE: Head, neck and underparts from throat to breast blackish streaked whitish, upper back and lesser wing coverts with broad whitish diamondshaped spots, feather shaft whitish, barbs next to the shaft buff and white and dark brown around the edge of the feather, wing coverts black with a large white patch, wing black, secondaries and coverts often have whitish diamond-shaped streaks, tail glossed bluish black, face brown with indistinct whitish streaks, chin whitish, breast and belly blackish or (on Halmahera) irregularly blotched

Goliath Coucal Centropus goliath 243 whitish and gray or blackish and gray; bill dark along culmen and the rest of bill pale, feet pale. NESTLING: Head with long white to buff hair-like down; iris dark gray, bill black above and pale gray below, feet pale (ZMA 48.175). SOURCES: AMNH, BMNH, MCZ, MSNG, MZB, RMNH, ROM, UMMZ, USNM, YPM, ZMA.

Measurements Wing, M (n ⫽ 10) 244–290 (261.6 ⫾ 14.6), F (n ⫽ 8) 255–298 (274.9 ⫾ 13.8); tail, M 390–434 (421.6 ⫾ 22.1), F 360–425 (394.5 ⫾ 26.6); bill, M 48.5–58 (50.6 ⫾ 4.0), F 48–60.4 (55.0 ⫾ 5.5); tarsus, M 46.5–55.5 (52.4 ⫾ 3.2), F 48–57 (55.1 ⫾ 4.0); hallux claw, M 22.5–28 (25.3 ⫾ 2.4), F 21–27 (24.3 ⫾ 2.2) (AMNH, USNM). Weight, M (n ⫽ 7) 340–453 (400.7), F (n ⫽ 2) 605, 638 (621.5) (MZB, ZMA). Wing formula: P 6 ⬎ 5 ⬎ 4 ⬎ 3 ⬎ 7 ⫽ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 62–70 cm. Very large coucal with black plumage, a large white wing patch, and a long graduated tail.

Voice Advertising call, a persistent series of deep “ooom” notes repeated about 2 / sec, the series lasting 4–12 sec. In alarm, one or more birds give a series of harsh, guttural monosyllabic or disyllabic notes, “kcau” or “kcau-kuc,” repeated at 1–4 sec intervals. Other calls include a deep, low, moaning roar (Ripley 1959, Coates and Bishop 1997).Voice is lower in pitch than C. sinensis and C. celebensis (Heinrich 1956).

Range and status Northern Moluccas (Morotai, Halmahera, Ternate, Tidore, Kasiruta, Bacan, Obi). Resident. Locally common and conspicuous on Halmahera, uncommon on Bacan (Shelley 1891, Lambert and Young 1989, Coates and Bishop 1997), not seen recently on Obi or Ternate (White and Bruce 1986, Lambert 1994). They perch on shaded branches in midstory

or lower crown canopy at forest edge where they give far-carrying calls (Coates and Bishop 1997). Tobelo people of Halmahera have different names for the two plumage morphs, “o ciungu” for the bird in normal black plumage and “o ciungu ma heheke” (ugly, dirty) for the whitish morph (Taylor 1990). Although they have been considered an allospecies with Greater Black Coucal C. menbeki of New Guinea (White and Bruce 1986), they are more closely related to Pheasant Coucal C. phasianinus of New Guinea and Australia.

Habitat and general habits Forests, including larger patches of scrub at edge of agricultural land; sea level to 450 m on Halmahera (Heinrich 1956), in primary and logged forest to 250 m on Bacan (Lambert 1994). They live in undergrowth to midstory; occasionally they perch in higher trees. Clamber slowly up vine- and epiphyte-covered trees. They feed in the understory and in lower forest subcanopy to 12 m, and they glean from tree trunks, the bases of Livistona palm fronds, vine tangles and epiphytes. Usually seen in pairs or small family groups of 3–4 birds (Rand and Gilliard 1967, Coates and Bishop 1997).

Food Large insects, including crickets, grasshoppers, cicadas and phasmid walking-sticks (Rand and Gilliard 1967).

Breeding Unknown.

244 Madagascar Coucal Centropus toulou

Madagascar Coucal Centropus toulou (P. L. S. Müller, 1776) Cuculus Toulou P. L. S. Müller, 1776, Des Ritters C. von Linne . . . vollständiges Natursystems . . . Supplement, 90. (Madagascar) Other common names: Malagasy Coucal. Polytypic. Two subspecies. Centropus toulou toulou (Müller, 1776); Centropus toulou insularis Ridgway, 1894, Aldabra I; Assumption I (extinct).

Description ADULT: Sexes similar. Breeding plumage, head and upper back glossy black, hackles black, wing unbarred rufous chestnut with dark brown tips, inner secondaries blackish chestnut, back, rump and upper tail coverts unbarred black, tail glossy black; underparts, chin to breast black, marginal under wing coverts black and gray, greater under wing coverts rufous chestnut; iris red, bill black, feet gray to black. Nonbreeding plumage dark and streaked, upperparts black finely streaked with long pale yellow shafts, the inner part of barbs graybrown, wing rufous, unbarred; underparts, throat and breast blackish streaked with yellow and gray, belly and under tail coverts black; bill rose-brown. Some birds retain a few unbarred juvenile rectrices in breeding plumage, and these may breed at a year of age. JUVENILE: Crown black with inconspicuous rufous streaks and yellow shafts, back black with indistinct buff bars and short pale shaft streaks, the buff marks forming short triangles; wing rufous chestnut with dark brown bars and tips, rump blackish barred with buff, upper tail coverts sometimes long (80 mm), black with narrow buff bars (bars of 3–4 mm black, 1.5 mm buff ), tail black with buff bars and tip; underparts, chin and throat variable, black with small buff spots to buff with irregular black bars, breast and belly sooty gray, feathers with indistinct black subterminal bars, rachis orange in the gray area and black in the barred area of the feather, under wing coverts dark gray with paler bars.

NESTLING: At hatching skin is black, dorsal surface covered with dense white hair-like shafts; bill black with white dorsal tip (remnant egg tooth), mouth lining and palate pink, papillated edge of choana white, base of tongue whitish with a black Ushaped band near the tip (Appert 1980). SOURCES: AMNH, BMNH, FMNH, MNHN, MSNG, ROM.

Subspecies Centropus toulou toulou (Müller, 1776); as above; Madagascar; Centropus toulou insularis Ridgway, 1894; underparts paler in nonbreeding plumage; Aldabra, formerly Assumption I where it now is extinct. Although it is paler on average, the nonbreeding plumage is not consistently paler than in Madagascar.

Measurements and weights Wing, M (n ⫽ 7) 135–156 (148.0 ⫾ 6.1), F (n ⫽ 13) 156–176 (167.8 ⫾ 6.5); tail, M 198–248 (218.0 ⫾ 21.8), F 218–270 (241.4 ⫾ 33.1); bill, M 24–28 (25.7 ⫾ 1.2), F 28–31.6 (29.5 ⫾ 1.4); tarsus, M 34–38 (35.5 ⫾ 1.6), F 35–42.7 (38.5 ⫾ 2.2); hallux claw, M 17.3–24 (20.8 ⫾ 2.0) F 20–24 (21.6 ⫾ 1.4) (AMNH, FMNH). Weight, Madagascar, M (n ⫽ 4) 125–150 (135), F (n ⫽ 3) 185–220 (205) (Goodman and Benstead 2003); Aldabra, M (n ⫽ 1) 117, F (n ⫽ 1) 131 (Benson et al. 1976). Wing formula, P 5 ⱖ 6 ⬎ 4 ⬎ 3 ⬎ 2 ⬎1 ⬎7 ⬎ 8⬎9⬎10.

Field characters Overall length 40 cm (male), 46 cm (female). Coucal with black head and underparts, rufous wings in the breeding season, back and breast finely streaked rufous and black in nonbreeding plumage, and with a long tail. The juvenile is dark and barred.This is the only coucal on Madagascar.

Madagascar Coucal Centropus toulou 245

Voice Muffled series of five to ten hoots decreasing in volume, short notes given c. 6 per sec at 0.6 kHz; the hoots are a self-naming “toulou, toulou . . .”. Also gives water-bottle call duets, the first bird in a “toulou” series, the second joining with a higherpitched series at 14 notes per sec dropping in pitch and slowing to 8–10 notes per sec; another call is a sudden guttural “coogoo” (Rand 1936, Benson and Penny 1971, Roché 1971, Milon et al. 1973, Frith 1975, Langrand 1990, Morris and Hawkins 1998, Randrianary et al. 1997).

Range and status Madagascar and Aldabra. Resident.Widespread and not uncommon on Madagascar, except on the denuded central plateau. Population is estimated at several hundred on Aldabra, where the birds are rather tame around humans (Benson and Penny 1971, Betts 2002).The population on Assumption I became extinct owing to disruption with the mining of bird guano and the destruction of the soil and vegetation (Stoddart 1984).

Habitat and general habits Dense vegetation or underbrush in forest, primary forest, recent clearings, second growth and scrub, eucalyptus woodlands, littoral forest, palms, gallery forest, mangroves, marshy reedbeds and grass, rice paddies, gardens. Occur from sea level to 1800 m; at PN de Marojejy from 450 to 1250 m (Goodman et al. 2000).They stay low in vegetation, feed on the

ground and in thick scrub.They chase grasshoppers and lizards in a “flush and rush” behavior and they rip bark from trees to get lizards beneath the bark. Coucals often feed in pairs, when one bird defends the other against theft of food by drongos (PrysJones and Diamond 1984). They are mobbed by small passerines. The coucals seldom fly, and when they do they give vigorous flaps, then glide into vegetation.

Food Invertebrates, large insects (beetles (Elateridae, Scarabaeidae, Tenebrionidae, Tettigoniidae), roaches (Blattodea), grasshoppers, crickets, mantids (Mantidae), caterpillars), ants (Formicidae), spiders (Salticidae), centipedes (Scolopendromorpha), mollusks, lizards (geckos, skinks), rats, chicks and eggs (Milon et al. 1973, Frith 1975, Goodman et al. 1997).

Displays and breeding behavior Male feeds the female with an insect in courtship. The pair wag their tails from side to side, calling occasionally; the male mounts as the female lowers and extends her wings, then passes the insect to her (Frith 1975).

Breeding Season uncertain. Adults are seen in black breeding plumage from July to September (Bangs 1918), and there are nesting records from September to March (Rand 1936, Milon et al. 1973, Frith 1975,Woodell 1976a, Benson et al. 1976, Appert 1980, Goodman et al. 1997, FMNH). Frith’s (1975) report that birds sometime breed in the brown “non-breeding” plumage on Aldabra suggests that the breeding season cannot be determined from the plumage alone, although other observations are of birds breeding only in the black plumage (Prys-Jones and Diamond 1984, R. Prys-Jones pers. comm.). In contrast to other birds but like some other coucals, the right testis is larger than the left testis (which is small in most coucals, and is sometimes absent) (Rand 1933). The nest is either a large spherical dome or a bulky open platform, woven from dry grass, twigs, creepers and other plant material longer than 10 cm, and lined with fresh green leaves. It is built

246 African Black Coucal Centropus grillii by the male, above the ground near the stem of a tree in dense vegetation such as a liana (Rand 1936). Eggs are white, 33 ⫻ 2 6 mm (Rand 1936). An egg is sometimes laid before the nest roof is built. Clutch size in Madagascar is 2–4, in Aldabra 2. Some eggs are laid several days apart. The male incubates more than the female, and he attends the nest. Incubation period is 14–16 days, the young hatch asynchronously when eggs are laid at intervals as long as 9 days apart (or as short as 3 days

apart, Woodell 1976b)—they hatch at intervals as long as 7–8 days apart.The parent feeds crickets and grasshoppers to the nestlings. The young beg with a churring sound.When disturbed in the nest they give a high-pitched hiss and excrete a foul-smelling black, sticky fluid. Nestlings grow to c. 90% of fledging weight by 15 days.The flight feathers and tail feathers are nearly half grown and most contour feathers open by 16 days, and the young fledge in 19 days (Frith 1975).

African Black Coucal Centropus grillii Hartlaub, 1861 Centropus Grillii Hartlaub, 1861, Journal für Ornithologie, 9, p. 13. (Gabon) Other common names: Black Coucal, Blackbellied Coucal Monotypic.

Description ADULT: Sexes alike. Breeding plumage, head and body black, black shaft streaks on the head, upper back and breast, lower back black barred buff, wing coverts rufous brown, wing rufous, tips of primaries and outer secondaries dark, inner secondaries dark brown, tail black; iris brown, bill black, legs and feet black. In nonbreeding plumage, upperparts dark brown with rufous barring, forehead to upper back streaked light rufous and black, wing rufous, tail black, no eye streak; underparts buff with whitish shaft streaks, belly with narrow brown bars, under wing coverts dark rufous.After the breeding season, black-plumaged adults molt into brown plumage and the bill becomes horn-gray. Birds in their first breeding season sometimes retain the barred wing coverts and flight feathers of the juvenile plumage in the black adult breeding plumage. JUVENILE: Upperparts light rufous and dark brown, head streaked and mottled with pale buff triangles on tip of feathers, nape and upper back streaked whitish along shaft, rufous along the white and dark brown along the feather edge, back barred light rufous and dark brown, wing coverts and

flight feathers rufous with dark brown bars, rump and upper tail-coverts dark brown with narrow buff bars, tail dark brown with fine rufous bars (dark bars 8 mm wide, buff bars 2 mm wide), under wing coverts pale chestnut with fine black bars; underparts, cheeks whitish, chin and throat whitish with fine black streaks, breast whitish buff with paler shaft streaks and barred black on the flanks, belly whitish, lower belly gray and under tail coverts blackish with buff bars; iris pale gray, bill pale yellowish, feet blue gray. NESTLING: At hatching, skin black with long white hair-like down above, forming a fringe over the forehead; bill black with a white egg tooth. SOURCES: AMNH, BMNH, BWYO, CM, FMNH, KMMA, NMM, ROM, SMF, USNM, ZMUC, ZSM.

Measurements and weights Wing, M (n ⫽ 14) 146–158 (152.3 ⫾ 3.9), F (n ⫽ 14) 164–173 (169.1 ⫾ 3.5); tail, M 122–170 (149.2 ⫾ 14.0), F 132–170 (159.0 ⫾ 10.1); bill, M 19–25 (22.29 ⫾ 1.6), F 22–27 (24.36 ⫾ 1.6); tarsus, M 29–40 (34.6 ⫾ 3.8), F 35–42 (38.9 ⫾ 2.5); hallux claw, M (24.6 ⫾ 3.6), F 23–31 (27.8 ⫾ 2.1) (BMNH (excludes specimens of A. Whyte, perhaps missexed), FMNH, ZSM). Weight, M (n ⫽ 6) 94–108 (100), F (n ⫽ 1) 151 (Irwin 1988). Wing formula, P 7 ⬎ 6 ⬎ 5 ⬎ 8 ⫽ 4 ⬎ 3 ⬎ 9 ⬎ 2 ⬎ 1 ⬎ 10.

African Black Coucal Centropus grillii 247

Field characters Overall length 30 cm (male), 34 cm (female). Small coucal in marshes and wet grass, breeding plumage black with rufous wings. Non-breeding and juvenile birds are streaked and barred, juveniles have barred wing and tail.The tail is shorter than in other coucals. They lack an eye streak (present in some plumages of White-browed Coucal C. superciliosus).

X

X

Voice The call usually heard is a double note “kop-kop” repeated at 2-sec intervals. Long song is a series of “kop” notes each rising and falling in pitch in the range 0.8–1.0 kHz, repeated 6 notes per sec for 2–3 sec with no change in pitch. When I watched the coucals at Ngaoundéré, Cameroon, in October, the two-note call was given in the early morning as the bird held its head down, with neck puffed out and bowed to the breast, a posture that other coucals use when they give the long water bottle call. Excitement or aggressive call is a series of harsh “shrehhh” and alarm call is a slow clucking “tucktuck” (Chappuis 1974, 2000, Stjernstedt 1993).

Range and status Africa from Senegal,The Gambia, Liberia and Ivory Coast eastward to northern Cameroon, Central Africa from Gabon and Angola through S and E Zaire to Ethiopia, Uganda and Kenya, also Zambia, Malawi, northern Zimbabwe, N and E Botswana, Mozambique and NE Natal. Resident or seasonal, depending on whether large regions or local sites are indicated. They appear and disappear with the rains, migrating into low-rainfall grasslands in the rainy season.They occur in both wet and dry seasons on the upper Niger River in Mali (Lamarche 1980). Seasonal with the rains from August to October in The Gambia; some remain into December (Barlow et al. 1997, Barlow 2003). In Ghana they occur in the south from May to August and in the north in the wet season from July to September (Greig-Smith 1976, Grimes 1987). In Togo they are seasonal, in the humid south in the dry season from November to July, in the north in breeding plumage in July to August (Cheke and Walsh 1996), and in southern Benin they occur from January to June

(Anciaux 2000). Resident in southern Nigeria and seasonal from April to October during the rains in northern Nigeria (Bannerman 1951, Elgood et al. 1994). In Gabon they occur in the dry months January and February (Rand et al. 1959, Brosset and Erard 1986). In Sudan and Ethiopia they are resident (Urban and Brown 1981, Nikolaus 1987), in East Africa they are seasonal with the rains (Lewis and Pomeroy 1989). On the Kafue River floodplain of Zambia they arrive in December or January and leave in April or May after breeding there (Osborne 1973). In the Okavango they may be resident (Skinner 1996) and in eastern Botswana they appear with heavy rains (Brewster 2000); in Zimbabwe they occur mainly from December to April (Irwin 1981). In South Africa they are uncommon irregular seasonal migrants in summer in Kruger NP and Natal (Vernon et al. 1997). The coucals are accidental on Zanzibar during the mainland dry season (Pakenham 1979). Common in suitable habitat, scarce over the total range. Seasonal disappearance of surface waters after rains and through the dry season and the largescale conversion of wetlands into sugarcane and the burning of old grasslands for grazing, all affect the local status of these coucals.

Habitat and general habits Tall rank grass and reedbeds, grassland near freshwater swamps in wide grassy river valleys and marshes, in floodplains and seasonally flooded grasslands (Irwin 1988, Lewis and Pomeroy 1989, Leonard 1998). In West Africa, they occur in high rank grass, in dry rice fields and forest clearings and coastal

248 Philippine Coucal Centropus viridis airstrips (Thiollay 1985, Gatter 1997, Barlow 2003). The coucals remain low in the grass under cover when not calling. Feeding behavior has not been seen as the bird creeps through the grass. Seen when it flies over the grass, alternating flaps and glides and flopping into cover (Winterbottom 1938).

Food Insects—mainly grasshoppers, also beetles, caterpillars, hemiptera, crickets, mantids, ants; spiders, small reptiles, and seeds (Chapin 1939, Irwin 1988).

Displays and breeding behavior Territorial, the pair defends its area against neighbors (Winterbottom 1938), the area 5–10 ha (Vernon 1971). In courtship, the female takes the initiative. Both sexes silently quiver their lowered wings before mating, male feeds female at copulation. Females are sometimes polyandrous (once 1 female observed with 3 males, each male tended a nest) and lay several clutches in a season (Vernon 1971, Rowan 1983); more often they are monogamous with two adults tending the young (Barlow 2003).

Breeding During rains when the grass is high in The Gambia, fledged young with short tails and a yellow gape were tended by two adults in eclipse plumage in December (Barlow 2003). Birds are in breeding plumage in Liberia from February to August (Gatter 1997), in northern Ghana birds in July and August, in coastal Ghana from April to July (Holman 1947,

Greig-Smith 1977, Grimes 1987), in northern Nigeria breeds in July and August (Elgood et al. 1994), in northern Tanzania (Serengeti, Mkomazi) from December to February (Brown and Britton 1980, Zimmerman et al. 1996), in Zambia from November to April (Aspinwall and Beel 1998), in Malawi from January to May (Benson and Benson 1977), in Botswana in January and February (Skinner 1996), and in Zimbabwe from December to April (Vincent 1946, Irwin 1981, 1988, Vernon 1971, Vernon et al. 1997). The nest is an oval ball of dry grass and sedge, concealed in grass or sedge 20–40 cm above the ground, with growing stems of grass wrapped around and over the nest chamber to form the outer covering, while the lower part of the nest is solid, substantial and lined with leaves ( Vincent 1946). The nest is lined with green leaves which are added before laying and during incubation. Eggs are white, 31 ⫻ 24 mm, clutch 3–6, eggs laid at irregular intervals (every day or with gaps up to 9 days). Incubation begins with the first egg.The incubation period is about 14 days. The nestling fledges when undisturbed at 18–20 days, or earlier when the nest is disturbed, then returns to the nest; the young bird flies at 28 days.The adult male builds the nest, incubates and provides all the care to the young, making 6–9 visits per hour to the nest, mainly in the early morning while the female keeps track of the nests and replaces clutches that are lost (Vernon 1971, Irwin 1988). Breeding success-a group of 4 adults reared c. 19 young in 6 breeding attempts (Vernon 1971).

Philippine Coucal Centropus viridis (Scopoli, 1786) Cuculus viridis Scopoli, 1786, Delicae Florae et Faunae Insubricae, 2, p. 89. (Antigua, Panay [Philippines]) Polytypic. Four subspecies. Centropus viridis viridis (Scopoli, 1786); Centropus viridis mindoroensis (Steere, 1890); Centropus viridis carpenteri Mearns, 1907; Centropus viridis major Parkes and Niles, 1988.

rufous, wing coverts rufous, wing rufous with dark brown tips, tail black, underparts with pale buff shaft streaks, under wing coverts dark gray with narrow whitish bars; other geographic forms are all black, though there is an uncommon white phase on Luzon; bare skin around eye gray, iris red to brown, bill gray to black, tarsus dark gray.

Description ADULT: Sexes alike, head, back and underparts black glossed blue to green, hackles with black shaft streaks from head to back and breast, lower back

JUVENILE: Above grayish brown streaked whitish buff, the whitish shaft streaks widening to elongate whitish spots on the upper back, back and rump

Philippine Coucal Centropus viridis 249 chestnut barred rufous, wing barred grayish brown and rufous, tail long, unbarred bronze green (tip sometimes barred), rectrices narrow; underparts irregularly barred whitish buff and dark gray, paler on throat and darker on belly, throat and upper breast sometimes streaked with whitish shaft streaks, under tail coverts slate gray, tail and body plumage soft texture. NESTLING: “Naked” at hatching, skin dark (Rabor 1977). SOURCES: AMNH, BMNH, CM, CMNH, FMNH, MVZ, ROM, SMTD, UMMZ, USNM, ZMUC.

Subspecies Centropus viridis viridis (Scopoli, 1786); wing chestnut, smaller; the Philippines (including Luzon, Mindanao, Masbate, Bohol, Negros, Cebu, Leyte, Samar, Catanduanes); Centropus viridis major Parkes and Niles, 1988; wing chestnut, larger; Babuyanes Islands group (the Philippines); Centropus viridis carpenteri Mearns, 1907; wing black ( juvenile and adult), larger; Batanes Islands group (the Philippines); Centropus viridis mindoroensis (Steere, 1890); wing black, smaller; Mindoro and Semirara (the Philippines). Comments: Ross and Ramos (1992) found a small bird on Camiguin (wing 162, tail 175), like C. v. viridis on Luzon rather than large like C. v. major on the other Babuyanes islands. Subspecies ranges here are as summarized by Parkes and Niles (1988), Dickinson et al. (1991) and Ross and Ramos (1992).

Measurements and weights C. v. viridis; Luzon: Wing, M (n ⫽ 8) 142.5–155.5 (148.4 ⫾ 4.7 66), F (n ⫽ 8) 151–178 (159.9 ⫾ 8.5); tail, M 223–234 (228.2 ⫾ 5.6), F 235–265 (247.6 ⫾ 9.8); bill, M 26.9–29.2 (28.1 ⫾ 1.0), F 227.6–31.5 (29.8 ⫾ 2.2); tarsus, M 35.5–38.5 (37.0 ⫾ 0.9), F 37.3–42 (40.0 ⫾ 1.8); hallux claw, M 16.6–21.2 (20.7 ⫾ 1.7), F 21.7–28.2 (23.5 ⫾ 3.2) (FMNH);

C. v. major: Wing, M (n ⫽ 5) 163–182 (171.0), F (n ⫽ 5) 175–190.5 (182.6) (Parkes and Niles 1988, Ross and Ramos 1992); C. v. mindoroensis: Wing, M (n ⫽ 9) 143–158 (153.7), F (n ⫽ 5) 163–175 (171.8) (Ripley and Rabor 1958); C. v. carpenteri: Wing, M (n ⫽ 1) 173, F (n ⫽ 4) 181.5–195 (189.5) (Parkes and Niles 1988, SMTD). Weight, C. v. viridis; Luzon: M (n ⫽ 8) 100.2–126.7 (112.1), F (n ⫽ 8) 108.4–152.1 (133.5) (FMNH); C. v. major: M (n ⫽ 2) 155–169 (162), F (n ⫽ 3) 212–223 (218) (Ross and Ramos 1992); C. v. carpenteri: M (n ⫽ 2) 153–170 (161.5), F (n ⫽ 6) 179–253 (209) (CMNH, USNM); C. v. mindoroensis: M (n ⫽ 6) 104–140 (123), F (n ⫽ 2) 150–165 (157.5) (Ripley and Rabor 1958). Wing formula, P5 ⬎ 6 ⬎ 4 ⬎ 3 ⬎ 7 ⬎ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 41–43 cm. Coucal with black head and body, and chestnut wings with dark brown tips. Birds on Mindoro and Batanes Islands are all black including the wings, the only all-black coucal in the Philippines (the larger C. steerii on Mindoro is dark brown).The birds are larger than the shortertailed Lesser Coucal C. bengalensis.

Voice A long, loud “boop boop boop . . .” given day and night, the notes about 0.6 kHz in pitch and given about 6 notes per sec; also a faster staccato “coo-coo-coo . . .” often descending in pitch, with 10 notes per sec, and “cha-coo” and “chi-go-go gook” alarm calls (Rabor 1977, Scharringa 1999, Kennedy et al. 2000).

Range and status Philippines, widespread except on Palawan and the Sulu Is, where Greater Coucal C. sinensis occurs. Resident. Common, it is the most common coucal in the Philippines (Stresemann 1939, Gonzales 1983, Dickinson et al. 1991).

Habitat and general habits Tall dense grass, mixed cultivation, disturbed second growth, thickets and vine tangles, bamboo, open

250 Lesser Coucal Centropus bengalensis

Food Insects, including beetles, grubs, grasshoppers and caterpillars, spiders, small lizards, sometimes carrion (Gonzales 1983, Goodman and Gonzales 1990, ZMUC).

Breeding

grass, clearings and edges of cultivation and gardens with dense vegetation; lowlands to mountains as high as 2000 m, on Mt Isarog uncommon at 450 m (Rabor 1977, Gonzales 1983, Goodman and Gonzales 1990, Kennedy et al. 2000, Fisher and Hicks 2000). Terrestrial, they live on or near the ground. When flushed out they rise and fly a short distance then plunge into thick bush or grass (Wolfe 1938, Gilliard 1950, Rabor 1977). Usually feed on or near the ground, hop under cover and climb through thickets of thorns and branches, seen in early morning or after a rain perched high while sunning and preening feathers to dry (Gonzales 1983).

Prolonged season; eggs are laid in late February (to account for an immature in April, Rand and Rabor 1960), a female had 3 ovulated follicles in April (CMNH 36135); other dates extend from May to July (Hachisuka 1934, Rabor 1977, Dickinson et al. 1991, Goodman et al. 1995, Robson 2000b, SMTD), and a large nestling was taken in September (ZMUC). Nest is a bulky globe of grass, about 20 cm h i g h ⫻ 13 cm wide or larger, built from wide grass blades woven around standing grass stems and incorporating live grasses pulled into the sides and over the top; entrance on one side; in tall grass usually 0.3–1 m above ground, lined with green grass and green leaves, which are added after the eggs are laid (Wolfe 1938). Eggs are dull white, chalky, 29.4 ⫻ 24.7 mm (Wolfe 1938), 30.4 ⫻ 25.0 mm (Rabor 1977), the clutch 2–3 (Wolfe 1938, Rabor 1977, CMNH 36135).The incubation period is about 2 weeks (Rabor 1977).The nestling period is unknown.

Lesser Coucal Centropus bengalensis (Gmelin, 1788) Cuculus bengalensis Gmelin, 1788, Systema Naturae 1, pt. 1, p. 12. (Bengal) Polytypic. Five subspecies. Centropus bengalensis bengalensis (Gmelin, 1788); Centropus bengalensis javanensis (Dumont, 1818); Centropus bengalensis medius Bonaparte, 1850; Centropus bengalensis lignator Swinhoe, 1861; Centropus bengalensis sarasinorum Stresemann, 1912.

Description ADULT: Sexes alike, breeding plumage, head and upper back black with green to blue gloss and black shaft streaks, lower back and rump brown, upper tail coverts black, tail glossy black sometimes tipped

whitish buff, wing coverts rufous brown with fine whitish shaft streaks, wing rufous brown with dark brown tips, underparts black from chin to under tail coverts, wing lining chestnut brown; iris red to reddish brown, bill black, legs black. Adult nonbreeding plumage is rufous brown with pale streaks. Head and upper back are brown with feathers having whitish streaks along the shaft, dark brown on the middle of the barbs and rufous on the edges; back unbarred brown; central upper tail coverts are barred and unusually long, in some birds nearly as long as the longest rectrices T1. Wing unbarred rufous brown with dark brown tips, tail

Lesser Coucal Centropus bengalensis 251 unbarred black or black with buff bars. Underparts buffy with fine straw-colored shaft streaks and indistinct incomplete black bars on throat and breast to belly, center of belly unmarked whitish, and flanks, lower belly and under tail coverts barred with black; under wing coverts pale chestnut; bill pale. Field observations indicate a seasonal change from black plumage to streaked brown plumage in India, southeast Asia, the Philippines and Taiwan (Deignan 1955, King and Dickinson 1975, Rabor 1977, Severinghaus and Blackshaw 1976, Grimmett et al. 1998); no museum specimens are in this molting plumage (Stresemann 1913b, Mees 1971). First-year nonbreeding (winter) plumage, head and back streaked brown, wing coverts and flight feathers rufous with blackish bars, central tail coverts barred and nearly as long as the tail, tail barred buff and dark brown. Underparts are as in older adult nonbreeding plumage, whitish to rich buff with streaks, dusky spots and bars, and under wing coverts pale chestnut. First winter plumage is attained by an incomplete molt of juvenile plumage, retains some barred feathers of the wing and tail; complete molt of juvenile body plumage. Birds molt directly from nonbreeding first-year plumage into adult breeding plumage, and some birds retain a few barred wing and tail feathers into the first breeding plumage. JUVENILE: Upperparts light rufous and dark brown, head irregularly streaked rufous with dark brown on the edge of feathers (lacks the white outlined by dark brown along the shaft of the adult nonbreeding plumage), back light rufous with dark brown bars, wing coverts and flight feathers light rufous with dark brown bars, rump and upper tail-coverts dark brown with narrow buff bars, tail dark brown with narrow rufous-buff bars (dark bars 6 mm wide, buff bars 3 mm wide); underparts, cheeks rufous, chin and throat whitish, breast whitish buff with fine paler shaft streaks and barred black on the flanks, belly whitish, lower belly gray and under tail coverts blackish with buff bars, under wing coverts pale chestnut with fine black bars; iris brown, bill pale horn, sometimes with a dark culmen. The iris develops a tan outer ring and darker brown inner ring during the time of postjuvenile molt.

NESTLING: Nearly naked at hatching, with thick, stiff, buffy hair-like down; iris gray, mouth lining dark red-pink, changing to pale flesh color with top of tongue black, bill flesh color to brown, feet grayviolet (Swinhoe 1863). Swinhoe identified the coucals on Taiwan, on which this description was based as C. viridis, but C. bengalensis is the only coucal on Taiwan. The yellow-white hair-like down remains attached to the upper back feathers until fledging. SOURCES: AMNH, BMNH, CM, CMNH, FMNH, MCZ, MVZ, MZB, RMNH, ROM, UMMZ, ZMUC, ZRC.

Subspecies Centropus bengalensis bengalensis (Gmelin, 1788), India, Nepal, Bangladesh, Burma south to Tenasserim,Thailand, southeast Asia; Centropus bengalensis lignator Swinhoe, 1861; larger than bengalensis; SE China, Hainan and Taiwan; Centropus bengalensis javanensis (Dumont, 1818); smaller than bengalensis; Malay Peninsula, Greater Sundas (Sumatra, Riau Archipelago, Lingga Archipelago, Bangka, Billiton; Java and Borneo) and the Philippines (Luzon, Mindanao, Mindoro, Cebu, Negros, Balabac, Palawan); Centropus bengalensis sarasinorum Stresemann, 1912; larger and darker than bengalensis; Sulawesi, Banggai Islands, Sula Islands, Lesser Sundas,Timor; Centropus bengalensis medius Bonaparte, 1850; larger than sarasinorum; the Moluccas (except SE), Tanimbar Islands (this form?). Philippine birds have sometimes been recognized as a subspecies C. b. philippinensis Mees 1971, said to differ from C. b. javanensis in the paler margin of the upper wing coverts, and it does in the type series of philippinensis in RMNH, and the form philippinensis was recognized in Dickinson et al. (1991). However, individual variation in adults is as great as betweenpopulation variation and may be related to wear or to age. C. b. philippinensis here is considered a synonym of C. b. javanensis. Birds in the central plains of Thailand, described as C. b. chamnongi Deignan 1955, are intermediate in size between C. b. bengalensis and lignator. Centropus bengalensis has been considered conspecific with African Black Coucal C. grillii, with

252 Lesser Coucal Centropus bengalensis which it is more similar in calls, size and habitat than either coucal is to Madagascar Coucal C. toulou (Dowsett and Dowsett-Lemaire 1993). Molecular genetics indicate that C. bengalensis and C. grillii share a more recent common ancestor than either does with C. toulou, while C. bengalensis is most closely related to Philippine Coucal C. viridis, a species that lives in the same range in the Philippines, than to C. toulou. These coucals each have distinct songs and they are distinct species.

Measurements and weights Centropus b. bengalensis; Assam: Wing, M (n ⫽ 7) 138–150 (145.0 ⫾ 6.0), F (n ⫽ 12) 158–174 (165.9 ⫾ 4.9); tail, M 161–188 (172.3 ⫾ 10.3), F 180–215 (194.0 ⫾ 11.9); bill, M 21.2–24.4 (22.8 ⫾ 1.2), F 22.6–26 (24.4 ⫾ 1.2); tarsus, M 34–38 (36.1 ⫾ 1.9), F 40–45 (42.7 ⫾ 2.2); hallux claw, M 21.2–24.4 (22.8 ⫾ 1.2), F 23.0–25.6 (24.6 ⫾ 1.2) (UMMZ); C. b. lignator; Taiwan, Amoy and Hong Kong: Wing, M (n ⫽ 5) 148–159 (153.2 ⫾ 4.8); F (n ⫽ 11) 165–174 (169.5 ⫾ 3.0) (Stresemann 1912); C. b. javanensis; Salanga, Sumatra, Bali, Natuna, Java, Malacca, the Philippines: Wing, M (n ⫽ 29) 125–147 (133.6 ⫾ 19.0); F (n ⫽ 25) 150–166 (158.9 ⫾ 4.7) (Stresemann 1912); C. b. sarasinorum; Sulawesi, Lesser Sundas,Timor: Wing, M (n ⫽ 5) 146–159 (153.4 ⫾ 5.3), F (n ⫽ 5) 175–188 (178.2 ⫾ 6.0) (RMNH); M (n ⫽ 41) 144–168 (152.9 ⫾ 23.5); F (n ⫽ 31) 169–190 (178.0 ⫾ 5.0); (Stresemann 1912); C. b. medius; the Moluccas: Wing, M (n ⫽ 9) 160–177 (171.7 ⫾ 5.4), F (n ⫽ 12) 190–205 (199.0 ⫾ 3.9) (Stresemann 1912). Weight, C. b. lignator;Taiwan: some mis-sexed?, M (n ⫽ 10) 95–187 (130.3 ⫾ 40.8), F (n ⫽ 10) 138–200 (176.8 ⫾ 17.4) (RMNH); C. b. javanensis, Philippines: M (n ⫽ 4) 82.7–90.0 (86.2), F (n ⫽ 5) 132–160.6 (148.3) (FMNH), C. b. sarasinorum, M (n ⫽ 5) 132–156 (145.6), F (n ⫽ 5) 227–290 (264.4) (ZMB). Wing formula, P5 ⬎ 6 ⬎ 7 ⬎ 4 ⫽ 8 ⬎ 3 ⬎ 9 ⬎ 2 ⬎ 1 ⬎ 10.

Field characters Overall length, male 31 cm, female 34 cm (India).A small coucal, in breeding plumage glossy black with

chestnut wings and a long black tail sometimes tipped white; the under wing coverts are pale not black. The bird is much smaller than Greater Coucal Centropus sinensis, which lives in much of its range; C. bengalensis looks streaked and scruffy due to the glossy shaffs of the head feathers, the chestnut color is paler, and the birds tend to be in grassy habitats, not forests. In the Philippines where it occurs with Philippine Coucal C. viridis, C. bengalensis is smaller and the wings are less rufous. In India the nonbreeding birds differ from Sirkeer Malkoha Taccocua leschenaultii in the plumage barring and the bill being brownish not red. In winter (first-year) nonbreeding plumage, above and below pale yellowish brown barred and streaked with dark brown extending onto the tail ( juvenile C. sinensis is darker and barred on the wings).

Voice Calls, (1) series of paired notes,“whoot, whoot”, then a series of “kurook, kurook, kurook” increasing in tempo and descending in pitch. The “whoot” notes are about 0.1 sec, shorter than notes of Greater Coucal C. sinensis. Calls are similar in India, southeast Asia and Malay Peninsula; (2) a series of short notes, “tu-dut” or “tu-dut-dut”, some pairs of notes low (0.4 kHz) and some pairs of notes high (1.0 kHz), in alternation, apparently given by mates (Scharringa 1999). Sometimes the members of a pair call together, each at its own rate, one bird giving a “whoot” and the other a “tu-dut-dut”.Variations on these themes are a series of hollow “booh-booh . . .” breaking into a tinkling cadence, a slowly accelerating series of short high notes “pok, pok, pok, po, po-popo-po-po . . .” descending in pitch, a rapid staccato call like a knock,“totok, totok, totopuk, totopuk . . .” and a coarse, harsh “krah, krah” (Meyer 1879, Coomans de Ruiter and Maurenbrecher 1948, Coomans de Ruiter 1950,Watling 1983, MacKinnon and Phillipps 1993, Smith 1993a,b, Coates and Bishop 1997). Calls are similar across the species range (Smythies 1960, Ali and Ripley 1969, Scharringa 1999, Wells 1999, Kennedy et al. 2000, Robson 2000a, Sheldon et al. 2001, Supari 2003). The common name “Dudut” in Indonesia is onomatopoeic (Bernstein 1859).The breeding adult approaches the nest with rapid short notes (Loke 1953).

Lesser Coucal Centropus bengalensis 253

Range and status Indian subcontinent (lower hills at base of the Himalayas and N Gangetic Plain from northern Uttar Pradesh east to Arunachal Pradesh, NE India in S Assam hills and the Sundarbans and in Bangladesh, and in peninsular India in the Western Ghats in Karnataka, Kerala and Tamil Nadu), southern China, Hainan, mainland southeast Asia, Malay Peninsula, Sumatra (also Riau Archipelago, Lingga Archipelago, Bangka, Belitung), Borneo, Lesser Sundas, Timor, Sulawesi, Moluccas, Tanimbar and the Philippines (Hume and Davison 1878, Rensch 1931, Smythies 1940, 1981, Hoogerwerf 1964, Mees 1971, Eck 1976, White and Bruce 1986, van Marle and Voous 1988, Cheng 1991, Dickinson et al. 1991, Lekagul and Round 1991, Coates and Bishop 1997, Grimmett et al. 1999, Wells 1999, Thomas and Foote 2003). In Sri Lanka the presence of this coucal is questionable; no records are mentioned by Legge (1880) or Baker (1927), and the record in Wait (1925) was based on a purchased skin (Harrison 1999, Thomas and Foote 2003). Resident in most of its range, migratory in parts of China, moving in winter onto the plains of Burma, Thailand and Vietnam (Smythies 1940, Deignan 1955, Robson 2000a), and one was mist netted at floodlights at Fraser’s Hill, in migration at night (Wells 1999).They colonize new sites where forests are cleared (Holmes and Burton 1987). After the 1883 volcanic eruption and devastation of all plant and animal life on Krakatau, these coucals had colonized the re-vegetated island by 1908 from a source at least 15 km distant over open water, and

they nested on the island soon afterwards (Dammerman 1922, Robinson and Kloss 1923b, Thornton 1996). Common in much of the range. Longest known survival from date of ringing, 74 months (McClure 1998).

Habitat and general habits Tall wet grass including elephant grass Imperata, reedbeds, standing rice fields, grasslands, agricultural plains, sugar cane fields, swamp, bamboo thickets, second growth forest, open country scrub and cultivation; mainly in lowland floodplains and lower valleys. They get into submontane tracts of the Himalayas, and they are widely distributed in the Indian Peninsula and in Bangladesh.They live from the lowlands to 365 m, and occasionally to 1400 m in Nepal, in isolated hill-station agriculture to 1500 m in the Malay Peninsula, to 1500 m in Sabah (most records below 700 m, once to 1500 m), 700 m in Sangihe, 1000 m in Sulawesi, 820 m in Buru, 650 ⫹ m in Seram, 900 m in Lombok, 900 m in Sumbawa, 1500 m in Flores and 500⫹ m in Sumba, rarely higher ( Jerdon 1862, Meyer 1879, Inskipp and Inskipp 1985, Coates and Bishop 1997, Riley 1997, Bishop and Brickle 1998, Wells 1999, Kennedy et al. 2000, Sheldon et al. 2001).Terrestrial, they do not climb about in trees (Smythies 1981). Heard often, they are seen in the early morning and late afternoon when the coucals are in the open on sunning perches, look-outs and song posts, or in labored flight low over the vegetation.

Food Insects, including grasshoppers (Acrididae), locusts, crickets, mantids, beetles, hemiptera, hairy caterpillars, Sphingid caterpillars; also spiders, lizards (Calotes), frogs and fruit (Ali and Whistler 1937, Loke 1953, Rabor 1977, Smythies 1981, Sody 1989,Wells 1999; ZMUC).

Displays and breeding behavior Monogamous, live in pairs. In courtship feeding, the male holds a large insect in his bill.The female approaches him, raises her head, neck and bill, and calls “tok tok tok tok” and the male approaches her, raises his tail and bows to her. Copulation lasts about 30 sec. In India, both sexes are said to build

254 Violaceous Coucal Centropus violaceus the nest, incubate and care for the young (Baker 1934, Ali 1953, 1996), whereas in Java, the male alone is said to build the nest, incubate and feed the young (Bernstein 1859, Spennemann 1928).

Breeding In India they breed from May to September, in the plains they nest after the rains begin in June, in the hills they begin by April and May, in Kerala they breed in August and September (Hume 1873, Baker 1934, Ali 1953); in Burma they nest in the rains (Oates 1877, Smythies 1981), in Thailand they breed from May to August (Riley 1938), in the Malay Peninsula eggs are laid from December to July and fledglings appear into October (Wells 1999), in Taiwan they nest in September (Swinhoe 1863), in the Philippines from March to May (Rabor 1977), in Java from January to July, and in November (Spennemann 1928, Hoogerwerf 1949, Hellebrekers and Hoogerwerf 1967), in Flores they

breed from February through September, mainly in March and April (Verheijen 1964), in Sarawak at the time of the rice harvest in February and in Sabah from December to May (Smythies 1999, Sheldon et al. 2001). The nest is built near the ground in dense vegetation, grass or low bushes, well concealed, c. 25 cm h i g h ⫻ 18 cm across. The round, ball-like nest is formed with live grasses pulled down and shaped into a nest which is then covered with a dome of grass, often lined with green leaves, and has a side entrance (Jerdon 1862, Swinhoe 1863, Hume and Oates 1890, Herbert 1924, Loke 1953, Wells 1999, Sheldon et al. 2001). Eggs are chalky white, nearly round, 28 ⫻ 24 mm (India), 30.3 ⫻ 24.8 mm (Malay Peninsula), 27.6 ⫻ 23.8 mm ( Java), 31 ⫻ 24 (Labuan) and 32 ⫻ 25 mm (Flores). The clutch is usually 3(2–6) in India, 2–4(5–6) in southeast Asia, 2(3) in Malay Peninsula, and 4 in Taiwan (Sharpe 1879, Riley 1938, other references above). Incubation and nestling periods are unknown.

Violaceous Coucal Centropus violaceus Quoy and Gaimard, 1830 Centropus violaceus Quoy and Gaimard, 1830, Observations zoologiques faites à board ‘de l’Astrolabe , 1: 229; Atlas, Oiseaux, pl. 19. (Carteret Harbor, New Ireland) Other common names: Giant Forest Coucal. Monotypic.

papillated ridge along the palate cleft, tongue pink with a papillated white rear edge and scattered white papillae on the dorsal surface, and a black U-shaped band near the tip, feet bluish gray (Gilliard 1961).

Description

Measurements and weights

ADULT: Sexes alike, upperparts and underparts black with violet gloss, wing black with violet gloss, tail long, graduated, broad and black with violet gloss; bare skin around eye white to light reddish, eye-ring black, iris red, bill black, feet pale slaty horn.

Wing, M (n ⫽ 6) 258–292 (271.2 ⫾ 12.5), F (n ⫽ 5) 245–288 (261.2 ⫾ 16.0); tail, M 340–405 (379.6 ⫾ 18.4), F 310–400 (356.0 ⫾ 34.9); bill, M 61.5–69 (65.3 ⫾ 2.9), F 62–77 (69.1 ⫾ 5.9); tarsus, M 61.5– 68 (65.3 ⫾ 2.9), F 62.5–70 (66.3 ⫾ 3.2), hallux claw, M 20.5–25 (22.3 ⫾ 1.7), F 20–23 (21.0 ⫾ 1.41) (AMNH, ZMUC). Weight, F (n ⫽ 1) “500 g” (Gilliard and LeCroy 1967a), label of specimen (AMNH 777857) says “500⫹ g”, perhaps the capacity of the weighing scale. Wing formula, P7 ⫽ 6 ⫽ 5 ⫽ 4 ⫽ 3 ⬎ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10.

JUVENILE: Upperparts dull black, wing and tail black glossed violet; underparts loose-webbed, sooty gray; iris light gray. NESTLING: Black skin, long white feather sheaths to 35 mm on head, back and wing, bill black with a white tip, mouth lining purplish pink with white

SOURCES: AMNH, ZMUC.

Lesser Black Coucal Centropus bernsteini 255

Field characters Overall length 64–70 cm. A very large blackish coucal with a violet gloss, a long graduated tail and a red eye.The bare white skin around the face and the whitish feet show clearly in the field.

Voice Deep hollow booming notes and loud booming duets “oo-OOMP”, given in a continuous series; the calls are loud and carry over several kilometers (Coates 2001, Mayr and Diamond 2001).

Range and status Bismarck Archipelago: New Britain and New Ireland, absent from the smaller islands (Rothschild and Hartert 1907, Hartert 1925c, Coates 1985, Mayr and Diamond 2001). Resident: More common in New Britain than in New Ireland (Coates 2001). Sparsely distributed both in lowlands and in the mountains; it is estimated to be one of the 25 rarest species in the Bismarck Archipelago, with fewer than 1000 pairs in 10,000 square kilometers (Mayr and Diamond 2001).

about in tree branches and vines to a height of 15 m, and they plane from the top of trees to lower levels (Hartert 1926). One bird walks to another adult carrying a large insect in the bill, in courtship feeding (Gilliard and LeCroy 1967a). Occur singly and in pairs; the members of a pair allopreen (Coates 2001).

Habitat and general habits

Breeding

Primary tropical rain forest, disturbed habitat, active in limbs and vines and ferns, from lowlands to 950 m. Call loudly back and forth in forest, with neck pulled down and tail hanging, as they give booming calls.They keep in cover of vegetation and are difficult to observe. They jump, hop and walk

Season is November to January. Nest is a loose structure of twigs built at the top of a tall tree in the forest. Eggs are white, somewhat glossy, 43 ⫻ 34 mm, the clutch is 2 or 3 (Hartert 1926, Meyer 1931, 1933, 1936, Gilliard and LeCroy 1967a). Incubation and nestling periods are unknown.

Food Large insects (stick insect as large as 18 cm), frogs, and tiny snails (Gilliard and LeCroy 1967a).

Lesser Black Coucal Centropus bernsteini Schlegel, 1866 Centropus Bernsteini Schlegel, 1866, Nederlandsch Tijdschrift voor de Dierkunde, 3, 251. (New Guinea ⫽ Vogelkop peninsula) Other common names: Black Scrub Coucal, Bernstein’s Coucal. Monotypic.

Description ADULT: Sexes alike, a small, short-tailed black coucal with black shaft streaks, upperparts glossed green, wing black, tail black, wing all black below, under

wing coverts black; iris dark brown, bill black, legs and feet black.A nonbreeding adult plumage is unknown. JUVENILE: Above blackish narrowly (1–2 mm) barred buff or rufous, wing and tail blackish with narrow (2.5 mm) buff to pale rufous bars, tail feathers narrow (32 mm wide, vs 42 mm in adult), throat whitish in midline, neck and breast marked with chestnut brown on sides, central belly gray with blackish bars, remaining underparts blackish brown with narrow pale barring.

256 Lesser Black Coucal Centropus bernsteini NESTLING: Short white tips on growing feathers of the face (AMNH 267142), not spiked with long hair-like down as in some other coucals. SOURCES: AMNH, ANSP, BMNH, FMNH, MSNG, MVZ, SMTD, ZFMK, ZMB, ZSM.

Subspecies No subspecies are recognized. In one sample (AMNH), the birds on Manam were smaller than birds in New Guinea and were described as a subspecies Centropus bernsteini manam, (Mayr 1937). In BMNH and MSNG samples, males on Manam and New Guinea overlap in size.

Measurements and weights New Guinea: Wing, M (n ⫽ 12) 164–193 (173.8 ⫾ 8.5), F (n ⫽ 8) 176–196 (180.0 ⫾ 8.0); tail, M 220–265 (247.4 ⫾ 13.6), F 230–265 (247.8 ⫾ 10.9); bill, M 27–33 (31.3 ⫾ 6.0), F 28–35 (31.8 ⫾ 6.2); tarsus, M 39–44 (41.2 ⫾ 1.9), F 38–44 (41.4 ⫾ 2.3); hallux claw, M 21–26 (23.9 ⫾ 1.4), F 24–26 (25.1 ⫾ 0.6) (AMNH, BMNH, ZSM); Manam: Wing, M (n ⫽ 4) 179–190 (182.8 ⫾ 5.0), F (n ⫽ 4) 188–200 (191.8 ⫾ 8.4); tail, M 248–275 (261.2 ⫾ 11.1), F 271–277 (274.5 ⫾ 2.5) (Mayr 1937 (males), AMNH, BMNH). Weight, M (n ⫽ 5) 130–160 (146), F (n ⫽ 2) 160–200 (180) (AMNH, FMNH, Hartert 1930). Wing formula, P5 ⫽ 4 ⬎ 3 ⱖ6 ⬎ 7 ⬎ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 46–52 cm. Small black coucal with a short tail, dark iris and a black bill. This species is smaller and more uniformly black than the Pheasant Coucal C. phasianinus with which it occurs in New Guinea, and has a dark iris (unlike yellow iris in Biak Coucal C. chalybeus).

Voice Three hoots, “woop woop woop” on a descending scale, sometimes given in longer series. Pairs often duet, one bird with “hui” and the other with “whui”, a lower note given at a slower rate. Call is weaker and flatter than the call of Greater Black Coucal C. men-

beki and shorter than that of Pheasant Coucal C. phasianinus in New Guinea (Gyldenstolpe 1955, Coates 1985, Beehler et al. 1986,Tolhurst 1992).

Range and status Western and central New Guinea; Manam I. In New Guinea they occur in the west from Vogelkop, and north from the Idenburg River and Humboldt Bay east along the coast to Sepik River, along the coast of the Huon Peninsula, Lae and Bulolo, and in the south from the Mikima and Setikwa rivers in the west and the Fly and lower Turama rivers near Kikori in the east. Resident. Largely allopatric with Pheasant Coucal C. phasianinus, but both occur in the north. Not uncommon in second growth and cane grass, they are shy and difficult to observe except when they perch in the open to call in the morning and late afternoon (Maxwell 1938, Beehler et al. 1986, Tolhurst 1992, Coates 1985, 2001).

Habitat and general habits Scrub, tall cane grass, rank vegetation and edge of forest along rivers and lakes; sea level to 500 m, rarely to 900 m (Mayr 1941, Rand 1942b, Gilliard and LeCroy 1966, Rand and Gilliard 1967, Coates 1985, Beehler et al. 1986).

Food Small lizards, small snakes, grass insects, butterflies (D’Albertis and Salvadori 1879, Gyldenstolpe 1955, FMNH).

Pheasant Coucal Centropus phasianinus 257

Breeding In New Guinea region the season is prolonged: a nest with eggs was found in early May along the Idenburg River (Rand 1942b), where a young bird with a short tail was found in December (AMNH 207147), a laying female was taken at Kaku in November (FMNH 267143), and a female had a large ovary in August at Mimika River, Wkatimi (BMNH 1911.12.20.961). Along the Sepik River, local people say the birds nest at the time of low water in July (FMNH 280092). Breeding males

sometimes have two large testes (FMNH), in contrast to some other coucals, while other males of this species have a rudimentary left testis (Rand 1942b).The nest is a covered mass of cane grass and has an irregular untidy opening in the side, and is built in grass and supported on all sides by the stems of the grass. One nest measured 21 ⫻ 38 cm high outside, 15 ⫻ 25 cm inside. Eggs are white, 32 ⫻ 25.5 mm, the clutch is 2 (Rand 1942b). The nest is attended to by the male (AMNH 339693). Incubation and nestling periods are unknown.

Pheasant Coucal Centropus phasianinus (Latham, 1801) Cuculus phasianinus Latham, 1801, Index Ornithologicus, Suppl.: xxx. (New Holland [⫽ New South Wales]) Polytypic. Six subspecies. Centropus phasianinus phasianinus (Latham, 1801); Centropus phasianinus spilopterus G. R. Gray, 1858; Centropus phasianinus nigricans (Salvadori, 1876); Centropus phasianinus thierfelderi Stresemann, 1927; Centropus phasianinus propinquus Mayr, 1937; Centropus phasianinus mui Mason et al. 1984. Other names: Australian Pheasant Coucal, Swamp Pheasant, Kai Coucal, New Guinea Pheasant Coucal, Common Coucal (New Guinea).

Description ADULT: (C. p. phasianinus): Sexes similar, breeding plumage, the head, upper back and underparts dull black, feather shafts glossy blacks, wing coverts rufous brown with black and cream barring and bold white streaks, wing flight feathers with inner webs rufous, outer webs barred buff, black and rufous brown, showing a broad dark trailing edge in flight, back, rump and upper tail coverts blackish brown with fine pale barring, tail long, black with buff bars, the broad bars sometimes tricolor with rufous bars inserted into the black bars, and the rufous bars sometimes spotted with black, tail feathers broad and rounded with a white tip; under wing coverts black (greater coverts) to rufous brown with fine blackish bars (lesser coverts).

Eye-ring narrow and blackish, iris red (breeding males) to pale brown (nonbreeding males), orange to yellow (breeding females), or whitish (nonbreeding females), bill black (breeding) to pale horn (nonbreeding season), feet black. Nonbreeding plumage, the upperparts rufous brown with buff streaks, tail brown with gray bars and feathers pointed at the tip, tail longer than in breeding plumage, underparts buff with paler streaks. In New Guinea and Kai Island subspecies, upperparts and underparts dull black, wing coverts and flight feathers sometimes with buff spots and bars, long tail black glossed green, under wing coverts black; iris red, bill black, feet black. Nonbreeding plumage, perhaps restricted to first year, upperparts dull black with buff streaks on head and upper back, wing blackish with narrow buff bars, tail black with narrow buff bars, chin whitish, throat and breast to belly whitish buff with straw-colored rachi, outer breast with fine black spots and bars, flanks black, lower belly and under tail coverts blackish with buff bars, bill pale yellowish (AMNH 628289). Birds molt from juvenile plumage into this plumage, and they molt into the black adult plumage. JUVENILE: Plumage black and rufous, head nearly all black (not streaked), back black with whitish buff spots or bars, wing rufous with narrow black bars, outer vane of primaries buff with black and rufous bars, inner vane rufous with whitish bars, tail

258 Pheasant Coucal Centropus phasianinus blackish with rufous bars, chin and center of throat whitish with dark shaft streaks, throat to belly blackish with whitish spots on throat and whitish mottled bars on breast and belly; iris brown, bill brown to pale below. Regional variation in plumage and size is less than in the adults in New Guinea and Kai Islands, plumage blackish with whitish buff spots and bars, feather shafts on head and upper back glossy black, tail black with narrow buff bars, chin white, underparts blackish with white bars, overall appearance black with whitish spots; iris gray to brown. NESTLING: Skin black with pink abdomen, upperparts with white hair-like natal down, most conspicuous on top of head and neck where the bristles are directed forward; palate red with pale patch of papillae on either side of the palate cleft, tongue red with a white papillated rear edge and a black U-shaped band near the tip. SOURCES: AMNH, BMNH, CM, FMNH, MSNG, MSNM, MZB, RMNH, ROM, SMTD, UMMZ,WAM, ZMB.

History and subspecies Centropus spilopterus was formerly considered a species restricted to the Kai Islands. Salvadori (1876, 1881) described the form nigricans of New Guinea as a distinct species as it had pale marks on the wing. New Guinea coucals were known as C. spilopterus beginning with Sharpe (1878); Salvadori recognized them as distinct species, C. nigricans. Later works considered the New Guinea forms (nigricans and others) conspecific with Australian Pheasant Coucal C. phasianinus but did not discuss the Kai Is birds (Stresemann 1927b, Mayr 1937, 1941, Mayr and Rand 1937, Mason et al. 1984). Most Kai Island and New Guinea birds are nearly identical in adult plumage pattern, black with pale wing marks (the marks are dark rufous in thierfelderi). White and Bruce (1986) likewise noted that some Kai birds have buffy barring on the flight feathers. Feathered nestlings and juveniles in the Kai Islands (AMNH, MSNG) have a whitish throat and are barred with buff and tan, like the young in New Guinea and Australia. Both spilopterus and nigricans are nearly

identical in mitochondrial gene sequence to C. phasianinus of Australia (Figure 5.3). C. p. spilopterus G. R. Gray, 1858; large size, tail unbarred black, primaries unbarred black, sometimes with pale bars, these narrower than the black bars on the wing, underwing coverts black; Kai Islands; C. p. nigricans (Salvadori, 1876) (includes C. s. obscuratus Mayr 1937); smaller than spilopterus, tail black with narrow indistinct buff bars, primaries black with narrow buff bars (buff bars 1–2 mm, black bars 7–9 mm); southeastern New Guinea, Yule I, D’Entrecasteaux Islands; C. p. propinquus Mayr, 1937; smaller than nigricans, plumage similar to nigricans; northern New Guinea; C. p. thierfelderi Stresemann, 1927; tail black or black with narrow rufous bars, primaries with rufous bars nearly as wide as black bars (4–6 mm, vs 5–8 mm), most birds have rufous to buff bars on the primary coverts, under wing coverts black with buff edge; southern New Guinea. C. p. phasianinus (Latham, 1801); tail black with broad rufous bars, primaries with rufous bars wider than the black bars, breeding plumage with breast black, nonbreeding plumage streaked brown and buff; Australia; C. p. mui Mason et al. 1984; breast white; East Timor (Los Palos district). In Australia, northern and northwestern birds are often recognized as a distinct subspecies, C. p. melanurus (Mason et al. 1984, Mason 1997). Goodwin (1974) and Storr (1977) questioned whether more than one subspecies occur in Australia; Mason et al. (1984) and D. Rogers (in Higgins 1999) found barring on the rectrices to differ between western and eastern birds. Birds in northeastern Australia intergrade with birds south of Ayr, Queensland, and intermediates occur in central coastal Queensland from 20° to 23°S (Mason et al. 1984, Ford 1987a,b, Higgins 1999). The form melanurus is not recognized, because northern and northwestern populations overlap those in eastern and southern Australia in size and in the pattern on the tail, which tends to have wider black bars in the north and northwest (AMNH). Australian birds have a seasonal change of plumage in the adult; birds in

Pheasant Coucal Centropus phasianinus 259 New Guinea and the Kai Islands are not known to have this, although they have a streaked nonbreeding plumage which is worn between the juvenile and first breeding adult plumage; these birds have been considered a distinct species, the New Guinea Pheasant Coucal C. spilopterus or C. nigricans. The Timor form C. p. mui is known only from the type specimen. Although White and Bruce (1986) suggest that mui is a distinct species, variation in the extent of white in plumage within other species of coucals, some with white color morphs (C. goliath, C. ateralbus, C. viridis) is consistent with the white breast in C. p. mui representing a variant within C. phasianinus. The nonbreeding adult plumage is well known in C. phasianinus in Australia; it has not previously been recognized in coucals in New Guinea and the Kai Islands (Mason et al. 1984). Juvenile spilopterus, nigricans and thierfelderi have black plumage marked below with spots and bars of white and buff, and head and upper back black or dark brown with dark feather shafts; the feathers of the crown are sometimes broadly spotted with buff. Nonbreeding adult plumage in these forms has straw-colored streaks on the head and upper back, the streaks being the shaft and base of the barbs; the underparts are pale with straw-colored shaft streaks on the breast, as in nonbreeding adult C. p. phasianinus in Australia. For C. p. thierfelderi, AMNH 425934 is a specimen in molt from mottled black plumage to straw-colored hackles; AMNH 425934 and 421865 are specimens in molt from straw-colored hackles to breeding plumage. For C. p. nigricans, AMNH 781441 is a worn juvenile in molt to non-breeding adult plumage, and AMNH 628289 and MSNG 13975 are birds in straw-hackled non-breeding adult plumage; breeding plumage adults include MSNG 13972, 213973, 13974 and 13976 and AMNH 628288, all in Salvadori’s type series. For C. p. spilopterus, fledged juveniles include AMNH 628279 and MSNG 13969, nonbreeding adults include AMNH 628276, and most birds in AMNH, MSNG, RMNH, SMTD, ZMA are in adult breeding plumage.A black spilopterus with pale marks on the wing, MSNG “d” ⫽ 13968, that Salvadori (1881) thought was a juvenile (‘giovani’), is adult.

Measurements and weights C. p. spilopterus: Wing, M (n ⫽ 9) 212–241 (228.9 ⫾ 11.2), F (n ⫽ 4) 249–268 (256.0); tail, M 271–330 (296.6 ⫾ 18.2), F 290–334 (324.3); bill, M 36–42 (38.5 ⫾ 1.8), F 36–42.5 (39.0); tarsus, M 44–52 (49.6 ⫾ 2.3), F 48–53 (50.8); hallux claw, M 23–27 (25.0), F 21.5–28 (27.3) (AMNH, BMNH, MSNG, RMNH, SMTD, ZMA); C. p. nigricans: Wing, M (n ⫽ 10) 200–229 (216.8 ⫾ 9.8), F (n ⫽ 10) 216, 220, 230–248 (236.3 ⫾ 11.5, including the two small birds, perhaps mis-sexed); tail, M 288–326 (299.0 ⫾ 22.4), F 280, 318, 326–354 (332.0 ⫾ 21.9, including the two small birds, perhaps mis-sexed) (AMNH, MSNG); C. p. propinquus: Wing, M (n ⫽ 5) 194–219 (202.4 ⫾ 10.1) (excludes “M” 248), F (n ⫽ 5) 207–237 (228.8 ⫾ 12.9) (excludes “F” 197); tail, M 260–301 (280.6 ⫾ 18.5), F 287–332 (308.6 ⫾ 17.8) (AMNH); C. p. thierfelderi: Wing, M (n ⫽ 10) 205–222 (214.3 ⫾ 5.8), F (n ⫽ 8) 222–251 (238.3 ⫾ 10.6) (excludes “F” 218); tail, M 264–338 (289.9 ⫾ 21.6), F 282–326 (317.3 ⫾ 19.2) (excludes “F” 278) (AMNH); C. p. phasianinus, Cape York: Wing, M (n ⫽ 9) 217–246 (237.3 ⫾ 9.6), F (n ⫽ 15) 247–274 (260.0 ⫾ 8.7); tail, M 330–378 (357.1 ⫾ 17.9), F 326–418 (376.1 ⫾ 29.3); hallux claw, M 20–27 (23.3 ⫾ 2.6), F 20–26 (23.2 ⫾ 1.4) (AMNH); NSW to central Queensland south of 21°S: Wing, M (n ⫽ 23) 205–248 (225.6 ⫾ 10.3), F (n ⫽ 31) 235–267 (249.3 ⫾ 8.3); tail, M (n ⫽ 13, breeding plumage) 282–342 (310.2 ⫾ 18.7), (n ⫽ 6, nonbreeding plumage) 323–356 (338.3 ⫾ 12.3), F (n ⫽ 12, breeding plumage) 305–361 (337.1 ⫾ 21.2), F (n ⫽ 9, nonbreeding plumage) 333–393 (360.2 ⫾ 22.1); bill, M 33.1–37.2 (35.5 ⫾ 1.3), F 35.2–42.6 (38.6 ⫾ 1.7); tarsus, M 46.3–53.8 (50.5 ⫾ 1.6), F 49.5–56.7 (53.2 ⫾ 1.8) (Higgins 1999); Western Australia: Wing, M (n ⫽ 10) 230–264 (245.2 ⫾ 10.2), F (n ⫽ 7) 260–283 (266.6 ⫾ 8.9); tail, M 348–376 (360.6 ⫾ 16.2), F 360–445 (384.4 ⫾ 26.5) (WAM); Pilbara to Cape York and tropical Queensland: Wing, M (n ⫽ 41) 227–254 (240.7 ⫾ 7.2), F (n ⫽ 40) 241–303 (268.8 ⫾ 12.9) (Higgins 1999);

260 Pheasant Coucal Centropus phasianinus C. p. mui: Wing, M (n ⫽ 1) 250; tail, M 377 (in partial molt) (Mason et al. 1984). In C. p. phasianinus, tail length differs between breeding and nonbreeding adults, longer in the nonbreeding plumage, due to a double annual molt of T1; the other rectrices molt only once a year (Mason et al. 1994, Higgins 1999). Weight, C. p. propinquus, M (n ⫽ 3) 180–225 (205), F (n ⫽ 1), 300 (Hartert 1930); C. p thierfelderi, M (n ⫽ 2) 250–281 (265.5), F (n ⫽ 1) 375 (Mees 1982); C. p. phasianinus: NSW to Central Queensland, M (n ⫽ 16) 200–364 (302.1), F (n ⫽ 22) 242–520 (444.9); Pilbara to Cape York, M (n ⫽ 13) 289–378 (327.6), F (n ⫽ 12) 390–600 (483.0) (Higgins 1999). Wing formula, C. p. spilopterus, P6 ⫽ 5 ⫽ 4 ⬎ 7 ⬎ 3 ⬎ 8 ⬎ 2 ⬎ 1 ⬎ 9 ⬎ 10; C. p. phasianinus, P5 ⫽ 6 ⬎ 7 ⫽ 4 ⬎ 3 ⬎ 8 ⬎ 2 ⬎ 1 ⬎ 9 ⬎ 1.

Field characters Overall length 60–80 cm. Large coucal, a skulking long-tailed pheasant-like bird, in Australia rufous above and the head, neck and underparts glossy black in breeding season, streaked above and below in nonbreeding plumage, long barred tail; in New Guinea and the Kai Islands, dull black plumage, often with pale spots on wing coverts and flight feathers, and a long, glossy green tail.

Voice Dull rapid resonant booming series of “hooh” notes, falling then rising in pitch, “coo-coo-coocoo-coo-coo-coo-coocal”, accelerates in tempo, with 20 or more notes in a series and lasting up to 6 sec. Members of a pair often sing in duet. Also gives a softer series of “coo” notes all on same pitch and lasting 3–4 sec. Other calls include harsh scolding “keouw”, a dull thud “pthuk”, a double note “nah-oo” used when danger appears, a grunt when moving through undergrowth and approaching a nest, and a hiss when alarmed. Begging calls of the young sound like shimmering, buzzing, or a loud clicking; and nestlings give a snake-like hiss “pssch” (Slater 1971, Mackness 1979, Buckingham and Jackson 1990, Taplin and Beurteaux 1992, Higgins 1999). Calls in New Guinea are like calls in

Australia (Diamond 1972, Coates 1985, 2001, Beehler et al. 1986, Coates and Bishop 1997).

Range and status New Guinea, Kai Islands, Australia and Timor. Resident. Conspicuous in breeding season. Kai Islands (Kai Kecil, Kai Besar, Roa). In New Guinea, local in the west in Irian Jaya (in the south at Merauke and Pulau Kinaan, in the north from Mamberano River to Jayapura at Humboldt Bay), in the east widespread in the lowlands; D’Entrecasteaux Islands (Goodenough, Fergusson and Normanby), Yule I and nearshore islands in the Torres Strait (Boigu, Daan, Saibai) (Hartert 1930, Mayr 1941, Mees 1982, Draffan et al. 1983, Beehler et al. 1986, White and Bruce 1986, Sibley and Monroe 1990, Coates and Bishop 1997, Mason 1997, Coates 2001). Common in lowlands. The Kai Islands form has been considered near-threatened (Collar et al. 1994). Not reported on the remote islands in the Torres Strait, and there is no evidence of dispersal between New Guinea and Australia. In Australia, common near the coast, uncommon in semiarid areas and scarce in the arid interior (Blakers et al. 1984, Ford 1987b, Mills 1987, Higgins 1999); in Western Australia occurs south to Pilbara Region (Storr 1984a); the last records in the Gascoyne Region were in the late 1800s (Storr 1985). In recent years, coucals have decreased in numbers in Australia (Seventy and Whittell 1976, Johnstone and Storr 1998, Higgins 1999). Population density, in southern New Guinea woodlands, c. 0.1 bird /ha (Bell

Pheasant Coucal Centropus phasianinus 261 1982a); in Australia, 0.02–0.4 birds / ha, highest in Kakadu NP (Higgins 1999).

Rand 1937, Cooper 1953, Frauca 1974, Coates 1985, Higgins 1999).

Habitat and general habits

Displays and breeding behavior

In New Guinea, coucals occur in the forest edge, lightly wooded cultivation, savanna, remnant forest patches and scrub, common in cane and grass, also in reed-bed; found in lowlands below 500 m, and locally in open highland valleys as high as 1500 m and 1800 m (Diamond 1972, Mason et al. 1984, Beehler et al. 1986, Coates 2001). In Australia, they live in dense riverine vegetation, long grass, rank herbage, coastal heathlands, margins of swamps, canefields, thickets of Melaleuca, Lantana and Pandanus, thickets of vines, mangroves, secondary forests, spinifex in sandstone country, sandstone gorges, dense grass along roads, and gardens. In Timor, they are in swampy grassland, monsoon forest and fringing grasslands (Mason et al. 1984). The coucals live singly and in pairs. They are mainly terrestrial, and run on the ground, with body low, neck stretched and head low and held forward. They climb into trees and perch, where they call from horizontal branches in midstory and lower crown at forest edge. They forage by clambering about on the ground and through thickets in search and pursuit of large insects. When disturbed they run or fly into cover (Bell 1970, Mackness 1979, Coates 1985, Coates and Bishop 1997, Higgins 1999). They search for food as they walk in low dense vegetation and ground cover, then run down the prey in “flush and rush” behavior, tail extended and body held low. Like other cuckoos, they are conspicuous when they sunbathe, spreading the back feathers to expose the skin, and spreading the wings and tail to dry after a rain or after foraging in wet vegetation. When feeding on large prey, such as a large snail, the bird holds the prey under its foot and tears it apart with its bill.

Socially monogamous, adults live in pairs, territorial in the breeding season. Coucals perch in the open when they call in the morning and late afternoon. Both members of a pair call in duet (Taplin and Beurteaux 1992, Higgins 1999). The male sometimes carries food and feeds the female as they mate. In courtship feeding, one bird holds a food item in its bill while it raises and spreads its tail, then moves onto the partner and mates, feeds the mate and then flies into tall grass. In copulations when no courtship feeding is observed, the male crouches behind the female, moves his head from side to side or up and down as he follows her; she runs ahead, keeping her body low, then the male straightens his body and drags his wings along the ground and spreads his tail, the female raises her body and the male mounts. Courtship feeding has been seen in New Guinea and Australia (Crawford 1972, Mackness 1979, Coates 1985, Higgins 1999). In Australia, both female and male incubate the clutch, and both feed the young; the male does most feeds (87% of 151 feeding visits). Both parents remove fecal sacs from the nest (Taplin and Beurteaux 1992).

Food Large insects (grasshoppers, crickets, mantids, stick insects, caterpillars, beetles, bugs), Sesarma mangrove mud crabs, spiders, scorpions, snails; frogs, lizards, snakes, nestling birds, small mammals including rodents and bandicoots (Mathews 1918, Mayr and

Breeding In the Kai Islands a nestling was taken in July (Hartert 1903) and fledglings with 12-mm short tails were taken in April and September (AMNH). In southern New Guinea (Fly River to Port Moresby lowlands and the Sogeri Plateau), adults carry nest material in May (Coates 1985), a nest had young in June (Bell 1969), nests have eggs in September (Coates 1985), tail-less fledged young appear in early November (Coates 1985) and a female had an egg in the oviduct in late December (AMNH 425929), all indicating a long period of nesting in the dry season. In Australia they breed near Sydney from October to December (Cooper 1953), in Queensland from September to May (Storr 1984b), in Kimberley Division from November to March (Storr 1980), in Northern Territory from December to April (Storr 1977).

262 Sumatran Ground-cuckoo Carpococcyx viridis The nest is a covered mass of grass, open at either end, like a shopping basket with a broad handle over the top. It is begun as an open platform, then the blades and stems are drawn together to form a cover, then a nest lining is added below with sticks as long as 25 cm, then fibers and leaves. It sometimes has an entrance tunnel or a platform of leaves near the entrance.The nest is built in a grass tussock near the ground or in a bush or pandanus. Eggs are white, becoming brown with stain; in New Guinea 36 ⫻ 27 mm, in Australia 38 ⫻ 29 mm; clutch 2–7, usually 3–5 (Hartert 1903, Hindwood 1942, 1957, Cooper 1953, Schönwetter 1964, Frauca 1967, Pratt 1972, Beruldsen 1980, Storr 1984a,b, Taplin and Beurteaux 1992). Adults incubate with the first egg, and they add green leafy twigs through incubation and after young have hatched. Incubation period is 15 days. Nestlings hatch asynchronously, a day apart, with long white

hair-like down on the top of head, also on back; skin black, gape bright red, edge of tongue black. Nestling pinfeathers break through the skin at 6 days when the pins are 2.5 mm long on the wing, at 11 days, feather brushes emerge from pins; the eyes open at 7 days (Frauca 1967, Pratt 1972, Hoppe 1988,Taplin and Beurteaux 1992). In New Guinea a short-tailed young was carried by the parent in flight as the adult held the body of the young in its claws (Bell 1969, 1970, 1984). Nestlings weigh 18.6 g at hatching and 123.5 g at fledging (male?), or 41% of adult male weight. Undisturbed nestlings fledge in 17 to 21 days (Cooper 1953). The natal down is gone within a week of fledging, and the young are independent at 40 days.A female may lay several clutches and rear three to four broods in a season (Mackness 1979). Breeding success, of 14 nestlings, 11 (79%) fledged, others were taken by a predator (Frauca 1967,Taplin and Beurteaux 1992).

Couinae Genus Carpococcyx G. R. Gray, 1840 Carpococcyx G. R. Gray, 1840. A List of the Genera of Birds, p. 56. Type, by monotypy, Calobates radiceus Temminck 1832.The genus name refers to the fruiteating behavior (Gr. karpos, fruit; kokkux, the cuckoo)

and the birds are also known as fruit-cuckoos. The genus is characterized by large body size, bare skin on the face, and gray plumage with iridescent gloss; the group is restricted to the Asian tropics.Three species.

Sumatran Ground-cuckoo Carpococcyx viridis Salvadori, 1879 Carpococcyx viridis Salvadori, 1879, Annali del Museo Civico di Storia Naturale di Genova, 14, p. 187. (Sumatra) Other common names: Sunda Ground-cuckoo (in part), Green-billed Ground-cuckoo (in part). Monotypic.

Description ADULT: Sexes alike, crown black shading to green on hind crown, upper back, wing coverts and secondaries dull green, lower back and rump dark rufous with fine black bars, wing and tail glossy

greenish-black; underparts, black patch from chin to under the eye extends to side of crown behind the bare facial skin, throat and upper breast dull greenish gray, lower breast and belly rufous buff finely barred with black, flanks and under tail coverts rufous barred black, under wing coverts brown; large area of bare skin on face, above the eye pale green, behind eye lilac and above the cheek crimson red, iris red, bill blackish above and pale green below, tarsi green. JUVENILE: Different from adult, upperparts dark rufous with indistinct black bars, rump paler rufous

Sumatran Ground-cuckoo Carpococcyx viridis 263 with black bars, wing coverts barred rufous and dark brown, alula black, feathers with a broad buffy brown edge, wing dark brown with feathers edged rufous brown, tail dark brown; underparts with the throat and breast rufous indistinctly barred with black, belly rufous indistinctly barred whitish and black, under tail coverts rufous barred black; bare skin around eye less extensive than in adult, iris dark gray, bill black, feet dark gray. NESTLING: Undescribed. SOURCES: MSNG, RMNH, ZRCNUS. The bird was first collected in Sumatra, a specimen that Schlegel identified as the previously unknown female of the Bornean Ground-cuckoo C. radiatus. When Salvadori obtained specimens taken by Odoardo Beccari on Gunung Singgalang in the Padang highlands, he described the bird as distinct from C. radiatus (Salvadori 1879, Finsch 1898). Sumatran Ground-cuckoos are known from only nine museum specimens (MSNG, RMNH, SMNS, ZRC; Arbocco et al. 1978, Collar and Long 1996, BirdLife International 2001) and had not been seen by western ornithologists since 1916, the last known specimens, when they were rediscovered in a forest in 1997 (Zetra et al. 2002).

Measurements Wing, M (n ⫽ 2) 210 (210), F (n ⫽ 1) 210, U (n ⫽ 2) 206, 211; tail, M 245–261 (253), F 264, U 256, 258; bill, M 40–44 (42), F 36, U 37, 34.1; tarsus, M 69–70 (69.5), F 78, U 68, 71 (MSNG, RMNH, ZRC); the birds in juvenile plumage are smaller. Wing formula, P5 ⫽ 4 ⫽ 3 ⫽ 2 ⱖ1 ⬎ 6 ⬎ 7 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 55 cm. Large ground-cuckoo with green back, black chin, gray throat and long tail. Face has less black than Bornean Ground-cuckoo C. radiatus, sides of neck and breast are greenish gray (gray in C. radiatus), back and wing coverts are darker green than in C. radiatus, bars on abdomen are narrower, sides of breast and vent rufous, and the bird is smaller than C. radiatus.

Voice Unknown.

Range and status Southwestern Sumatra. Occurs in foothills of the Barisan Range in the southern half of the island; its known localities are Gunung Singgalang, Padang highlands, Muara Sako, Rimbopengadang and Gunung Dempo (BirdLife International 2001). Originally described and recognized as a species, then as a subspecies of Bornean Ground-cuckoo, and recently again a distinct species (Shelley 1891, Finsch 1898, 1901; Robinson and Kloss 1923b, Peters 1940, van Marle and Voous 1988; Collar and Long 1996). Resident. Ground-cuckoos are hard to see in the forests, and although biologists looked for the birds for many years, they were not seen again until November 1997, when one was found alive in a mammal trap, photographed and released, in the Bukit Barisan Selatan NP, Tanggamus District, Lampung Province (Zetra et al. 2002). Later that day a ground-cuckoo seen running on the ground was probably the same bird. Another was reported in Bukit Rimbang-Baling Wildlife Sanctuary in Riau Province in October 2000. The observations let us take an optimistic view that they are survivors. At least a third of the montane forest and most lowland forest in Sumatra has been lost in the past century, and forest loss together with large-scale movement of human populations into its range is the most serious threat to the existence of this cuckoo. The Sumatran Ground-cuckoo is critically endangered (BirdLife International 2001).

264 Bornean Ground-cuckoo Carpococcyx radiatus

Habitat and general habits Forest of hilly areas, between 300 and 1400 m. Sites where specimens had been collected were tall forest; at Gunung Dempo tall trees with fairly open undergrowth, at Muara Sako extensive, and Rimbo Pengadang included both primary and secondary forest (Robinson and Kloss 1923b). Sumatran forest habitat is currently being degraded at an alarming rate. The 1997 sighting was in primary tropical evergreen rainforest on sloping hills c. 500 m above sea level, the understory with palms, pandanus, large ferns and rattan, and the most numerous trees Stemonurus secundiflora, Pterocymbium tubulatus, Dillenia indica, Dipterocarpus kunstleri, Xanthophyllum excelsum, Cinnamomum parthenoxylon, Adina polycephala and Pometia pin-

nata.The site is near a coffee plantation.The nominally protected site is losing ground to widespread poaching, human encroachment, and timber exploitation; both commercial and smallscale logging is ongoing. On the rich volcanic soils of these bird habitats, cinnamon plantations and cabbages are replacing forests (BirdLife International 2001, Zetra et al. 2002).

Food Insects (Finsch 1898, Robinson and Kloss 1923b).

Breeding A juvenile was taken in September (MSNG), suggesting a successful breeding in the first half of the year. No other breeding information is known.

Bornean Ground-cuckoo Carpococcyx radiatus (Temminck, 1832) Calobates radiatus Temminck, 1832, Nouveau recueil de planches coloriées d’oiseaux, livre 91, pl. 538. (Pontianak district, western Borneo). Other common names: Sunda Ground-cuckoo (part), Green-billed Ground-cuckoo ( part), Radiated Fruit-cuckoo. Monotypic.

barred buff, underparts unbarred pale rufous, flanks rufous with black bars. NESTLING: Undescribed. SOURCES: AMNH, BMNH, MCZ, MZB, RMNH, ROM, SMTD, UMMZ, USNM, ZMA, ZMB.

Description ADULT: Sexes alike, head black glossed purple, back and wing coverts dull green with purple gloss and coppery red reflections, rump dark rufous with indistinct black bars, wing and tail unbarred coppery violet, sides of neck gray, lower back and rump rufous with blackish bars; underparts, chin and throat black, upper breast gray, breast and belly to under tail coverts white finely barred black, under wing coverts dark rufous; large area of bare skin on the face around eye bluish green, iris brown or gray, or gray with a narrow brown outer circle, bill green, mouth lining pale pink, tarsi and toes green. JUVENILE: Similar to adult but crown dark brown, fluffy at base of feathers, back brown, alula black, feathers have a broad buffy brown edge, wing coverts

The species was first collected in Borneo in 1826 by French naturalist P. Diard. It was described by Temminck and figured in a color plate as Calobates radiceus. Temminck himself corrected the name to Calobates radiatus in his 1838 edition of Nouveau recueil de planches coloriées d’oiseaux in the Tableau Méthodique index,Vol. 1, p. 53.

Measurements and weights Wing, M (n ⫽ 15) 240–272 (258.9 ⫾ 10.4), F (n ⫽ 8) 242–272 (252.5 ⫾ 8.9); tail, M 290–326 (303.7 ⫾ 9.8), F 286–326 (299.3 ⫾ 12.0); bill, M 44–56 (47.5 ⫾ 3.0), F 43–53 (47.6 ⫾ 3.5); tarsus, M 78–94 (85.5 ⫾ 4.3), F 82–95 (87.9 ⫾ 5.2) (AMNH, BMNH, USNM). Weight, M (n ⫽ 1) 455, F (n ⫽ 1) 540 (MZB).

Bornean Ground-cuckoo Carpococcyx radiatus 265 Wing formula, P7 ⫽ 6 ⫽ 5 ⫽ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⫽ 8 ⫽ 9 ⬎ 10.

Field characters Overall length 60 cm. A large ground cuckoo with green back, black chin and throat and gray breast, barred whitish underparts, and long tail. Juvenile is brown above and pale rufous below.

Voice Four calls are known for this ground-cuckoo: (1) the main long-distance call, a repeated lowpitched double note, “thook-torr” or “koohoo”, nearly on one pitch with the first note rising and the second note falling, first note about 0.5 kHz, second note longer rising from 0.4 to 0.5 kHz, then drops to 0.3 kHz, sounding like a lowpitched Koel; (2) a rolling one-note “torrmmm” given in response to playback of its long-distance call; (3) a sharp alarm,“ark”, given along with vertical movements of the tail, perhaps the same as the coughing call “khaaa” or “heh heh heh”, given when a bird is disturbed, an apparent alarm call; and (4) a complex call given when two birds are together, a harsh call followed by dove-like descending cooing and lamb-like bleating, perhaps a combination of calls (3), (1) and the bleat (Davison 1979, MacKinnon and Phillipps 1993, Laman et al. 1997, Holmes 1997, Long and Collar 2002, Ben King).

Range and status Borneo. Resident. Ground-cuckoos are widespread but they keep a low profile. They are wary, hard to observe, and occur at low densities (Finsch 1898, Büttikofer 1900, Davison 1979, Smythies 1981, Sheldon et al. 2001). They are known from 49 localities in Borneo, including 10 in Sabah, 15 in Sarawak, four in Brunei, and 20 in Kalimantan. In Sabah they live from sea level to 50 m in low-elevation primary forest, (Sheldon et al. 2001). Recent observations of ground-cuckoos are known from Danum Valley in Sabah, Gunung Mulu NP and Similajau NP in Sarawak and Gunung Palung NP in West Kalimantan. Most recent observations are of cuckoos responding to playbacks of their recorded calls and cuckoos

captured in snares set for galliform birds (Long and Collar 2002). During seven years of forest fieldwork and surveys by three observers in Gunung Palung NP, the ground-cuckoos were seen only six times (Laman et al. 1997). They are generally absent from fragmented patches of forest (Fogden 1976, Long and Collar 2002).Vulnerable to loss of large stands of forest and to hunting, they are considered near-threatened (BirdLife International 2001).

Habitat and general habits Forest, floodplain alluvial forest in Sarawak, lowland and hill forest in Brunei, primary lowland forest and perhaps on limestone soils in Sabah, pristine lowland riverine dipterocarp forest and low swampy forest in West Kalimantan. In Sarawak they live in dry, level forest in Gunung Mulu NP (Davison 1979) and Similajau NP (Duckworth et al. 1996). In Sabah they have been seen near Baturong Caves (Sheldon et al. 2001). Although nearly all records are in lowland forest, they may occur in upslope forests as well (Long and Collar 2002). Terrestrial, they live on the ground.They run and jump, perch on logs and in trees to call, and roost in trees. Ground-cuckoos sometimes follow feeding groups of bearded pig Sus barbatus that dig in soil; the bird snatches food items from the upturned earth (Smythies 1981, Davison 1979, Laman et al. 1997). Three of five local names of ground-cuckoos translate as “pig birds” (Banks 1935, Smythies and Davison 1999). Ground-cuckoos also follow foraging sun bear Helarctos malayanus in search of termites (Long

266 Coral-billed Ground-cuckoo Carpococcyx renauldi and Collar 2002). In threat display the birds gape widely and spread both wings and the tail (Davison 1979), much as in C. renauldi (Rinke 1999).

Food Mainly insects, including beetles, giant ants and termites; they also take fruit and are known as fruitcuckoos (Finsch 1898, Davison 1979). Captive birds take mixed vegetables, raw meat, insects, dead mice, cockroaches and fish (Beddard 1901, Long and Collar 2002).

Breeding Season unknown.The nest is undescribed, although birds have been reported to build a nest (Shelford 1916). Eggs are white, 47 ⫻ 35 mm (Schönwetter 1964), laid by a captive bird, perhaps the one kept in the London Zoo and later examined by Beddard (1901). The clutch size, incubation and nestling periods are unknown. A juvenile was seen accompanied by an adult ground-cuckoo in a party of bearded pigs (Laman et al. 1997), and this is the best evidence that these ground-cuckoos provide parental care for their young.

Coral-billed Ground-cuckoo Carpococcyx renauldi Oustalet, 1896 Carpococcyx Renauldi Oustalet, 1896, Bulletin du Museum National d’Histoire Naturelle, Paris, 2, p. 314. (Province of Quang-tri, Annam) Monotypic. Other common names: Renauld’s Groundcuckoo, Annam Ground-cuckoo.

Description ADULT: Sexes alike, head, neck and throat to upper breast black; back gray, indistinct whitish band behind the black neck, rump unbarred or vermiculated dark rufous, primaries and tail black glossed purplish, secondaries grayish green, secondaries 1–6 dull purplish near tip; below, center of lower breast to belly white, flanks and legs whitish finely vermiculated with dark gray, under tail coverts gray, under wing coverts dark rufous; bare skin around eye red and violet, iris dull orange to yellowish or pink, bill dusky red, rictal area violaceous blue, feet red. JUVENILE: Forehead and crown brown, back dark gray barred rufous, rump brown, tail unbarred black, primaries black glossed purplish, secondaries and upper wing coverts dark gray with tips rufous, face brown (gray around eye), chin to upper breast unbarred gray to rufous brown, belly gray, under tail coverts rufous finely barred with black; bare facial skin gray, iris light brown, bill blackish brown, feet blackish brown.

NESTLING: Naked at hatching, skin brown, at three or four days feather quills show on wing and body, feather sheaths dark gray; palate red with white markings, comprising two raised, broad white marks on either side of the palate and a raised U-shaped shield behind that, and an arc of small white papillae anterior to the broad white marks, tongue red with a black mark in the middle; palate colors fade by the age when the bird is independent of parental care. SOURCES: AMNH, BMNH, MCZ, RMNH, UMMZ, USNM.

History The species was discovered in 1895 by Père R. P. Renauld, a missionary in Quangtri Province, Annam (Vietnam). He sent three skins to the Paris museum, where Oustalet described the species and later gave a descriptive account (Oustalet 1899). Delacour heard that the cuckoos were being captured and reared for food, and they did well in captivity. His expedition from 1926–1927 (Delacour 1927) was the first to observe the birds in the field and watch their feeding behavior and displays, and some were taken to his aviaries for further behavior observations.

Measurements and weights Wing, M (n ⫽ 12) 260–290 (278.4 ⫾ 9.5), F (n ⫽ 7) 266–294 (278.2 ⫾ 9.7); tail, M 294–348 (325.1 ⫾ 16.2), F 310–362 (330 ⫾ 17.0); bill, M 44–49

Coral-billed Ground-cuckoo Carpococcyx renauldi 267 (46.3 ⫾ 1.7), F 43–47 (44.8 ⫾ 1.7); tarsus, M 87–95 (90.4 ⫾ 2.6), F 82–95 (89.0 ⫾ 5.2) (AMNH, BMNH, MCZ, USNM). Weight, (n ⫽ 1) 400 (captive, nearly grown fledged young 44 days after hatching) (Robiller et al. 1992). Wing formula, P5 ⫽ 4 ⫽ 3 ⫽ 2 ⬎ 1 ⬎ 7 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 65 cm. Large ground cuckoo with black head and throat and gray back, and long tail glossed violet, below white finely barred with black.

Voice Territorial call is a loud, mellow, moaning whistle, “woaaaah”, “wooaa” or “wohaaau” which rises from 0.9 kHz to 1.0 kHz then drops to 0.9 kHz and lasts 1 sec, repeated again in 2–3 sec (White 1984, Scharringa 1999, Robson 2000a, NSA). In calling, the male perches on a shrub or tree, opens and closes his wings repeatedly and nods his head up and down (Wylie and Shelton 1982). The moaning call gives the bird its Lao name of “ghost chicken” (P. Round). Another call is given in duet with one bird giving a 3-note call “whup, whooup” the other bird a rolling gargle (Lekagul and Round 1991), or with male and female alternating similar calls (Hughes 1997c). The male gives a bill clack as the last sound of “coo-cuh-clack” (Atkinson 1982), and the female gives a quiet “blirrrrrrrrr” while vibrating the body and pumping the tail (Hughes 1997c).

Range and status Southeast Asia: local in continental Thailand, Laos, Cambodia, Vietnam (north and central Annam). Resident. Scarce to locally common (Evans et al. 2000, Robson 2000a), they were considered nearthreatened by Collar et al. (1994).

Habitat and general habits Broadleaf evergreen forests (lower and middle strata), second growth, scrub, vine, dense cover on ground, rocky areas.They are terrestrial and diurnal. They are shy, run from disturbance and sometimes

fly from cover. Best seen when they come to feed in certain favored sites, year after year (Ben King, Phil Round). In Khao Yai, Thailand, ground-cuckoos come to refuse tip behind a local restaurant, feed on noodles, rice and perhaps insects. Another site is an insect-rich drainage pit below a camp latrine. In observation at feeding site, a bird stands motionless for 5-10 minutes, then leaps and jumps with spread wings into the drainage bed. In feeding, it reaches neck slowly forward and works it into litter, turning leaves with the bill. On guard near the feeding site, it takes short rapid runs at small birds on the ground, supplanting them. Ground-cuckoos are quick and agile in movement, or they walk and stalk, reaching neck forward and low, extending a leg to the front of the head, then move the body forward while it holds the head in place over the ground, in slow, stealthy slides, like a rail, toes splayed apart on the ground (RBP). At other times, several ground-cuckoos share a feeding site (Ben King). They live in lowlands to 1000 m, exceptionally as high as 1500 m (Robson 2000a). Uncommon in Thailand and Vietnam, where they are trapped for food and the bird trade (Round 1988, Eames 1996).

Food Insects, small reptiles, mammals and birds (Delacour and Jabouille 1931). Captive birds require high protein diets. The birds survive and breed successfully on a diet of pigeon grain, mice, crickets, mealworms and earthworms (Delacour 1927, Ezra 1927,Wylie and Shelton 1982, Robiller et al. 1992, Rinke 1999).

268 Crested Coua Coua cristata

Breeding May to August (Robson 2000a). In the field, the nest has not been described. In captivity, it is an open platform or cup, built of twigs and branches and lined with leaves. It measures 36 cm wide and 8 cm deep, and is built on the ground or in a tree, 1–4 m above the ground. Both sexes build the nest, incubate and feed the young. Eggs are white, 44 ⫻ 34 mm (Walters 1996), the clutch is 2–6, the eggs are laid 2–3 days apart. The eggs hatch in 18–19 days. By day 10 the young have open eyes and are partly feathered, they fledge in 18–24 days and feed themselves from 28 days onwards. Growth continues

until day 40, and the young are independent at 50–60 days, when they are driven off by the male as the parents nest again. Plumage of the young bird at 45 days is mixed juvenile and adult, plumage at 90 days is like the adult. Birds are sexually mature in a year and have lived for 10 and 16 years in captivity (Ezra 1927,Atkinson 1982,Wylie and Shelton 1982, Robiller et al. 1992, Hughes 1997c). In defending the nest, the parent spreads its wings out and over the back, the large alulae spread and extending the apparent size of the bird, presenting a large front as it faces danger (photograph in Rinke 1999).

Genus Coua Schinz, 1821 Coua Schinz, 1821, Das Thierreich eingetheilt nach dem Bau der Thiere . . . , 1, p. 661. Large cuckoos of Madagascar with long tails, long legs and often a crest. Type, by monotypy, Cuculus madagascariensis Gmelin ⫽ Cuculus gigas Boddaert (Peters 1940). The German term for couas,“Seidenkuckucke” (⫽ silk cuckoos, Appert 1980), describes the soft silky sheen of the plumage, the pastel purples, pinks and

buffs. Couas have long eye-lashes. The nostril is an open slit, lower near the base of the bill than near the tip. “Coua” is a native Malagasy name for these birds.Appert (1970) recognized two species groups, the tree couas (C. cristata, C. verreauxi and C. caerulea) and the running couas (the other species), while Milne-Edwards and Grandidier (1879) noted that C. reynaudii is also a climber.Ten species.

Crested Coua Coua cristata (Linnaeus, 1766) Cuculus cristatus Linnaeus, 1766, Systema Naturae, ed. 12, 1, p. 171. (Madagascar) Polytypic. Four subspecies. Coua cristata cristata (Linné, 1766); Coua cristata pyropyga (Grandidier, 1867); Coua cristata dumonti Delacour, 1931; Coua cristata maxima Milon, 1950.

Description ADULT: Sexes alike, head gray, crest gray, back green gray, wing slightly darker, upper tail coverts washed purple, tail long, purplish blue with feathers tipped white on all but the central tail feathers; underparts, throat gray, breast pinkish gray grading into pinkish orange on lower breast, belly white, under tail coverts ochre-buff; bare skin around eye blue, anterior to eye cobalt blue, bright purplish blue or purplish brown, behind the eye mixed

sky blue or greenish blue and brilliant green, bare skin narrowly outlined in black with feathers, eyering blue, iris brown, bill black, legs and feet black. JUVENILE: Plumage loose-webbed, head gray, back and wing coverts green grey edged rufous, breast unbarred; skin around eye mostly feathered not bare, bill pale gray with a pink base and lower mandible.The palate rosette is still apparent at independence and fades by the time of postjuvenile molt.Young at four weeks are similar in appearance to their parents, but paler and the bare skin around the eye is dull blue, not bright as in the adult (Marcordes and Rinke 2000). NESTLING: Hatchlings are naked, without a trace of down, skin purplish black, darker above; bill reddish.

Crested Coua Coua cristata 269 Palate bright pinkish red, on each side the palate fold supports a raised rosette of white, forming a raised white ring around a raised white spot, contrasting in color with the bright palate.Tongue black on dorsal surface, supporting a raised outer shield outline with the posterior corners white and the shield grading to light blue around the middle of the tongue, and a central ring linked to the outer shield by a lateral spoke on each side.The nestlings are well feathered on the back by 6–8 days (Bluntschli 1938, Appert 1980, Marcordes and Rinke 2000). SOURCES: AMNH, BMNH, FMNH, MNHN, MSNG, MVZ, ROM, SMF, UMMZ.

Subspecies Coua cristata cristata (Linné, 1766); as above, crest short; Madagascar E, N and W to Mahajanga; Coua cristata dumonti (Delacour, 1931); larger, crest longer, plumage paler, under tail coverts pale rufous, white tail tips broader than in C. c. cristata; Madagascar W from Mahajanga to Morondava; Coua cristata pyropyga (Grandidier, 1867); larger, paler, under tail coverts bright rufous, white tail tip broad as in dumonti, crest intermediate in length; SW Madagascar between Morondava and Toliara, also south to Amboasary; Coua cristata maxima (Milon, 1950); larger, darker, under tail coverts lack rufous, back and inner primaries washed blue not green, known from only one specimen, perhaps a hybrid (Goodman et al. 1997); SE Madagascar (Tolagnaro).

Measurements and weights Coua c. cristata: Wing, M (n ⫽ 6) 134–148 (139.2 ⫾ 6.1), F (n ⫽ 6) 136–152 (145.9 ⫾ 7.4); tail, M 194–208 (198.4 ⫾ 5.6), F 197–202 (198.8 ⫾ 2.3); bill, M 18.6–20.9 (19.8 ⫾ 0.9), F 17.1–20.9 (19.3 ⫾ 1.6); tarsus, M 37–42 (39.3 ⫾ 2.0), F 40–44 (41.3 ⫾ 1.5) (AMNH); C. c. dumonti: U (n ⫽ 10), Wing 132–148 (140.4), tail 181–209 (195.4); C. c. pyropyga: U (n ⫽ 10), wing 157–168 (161.2), tail 197–222 (206.8); C. c. maxima: U (n ⫽ 1), wing 175, tail 232.5 (Benson et al. 1976). Weight, C. c. pyrrhopyga, M (n ⫽ 4) 135–150 (143.7), F (n ⫽ 3) 131–150 (144.7) (FMNH); all

(n ⫽ 14) 131–168 (146.9) (Goodman and Benstead 2003). Wing formula, P7 ⫽ 6 ⫽ 5 ⫽ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 40–44 cm. Gray arboreal coua, white below, with a pale gray crest (small dark crest in Verreaux’s Coua) and a black outline around the blue facial skin (black outline lacking in Verreaux’s Coua C. verreauxi).

Voice Loud, clear “coy coy coy . . .” with notes distinct and decreasing in volume, a chicken-like “wukwuk-wuk” given in alarm, and coos and grunts. Birds respond to calls of their neighbors, when at dawn and dusk as many as 15 birds countercall (Rand 1936, Appert 1970, Langrand 1990, Morris and Hawkins 1998, Randrianary et al. 1997).

Range and status Madagascar. Resident.Widespread, with the widest geographic and habitat range of any coua, throughout Madagascar except in the central highlands. Common around Ampijoroa and in the south near Berenty, and remarkably common in secondary forest in certain sites (Lavauden 1937, Milon 1950, 1952, Milon et al. 1973, Morris and Hawkins 1998). In the remaining blocks of forest around Tolagnaro, where the sole specimen of C. c. maxima was taken, the bird has not been seen again and is thought to be extinct (Goodman and Wilmé 2003).

270 Verreaux’s Coua Coua verreauxi

Habitat and general habits Primary and secondary forest, savanna, brushland, littoral forest and calcareous hills of southwest coast, palms, and mangroves, mainly in deciduous forest, uncommon in evergreen forest in the east; sea level to 1000 m. Arboreal, they feed in canopy foliage of the forest, move from tree to tree with long glides; they also feed on the ground, and birds and in mixed-species foraging groups. Often seen in the morning hours sunning on exposed branches with their wings drooped to absorb solar radiation (Rand 1936, Langrand 1990, Urano et al. 1994, Morris and Hawkins 1998, Goodman and Benstead 2003).

Food Large insects (caterpillars, grasshoppers and crickets (Acrididae, Euschmidtidae, Gryllidae, Pyrgomorphidae), beetles, cicadas, Phasmatodea (Phyllididae)); snails, chameleons, geckos, berries, fruits, seeds, tree flower buds; takes gum from trees (Rand 1936, Benson et al. 1976, Charles-Dominique 1976, Goodman et al. 1997). In captivity, it breeds when maintained on a diet of dog pellets, crickets and grasshoppers, with cooked cow meat and heart, and takes fruit especially mangos (Marcodes and Rinke 2000).

Displays and breeding behavior Monogamous, couas occur in pairs and both parents feed the young (Appert 1970). In captiv-

ity, a female sometimes mates with two males in rapid succession, even when another female is present on her own nest (Marcodes and Rinke 2000).

Breeding Nestbuilding is seen in November and December, also in April and May; oviduct eggs are known in October and March, fledglings are seen from late October to early February, and in June a fledged young was fed by both parents (Rand 1936, Appert 1970, 1980, Benson et al. 1976, Goodman et al. 1997). Nest is a bulky shallow bowl of twigs and rootlets, built in a tree, 4–15 m above ground, mean 8 m. At some sites both sexes build the nest, one bringing twigs and the other remaining in the nest where it arranges the material into place; at other sites only one bird is said to build the nest. Eggs are dull white, 35 ⫻ 26.5 mm, clutch 2 (1). Fledglings flutter the wings and call with a hum or hiss, and the young are attended to and fed by two adults (Appert 1970, Masuda and Ramanampamonjy 1996). Incubation and nestling periods are unknown. Like other coua species, when the nestlings beg they display the palate with open mouth, vibrate the large alula and give hissing calls. The birds have bred successfully in captivity when reared on a diet of dog pellets, crickets, and grasshoppers (Marcodes and Rinke 2000).

Verreaux’s Coua Coua verreauxi Grandidier, 1867 Coua verreauxi, Grandidier, 1867, Revue et Magasin de Zoologie Pure et Appliquée, (2), 19, p. 86. (Cape Sainte-Marie, Madagascar) Other common names: Southern Crested Coua.

Description ADULT: Sexes alike, upperparts gray, crest gray with dark tip, crest held backward, upright or forward, wing slightly darker, gray with a greenish tinge, tail dark gray with tinges of blue and purple and with broad white terminal band on all but the central tail feathers, head green-gray; underparts whitish, gray on

throat and breast, and breast white, sometimes buff on sides, the tail from below has the outer rectrices T3–5 with broad white band (42 mm or longer),T2 with a short white band (24 mm); bare skin around eye bicolored, ultramarine blue around and in front of eye and bright sky blue behind the eye, no black outline around the bare skin, eye-ring blue, iris brown to red, bill small and black, legs and feet black. JUVENILE: Similar to adult but crest short, and tail feathers narrower and more pointed (not broadly rounded as in adult) and with shorter white band

Blue Coua Coua caerulea 271 (e.g., T4 is 32–36 mm in juvenile, vs 42–52 mm in adult); bill pale. NESTLING: Undescribed. SOURCES: AMNH, BMNH, MCZ.

Measurements Wing, M (n ⫽ 9) 129–137 (133.2 ⫾ 3.1), F (n ⫽ 6) 128–132 (130.3 ⫾ 1.6); tail, M 173–195 (187.9 ⫾ 3.7), F 182–198 (189.4 ⫾ 4.8); bill, M 15.2–18.4 (16.5 ⫾ 1.0), F 16.4–17.3 (16.9 ⫾ 0.4); tarsus, M 38.3–41.1 (38.3 ⫾ 1.6), F 31.7–42.7 (37.8 ⫾ 4.6) (AMNH, BMNH, MCZ). Wing formula, P8 ⫽ 7–6 ⬎ 9 ⬎ 5 ⬎ 4 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 10.

Field characters Overall length 34 cm. Small arboreal coua with a small dark spiked crest, a pale blue spot behind the eye, and no outline of black around the bare skin of the face.

Voice Loud descending series of squawking notes, often given at dusk, “crick-crick-crick-corick-corick”, higher pitched and more rasping than the corresponding “coy coy” call of Crested Coua, also a loud growl “quark quark” followed by soft “coo coo,” descending in pitch (Rand 1936, Langrand 1990, Morris and Hawkins 1998).

Toliara. Resident. The species range is complementary to that of Coua cristata; with C. verreauxi in drier habitat. Near-threatened (Collar et al. 1994).

Habitat and general habits Subarid thorn scrub (Euphorbia and Didierea brush) and subdesert on sandy or calcareous soil; sea level to 200 m (Rand 1936, Benson et al. 1976, Langrand 1990, Morris and Hawkins 1998). These couas are mainly arboreal.

Food Insects, gekkos and small chameleons, also fruit of Cassia, taken mainly in trees (Rand 1936, Benson et al. 1976, Goodman and Benstead 2003).

Range and status SW Madagascar between Onilahy and Menarandra rivers, east of Menarandra, in coral rag scrub near

Breeding Unknown.

Blue Coua Coua caerulea (Linnaeus, 1766) Cuculus caeruleus Linnaeus, 1766, Systema Naturae, ed. 12, 1, p. 171. (Madagascar) Monotypic.

around eye blue (sometimes bicolored with anterior to eye purplish blue, behind eye cobalt blue), the bare skin around eye narrowly outlined by black feathers, iris brown, bill black, feet black.

Description ADULT: Sexes alike, upperparts and underparts dark blue, not glossy, wing and tail dark blue with violet sheen, tail with no white tip; bare skin

JUVENILE: Back and lower belly sooty black, wing dull blue, tail lacks violet; skin around eye feathered not bare, bill and feet black.

272 Blue Coua Coua caerulea NESTLING: Undescribed. SOURCES: AMNH, BMNH, CUM, FMNH, ROM, SMF, UMMZ, USNM.

Measurements and weights Wing, M (n ⫽ 6) 182–206 (195.2 ⫾ 8.0), F (n ⫽ 6) 194–213 (199.5 ⫾ 7.3); tail, M 238–276 (253.7 ⫾ 13.9), F 248–273 (258.5 ⫾ 9.5); bill, M 21.6–23.1 (22.4 ⫾ 0.6), F 19.4–25.0 (22.6 ⫾ 1.9); tarsus, M 52–57 (55.0 ⫾ 1.8), F 51–59 (54.8 ⫾ 3.1) (AMNH). Weight, M (n ⫽ 3) 225–257 (235.5), F (n ⫽ 2) 240–268 (254) (FMNH, UMMZ). Wing formula, P6 ⫽ 5 ⬎ 4 ⬎ 7 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 48 cm. An arboreal all-blue coua, with blue-violet wings and tail.

Voice Brief trilled “brrreee-ee” increasing in volume, also loud series of “coy coy coy coy” decreasing in intensity, with low pitch; also grunt “kroo kroo” (Rand 1936, Langrand 1990, Randrianary et al. 1997).

Range and status Madagascar, in most of the east and northwest. Resident, with local seasonal movements between littoral forest and nearby humid forest on laterite soils (Goodman et al. 1997). Common in suitable habitat through their range, mainly in eastern evergreen forest, seen in PN de Ranomafana, also locally in the north and northwest including Île Sainte-Marie. They are hunted and their forest habitat has become degraded (Benson et al. 1976, Goodman 1993, Morris and Hawkins 1998). Ranges of the arboreal C. caerulea and Crested Coua C. cristata are largely exclusive (Goodman et al. 1997).

Habitat and general habits Primary forest, second growth and dense clove plantations, local in deciduous forest, littoral forest, mangroves, from treetops to undergrowth, mainly in mid-strata.They occur from sea level to 1800 m; most are between 440 to 1200 m. In the PN d’Andohahela, they occur from 440 to 1875 m and are most numerous around 810 m (Hawkins and Goodman 1999); at PN de Marojejy from 450 to

1625 m, rarely higher (Goodman et al. 2000). An arboreal coua, seen when moving from tree to tree or crossing an opening, flying from higher perch and gliding lower to the next site. Slow and deliberate in movement in the trees and clumsy in flight (Rand 1936). Active from forest canopy to the undergrowth, and they run on the ground. Blue Couas, usually single birds, follow foraging troops of lemurs Eulemur fulvus (Goodman et al. 1997).

Food Insects (cicadas, locusts, crickets, katydids, bees, beetles, caterpillars, large homoptera), small reptiles (chameleons), frogs, crabs, fruits, flower buds of tree Symphonia, takes gum resins from Sloanea rhodantha trees in humid forests, the gums being rich in polysaccharides (Rand 1936, Charles-Dominique 1976, Goodman et al. 1997).

Displays and breeding behavior In courtship feeding, a bird held a frog, then opened its wings and fanned its tail, and gave a bow to the other bird (Goodman et al. 1997).

Breeding Mainly in the rains from July to December (Rand 1936, Langrand 1990). Nest is either covered above (Langrand 1990) or an open bowl-shaped clump, built of interlaced dry plant material, in dense foliage, 3.5–10 m above ground. Eggs are white, 37 ⫻ 28.5 mm, clutch 1 (Schönwetter 1964, Hawkins et al. 1998, Morris and Hawkins 1998). Incubation and nestling periods are unknown.The adult feeds insects to the young.

Red-capped Coua Coua ruficeps 273

Red-capped Coua Coua ruficeps G. R. Gray, 1846 Coua ruficeps G. R. Gray, 1846, The Genera of Birds, 2, p. 454, col. pl. cxv. (Madagascar and the eastern side of Africa ⫽ Madagascar) Polytypic. Two subspecies. Coua ruficeps ruficeps (G. R. Gray, 1846); Coua ruficeps olivaceiceps (Sharpe, 1873).

Description ADULT: Sexes alike, crown rufous or light brown green, upperparts and wing uniform light brown with green wash, upper tail coverts rufous brown, tail dark with purple cast, tail feathers broad, T1 and T2 uniform brown, T3 to T5 black tipped white; underparts, throat buffy white, breast light purplish, belly buff, center of belly whitish, under tail coverts rufous-buff; bare skin around eye indigo blue outlined in black feathers, the outline thicker behind and below the eye, the black feathers forming a band around the nape, iris dark brown, bill black, feet black. JUVENILE: Plumage dull, crown rufous buff with black feather edges, back greenish brown barred with blackish subterminal band and light brown feather edges, wing barred with gray to buff tips on coverts, primaries and secondaries black with buff terminal band, tail feathers narrow and pointed,T3 to T5 tipped whitish, breast rufous with dark gray bars; bill pale, color changes from reddish to black. NESTLING: Hatchlings naked, without a trace of down, skin purplish black, bill reddish, eyes closed. Palate bright pinkish red, on each side the palate fold supports a raised rosette of white, forming a raised white ring around a white central raised spot, contrasting in color with the bright palate.Tongue black on dorsal surface, supporting a raised shield outline, a ring around a central boss, linked to the white edge of the shield by four radial spokes. In older nestlings the white shield regresses and the red spaces enlarge, losing the anterior spoke (Berger and Lunk 1954, Appert 1967, 1970, 1980). SOURCES: AMNH, BMNH, FMNH, MVZ, ROM, SMF, UMMZ.

Subspecies Coua ruficeps ruficeps (G. R. Gray, 1846); adult, crown bright rufous; Madagascar,W and S from Mahajanga to near Morondava; Coua ruficeps olivaceiceps (Sharpe, 1873); adult, crown light brown green, upperparts and underparts paler than in C. r. ruficeps; juvenile, crown blackish gray slightly barred buff at tips of feathers, back barred, breast barred gray on white, belly white, under tail coverts gray, wing coverts with subterminal black band and buff tips, bill pale pink at base; south from Morondava to southern Madagascar.

Measurements and weights C. r. olivaceiceps: Wing, M (n ⫽ 6) 164–172 (167.5 ⫾ 2.8), F (n ⫽ 7) 161–173 (167.9 ⫾ 4.9); tail, M 214–240 (225.7 ⫾ 9.5), F 222–242 (234.2 ⫾ 6.7); bill, M 21.0–22.5 (22.0 ⫾ 0.7), F 20.4–23.8 (22.5 ⫾ 1.2); tarsus, M 54–59 (55.8 ⫾ 1.9), F 53–58 (55.4 ⫾ 1.6) (AMNH); C. r. ruficeps: Wing, M (n ⫽ 6) 161–168 (166.0 ⫾ 2.9), F (n ⫽ 3) 154–167 (160); tail, M 233–252 (239.8 ⫾ 12.4), F 231–243 (236); bill, M 22.7–24.6 (23.4 ⫾ 1.0), F 21.7–22.8 (22.4); tarsus, M 51–55.9 (53.3 ⫾ 2.1), F 48.5–53.5 (50.8) (AMNH, BMNH). Weight, C. r. ruficeps M (n ⫽ 1) 202 (BMNH); C. r. olivaceiceps U (n ⫽ 1) 182 (Goodman and Benstead 2003). Wing formula, P5 ⫽ 4 ⫽ 3 ⬎ 6 ⬎ 2 ⬎ 1 ⬎ 7 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 42 cm. Terrestrial coua with long legs, long neck and small head, slender bill, long tail and slim body which give the bird a characteristic shape, and the thick black outline behind the bare blue face is distinctive.

Voice A loud call “coy coy coy coy” rising up the scale unlike that of other couas, a loud “hug yew yew yew kuh kuh” with the last two notes lower; also grunts (Appert 1970, Langrand 1990, Morris and Hawkins 1998).

274 Red-fronted Coua Coua reynaudii forest areas (Urano et al. 1994).They live alone or in pairs (Delacour 1932,Appert 1967, 1970, Morris and Hawkins 1998). Size of territory or home range of nesting pair, 1–4 ha (Masuda and Ramanampamonjy 1996). They sunbathe on the forest floor, drooping their wings to absorb the heat, and they visit forest camp sites and feed on discorded rice (Goodman and Benstead 2003).

Food Insects (orthoptera, beetles); fruits, seeds, and rice (Milne-Edwards and Grandidier 1879, Rand 1936).

Range and status Madagascar in the west, southwest and south to Ankaoabo,Toliara,Ampotaka and Berenty. Resident. This is the only common terrestrial coua in the seasonally dry woodlands.

Habitat and general habits Dry deciduous forest, thorn scrub, spiny desert, second growth, degraded open woodland, gallery forest on edge of rivers and in forested bottomlands, in coral rag scrub near Toliara and arid habitat near Berenty; sea level to 900 m. Terrestrial, walk on the ground, usually run to escape danger, fly short distances, and perch in trees to call and to sun in the early morning. They also walk along branches. On the ground they usually walk, but when disturbed they move quickly in a jumping run (Berger 1960). They feed mainly on the ground in open

Breeding Season: October to January, females have enlarged ovary in February and April (Rand 1936, Appert 1970, 1980, Goodman et al. 1997; SMF). The nest is a shallow bowl of thin branches, bark and creepers, 15 ⫻ 25 cm and 5–12 cm thick, built in a tree 2–6 m above ground, with 19 of 37 nest sites found in the canopy and most others in the first fork of a tree or in a bush. Eggs are white, with a slightly chalky outer layer, the inner layer tinged dull blue, 34.4 ⫻ 28 mm, clutch most often 2 (1–3). Incubation period is unknown, and the nestling period is 12 days. Day-old nestlings beg in upright posture, neck stretched upward, mouth held open exposing the patterned palate, wings held straight out at the side (Appert 1970). Most nests are lost due to predation (Milon 1952, Appert 1967, 1970, 1980, Langrand 1990, Masuda and Ramanampamonjy 1996).

Red-fronted Coua Coua reynaudii Pucheran, 1845 Coua reynaudii Pucheran, 1845, Revue et Magasin de Zoologie Pure et Appliquee, 1845, p. 51. (Madagascar) Other common names: Reynaud’s Coua. Monotypic.

Description ADULT: Sexes alike, crown bright rufous, back and wing coverts dark olive green, wing green with blue wash, tail long,T1 with green gloss and T2 with blue gloss and dark tip, face blackish; underparts dark gray, darker on lower belly and under tail coverts; bare skin

around eye bicolored, ultramarine blue around and in front of eye and sky blue behind the eye, eye-ring blue, iris brown, bill black, legs and feet dark gray. JUVENILE: Crown slightly dull rufous brown, feathers loose-webbed with gray base, nape dull brown, back dull olive brown, rump rufous with the fluffy base of the feathers black, wing coverts and flight feathers tipped rufous with black subterminal band, tail dull green-brown T1, outer T2 to T5 glossed blue; underparts, throat gray, breast rufous-buff,

Red-fronted Coua Coua reynaudii 275 lower breast and belly slate brown; facial skin dull, bill yellow. NESTLING: Undescribed. SOURCES: AMNH, FMNH, ROM, UMMZ.

Measurements and weights Wing, M (n ⫽ 7) 129–140 (136.9 ⫾ 6.0), F (n ⫽ 7) 131–142 (139.1 ⫾ 3.8); tail, M 200–223 (210.1 ⫾ 6.8), F 198–222 (207.1 ⫾ 6.6); bill, M 18.8–21.0 (19.5 ⫾ 0.8), F 17.3–21.4 (19.9 ⫾ 1.4); tarsus, M 41–47 (44.0 ⫾ 2.4), F 42–46 (44.6 ⫾ 1.5) (AMNH). Weight, M (n ⫽ 2) 128–151 (139.5), F (n ⫽ 1) 163 (FMNH, UMMZ); all (n ⫽ 5) 128–175 (153.2) (Goodman and Benstead 2003). Wing formula, P6 ⫽ 5 ⫽ 4 ⬎ 7 ⬎ 3 ⬎ 2 ⬎ 1 ⬎ 8 ⬎ 9 ⬎ 10.

Field characters Overall length 38 cm. A terrestrial coua with dark plumage, no reddish color below, a rufous crown and a pale blue skin behind the eye.

Mainly terrestrial, they move slowly, walk on the forest floor and on low herbs, in forest clearings and on trails near vine tangles, move and feed on low branches and creepers, walk up sloping trunks and perch in trees (Rand 1936, Goodman et al. 1997, 2000, Hawkins et al. 1998, Morris and Hawkins 1998, Hawkins and Goodman 1999).

Food

Brief, raucous, plaintive “koo-ah” repeated several times, a sharp “jick”, and a long guttural rattle or chatter that increases in amplitude (Rand 1936, Langrand 1990, Morris and Hawkins 1998).

Insects, including beetles (Cerambycidae, Curculionidae, Elateridae, Scarabaeidae), grasshoppers (Euschmidtiidae, Tetrigidae), locusts, cicadas, stick insects (Phasmatodea), caterpillars; centipedes and spiders (Areneae) (Rand 1936, Goodman et al. 1997, Hawkins et al. 1998), also fruits and seeds (vs Coua serriana, which are