‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects
Tim R. New
‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects
Tim R. New Department of Zoology La Trobe University Melbourne, Victoria Australia
[email protected]
ISBN 978-94-007-1779-4 e-ISBN 978-94-007-1780-0 DOI 10.1007/978-94-007-1780-0 Springer Dordrecht Heidelberg London New York Library of Congress Control Number: 2011934968 © Springer Science+Business Media B.V. 2011 No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
Preface
H.M.S. Endeavour, captained by James Cook, visited the east coast of Australia from April to August 1770. Amongst the far-reaching accomplishments from that visit, Joseph Banks and Daniel Solander initiated study of the animals and plants of this island continent. Banks, and his assistants, collected the first suite of insects to be taken back to Europe for study – from several localities from Botany Bay northward. Some insects clearly impressed Banks – he reported the towering mounds of termites he saw at Cooktown as reminding him of ‘the Druidic monuments I have seen in England’. Almost 240 new species from this collection were described formally in 1775 by J.C. Fabricius, a leading disciple of Linnaeus and one of the most influential entomologists of the era. His later ‘Philosophia Entomologica’, published in 1778, is regarded by many people as the first real entomology textbook and Fabricius was perhaps the first entomologist to appreciate the massive variety of insects. In the ‘Philosophia’ he noted that the number of species ‘is almost infinite’ and that ‘if they are not brought in order, entomology will always be in chaos’. Before the Endeavour voyage, Fabricius had met Banks and Solander in London in 1767–1768 and, no doubt, urged them to bring back insects from the voyage. Several of those first-collected Australian insects are common species that are easily recognisable today – the common brown butterfly (Heteronympha merope), the yellow-winged grasshopper (Gastrimargus musicus), and the green mantis (Orthodera ministralis) are examples. Another was a widely-distributed bull ant, Myrmecia gulosa, collected first at Botany Bay, and it is tempting to speculate that Banks might even have been the first European to suffer the pain of their attack! We know that members of his party were stung by the green tree ant (Oecophylla smaragdina) further north – for, in the Endeavour Journal, Banks recorded ‘their stings were by some esteemed not much less painfull than those of a bee’. These insects and many others, named under Linnaeus’ then radical binomial system, were allocated to genera described from other parts of the world and with which European workers were familiar: most have since then proved to be far more distinctive and transferred to more recently-described genera, many of them restricted to Australia and, sometimes, also its neighbours. For example, the very first Australian insect (and, indeed, animal) to be named was a scarab beetle that Fabricius called v
vi
Preface
‘Scarabaeus barbarossa’, but is now referred to the genus Haploscapanes, which contains only a handful of named species, all from the Australian region. Numerous similar revisions have helped to emphasise the unusual nature of Australia’s insects – and, indeed, they are no less characteristic and endemic than the better-publicised mammals and birds. As further exploration occurred, the richness and complexity of insects became gradually clearer, with few species from Australia even remotely familiar to their describers in the northern hemisphere. It seems that no insects were collected on Cook’s second voyage, but the visit to Adventure Bay, Tasmania in January 1777 (on the third voyage, with H.M.S. Resolution) yielded more, with ten species described by Fabricius in 1787. Cook recorded that the insects seen there were ‘in considerable variety’. Insects are the most diverse of all animal groups, and characterising and understanding the Australian insect fauna is ‘a work in progress’. From the First Fleet onward, changes began to occur, with arrivals of alien animals and plants either accidentally or being introduced for settler commodities and agriculture. The insect fauna was no longer pristine, with progressive arrivals from overseas of insects (including fleas and lice as parasites of domestic stock and companion animals), some having substantial impacts on human welfare as consumers of crops and stored products. Those early arrivals were not documented, of course, but many later introductions have been – honeybees, for example, were brought to Australia in 1822 as amongst the first of suites of insects deliberately imported, for a variety of purposes but without consideration of any future impacts in the Australian environment. In parallel, however, visiting and resident naturalists had greater opportunity to collect and study Australia’s native fauna, and are still doing so. More than 200 years later, we still have only vague ideas about the diversity of many groups of our insects, with various ‘scientific guesstimates’ based on collection contents and expert opinions. Many species have not been studied or, even, collected and it is common for any visiting specialist working on a particular family of beetles, flies, wasps or other large group to discover large proportions of hitherto undescribed species to augment the total. Many surprises remain. Our largest known stick insects, giants of the order with one given the appropriate species name ‘gargantua’, have been described only in the last few years – the female of this particular giant, with body length exceeding 30 cm (and spanning more than twice this with legs extended) is known only from a small area of tropical rainforest in northern Queensland, and is one of the world’s largest insects. But at the other extreme, many minute insects are amongst the ‘black holes’ of our formal know ledge. Enormously diverse, of serious interest to only a few specialists (most of them based far from Australia), they remain undercollected and difficult to appraise. Some tiny wasps, that pass their early life within a single egg of a small barklouse or leafhopper, are only about one fifth of a millimetre long: considerably smaller than a large single-celled Amoeba but with all the structural complexity of much larger insects miniaturised into this speck of life. This variety of size was familiar to early entomologists, but is still surprising to many other people, and the practical difficulties of studying the richness of insect life renders estimates of their diversity somewhat intangible.
Preface
vii
We simply do not know how many native insect species occur in Australia. Recent suggestions of more than 200,000 different species assure them an easy top place for diversity amongst the entire fauna; and the figure is debated – it cannot be refuted, and may even be an underestimate. Methods of collecting and studying insects have advanced from Banks’ day, but the principles of needs for capture, preservation, curation, and expert examination and diagnosis are constant. Fabricius initiated the taxonomic foundation on which we must still build, with the realisation that even now perhaps only a quarter or fewer of our native insects have been given formal names. Early descriptions of species are brief, commonly only one or two lines of Latin and addressing a very limited range of characters. They contrast with the lengthy diagnoses now the norm for differentiating similar or allied forms. Fabricius’ contention (in Mantissa Insectorum 1787) that ‘Too many words are the real trouble of entomology’, was founded in an era when recognising entities regarded as species was altogether a simpler exercise than it is now. For example detailed measurements and good illustrations, involving morphological details often necessarily based in delicate dissections and microscopical examination, are now almost mandatory in describing insect species and differentiating related forms. Differences based on structural features are increasingly being augmented by molecular data and statistical analyses to clarify relationships. This book is about this ‘considerable variety’: what it comprises, how and where insects live, their peculiarities and roles in Australian environments, and their interactions with humanity. It is an introduction to the natural history of insects in Australia, and some of the remarkable features of the fauna that render insects the richest and most successful animals with which we, sometimes uneasily, share the planet. I hope to introduce the study of insects, entomology, through their evolution and adaptations to the variety of Australia’s terrestrial and freshwater environments they so capably dominate. This is not a formal textbook, but covers much of the ground that an elementary entomology text may include, in a framework intended to help people lacking formal biological training or knowledge of insects to begin to understand the major general features and causes of insect variety, and emphasising the importance of Australia’s insects, how they ‘work’, and the needs for conserving their diversity and sustaining their participation in ecological processes and systems. The sequence commences with several general introductory chapters on insect structure, evolution, biology and ecology, helping to illustrate the richness, variety and peculiarities of the Australian fauna. Later chapters summarise the main entomological features of some of Australia’s key environments, and the final chapters address aspects of interactions between people and insects and the importance of increasing efforts for documentation and conservation. I have tried to avoid much of the technical ‘jargon’ that readers can find so offputting and an impediment to understanding, and the sequence of general themes are each treated from basic principles; an Appendix summarises main features of the different insect orders. Each chapter contains suggestions for further reading but, except in a few cases in which I have referred directly to specific papers, close referencing is not given, as likely to disrupt the book. I hope that biologists who recognise allusions to their work without direct citation will forgive this approach. Many of the references cited
viii
Preface
are classics, and for many I have indicated their relevance: all are available readily. With similar intent, I have not attempted to provide a full illustrated synopsis of Australia’s insects; more comprehensively illustrated books are cited in Chap. 1, for example. My purpose has been, rather, to provide a limited range of illustrations of some representatives of major insect groups that help to ‘tell a story’, to consolidate points in the text, and that indicate particular features or habits that aid understanding of insect variety. Deliberately, many are of common or widespread species that can be discovered easily, some in home gardens, and so that can become familiar with relatively little effort. Much of the book’s content draws on basic information and principles, and so transfers easily to the insects of any other part of the world. The book is, I hope, based in good science and is intended to be accessible to non-entomologists as a means of introducing insects to a wide non-specialist rea dership and, in particular, of demonstrating the bases of the immense – indeed ‘considerable’ – ecological, functional and taxonomic variety that renders the Australian insect fauna so intriguing, and also so important to sustain. Two major strands of modern conservation are education (linked with informed understanding and advocacy) and scientific knowledge. Insects have long suffered from both image problems and that non-entomologists, including the great majority of ecologists and managers charged with conserving Australia’s unique biota and ecosystems, do not appreciate their taxonomic and biological subtleties and complexity that influence the scale of attention needed to sustain them. The book indicates some of the ambiguities and complexities of documenting insect diversity in Australia and discovering how insects have exploited this vast geographical arena – and so contributes to defining the steps needed to assure the wellbeing of this unique biological heritage. Acknowledgments The contents of this book are derived from many sources, only a few of which are cited specifically. New information on Australia’s insects is published in a range of relevant journals, such as Australian Journal of Entomology, Australian Entomologist, Australian Journal of Zoology, Austral Ecology, and Invertebrate Systematics, all of which focus on research in the region, and many syntheses of relevant topics can be found in the Annual Review of Entomology and elsewhere. Selection of examples to include or omit for limitations of space has been a complex and idiosyncratic exercise, and informed readers may justifiably consider some suboptimal and would opt for a different array from which to discuss general themes and principles. Photographs supplied by colleagues are acknowledged individually in the legends, and it is a pleasure to reiterate my thanks to these friends who responded so generously to my requests for this use. I appreciate comments from reviewers of the original proposal, and the continuing support of Zuzana Bernhart at Springer, together with the friendly cooperation and advice during production from Elisabete Machado. Production has been facilitated immensely by the careful help of Ms. Juno Martina George. La Trobe University Victoria 3086 Australia
T.R. New
Contents
1 The Basic Insect Pattern: Theme and Variations.................................. Introduction: Insects and Their Close Relatives........................................ The Insect Body Plan................................................................................. Inside Insects.............................................................................................. Further Reading.........................................................................................
1 1 3 7 12
2 Fossils and Major Insect Adaptations.................................................... Introduction: The Process of Insect Evolution........................................... Wings and Flight........................................................................................ Wings and Ecology.................................................................................... Insect Diversification................................................................................. Further Reading.........................................................................................
13 13 14 18 19 22
3 Insect Life Histories................................................................................. Introduction: Modes of Development........................................................ Diversification Within Metamorphosis...................................................... Seasonal Development............................................................................... Further Reading.........................................................................................
23 23 25 32 36
4 Origins, Distributions and Diversity...................................................... Introduction: Australia as an Environment for Insects.............................. Insect Species?........................................................................................... The Yellowish Skipper and Donnysa Skipper, Hesperilla flavescens and H. donnysa................................................... The Swordgrass Brown, Tisiphone abeona........................................... Further Reading.........................................................................................
37 37 43
5 Environments and Habitat for Insects in Australia.............................. Introduction: Places to Live....................................................................... Resources for Insects................................................................................. Further Reading.........................................................................................
55 55 61 66
44 44 53
ix
x
Contents
6 Foods and Feeding Biology...................................................................... Introduction: The Variety of Food and Feeding Habits............................. Exploitation or Partnerships?..................................................................... Searching for Food..................................................................................... Insect Herbivores....................................................................................... Insect Carnivores........................................................................................ Insect Parasitoids................................................................................... Insect Parasites...................................................................................... Insect Decomposers................................................................................... Further Reading.........................................................................................
69 69 70 71 75 79 85 87 88 91
7 Insect Behaviour and Lifestyles.............................................................. 93 Introduction: Behaviour and Adaptation................................................... 93 Sensory Mechanisms................................................................................. 95 Social Existence......................................................................................... 101 Further Reading......................................................................................... 111 8 Insect Communities................................................................................. Introduction: Living Together.................................................................... Richness and Variety.................................................................................. Evolutionary Radiations............................................................................. Assessing Richness.................................................................................... Further Reading.........................................................................................
113 113 116 119 125 127
9 Insect Populations.................................................................................... Introduction: Population Size and Structure.............................................. Population Fluctuations............................................................................. Further Reading.........................................................................................
129 129 131 137
10 Insects in Inland Water Environments.................................................. Introduction: Inland Aquatic Habitats for Insects...................................... Insect Variety............................................................................................. Further Reading.........................................................................................
139 139 141 147
11 Australia’s Alpine Insects........................................................................ Introduction: Environmental Extremes...................................................... Alpine Insects............................................................................................. Further Reading.........................................................................................
149 149 150 155
12 Lowland Insects and Their Environments: Non-forest Habitats.................................................................................. Introduction: Terrestrial Open Habitats..................................................... Grasslands.................................................................................................. Arid Environments..................................................................................... Mallee Environments................................................................................. Further Reading.........................................................................................
157 157 157 162 163 165
Contents
xi
13 Forest Insects............................................................................................ Introduction: Forest Habitats..................................................................... Forest Insects............................................................................................. Further Reading.........................................................................................
167 167 168 174
14 Insects and People in Australia............................................................... Introduction: Interest and Involvement...................................................... Pest Insects................................................................................................. Further Reading.........................................................................................
177 177 179 185
15 Australia’s Alien Insects.......................................................................... Introduction: Variety and Impacts.............................................................. Importation and Establishment.................................................................. Consequences............................................................................................. Australian Insects Abroad.......................................................................... Further Reading.........................................................................................
187 187 188 190 194 197
16 Conservation............................................................................................. Introduction: Rationale and Needs for Insect Conservation...................... Species Conservation................................................................................. Habitat Conservation................................................................................. Wider Views............................................................................................... Further Reading.........................................................................................
199 199 201 206 208 210
17 Diversity and Its Implications for Understanding Australia’s Insects.................................................................................... Introduction: Relevance of Basic Documentation..................................... Surveying Diversity................................................................................... Increasing Understanding.......................................................................... References.................................................................................................. Further Reading.........................................................................................
211 211 212 214 221 222
Appendix........................................................................................................... Australia’s Insects: The Players................................................................. Apterygota............................................................................................. Pterygota............................................................................................... Palaeoptera................................................................................................. Neoptera..................................................................................................... Polyneptera............................................................................................ Paraneoptera.......................................................................................... Oligoneoptera........................................................................................
223 223 225 226 226 229 229 235 239
Index.................................................................................................................. 249
Chapter 1
The Basic Insect Pattern: Theme and Variations
Introduction: Insects and Their Close Relatives Even defining ‘an insect’ can be difficult! But any understanding of their massive variety must start from a clear picture of the basic structural template that forms the foundation for any such definition and later diversification. That progressive differentiation has taken place through adaptive modification of almost any structure present, and defining that body plan is vital in distinguishing true insects from other animals. Examining some of the evolutionarily older kinds of insects helps us to characterise that pattern, as well as to suggest some of the reasons why insects as a group appear so successful and have persisted largely unchanged in their fundamental design for so long. Insects are arthropods, members of that vast phylum of invertebrate animals that share a hard external skeleton and have jointed limbs, ancestrally a pair for each body segment. Within the arthropods, they are accompanied by spiders, mites, crabs and other crustaceans, myriapods such as centipedes and millipedes, and a host of others, each of which has a reasonably consistent body plan that enables us to recognise them. So, also, with insects. Characteristically, insects have six legs and their body is divided into three major regions, the anterior head, central thorax and posterior abdomen. Sometimes these regions are clearly separated – as in a ‘waist’ (although, paradoxically, in wasps and their relatives, that waist is actually after the first bit of the abdomen!) between thorax and abdomen; hence the name ‘insect’ (cut into), in marked contrast to the body of many other arthropods. This much is straightforward, but the integrity of defining insects in this way is disrupted by the existence of several other groups of small arthropods that share this pattern and so join them in the Hexapoda, the six-legged arthropods. These, the springtails (Collembola), proturans (Protura), and diplurans (Diplura) have historically all been placed in the class Insecta, but each is now considered an entire independent class equivalent to the whole of the true insects. The reasoning for this is complicated, and rests on the form of the mouthparts. The three small groups are collectively called ‘Entognatha’ (or entognathous hexapods) to emphasise that their mouthparts
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_1, © Springer Science+Business Media B.V. 2011
1
2
1 The Basic Insect Pattern: Theme and Variations
Fig. 1.1 Broad groupings of insects and other hexapod classes. The three classes of Entognatha were all earlier considered to be insects, but all ‘true insects’ are ectognathous, with the winged forms (Pterygota) comprising several major lineages (see text)
are enclosed in extended folds from the front of the head, in contrast to the exposed mouthparts of the true insects, which are thereby ‘Ectognatha’. However, even Entognatha is probably not a single lineage, and Diplura are probably nearer to the basal insect line of evolution than the other two, again assessed on the articulation of the mouthparts. Even within the unambiguous Insecta, the most ancestral forms include one small order, bristletails (Archaeognatha), with jaw (mandible) articulation different from all others. These broad groupings are summarised in Fig. 1.1, and at least enable us to define ‘insects’ in discrete taxonomic terms that are universally accepted. However, simple observation of many insects reveals many departures from the above basic pattern used so far to define them. Many butterflies have only four legs; some adult insects have even lost all their legs (as in female Strepsiptera, living inside their hosts) and the traditional body divisions may be difficult to discern; when we include highly modified immature stages (larvae), of which more later, the appearance is often very different. A typical ‘maggot’, the larva of many true flies (members of one of the largest orders of insects, Diptera, see Chap. 3) is basically a tapered cylinder, without any head, the modified mouthparts enclosed in the anterior end, no obvious differentiation between any body regions, and no legs. Such radical departures emphasise the extent of modifications that insects may undergo to exploit different environments and ways of life, and also make it very difficult for field biologists to associate early stages of many insects with the corresponding adults. Nevertheless, the basic pattern forming the foundation for these is at least reasonably consistent, and the basis also for classification of insects into their major
The Insect Body Plan
3
groups, orders. Insect systematics and recognition is based largely on external structural features, with relationships inferred from patterns of change and transition that occur. Nowadays, this information can be augmented, and in some cases questioned, from results of molecular analyses, but each of the approximately 30 orders of insects alive today can be recognised, and diagnosed formally, on a particular combination of structural features common to all its members and differentiating the order from all others (Appendix, p. 223). Simplistically, recognition of a dragonfly, beetle, moth, grasshopper, or many others is generally straightforward, even though assessing the relationships between the orders may not always be so. Even experienced entomologists can be misled by the bizarre appearance of some forms. Just as the first specimens of the platypus sent back to England caused naturalists to speculate that they had been manufactured, or birds of paradise were presumed to lack feet, some Australian insects have at first seemed not to fit any conventional ideas. The initial formal description of the orthopteran known as the ‘Cooloola monster’ (Cooloola propator, from Queensland) was introduced by the following comment: ‘After some amusement at the technical excellence of the apparently manufactured monster, it was determined that it was a genuine complete cricket-like insect’. Occasional other oddities have proved difficult to allocate even to order, but discovery of entirely new orders is unlikely to occur very often, although specialists continue to debate whether some of the long-recognised orders should retain their current boundaries or divisions. The most recently erected insect order, the southern African Mantophasmatodea (heelwalkers, rock crawlers, with features of both praying mantids and stick insects), was named following discovery of living insects in 2002. However, since then it has been relegated to a suborder and combined with another small non-Australian group within the existing order Notoptera. But, in short, the basic body plan of insects is both definable and has become differentiated to produce the largest suites of animals ever to grace Earth.
The Insect Body Plan The structural plan of true insects is exemplified well by a rather ‘basic’ insect such as an adult of the field cricket (Teleogryllus commodus) or the Australian plague locust (Chortoicetes terminifera), both members of the order Orthoptera and notable for their intrusions into pasture and cropping systems. This ‘basic body’ (Fig. 1.2) comprises three very different-looking regions. The head is a solid capsule, with the antennae (chemosensory structures), three pairs of mouthparts, and large compound eyes the major features. Small ‘simple eyes’ (ocelli) are also present. The thorax is also a solid box, with three pairs of legs and in most crickets and other insects, two pairs of wings (p. 14). The abdomen is elongate, more delicate in appearance and with posterior elaboration used in mating by both sexes and egg-laying by females. All of the appendages noted are derived from the serial appendages of the basic arthropod, so also help to indicate the number of segments in the theoretically complete insect body. The pattern is clearest for the thorax, where the three segments
4
1 The Basic Insect Pattern: Theme and Variations
Fig. 1.2 Generalised adult body pattern of an insect, indicating the three main regions; head (with large compound eyes, antennae and mouthparts), thorax (dotted, with legs on all three segments and two pairs of wings), and abdomen: insets show some variations on the form of antennae (a–d), legs (e–i) and mouthparts (j–l) to indicate some of the extensive modifications that enable insects to adopt different ways of life. Main diagram shown with slender ‘filiform’ antennae: others are (a) clubbed, or ‘clavate’, as in butterflies; (b) flanged or ‘flabellate’, as in some beetles; (c) feathery, ‘plumose’ as in some moths; (d) elbowed, ‘geniculate’, as in ants. Legs: (e) fore leg of mantid, adapted for grasping prey (cf Fig. 6.3); (f) fore leg of mole cricket, broadened for digging; (g) parasitic louse, for firm gripping of host hair or feather; (h) enlarged hind leg of grasshopper, for jumping; (i) hind leg of water beetle, flattened and with fringe of long hairs, as paddle for swimming. Heads; main diagram with chewing mouthparts, as in a grasshopper: (j) fly, with mouthparts composed of labium, expanded as ‘sponge’ for semiliquid diets; (k) butterfly or moth, a long coiled proboscis adapted for sucking nectar from flowers; (l) plant bug, with long slender stylets for piercing vegetation and ingesting plant sap
each retain the pair of appendages, as legs and contributing also to construction of wings on the second and third segments. The six segments of the head are fused together, and appendages modified to form antennae (segment 2), and the mouthparts (segments 4–6), and the abdomen appendages are lost other than for those
The Insect Body Plan
5
constituting genitalic structures on segments 8 and 9 of the 11 total. Remnants of the limbs are present on the more anterior segments in some very primitive wingless insects, giving us clues to the derivation of the more posterior structures in crickets and others. The insect body as a whole thus comprises 20 segments arranged into three ‘blocks’ (tagmata) adapted primarily for rather different roles: the head for sensory perception of new environments as the insect encounters them, and for feeding, the thorax for locomotion, and the abdomen for reproduction. Because solid muscle attachments are needed for mouthparts, legs and wings to operate effectively, the head and thorax are indeed usually very solid structures. Conversely, the abdomen is flexible, with a lateral membrane rather than solid ‘wall’. It can be distended for storing food, food reserves, or eggs, and also allows versatility in mating – some insects adopt postures that could be envied by compilor(s) of the Kama Sutra, but which usually involve juxtaposition of the abdominal tip of the two sexes – and depositing eggs in many hidden habitats such as under bark, in the ground, or in other insects or even other animals. In some insects, the abdomen also has two posterior ‘cerci’, filamentous sensory organs, broadly analogous to ‘posterior antennae’ and which tend to be reduced or lost in the most advanced insects. Any or every appendage and other structure can be changed massively from this basic form, and it is useful to look at some of these divergences here, as a prelude to seeing how they operate in different kinds of insects with differing ways of life, and keeping in mind that the variations have both functional roles, and value in diagnosing and recognising different kinds of insects. Several representative variations of antennae and mouthpart form are shown in Fig. 1.2, each condition diagnostic for some insect group(s), and collectively helping to emphasise the strong linkages between structure and function, the last reflecting ‘way of life’. Again, the cricket provides basis for comparison. The cricket’s antennae are long (although not as long as in many other similar Orthoptera, in which they can reach several times the body length) and slender. They are technically ‘filiform’ or threadlike, and are made up of numerous small jointed lengths, commonly (but not embryologically) termed ‘segments’ but more properly ‘antennomeres.’ Many other insects have much shorter antennae, and the basic appearance can be much different, with branches, flanges (Fig. 1.2b), numerous lateral projections, or apical thickenings so that they can seem feathery (plumose: Fig. 1.2c) or clubbed (clavate: Fig. 1.2a). The effect of these ornamentations is to increase the surface area available for chemical receptors. In some moths, the form of the antennae differs markedly between the sexes within the same species: those of males are strongly feathery, and of corresponding females, slender. Such differences indicate rather dissimilar needs. In this case, for example, females of many moths, such as codling moth (Cydia pomonella) and Oriental fruit moth (Grapholita molesta), both pests of orchard crops, do not fly but attract mates by emitting a highly specific pheromone scent. The males detect the scent through their antennae and respond by flying upwind along the increasing concentration gradient to encounter a potential mate. This behaviour, by which males of some species may be attracted from up to several kilometres away, has been used in aspects of pest management for these species, by attracting males to artificial pheromones on crops and so keeping them
6
1 The Basic Insect Pattern: Theme and Variations
from mating, thereby reducing the next generation of the pest. Of less economic importance, collectors may use a female moth to attract males as specimens. Not all moths ‘work’ in this way, but this example illustrates well how the appearance and structure of an insect appendage allows us to interpret or infer some important aspect of its biology. Likewise for mouthparts. Diversification of feeding habits is a major component of insect evolution and their spread of ecological roles and interactions, and is reflected in modification of any or all of the mouthparts. In the cricket, again regarded as a ‘basic’ representation, these comprise a pair of strong jaws (mandibles, on segment 4) that dismember food – in this case, predominantly vegetation; behind these are the paired maxillae (on head segment 5), very different in appearance from the tough mandibles and including a sensory structure (maxillary palp) and ‘accessory jaws’; and the third pair on segment 6 are structurally similar maxillae with reduced palps but fused in the midline to constitute a single structure, the labium. This arrangement is found in many different insects with chewing habits, whether herbivores or carnivores, but this structure is clearly not well-adapted to ingest liquid diets, such as plant sap or blood. The functional need is then for some structure with a role equivalent to that of a drinking straw or hypodermic syringe that can probe or penetrate the plant or animal surface and imbibe the liquid. This is accomplished independently, and by rather different modifications in several widely disparate groups of insects. In some it involves transformation of the cricket-like mandibles and maxillae into slender piercing ‘stylets’, each forming part of the circumference of a tube (proboscis) through which liquid is passed. The whole of this delicate structure in sucking bugs (Hemiptera, Fig. 1.2l) and some Diptera (such as mosquitoes) is supported by a broadened protective labium. The functionally similar structure of Lepidoptera is formed from maxillae alone, and can be coiled under the head when not in use (Fig. 1.2k), so not impeding manouverability: because of their need to take nectar from flowers, some Lepidoptera have a proboscis several cm or more long – the record is perhaps of a hawk moth from Madagascar, in which this structure extends about 30 cm, enabling it to gain nectar from orchids with very deep flowers and act as a pollinator for these! Other mouthpart variations occur. In bushflies and other advanced Diptera, the major structure is from the labium alone, flattened and expanded at the apex which is adpressed to surfaces of dung, carrion, plants or other foodstuffs to sponge up semiliquid materials (Fig. 1.2j). As with many other structures, mouthpart form can be taxonomically diagnostic as well as functionally informative. Moving to the thorax appendages, any or all the parts of the typical insect leg (Fig. 1.2e–i) can also be changed – whereas we naturally think of walking or running as their primary function, other roles are common. In our cricket/locust examples, the hind legs are conspicuously elongated and strengthened for jumping; many aquatic insects have legs broadened and/or fringed with long hairs to increase their surface area for ‘rowing’ on or under water; mole crickets have unusually strong and broad front legs for digging into soil; and the spined and grasping front legs of mantids are used to capture other insects (and on occasion other animals, even small birds) as prey – in an adaptation paralleled by some predatory bugs, lacewings and
Inside Insects
7
flies in which the forelegs have assumed similar form. Insect courtship may involve elaborate displays including ‘leg-waving’, and legs are also involved in the sound production (stridulation) and reception in many Orthoptera. Wings are discussed more fully later, as a key feature of insects, but the basic principle of multiple modifications from a single basic form is common to all the structures we have noted, with numerous cases of parallels – as in the grasping forelegs of some predatory insects, above. Adoption of a similar habit or way of life by groups of insects that are only distantly related can commonly lead to a problem of capability being solved by the same basic adaptation evolving independently. But, however unusual or bizarre any insect may appear, it has a fundamental structure derived from a pattern similar to that of the cricket with which we started this section. The abdomen is more uniform, largely reflecting that the serial appendages so prone to modification are restricted to the posterior end. However the most ancestral groups of true insects, the silverfish (Zygentoma) and bristletails (Archaeognatha) show us how these reproductive structures may be derived from the same basic limb form. In these very primitive lineages the underside of some (even, most) of the abdominal segments have paired narrow ‘styles’ projecting rearwards from the posterior margin. These represent part of the base (coxa) of the leg, and have disappeared from the more anterior abdominal segments in advanced insects. The coxae of normal thoracic legs of some of these insects also bear a style, clearly indicating he homology described above.
Inside Insects The internal structure of insects, also, follows a rather basic and consistent pattern to accommodate the needs for digestion, respiration, reproduction, movement and the variety of other metabolic and developmental processes and responses to the local environment. This pattern, summarised in Fig. 1.3, shows the characteristic relative positions of the major anatomical systems. Thus (1) the alimentary system (dotted in Fig. 1.3a) is the most conspicuous as a continuous tubular ‘gut’ from the anterior mouth to the posterior anus, along the whole length of the body; (2) the circulatory system is predominantly dorsal to this, with a mid-dorsal vessel (sometimes called the aorta anteriorly and the ‘heart’, posteriorly, but a single tube) along the midline; (3) the central nervous system, in contrast, is ventral with anterior concentration of nervous tissue as a brain, encircling the gut in the head; (4) the respiratory system, as a series of tubular tracheae, extends throughout the body, with air admitted from the exterior through a series of paired lateral openings, spiracles; and (5) the reproductive system is predominantly posterior, with reproductive openings for mating and oviposition situated below the anus. All these systems are within a body cavity, the haemocoel, and can be displayed easily in dissection of a freshly killed cockroach or grasshopper. The body cavity contains haemolymph, in which all the above structures are bathed. Haemolymph is
8
1 The Basic Insect Pattern: Theme and Variations
Fig. 1.3 (a) General internal anatomy of an insect, to show organisation of various organ systems. (b) Digestive tract, indicating different regions
the blood of insects but, unlike vertebrate blood, has only a minor role in respiration and is involved more in nutrient and waste metabolite transport and in some immune interactions. Haemolymph can provide defence against disease, parasites, or physical injury, such as by forming clots that can seal wounds through the cuticle, or cellular reactions that encapsulate disease-causing organisms (including eggs of parasitoid wasps, p. 85) and isolate them. Waste products are filtered from the haemolymph by special structures, the Malpighian tubules, and discharged into the gut and, thence, to the exterior. The haemocoel also contains material loosely termed the ‘fat body’, a more-or-less evident layer of fat around the gut or lining the body cavity and which has complex metabolic roles in storage and reorganisation of nutrients and regulating their supply to the insect. Both the haemocoel and the gut of insects contain microorganisms of various kinds. Some are clearly needed by the insects, and are mutualistic. Wood-feeding termites depend on single-celled protistans to break down cellulose into digestible components, for example, but the true roles of many of these characteristic symbionts are still unclear. Each of the major anatomical systems can undergo modifications for particular ways of life, but they are used only to a very limited extent in classifying insects – not least because the hard external structures are much more accessible for study, and remain available and unchanged in long-dead specimens in collections.
Inside Insects
9
However, some understanding of the variety of internal structure helps in interpreting the variety of life styles that insects may exhibit. Thus, the alimentary canal is divisible into several distinct regions, differing in appearance and primary function The relative development of these may reflect the diet of the insects, predominantly whether it is solid or liquid, or plant or animal material, so that the gut may be generalised, or specialised to deal with particular food materials. Insects such as grasshoppers and caterpillars that ingest solid vegetable food tend to have simple, short, muscular guts, strong enough to resist abrasion from plant or animal fragments and wide enough for those particles to pass easily. In contrast, many liquid-feeders have long, more convoluted narrow intestines that allow greater surface contact with the liquid, and protection from abrasion is not an issue. Many sap-sucking bugs and other liquid-feeders have to take in large volumes of food, because the nutrients they need are very dilute in excess water; some bugs have a special ‘filter chamber’ to eliminate excess water and concentrate the food for digestion. Whatever the diet, the gut may also need to store food at times, because many insects can feed only intermittently. The basic pattern of the gut divisions is shown in Fig. 1.3b. Food enters through the mouth, near which the salivary glands open. The saliva is sometimes used to commence digestion of food outside the body, and may be injected by liquid-feeding insects for this purpose and to aid ingestion. The gut itself is conventionally divided into three main regions, the anterior foregut, central mid-gut and posterior hind gut. The fore gut comprises the oesophagus, through which food passes to the crop (in which food may be stored) and insects taking solid food have a proventriculus (or gizzard, often muscular and with internal hardened spines or ridges by which food particles are broken down). The mid-gut is the major region for digestion, with the surface area for the enzymes produced in this region to interact with food sometimes increased markedly by pouches or gastric caeca from the central ventriculus. In many insects it is lined by a peritrophic membrane separating the food from the gut wall and increasing circulation of enzymes and which is shed at intervals together with any post-digestion food residue. Thus, the diet of dragonfly larvae can best be studied by allowing wild-caught larvae to eject their faecal pellets surrounded by peritrophic membrane, and which encapsulate the remains of their arthropod prey from their recent meals. Dissection of these pellets provides many characteristic fragments (such as hard parts of arthropod mouthparts or legs, that cannot be digested) that can be identified, often quite precisely. The Malpighian tubules, carrying metabolic wastes as noted above, open to the intestine at the junction between mid-gut and hind gut, and their number and form can also characterise particular insect groups. Major functions of the hind gut are absorption of useful materials from faeces and urine before these are egested, and the three successive regions, commonly differentiated as ileum, colon and rectum, differ in relative extent across taxa. In some insects, such as larvae of antlions and other lacewings (Neuroptera), the hind gut is blocked, so no faecal material can be passed until the insect matures. A liquid-filled body cavity also provides an internal ‘hydrostatic skeleton’ that enables soft-bodied insects such as maggots and similar larvae to crawl, through
10
1 The Basic Insect Pattern: Theme and Variations
waves of muscular contraction being transmitted along the body. Insects with a hard exoskeleton move mainly through direct action of muscles attached to this. Insects have only striated muscles, and those associated with the wings and powering flight or other strenuous movement are particularly well developed. Circulation of haemolymph occurs through passing it along the dorsal vessel, with segmental openings with valves to help ensure a one-way flow toward the anterior. The ventral nerve cord consists of a series of segmental concentrations of nervous tissue (‘ganglia’) linked by paired longitudinal connective nerves. Their basic pattern is of a pair of ganglia for each body segment, but considerable modifications have occurred through these becoming fused or concentrated to varying extents. Most consistently, all the head ganglia are joined to form the brain (dorsal) and suboesophageal ganglion (ventral) around the fore gut, and the numerous other patterns range from all thoracic and abdominal ganglia being distinct to, at the other extreme, all forming a single mass within the thorax. However, nerves radiate from the ventral cord to all muscles and sensory structures to govern the insect’s responses and behaviour. In addition to the conspicuous sense organs, such as eyes and antennae noted earlier, the variety of less obvious structures reflect the needs to respond to both internal and external changes, and to a great variety of environmental cues. Numerous specialised hairs, bristles and related structures on the body surface are linked with individual nerve fibres and are highly adapted receptors for mechanical, positional, chemical, audial or temperature or humidity cues – so that sounds produced by other insects, (whether mates or antagonists), and chemicals such as pheromones, and changes in the external environment can all be detected effectively and appropriate responses be made. The repertoire of sensory structures and responses for any insect may include many individualistic components that facilitate precise responses. Insects obtain oxygen and eliminate waste carbon dioxide through a system of ramifying internal tubes, tracheae and smaller tracheoles, with external openings (spiracles) through which air is taken from or eliminated to the outside environment. One pair of spiracles opening laterally from each thoracic and abdominal segment is the primitive pattern, but Recent insects never have more than two thoracic and eight abdominal spiracles, and many have far fewer. Dragonfly larvae (‘mudeyes’) and some others have no spiracles at all – in larval dragonflies, gaseous exchange takes place across the wall of the rectum, into which water is pumped, and mayfly larvae obtain oxygen by diffusion across the lateral abdominal gills. Likewise larvae of some internal parasitoids (p. 86) also lack spiracles and have finely ramifying tracheoles over much of the body surface to enable gas exchange across the body wall. Many insects with spiracles can close them, reducing water loss in more arid environments. Reproductive structures also follow a rather basic pattern, as a foundation for innumerable variations in size, shape, complexity and development in both sexes, in relation to functions and reproductive behaviour. These functions are complex. A female insect of a bisexual species needs to mate, store and transmit sperm, produce varying numbers of eggs (from few to many, all together or over an extended period and perhaps store these over many weeks or months) as well as lay them, perhaps in precisely selected locales. Some taxa are viviparous, so that larvae are the
Inside Insects
11
first stage to be released to the outside world, and many insects are parthenogenetic. Males need to produce and store sperm and transmit it to the female. They may also have adaptations to enhance their chances of success in competition for mates: males of some butterflies (such as the big greasy, Cressida cressida, a swallowtail from Queensland) actually plaster the female reproductive opening with a secretion that hardens to form a ‘chastity belt’ (technically, a ‘sphragis’) to prevent subsequent matings with other males. And some male dragonflies use spines on their reproductive appendages to ‘rake out’ sperm from any previous matings before depositing their own. Much of the intricate behaviour of insects relates to increasing chances of reproductive success and larger numbers of progeny, and their often complex behavioural strategies link strongly with structural adaptations in the reproductive system. As in other animals, hormones in insects play pervasive roles in moulting, development and many aspects of reproduction and metabolic regulation. They are produced from various internal organs and transported by the haemolymph. Three major groups of hormones are usually distinguished as central to reproductive and growth functions, and these are termed the ecdyosteroids, the juvenile hormones and the neurohormones (or, more commonly, neuropeptides). The first are concerned with moulting as a critical process in insect growth and maturation; the second are involved with control of metamorphosis and reproductive development; and neuropeptides influence almost all other aspects of metabolism as well as reproduction and the regulation of juvenile hormone production. They are integral drivers of ‘how insects work’. Entire texts have been written on almost every aspect of insect structure and physiology, and numerous articles in scientific journals and reviews continually present new information and interpretation of their functions and evolution. Part of the story of insect variety is linking structure and function, and appraising how all aspects of insect morphology and metabolism enable the insect to cope with (and capitalise on) its environment, to fit it to develop, disperse, find and use the resources it needs throughout its life, and to regulate its behaviour and lifestyle to persist and cope with changes in that environment in both space and time. The idiosyncrasies of any insect species or group reflect these needs. With a few exceptions of apparent or relative environmental uniformity (such as the flour or grain storage environments of some stored products beetles) insects live in environments that are patchy and variable, and within these they may encounter a range of conditions of humidity and temperature, of food supply, and of other species that may facilitate or oppose their own wellbeing. But, even within a warehouse or sack of flour, conditions change – in aeration, nutritional quality and in the number of individual insects present as populations increase with little initial opposition, so that density of insects may lead to crowding and competition for food and space, and change the interactions between individuals (and species) as more frequent and less easily avoided encounters occur between them. Such situations can induce changes in hormone balance that, in turn, induce behavioural or other changes. Parallels are numerous in more open environments, but there may be a greater variety of ‘escapes’ possible. However, increased density is associated with, for example, changes in
12
1 The Basic Insect Pattern: Theme and Variations
some normally solitary grasshoppers to induce them to enter a gregarious phase, as ‘locusts’, in which changed behaviour reflects modifications in both hormonal balance and sensory responses. Sensory capability to select food, oviposition sites, mates and other necessities, and how and when to disperse are all critical components in an insect’s life. Not all such decisions may be positive, particularly in interactions with plants or other animals, from which a wide array of outcomes may be possible, as we see later. Interactions between individuals and different species are mediated largely by sensory mechanisms.
Further Reading The first three references below are to entomology texts of varying complexity, and provide more formal information on insect structure and biology. The next two are amongst several wellillustrated guides to insect recognition and biology in Australia. The last is a major, indispensable, source of information on published knowledge on Australian insects CSIRO (1991) The insects of Australia. Melbourne University Press, Melbourne (The most comprehensive, two-volume, text on Australian insects) Gullan PJ, Cranston PS (2010) The insects. An outline of entomology, 4th edn. Wiley/Blackwell, Oxford (Latest edition of a very successful general entomology text) (Note that various chapters of either of the above are valuable ‘further reading’ to most chapters of the current book) New TR (1996) Name that insect. A guide to the insects of southeastern Australia. Oxford University Press, Melbourne (An introduction to the regional fauna) Brunet B (2010) Australian insects: a natural history. French’s Forest, New Holland Zborowski P, Storey R (2010) A field guide to insects in Australia. French’s Forest, New Holland Daniels G (2004) Bibliography of Australian entomology, 1687–2000 (2 volumes). Privately published, Mt Ommaney
Chapter 2
Fossils and Major Insect Adaptations
Introduction: The Process of Insect Evolution This structural diversity of insects, and the biological variety it reflects, did not develop all at once. Insects as we would recognise them from modern forms have been around for at least 300 million years, as amongst the first major diversifications of arthropods on land. Over this vast period, we can detect several changes and transitions in structure that appear now to have been ‘pivotal events’ in leading to their success and fostering their recent abundance. However, the fossil record from which we infer those changes remains cryptic in places: assembling unambiguous evidence from ancient insect fossils is not always easy, and it is not surprising that uncertainties persist – or that the opinions of various specialists may differ widely over how particular fossils may be interpreted! In this chapter, some background to the information on insect evolution derived from the fossil record is outlined, together with its relevance to study of the insects around us today. The conventional belief is that insects evolved during the Devonian period, about 360 million years ago in the middle of the Palaeozoic era. Until recently, the few relevant fossils available from that early time, or even earlier, are Archaeognatha (from about 380 to 390 million years ago) and the non-insect Collembola. They have no trace of wings. The oldest of these early fossils, from Scotland, are almost 400 million years old and the collembolan Rhyniella praecursor was long believed to be the world’s oldest hexapod. The ‘true insects’ appeared first in the Devonian of North America. Very recently, however, a fossil from the Scotland deposits (Rhyniognatha hirsti) has been reappraised and is now considered a ‘real’ insect, leading to the implication that insects actually originated in the Silurian period and that wings may have evolved considerably earlier than is commonly supposed. The challenge to find supporting fossil evidence for this remains. Modern Archaeognatha are still wingless, and exemplify the insects termed ‘Apterygota’ (non wing-bearing) as the ancestral condition from which other insects arose. They and silverfish are the modern representatives of this truly ancient lineage.
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_2, © Springer Science+Business Media B.V. 2011
13
14
2 Fossils and Major Insect Adaptations
Wings and Flight The subsequent development of wings, conventionally presumed to have taken place during the Carboniferous period (extending between about 360 and 285 million years ago), is the most significant single event in insect evolution. Insects are the only winged invertebrates, and the ability to fly probably gave them enormous advantage in escaping from local threats (such as being eaten by ground-dwelling predators) and exploiting new habitats and resources as these arose. Wings remain the most obvious distinguishing feature of recent insects and all but a tiny fraction of modern insect species belong to the ‘winged insects’ (Pterygota: wing-bearing). The Carboniferous, the period during which most modern coal measures were laid down – largely from tree fern vegetation, is widely depicted as moist, with predominantly swampy environments in which these plants flourished. It has been suggested to mark the period when insects moved from the ground to exploit this taller vegetation, and from which the complex suites of interactions between insects and plants have developed. Tall vegetation was present some 30 million years before the Carboniferous, so that the earlier development of wings suggested above might have had similar evolutionary stimulation. However, simply climbing up vegetation poses hazards – falling off, or being dislodged by rough weather, for example, may expose insects to ground-dwelling predators, particularly should they ‘crash land’. One possible way to avoid this, and survive, is through wings. Hypothetically – for there were no human observers to record the process – development of lateral processes on the body might facilitate posture control in our falling insect: if it lands the right way up in a hostile environment, it might be able to run away and hide quickly. If not, it may perish. Should those processes become larger during evolution they might form gliding planes (whereby a dislodged insect might reach another tree fern and avoid the ground altogether, using principles similar to the ‘webbing’ of gliding possums, for example), and from there it is a relatively small (but significant) step to progressive enlargement and eventual hinging of these structures at the base to constitute wings with both directional and postural controls – as true locomotor structures affording aerial capability. The Carboniferous also marks the era of gigantism in the early winged insects. Some fossil dragonflies from that time are the largest insects that have ever existed. They span around 60 cm from wingtip to wingtip, and some mayflies were also much larger than any modern species. It is tempting to suggest that – because there were no other aerial organisms at that time – the exuberance of these forms was facilitated by the innovation of flight occurring in an environment in which there was little risk or competition. Gigantism may also have been facilitated by the high atmospheric oxygen levels during the later Carboniferous and much of the following Permian era, during which oxygen levels reached almost 30% – far higher than they have ever been since then. However attractively simple the above hypothesis on wing origins may be, as based on interpretation of wings and their possible precursors from fossils, two rather different origins for insect wings have been discussed extensively – and each
Wings and Flight
15
has its proponents. First, that wings developed as above from the two more posterior thoracic segments as dorsolateral ‘paranotal lobes’, initially as small flanges and then becoming more elaborate as they are enlarged. Second, that structures equivalent to the lateral respiratory gills still found in larvae of some primitive winged insects (the mayflies) may have led to wings – this assumes that the insect may be ‘lifted’ by winds when left exposed and moved passively, with subsequent elaboration of the flaps leading to control. The ‘gill theory’ is now accepted more widely, with the ancestral structure being one of the two branches of the so-called ‘biramous’ arthropod legs still widespread amongst modern crustaceans. With either scenario, however, we may be endorsing an entirely false presumption, that the role of any lateral process was indeed evolved as a wing precursor: this is not certain or perhaps even likely. Simply because wings are used so widely for flight, the need for flight was not necessarily the impetus for initiation of adaptations leading to wing development. Other possible roles of lateral processes, besides being gills, could include camouflage (through colour, ornamentation, or decreasing shadows), defence (by making the insect effectively larger as a refuge from predators, appear more fearsome, or simply tougher), providing extended bases for muscle attachment for locomotion (perhaps aiding sudden movement), or reproduction (elaboration for attracting mates, and perhaps found in one sex only). Each of these has parallels in the forms of modern insects and, as examples, indicate the considerable variety of adaptive advantages any such structure might convey. Once present, however, the structures could form a basis for wings. Some early fossil insects (in the extinct order Palaeodictyoptera) have lateral lobes on all three thoracic segments, but only two pairs of wings have ever developed. These are always on the second segment (mesothorax: fore wings) and third segment (metathorax: hind wings), and the two segments together are often termed the pterothorax (wing-bearing thorax) which, because of need for large flight muscles, can become large. Wings have fundamental importance in tracing evolution of insects and inferring some relationships between orders, because it is almost certain that they have evolved only once – so that all winged insects (the vast majority of those alive today) are derived from a common ancestor, and the enormous variety of wing forms all relate to the same basic pattern. This pattern is consistency of longitudinal struts (veins) radiating along the wing from the base, with the pattern (wing venation) of immense value in classifying insects. The veins form a consistent sequence from the anterior margin of the wing towards the posterior; each vein has a name, with equivalent notational shorthand based on initial letters, as shown in Fig. 2.1. Each, with the pattern consistent in the two pairs of wings, may change in track, branching pattern, and extent of development so that different insects can differ markedly in the numbers and intensity of venation, and differences between fore wing and hind wing. The longitudinal veins are linked transversely by variable numbers of cross-veins, very numerous in insects such as dragonflies, grasshoppers and crickets, many fewer in some advanced insects, but all helping to delimit closed areas (cells) on the wing, also of taxonomic significance. As one example, most members of two major families of parasitoid wasps (p. 86), can be differentiated on the absence (Braconidae)
16
2 Fossils and Major Insect Adaptations
Fig. 2.1 Wing pattern and venation. The general pattern and nomenclature of the main longitudinal wing ‘veins’ of an insect, as a series of struts from the base to apex of the wing, and standardised from anterior to posterior; the veins may branch as shown, and are named from the initial letters of each, with branches of the same vein numbered sequentially from the anterior. Any of these veins may be modified or lost and linked by crossveins to produce characteristic patterns of wing venation important in diagnosis and classification of insects
Fig. 2.2 Example of use of wing venation in classification: the two large related families of parasitoid wasps, Ichneumonidae and Braconidae. Most members of these families are separated by the presence (a, Ichneumonidae) or absence (b, Braconidae) of a single fore wing crossvein between veins R and M, furnishing a useful ‘spot character’ for recognition; note the highly reduced venation from the general pattern shown in Fig. 2.1
or presence (Ichneumonidae) of one particular fore wing crossvein (Fig. 2.2). A typical ichneumonid is shown in Fig. 2.3. Even small differences in wing venation may be characteristic diagnostic features for groups or species. The qualifying ‘almost’ in ‘almost certain’ above refers to ambiguity over the independent evolution of the mayflies and dragonflies, as the two most ancient insect orders with functional wings. Many authorities accept that there are in fact two insect lineages with wings, but there is still some uncertainty over whether
Wings and Flight
17
Fig. 2.3 A living ichneumonid wasp to indicate venation outlined in Fig. 2.2a
these are ‘mayflies and (dragonflies and all the rest)’ or ‘dragonflies and (mayflies and all the rest)’. Mayflies (Ephemeroptera) are the only insects in which there are two winged instars (growth stages, p. 23), with a winged ‘subadult’ (subimago) preceding the adult (imago) stage. Whichever association is accepted, it is clear that mayflies and dragonflies are the earliest winged insects that have persisted to the present, albeit as rather smaller forms than their enormous ancestors. Although the term is commonly discarded, reflecting their possible independent origins noted above, they are still commonly referred to as Palaeoptera (‘ancient-winged insects’, to distinguish them from the later ‘Neoptera’). The main unifying feature for this juxtaposition is that their wings cannot be flexed – folded back to rest along the insect’s body when not in use – so that all modern mayflies have the wings held at right angles, upward from the body, and the two suborders of Odonata have the wings held horizontally extended (dragonflies, Epiproctophora, often noted under the earlier less-embracing but more familiar name Anisoptera) or vertical (damselflies, Zygoptera). The flexing is an important change, found in all the more recent groups of winged insects – although lost in butterflies, that is a secondary specialised development. Wing flexing means that the insects can reduce their effective body size, much as many naval aircraft are modified for hangar storage on board by ‘folding their wings’. In locusts and many others, this habit may be accompanied by ‘pleating’ of the hind wing, which is much larger than the forewing, so that it folds fanwise. The adaptation may aid rapid movement on the substrate, as escape from predators and to reach refuges (such as under vegetation, or in cracks in the ground or under bark); the smaller effective size enables insects to crawl into small spaces more easily; wings are not as likely to get damaged on vegetation, and so on – in short, one can postulate advantages for this structural change in increasing chances of the insects surviving.
18
2 Fossils and Major Insect Adaptations
Wings and Ecology Numerous subsequent adaptations of wings have taken place – many insects have highly modified structures derived from wings, have lost one or other pair, or have even become secondarily wingless. Or wings may be used for purposes other than flight – for example, a few stoneflies raise their wings as sails enabling them to be blown across water surfaces. Variations within a species are also frequent: many aphids have winged and wingless individuals in a population, either together or in different generations, and some moths have fully winged males and flightless females with rudimentary, stunted wings. As examples of changes to the basic pattern of two pairs of typical wings used for flight, beetles have the fore wings hardened as elytra, to constitute protective ‘shells’ and give Coleoptera their characteristic appearance so that only the hind wings are employed in flight; flies (Diptera, the name meaning ‘two-winged’) rely on fore wings alone for flight, with the hind wings transformed into small gyroscope-like structures (halteres) that aid aerodynamic manouverability; and ectoparasitic insects such as lice and fleas have lost their wings completely. They can rely on their mammal or bird hosts for transport, and do not ‘need to fly’. Reasons for secondary flightlessness in so many insects have been debated extensively. It has evolved many times independently in different groups and, with the exception of ectoparasites, loss of wings is likely to restrict possibilities of long distance dispersal, and it is indeed common in insects living in very restricted but for long suitable habitats (such as caves, on isolated islands, and amongst denizens of social insect colonies) from which dispersal might be very risky. Wings may be lost in only one sex – most commonly the female. In some moths wingless females can attract flying males by pheromones (p. 5), and one evolutionary suggestion is that , because they do not have to allocate energy to wing development during growth, that energy is then available for enhanced reproduction such as producing more or larger eggs. Some other insects undergo generations of winged and wingless individuals. Some aphids disperse by flight early in the warmer periods of the year, at the start of a breeding season, and the populations then ‘stabilise’ in a habitat by offspring being wingless, perhaps over several generations, before winged individuals are again produced. Wings can be textured, haired or naked, coloured to facilitate camouflage, display, defence or other purpose. The elaborate colour patterns that render Lepidoptera so attractive to collectors are formed from the covering of flattened scales on the wings, and the pattern may be highly characteristic for particular taxa, but also influenced by temperature and other factors. The well-known phenomenon of ‘industrial melanism’ in the peppered moth (Biston betularia) in Britain has been discussed extensively – with rise of heavy industry, and sooty emissions blackening tree trunks on which the nocturnal moths were purported to rest by day, proportions of black (melanic) moths increased, and the normal paler forms declined in frequency, as they were more conspicuous and vulnerable to bird predators. Somewhat intri guingly, in view of so much of Australia’s forest being burned at intervals and the trunks of eucalypts commonly blackened by charring, we have very few black
Insect Diversification
19
moths. One southern geometrid moth, Melanodes anthracitaria, however, does have a wholly black form as one of its two colour patterns: as a forest species it may gain some benefit from crypsis, but this has not been investigated experimentally. A few other blackish moths, including some Noctuidae, are not primarily forest species. Wings are the most numerous insect fossils from the older geological periods, from the Carboniferous on. They are tough, decay-resistant, and commonly discarded by consumers, and their essentially two-dimensional structure renders them far easier to study than the more complex fossils of insect bodies – even though interpreting the patterns they show can be much harder. That wings have persisted so long and diversified so widely attests to their contribution to insect success; they are regarded widely as the single major advance that has facilitated insect domination of so many ecosystems. They have been pivotal in demonstrating the increasing diversification of insects from the Carboniferous onward, as global conditions changed. With the rise of gymnosperms during the Permian period, for example, some herbivorous groups proliferated – and the Palaeoptera began to decline from their Carboniferous glory.
Insect Diversification Fossils laid down over the next 150 million years or so demonstrate the changing balance in global insect fauna, and responses to environmental changes – essentially, as new resources arose, insects developed to exploit them – but also that many basic insect types (orders) have actually changed very little in form from those times. Many present-day cockroaches (Blattodea), for example, closely resemble their ancient ancestors in general appearance. Much later, with evolution of flowering plants and warm-blooded vertebrates in the Cretaceous period, the most recent insect orders became distinctive. The nectar-sucking Lepidoptera did not appear in the fossil record until flowering plants were present, and parasites such as fleas could not develop in their modern forms until mammals and birds became available as warm-blooded hosts. Other new resources such as warm dung also fostered specialist insect consumers – and strong suggestion of parallels with some adult beetle and flies using nectar, whilst breeding in dung or vertebrate carrion aiding the ecological divergences between adult and larval stages of the same species. Few such opportunities on land or in freshwater have been ignored during insect evolution. However, a major anomaly arises from this picture – simply, if insects are so successful in these terrestrial and freshwater environments, why are they not also predominant in marine environments? Marine insects do exist, and range from bugs that skate on the water surface in parts of the Pacific (some passing their whole life up to thousands of kilometres from land), to a variety of species on the shorelines – some flies and beetles exploit debris such as seaweeds and dead fish or birds washed up on the strand, for example. But they have indeed not fully exploited the ocean environments. Physiologically, some are certainly capable of doing so – insects regularly exploit highly saline waters elsewhere (p. 139) – and the most
20
2 Fossils and Major Insect Adaptations
likely suggestions reflect that insects arose on land, from immediate terrestrial ancestors, at a time when the land environment was not overly crowded and where they could establish their evolutionary claims without undue competition. However, at the same times, other arthropods were proliferating in the seas – so that land might have been the ‘easier option’, whilst possible forays into marine systems were thwarted by the organisms already present. Oceans were perhaps just too difficult for insects to exploit, but whether this was because of competition or the complex physiological demands of respiration and osmoregulation – or both of these – must remain speculative. Some entomologists suggest that simply the physical problems of coastal wave actions, turbulence and tides may effectively prevent land insects invading the seas. Most marine insects occupy coastal habitats such as intertidal zones, salt marshes, mangrove swamps or others that may be regarded as intermediate between sea and land. One caddisfly, Philanisus plebeius, representing an order in which larvae are most typically found in fresh water, lives in rockpools in south east Australia and New Zealand. Some insect orders are known only from fossils, and have clearly become extinct as they were replaced by ever more efficient forms adapted to more recent conditions. The older insect fossils, from which most of this long-term story is inferred, are shale impressions in two dimensions, and the amount of detailed information derived from them varies considerably, depending on the individual condition and orientation of structures. Much later, the record was augmented by a second category of fossils, those in amber, most commonly from the Baltic region and with insect inclusions highly desirable for jewelry. Amber is the fossilised resin of conifers, in which insects were trapped soon after it was exuded and glue-like – modern functional equivalents are common in Australia, where the ‘gum’ oozing from wounded or stressed acacias or gum trees often captures visiting insects attracted by scent, by the promise of a meal by feeding on insects already trapped, or simply blundering in. More resin (sometimes referred to as ‘sap’, but in reality somewhat different in origin) eventually covers the insect, to enclose it, in course hardening and persisting indefinitely. Amber fossils are thus whole insects, preserved in manner equivalent to mounting in gum on a microscope slide and available for very detailed examination. Some are up to about 100 million years old, from the Cretaceous period. Amber from different periods and widely separated locations gives us snapshots of the insect assemblages present. Older Cretaceous deposits include those from Canada and Lebanon. The Baltic amber is much more recent, ‘only’ 40–50 million years old. Particularly for smaller insects (essentially those not available in the sedimentary fossil record, from which most specimens detected are relatively large) very fine details can be discerned, to the extent of being able to measure individual hairs and leg parts after careful polishing and grinding of the amber bead to provide clear viewing from a variety of angles. In general, Cretaceous amber insects are all referable to extant orders, and most to families and genera that can be related easily to modern insects and, in many cases are identical to these. Baltic amber insects are even closer to modern forms. Many, indeed, are apparently of the same species as insects found today, although usually far from northern Europe. Amber fossils also
Insect Diversification
21
reveal the long term stability and ancient nature of some intricate ecological relationships – for example, they include a number of tiny wasps likely to share the habit of their modern relatives of developing inside other insects, and we can sometimes even postulate what kind of insects their hosts may have been. Aggregations of fossils can allow at least tentative ecological interpretations. Australian insect fossils have contributed to this framework – several hundred species, spanning about 20 orders, have been described, and several classic fossil localities are well known to palaeontologists. The earliest undoubted insect fossil from Australia is a palaeodictyopteran from an Upper Carboniferous site in Tasmania. The Permian deposits in New South Wales (notably from Belmont and Warner’s Bay) yielded a considerable variety of taxa, very distinct from those of the northern hemisphere. Perhaps the two most fruitful insect fossil sites in the country are (1) the Ipswich Triassic series in Queensland, with more than a thousand specimens, some (such as the oldest known aphids) especially notable, and (2) the freshwater lake deposit from the lower Cretaceous at Koonwarra, Victoria, discovered during road construction activities. The latter gained fame also as yielding the first bird feathers from Australia, in conjunction with the earliest known fleas. It includes more than 80 species, in 12 orders, and is notable for the array of aquatic larval forms represented. More recent Tertiary fossils are known from Queensland and New South Wales. The most exciting insect fossil discovery in Australia in recent years (first publicised in 2006) has been that of amber on beach strand washes in northern Queensland, with subsequent findings also elsewhere in the country, and the first knowledge of amber being present in the country. Australian amber is thought to be about 15 million years old, on initial estimation, and derived from kauri trees (Agathis, still found in rainforests of Queensland). Little has so far been published on its insect inclusions, but they are considerable and diverse, with flies, termites, ant, wasps, beetles and others signaled, together with feathers, hairs, and a wealth of other fossils. The first formal description of an Australian amber insect, of a dolichopodid fly in 2009, is surely the precursor of many other exciting discoveries from this amber to help understanding the origins of Australia’s insects. So far we have seen three of the major features of insect evolution, in conjunction with changing representation in the fossil record: development of wings, the basal articulation of wings with the thorax changing to allow wing flexing, and the progressive diversification of feeding habits and mouthpart structure. The fourth critical feature is the development of different developmental patterns within the life cycle, to achieve a ‘complete metamorphosis’. This is outlined next, but for completeness here, the development by some insects of a true social existence (Chap. 7), whereby some aspects of environmental variation may be buffered or countered, has also frequently been noted as a further factor contributing to their success. ‘Success’ in insects can be considered superlative in both evolutionary and ecological terms – reflecting long persistence of early origins and well-established lineages, many of which have generated vast numbers of species but within which the basic pattern has essentially changed little. The second category also reflects that some insect groups have gained predominant ‘key’ roles in ecosystems. The social insects
22
2 Fossils and Major Insect Adaptations
(Chap. 7) could be considered amongst the paramount evolutionary successes of insects. Only modestly rich in species compared to many far more diverse groups, termites (globally with around 3,000 described species), have ecological importance far beyond this tiny proportion of insect richness, as the major consumers of cellulose in dead plant biomass. The ubiquity and ecological importance of ants also confers major influences in many terrestrial ecosystems, both in Australia and elsewhere. But these two social insect groups have very different developmental pathways, and understanding the major life history patterns of insects is a critical aspect of interpreting insect evolution and diversity.
Further Reading Bickel DJ (2009) The first species described from Cape York amber, Australia: Chaetopogonopteron bethnorrisae n.sp. (Diptera: Dolichopodidae). In Berning B, Podenas S (eds.) Amber – archive of deep time. pp 35–39. Denisia 26 Grimaldi D, Engel MS (2005) Evolution of the insects. Cambridge University Press, Cambridge (comprehensive, fully-illustrated, recent survey of the insect fossil record and the features and evolution of Recent insects) Jell PA, Roberts J (eds) (1986) Plants and invertebrates from the Lower Cretaceous Koonwarra fossil bed, South Gippsland, Victoria. Association of Australasian Palaeontologists, Sydney (comprehensive, well-illustrated descriptions and appraisal of a major insect fossil deposit in southern Australia) Poinar GO (1993) Insects in amber. Annu Rev Entomol 46:145–159 (overview of information then available) Wootton RJ (2001) How insect wings evolved. In: Woiwod IP, Reynolds DR, Thomas CD (eds.) Insect movement: mechanisms and consequences. CABI Publishing, Wallingford, pp 43–64
Chapter 3
Insect Life Histories
Introduction: Modes of Development In common with other arthropods, being bounded by the confines of a hard external shell means that insects cannot grow continuously but must undergo a series of moults in order to increase in size. The insect body inside separates from the exoskeleton, which is then split and shed, leaving a larger new and initially soft exoskeleton for the insect to expand into. Successive moults separate different stages (stadia, instars) that approach the adult stage with reproductive capability. Along this path, the body increases in size and may change in appearance in various ways. Most commonly, an insect life cycle starts from an egg, hatching to a larva (initially a ‘first instar’ that eventually moults to a ‘second instar’ and so on in a sequence that can be numbered to denote the individual stage). Eventually the adult stage is reached and in most insects no further moults occur and the cycle is completed as adults mate and reproduce. However, the changes between egg and adult differ markedly in extent in different insects and define several widely adopted patterns of insect life cycle. The simplest pattern is that found in the primitive Apterygota which, by definition, do not develop wings. A hatchling silverfish resembles its parent in most details – it is simply much smaller, and cannot reproduce. At each successive moult, it becomes larger but changes very little in appearance – but an adult silverfish continues to moult at intervals throughout its life. Presence of wings in more advanced insects constrains this – in part reflecting that fully developed wings are large delicate structures that can be damaged easily by moulting and may render the insect especially vulnerable whilst doing so – so that (with the single exception of the primitive mayflies, amongst the ancestral forms of winged insects, p. 17) insects with fully developed functional wings do not moult. The more primitive orders of Neoptera develop wings gradually, on the outside of the body. The field cricket egg hatches to a wingless active miniature cricket, feeding in the same way as the adult. After the second or third moult, the appearance changes somewhat, with development of small posterior lobes, a pair each on the
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_3, © Springer Science+Business Media B.V. 2011
23
24
3 Insect Life Histories
upper hind angle of the mesothorax and metathorax. These lobes (‘wing buds’) enlarge with each successive moult and at the final moult are transformed to proper wings by becoming hinged at the base. Crickets thereby show a progressive stepwise transition from wingless to winged form, with the different instars broadly similar in their life style and occurring in the same habitats. This can be contrasted with the far more dramatic changes developed in the more advanced insects, in which the appearance of the larva and adult differ greatly – consider a caterpillar and a moth, or a maggot and a fly, for example. In these, the larva and corresponding adult occur in different environments and have very different feeding habits and biology. A typical caterpillar is a chewing plant-feeder, the adult a nectar-feeding butterfly or moth exploiting this food by means of a long sucking proboscis. Further, the caterpillar or maggot has no external trace of wings. The two phases of larva and adult appear as discrete biological entities, and the transition between these represents one of the major adaptive advances in insect evolution. This transformation results from establishment in the life cycle of the pupa, derived from the last instar larva as a stage that (1) does not feed, and so allows the insect to survive periods when food is not available, (2) in consequence provides a refuge to survive seasons that may not be suitable for continued development because of being too cold, too hot or from other environmental factors, and (3) allows for the larval body to be broken down whilst enclosed in a protective case, and reorganised into the adult form (including formation of external wings) before the latter resumes development in the outer world. Thus, a caterpillar ‘pupates’, and a moth or butterfly subsequently emerges from the pupa (Fig. 3.1). But one evolutionary principle is important to clarify here – whereas a larval silverfish or cricket has always resembled the corresponding adult insect in general form, a caterpillar has never resembled a moth, nor a maggot resembled a fly. The pupa stage facilitates generating changes in both ‘directions’ – so that, simplistically, both the caterpillar and the moth have been derived from the same basic pattern but diverged as they exploit different lifestyles, so that their various distinguishing characters have largely been acquired independently. From this, we have three broad patterns of insect development, in summary: 1 . Silverfish and their allies: little or no change in form as they grow. 2. Crickets and many others: gradual change of form, with development of wings as projections on the outside of the body. 3. Most more advanced insects: considerable change of form from larva to adult, with an intermediate pupa stage and internal development of wings. Post-embryonic change of body form is termed ‘metamorphosis’, and the above categories therefore show different kinds of metamorphosis. With no clear change of form, Apterygota are termed ‘ametabolous’; external development of wings, literally ‘exopterygote’, is referred to as an incomplete metamorphosis, more formally the insects are ‘hemimetabolous’; and the pupa results in internal wing development (‘endopterygote’) with a complete metamorphosis, and the insects designated as ‘holometabolous’. Holometabolous insects are by far the predominant group, with close to 90% of living insects showing this pattern, essentially of having two different ways of life
Diversification Within Metamorphosis
25
Fig. 3.1 Emergence from the pupa and expansion of wings is a vulnerable phase in the life cycle of many insects. This female common brown butterfly (Heteronympha merope, Nymphalidae) is completing this phase whilst hanging from vegetation and with the mottled brown underside possibly aiding its camouflage
within the same species. But, as we would expect, some complications occur, with needs to operate in different milieux leading to other divergences between adult and larval forms. The primitive Palaeoptera arose long before any complete metamorphosis – yet the transition between aquatic larvae and terrestrial adults, with accompanying changes in feeding habits and other ecological features also results in very different life forms without the intervention of a pupa. The final instar larva (having developed external wingbuds as in early terrestrial winged insects) leaves the water and moults directly to the land-based adult stage. A similar pattern of highly derived larval forms for aquatic life is found also in the stoneflies (Plecoptera). All of these larval forms face the same problems – they need gills for aquatic respiration, for example – but their transitions do not represent holometabolous development. The larvae of mayflies and dragonflies are sometimes referred to as ‘naiads’ (water nymphs), and all larvae of exopterygote insects are sometimes termed ‘nymphs’.
Diversification Within Metamorphosis In later chapters we will see many individual departures from the generalities implied by these patterns – emphasising the difficulties of attempting any sweeping statements to encompass insect variety! Not all insects have an egg stage,
26
3 Insect Life Histories
Fig. 3.2 The egg-case (ootheca) of a large mantis, Archimantis latistylus (Mantodea), is formed from proteinaceous foam exuded by the female, that hardens to enclose her eggs as they are laid, and is attached to vegetation. The eggs are protected from desiccation and some attack by predators and parasitoids, but can still be attacked heavily by small wasps (Podagrion), whose females have a long slender ovipositor that can reach the eggs through the foam
for example, as some are viviparous. Many aphids, some flies and others (so both exopterygote and endopterygote insects may be represented) produce active larvae as the first stage and the egg retained inside the parent’s body to (1) shorten the life cycle – so that aphids can undergo generations in very short periods and ‘capitalise’ on suitable environments by building up numbers quickly and (2) increasing precision, so female parasitic flies can deposit individual larvae, as an active feeding stage, directly onto a suitable host to exploit it without delay and avoiding possibly hazardous exposure. And insects may lay from few to many offspring, and either leave these exposed, or protect them in some way from harm (Fig. 3.2). The numbers of larval instars also differ substantially in different insect groups. From an indefinite number with silverfish, many Palaeoptera undergo 20–25 larval instars but this number can decrease, characteristically to only 3–5 in many holo metabolous insects. The precise number may be influenced by food supply and other environmental conditions, together with the length of the developmental period. Larvae of holometabolous insects differ greatly in appearance across different orders and, because of their extensive adaptive characters, they may be just as diagnostic as the corresponding adult stages. However, most Australian insects have been described and named from the adult alone (many from only one sex, so that even associating males and females correctly may be difficult in interpreting samples or survey results) and only a small proportion have been associated clearly with larval stages, by rearing from them. If we are confronted with a beetle larva or caterpillar from a field collection, we may only be able to infer its identity from
Diversification Within Metamorphosis
27
what we know of identified adults found in the vicinity, with that knowledge usually highly incomplete. In contrast, the extensive differentiation between larvae and adults of exopterygote aquatic insects sometimes renders the immature stages better studied than adults: they are often abundant and easy to collect, many are long-lived, and are important components of freshwater fauna. Larvae of Ephemeroptera, Odonata, Plecoptera and holometabolous Trichoptera (caddisflies) have all been studied well because of their practical importance in helping to assess water health. Water quality can sometimes be appraised by looking at the number of species and relative abundance of these insect groups. In North America, the ‘EPT Index’ (for ‘Ephemeroptera, Plecoptera, Trichoptera’) is a commonly used measure and, in general, larvae of these orders may be amongst the better known groups of immature insects. Three very broad categories of endopterygote larvae have been a useful framework in helping to describe their variety, and each of these reflects capability and needs of a particular way of life – primarily the extent of mobility and features of the environment in which they must feed. Historically, these categories have been likened to different embryonic stages, in relation to a sequence of development and later suppression of the paired limbs on each body segment, and the names themselves are derived from embryological terms use to denote stages of development. The parallel is simplistic because the ‘simplest’ larval form is in fact a highly specialised derivative, rather than a primarily early stage, within the most advanced of all insect orders, Hymenoptera. This, the ‘protopod’ larva, appears rather simple in structure, as if at an early stage of development. The body may be incompletely segmented with the abdomen sac-like and sometimes scarcely distinct from the thorax, no thoracic limbs are present, the head lacks eyes and antennae, and only the mouthparts are well-developed. At first impression, this seems an odd and seemingly incapable animal, but it is one whose features are in fact highly adapted to a special way of life. It is typically the offspring of some parasitic (strictly, parasitoid, p. 85) wasps that lay their eggs inside a host (such as a caterpillar) in which the whole larval development takes place. The larva is thus bathed in abundant food (caterpillar body fluids and tissues), does not need to move around much or to hold on in its small operating universe, has no need of vision or other major sensory systems – it simply persists and feeds in an environment that is sheltered, nutritious and largely predictable. The countering possible disadvantage is that it cannot escape – if anything happens to the host (for example, by it being eaten by another arthropod or a bird, or killed by a virus or other pathogen), the internal parasitic wasp is assuredly also lost, and the precise placement by the parent female may then be futile. One evolutionary way to, in turn, counter this eventuality is for the wasp to invest rather little energy in each individual offspring but compensate by producing large numbers, distributed amongst many different individual hosts, some of which may survive. Many such wasps are highly dispersive and search efficiently for hosts – and energy spent in searching activity is then not available for reproduction; but with the sheltered environment, small protopod larvae can be produced at this early stage with their necessary capabilities assured. A related feature of some parasitic wasps is ‘polyembryony’ whereby a single egg or embryo deposited in a
28
3 Insect Life Histories
Fig. 3.3 Caterpillar of the vine moth (Phalaenoides glycine) (Noctuidae, Agaristinae), experimentally attacked by two polyembryonic wasp parasitoids at different times. Each single oviposition has led to proliferation of larvae, as exposed here by pricking the caterpillar skin, so that high numbers of offspring are produced with minimal expenditure by the parent wasp
host divides to produce large numbers of offspring (Fig. 3.3) – so that a host that survives until the parasitoids have developed can supply a whole cohort of wasps, perhaps of several hundred individuals, to the next generation (Fig. 3.4): a very effective counter to some other hosts being lost. Surviving caterpillars eventully proceed to adulthood (Fig. 3.5). The second of the major categories of larvae (Fig. 3.6a) is associated with living on vegetation, with needs to feed by chewing, to move around and to be able to hang on during wet and windy weather – with some exceptions due to mobility, a caterpillar dislodged from a tall tree is doomed just as surely as a wasp inside an eaten caterpillar! These larvae are termed ‘polypod’ (many-legged). They have a well-developed head, with chewing mouthparts and a group of simple eyes replacing the normal adult compound eyes, and the usual insect pattern of three pairs of thoracic legs – but, in addition, they have a variable number of fleshy lobelike ‘prolegs’ on the abdomen, and a pair of these constituting posterior ‘claspers’. The prolegs are used to grip the substrate, such as to clamp the edges of leaves, and their broad surface can have circlets of small hooks or spines, as a natural-world precursor to some human ‘fastener’ inventions, such as ‘velcro’®. Lepidoptera and plant – feeding sawflies (Hymenoptera: the black or dark green ‘spitfire’ grubs found on eucalypts, sometimes in large numbers or dense clumps, are familiar to many people: Fig. 3.7) exemplify this category. Finally, but with much more extensive diversification from its basic form, the ‘oligopod’ larva (Fig. 3.6b–d) fits the typical insect pattern of three body regions, primarily chewing mouthparts, and limbs restricted to the thorax. As in polypods,
Diversification Within Metamorphosis
29
Fig. 3.4 Caterpillar of the vine moth, above, from which fully grown parasitoid larvae have emerged, each to spin a small silken cocoon in which to pupate
Fig. 3.5 The adult Phalaenoides glycine is a colourful, day-flying moth with aposematic colouring, perhaps deterring vertebrate predators from attacking it. It is an important pest of vines in southern Australia
eyes are simple (other than in scorpionflies, Mecoptera, in which compound eyes are retained) and comprise a small group of individual stemmata, commonly six or seven in number. The variety of oligopods can be demonstrated by contrasting two forms of beetle larvae, representing very different ways of life. Active predatory
Fig. 3.6 Representative larvae of holometabolous insects to indicate some broad functional categories. (a) polypod larva of Lepidoptera with normal thoracic legs, several pairs of abdominal prologs, and posterior claspers; (b) oligopod larva, a typical beetle to indicate basic pattern of thoracic legs and no prolegs; (c) modification of this as active predatory ‘campodeiform’ larva as in ground beetles – head orientated forward so mouthparts anterior, legs long, and body streamlined; (d) a contrasting sedentary ‘scarabaeiform’ beetle larva adapted for an inactive life style spent underground and feeding on plant roots; (e) a ‘maggot’ of higher Diptera, lacking limbs or a distinct head, and with mouthparts as small hooks retracted into thorax, lateral spiracles absent and large posterior spiracle present
Fig. 3.7 Representative larvae of polypod, the ‘spitfire larvae’ of a pergid sawfly on Eucalyptus
Diversification Within Metamorphosis
31
Fig. 3.8 Scarabaeiform larva of a stag beetle, Lucanidae
beetle larvae such as those of many rove beetles and ground beetles chase their prey, detect them by sight, and capture them with their mandibles. They tend to have slender bodies, raised above the ground on long legs, used for running, and the head axis ‘tilted’ so that jaws are orientated forward and encounter prey organisms in front of the insect, rather than underneath it. Pasture scarab beetles differ greatly from this appearance and have a very characteristic body form (giving them a common name of ‘curl grubs’). The head is small, with mouthparts in the usual ventral position, thoracic legs are short, and the abdomen is greatly enlarged and ‘curled’. They live underground, feeding on plant roots, so do not need to move fast, but must accommodate and process considerable volumes of food material (Fig. 3.8). These two contrasting kinds of larvae retain the full oligopod complement of features. But, in response to exploiting different feeding habits and habitats, many other oligopod larvae have undergone considerable modification, with the major trends including loss of limbs and heads to varying extents. A bushfly maggot (Fig. 3.6e) living in dung has no distinct head, and its mouthparts are reduced to small hooks retracted into the front of the body; likewise it has no legs. Larvae of some timber-boring beetles retain the head, needing strong mandibles to chew open their tunnels and fragment the wood on which they feed: again, they lack legs but may have various processes on the body, these aiding purchase on the sides of their excavations. Many such variations occur. As we might expect, insect pupae also provide similar structural variety related to adaptive need. They can be long-lasting, providing a refuge for the insect over much of the year, and can thus be vulnerable to attack or loss. Several trends toward apparent protection occur, sometimes in conjunction – crypsis (camouflage if exposed, or formed in less exposed sites such as under bark or in holes in the ground), toughening (with the outer case hard, ornamented or ‘polished’ to resist bites), or a protective covering (cocoon). The large pupae of the gum emperor moths
32
3 Insect Life Histories
Fig. 3.9 Representative pupae: some basic forms. (a) decticous exarate pupa (e.g. Megaloptera) with limbs and jaws free; (b) adecticous obtect pupa, with all appendages sealed to the body, and usually with a toughened outer covering (e.g. Lepidoptera); (c) adecticous exarate pupa, with legs and antennae free (e.g. some Hymenoptera)
(Opidiphthera) are enclosed in a hard egg-shaped cocoon, constructed of bark fragments and silk produced by the caterpillar before it pupates, and this is a very tough structure formed usually on or under eucalypt tree bark. Some other moths incorporate stinging body hairs into a more delicate silken cocoon. However, becoming captive in this way poses the problem of emerging later; being enclosed in a barrier sufficient to repel intruders also necessitates being able to break through it from inside. The more primitive endopterygotes such as Neuroptera have pupae with functional mandibles and some mobile legs – they are able to bite their way out of a cocoon, or move to more exposed areas before moulting to the adult. This mobility is lost in more advanced orders, in many of which the jaws are non-functional and all limbs are sealed to the body surface – the basic pupal forms (Fig. 3.9) are thereby termed (1) decticous (jaw-bearing) and adecticous (without jaws) with the second divided into ‘obtect’ (all limbs cemented to the body and confined within the pupal case) and ‘exarate’ (with some use of legs to aid limited movement). The adecticous pupae commonly have spines and processes on the body, to help the emerging (pharate) adult push out of the pupa case, sometimes through predetermined lines of weakness, or through ‘one way funnels’ in the pupal case or cocoon.
Seasonal Development The various growth stages integrate into sequences of development that fit any insect species for its environment, and may have strongly seasonal patterns that reflect the stability and predictability of the resources needed, and the physical environment in
Seasonal Development
33
which these occur. Some environments vary much more than others, of course. Domestic environments – insides of houses, warehouses or other food storage areas may be well-buffered against climatic variations, so that conditions vary little throughout the year. Insects living there may be able to breed year-round. At least part of the reason why some beetles found in stored grains or termites in buildings gain pest status is simply that the enclosed conditions allow them to build up vast populations continuously, unhampered by climate or food shortage. Most insects living out-of-doors cannot do this. Seasonal timing of insect development has two main constraints, both intuitively obvious but which are fundamental considerations in the evolution of life history patterns. First, feeding stages (larvae, adults – the latter with some exceptions that do not feed) must occur at times of the year when suitable food is available. Second, particular seasons may be too cold or too hot to permit development, so that ‘resting’ (non-feeding, inactive) stages may be needed to overcome them. The times optimal for development may be short – perhaps only a few months in spring and summer when temperatures are warm enough and a good supply of food is present – and in the most clearcut cases, the rest of the year may be too cold and food in short supply or of low quality. Insects have collectively evolved two major strategies to help them cope with this variation. They can take refuge in space, or in time. The first necessitates movement, most commonly migration by flight to track resources in the landscape – as one area becomes unsuitable, another is exploited. The second involves some form of dormancy, with the insects in non-feeding stages or protected sites where development ceases (or is reduced markedly) until conditions improve. The two can be combined, because migration only ‘works’ in tracking resources as long as physical conditions within the dispersal range allow continued feeding or if the insect does not need to feed whilst migrating. Movement may also be made to dormancy sites. The Bogong moth (Agrotis infusa) is one of Australia’s most iconic migrant insects. It breeds in lowland areas of the southeast, where caterpillars eat grasses, cereal crops, and similar vegetation. However, as conditions warm in early summer the moths migrate to higher ground and spend the summer sheltering gregariously in caves or under rocky overhangs in the Australian Alps in a dormant state of ‘aestivation’ (the summer equivalent to winter hibernation). Their migrations (p. 150) involve many thousands of moths, and cause comment in most years – vast numbers of moths are attracted to the lights of evening sporting events, and a parliamentary report in 2006 discussed methods of preventing the nuisance they create by entering Parliament House in Canberra (for which the recommendations included closing doors and turning off lights late at night!). In autumn, as conditions become cooler, moths return to lowland areas and resume breeding on fresh growth of the food plants. This migration has other ecological interests but contrasts with that of many other insects in which winter refuge is needed. Inclement periods are most commonly passed in non-feeding stages, so that the Bogong moth (in which the adult becomes dormant) is relatively unusual – overwintering eggs or pupae are much more commonly adopted, with each produ cing feeding stages as conditions again become suitable. This broad dormancy is
34
3 Insect Life Histories
undertaken in many insects by the particular situation known as diapause. The term was coined (in the 1890s) to describe a period of embryonic pause within the egg, but is now used to denote arrested development in any life stage. Within different insects, it may be obligatory (the most common condition, affording a consistent feature in the life cycle) or facultative. The seasonal regulation of diapause, both its onset and termination, necessitates some recurrent or cyclic environmental cues as ‘triggers’. The most predictable of all these factors is photoperiod (day length), commonly with temperature also involved and mediated through changing hormonal balances within the insects. In temperate regions such as southern Australia, day length varies considerably throughout the year – between the shortest day in June it increases to the longest at the summer solstice in December, and thereafter decreases again so that day length varies in a wholly predictable cyclic pattern. Insects may cue in to these changes to ‘anticipate’ (in evolutionary terms) when living conditions become adverse and subsequently suitable. Thus, many insects active during the summer period respond to shortening days to enter diapause at or near a particular day length threshold in autumn, so that their refuge is to reach a safe stage in their life cycle by that date. In spring, those same insects respond to increasing daylength to resume development: eggs and pupae hatch (breaking developmental diapause), diapausing reproductively immature adults mature and reproduce (breaking reproductive diapause), and so on. The strategy is then to exploit the suitable environment as effectively as possible before conditions again degenerate. The role of diapause is to maximise survival in circumstances in which seasonal conditions do not permit normal development to continue. Whereas photoperiod is a very common regulator, reflecting its predictability, temperature is also important in modifying the insect’s response to critical photoperiods – high temperatures may prevent or delay diapause, and cooler temperature hasten its onset. Likewise a minimum temperature, even a cold snap, may be needed to force ending of diapause. Dietary and genetic influences may also occur. Broadly, diapause is associated with (1) reduced metabolism and development, (2) reduced activity and (3) resisting environmental conditions that are unsuitable for the insect. With obligatory diapause, the most common syndrome, a very well-defined seasonal pattern can be established, particularly when only one generation each year is undertaken. Greater developmental flexibility occurs with more numerous generations, because conditions within the favourable period may affect these in different ways. The adoption of holometabolous development, allowing insects massive opportunity to diversify (even within individual species), vies with evolution of wings as one of the most critical phases of their evolution. Its importance is indicated by all of the ‘hyperdiverse’ orders exhibiting this pattern. The major lineages of Recent insects can be summarised as follows, and the constituent orders are listed in Table A.1 (p. 224), simply to introduce them: some will be very familiar to most readers as components of ‘nature’ that impinge on our daily lives in various ways – others will probably be less familiar, but also just as important in the roles they pursue. 1. Apterygota Primitive wingless insects with ametabolous development, and continued moulting as adults.
Seasonal Development
35
2. Pterygota (all winged insects) Palaeoptera (‘ancient-winged insects’). The most primitive winged insects, lacking ability to flex wings; exopterygote but with strong secondary specialisations for aquatic larval life. Neoptera (‘new-winged insects’). Can flex wings. Polyneoptera. The most primitive series of orders; chewing mouthparts, exopterygote. Paraneoptera. Based on sucking mouthparts for liquid food, and derivations from these, exopterygote. Oligoneoptera. Most advanced insects, very variable orders, united by being endopterygote. Of these series, Apterygota are a tiny remnant fraction of modern insects, and Palaeoptera have decreased markedly since their Carboniferous predominance. The three series of Neoptera are noted here: Polyneoptera are the first-evolved, and include those insects most similar to the basic pattern, such as crickets; almost all are terrestrial. Paraneoptera are dominated by the sucking bugs (Hemiptera) and apparently arose first as plants with sap became available. Later evolution of a complete metamorphosis allowed the Oligoneoptera to diversify immensely and become by far the predominant modern insects. Several of the extant orders of insects do not even occur in Australia, and these are noted here to aid clearer focus on those that are present. As noted earlier, Mantophasmatodea is the most recently diagnosed insect order, discovered alive only in 2002, and testament to the novelties that may still await discovery. They are a curious intermediate group within the Polyneoptera, native to southern and eastern Africa where they occur in a variety of humid to arid semidesert habitats, as predators. They seem to be closely related to another group absent from Australia, the Grylloblattodea (ice crawlers, rock crawlers), small cricket-like organisms confined to the temperate northern hemisphere zones – indeed, as noted earlier, there is now some consensus amongst entomologists that these should be treated as the same order, under the name Notoptera. Grylloblattids are most commonly scavengers, and have been reported from caves and along the edges of retreating snowlines feeding on insect carrion as it becomes exposed. The other major absence from Australia is the Raphidioptera, snakeflies (named for their elongated thorax), related closely to Neuroptera amongst primitive holometabolous insects. They are predo minantly northern hemisphere insects, feeding on small arthropods on vegetation and larvae commonly living under bark, but extend into south eastern Asia as far as Thailand. Finally, Zoraptera are included here but in a slightly different context. They are unknown from the Australian mainland and immediate region but a species has been described from Christmas Island – politically Australian, but biogeographically a part of Indonesia. These small Polyneoptera occur in leaf litter and under bark, and have been reported also from New Guinea, so it is still possible that they might be found in the forests of Cape York Peninsula. The other orders listed all occur in Australia, but to varying extents and reflecting different origins and distribution patterns. Some background to these themes aids
36
3 Insect Life Histories
understanding their ecology, as well as characterising the peculiarities of Australia’s insects a little further. A brief systematic synopsis of Australia’s insect orders is included in an Appendix to this book.
Further Reading (The texts listed under Chap. 1 are all useful adjuncts to this chapter) Tauber MJ, Tauber CA, Masaki S (1986) Seasonal adaptations of insects. Oxford University Press, New York (comprehensive synthesis of early work on diapause and seasonal patterns of life histories in insects)
Chapter 4
Origins, Distributions and Diversity
Introduction: Australia as an Environment for Insects Unlike many other parts of the world, we still have the privilege of being able to evaluate and study much of Australia’s insect life under reasonably natural conditions. Despite the extensive and severe changes to Australian environments over the rather short period since European settlement, and continuing human pressures on natural ecosystems, many of them – at least in remnant form – still maintain sufficient naturalness to support understanding of patterns of insect distribution and evolution, species richness, endemism, genetic and ecological variety within species, and other fundamental aspects of ‘diversity’. Changes to Australia’s natural ecosystems have almost certainly already led to the demise of many insects and other invertebrate species, so that there is some urgency about how we may be able to proceed on such study within the island continent. Without increased commitment to conserving Australia’s remaining natural environments, and enhanced resources to survey and study our biota, the opportunity to do so will be lost. It can never be regained. The array of insect orders in Australia differ enormously in their size, variety, distributions, and relationships with insects in other parts of the world – but a predominant feature of almost any group is that most species do not occur naturally outside the Australian region, so are endemic. This area has long been recognised as one of Earth’s major biogeographical regions but, reflecting the size and climatic variety of Australia, most native insects do not occur throughout the country. They are restricted by environmental factors and dispersal prowess, as well as by their geographical origins – and the last helps to delimit a number of ‘faunal elements’ that are highly characteristic of parts of the continent. However, interpreting the patterns of insect distribution can really only be accomplished if we know something of their relationships – so that we can infer the ways in which the species of – say – a particular family of beetles or moths have differentiated by looking at character changes as gradual transitions that may show taxonomic relationships, and also know how the group concerned reached or has evolved in Australia.
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_4, © Springer Science+Business Media B.V. 2011
37
38
4 Origins, Distributions and Diversity
Sound biogeography includes considerations of both dispersal and of evolution, and what influences these. Modern Australia is thought of as an isolated island but although it now has no contiguous boundaries with any other place, this has not always been the case. To the north, Torres Strait has existed only for about 8,000 years, to separate Australia from New Guinea, and Australia itself is a fragment of the former great southern supercontinent Gondwana. This broke up progressively since the Jurassic period (somewhat more than 150 million years ago), but with the fragment comprising Australia, Antarctica and South America dividing later – so that Australia became isolated ‘only’ about 40 million years ago (Fig. 4.1). The Australian continental plate continued to drift northward through more than 20° of latitude, meeting south east Asia with its very different origins and biotic remnants. Origins of the Oriental realm fauna are also complex, and biogeographers continue to debate the boundaries between this and Australasia through the complex region of the Indonesian archipelago. Over this period, Australia bore with it parts of the biota shared with other components of Gondwana, and subsequently related to those developing independently on the other ‘southern continents’ having the same ancient origin. These insects thereby have close affinities with taxa in temperate South America, southern Africa or New Zealand, and are ancient lineages largely restricted to – usually – some of these areas. Many freshwater insects, several groups confined largely to cooler streams in the south east, fall into this category. As Australia approached other more northerly places by continental drift, elements of the insect fauna of south eastern Asia were able to disperse to meet it, so that much of the northern fauna now has strong relationships with these taxa, which are often very easily distinguished from more southern forms. Some, however, indeed extend to the south east: a number of dragonflies, mostly strong-flying insects, appear to have originated in the Oriental region, having progressively migrated from ancestral stocks in the Indonesian archipelago or beyond. Many insects in the tropical north are closely related to those of Indonesia and New Guinea – the islands of Torres Strait form a series of ‘stepping stones’ for dispersal of organisms in both directions, so that Cape York Peninsula is a region of faunal interdigitation. As with Gondwanan fauna, elements from the north and northwest have undergone extensive differentiation in the new environments – but northern invasions have continued over long periods, with some separation of older and more recent elements, reflected in the extent of distributions within Australia and the degree of taxonomic radiation and diversification they show. Simplistically, several ‘waves’ of invaders can be inferred from the extent to which they have diversified since arrival, but the process is clearly continuous. Two other components of the insect fauna are important to note – one that is understood quite well and one that is decidedly puzzling. The first, with major influences on Australia’s ecology and economy are the ‘recent insects’, those that have joined the fauna since European settlement and widely termed ‘alien’ or ‘exotic’. These have either arrived naturally or been brought (deliberately or unintentionally) by people – an activity that still occurs, with the vigilance of quarantine authorities helping to prevent arrivals of many further species at our ports and airports every year. Representatives of this category are the best-known insects to many people, and some
Introduction: Australia as an Environment for Insects
39
Fig. 4.1 The gradual isolation of Australia from the breakup of Gondwana, to indicate origins and geographical relationships of ancient Gondwanan biota now found in Australia: (a) Gondwana at 150 million years before present (MYBP); (b) 70 MYBP, Australia (Au) narrowly separated from Antarctica (An), Africa (A) and India (I) more so; (c) 40 MYBP, Australia more distinctly separated from Antarctica, New Guinea not yet distinct; (d) New Guinea separated from Australia; (e) Present, with Australia considerably further north, having moved about 20° of latitude since initial separation. Present south pole shown (black spot) for reference; in ‘b’, Nothofagus forests from South America, across eastern Antarctica to south eastern Australia; in ‘c’ Nothofagus throughout Antarctica and Australia; by ‘d’ Eucalyptus widespread through central and western Australia (After Attiwill and Wilson 2006)
40
4 Origins, Distributions and Diversity
have become widespread and naturalized. They include species variously regarded as beneficial (honey bees, Apis mellifera, but see Chap. 15) and highly damaging pests of crops, commodities and livestock. Species in the second category remain anomalous – their origins are often entirely speculative. This Archaic element is small and comprises a number of taxonomically isolated ancient insects that occur in highly disjunct parts of the world. The scorpion fly family Meropeidae, for example, is known globally from single species in different genera in Western Australia and North America. This ancient family might have a Gondwanan origin, and one authority has speculated this origin, with movement into North America from South America and its loss (or non-discovery: it might really be there!) in the latter continent. Whereas we can infer possible origins of many of our characteristic insects, and trace their evolution and diversification in the country, others remain anomalous – but remarkable – elements of our fauna. However, within Australia, distribution patterns give many clues to origins. Rather few species or genera are found Australiawide, and distributions of individual species range from regional to highly localised. Many are known from only single sites or regions, and some are supposed strongly to be narrow-range or point endemics, found nowhere else on Earth and almost inevitably of high conservation interest. Assessment of broad distribution patterns starts with the delimitation of several broad biogeographical regions as a guide to distinguishing meaningful general elements. Our major focus is on the mainland and Tasmania, but ‘political Australia’ includes also some distant outliers such as Christmas Island (to the south of Java, Indonesia), Lord Howe and Norfolk Islands in the Pacific, and the subantarctic Macquarie Island, some 1,500 km south-south-east of Tasmania. The main scheme for regional division of ‘Australia proper’ flows from that initiated by Baldwin Spencer in 1896, but is still a valuable framework to consider. The 500 mm isohyet was used to distinguish the more humid coastal perimeter of much of Australia from the semiarid/arid interior and western Eyrean region (Fig. 4.2). The northerly Torresian region is the tropical/subtropical monsoon-dominated area to which many of the more recent northerly incursives from the Oriental region are limited, whilst the southern cool temperate Bassian region harbours most of the Gondwanan elements. However, the Bassian region is now divided into three disjunct ‘provinces’, each with its peculiarities. Thus many insects found in the West Bassian province (the south west of Western Australia) occur only in that region, but are related to those in the southeast of the continent, long separated by the dryer Nullarbor Plain. Converse examples include several families of lacewings (Neuroptera) found in the southeast but not in the western province, and parallels can be found in many insect groups. The East Bassian province is itself divided, into the mainland area and the island state of Tasmania – with the intervening Bass Strait formed most recently around 12,000 years ago. Each part supports numerous endemic insects, again many of them highly localised. Biogeographers have sometimes sought to divide these broad regions into finer levels, and any of a substantial number of features may be used to do so – vegetation type, altitude, and humidity are cited widely, but much subdivision has reflected the distribution patterns of the individual animal or plant groups assessed. For our purpose it is sufficient to recognise that different distribution
Introduction: Australia as an Environment for Insects
41
Fig. 4.2 Map of Australia to indicate political divisions (States and Territories) indicated by initial letters as ACT [Australian Capital Territory], NSW ([New South Wales], NT [Northern Territory], Q [Queensland], SA [South Australia], T [Tasmania], V [Victoria], WA [Western Australia]) and main biogeographical regions (Bassian, Torresian, Eyrean); the 500 mm isohyet separates the inland semiarid/arid Eyrean region from the more humid coastal areas; the Bassian region comprises three separate areas, the south west, the south east, and Tasmania; the transition between Bassian and Torresian along the east coast is gradual and various points of demarcation have been proposed
patterns of insects occur, that many may be categorised broadly into origin elements, and that insects are continually mobile and evolving, so that distributions may contract or expand as conditions (such as climate) alter – and, importantly, that insects are sufficiently varied to provide distributional exceptions to almost any generality we seek to define or impose! One important faunal consideration, however, is the interdigitation of Torresian and Bassian insects along the east coast: in short, ‘southern’ insect elements can extend far to the north, some reaching New Guinea, and the converse occurs with some more Torresian taxa extending far down the east coast, particularly along the lowlands east of the Great Dividing Range. Substantial intermingling occurs in the far northern areas of New South Wales and far south east Queensland, with increased species richness of some insect groups there. For many insect groups, these broad zones give us a picture of their major centres of distribution and evolution and indicate hotspots of higher species numbers and biogeographical interest, together with clues to the environmental gradients (such as latitude or elevation) affecting these. High species richness may reflect a source area, as well as particularly suitable conditions for the resources they need. But one possible bias must be mentioned here – simply, much of Australia is still very poorly explored for many insect groups. Few higher taxa are well known, and the collecting intensity may create false impressions of true distributions. Even for the best-known insect group, butterflies – long popular and the focus of hobbyist collectors for well over a century – many gaps in detailed knowledge of species’ distributions remain. Collection intensity has been biased toward regions accessible to major cities and more distant ‘classic’ localities where collectors with limited time available can reasonably expect to find the rarer species they covet,
42
4 Origins, Distributions and Diversity
Fig. 4.3 The tiny (fore wing length about 22 mm) ‘living fossil’ damselfly Hemiphlebia mirabilis (Odonata, Hemiphlebiidae) is known from several widely-separated localities in south eastern Australia, and is of substantial conservation interest. Males have an elaborate courtship display, apparently unique in the order, in which they repeatedly elevate the abdomen, tipped with expanded white flanges, whilst at rest on vegetation
rather than incur the extra time, costs, and risk of exploring new terrain. Information on many other groups is far more tenuous – but the caveats for butterflies, as the most informative group, are clearly of wider relevance. However, for our butterflies, three major centres of higher diversity occur along the east coast: the far northern tip of Cape York with its proximity to Papua New Guinea; the ‘Wet Tropics’ forested region between Cooktown and Townsville; and the region encompassing south east Queensland and northern New South Wales. The last may largely represent faunal mixing of the Bassian and Torresian taxa – but its diversity has also fostered high collector interest, as being easily accessible, particularly, from Brisbane and Sydney. The most diverse groups of insects commonly are those that have been here longest, although there are many exceptions to this, in which very primitive groups have scarcely diversified. Some ancient insects remain isolated single species of ‘living fossils’: the Western Australian scorpion fly noted above (Austromerope poultoni) is one such example. The tiny damselfly Hemiphlebia mirabilis (known as the ‘Ancient Greenling’) in Victoria, South Australia and Tasmania has long been considered an isolated ‘living fossil’ within the Odonata (Fig. 4.3). It seemingly has no close relatives at all and is conventionally placed in a major taxonomic division (superfamily) of its own. It now persists in a few small swampy areas and, although some authorities suggest that it has been derived from one or other of two much more widepread superfamilies, it remains both anomalous and peculiar. The insect fauna is characterised by many such oddities, from isolated single species to more diverse lineages found wholly or almost wholly in Australia. Some of the aquatic insect orders, such as Odonata (with three other families – Hypolestidae, Diphlebiidae, Cordulephyidae – also Australian endemics, and the three species of Petalura, although
Insect Species?
43
dwarfed by their Carboniferous antecedents, are the world’s largest modern dragonflies) and Ephemeroptera, contain many such taxa, but they are widespread also in terrestrial groups. The other major consideration is the occurrence of ‘radiations’ (proliferations of closely related taxa to form many very similar species – some of these taking place in parallel with radiations of major groups of plants: Chap. 8), producing complexes of daunting taxonomic complexity and often highly localised. How these are interpreted has major influences on how we enumerate insect diversity, so some comment is needed here.
Insect Species? Within many such radiations, the diagnostic boundaries between easily-defined entities (normally, species) are often very difficult to discern, and lower level populationbased or geographically discrete units are commonly designated. These are formally ‘subspecies’ (with a third formal name, so ‘Genus, species, subspecies’ augmenting the usual binomial ‘Genus, species’ combination), with the presumption that some consistently recognisable feature(s) can be used to diagnose them. Nevertheless, much of the problem of defining the richness of Australia’s insect fauna rests on our inabilities to define ‘species’ (as the main entities we wish to enumerate) by fully objective criteria. Not surprisingly, understanding the problems involved is both most advanced, and most controversial, amongst the insects we know best – the butterflies, amongst which many subspecies have been named (so that the approximately 420 full species in Australia increases to about 650 named taxa once subspecies are also included). Many local forms, some differing in adult appearance very little from others, attest to the variation that occurs – but often the causes of that variation, its consistency, and the biology leading to and accompanying it are wholly unknown, so that the validity of many butterfly subspecies is hotly debated amongst lepidopterists. In contrast, equivalent debate cannot occur for most insect groups, simply because not enough people are sufficiently interested in them and they are more poorly known, and the opinion of (often) a single specialist cannot be refuted authoritatively and becomes dogma. Although ‘species’ are the most commonly recognised units for enumerating biodiversity, it is worth here emphasising that there is no single universally accepted definition of the term, and that biologists and philosophers alike continue to debate the merits of various ‘species concepts’ across the whole gamut of life forms. The pithy summary that a species is ‘whatever a taxonomist says it is’ as a definition (as ‘taxonomic species’) is pervasive amongst many insect groups, often as a refuge to mask our lack of biological knowledge. Studies of variety, however, encompass ‘biological species’ (drawing on the widespread definition of reproductively isolated biological entities), ‘ecological species’ (occupying different ecological niches and so isolated), ‘evolutionary species’ (differing evolutionary lineages), ‘genetic species’ (within a common gene pool), and ‘morphospecies’ (the typological approach whereby species are defined largely on form and appearance). The last is by far the most common application in insects.
44
4 Origins, Distributions and Diversity
It is thus not surprising that ambiguities occur within treatments by different authorities dealing with the same insects. Within-species (or within ‘species-group’) to enlarge the scope realistically with the hierarchy expressed in Table A.2 (p. 225), several different contexts for variations occur, spanning distribution and isolation, ecology, and the boundaries between related taxa. Two butterfly examples from the south east mainland illustrate some of these contexts, with the initial premises including (1) variations in the appearance of the adult insect are the major basis for naming subspecies; (2) the biological causes of those variations are usually unknown; (3) populations differ in appearance in different parts of the species’ range; and (4) the various populations either intergrade gradually or are geographically disjunct, now occupying isolated fragments of a formerly more continuous range, possibly along one or more environmental gradients. These might originally have been parts of a pattern of continuous gradual variation, or cline. These examples indicate some of the interpretative problems that may arise.
The Yellowish Skipper and Donnysa Skipper, Hesperilla flavescens and H. donnysa Hesperilla donnysa is sometimes called the ‘Varied sedge-skipper’, a very suitable name in view of the considerable variations in colour pattern the butterfly displays across its range. It is very closely related to H. flavescens, with this skipper restricted to small areas of southern Victoria and South Australia. Further confusion and debate continues over the validity of various named subspecies based on this variety. H. flavescens includes two geographically disjunct subspecies on a few sites near Melbourne, Victoria (H. f. flavescens) and the very similar H. f. flavia in South Australia. It is typically confined to near-coastal saline sedgelands, where caterpillars feed on a single species of sedge, Gahnia filum. Whether the two subspecies arose independently in parallel, or are fragments of a formerly more continuous coastal range remains a tantalising problem. H. donnysa mainly occurs further inland with its caterpillars feeding on a wide range of Gahnia sedges. However, uncertainties occur over precise separation of the two species on the adult appearance, as the two intergrade substantially, so that H. flavescens is sometimes considered an extreme yellowish form of H. donnysa, with its range and biology a subset of that of the more widespread, and possibly parental, taxon.
The Swordgrass Brown, Tisiphone abeona Tisiphone abeona is one of the most intriguing and well-studied complexes of a variable butterfly in Australia, with members of each of the coastal populations in the south east (Fig. 4.4) each relatively constant in appearance, but sufficiently distinct to have given rise to seven subspecific names, with the additional complication
Insect Species?
45
Fig. 4.4 Range of Tisiphone abeona subspecies. The various named subspecies of this variable butterfly are each distributed in a circumscribed part of the south east; T. a. ‘joanna’ is a hybrid form occurring where the ranges of two parental subspecies overlap narrowly; the western outliers of T. a. albifascia have sometimes been deemed a separate subspecies, T. a. antoni, but this is not now accepted widely
of an eighth named form (T. a. joanna) near Port Macquarie highly variable and almost certainly a hybrid maintained by meeting of populations to the north (T. a. morrisi) and south (T. a. aurelia). A further form from this area of New South Wales may represent yet another (unnamed) taxon in this group. The southern subspecies are alike in sharing a broad orange band on the fore wing, and the northern ones have more extensive creamy markings. Both series are clinal in character, changing gradually along their range, and the two Victorian subspecies are regarded widely as not distinct, despite their disjunct distributions. Similar patterns and difficulties are almost inevitable in many other insects for which at present we have seen (or distinguished) only the most obvious forms. Insect species and subordinate categories have most traditionally been diagnosed from morphological features alone, usually by people who had never seen the living animal and its environment, so that biological information was almost inevitably fragmentary. Fabricius could have had little idea of the habits of Myrmecia gulosa, of how it might differ from other bull ants (or even whether any such others existed), or how these fit into the panoply of ants in Australia. Myrmecia as now understood contains about 90 described species with only one of these found outside Australia, in New Caledonia. Within the genus, many of the species can be differentiated only on rather small features, such as the arrangement of projections (‘teeth’) on the mandibles. It is not uncommon for specialists in any insect group to find individuals that in some way fall intermediate between described species, and to refer these to a ‘species group’ designated by the name of the nearest similar species to indicate their possible affinity: the currently delimited ‘Myrmecia gulosa species-group’ includes around half the described species of Myrmecia, and an array of subspecies names. Two other examples extend the range of contexts and problems of enumerating insect species and assessing how they have evolved and spread. Brightly coloured jewel beetles (Buprestidae) have long been a focus for collectors. The genus
46
4 Origins, Distributions and Diversity
Fig. 4.5 Different species within a genus may have very different distributions that may overlap in places; the example of four species of fruit flies within the genus Bactrocera (Tephritidae, Dacinae) in eastern Australia, as representatives of a group of important economic pests: all occur in the far north, and the ranges shown for B. musae, B. neohumeralis and B. tryoni all include more northerly and easterly regions occupied by other species (Information from Drew et al. 1982)
Castiarina, by far the most diverse in the family, is restricted to the Australia/ New Guinea region, and some 480 species have been recognised in Australia. Most of these have been described from adults collected on blossom, where they can be numerous and important pollinators. Larvae occur in wood, developing within the trunks of trees and shrubs, and the biology and host plant relations of most species are still unknown. Adult colour pattern has traditionally been a major component of species separations. However, numerous minor variations in this occur, and the causes of such differences have not been comprehensively examined; genetic information on the genus is sparse. Even in groups of insects in which species are relatively easily recognisable, distribution patterns can differ markedly in related species. Figure 4.5 shows the distribution ranges published some years ago for four representative species of dacine fruit flies. Fruit flies are amongst the most serious economic pests in Australia, and their depredations on fruit crops have led to long term investigations on distribution, biology and control, as well as species recognition. Most species are now included in the genus Bactrocera, but numerous species complexes (including more than 20 ‘subgenera’) occur amongst the more than 330 species recorded from Australia and nearby parts of the Pacific. Their differing ‘ecological tolerances’ are reflected in the differing amplitudes – from tropical to cool temperate – reflected in the
Insect Species?
47
Fig. 4.6 Caledia captiva range. The distribution of some major intraspecific categories in eastern Australia as assessed by karyotype differences (see text). The ‘Moreton’ form forms a cline from south east Victoria northward to its main centre on the central east coast, where it meets the northern ‘Torresian’ form; the position of the dynamic hybrid one is indicated by a dashed line (After Marchant and Shaw 1993)
distributions shown and, as climates change, together with areas of fruit cultivation, all such plots are also subject to substantial changes (p. 183). Wherever two closely related insect forms meet in distribution, and create even a narrow zone where they overlap, hybridisation may occur – as in the Tisiphone butterflies noted above – but the outcomes can be very difficult to interpret meaningfully. The grasshopper Caledia captiva is very widely distributed in grasslands in eastern and northern Australia, and was also amongst those insects first collected in 1770, although not described until considerably later. Although formally a single named species in Australia, it is in fact very complex, and illustrates some of the problems involved in approaching the question of ‘what is a species?’, and how such extensive variation may be explained properly. Found also in Papua New Guinea, the range of C. captiva spans about 35° of latitude (Fig. 4.6) and it is common throughout both tropical and temperate zones, apparently having gradually moved southward from the tropics. Over this distance, it can be subdivided into several distinct entities that differ substantially in DNA features, the organisation of the chromosomes, and extent of reproductive isolation, as assessed through comprehensive series of laboratory trials. Two main geographical groups have been referred to as ‘Torresian’ (occurring in Papua New Guinea, Northern Territory and coastal Queensland) and ‘Moreton’ (south east Queensland to Victoria) and these two meet to form a hybrid zone in south east Queensland. That zone is only about a kilometre wide, but more than 250 km long, and is of considerable evolutionary interest. The open forest to improved pasture region supports grasshoppers at densities of up to about 2,800 individuals per hectare, and chromosome characteristics have been
48
4 Origins, Distributions and Diversity
assessed in samples taken at small intervals (200 m) to attempt to characterise the purported hybrid zone. Only a kilometre separates the two distinctive types, with further abrupt transitions occurring within only 200 m, and with climate studies suggesting that the zone itself may move during periods of climate stress. Many questions remain, but this carefully documented example illustrates the extensive genetic and ecological changes that can occur over very short distances and, equally, not be revealed without equivalent fine-grain study, with – in this case – the two protagonist forms only partly isolated reproductively but with a further member of the complex (‘Daintree’, Fig. 4.6) fully isolated and does not produce hybrids. More recent studies suggest that this variation within C. captiva is really a suite of clines, and that it comprises only one real species. The chromosomes of the two main forms differ substantially, with ‘Moreton’ showing a consistently variable pattern along the latitudinal gradient: more than 600 different chromosome forms (transcriptions) have now been recognised in this grasshopper! Such variations are not necessarily worthy of formal recognition by naming, of course, but indicate areas in which debate over formal status may occur – and for numerous other species for which the picture is far more incomplete, scientists may well be tempted to elevate sporadic varieties of this nature to subspecific or specific level. As another complex example, the eucalypt-frequenting stick insect Didymuria violescens includes at least ten different distinct chromosome forms, each of them uniform within a particular part of the species’ range. These have been called ‘races’ and overlap only very narrowly in places. D. violescens is very variable in appearance and morphology, but variations in appearance are not correlated with patterns of chromosome differences, so any individual cannot be allocated to population/ race simply on what it looks like. Hybridisation can occur between parapatric races, so that all ‘races’ are referred to the same species as not being reproductively isolated. However, the varying levels of reproductive incompatability imply a process of incipient speciation, with the narrow hybrid zones constituting partial genetic barriers between dynamically interacting ‘potential species’. Both these examples, and others (Fig. 4.7) such as the so-called ‘tension zones’ designated for areas where putatively different races of some other species of grasshoppers come into contact, emphasise the difficulties of hard categorisation, and a conservative approach to enumerating species has been advocated repeatedly because such forms in the transition of becoming more distinct and reproductively isolated are open to very subjective opinion on what they really represent, and much of the ‘evidence’ cannot be resolved objectively. Different chromosomal forms of plant-feeding insects (as those most commonly studied) may be linked with host plant-specificity. The complex gall-forming genus of coccoid bugs, Apiomorpha (Eriococcidae), with nearly 40 species described, predomnantly infests Eucalyptus. The gall form is often very characteristic for a given species (Fig. 4.8), although some are more variable. A. munita has diploid chromosome counts ranging from 6 to >100 in different populations, with considerable variety within each of the three defined subspecies, so that considerable caution is needed in using genetic information to categorise taxa. All three subspecies of A. munita are found only on the series Symphyomyrtus, with no overlap of host plant species across them.
Insect Species?
49
Fig. 4.7 The common grass-blue, Zizina labradus (Lycaenidae), is one of the most abundant and widespread butterflies in Australia, occurring in a variety of open grassy habitats. However, its nomenclature is still not wholly resolved across its wider range in the western Pacific region. A closely related New Zealand form, Z. oxleyi, has been claimed to be threatened by hybridisation with the introduced or invasive Australian taxon. A recent account concluded that Z. labradus is best treated as a subspecies of the widespread Z. otis, rather than as a distinct species
‘Pattern’ (here, the genetic and/or morphological peculiarity of populations of an insect) and ‘process’ (how that variety has arisen and the extent to which it represents functional or reproductive isolation) are easy to confuse in studying insect evolution. Most practising entomologists have a realistic idea of what constitutes ‘a species’ in their particular group of interest, and of how to appraise variations. However, insect taxonomy has a history of two camps of people who describe insect species. They are referred to traditionally as ‘splitters’ (those who name new species on small, sometimes individual, character differences) and ‘lumpers’ (who tend to amalgamate variety of probable specific value into the same taxon, rather than proliferate formal names). Others, of course, tend to adopt a more balanced viewpoint than either of these, but without close independent study of any insect group, there is no infallible guideline to what any given name may really mean, however useful it may be as a descriptive epithet. But much insect variety is never acknowledged formally in that way, and the characters contributing to that variety, often marking consistent variants, will continue to proliferate as new approaches to comparisons arise. How we interpret this consistently, or even whether consistency is possible across different insect groups diversifying at different rates and in different ways, remains largely an open question. Genetic studies of various insects have sometimes confirmed the validity of subspecies as discrete, and in many cases led to more intensive study resulting in some being elevated to full species on a greater array of characters than initially examined.
50
4 Origins, Distributions and Diversity
Fig. 4.8 Australia has a variety of gall-inducing coccids (Hemiptera: Coccoidea). Each gall supports the development of a single individual and may persist for many months. Eriococcidae are a major family involved, with species of Apiomorpha a notable radiation on eucalypts, often with galls species-specific in appearance: female galls of two taxa are shown to illustrate this. Both are about 2.5 cm long, and male galls are much smaller: (a) the 3–4 flanged gall of A. munita; (b) the cylindrical gall of A. conica
In other cases the variation may be shown to be continuous, even if modified locally by environmental factors (such as the food plants of herbivores), or temperature, so that any separate name is spurious. So-called ‘cryptic species’, however, are undeniably numerous and reflect situations in which even experienced specialists may fail to recognise that a series of specimens confronting them includes more than one species.
Insect Species?
51
Consistent biological differences (such as restriction to particular food plants) may lead to biological differentiation without any obvious morphological change. Part of the debate over the size of Australia’s insect fauna thus devolves simply on how to define a species, and whether this can be achieved consistently and acceptably across the various orders. Species have been defined in various ways, but some clear consistent definition is obviously needed for comparisons and absolute evaluation of diversity present. This is not easy, not least because different groups of insects may be evaluated by specialists in different ways, so that differences between species are not always obvious – in evolutionary terms, they form continua which we are trying to partition and label into discrete components: to impose firm, but sometimes artificial, boundaries. Taxonomy involves ‘pigeon-holing’; but for variable entities the borders of the ‘holes’ blur, and different specialists may proffer different sized ‘pigeons’ for consideration. Suggestions that cryptic species may outnumber recognisable species severalfold provide considerable uncertainty, but some recent molecular studies of insects – using DNA extracted from tiny parts of a body, such as a single leg or small area of wing, as techniques continually gaining in efficiency and subtlety – indeed reveal massive unsuspected variety. ‘One’ South American butterfly, for example, was revealed to comprise ten recognisable species! Our challenge for the future is to interpret, enumerate, and document this variety. Bar-coding is at present receiving much attention as a means of helping to diagnose species or other ‘evolutionary units’ and, as in the above case, may reveal high levels of previously unsuspected diversity: one recently- described Australian geometrid moth, earlier hidden within the diagnosis of a previously-diagnosed species of Oenochroma, has actually been named ‘barcodificata’ as a homage to the technique! However, many scientists urge the combination of traditional and such more modern techniques in delimiting species, because of the ambiguities that arise from the latter used in isolation. For another geometrid moth, the consistent but small differences in male structures led to the Tasmanian form being named as distinct from the mainland species – but accompanying DNA data were ambivalent, even though implying this was the correct decision. But – despite advocacy to the contrary – bar-coding is viewed widely as not being an alternative to good traditional taxonomy based on structural characters: some workers seem to regard it simply (and, perhaps, cynically!) more as a ‘quick fix’ that can make any differences seem significant, and lead to subjective evaluations that are not then testable objectively. The status of a structurally-defined species, by contrast, can be re-evaluated as more material comes to hand, and the character states revised or augmented. Up to the present, a high proportion of named insect species have been described from appearance and structural features of dead specimens, often from very few individuals and no biological knowledge other than what can be inferred from incomplete knowledge of related forms. Whilst analytical techniques are improving rapidly, DNA analysis from long dead insects is still inconsistent. Often, the extent of individual variation within a species is simply unknown, and specialists may elect to designate ‘new species’ on small morphological features alone. Such classical taxonomy is thereby typological, rather than reflecting any boundaries between biological species, and based on the rationale that if the insect has a given suite of
52
4 Origins, Distributions and Diversity
structural features it is ‘species x’, and if it does not it is ‘species y’. This may be entirely valid, of course, but inevitably has led to a substantial number of insects each being named on repeated occasions and, as the butterflies noted above illustrate, realistically interpreting natural variation is fraught with difficulty. One species of thrips (not Australian, but equivalent examples are almost certain to be found in other orders) has received 18 different names in four different genera! Such situations can arise through differences between males and females, and between small and large individuals, between individuals from different hosts, natural variations in colour and form, seasonal differences, and so on. Ideally, taxonomic evaluation should be able to appraise this variation within populations and across the putative species’ range – but this has rarely been possible in the past, so that confusion over ‘real numbers’ of species has two main sources: that many real species remain undescribed and often undetected, and that many of the species’ names available as a basis for species counts represent synonyms, also undetected. For thrips, to follow the example introduced above, it has been estimated that around 1,500 of the 7,000 species names recognised worldwide in 2004 are synonyms, because variation was not recognised or material was insufficient for detailed comparative study. An insect taxonomist rarely knows that two partially similar dead individuals are (or are not) the same species. He or she uses knowledge and experience, often acquired over many years of study of the insect group involved, to predict that a specimen belongs to a particular species, but this is not the same as knowing that it could interact with other individuals as a biological species. A purported species is at one level simply a scientific hypothesis, whose validity may eventually be proved (or disproved) by collection of more specimens or more data. Whereas ‘counting species’ is a central theme in insect diversity, the recognition and designation of individual insect species is of far more than theoretical interest alone, because different entities can, and usually do, have different biology. As suggested above, we may in some cases be counting names, rather than species! If our insect has a name that can be applied consistently (implying that the insect is recognisable, commonly not the case in Australia) that name becomes a key to retrieving information on that species, should any have been published in the past. Conversely, a wrong name can easily generate false information which sometimes has far-reaching implications. Confusion between closely-similar pest species, for example, can have important and expensive consequences. The two major species of Helicoverpa (formerly Heliothis) moths are amongst the most severe agricultural pests in Australia, but have very different life cycles and diapause regimes, so that confusion may lead to inadequate control of their depredations and substantial crop losses. In short, even closely–related species can differ substantially in their biology and for good ecological understanding it is emphatically not sufficient to categorise insects from surveys to assess diversity simply to broad categories such as beetles, flies, moths or, in many cases, even just to families and genera. These broad groupings can each mask substantial biological variety. The extent of this generalisation is not always appreciated, and accepted uncritically, but consider parallels amongst animals that are more popular and better-known. The limits to individual species knowledge that would flow from designation of various vertebrates only as ‘snakes’,
Further Reading
53
‘possums’ or ‘parrots’ would be widely obvious to naturalists – simply because each of these includes massive variety of detail – yet many of the people who would ridicule reliance on such broad categorisations in the vertebrate literature readily accept terms such as ‘beetles’ or ‘wasps’, perhaps taken to the family levels as (for example) Carabidae or Ichneumonidae as impressive names for families which each include as many species as the whole vertebrate fauna, and manifest equivalent variety of biological detail. At the other extreme, of course, designating individual zebra as separate species simply on small individual differences in stripe pattern would be equally misleading in assessing species diversity! In the meantime, the somewhat uneven information available allows some comments on Australia’s insect diversity as we currently understand it. Points over which there is little debate include (1) that Australia has an enormous number of insect species; (2) most of these are endemic and many have evolved in Australia, collectively from a considerable variety of founder elements – some of them ancient; (3) we are far from being able to provide an accurate estimate of numbers or inventory of variety, and most species are still unnamed; (4) that there are too few specialists to remedy this situation in the foreseeable future; and (5) insects are amongst the most ecologically and economically important components of Australia’s biodiversity, and many lineages are increasingly at risk of loss from human activities. Of the orders present, and reflecting a wider global picture, the ‘big four’ are Coleoptera, Diptera, Hymenoptera and Lepidoptera, all with a complete metamorphosis. Beetles are probably the richest of these, with perhaps a quarter of the possibly 80,000 species described. Several other orders – the plant-feeding sucking bugs (Hemiptera) and the grasshoppers and their allies (Orthoptera) each contain well over a thousand species and are the largest exopterygote orders whilst, at the other extreme, Archaeognatha and Zygentoma (the remnant Apterygota) are sparse and amongst Pterygota – not counting the Zoraptera, as noted earlier- only Megaloptera and Mecoptera have fewer than about 30 species. Spread across about 27 orders, this wealth of species is generated and supported by the variety of environments suitable for insects in Australia, and the resources they furnish. Understanding their ‘working environments’ is central to understanding the insects themselves. Linked intricately with this, knowledge of feeding habits and interactions with other species are key aspects of assessing how insects fit in to wider communities and how their individual populations vary in time and space. Behaviour is a major determinant of capability and adaptive potential in any such context.
Further Reading Attiwill P, Wilson B (2006) Ecology in Australia. In: Attiwill P, Wilson B (eds.) Ecology; an Australian Perspective. Oxford University Press, Melbourne, pp 3–14 Barker S (2006) Castiarina: Australia’s richest jewel beetle genus. Australian Biological Resources Study, Canberra (survey and taxonomic review of the major component of Buprestidae in Australia) Braby MF (2000) Butterflies of Australia, CSIRO Publishing, Collingwood (the standard work on these insects, with much more detail on the examples noted in this chapter, and many others)
54
4 Origins, Distributions and Diversity
Cook LG (2001) Extensive chromosomal variation associated with taxon divergence and host specificity in the gall-inducing scale insect Apiomorpha munita (Schrader) (Hemiptera: Sternorrhyncha: Coccoidea: Eriococcidae). Biol J Linn Soc 72:265–278 Drew RAI (1989) The tropical fruit flies (Diptera: Tephritidae: Dacinae) of the Australasian and Oceanian regions. Memoirs of the Queensland Museum 26 (Detailed taxonomic review of 209 species, as an example of richness within a native pest insect group) Drew RAI, Hooper GS, Bateman MA (1982) Economic fruit Flies of the South Pacific Region, 2nd edn. Queensland Department of Primary Industries, Brisbane Marchant AD, Shaw DD (1993) Contrasting patterns of geographic variation shown by mt DNA and karyotype organisation in two subspecies of Caledia captiva (Orthoptera). Mol Biol Evol 10:855–872 Ogata K, Taylor RW (1991) Ants of the genus Myrmecia Fabricius: a preliminary review and key to the named species (Hymenoptera: Formicidae; Myrmeciinae). J Nat Hist 25:1623–1673 (Technical account of bullant taxonomy, indicating complexities of species delimitation and recognition) Shaw DD, Marchant AD, Conteras N, Arnold ML, Groeters F, Kohlman BC (1993) Genomic and environmental determinants of a narrow hybrid zone: cause or coincidence. In: Harrison RG (ed.) Hybrid zones and the evolutionary process. Oxford University Press, New York, pp 165–195 (Overview of the detailed studies on Caledia captiva) Yeates DK, Harvey MS, Austin AD (2003) New estimates of terrestrial arthropod species-richness in Australia. Records of the South Australian Museum, Monograph series 7:231–241
Chapter 5
Environments and Habitat for Insects in Australia
Introduction: Places to Live ‘Places to live’ can be considered at various scales, from broad biogeographical regions extending over thousands of kilometres, to part of a single small plant or individual animal host. Those scales form a hierarchy, so that an environment for any given insect can be considered as a series spanning these extremes, with each subordinate level representing increased specificity and ecological specialisation. One leading European ecologist recently used the apposite simile of the Russian matrioschka dolls to illustrate this, with each successive doll inside a larger one representing finer detail of need but still depending on the protection of the enveloping covers. A moth caterpillar that can feed as a generalist herbivore on a variety of different plant species, perhaps across several different plant families could be considered to need a ‘larger minimum doll size’ than (for example) a caterpillar that can feed only on the young foliage of a single plant species that is itself restricted in distribution to a small part of a region. In general, many insects thrive on rather limited resources – one advantage of being small is that each individual does not need much food. The lifetime consumption of either of our caterpillars, above, may be only a few grams – very substantially less than that for kangaroos or cattle, which also eat for much longer than the few weeks or months many caterpillars need to develop. Such limited needs foster specialisation and fine-grain division of the environment – simply, dividing up a larger resource in some way means that more species can share it. Being small can be functionally advantageous in using key resources. The major environments, as the largest units defined and influencing biota, are each complex and very varied. Much of Australia’s distinctive character reflects its vegetation, recognised as diverse and unique from 1770, as acknowledged initially by the naming of Botany Bay, and with many of the plants differing markedly from the flora of any other region. That vegetation can be classified or categorised in various ways to display its variety, and this is done usually by a combination of structure and floristic composition – the particular plants present there. Each such
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_5, © Springer Science+Business Media B.V. 2011
55
56
5 Environments and Habitat for Insects in Australia
Fig. 5.1 Simplified distribution of major vegetation types in Australia (cf. Fig. 5.2). Key: black, closed forest; dense dotting, open forest; coarse dotting, woodland; unmarked, shrubland; vertical hatching, scrub and heath; horizontal hatching, herbland (After Carnahan 1977)
vegetation type – however it is delimited or defined – may also support specialised insects restricted to (or predominant in) it, and a high proportion of native insects are associated in some way with plants, many of them to individual plant species or genera. Evolution together (so, ‘coevolution’) over tens of millions of years has produced intricate associations and interdependent relationships between plants and insects in Australia. Terrestrial environments have traditionally been characterised by vegetation type, with the two axes of ‘life form and height of tallest plant layer’ and ‘percentage foliage cover of tallest plant layer’ most commonly employed to designate a series with the extremes of ‘tall closed forest’ (trees >30 m tall; dense cover of 70–100%) to ‘low open shrubland’ (shrubs 0–2 m tall; very sparse cover, <10%). A broad superimposition of major vegetation systems (Fig. 5.1) or more recently devised ‘ecoregions’ (Table 5.1, Fig. 5.2) on the biogeographical divisions (Fig. 4.2) helps initial recognition of some of these. Thus, the simplest descriptor for ‘forests’ is simply having trees as the tallest layer of vegetation, with a series of categories of height and foliage cover delimiting some major categories of structure – such as ‘closed forests’ with much canopy foliage cover, and ‘open forests’ with much sky exposed – and the tree species present differing between categories such as species-rich rainforest, and the eucalypt-dominated more open forests, which intergrade to even more open woodlands. Some forests are characterised by predominance of single tree species as a major descriptor – Nothofagus cunninghamii (myrtle beech) forest (Fig. 5.3) in the southern temperate rainforests, for example, itself characterises ancient Gondwanan associations.
Introduction: Places to Live
57
Table 5.1 The eight major ecoregion systems found in Australia. Globally 14 ecoregions occur, and are differentiated by containing geographically distinct groups of animals and plants that have evolved in relative isolation. The system is thereby more embracing than classification systems based on vegetation type alone, as has been the most frequent approach in the past (Information from Department of Environment, Water, Heritage and the Arts 2011) Ecoregion Comment Tropical and subtropical Large, discontinuous patches in north and north east; low moist broadleaf forests variability in temperature and high levels of rainfall. High floral and accompanying animal diversity Temperate broadleaf Extensive occurrence along the east coast, with wider variability and mixed forests in temperature and rainfall than above. Eucalyptus and Acacia typically present, unlike above. Ancient Gondwanan elements in Tasmania Characterised by rainfall within range of 90–150 cm/year. Tropical and subtropical Savanna systems, in particular restricted in extent. Patches of grassland savannas dry rainforest are rich in species, and grasslands are extensive and shrublands Temperate grasslands, Cooler than above and with wider temperature range; generally savannas and shrublands devoid of trees – except close to watercourses – and separate temperate forests and the arid interior of the continent; much of region now heavily disturbed, leaving only small remnants of natural vegetation High elevation (montane, alpine) environments, restricted in Montane grasslands and shrublands Australia to elevations above 1,300 m in the south east; small in extent and regarded widely as very vulnerable Characterised by hot, dry summers and cool, moist winters. Mediterranean forests, woodlands and shrubs Globally rare, but the south west corner of Australia is a globally leading example with rich and highly endemic flora and associated fauna Deserts and xeric shrublands Temperature extremes characteristic; rich array of habitats with high levels of local endemism amongst species Tundra Treeless polar desert, occurring in Australia only on some sub-Antarctic islands
This basis can, of course, be subdivided much more finely, so that in Victoria alone, a plan instigated a few years ago recognised around 300 ‘Ecological Vegetation Classes’ that could be bulked into about 20 groups and their relative incidence is used to delimit 28 ‘bioregions’ in the state, combining considerations of vegetation structure, species composition and influencing climate. Each of these is still very varied, but defined in somewhat more limiting terms than the State alone, and progressively perceptive or confusing subdivisions lead, almost inevitably, to any particular small patch of vegetation or small waterbody being deemed unique on fine details of community composition of resident plant or animal species. This extremism – for example the number of Ecological Vegetation Classes nominated in Victoria is currently around 1,000 and rising, with need for considerable botanical ability to tell many of them apart – is rather pointless in practice, as complicating rather than simplifying clear interpretations. However, it still helps to emphasise the massive variety of a ‘fine grain world’ available to insects, each of which must live in a local community and interact with its neighbours in various
58
5 Environments and Habitat for Insects in Australia
Fig. 5.2 Distribution of major ‘ecoregions’ in Australia. Broad separation of the seven ecoregions indicates some overlap with biogeographic provinces (cf. Fig. 4.2, p. 41), and helps to indicate some major biomes for insect life somewhat more broadly than from vegetation alone (cf. Fig. 5.1). Key: unmarked, deserts and xeric shrublands; vertical hatching, Mediterranean forests , woodlands and scrub; dense dotting, temperate broadleaf and mixed forest; horizontal hatching, temperate grasslands, savannas and shrublands; coarse dotting, tropical and subtropical grasslands, savannas and shrublands; black, tropical and subtropical moist broadleaf forests; arrow, montane grasslands and shrublands in south east (see text, after Department of Environment, Water, Heritage and the Arts 2011)
ways – whether on land or in water – with that spectrum of co-occurring species influenced and determined by other factors. Vegetation may also be a strong influence in classifying aquatic communities, but many other factors – still or running water, substrate form and texture, water temperature and chemistry, and others can also determine habitat suitability. Indeed, the variety of aquatic habitats for insects in Australia parallels that for terrestrial environments, with schemes of subdivision based on permanent versus temporary water bodies, and still or flowing water, each with numerous intermediate conditions (p. 140). Whatever combination of features is present in a given place, the versatility of insects as a group almost always guarantees that some will exploit it, and that the array of resident insect assemblages will progressively change in response to vegetational succession, water flow or chemistry, or other site changes – wherever that site may be – as well as differences that occur along flowing waters, so that insects in river headwaters may be far different from those living in its lower reaches. In essence these variations constitute a complex mosaic of insect habitats which intergrade in very subtle ways along environmental gradients, so that any ‘hard boundaries’ are difficult to define or even detect and the features influencing the dynamics of any more obvious ‘edges’ not always understood. Small differences in elevation, latitude, aridity, salinity, soil features – as
Introduction: Places to Live
59
Fig. 5.3 Nothofagus fringing a subalpine woodland edge in Victoria
well as vegetation – may restrict occupancy by a particular insect species and be determinants of its range. This manifests, for examples, by presence of clear groups of alpine insects confined to higher elevations in the southeast, distinctive desert faunas across the arid to semiarid regions, and a number of species restricted to saltmarshes and similar habitats transitional between water and land. Broad descriptive terms (such as the above alpine, desert, saltmarsh, pond, river) for habitats may convey much of the information needed to initiate understanding the insects that live there, by suggesting limits to their needs. Desert insects, for example, are clearly suited to living in areas with low and unpredictable rainfall, high summer temperatures, low nutrition soils and little or no tree cover; alpine insects must cope with colder winters, snow cover, a limited array of taller vegetation and a short summer season – and so on. On a broad initial characterisation of Australian terrestrial habitats along these lines, most ecologists recognise the very broad categories noted earlier as differing in gross vegetation composition and emphasising differences between ‘more open’ and ‘more treed’ environments. Two aspects of vegetation composition must be distinguished carefully in influencing insect diversity in somewhat different ways. A grassland may comprise few or many plant species, each of which may foster particular insects, but all are low-growing plants. A forest may provide similar plant species variety, but has markedly different structure, with several plant growth forms as trees, understorey shrubs, low herbs and grasses: it has a more complex ‘architecture’, with more three-dimensional complexity for insects to exploit. A single tree, for example, may support different species of insects chewing foliage, sucking sap, tunneling into bark, feeding on seeds,
60
5 Environments and Habitat for Insects in Australia
flowers, roots and so on, as opportunities that are simply not available on a small grass stem. Even closely related species of plants growing close together commonly support different insect species, and the numerous interactions between insects and plants are a key theme in insect natural history. In turn the insects feeding on these plant-frequenters may also show marked site and food-species specificity as an outcome of habitat features, so that the insect community of any given habitat may have characteristic constituents at several trophic levels. Increased plant diversity, both in species richness and physical structure, provides increased opportunities for insects of many kinds beyond those obviously feeding on the plants. The natural environments emphasised above are only part of the scope of insect habitats, albeit the major one. Human constructs, introduced animals and plants, and the numerous changes imposed on natural biotopes by felling forests, clearing grasslands for agriculture or urban development, and many other changes create ranges of new opportunity that facilitate wellbeing of many insects from overseas which could not otherwise thrive in Australia. At the same time, such changes can be highly detrimental to some native species that cannot cope with the disturbance or change. Some do so, and the collective variety and ecological ingenuity of the insects ensures that no exploitable place or resource is likely to remain insect-free for long. Hence, our homes and storage barns are associated with an array of ‘domestic’ insects, and most of our crops and other plantings suffer some level of insect feeding by species we regard as pests. Those of introduced plants may be attacked by native insect herbivores that can ‘switch’ to these, previously unavailable, food supplies. For example, a number of native Lepidoptera have been able to exploit Monterey pine, Pinus radiata, to the extent that some are sporadically important pests of plantation trees. These resources provide opportunities and conditions not generally available in many more natural environments. Living inside buildings provides some buffer against climatic extremes, and this may even be imposed for uniform temperatures or humidities to protect stored goods or provide equable living conditions for people. This constancy may enable insects to breed throughout the year, rather than be highly seasonal. Outside, a ‘crop’ often provides a large ‘target’ for insects – typically it may be many hectares of conspicuous uniform monoculture of highly nutritious food, guaranteed as such by plant variety selection, fertiliser applications and irrigation, and which in many cases also lacks any major background natural insect community that may oppose insects as they arrive. The above emphasises the most conventional meaning of habitat as ‘a place to live’, and this is by far the most common sense in which the term is applied by insect ecologists. In this sense, it often equates to a ‘site’, sometimes characterised by details of vegetation, climate or topography (singly or in combination) to define it further. But for many insects the determinants of their occurrence extend well beyond these attributes of place alone, to the main issues of incidence, accessibility and quality of particular, critical, resources – those that an insect needs in order to persist and thrive. From this perspective, an insect’s habitat may be defined better in these more specific terms to indicate suitability, rather than as a broad vegetation class or biotope. Thus, a ‘place’ may appear broadly suitable for a given insect, but without the correct combination and balance of critical resources may not sustain it.
Resources for Insects
61
Resources for Insects Critical resources must cover the needs of all stages of the insect (such as the disparate needs of a caterpillar and a butterfly), and encompass two main groups of factors, clarified recently in detail for the British butterflies but of universal relevance. The first is ‘consumables’, the often very specific food requirements needed by each feeding stage in order to grow, mature and reproduce successfully, with due needs for amounts, quality, and continuity of supply, as well as density of resources and their distribution in the landscape. The second category, ‘utilities’, is more complex because it brings in wider life style aspects of the insect’s biology. It comprises the non-consumable resources needed for the insect to live and behave normally throughout its residency: and a complete picture is very difficult to obtain, simply because it is extremely easy to overlook factors that may indeed be critical. Topography, for example, may include extent and direction of slope as influencing how much sunlight reaches the insect or its immediate surrounds and affects the temperature and rate of activity or development; bare ground may be needed for display or hunting; soil texture may influence capability to burrow or dig tunnels in which to pupate; cleared areas may be needed for courtship flight or territorial display – and so on. The density, dispersion and complementarity of utilities may differ greatly within a landscape or region. A full compendium of possible utilities is probably not possible for any but the best-studied insect species, but can include many such factors, as exemplified in Table 5.2. These resources, including the microclimate regimes in which they occur (a category that some entomologists prefer to separate from utilities as ‘conditioners’, influencing accessibility and availability of all other resources) are major influences on insect incidence, distribution and abundance in any place, and discovering how insects exploit them in space and time brings out more detail of some themes introduced earlier. An insect species may be present in any place as a permanent resident, or it may need to reach it, colonise and breed there at intervals – perhaps every year – and eventually leave. The critical resources present may be abundant and widely distributed, or scarce and highly scattered, and variously present for the whole time or during only particular seasons so that availability of food may be a strong seasonal determinant of when an insect can feed and develop. Very simplistically,
Table 5.2 Part of the variety of ‘utilities’ that may be critical resources for an insect, in part as summarised for butterflies in Britain (see Dennis 2010) Adult basking sites: for thermoregulation, such as on bare ground or shrubs/trees Mate location sites: territorial behaviour, and flyways for patrolling Egg-laying sites Adult roosts, or shelter from wind or rain Larval sites for resting and moulting Pupation sites Refuges from predators and parasitoids Hibernation and aestivation sites
62
5 Environments and Habitat for Insects in Australia
insects that can feed on many kinds of plant or other broad food array present throughout the year in large amounts may be faced with problems far different from those restricted to a single food species that is rare, distributed sporadically and available for only a few weeks or months each year. In addition, habitats change in both short term cycles (such as seasonally) and directionally over longer periods, such as by succession, so that the attributes of place that render it suitable for an insect at present or at particular times of the year may change. With seasonal changes, a particular short lived food plant or animal may ‘predictably’ reappear at around the same time next year, so that evolutionary pressures are for the life cycle to be adjusted to exploit that resource effectively. With successional change, that food may be eliminated or displaced in space and become unavailable – extent and rate of changes are both important. In general, persistent resources (forest trees may live for up to several centuries if left undisturbed) provide enveloping environments that are also ‘stable’ over long periods, and some may not develop until the system itself is mature – development of tree holes in mature trees as nesting cavities for marsupials and birds is a frequently cited example, but is paralleled by the needs of vast numbers of insects that feed on wood at different stages of its decay. Early successional stages of vegetation, by contrast, may pass rapidly and be replaced by others, so that the environment is much more ‘unstable’ and unsuitable for any long residency. At extremes, an insect species or population may be supported for up to several hundred generations in a long-lived forest or a permanent river or lake, or for no more than a single generation in many more transient habitats. The latter span both largely predictable seasonal incidences (examples include annual plants; particular plant stages such as flush growth, seeds or fruits; rain pools or streams) and aseasonal habitats (such as animal dung; carrion; casual rainfall pools; dead fallen branches). In the latter, the pool of potential colonists may change at different times of the year – but any such unit may itself support an individual succession of insect colonists, with the earliest arrivals replaced by other species. One well-studied example is the succession of fly species that invade human corpses, used widely in forensic investigations because the particular species present may help to determine time since death: a dead body attractive to one fly species may not appeal to another at the same time. Australian studies, identifying the flies that successively invade pig or other carcasses as they decay, have revealed significant seasonal and weatherinfluenced changes in patterns of species invasions on cadavres, as ‘microseral stages’, and reflecting the local species pool available to participate. This itself can differ between, for example, forest and pasture environments, but many such insects depending on shortlasting habitats are extraordinarily adept at finding them. Another example is the succession of insects colonising dead wood – but here the resource may last for considerably longer: up to a few decades, rather than just a few weeks or months for carrion or dung. The major generalised correlation is that high migration levels within a species are associated with dependence on such ‘temporary habitats’, whereas migration levels are characteristically low for insects in ‘permanent’ habitats. But this dichotomy (which is clearly in reality a continuum: the extremes simply illustrate the principle) has further important ramifications for the insect’s way of life. The environment of a
Resources for Insects
63
‘permanent’ habitat is likely to include many other insect species, each also needing to persist but collectively rendering the environment complex and possibly hostile, particularly to any newcomer, so any member of the local fauna may have to adjust to the presence of these possible competitors or antagonists – perhaps hiding at times, maintaining vigilance, or defending itself energetically. The environment of a ‘temporary habitat’ may be much less crowded, with few species present at any time – particularly in the earliest stages of exploitation. Under this idealised scenario, a colonising insect may be able to invest its effort into feeding and reproduction, rather than contending with other species. Conditions for this may pass quickly, so that many such species are characterised by high reproductive rates, as both fecundity and developmental rate, before the environment becomes untenable. At that stage, the level of feeding specificity may become even more important. Some polyphagous aphids, for example, are able to migrate to track a succession of shortlived resources (in this case annual plants) over a few months of summer as each becomes available. Very rapid breeding allows population buildup on each, and dispersal of offspring of successive generations occurs. Many individuals may not find a new habitat, and perish, and high fecundity is in part a counter to the risks involved in migration. A feeding specialist, in contrast, may lack this ability and not feed on any other species – so that its seasonality is very restricted, and its life cycle constrained accordingly. As noted above, persistence in a habitat and reliance on movement to another demand rather different strategies. Each may involve considerable specialisation of behaviour and life style but, again, the two extremes intergrade in many ways. Defining the habitat needs of an insect generally starts with finding the insect in a particular place and attempting to discover what it is doing there, and if it really lives there. Its presence indicates possible residency, but does not assure it. Many insects found in almost any biome are ‘tourists’ in that they are there by accident or opportunism, rather than having any regular association with the place or resources present but, nevertheless, may participate – at least temporarily – in the processes within the biome, such as by eating or being eaten. Other species may be ‘temporary residents’, not breeding but using utilities resources in some way. Female dragonflies of many species, for example, fly far from their freshwater natal sites and undergo a period of maturation before returning to water bodies to mate and reproduce. In short, insects found in any place are not always resident there or adapted to the conditions present; often we have no initial idea of this, whence they have come, and how far they may have travelled. Careful sampling and study may be needed to clarify this in any attempt to enumerate insects depending on any habitat Although far less tangible than focusing on the biology of individual species, studies of ‘insect diversity’ provide important background on habitat relationships by giving perspective on themes such as how many insect species occur, what they are, their relative abundance, and how they interact with each other and with the array of resources in the biome. Collectively, these indicate ‘importance’ of the habitat and the complexity of the community present, and comparative studies across broadly similar areas may reveal some ‘rankings’ of quality or suitability, and how particular features influence insect diversity. Many approaches to sampling insects have been
64
5 Environments and Habitat for Insects in Australia
Table 5.3 Some published sources of information on collecting, sampling and preserving insects Endersby I (2009) Collecting and sampling insects. Entomological Society of Victoria, Melbourne (short, introductory overview of range of approaches and methods) New TR (1998) Invertebrate surveys for conservation. Oxford University Press, Oxford (broad survey of techniques and approaches) Samways MJ, McGeoch MA, New TR (2010) Insect conservation. A handbook of approaches and methods. Oxford University Press, Oxford (wide discussion and survey) Smithers C (1981) Handbook of insect collecting. AH & AW Reed, Sydney (Australian focus, well-illustrated survey) Southwood TRE, Henderson PA (2000) Ecological methods, 3rd edn. Blackwell, Oxford (most recent edition of the classic work published first in 1966) Upton MS, Mantle BL (2010) Methods for collecting, preserving and studying insects and other terrestrial arthropods, 5th edn. Australian Entomological Society, Canberra (earlier editions also remain very useful)
devised – ranging from simple visual inspection of conspicuous forms to the mass trapping and killing of insects for preservation and later examination. The latter approach, ethically unpalatable though it may be to kill hundreds to thousands of specimens in ecological investigations, is largely inevitable if we need to gain quantitative and taxonomically accurate information on most groups of insects. Sources of information on some of the methods commonly used in insect sampling and collecting are listed in Table 5.3 but, broadly, few insect groups are amenable to species-level identifications without close examination and, in many cases, study by specialists in the group involved, and it is important that they are preserved in sufficiently good condition for such study. The two main exceptions are butterflies and dragonflies (both sometimes referred to as ‘birdwatcher’s bugs’ because many species can be identified through binoculars in conjunction with well-illustrated handbooks) but even these include taxa for which separation depends on fine structural characters and microscopical appraisal. Needs to document insect faunas from trapped samples and to conserve scarce or threatened species present by not capturing them are not always easy to harmonise (Chap. 16). Any reasonably complete inventory (count of the species present) of insects must be based on samples taken at intervals over at least a year, to cover seasonal variations. Many butterflies, for example, may be present as identifiable or conspicuous adults for only a few weeks each year, with their developmental patterns leading to a characteristic flight season. The series of ‘brown’ butterflies (Nymphalidae, Satyrinae) in temperate south eastern Australia exemplify such differences well, with the characteristic flight period for each summarised in Fig. 5.4; similar sequences occur in other insects as one dimension in which they may share the environment across time – and many of the characteristic seasons are well known to collectors or ecologists who study that group. For widely-distributed insects, the season can differ in different places, reflecting temperature regimes – in some, even the number of generations each year may be influenced. Further subtleties can involve different appearances of the two sexes. The most widely distributed of the above Satyrinae, for example (Heteronympha merope, the common brown) is also one of the longest-lived and often appears to be
Resources for Insects
65
Fig. 5.4 Patterns of seasonal appearance. The sequence of flight seasons of the south eastern species of three genera of satyrine butterflies, the ‘browns’. Months from September to May (shown by initial letter) span the austral spring and summer; characteristic main flight season of each species shown by horizontal line: both time and duration differ amongst species
protandrous – the males emerge before the females in early summer. Once mated, females delay reproduction and shelter for several months (so are not commonly seen) before they lay eggs in late summer, at which time some males are still active and seeking mates. Eggs and young caterpillars are thereby not exposed to the highest summer temperatures, and eggs hatch at a time when young fresh grass growth may be available as food. A general recommendation for surveying insects initially at any site is to sample at 4–6 intervals over at least one year, preferably also for a second year to counter any vagaries caused by unusual weather. However, a complete inventory of the insects of any reasonably complex natural environment is almost impossible, and additional species can continue to accumulate in samples over many years. Nevertheless, most of the more characteristic and ecologically significant denizens will usually be revealed within the first year or so, and associated by their relatively large numbers and consistent presence across samples from different sites within a habitat. Many of the insects thought of as ‘rare’ are ecological specialists, some depending on resources that are always in short supply so that populations of the insects can never become large. In collecting samples of insects in any habitat for study, by whatever methods, the almost invariable result is that very few species are found in very large numbers, and a large proportion of the species are represented by only one or two individuals (Fig. 5.5). Enlarging the sample (by more sample units or greater period of sampling) simply endorses this picture, typically furnishing yet more individuals of the abundant
66
5 Environments and Habitat for Insects in Australia
Fig. 5.5 Illustration of the typical relative abundance of species in communities: few species are abundant, and numerous species are represented by few individuals; in many samples of insects, a high proportion of species are found only as one or few individuals
species and increasing the number of low abundance ones. They may be genuinely scarce, or may be elusive or inconspicuous (essentially, difficult to find or sample), either through their short periods of appearance or through avoiding other organisms in complex environments. The inevitable impression is that it is extraordinarily difficult to count all the insect species present on even a small local site, let alone on a large and varied region encompassing a variety of habitats. ‘Rarity’ has connotations of low abundance, ecological specialisation and narrow distributions and – either in reality or reflecting our ignorance – substantial numbers of species in Australia fit one or more of these criteria. Many of those considered common show the opposing features, although what we might consider ecological generalisation almost always reflects long periods of adaptive selection: although rarely considered as such, polyphagy (feeding on many kinds of food) is itself highly adaptive. Its contrast, monophagy (feeding on only one food kind or species), is indeed high specialisation, but has evolutionary dangers as well. Should an insect (or other consumer) come to depend completely on a single food plant, for example, should anything happen to the eaten species, the consumer is doomed also. Feeding habits are the next major theme we need to consider, as a major influence on insect diversification and diversity.
Further Reading Attiwill P, Wilson B (eds.) (2006) Ecology. An Australian perspective, 2nd edn. Oxford University Press, Melbourne (much background to the variety of biotopes and environments within Australia, and now a standard text) Carnahan JA (1977) Vegetation. In: Jeans DN (ed.) Australia: a geography. Sydney University Press, Sydney, pp 175–195 Dennis RLH (2010) A resource-based habitat view for conservation. Butterflies in the British landscape. Wiley-Blackwell, Oxford
Further Reading
67
Department of Environment, Water, Heritage and the Arts, Canberra: www.environment.gov.au/ parks/nrs/science/bioregion-framework/terrestrial-habitats.html. Accessed Jan 2011 Hughes RD (1974) Living insects. Collins, Sydney (an informal introduction to biology of insects in Australia and their major habitats, with much classic work included) Matthews EG, Kitching RL (1984) Insect ecology, 2nd edn. University of Queensland Press, Brisbane (a more formal introduction to insect ecology in Australia)
Chapter 6
Foods and Feeding Biology
Introduction: The Variety of Food and Feeding Habits Almost any organic material is food for one or more kinds of insect, so that the collective roles of insects in ecological processes involving material and energy pathways in any biotope are enormous. Few organic materials escape insect attention entirely – even the individual droppings of koalas and some possums can support development of caterpillars feeding inside them, whilst other insects complete development inside a single leaf or seed, or other living or dead insect individual, other animal or plant. An individual insect does not eat much, but amplifying this to the thousands, even millions, of individuals in some insect populations dramatically changes perceptions of their impacts. Locusts, for example, can strip large areas of crops bare, and caterpillars on plantation trees can also cause complete defoliation. Both are viewed as major pests in such contexts of causing massive economic losses. The rain of faeces from caterpillars feeding in a forest canopy is a major nutrient input to the soils beneath. Measuring the amounts of faeces dropped by caterpillars of some forestry pest moths in North America has been a standard monitoring approach for pest management, to help assess when suppression measures should be instigated. Foods and questions of food supply are a central theme in insect ecology and diversification. The quantity, quality and accessibility of foods, and how they are partitioned amongst consumer species affects activity, and patterns of diversity and coexistence, of all insects. Insect herbivores are our major competitors for vegetable resources, and understanding their depredations on crops is a fertile arena for increasing our awareness of insect behaviour and biology. Carnivores eating other insects as predators are amongst the most potent agents for keeping such herbivores ‘in balance’ in natural communities, and some are manipulated by people as biological control agents to suppress pests. Both the breadth and versatility of insect feeding habits and foods contribute to their ecological roles, and much of the behavioural repertoire of insects has developed to facilitate finding, selecting, and exploiting their foods, defending their food against other consumers, and countering the defences their food (whether animal or plant) may raise that prevent it being eaten.
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_6, © Springer Science+Business Media B.V. 2011
69
70
6 Foods and Feeding Biology
Exploitation or Partnerships? However, many associations between the eater and the eaten species are regular ones, with the two persisting together over long periods, during which they may in some ways become ‘co-adapted’. Many specific feeding relationships between insects and other living species are in some ways ‘partnerships’, through which both species may benefit – many, indeed, are mutualisms in which the relationship may be both specific and necessary. The numerous flowering plants that depend on particular insect vectors for cross-pollination are one clear example, which – as in many similar feeding relationships – ranges from very general (a variety of different insect vectors feeding on nectar and/or pollen) to highly specific (many Australian orchids are thought to be pollinated by single species of wasps) with very specialised adaptations to each other – for example, in some orchids the flower has evolved to ‘mimic’ the wasp in both appearance and scent, so that a male wasp is deluded into attempting a copulation with the flower. Some such orchids have been described as ‘prostitutes’, in providing strong sexual stimuli that induce male wasps to mate with them. The richly diverse pollination strategies of orchids have intrigued naturalists for centuries – with Charles Darwin writing a notable book on this theme in 1862. Insect–plant relationships, as we see later, include many intricate examples of such ‘co-evolution’, the long history of mutual responses that has led to highly specific and finely-honed intimacies between different species. Perhaps none is more intriguing than the strong mutualisms formed between figs and fig wasps, as amongst the best-documented cases of specialised interactions between species. Figs (Ficus, with about 750 species worldwide and around 500 of these in the Australian region) and members of a family of chalcidoid wasps (Agaonidae) as specific pollinators often have one-to-one mutualisms, with all 20 genera of fig wasps having this role. Each fig species has, until recently, been believed to depend on its own wasp species, with the wasp specific to that fig but, whilst fig-pollinating by wasps seems to have evolved only once, much subsequent coevolution is indeed more complex than this simple scenario suggests. As with some other insect groups, recent molecular studies have helped to display previously unrecognised fig wasp variety. A major genus of fig wasps in Australia is Pleistodontes, with some of its 17 species described only recently. However, studies on only one of these (P. imperialis) revealed several substantially different groups on DNA differences, so that ‘it’ might comprise several distinct cryptic species. Many possible parallels may occur. The life history of fig wasps involves mutual dependence with the fig. Particular Ficus species may be either monoecious (male and female florets in the same inflorescence – the closed syconium referred to as a fig) or dioecious (separate male and female trees). An individual fig tree can produce enormous numbers of figs, some species doing so twice a year. As one example, P. froggatti occurs with the monoecious Moreton Bay fig, Ficus macrophylla. Winged female wasps enter the fig through a narrow open tunnel (the ostiole) (Fig. 6.1) which may then contract to trap the wasps inside. Wasps attempt to lay eggs in the developing florets, pollinating the
Searching for Food
71
Fig. 6.1 Life history of fig wasp. Pleistodontes froggatti on Moreton Bay fig, Ficus macrophylla (see text)
female florets in the process, and allowing male florets to become ‘gall flowers’ with wasp larvae developing within them. Others florets, pollinated, develop to seeds. Male wasps are wingless, and do not leave the syconium in which they develop. They mate with females in the same syconium, sometimes whilst they are still inside their galls. Fertilised females then emerge, acquire pollen by contact as they leave the parent syconium, and fly to seek other developing figs – in some instances aided by chemical attractants, and the cycle is completed as they reach them. In F. macrophylla, all figs on any one tree tend to have similar seasonality, but each tree can be different, so that their collective maturation occurs over an extended period and wasps are continually cross-pollinating between different trees. Many other Ficus species are dioecious, so that different trees produce only wasps (male trees) or only seeds (female trees, with the female wasps dying after pollination and attempted, futile, reproduction).
Searching for Food Adult insects may or may not need food. If they are very short-lived (such as mayflies and some moths) they can rely wholly on food reserves carried over from larval feeding to sustain them. Most longer-lived insects need to feed to gain an energy supply and nutrients to mature reproductive products, perhaps continually over a longer period. Some entomologists draw a distinction between these categories as ‘capital reproducers’ (by analogy with economic terms: they draw on the capital of larval food reserves acquired earlier in life, and deplete this until it has gone) and ‘interest reproducers’ (using acquired benefit, food, to provide for their needs over a longer period). Many may indeed reproduce soon after reaching adulthood, but others may disperse or undergo a period of diapause before becoming reproductively mature. Insects with complete metamorphosis have two rather different tasks in obtaining food, because of the different needs of larvae and adults. Adult butterflies, beetles,
72
6 Foods and Feeding Biology
flies and others may need food themselves in order to mature and reproduce and must then deposit their eggs in places where the larvae are provided with, often markedly different, food. The nectar needed by an adult butterfly is commonly not provided by the same plant species that the caterpillar must eat, so that the butterfly must be capable of encountering both, perhaps by using different sensory mechanisms, and with the different plant species not necessarily even growing close together. The nectar from a variety of different plants may be accessible within a vegetation association, so that a more generalised search pattern may suffice, although many adult insects indeed exhibit strong preference for particular flower colours or structures. Consider two contrasting contexts of how insects may encounter food for their offspring, and how their general level of ecological specificity might be involved in how sites for egg-laying are selected. The large ‘swift moths’ (Hepialidae) are a spectacular feature of Australia’s more primitive moth fauna with close to a hundred species described, some of them very variable in appearance. The large silver grey/ brown adults of some species are familiar as attracted to light in late summer, and the caterpillars of many feed on roots of plants after tunneling underground. Females do not actively select where to lay their eggs, but simply fly across areas of open woodland or other vegetation and continually eject their eggs, as if on a ‘bombing run’ with rather imprecise targetting. This is in some ways risky – many caterpillars will almost certainly starve to death as unable to find suitable food – but risk can be compensated by large numbers. Many years ago, patient counts of eggs produced by a single large female of a Trictena species showed it to lay more than 50,000 eggs! Only two of these, of course, need to complete development in order to sustain population numbers in the next generation. A caterpillar hatching on the ground and wandering whilst seeking food may encounter any of a considerable variety of plant species. Should it ‘test’ one of these plants by biting it, three possible options arise – acceptance, rejection or tolerance. If the plant is ‘acceptable’, perhaps by texture or the presence of chemicals (and their detection by the caterpillar) that foster feeding activity, feeding can continue. Conversely, the insect might encounter repellent chemicals or, worse, toxins that poison it, or find the plant too tough to chew, in which case the caterpillar (should it survive the encounter!) is committed to trying again on another plant, and, unless it finds suitable food, may eventually starve to death. The third option is one of neutrality between the others. The plant is acceptable in not actively repelling the consumer and thwarting feeding, but supports growth and development only to some limited extent, often well below the optimal performance level afforded by a very suitable food. Much of the natural variation in developmental rate, adult size, fecundity and other measures of ‘quality’ of polyphagous insects is the outcome of food suitability. Contrast this somewhat chancy approach with the much more precise needs of highly specialised feeders that must encounter a specific food species in a similarly complex environment, in which the food resource needed may be both scarce and widely dispersed. Finding a particular species of small herb in a dense multispecies vegetation association, or a particular species of insect prey or host can be likened to search for the proverbial ‘needle in a haystack’, but is a problem that
Searching for Food
73
many such insects are remarkably good at solving, largely through using intricate chemical and visual cues and well-timed searching activity. Our insect – such as one of many butterfly species with broadly similar needs for specific plants – may spend considerable effort in simply finding somewhere suitable to reproduce, and may need to repeat the process many times in order to disperse its offspring widely. It may need to place its eggs very precisely, such as on foliage of a particular age, perhaps on the underside of leaves to avoid detection or rain, perhaps on buds or flowers for proximity of such food for caterpillars, or on the stem near the ground. For predatory insects, the presence nearby of prey may be an important criterion, so that some hoverflies lay eggs near visually-detected colonies of aphids on which the larvae can feed; and specific parasitoid wasps may need to initially locate a plant species on which their host feeds, and then the host itself. The host in some instances is not even visible, but may be tunnelling inside wood, so that parasitoids of woodwasps or timber beetles have to undertake close range investigations by probing with a long ovipositor to find them. The subtleties are endless, but the unifying principle demonstrates increased care and effort to assure survival chances of offspring in association with feeding specialisations in many contexts. Often, this increased security over polyphagous generalists is associated with lower fecundity, simply because energy is spent on searching behaviour rather than on egg development. Feeding specialisation may allow an insect to exploit a food or feeding environment unlikely to be used by other species, whilst lesser discrimination will almost always lead to it having to share that food with other species. The small ant-blue butterfly, Acrodipsas myrmecophila, is associated with a particular genus of ants, so-called ‘coconut ants’ (Papyrius spp.), and is unusual amongst Lepidoptera in that caterpillars do not feed on plants at all but live within the ant nests in dead wood, where they feed on ant brood. The ants’ common name refers to the strong odour they produce, reminiscent of ripe coconuts. It seems that female butterflies respond to this unusual and characteristic airborne odour as a very specific cue to locate the ant nest, and lay their eggs close to the entrance of the nest, whence young caterpillars are carried or escorted into the nest by ants. This case exemplifies many in which a very specific attractant (technically a ‘kairomone’, a chemical produced by one species and that benefits another species that can detect it) is used to find an equally specific scarce resource, and in which any confusion with other chemicals could not be tolerated, as harmful. The result is exploiting a very specialised food source in ways not easily paralleled by other consumers. For some other insects seeking scattered food, more general chemical stimuli may suffice. Dung beetles need to find freshly deposited dung in which to lay: some Australian species even cling to the fur around the anus of wallabies, and accompany dung pellets as they drop to the ground. Other species fly, commonly around dusk, and respond to the scent of fresh dung – and several species may arrive together on the same faecal deposit, so that elaborate competitive interactions may eventuate. Early arrival (efficient searching) may confer advantage, both individually and for the species, but another aspect of coevolution and habitat suitability has been important in the study of dung beetles in Australia. Many native species have evolved in association with
74
6 Foods and Feeding Biology
marsupials in woodland areas, with these mammals producing dung in small discrete pellets. They have largely been unable to adapt to the large amorphous dung pads produced by cattle in open areas, with the consequence that persistent dung has led to extensive pasture staling (such as by growth of rank grasses) and provided breeding sites for pest flies. The extent of the problem was stated in the mid 1970s as the amount of dung produced by cattle in Australia being 350–400 million pads each day! From 1968 onward, other species of dung beetles from southern Africa and Europe – where they evolved in conjunction with such large mammals – have been progressively introduced to Australia. Some of the more than 45 such species have been very successful in breaking down cattle dung rapidly, and several hundred individuals may develop in a single dung pad. In this example, understanding the relationships between native insects and their food supply was instrumental in solving a major agricultural and social problem. Either an ant nest or a dung pad may be viewed as a small ‘island’ of suitable habitat in a large area that could not otherwise support the insect adequately. The same applies to an animal prey, a host for a parasitoid, an animal carcass such as a dead snail (used by some flies) or small mammal (for ‘burying beetles’), or an orchid or other plant, or a small pool of water. An insect needing any such food or localised environment must find it, exploit it (perhaps breeding there for more than one generation) and in many cases cope with other individuals or other species wanting the same resource. Interactions between insects and their food have become very intricate and complex, and can change over time. Thus, the first releases of overseas dung beetles in Australia encountered virtually no competition from other species in cattle dung – none was there! Later, the cattle dung environment became more crowded, with five or six introduced species common in the same individual dung pad reported in Queensland. Later arrivals or introductions had to contend with possibly strong competition for resources in a habitat sought also by these already established beetles. Sharing such a resource may necessitate each species adopting different behavioural strategies and ecological needs – and the precise combination of species occurring together will be influenced by local climate, and so differ geographically and at different times of the year. The way in which each species uses the resource can also differ. Some of the dung beetles in cattle droppings are typical ‘brood ball rollers’, excavating balls of dung and removing them from the original mass These include species of Sisyphus, a genus named for the legendary founder of Corinth who was condemned to roll a huge rock uphill, with this continually rolling down again so that his task never ended. Many individual Sisyphus can together remove an entire dung pad within a day. Many other beetles are ‘tunnelers’ that bury dung in tunnels excavated under the dung pad and (as in some species of Onitis) do not do this for several days after arrival. ‘Sharing’ a resource to avoid interference or other competition, whilst continuing to live together, may involve very subtle levels of differentiation between the species involved. Various members of the above tunneling dung beetle fauna, for example, each typically excavate tunnels in different sites – at different depths in soil beneath the dung, or even within the dung mass itself.
Insect Herbivores
75
Insect Herbivores The same principle of ‘habitat division’ applies in any context, and has been studied extensively in plant-feeding insects, with the feeding site of each species reflecting the quality of the food, the ease of reaching and using it and, at another level, ways of the herbivore avoiding getting eaten itself in view of equally specialised pressure from ‘above’ by predators and parasitoids. So-called ‘enemy-free space’ becomes very important if a herbivore is to persist. In short, an insect feeding on a plant must cope with the plant itself, its immediate competitors and potential competitors, and its animal enemies as a wide collective portfolio of influences on its biology and wellbeing. High levels of feeding specialisation, at any feeding level, are very common, and also help to emphasise some advantages of small size. Basically, small individual size equates to little food needed to develop and mature, so that plant species, tissues or organs that are themselves scarce, small or few in number are not immune to attack by an insect capable of finding them. A single leaf, flowerbud or seed, or a single host caterpillar or beetle egg, may provide all the food that another insect needs. One consequence of this is that many different insect consumers may be able to ‘share’ a food species by specialising in different ways, or being present at different times of the year. Many small insects occupy feeding sites that are inconspicuous or ‘hidden’ in some way. Species of Melanterius are seed-eating weevils (beetles), some found on Acacia trees. Female weevils insert eggs singly through the wall of developing seedpods onto a single soft seed, in which the larva feeds. As a second example, many moths, and some Hymenoptera, Diptera and Coleoptera, are known as ‘leaf miners’ because the larvae develop within leaves rather than feed externally on them. The native jarrah leaf miner (the moth Perthida glyphopa) occurs in vast numbers on Eucalyptus marginata in south western Australia, and is regarded as a serious pest because it causes damage to tree crowns and reduces increment of wood. Its effect is to cause large areas of forest to appear ‘scorched’, as if by fire. From eggs laid on the underside of leaves, caterpillars tunnel inward to ‘mine’ inside, eventually using the upper and lower leaf surfaces as a shelter that they cut out and within which they drop to the ground before pupation in the leaf litter, leaving the leaves with a shot-holed appearance with, sometimes many, holes about 3–4 mm long. An even more restricted feeding habit occurs in many other moths – the introduced European oak leaf miner (Phyllonorycter messaniella) has caused concerns for its abundance on ornamental oak trees in some major cities, where it mines from the underside of leaves to produce blotch mines that appear silvery because of air within the leaf cuticle (Fig. 6.2), and are sometimes considered unsightly. The hatchling caterpillar is flattened, and feeds solely within the single cell layer of the leaf epidermis. As it develops, it moves deeper into the leaf, and becomes cylindrical – a ‘typical caterpillar’ – but feeds only in the mesophyll tissue, avoiding the upper pallisade tissue entirely, so that the mine is invisible from the upper surface. The caterpillar pupates within its mine, and the moth emerges directly from it. Exploring why such extreme specialisation occurs, when suitable food is apparently plentiful, raises many ecological questions.
76
6 Foods and Feeding Biology
Fig. 6.2 The leafmines of the introduced European oak leaf miner, Phyllonorycter messaniella (Gracilariidae), on Quercus. The ‘blotch’ mines (about 1 cm in diameter) on the underside of the leaf appear silvery because of air inside, and the caterpillar pupates within the mine, from which the moth later emerges. Two mines are shown, one old and the other occupied
Amongst these, we need to consider the theme of whether, and how, plants may be able to defend themselves against being eaten, or in some way influence the food available to insect consumers and other small herbivores. Plants may offer physical defences, such as being tough (high silica content in some grasses, for example, wears down the mandibles of caterpillars or grasshoppers), or in having coverings of waxes (that clog up the sucking mouthparts of small bugs such as aphids or leafhoppers) or leaf hairs (preventing the insect from reaching the leaf surface to feed), or more obvious spines. Alternatively, or in addition, chemical influences play major roles in moulding interactions between insects and plants. The traditional picture of plant and insect interacting is one of the insect attacking the plant, the plant thwarting this in some way, the insect in turn overcoming the plant’s defence, and the sequence repeating with continual modification and innovation in a stepwise manner. However attractive, this is in many cases overly simplistic – in particular, plant chemicals are much more variable in their effects, and reflect also the plant’s (or plant part’s) life style and vulnerability. Chemical defences by plants can be viewed as a continuous suite of variations that extend between two scenarios posed by different plant groups. Consider the possible impacts of insect herbivory on a small, short-lived annual herb and on a large, long-lived perennial tree, as highly contrasting life styles. Simplistically, the annual may need to grow rapidly, mature and set seed within a few weeks, and even a small amount of loss to any herbivorous insects may prevent this from happening – it may, for example, produce only a few leaves, flower-buds or seeds, and loss of
Insect Herbivores
77
any may be detrimental. The defence needed is ideally one that provides immediate and effective response to insect attack when feeding is initiated, to deter further attempts. Any such defence should also be ‘cheap’ so as not to deflect plant energy from growth. The ‘best’ effect may be for toxins that immediately poison (kill or weaken) any attacking herbivore or repel it strongly, and that are contacted as soon as the insect initiates feeding: the first bite or probe is then all that the plant may experience. A large tree faces a rather different problem. Massive amounts of foliage are produced, either synchronously as seasonal flush growth or more continuously, and the plant may mature for decades before any reproduction occurs. Protection from herbivory thereby assumes long term significance – individual leaves may need to function for months or, as in many eucalypts, for 2–3 years. Over such a period, foliage may change substantially in nutritive value and toughness. Young fresh foliage is generally highly nutritious, with high protein content, but the sheer quantity produced constitutes a defence not available to small herbs – simply, the ‘safety in numbers’ approach tends to ensure that much flush foliage will often escape attack over the short period that it is especially suitable as insect food. It may have to tolerate attack only over a single generation of the insect feeding. However, the surviving foliage must then be protected from continuing depredations, with insects likely to attack at any time and their target obvious. Rather than toxins, many trees produce chemicals that influence an insect’s feeding behaviour without poisoning it. Substances such as tannins are called ‘digestibility reducers’ and act by forming complexes with leaf proteins and so rendering them unavailable to insect feeders. The herbivores are not repelled, but continue to feed on low quality food, which may not allow them to develop or fulfil their optimal reproductive potential. They may form persistent associations with the tree, but cause it little harm, and the progressive buildup of tannins in older foliage augments its protection without exerting strong selection pressure as toxins do. Tannins are, rather, a generalised form of defence effective against numerous species, and regarded as very difficult for insects to overcome. A specific toxin, on the other hand, may be overcome by natural selection – and in some cases, these chemicals have been functionally transformed into specific search cues for specialised feeders that then adopt that foodplant in a monophagous (or near-monophagous) association. Most such cases involve plants that are highly toxic to most other potential feeders. Reverting to the two leaf miners noted above, examining the distribution of tannins helps to understand the feeding patterns they show. Eucalyptus leaves are symmetrical in structure, with pallisade tissue on upper and lower sides, whereas oak leaves have pallisade (where tannins are accumulated) only on the upper side, so that Phyllonorycter can avoid these completely by feeding on the mesophyll below this. More generally, ‘microsite’ selection for insect feeding is often related to palatability of the food, so that many insect life cycles are predicated on seasonal food suitability, not merely its presence. Many herbivores, for example, are specialised flush growth feeders that cease to reproduce once nitrogen content lessens in older leaves. Some may persist on trees, but with much lower reproduction and progressively slower development.
78
6 Foods and Feeding Biology
Table 6.1 Categorisation of major feeding groups of aquatic insects. The major groups are based in the feeding mechanism and the size of food particles ingested, with initial division as detritivores (feeding on non-living organic food), herbivores (feeding on vascular plant or algal tissue) and carnivores (feeding on living animal tissue). The categories are functional groups, valuable in assessing freshwater environmental condition (After Cummins 1973) Category Particle size range (mm) Subdivisions Chewers and miners: herbivores Shredders >103 Chewers and miners: detritivores on plant tissue Collectors <103 Filter/suspension feeders: herbivores-detritivores on living algal cells and decomposing organic matter Sediment/deposit feeders: detritivores on fine particles Scrapers <103 Mineral scrapers: herbivores on algae and associated microflora on substrates Organic scrapers: herbivores on algae and associated microflora Predators >103 Swallowers: carnivores on whole animals or parts Piercers: carnivores on cell and tissue fluids
This is only one of several different perspectives on plant defences involving chemicals, but emphasises that the conspicuousness of plants (their ‘apparency’, in ecological parlance) may be related to both severity and diversity of insect attack. A short-lived plant is fundamentally not available for much of the time and, particularly if rare, small, or highly scattered in its environment, may be difficult for insects to find, or depend on – it has low apparency or detectability. On the other hand, a large tree, thriving for centuries, is difficult to avoid: its high apparency renders it easily detectable over a long period, without seasonal restriction. It is almost inevitable that it will be encountered by numerous insect species, and that many of these will be maintained at rather low abundance levels but continue to coexist on the tree, many in specific sites or associations. Numerous general relationships between insects and plants allow categorisations such as foliage-chewers, sap-suckers, seed-eaters, pollen feeders, bark feeders, timber borers, root borers and many others, collectively encompassing the full array of feeding materials and sites available on or in any given plant species, their variations in seasonal availability and conditions, and insect feeding habits. These primary attackers, direct plant feeders or ‘herbivores’, are each in turn the food for an array of predators and parasitoids, many of them also highly specific in feeding habits and site, and emphasise the complexity of the food webs that can be present based on trophic level. A large plant, such as river red gum (Eucalyptus camaldulensis), is essentially a hub for a community of many hundreds of insect species, many of which in some way depend on that host species or, for the present, that individual tree on or in which they occur. However, some biologists working with freshwater insects have adopted a somewhat different approach, whereby the feeding method used – rather than the foodstuff itself – is used to define primary categories (Table 6.1). Thus ‘shredders’ (such as some mayfly nymphs and caddis fly larvae) feed by shredding coarse debris such as bark and foliage; ‘collectors’ (predominantly various fly larvae) filter fine organic material from the water or substrate; ‘scrapers’
Insect Carnivores
79
(exemplified by some further caddis flies) feed on algae and other material rasped or dislodged from submerged plant or substrate surfaces; and predators parallel the usage below, of feeding on other animals.
Insect Carnivores Higher trophic level relationships, involving insects feeding on herbivores and other animals, are also intricate, varied and complex – only rarely is a specific food chain of the form ‘carnivore-herbivore-plant’ easily defined, and much more complex food webs are more general. The most widespread of the higher trophic level are ‘predators’, those insects which feed directly on other animals (predominantly other insects) and so must undertake a sequence of discovery and exploitation paralleling that of herbivore insects seeking plants. Other feeding habits involved include ‘parasitoids’ (below) a particularly important category of feeding guild amongst insects. Perhaps the most notable difference in these interactions from those involving only herbivores and plants is that animal prey usually have far greater movement capability than a rooted plant, so that they have potential to escape in space by such means as running or flying away! Predatory insects fall into two very broad categories, reflecting needs for activity. These are ambush or ‘sit-and-wait’ predators, and active hunters that pursue their prey. Predators in either category may have very specific prey requirements, and others can eat anything they are able to capture, subdue and handle. Some large mantids in northern Australia have been reported capturing small birds, for example, but such cases appear to represent opportunism rather than any regular intent. Ambush predators succeed in large part because they are not detected and actively avoided by their prey. Mantids, above, may be very well camouflaged and highly cryptic when resting on vegetation (Fig. 6.3). Many of the Australian species resemble twigs or foliage in appearance, and even the more actively hunting species on bark are cryptic and capture their prey by stealth, using their highly modified grasping fore legs. Similar raptorial limbs have been developed independently in several other predatory insect groups: Mantispidae (Neuroptera) are named for their resemblance to mantids; some bugs (both in aquatic groups such as ‘water scorpions’ and some terrestrial ‘assassin bugs’, Reduviidae), and flies (Empididae). Mouthparts are also important ‘prey traps’ amongst many insect ambush predators. Aquatic larvae of Odonata have a unique structure, the labial mask, hinged under the head when the insect is at rest, but extended rapidly to impale prey that venture into range. Larvae of some lacewing families, predominantly those of Ascalaphidae (owlflies) and Nymphidae (a small family largely restricted to the Australian region) rest on tree trunks or foliage – some with bodies flattened and with elaborate lateral processes that help to hide them and reduce shadows – with their spined jaws held open at 180° or even more, to snap closed, like a trap, on prey animals (Fig. 6.5).
80
6 Foods and Feeding Biology
Fig. 6.3 A representative praying mantid (Mantodea) showing raptorial fore legs, large eyes and elongate prothorax, together with general cryptic appearance as an ambush predator on vegetation (Photo courtesy Mallee Fire and Diversity Project, La Trobe University/Deakin University)
Their close relatives, antlions (Myrmeleontidae), are easily the largest family of lacewings in Australia, and are best known for the habit in some species of larvae excavating conical pits in sand and resting with only their jaws exposed at the bottom of these, to feed on insects falling in (Fig. 6.4). Once captured, prey (ants or other) are dragged under the sand – a habit aided by a modification of the hind legs that ensures that the antlion larvae can move only backward. Somewhat similar traps are constructed by some tiger beetle larvae. Camouflage is widespread also in hunting predators, enabling them to avoid detection and retaliation as they approach prey. Some actually don disguise, as in some green lacewing larvae covering themselves with debris or remains of previous prey as they approach aphids, but a wide array of adaptations is involved; several structural features are reasonably good clues to their life styles. Many adult active predatory insects are predominantly visual hunters. They have very large eyes, in dragonflies, owlflies and some Diptera occupying most of the head surface. As other correlated predatory features, mouthparts may be orientated toward the front of the head, and legs may be modified for prey capture or elongated for rapid pursuit. Some may adopt territorial behaviour, with elevated perches giving them high visibility over surrounding terrain – one colleague has told me of owlflies patrolling from his garden clothes hoist, for example. In some ways, they are really also ‘sit-and-wait’ predators – rather than cruising around to look for prey, they detect it visually before making short darting flights from their perch to capture it – a much more efficient use of energy. Predation has several clear stages, each of which may demand very specific behaviour – searching for prey, capturing and subduing it, and eating it, against a background of the prey proffering defences that parallel the variety of those of
Insect Carnivores
81
Fig. 6.4 Pits in shaded sandy soil constructed by larvae of an ‘antlion’, in this case Myrmeleon acer (Neuroptera, Myrmeleontidae), one of the most widespread species in Australia. The conical pits vary in size with larval instar; the coin, for scale, is about 2.5 cm diameter
Fig. 6.5 A larva of Osmylops sejunctus (Neuroptera, Nymphidae) on a eucalypt leaf. Nymphidae is a small family of lacewings found only in the Australia/New Guinea region. The larva is discoidal, semi-translucent and strongly flattened, and the long, haired lateral processes eliminate any shadow production as it rests, as an ambush predator, for prey to venture within capture range. The ‘jaws’ (mandible and maxilla on each side) are held open at more than 180° and snap shut to trap prey
82
6 Foods and Feeding Biology
Fig. 6.6 The yellow-winged grasshopper, Gastrimargus musicus, exposes its brightly coloured hind wings as a ‘startle display’ to deflect or delay predator attack. Such defences are not always effective: blundering into the orb web of an araneid spider has here proved fatal
plants against herbivores. Searching behaviour may have to counter prey hiding, and deterring them from simply running or flying away when they detect predators, but insect species susceptible to being eaten have collectively developed numerous behavioural and ecological defence strategies to thwart attack. Crypsis is widespread amongst potential prey, with numerous insect herbivores resembling vegetation on which they rest or feed. Treehoppers (Membracidae) and other bugs may have elaborate thoracic processes and spines, for example, and appear like thorns or other projections on plants. The elongate ‘sticklike’ form of phasmatids and many grasshoppers is thought to help them avoid detection, and the appearance may be augmented by behaviours such as swaying in the wind. However, many features of vulnerable insects have developed in response to vertebrate predation, rather than to other insects alone. Thus the yellow-winged grasshopper mentioned in the preface is normally cryptic on vegetation but if disturbed, suddenly exposes the bright yellow (but normally hidden) hind wings in taking flight (Fig. 6.6). The ‘surprise’ may cause a predator such as a lizard or bird to delay attack and allow the insect to escape – numerous similar cases occur, where exposure of coloured or strikingly marked body parts creates (or is suggested to create) confusion – the ‘eyespot’ markings on the hind wings of some moths are another example. Functionally different, but also involving changes in body shape, ‘decoy’ structures may help to deflect attack. The ‘tails’ on the hind wings of small butterflies, such as the imperial blue (Jalmenus evagoras) and its relatives help to create the appearance of a ‘false head’, together with other wing markings (Fig. 6.7). When the butterfly is resting or feeding, those projections may be moved, and the movements attract the attention of predators and lead them to strike at them. The attack is
Insect Carnivores
83
Fig. 6.7 The Imperial blue butterfly, Jalmenus evagoras: a worn male at rest. The ‘false head’ impression is conveyed by the tailed hind wings subtended by brightly coloured orange spots, to which attention is directed by dark lines on the wings (see also Fig. 7.4)
thus directed to a non-vital part of the body, and, although perhaps losing part of its wings, the insect may survive and escape – it is not unusual to find butterflies with beak marks on the wings, for example, or parts of the wings torn away, suggesting escapes of this nature. Other modifications that deter vertebrate predators include combination of structural changes and markings so that some hawk moth caterpillars resemble snakes, by having their abdomen expanded and with large eyespots, resembling a reptile’s head. However, bright colours can have another important defence context. For many insects, they advertise that the insect is distasteful or toxic, so that combinations such as black and red, black and yellow (as in some stinging wasps) or black and white are ‘warning signals’, technically aposematic colours. Predatory vertebrates such as birds, lizards or insectivorous small mammals may initially grab such an insect but be repelled, poisoned, or induced to vomit by its chemicals – sometimes accumulated directly as plant-produced toxins, denatured and stored in the prey’s body and employed in its own defence. An individual predator learns from this experience and thereafter may avoid any insect with a similar appearance. Being very conspicuous – brightly coloured and active by day – may be an effective defence, so that what might initially be a potent search cue for predators to find their prey assumes a very different role. Many highly coloured insects thereby depart from the more familiar guise of becoming cryptic. The caterpillars of the cinnabar moth (Tyria jacobaeae), a species introduced to Victoria from Europe in effort to control the alien European ragwort weed in pasture, are banded in black and yellow and (even though compared with the livery of a well-known Melbourne football club) this widespread general aposematic pattern transcends adult and larval insects.
84
6 Foods and Feeding Biology
Fig. 6.8 Robberflies (Diptera, Asilidae) are diurnal visual predators, streamlined and with large eyes that detect prey. Many species rest on vegetation and make short darting flights to capture prey with their long, strongly spined legs. These structural and behavioural traits are paralleled closely in some other predatory insects
Plant chemicals are deployed widely in herbivore insect defences, occasionally augmented by a collective strategy, rather than by individual insects alone. One such case is for the common ‘spitfire grubs’ of certain sawflies on Eucalyptus, which can occur in large aggregations of black or dark green larvae feeding on foliage, and rest in clumps. They are able to separate the noxious eucalypt oils from their food, and store these in a pouch near the front end of the gut. If disturbed, the grubs raise the end of their abdomen and wave this, whilst simultaneously regurgitating the strongsmelling sticky oils – so that the collective defence here involves components of appearance (increased size conveyed by aggregation), movement, repellent scent, physical ‘sticking’ and poison. Simply operating in groups, rather than alone, may help confer protection. More active physical defence (such as kicking or biting back) is not uncommon, and may be associated with structures – the long jumping hind legs of many grasshoppers are strongly spined and can inflict wounds on attackers, and strong mandibles and stings can both be used for attack and defence. Predatory insects may have features that appear to help protect them from such retaliations. Robberflies (Diptera, Asilidae, Fig. 6.8) and owlflies (Neuroptera, Ascalaphidae) are amongst the many insects that capture prey whilst flying. They often have very hairy faces (some appearing as with a ‘beard’), and this has been suggested to help protect the eyes, in particular, from physical injury from prey. Bright colours and distastefulness culminate in strategies of ‘mimicry’, whereby other species of insects may also become protected through being conspicuous. A vertebrate predator that attacks a distasteful insect, as above, may carry the search image of ‘something to avoid’ for the rest of its life, so that any other species with similar appearance may also be avoided, even though it may be palatable, so that convergence in colour pattern or conspicuousness may be beneficial. This, mimicry,
Insect Carnivores
85
takes several forms, two of which have influenced much thinking on evolution of insect form and diversity. Pioneered through studies of butterflies in the tropics, Batesian mimicry (named for the famous nineteenth century explorer-naturalist, Henry Walter Bates) involves one species (the mimic) avoiding predation by resembling a toxic species (the model). For this to be effective, several other conditions must prevail – the mimic must co-occur with the model, and be sufficiently similar to it in appearance and behaviour to confuse the predator, and it must be less abundant; otherwise, predators might form a positive image for the palatable prey item. A second category, Müllerian mimicry (named for ‘Fritz’ Müller, who also worked in Brazil and published some classic studies on insect coloration in the nineteenth century), operates somewhat differently, in that an array of distasteful vulnerable species increase chances of individual protection by converging in appearance so that a predator cannot differentiate them, and a single search image suffices. Moths, beetles, flies, wasps, bugs and others may participate in such a complex of co-occurring and diurnally-active species by exhibiting a common pattern of bright colours or other aposematic features. As noted above, black and yellow bands are one such ‘warning pattern’. Appearance, time of activity, and many aspects of behaviour may enhance the collective effect. ‘Buzzing’, for example, may convey warning because it is associated with stinging Hymenoptera, and be adopted by other insects as a component of their mimicry.
Insect Parasitoids Parasitoids, mostly small wasps but also including flies and a variety of other insects, must also trace their food in a diverse environment, but in this context, the searching female adult is seeking a food supply for its offspring rather than for itself. An adult parasitoid wasp may need to find the egg, larva or pupa of a specific insect host in which to lay one or more eggs, but itself primarily feed only on nectar or pollen. The ‘host’ is killed, and the term ‘parasitoid’ refers to this rather specialised relationship, differentiating the protagonists functionally from true parasites that do not kill their hosts. The two terms are often used synonymously, with ‘parasitoid’ increasingly common over recent decades. It is all too easy to become both confused and distracted by semantics as we try to partition insect feeding habits into neat categories! For sanity, accept that experts will continue to disagree and that the exuberant variety of behaviour and ecology involved largely prevents full consensus in attempts to partition these neatly and unambiguously. Most simply, a ‘parasitoid’ in many ways is intermediate between a predator and a parasite – it eventually kills its host, only a single host is needed for it to complete development, and the host – once attacked – is not stored, as are the paralysed prey of many solitary wasps regarded conventionally as predators. Nevertheless, two wider uses of ‘parasitoid’ need to be mentioned here, simply because they arise occasionally in similar functional terms. First, some entomologists include some insects feeding on plant hosts – such as seed weevils developing on a single seed; and, second, it has been argued more
86
6 Foods and Feeding Biology
widely that all organisms that develop on, and kill, a single animal host fall into this functional category. Here, as most commonly, the term is confined to holometabolous insects that do this. Parasitoid wasps are amongst the most numerous of all insects, and are also amongst the poorest known groups, reflecting that many are very small and difficult to study; they also span an enormous variety of biological patterns, as an eloquent testament to the variety that can develop within a basic feeding habit. Much of the interest in parasitoids has developed because of their importance (together with predatory insects) as biological agents that can be manipulated to suppress pest insects. Indeed, most insect groups are attacked by one or more parasitoids. Some tiny ‘fairy flies’ (which, despite their common name, are true wasps, Mymaridae) may even swim under water, using their wings as oars, to find water beetle eggs in pond weeds in which to lay their eggs. One level of specialisation, therefore, is the host stage susceptible to the wasp – different parasitoids may attack the eggs, larvae or pupae of the same host species; other differentiations include the host taxon (ranging from very general to highly host-specific parasitoids), habitat, and mode of attack. Some are nocturnal – large pale brown ichneumon wasps are common visitors to lights in Australia – reflecting night time activity of their hosts. However, knowledge of the host range of most parasitoids is fragmentary, and even much of the published record is of dubious value, because of the difficulties of identifying and differentiating species in a fauna with high numbers of taxa undescribed. Australian workers can take little comfort from recent comments on the best–known of all parasitoid faunas, those of Britain and western Europe, such as particular groups of these wasps being, even there, still ‘extremely poorly known’ (British Ichneumonidae), having ‘numerous cryptic species’ (British aphid parasitoids), ‘difficult to identify’ (British Figitidae), ‘frequently posing problems with their identification’ (British Trichogrammatidae), and so on. Collectively, parasitoids were described in a recent text as ‘exhibiting incredible levels of species richness, accompanied by an equally high level of diversity in biological habits’. Host records are frequently inaccurate, but even a ‘correct’ name allocated on morphological features alone may simply designate a complex of different biological or genetic entities masquerading under a similar structure. This richness and diversification is reflected by parasitoids occurring in more than 40 families of Australian Hymenoptera, distributed amongst about 10 superfamilies. Some, indeed, have become ‘hyperparasitoids’, extending the food webs by adopting parasitoids as their hosts. Without detailed rearing records, it is often not possible to differentiate parasitoids from hyperparasitoids, and – almost by necessity – many of these organisms confined to higher trophic levels may be scarce, particularly if confined to hosts that are also rare. By comparison with wasps, parasitoid flies are far less varied but the habit has still evolved in distantly related fly families such as the Bombyliidae (bee flies) and Tachinidae (one of the most diverse families of Diptera in Australia, attacking representatives of many other insect orders). Two very broad categories are often differentiated, by their relationship with the host. Parasitoids that cause their host to cease development once attacked, so that the host does not then moult or increase in size (sometimes it is paralysed by toxins, although still living) are known as ‘idiobionts’. Others (‘koinobionts’) have hosts
Insect Carnivores
87
that remain active and continue to develop – although often slowed from their usual healthy rate. In these circumstances, the parasitoids gain extra time to develop – so that an infested caterpillar may continue to feed and moult, but is eventually killed. The parasitoid adult may emerge from a later host stage than that originally attacked, for example as a larva-pupa parasitoid. Adult parasitoids use many different environmental cues in finding hosts. They may respond to plant chemicals or structures (such as galls or leaf mines), the scent or frass of their hosts, and as in other feeder-food interactions the host may develop countermeasures to avoid attack. As with searching insect herbivores, many parasitoids fall into one of two searching strategies. These, although they are simply the extremes of a continuous range of options, are sometimes designated as more discrete categories. ‘Time-limited’ parasitoid species can mature further eggs once they have deposited their present complement, so that reproductive output reflects the number of hosts they are able to encounter over their life. ‘Egg-limited’ species have a finite number of eggs, so a maximum reproductive capacity with the implication that very careful host selection may be critical.
Insect Parasites As noted above, parasitic insects represent yet another category of feeding specialists. They are amongst the most specialised and highly modified of all insects, and several orders have adopted this way of life exclusively, as external parasites (ectoparasites) of vertebrates. Their modifications extend well beyond feeding habits, but reflect also that they must remain in close association with hosts that are often highly mobile or disperse over large distances. Two insect orders are invariably associated with warmblooded vertebrates, and have very different life styles, but these feeding associations give some members immense importance as vectors of diseases of livestock, wild mammals and birds, and people – the historical impacts of some lice (as vectors of typhus) and the plague flea (Xenopsylla cheopis) have indeed been substantial, for examples. Parasitic lice (Phthiraptera) pass their whole life on the host. They glue their eggs to hairs or feather plumules, and the nymphs and adults grasp these firmly with highly modified legs. Different groups of lice feed on scurf and general skin debris, or on blood. Adults of the other order, fleas (Siphonaptera) feed on blood, but their maggot-like larvae feed more widely on organic debris in nests or burrows of the host. Both orders are wingless – they are carried around on their hosts (even internationally on migrating individuals), and fleas are renowned for jumping, so can regain their hosts if dislodged temporarily. Lice and fleas are also transferred between host individuals during grooming, communal roosting, mating and other contacts. Loss of wings, and modifications of legs and body form, is common also in those highly specialised flies that have become ectoparasites, and in some other vertebrate-infesting insects. Blood-feeding bedbugs (Hemiptera, Cimicidae), for example, leave the host after nocturnal feeding, and secrete themselves by day in crevices or other nearby retreats. However, with complete reliance on the host for its survival, a parasite’s life
88
6 Foods and Feeding Biology
has been described as ‘an impressive gamble’. Insect parasite life styles and wider adaptations help increase the odds of survival by facilitating close associations – in much the same way as obligatory feeding relationships throughout the Insecta are assured. Adult fleas can starve for up to many months should their host die, and some species are renowned for the pupae undergoing long periods of dormancy within the cocoons. They respond to disturbance, and can then hatch rapidly – so, in situations where migratory hosts leave a nest or burrow for months on end and return the following breeding season, fleas do not emerge until the host returns and blood food is once again available. The commonly reported incidence of ‘a plague of fleas’ in long-empty houses or chicken huts simply reflects this previous occupancy by flea hosts, and mechanical disturbance by current visitors, not necessarily the hosts. The distinction between parasitism and predation is sometimes fine, because mortality from parasite attack can occur, and confuse the usual bland definition of non-killing. A leading British ecologist, Charles Elton, in noting the difference that a parasite is in some ways a form of predator that does not habitually kill its prey, described the difference as a predator living on capital and a parasite living on income. Two somewhat intermediate examples illustrate the ambiguities that can arise. The Strepsiptera are tiny insects that parasitise (or, perhaps, ‘parasitoidise’!) other insects such as leafhoppers and some wasps, feeding on the body tissues of their hosts, usually without killing them – although the hosts may be sterilised and undergo severely protracted development and, sometimes, morphological changes. As endoparasites, Strepsiptera feed within the host, and most of them pupate there. Adult males leave the host and fly, whilst the grublike wingless females remain in the host, simply extruding their anterior end and emitting pheromones to which the males respond. Unusually, sperm are inserted into a ‘brood chamber’ near the anterior of the female’s body, and move to the genital opening. Host relationships of the Australian species are poorly known. Hosts die after parasite emergence, so do not survive to reproduce. Our second anomaly is a group of flies whose larvae eat bird nestlings. Species of Passeromyia are closely related to houseflies and bushflies in the family Muscidae, but females lay eggs in active nests of Australian honeyeaters and other birds; the carnivorous maggots burrow into chicks, and sometimes kill them – although by no means universally, so that individual variation in outcome may here affect our definition.
Insect Decomposers The final broad feeding category of insect consumers we need to note here are those that feed on dead organisms, either animal or plant, as ‘scavengers’ or ‘decomposers’. They thus do not affect the mortality or reproduction of their food species, but play important roles in recycling activities. One significant group, dung beetles (including the major cohort of species introduced from southern Europe and Africa to aid breakdown of cattle dung), has already been mentioned. They are paralleled by insects such as carrion beetles, disposing of small mammal carcasses, and numerous
Insect Decomposers
89
herbivores that specialise on dead and decaying vegetation. As with all the other feeding habits mentioned, seasonal effects can influence decomposer life cycles and activity – autumnal leaf fall, for example, can be a strong bias to seasonally increasing dead leaf litter as food; the duration of food ranges from a few days (small carcasses, some fleshy fruits) to decades or longer (large dead trees) and may host many generations of insects that feed on dead wood, and involve complex successions of community change as the wood gradually rots or is broken down by earlier inhabitants. Many decomposers are not very conspicuous insects, and often escape notice from casual observers, unless structural damage becomes evident – as in the depredations of wood-feeding termites on timber in service, such as houses. Notwithstanding Ogden Nash’s observation (‘Some primal termite knocked on wood; And tasted it and found it good. And that is why your Cousin May fell through the parlor floor today’), many termites in Australia feed mainly on dead grasses and are amongst the most important decomposers in savanna ecosystems, and the two major sources of bulk cellulose (grasses, wood) are exploited largely by different termite taxa. However, rather few species may be involved in any locality. In contrast, dead foliage of eucalypts is a major food resource for many caterpillars and beetle larvae, without which the tough decay-resistant leaves can persist for up to several years on the ground. The enormous moth family Oecophoridae (‘mallee moths’, arguably the largest insect family in the country with well over 5,000 species) includes hundreds – perhaps thousands – of Australian species whose caterpillars are specialist feeders on dead eucalypt foliage, and which can be found at densities of up to several hundred a square metre, with many species occurring together. Each of these major categories of feeders, and their intergrades – we should remember, also, that many insects are omnivorous and transcend roles by being, perhaps, both predators and herbivores, or feed differently at different stages of their life – may be specialised at different scales of ecological interpretation. Grouping like with like in ‘feeding guilds’ must therefore incorporate some indication of scale; a ‘herbivore guild’ is clearly far more encompassing than a ‘leaf-eating guild’ or a ‘sap-sucking guild’ as subcategories of this, for example, and many such restrictions involving subdivision of the total potential resources available are highly characteristic of particular insect groups, together with the range of plants on which each may feed. A herbivore might be considered a specialist because it feeds on Eucalyptus foliage and cannot eat leaves of any other tree genus, even when available easily. However, it might live in a forest where several species of Eucalyptus occur together. From this finer viewpoint, it might be considered a generalist if it feeds on all of these and a specialist only if is even more selective and feeds on only one (or some other subset) of them. Even then, it may be restricted to foliage of a particular size or age class. For a pergid sawfly studied some years ago, previous defoliation of some eucalypt trees (by psyllid bugs and beetles) resulted in reduced leaf size, so that female sawflies could not then grasp the leaves to lay in them. Conversely, leaves may become too large for the females to handle! Apparent specialisation may reflect a considerable variety of evolutionary and more locally influenced outcomes so that simply defining an insect as a feeding specialist may be difficult without prolonged
90
6 Foods and Feeding Biology
observations and experimental trials. Further complications can arise because of differences across the insect’s distributional range, in part reflecting the availability of different foods. Some swallowtail butterflies with different feeding habits in different places illustrate some of the situations that arise. The chequered swallowtail, Papilio demoleus, can be a serious pest of citrus trees, but in some places the same citrus foliage is toxic to caterpillars and prevents them maturing. Another species, Macleay’s swallowtail (Graphium macleayanum) uses only a single host plant species (southern sassafras, Antherosperma moschatum) in Tasmania but further north in eastern Australia, its caterpillars eat foliage from at least 13 plant species spanning seven genera and four plant families. Such local specialisations are sometimes called ‘ecological monophagy’, but they are usually very difficult to explain. Local feeding restrictions and differences have been attributed to many different influences, including the availability of particular foods – and of suitable, acceptable alternatives – and influences of natural enemies, so that geographically restricted ‘host races’ amongst herbivores are frequent, even when edible alternatives exist. The suitability of any particular food plant (or, by extension, other food such as prey or host) may be consistent or influenced by local conditions, notably by other members of the community in which the species of interest occur. Finding suitable foods, or avoiding harmful foods, are important facets of how insects respond to stimuli within their environment. In some contexts, exaggerated stimuli may be emitted by either plants or animals that ‘lure’ insects within capture range or otherwise harm them, or provide the mechanisms that assure mutually beneficial interactions. The main basis for cross-pollination by insects, for example, is that flowering plants produce brightly coloured, strongly scented and nectarproviding flowers that attract pollinators, often specifically, and which subsequently carry pollen to other flowers. It is a relatively small step from this mutualism to a more one-sided relationship. Insectivorous plants, whether by the sticky globules produced by sundews (Drosera) (Fig. 6.9) or the colour and nectar glands of pitcher plants (such as the Western Australian Cephalotus), attract insects that are then used for their own nutrition. And some predatory insects (such as some mantids, as convincing mimics of flowers, above) are brightly coloured, resembling flowers, and rest on flowers where they feed on visitors. In some cases, some involving interactions with alien species, conservation concerns can arise from behaviour resulting from these relatively novel associations. An introduced South American vine, Dutchman’s pipe (Aristolochia elegans), is highly attractive – far more so than the native vine (Pararistolochia praevenosa) used by caterpillars for food in rainforests in southern Queensland and northern New South Wales – to females of the Richmond birdwing butterfly (Ornithoptera richmondia), and they select this alien species on which to lay their eggs. However the foliage of Dutchman’s pipe is toxic to the young caterpillars, and kills them, so that its spread into forest environments is a major threat to the butterfly. Studies of insect natural history and diversity include considering how species may live together and respond to each other within communities, and the means by which the available resources there can be partitioned or shared. Feeding relationships, and the ways in which they proliferate and vary, are integral to this. Clarifying
Further Reading
91
Fig. 6.9 The small scorpionfly Nannochorista (Mecoptera) is found near wetland areas in the south east. It is shown here on a co-occurring insectivorous sundew, Drosera
the feeding role of an insect species in general terms such as herbivore or predator is often relatively straightforward, but estimating its full ecological role and variation is not. We get strong inferences on feeding habits from where the insect is found regularly, but often can only guess why it does not also occur elsewhere. Detecting the full prey range of mobile lacewing larvae or ground beetles may necessitate sampling the full variety potentially available and, as with herbivores, trials in controlled conditions to see which possible foods are acceptable and support development. The community context of any such study may be highly important in clarifying any such relationships.
Further Reading (Note that none of the following is peculiarly Australian in coverage, but each helps to display the variety and intricacies of associations between insects and their foods) Barth FG (1991, translation by Bieder-Thorson MA) Insects and flowers: the biology of a partnership. Princeton University Press, Princeton, New Jersey (wide-ranging discussion of many aspects of insect-flower relationships and their development and sensory basis: a classic work) Cummins KW (1973) Trophic relations of aquatic insects. Annu Rev Entomol 18:183–206 Godfray HJC (1994) Parasitoids. Behavioral and evolutionary ecology. Princeton University Press, Princeton (broad survey of parasitoid variety and biology)
92
6 Foods and Feeding Biology
Hanski I, Cambefort Y (eds.) (1991) Dung beetle ecology. Princeton University Press, Princeton (includes Australian species in global surveys of dung beetle diversity and ecology) Jolivet P (1992) Insects and plants. Parallel evolution and adaptations, 2nd edn, Flora and Fauna Handbook no 2. Sandhill Crane Press, Gainesville (as with Barth, above, a readable and stimulating book on the complexities of insects interacting with plants) New TR (1991) Insects as predators. New South Wales University Press, Kensington (broad survey of biology and variety of predatory insects) Roslin T, Kotze DJ (eds.) (2005) Spatial ecology of hebivorous insects. Annales Zoologici Fennici 42:291–475 (Proceedings of a symposium, containing many recent studies and interpretations on the biology of plant-feeding insects)
Chapter 7
Insect Behaviour and Lifestyles
Introduction: Behaviour and Adaptation Whether living in water or on land, in or on plants or other animals, every insect individual and species has to face – and solve – a similar suite of problems in order to survive and thrive. Many of these demands have already been noted – how to find and exploit food, how to locate and court mates, how to respire under water, and how to coexist with other, often competing or antagonistic species in ‘ecological space’, and many others. The examples noted earlier, such as intricate behaviour of predators to seek prey and correspondingly of prey to avoid being detected or captured, help to emphasise the critical importance of ‘correct’ responses to environmental cues or stimuli in survival and in an insect fitting in to its ‘niche’ in nature. They link with development of finely honed ecological and physiological attributes influencing the responses that adjust an insect to its environment, and also with intricate behaviour patterns of individual insects, between individuals within a species, and between species in assemblages or communities. Both the cues involved and the activity itself may change with gender, age, stage, or physiological condition (such as hunger level – so that a hungry predator may search more avidly for food than one that has fed recently – or reproductive state), with much insect behaviour innate (instinctive). Dispersing bean aphids (Aphis fabae) respond to different wavelengths of light as they age – initially they respond to blue wavelengths, so that they fly upward into the sky to join the aerial plankton, that mass of mostly small insects dispersed by wind currents, rather than flying actively. As they age, their orientation changes, and aphids respond to green and yellow wavelengths, those of the plants on which they must settle in order to reproduce. Aphids then fold their wings and gradually ‘sink’ to the ground. Most such behaviour is indeed instinctive and invariable, and this has long been presumed to be the major category of insect behaviour. However, by the end of the twentieth century it had become abundantly clear that ‘learning’ (that is modifying behaviour in a repeatable manner as a result of experience, with the implication that insects can associate stimuli with environmental condition and process ‘memory’)
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_7, © Springer Science+Business Media B.V. 2011
93
94
7 Insect Behaviour and Lifestyles
Fig. 7.1 The jewel beetle Julodimorpha bakewelli, a large member of the mallee Buprestidae, here recovered from a pitfall trap set for vertebrates (Photo courtesy Mallee Fire and Diversity Project, La Trobe University/Deakin University)
is also widespread, although commonly difficult to prove. Historically, the value of learning to many insects had been doubted, because their short lives and small size suggested that innate behaviour alone would be sufficient for their needs. Learning had long been known in the behaviour of social insects – such as foraging bees repeatedly visiting the same kinds of flower to collect nectar on their trips, and responding to colours and scents to do this and communicating information on nectar sources to other members of the colony by intricate ‘dances’. Learning can be very rapid, and many ethologists now think that it is probably universal amongst insects, notwithstanding the intricacies of innate responses. Learning itself can take many forms, and one of the most common amongst insects is ‘associative learning’ whereby insects learn to associate reward with some particular syimulus in the environment. Not all responses may be rewarding, and the incidence of ‘supernormal stimuli’ may be a strategy of predators or plants as ‘decoys’ that aid their own wellbeing by attracting food or pollinators. Superficial structure and colour may also provide powerful attractive stimuli in more innocuous contexts. One large Australian jewel beetle (Julodimorpha bakewelli, Fig. 7.1) is infamous through its reported predeliction for attempting to mate with empty beer bottles!. Male Julodimorpha are attracted to discarded brown ‘stubbies’ by their colour and marbled texture, and presumably perceive these as ‘supernormal’ female beetles. Despite the precision of many cues used by insects for searching, ‘mistakes’ such as this are not uncommon – many water beetles are attracted to shiny bonnets of cars, as reflective surfaces mistaken for water bodies, for example. And responding to the wrong cues may lead to trouble – the brightly-coloured but well-camouflaged mantids sitting on flowers and resembling them capture and eat insects attracted to them, and chemical cues may induce insects to select non-suitable plants on which to lay.
Sensory Mechanisms
95
Part of the conservation interest in the Richmond birdwing butterfly (Ornithoptera richmondia) in Queensland and New South Wales results from its decline flowing from its attraction to an introduced vine (p. 90). Sensitive sensory responses to such decoys can be fatal – so that the ‘correct’ behaviour is honed and refined by evolution over long periods as environments and communities change as the milieux in which insects must continue to thrive. More broadly, any insect must ideally (1) survive and complete development; (2) find and exploit food and other critical resources; (3) find, or be found by, a mate, and reproduce; as well as (4) contend with all other species, including natural enemies, in its environment, (5) be able to disperse or be dispersed to track suitable resources and/or refuges in the landscape, and (6) resist adverse changes in the environment. ‘Correct’ behaviour is part of the adaptive armory that enables insects to survive and their species to persist and evolve.
Sensory Mechanisms Many insects respond very precisely to environmental cues, with behavioural responses predominantly to visual, chemical, or tactile cues – but no senses can be excluded fully from influencing the repertoire of influences. Many orthopteroids, as well as cicadas and a few others, are renowned for sound production, with the massed daytime or evening choruses of suburban cicadas or crickets a much anticipated harbinger of warm summer weather. The sounds produced, species-specific and sometimes with distinct regional ‘dialects’ across a distributional range and constituting extremely specific mating signals, have a variety of different roles. They can be very penetrating: noise of the familiar cicada Cyclochila australasiae (the ‘Greengrocer’, with different colour forms including the ‘Yellow Monday’ and others) in the south east can near the pain threshold of many people, and continue unbroken for an hour or more. It is one of the loudest of all insects. Most commonly, sounds are used to attract mates, but can alternatively lead to aggregations, mark territory or be used in defence. Even time of ‘singing’ may be highly important – the very loud noises produced by some strongly flying day-singing cicadas, sometimes from numerous males sitting close together, may confuse potential bird predators, and even repel them. Some dusk-singing cicadas, in contrast, are weaker flyers, and their activity period may then coincide with less activity by insectivorous birds. They are yet another symptom of insect variety – crickets, katydids or grasshoppers that appear structurally identical, even under close examination by specialists, may have very different ‘songs’ that assure matings are between members of the same species: the insects can tell the species apart far more ably than mere humans can do! In practice, people studying the taxonomy and relationships of crickets, katydids or grasshoppers try, wherever possible, to make recordings of their stridulations as an important taxonomic tool, much as ornithologists may do for calls of some groups of birds. In many species, the sounds produced are not audible to people.
96
7 Insect Behaviour and Lifestyles
Male insects are the usual sound-producers, with a ‘calling song’ often loud and characteristic and inducing females to respond and answer. The two broad categories of mechanisms for sound production involve either body parts alone, or knocking body parts onto the substrates, and sound perception also differs across insect groups. Many orthopteroids have ‘eardrums’ (properly, tympana) or special hairs that pick up sounds. Tympana occur on various sites – on the fore tibia of many crickets and katydids, or on the sides of the first abdominal segment in grasshoppers, as examples. Likewise, sounds are produced in various ways, commonly involving modifications to legs or wings to produce rasps or stridulatory files as rows of small pegs, in association with resonating structures to amplify the noise generated. Thus, all of the thousand or so species of Australian katydids produce sounds, most commonly from a ‘stridulatory file’ (a row of small pegs on the underside of the left fore wing, near its base, rasped against a raised vein on the right fore wing) and detected by females by either tibial tympana or an auditory structure on the thorax. Females of some katydids can respond by sound production, but many others seem to remain silent and respond only to the sound of males of their own species. In cicadas sound production is almost wholly by males (long commemorated in the adage ‘Happy are cicadas lives. Because they all have voiceless wives!’ attributed to the Greek philosopher-poet Xenarchus). The main exception to this is the primitive hairy cicadas (the two species of Tettigarctidae found in south eastern Australia), in which audial communication is initiated by either sex, by tymbals (which are flexible membranes on either side of the first abdominal segment) and amplification through airsacs in the abdomen. Both sexes of Tettigarcta transmit low intensity vibrations (the sounds are not audible to people) through the substrate, and these are thought to help locate mates from nearby. They lack tympanal organs, so cannot ‘hear’ in the same way as the more advanced cicadas. Some other insects simply bang their abdomen on the substrate to produce audible noise or vibration – with the pattern sometimes even then a useful taxonomic character. The death watch beetle’s ‘knocking at the door’ in Europe is produced in this way, and some booklice (Psocoptera) have a similar system. Even out-of-doors, characteristic patterns of vibration produced by tapping the abdomen on leaf surfaces can be part of a courtship ritual, with closely similar species of some green lacewings differing in the pattern they produce. Many insects not usually thought of as producing sounds actually do so, either within the human audible range or ultrasonically. Moths are one such case. Males of the pyralid Syntonarcha iriastis (a moth widespread in the western Pacific and Australia) produce sounds from their highly modified genitalia, and these can be heard over at least 20 m. This behaviour occurs during evenings, with the moths resting on top of bushes or trees with their genitalia exposed, and is thought to combine with pheromones produced from specialised haired areas on the abdomen. Another unusual example is of Australia’s endemic day-flying ‘whistling moths’ (Hecatesia, Noctuidae), in which sounds are produced, voluntarily, during flight. Their clicking noise is made as the toughened coastal areas of the fore wings clash during flight, and gives the effect of whistling – it has also been likened to castanets – and can be heard more than 100 m away. Hecatesia whistles are believed to be
Sensory Mechanisms
97
involved in courtship. In contrast ultrasonic sounds in many nocturnal moths are thought to be defensive – production of such noise can be stimulated by squeaks of bats, and (particularly in distasteful tiger moths, Arctiidae) may help to warn bats of their true unpalatability. Acoustic signalling is important in both species recognition (and diagnosis) and individual recognition (so that in some psyllids and others, the two sexes ‘duet’ through audible or substrate-borne signals in order to meet and convey information on reproductive condidion), and changes in pattern have potential to provide useful information on processes of species formation and divergence. Use of sound is only one of the range of intricate behavioural procedures by which mates are sought. In Chap. 1, the topic of attractant scents, pheromones, was introduced, exemplified by those produced by some female moths and detected by males flying upwind along increasing concentration gradients to discover the source. Very subtle chemical differences can determine effectiveness, but pheromones generally are very specific in action. They can be defined as chemicals produced by one individual and which, when perceived by other individuals of the same species, elicit a specific response – usually behavioural but sometimes developmental. This definition is very broad and emphasises that (as for sound production) sex attraction is only one of a considerable range of responses that may occur. In particular insects, pheromones can parallel sounds in leading to aggregation, signalling alarm, marking territory or foraging trails, or others. They are most frequently discussed as moth sex attractants, as here, but rather few moth families have adopted them widely although a range of different hydrocarbons are involved. Many male moths also produce pheromones, with a less well-defined variety of influences – suggestions in particular taxa include inhibiting female escape behaviour, repelling other males, and deterring them from mating. Aggregation pheromones are best-known in some Coleoptera, particularly bark beetles in which pheromones can both increase mating frequencies and ensure that numbers of beetles are sufficient to condition the tree on which they occur for colonisation. As in many other aspects of insect behaviour the path from ‘stimulus’ to ‘outcome’ follows a strict sequence of steps, for sex pheromones being ‘pheromone emission – detection – flight/movement initiation – dispersal – progressive orientation toward pheromone source – landing – discovery of mate – mating’. Various intermediate levels, such as a courtship display, may be interposed, but each of these steps may be affected by the local environment. Pheromones facilitate mate detection from considerable distances, and enable genetic interchange to occur with mates that may be highly cryptic or with very low dispersal capability – female Strepsiptera within their leafhopper hosts (p. 88), and flightless moths (p. 18) are two such examples of insects that must depend on such mechanisms for discovery by a mate. Visual signals, necessarily effective mostly at quite close range – perhaps following pheromone attraction from further away, are predominant components in the mating rituals of many insects, with structural, postural and colour features associated with this. All are major contributors to insect variation, particularly within species: ‘bigger’ may indeed be ‘better’. Individual variation in size, ornamentation or colour, for example, may confer selective preference over less elaborate individuals
98
7 Insect Behaviour and Lifestyles
to potential mates. But, again, duality of function may be very common, and also strongly influence interactions between rival males for mates. Thus, the lengths of the eye-stalks of the bizarre-looking ‘stalk-eyed flies’ (Achias, Platystomatidae) are used in jousting during territorial fighting between males, but many visual features of insects have been described broadly as ‘visual propaganda’ in eliciting preference from females needing a mate. Sexual selection, whereby (usually) females respond to individual variation amongst males can involve almost any colour, pattern, size, shape, posture, signal or display. Any of these cannot be considered in isolation because any may also be involved in repelling or deterring rivals or natural enemies, in crypsis or by advertising distastefulness, and so on, as well as adaptations to local environmental extremes – such as by absorbing or reflecting heat. The place of mating, the daily activity regime, the feeding habit, and many other factors may influence how partners meet or are selected. Simplistically, many visual signals based on colour may not be effective at night, whereas signals transmitted by sound or light may operate well, or even better than by day. Movement by individuals may also be critical – a ‘moving target’ may be either more conspicuous (providing additional visual cues) or harder to locate (by confusing pheromone orientation to a stationary source). Visual attraction or signalling can also become a group effect. The dense swarms of some mayflies and small Diptera can comprise hundreds to many thousands of individuals ‘dancing’ together in clouds and creating a collective, very conspicuous entity. For mayflies, this behaviour may be a very effective enhancement for mating successfully during a very short adult life of only hours to a day or so. Males of some insects may establish ‘territory’ as an arena for finding mates, and chase off intruders as possible competitors. Some butterflies such as the Australian admiral (Vanessa itea) perch on bushes from which they fly rapidly to investigate other insects, or patrol short stretches of ground. A number of others may increase detectability through a behaviour known as ‘hilltopping’, whereby individuals congregate – perhaps moving from up to several kilometres around – on the summits of isolated hills or other topographic prominences, in the landscape. These prominences need not be large – low sand dunes only a few metres high may serve in flat desert country. Gathering in this way is thought to facilitate finding mates, and the habit is found in many insect species that are scattered widely and in low densities in the countryside and which, without such focus, might never encounter potential mates. Most insects found on hilltops are males, which may move there early in adult life and establish patrolling areas or territories. Hilltops are visited more briefly by females that, once they have mated, leave to reproduce in suitable places; males stay and may mate on multiple occasions. Different patterns of activity may be evident on a hilltop, so that particular butterfly species may be active characteristically in the morning, afternoon or early evening, but collectors value such sites as places where they may encounter rare or elusive butterflies that are otherwise very difficult to find. Some Australian butterflies are actually known only from a small number of ‘classic’ hilltops, and it is still speculative where they actually breed. Other insects also hilltop – representatives of many orders have been reported, but most species are of Diptera, Hymenoptera and (as above) Lepidoptera. Mount Moffatt in Queensland rises about 350 m above the
Sensory Mechanisms
99
surrounding lowland and well over a thousand species of insects have been collected hilltopping there. More than a hundred species of the parasitoid fly family Tachinidae alone were collected on only three spring days in 2002! In essence, hilltopping is one of several behaviourally founded mating strategies for insects, and that relies on finding mates far more efficiently through the behaviour than might be possible without it. Many other insects aggregate in seeking mates – males may gather to display, but massing of individuals also has defensive roles. Even in larvae, aggregations can constitute a well-founded antipredator strategy, as for the spitfire grubs noted on p. 84. They can also improve feeding efficiency, and many caterpillars move to a solitary existence only after the first one or more instars passed together. Some remain aggregated throughout life, such as the silken ‘nests’ of Ochrogaster and some other moths. Ochrogaster caterpillars (Thaumatopoeidae, ‘processionary caterpillars’) collectively spin large silken ‘bags’as shelters, and these often incorporate the stinging hairs of larval skins. Caterpillars move out to feed by night, often moving in a characteristic single file – hence their common name. A related scenario in mating strategies is that involving so-called ‘nuptial gifts’, whereby a male gains preference, or safety, by presenting a female with an edible gift that occupies her attention whilst mating occurs and, in some generalist predators, may prevent him from being devoured. Experiments on some scorpionflies, and others, have shown that the size and quality of the gift may confer preference in female choice – as indicating the most vigorous potential mate. Australian Harpobittacus scorpionflies are one such example, and capture a prey insect, such as a fly, to present to a potential mate. As with many behavioural traits, nuptial gift-giving has developed independently in a number of different insect groups, with counter to potential cannibalism seemingly involved in these. Some empidid flies wrap their gift (usually a smaller insect, such as another fly) in silk, so occupying the female for longer as she unwraps it, and a few indeed have gone even further and simply present an empty packet of silk, so saving the energy of actually ‘going out shopping’! The variety of nuptial gifts also includes nutritious ejaculate from spermatophores, again eaten by females (of many Orthoptera, for instance) after mating. It is also claimed that sexual cannibalism, renowned (although sometimes apocryphally, but in which the female devours the male during copulation) in many praying mantids is, in fact, the ‘ultimate nuptial gift’! Most of the foregoing refers predominantly to within-species interactions. Insect chemicals and other signals are often, however, directed mainly at other species. The broad chemical category of ‘kairomones’ are defined as chemicals produced by one species that are of benefit to other species receiving them – as in parasitoids and predators responding to volatile chemicals produced by hosts or prey, as search cues. Conversely, insects may produce ‘allomones’, repellent or deterrent to natural enemies. The specificity of many pheromones and kairomones has rendered them valuable in some aspects of managing insect pests (Chap. 14). A searching insect, whether seeking food or a mate, must usually operate on at least two different spatial scales – the wider environment in which its target occurs, such as a particular plant species in mixed vegetation, and the physically smaller resource within that environment, such as a caterpillar living on a leaf of that plant.
100
7 Insect Behaviour and Lifestyles
In behavioural terms, these constitute a hierarchy of, here, two sequential stages. They necessitate separate ‘decisions’ by the searcher, namely (1) which environment to select for investigation and how to find and reach it, and (2) how to search effectively within that environment for the resource, and how to exploit it. In many situations these are linked further – if searching proves fruitless or ‘expensive’ (in terms of low returns for high energy used in activity), the insect may then need to ‘decide’ when to give up searching within that particular environment patch and find another. In ecological parlance, this topic of ‘optimal foraging’ has been studied extensively in looking at how parasitoids and predators can be used against pests, and much of our understanding of how insects search has come from this applied need. Increasingly sophisticated mathematical models designed by ecologists have been used to try to predict the patterns of searching and when to give up searching – and explain what insects should do – and experimental studies help to determine what they do do. An insect’s individual success in reproducing will depend largely on the outcomes of searching for critical resources, drawing on an array of sensory cues to detect these, against a background of vast numbers of competing or confusing sensory cues presented in the same arena at the same time – and amongst which the insect must discriminate accurately. In general, volatile chemical cues operate at broader scales and, once within the suitable local environment, contact chemical and physical cues become more important. A parasitoid or predator (or a herbivore seeking a particular plant) may rely wholly on this sensory environment through innate behaviour, and perception of the particular stimuli of importance may be affected by factors such as humidity, wind and temperature, in addition to other sensory factors. The intensity of the cue may also be influential – as a deceptively simple example, a large plant may produce more volatile chemicals than a small one, and a single aphid may emit far smaller amounts of a specific chemical than the plant on which it feeds. Even tiny amounts of such chemicals can be highly effective communicants if they can be detected and discriminated by the receiving insect. Once close-range cues have been perceived, the searching behaviour of an insect for food or host often becomes more concentrated. Rather than wandering widely to traverse the widest possibly rewarding environment with least energy expenditure, many species then slow their walking speed and increase their rates of turning, so that the local environment is searched very effectively, and small prey or hosts in the vicinity found. Additional problems may arise – for example, a parasitoid needs to discriminate between available hosts and those that have already been attacked by a previous visitor: in some taxa, the first one deposits a ‘marking pheromone’ that conveys this message. Another strategy used by some parasitoids is, once in a favourable environment close to likely hosts, simply to lay their eggs and transfer the final host-finding phase to the emerging larvae. Some tachinid flies even depend on the eggs being eaten unharmed by potential plant-feeding hosts. However, perhaps the facets of behaviour that have attracted most interests (over several centuries) and intrigued people through their intricacy and integration, have been those exhibited by the ‘social insects’. In these, numerous individuals may live together over sequences of generations, and their persistence convey the impression
Social Existence
101
of effective coordination through which the wellbeing of the individual becomes secondary to that of the colony, sometimes of many thousand, even a million or more, individuals.
Social Existence Most insects live solitary lives. The two sexes must meet for mating, of course, and various forms of aggregation for feeding, aestivation, hibernation or defence are not uncommon. And, as with cicadas grouped and singing, the survival of individual insects may be enhanced through aggregations; ‘safety in numbers’ is quite common in traits such as escaping predators. However, with few exceptions, these insects still act alone and without any organised cooperation – even though an individual’s behaviour may be influenced by that of others, or their proximity. Many insects occur naturally in ‘clumps’ or large aggregations, but these do not always involve marked social interactions or mutual responses. They may simply be the outcome of local reproduction and limited dispersal, such as in aphids exploiting a short-lived food source (Fig. 7.2). However, more effective and integrated long-term cooperative behaviour is a notable feature of some insects and can evolve a true ‘social existence’; the transition from solitary to social living in insects has been called ‘one of the key events in evolution’. This seems to be anticipated in numerous cases of ‘presocial behaviour’ in which female insects, such as those of the common European earwig (Forficula auricularia, now widespread in Australia) show extended care of their offspring by brooding their eggs and young. Another example is of some giant water bugs, in which the eggs are carried on the
Fig. 7.2 ‘Group living’ is exemplified well by several groups of plant feeding Hemiptera; Homoptera, as exemplified by (a) aphids which can occur in enormous numbers in ‘colonies’ derived from a single female and (b) the wattle tick scale, Cryptes baccatus, on Acacia
102
7 Insect Behaviour and Lifestyles
Table 7.1 Insect behaviour: some of the categories or states reflecting the development of true social behaviour (Based on Michener 1969) Category Features Solitary Show none of the three traits of (a) cooperation between individuals in caring for offspring, (b) reproductive division of labour with non-reproductive individuals working on behalf of reproductive individuals, and (c) overlap of generations so that offspring coexist with parents for some period Adults care for their own offspring as larvae for some period Subsociala Communala,b Members of the same generation use the same nest without cooperation in brood care Quasisociala,b Members of the same generation use the same nest and cooperate in brood care Semisociala,b As quasisocial but also with reproductive division of labour, so that a ‘worker’ caste cares for young of a ‘reproductive’ caste Eusocial As in semisocial, but also with overlap in generations so that offspring assist parents a Collectively ‘Presocial’, all social stages before Eusocial b Collectively ‘Parasocial’, restricted to one generation
male’s back, rather than simply being abandoned when laid. Various formal ‘grades’ have been recognised, and the terminology of this evolutionary continuum can become confusing: some of these terms are noted in Table 7.1, simply to illustrate some of the subtleties that have been discerned by evolutionary biologists. But, even within an individual colony, of bees for example, the grade of sociality can change during development. More intimate associations, transcending generations, lead to true (eusocial) social life, and two of the insect orders adopting this are amongst the most abundant and ecologically predominant of all insects. Lesser known examples include the gall-forming thrips on Acacia (p. 121), and some social aphids, most of which have been studied intensively only during recent decades, and have helped considerably to augment understanding developed from classic work on termites and Hymenoptera. It is, of course, difficult to describe fully the processes affecting the transition from solitary to eusocial existence, particularly within the groups that are now wholly eusocial. The two predominant orders with eusocial existence are the Isoptera (termites, with all species eusocial) and Hymenoptera (all the ants, and some bees and wasps being eusocial). The last group have particular evolutionary interest in that in some groups of wasps we can postulate aspects of the development of social existence through various transitions from solitary to social life. Social wasps are almost wholly in the family Vespidae, many making the well-known ‘paper’or ‘carton’ nests (the small suspended nests of Polistes wasps are common under house eaves and in similar sites in eastern Australia), and within the family all stages from solitary to social existence have been postulated as the most complete transition series assembled. This sequence is outlined in Table 7.2, derived from an early account by a noted American authority on social insects, W.M. Wheeler. However, sociality has evolved independently on numerous occasions, so a generalised pathway may not be universal. Eusociality has developed at least five times amongst bees, for example, and even closely related species may have very different ways of life.
Social Existence
103
Table 7.2 The seven hypothetical stages leading to the development of social existence, based on wasps (After Wheeler 1923) 1. Female scatters eggs in usual living environment; in some cases eggs placed near larval food 2. Placement of eggs in part of environment that serves as food for larvae 3. Supplies eggs with protective covering (with either stage 1 or stage 2) 4. Remains with eggs or young larvae, and protects them 5. Deposits eggs in safe/selected/prepared situation (nest) with supply of food and made easily accessible to hatchlings, as ‘mass provisioning’ 6. Remains with eggs and young and protects and continuously feeds with prepared food (‘progressive provisioning’) (this is ‘subsocial behaviour’) 7. Progeny not only protected and fed by mother but also cooperate with her in rearing additional broods, so that parents and offspring live together (‘eusocial’)
And, even though only around 1% of aphid species are social, these may exhibit at least 17 independent origins of sociality, with the added intrigue that some of the species involved are clonal, so that individuals within a colony are genetically identical, despite their different appearance and roles. Different stages along this gradation are sometimes difficult to categorise – not least because almost any aggregation of individuals involves some form of interaction between individual insects. Aestivating Bogong moths (p. 211) assemble like roof tiles, for example, with overlapping bodies and wings whilst at rest, and respond to other individuals in forming these massed roosts. Eusociality is believed to have evolved from solitary insects in response to benefits of increased longevity, communal or cooperative endeavour, and defence or protection of offspring. For the term ‘eusocial’ to apply, three main criteria must be met. First, there is overlap of generations, with mothers living sufficiently long to overlap with their own offspring as adults – a very different scenario from most insects that die well before the next generation is completed and can have no contact with their matured offspring: Bogong moths are thereby not ‘eusocial’. Short-lived adult insects, such as many mayflies, would not survive until their eggs hatched, and most insects must simply abandon their progeny to their independent fate. A further implication of this temporal overlap is that there must be a relatively long-lived domicile, shared by members of the family. This may be a leaf gall, a termite mound or a wasp or ant nest, or more artificially, a bee hive. Usually, these are constructed, or based on modified natural features, by the insects; some social aphids in Japan live in hollow bamboo, for instance. For any such domicile, a key need is for it to be protected and, if necessary, defended, throughout its existence of up to many years. Second, there is cooperative care of offspring, so that individuals forage and supply food and husbandry for young that they have not themselves produced, but with which they share a gene pool through the same (or closely related) parents; altruism, whereby some individuals forgo reproduction and care for their kin, is a key element of eusociality. This immediately implies that individuals within the population operate in different ways, commonly as differentiated castes of different appearance and differing roles. Development of castes, commonly reproductives, workers and soldiers, constitutes a division of labour amongst individuals. Third, because not all
104
7 Insect Behaviour and Lifestyles
Table 7.3 Analyses of behaviour of primitive ants, Myrmecia, indicating primitive traits and more advanced traits, the latter shared with higher ants (After Wilson 1971) Primitive traits Multiple queens in many nests Eggs spherical and deposited separately on the nest floor Larvae fed directly with fragments of freshly caught insects Adults mainly nectarivorous, but catch insects as food for larvae Transport of an adult by another is rare No recruitment among workers to food sources, and no other apparent cooperation during foraging Alarm communication not advanced: described as ‘slow and inefficient’ Colony founding only partly secluded Nest queens, if deprived of workers, can revert to colony-founding behaviour Advanced traits Queen and worker castes very distinct from each other; intermediates rare Worker polymorphism in many species, with coexistence of two distinct worker subcastes Colonies moderately large, nests regular in construction Regurgitation occurs among adults and between adults and larvae Adults groom one another, and groom brood Trophic eggs laid by workers, fed to other workers and to queen Workers cover larvae with soil immediately before pupation, and assist eclosing adults in emerging from cocoons Nest odours exist, and territorial behaviour among colonies well-developed Workers carry dead nestmates out of nest
females reproduce they contribute unequally to the next generation. Some individuals provide numerous offspring, and others, none. Polymorphism is characteristic in caste formation, so that the different castes may be very different in appearance, as well as in their roles or functions within the colony. Within Hymenoptera, in particular, we can see several possible ways in which solitariness makes the transition to eusociality. Through the vespid wasps, above, and to a lesser extent with ants, thrips and aphids, we can gain clues through studying behavioural transitions in relation to their taxonomic position. The primitive bull ants (Myrmecia) were described by the founder of the science of sociobiology, E.O. Wilson, as ‘exceptionally rewarding subjects for the study of social biology’, not least because of their large size, that they are amongst the most primitive of all living ants, and they form rather small colonies (of up to a thousand or so individuals). These colonies are formed in chambers and galleries excavated in soil. The ants themselves are solitary roaming predators, and also collect floral nectar. Workers vary in size – large ones forage and the smaller ones tend brood within the colonies. Wilson tabulated the various ‘primitive’ and ‘advanced’ (that is, found in higher ants) traits of Myrmecia to illustrate the peculiarities of bull ant biology, and this variety (Table 7.3) helps to indicate the many subtleties of behaviour amongst social insects. Parallels could be suggested amongst termites, with the northern Australian Mastotermes darwiniensis generally accepted as the most primitive living termite. It is the sole surviving member of the once widespread family Mastotermitidae.
Social Existence
105
In contrast to Myrmecia, Mastotermes colonies can be vast, with more than a million individuals, and this termite is renowned for its wide range of foods – including such apparently unpalatable and unprepossessing items as billiard balls, ivory, and plastic cable coatings! However, because all living termites are eusocial, we cannot be sure how the habit arose. Their closest relatives are the cockroaches, and some workers have suggested that the two orders are not really distinct (p. 231). As in any other features of insect classification, various perspectives are possible on where such bounds occur. Some recent analyses imply that the wood-feeding Cryptocercus (wood roaches, not found in Australia) are the closest relatives of typical termites. Nevertheless, Cryptocercus aggregate in family groups, suggesting presocial behaviour, but no obvious caste differentiation occurs. Despite the advantages posed for social existence the process has apparently been reversed repeatedly, with evolutionary transitions also from eusociality to much looser social associations and, even, to regaining solitary ways of life. The three criteria noted above can form the basis for evolutionary pathways to gaining eusociality, as follows.The first generalised pathway simply extends the idea of maternal care through increased longevity enabling the parent to remain with her progeny until after they reach maturity. The second pathway involves several unrelated females establishing a communal nest, so that some division of labour becomes possible in rearing offspring. A third pathway in part overlaps these – but with increased longevity enabling related females (reproductive or not) to accumulate and participate in colony growth and maintenance, with progressive functional differentiation to achieve this. True social insects can form very large populations, commonly focused on a single nest (commonly termed ‘colony’, reflecting that it may have been initiated from a previous similar entity by a pioneering female or couple). They represent a major success in insect evolutionary history, their biology is the foundation of the discipline of sociobiology, and social insects have long captured human imagination for the intricacy of their interactions and perceived parallels with human societies as a ‘superorganism’ – with a large and fanciful accompanying literature on such themes. Biblical references to the industry of ants have parallels in many other cultures. The facts themselves are impressive enough, without recourse to imagination or fiction. Some ants form ‘supercolonies’ coordinated well beyond the confines of an individual nest – all the various colonies of the alien Argentine ant (Linepithema humile) around Melbourne appear to be genetically identical, for example. The constructions of some termites in northern Australia – such as the so-called ‘magnetic termite mounds’ of Amitermes or those of Nasutitermes (which can be more than 7 metres high) are the largest structures made by any Australian insects. They are probably also the most persistent insect features – some termite colonies are known to be more than 50 years old, and the founding females (queens) are perhaps the longest-lived insects, laying up to 2,000–3,000 eggs every day for several decades. Because many termite colonies can rapidly recruit other (previously non-functioning) reproductive individuals to take the place of the original queen, should anything happen to her, some entomologists have noted the colonies as ‘potentially immortal’. Such secondary reproductives can be generated as needed to sustain the colony.
106
7 Insect Behaviour and Lifestyles
Fig. 7.3 Termites (Isoptera), workers and soldier (dark head) of Coptotermes in wood
New termite colonies are formed typically following massed flights of reproductive adults (the only winged caste) from a parent colony, with the onset of flight induced by particular weather features – often on warm, calm evenings following rain (which may help soften the ground for new individuals to excavate a nest chamber, if they are soil-nesters). This is the only occasion on which the termites fly. On landing, the wings are shed, breaking off along a transverse line of weakness near the wing base, and females start to emit a pheromone, so attracting a mate. A mated male and female founder couple (in common parlance, the ‘king’ and ‘queen’) establish a nest site in soil or wood. The eggs develop into different castes, at first only wingless workers and soldiers (Fig. 7.3) but as the colony develops, also alates (winged individuals) with potential reproductive capacity. The effective division of labour – workers for foraging, construction, tending young, and soldiers for defence (aided by either or both of strong biting mandibles and sprayable acid secretions from specialised head glands) – fosters considerable competitive ability and persistence in termites. Mechanisms and pathways of caste differentiation during development can be complex. Even in the primitive Mastotermes, two rather different trajectories occur from the second instar onward. One proceeds through six instars to the first worker stage, which may then continue to moult indefinitely but without differentiating any further, and may give rise to defending soldiers.The second pathway goes through 11 instars to the winged adult. Details of such caste differentiation differ in higher termites, but intricate hormonal influences are widespread. The castes are very different in appearance in termites, but in some ants there is a continual gradation of ‘subcastes’, so that the workers may be small, medium-sized or large, with the largest individuals often operating as ‘soldiers’. In parallel, similar differences between individuals of some bees lead to presence of ‘megacephalic’ (‘big-headed’) insects.
Social Existence
107
In most social insects, the reproductive queens are the largest individuals, and queen termites can become swollen with eggs and reach several cm in length over their long life. The most advanced lineages of termites form much larger colonies and have greater levels of caste differentiation than more primitive lineages. Each taxonomic group of insects seems to have its own set of conditions that have affected the course of social evolution. Whereas all termites feed on cellulose (or lignocellulose), many eusocial Hymenoptera are predators (wasps, some ants) or herbivores (bees, some ants – sometimes as specialised nectar-feeders or seed–feeders, respectively). All Hymenoptera, of course, have complete metamorphosis, so that larvae must be tended by adults and do not participate in colony activities. Their food needs may differ. Polistes larvae, for instance, feed on insects (protein) brought into the nest by foraging adults, whereas adult wasps are nectar/liquid feeders. However, if nectar is not available, the adults can solicit food from the larvae to counter risk of starvation. As with termites, worker Hymenoptera (sterile females, rather than both sexes as in some termites) may change ‘duties’ as they age, often from more more nestbased activities to active foragers, as in honeybees. Ants are by far the most diverse group of eusocial Hymenoptera, whilst the habit has evolved on a number of different occasions amongst bees and wasps. In social Hymenoptera such as honeybees (by far the most intensively studied members of the order, reflecting their long importance to humanity), caste differentiation within females is predominantly through nutrition. Unfertilised eggs produce males. Females, from fertilised eggs, develop either to workers (if larvae are fed on nectar and pollen within their individual cells) or to queens (if given ‘royal jelly’, rich food derived from mandibular secretions). New colonies of honeybees are initiated as existing ones become crowded, and typically by a queen and an accompanying swarm of workers leaving and seeking a new site, such as a tree hollow. The famous communication dances of the bees include ability for scouting workers to communicate finding of a potential suitable new site. Several species of one major lineage of Acacia-galling thrips, Kladothrips, have developed distinctive soldier individuals which remain within the gall but have reduced wings and greatly enlarged ‘clublike’ fore legs. Both sexes may have this form, but females appear to reproduce far less than the colony foundress. Male thrips soldiers do breed, and the primary role of soldiers is to defend the gall against invaders, including other species of thrips. Galls are a scarce and vital resource, particularly in arid areas, and defence is important for the resident Kladothrips population. A habitat, whether a gall extending over several months or a termite mound or ant nest lasting for decades or more, supporting successive generations of an insect is obviously an important resource that must be protected in the interests of the population. The existence of specialised ‘soldiers’ and ‘swarming aggressors’ has captured the imagination of many science fiction writers, and others seeking to promote anthropomorphic parallels between insect societies and human ones, however fanciful these may be. Nevertheless, some dramatic stories of mass attacks by ‘killer bees’ and the like are not far removed from reality. Defence can be physical or, commonly,
108
7 Insect Behaviour and Lifestyles
Fig. 7.4 Bull ants, Myrmecia: (a) the ‘jumping jack’ ant, Myrmecia pilosula, the name applied to a group of species sharing the appearance of black body and contrasting yellow jaws; (b) a member of the M. gulosa group, with more uniform dark reddish appearance
chemical, through biting, spraying of toxins, stinging, or various combinations of these. Many species can mobilise a collective defence through use of pheromones that induce large numbers of individuals to aggregate for some ‘massed attack’. Honey bees, for example, release an alarm pheromone when the sting is extended and, if stinging, the barbed shaft is torn out so that the pheromone release is enhanced and attracts other individual worker bees to the vicinity. In many cases, biting or stinging effectively repels the victims, but the consequences vary with the susceptibility of the individual victim, and anaphylaxis and even death can result in some people. Six ant sting-related deaths, for example, were found over a survey from 1980 to 1999 in Australia, five from Tasmania and the other from New South Wales. In most such tragedies the precise ant responsible was not identified, but the most likely candidates are two bullants, Myrmecia pilosula (jumping jack: Fig. 7.4) and
Social Existence
109
M. pyriformis, both noted for their allergenic stings. Most victims died within 20 min of a single sting: the wisdom of carrying injectable adrenaline in the field if suspected to be allergic to the animals is obvious. Numbers of deaths from bee or wasp stings are apparently low, but records suggest that one or more fatalities occur in Australia in most years. One of the widely suggested benefits of shelter, mound or burrow construction by social insects is that these can ‘buffer’ the insects against environmental fluctuations in ways not available to solitary, exposed insects. Many other insects are associated with eusocial insects, or their domiciles, as predictable and secure places to live and in which food – either the host insects themselves, their food or products, or other species – is assured, together with refuge from climatic extremes. Physical structures such as termite mounds and wasp nests provide substantial barriers against outside seasonal climate variations – their inhabitants are protected from precipitation, and humidity may also be controlled. The so-called ‘magnetic termites’ (Amitermes) of northern Australia orientate the large above-ground nest mounds in a north-south direction, with the broad sides facing east and west – this appearance is remarkably constant in these species and is believed to prevent overheating of the colony, because the hot mid-day sun strikes only the narrowest profile of the nest. Changes in termite metabolism and activity may also be involved in temperature regulation within the colonies. The internal structure of different termite mounds may be very characteristic. Termite mounds, for example, may have air-circulation and temperature control mechanisms that could be envied by many urban tower-block apartment-dwellers. Ventilation systems in mounds may depend on open vents – some with raised rims for outflow, and other with flush rims (inflow) to maintain unidirectional air currents. Others have closed systems, sometimes aided by workers carrying water upward, perhaps from considerable depths, so that evaporative cooling occurs for the circulating air. Such influences on local microclimates are held widely to be part of the rationale for success of eusocial insects, in providing stability – and this protection can then also apply to other denizens of the colonies. These include a host of small beetles, flies, silverfish and others, many of which are never found elsewhere except in the nests of particular ants or termites, as obligatory ‘termitophiles’ or ‘myrmecophiles’, and for which the biology may be correspondingly complex. Those living with ants are, perhaps, the better studied. Many are very odd-looking insects, with dramatic physical or chemical mimetic resemblances to their hosts – many, indeed, are called ‘myrmecomorphic’ because of their shape resemblance to ants, and ant mimicry is thought to have evolved at least 70 times in different arthropod lineages!. Physical resemblances are commonly enhanced by behaviour, whereby mimics ‘act’ like their hosts and move in a similar way, at similar times – often sufficiently well to confuse entomologists seeking them. However, many other myrmecophiles do not mimic ants in this way but, commonly through chemical characters, are able to have close and intricate relationships with them. Perhaps the best and most general interpretation of such inhabitants of social insect colonies is to think of them as ‘guests’ of the ants or termites (and smaller numbers with bees and wasps) with which they regularly co-occur.
110
7 Insect Behaviour and Lifestyles
Fig. 7.5 The common imperial blue butterfly, Jalmenus evagoras (Lycaenidae) (see Fig. 6.7) is associated with Acacia, and caterpillars and pupae are found in groups. They are both tended by large numbers of ants (most commonly Iridomyrmex) attracted by the sugar-rich secretions of the caterpillars. Unusually, both caterpillars and pupae produce sounds, one suggested role of these being to attract ants to help protect the Jalmenus from natural enemies. Towards emergence, pupae produce a pheromone scent, attracting male butterflies to female pupae, where they compete for mating as soon as the females emerge
Many myrmecophiles illustrate the ecological subtleties of such associations. Many are ‘nest parasites’ that obtain their food by feeding on ant brood or from worker ants tending that brood. One or more such species probably occur with almost every ant species. However, more broadly, three kinds of association are recognised widely as (1) symphilic, found only with single ant species, on which they depend wholly for food; (2) synoeketes, tolerated indifferently by their host(s) and feeding on nest materials and wastes rather than on the host; and (3) synechthrans, occasional invaders not normally living with social insects, but intruding into nests for food. The last two categories are sometimes called ‘non-integrated’, because their biology is not fully interlinked with that of the host. Much of the attention to myrmecophiles in Australia has arisen from butterfly enthusiasts, but numerous tiny rove beetles, flies, silverfish and others also occur. The impetus from butterfly hobbyists results from the biology of many scarce (and desirable to collectors) species of our largest butterfly family, the ‘blues’ (Lycaenidae). Many have intricate specific relationships with ants, which have aroused interest from entomologists, evolutionary biologists and conservationists alike (Fig. 7.5). Two examples, one representing a genus, Acrodipsas, with some species wholly feeding on ant brood, but with the common epithet of ‘ant blues’ marking their obligate need for ants, and another in which the association, although obligatory, is very different in intensity, are noted elsewhere (pp. 73, 205) and the likely mutualistic relationships that benefit both parties noted. A remarkable – and perhaps more one-sided – third case is of the unusual tropical ‘moth butterfly’ (Liphyra brassolis) which, like Acrodipsas myrmecophila, feeds on ant brood but in this case of the green
Further Reading
111
tree ant, Oecophylla smaragdina, on trees. The caterpillars have been described as ‘tanklike’ – with a broad tough ‘shell’ extending around the body and protecting them from ant attack. To feed, the caterpillar seizes an ant larva with its jaws, drags it under its carapace, and eats it beneath this protective rim. Altogether, about 700 species of Lycaenidae throughout the world have been reported to have associations with ants in some way, and Australia’s unusually diverse and complex ant fauna has assuredly fostered diversity of myrmecophiles of many kinds as a largely unseen component of the considerable variety of insect life. Most are not yet as well documented as the butterfly caterpillars, but may prove to exhibit parallel high variety in their interactions and levels of dependency.
Further Reading Drosopoulos S, Claridge MF (2006) Insect sounds and communication. Physiology, behaviour, ecology and evolution. CRC Press, Gainesville Eastwood R, Fraser AM (1999) Associations between lycaenid butterflies and ants in Australia. Aust J Ecol 24:503–537 (survey of diversity and associations involved) Michener CD (1969) Comparative social behavior of bees. Annu Rev Entomol 14:299–342 Moulds MS (1990) Australian cicadas. New South Wales Univerity Press, Kensington (includes section on sound production and ecology in a broad overview, with illustrations and keys to all taxa then known) Percy DM, Taylor GS, Kennedy M (2006) Psyllid communication: acoustic diversity, mate recognition and phylogenetic signal. Inver Syst 20:431–445 (significance of sound production in some Australian psyllids, and its contribution to taxonomic analysis) Wheeler WM (1923) Social life among the insects. Harcourt, Brace & Co, New York Wilson EO (1971) The insect societies. Belknap Press of Harvard University Press, Cambridge (the first major modern compendium of insect social existence, its origins and significance in evolution)
Chapter 8
Insect Communities
Introduction: Living Together Whether in water or on land, any environment provides suites of resources that insects may exploit. Those resources most commonly vary across space and time (both seasonally and in the longer term of more permanent change), as does the suitability of the host environment, and differ markedly in kind, amount, and how they can be reached and used. From an entomological perspective, diversity and amount of resources (including plants and animals as consumables, with suggestions that limits to the variety available may be imposed by climate and space) is associated with insect diversity. Simplistically, more resources can support more insect taxa and individuals, and can be associated with greater levels of specialisation as the various species partition their uses of these. Indeed, many scientists would support a stronger causative link, along the lines of ‘resource diversity begets insect diversity’. This chapter is about the structure of insect communities (the variety of species found together in a habitat and the interactions that occur between them) and how this is influenced by features of the local environment, and develops the themes introduced in the last two chapters. Many observations have led to widely-held generalisations that help to give a foundation for discussion. First, resource-rich and complex (productive) environments often harbour large numbers of insect species, and the converse. Some insect faunas in extreme ‘harsh’ environments are genuinely sparse – one of the best documented such areas of Australia is Macquarie Island, far to the south of the mainland, remote and with an extreme climate. Only nine orders of insects have been recorded there, most of them with very few resident species; only three families of beetles, four of Lepidoptera and a single species of Hymenoptera (probably a parasitoid of kelp flies) occur there with the total insect fauna (in 2006) only 97 species, some of which appear to be casual occurrences and not yet confirmed as residents. More than half (53 species) are parasitic lice, none of them endemic and 48 associated with birds – such preponderance of lice in any regional insect fauna is unusual, possibly unique.
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_8, © Springer Science+Business Media B.V. 2011
113
114
8 Insect Communities
Fig. 8.1 Acacia species are attacked by Uromycladium fungus, producing woody galls, in turn habitats for invading insects including beetles, caterpillars and a wide variety of their predators and parasitoids, comprising well-defined successional communities in these individual galls. Many hundreds of galls of varying ages may occur on one tree
Second, the spectrum of natural environments inhabited by insects each harbours resident species that are characteristic of that environment and may be largely or wholly restricted to it, as well as more widespread species. This, third, leads to parallels that each such community includes insects that are in some way ‘specialists’ to the conditions provided, or are ‘core’ members of communities typical of that environment, and more generalised (and, so, less characteristic or restricted) species. And, fourth, the overall size of the environment may be influential in itself – larger areas can, at least in principle, hold more species, and more resources for the pool of inhabitants to utilise. Any insect species, of course, may itself be a resource for others. However, because most insects are small, diversity can be high in even rather small habitats – a single cattle dung pad (p. 74) can support a successional community of surprising complexity, and the microcosm of a single fungus gall on a wattle (Fig. 8.1) can yield an array of different insects. The resources in any place, of course, include other insects, which may share the same foods and so potentially provide inter-species competition and other interactions. High species richness within a community or feeding guild often implies specialisations to avoid such effects, and sufficient stability in which the various species have in some way come to terms with each other and can coexist. All these habitats and resources are scattered, as a mosaic, in the wider landscape (or waterscape), and are to some extent islands suitable for occupancy in a wider and generally inhospitable arena (sometimes referred to as the ‘matrix’), so that ecology has drawn heavily on parallels with ‘real’ islands for understanding. A fungus gall may be just as remote, and as important to survival of a small wandering insect, as Pitcairn Island was to the Bounty mutineers or Juan Fernandez to the marooned Alexander Selkirk.
Introduction: Living Together
115
Modern island biogeography has produced a number of working paradigms relating diversity and area, in part related also to distance from possible source areas and the size of the island ‘target’. Small, highly scattered islands far from the homeless insects are more difficult to find (other than by chance) than large, close together, nearby islands. Some insects do indeed disperse over large distances of uninhabitable terrain, equivalent to oceans – but perhaps not as extreme in that an insect over land may at least be able to settle (and perhaps feed) at intervals. Even remote ‘real islands’ (particularly when vegetated or in warmer climes) develop complex insect faunas over time, as suitable resources develop. Every insect ‘community’ differs in detail from every other one, but those found in any characteristic environment can be defined in part by equally characteristic core species that commonly do not occur elsewhere. Thus, an ‘alpine environment’ in south eastern Australia can support many insects, including some butterflies, moths, grasshoppers and others, not usually found elsewhere and so specialised to live in the high elevation, cool climates with the restricted vegetation (including treeless areas) present. However, because alpine environments intergrade with others along elevational gradients, characterising them absolutely is very difficult and uncertain – as is the case also with almost any habitat and the commonly mobile insects that live there – so that defining the core species, and preparing an inventory (or species list, to indicate diversity, below), are exercises that may need careful thought and qualification. A categorising term such as ‘water beetles’ initially seems clear and to convey an unambiguous meaning. However, it is almost impossible to compare the numbers of water beetles across different published surveys and different habitats because this descriptive term transcends taxonomic boundaries, and is applied by different authorities in very different ways. Some beetles spend their whole existence in water; others have aquatic larvae and adults that roam far from water; others occupy the riparian fringe, such as vegetation or wet sediments along shorelines, and so on. In short, water beetles have arisen from terrestrial forebears on many independent occasions, so do not represent any single taxonomic group, and collectively have numerous different associations with water bodies, with many beetles more properly ‘semiaquatic’. Different coleopterists can easily opt to include or exclude particular taxa or associations from such an embracing descriptor, and their advice influence delimitation by ecologists or planners undertaking surveys in any particular context. Nevertheless, overcoming such interpretative problems is important in gaining consistency for describing the patterns of insect distribution, and how these may be influenced within communities by both the resources available and the other species present. A related term, ‘assemblage’, is used mainly in a taxonomic sense rather than to imply any linkages – so that the ‘assemblage of Formicidae’, the ‘assemblage of Lepidoptera’, or the ‘assemblage of Hemiptera’ simply refers to the variety of ants, moths and butterflies, and bugs present. It may be qualified by feeding guild or other habitat limitation, so ‘the assemblage of predatory Coleoptera’ or ‘the assemblage of gall-forming Hemiptera’ gives us functionally restricted categories for finer comparisons or evaluations. Assemblages are, of course, a core component of wider communities, and documenting assemblages helps us to appreciate the
116
8 Insect Communities
foundations of many insect communities. Indeed, most surveys of insects have focused on one or few taxonomic groups, simply because entomologists almost always themselves specialise at this level – many entomologists have passed their whole working careers studying a single family of insects, progressively gaining expertise in recognising the species involved, but associated in practice with widespread lack of capability to enumerate entire insect communities.
Richness and Variety Whichever focus is adopted, in communications about insect communities or assemblages two parameters are used very widely – the ‘richness’ (number of species present), and its composition (the relative abundance of those species), together helping to define the diversity. Both terms help to describe any such entity, and the similarities and differences are fundamental tools in comparing different communities and assemblages. The aim of many ecological studies, and in which the diversity of insects renders them vital components, is to characterise and define the community in structural and functional terms, and to determine how it has been changed – or may be changed in the future – by human activities. The ecological variety and sensitivity of insects give them massive value in such evaluations both in defining the baseline of a ‘healthy community’ and in monitoring changes that may be imposed, or occur through processes such as climate change or natural succession. Understanding communities brings to the fore the functional roles of the species present in sustaining ecological processes and functions. In many contexts, to do this effectively necessitates (or can benefit markedly from) recognising and identifying the insects present and knowing (or reasonably inferring) their likely feeding habits and interactions. This is not always easy. Conversely, it IS easy to overgeneralise about many groups of insects as ‘herbivores’, ‘predators’ or other guild without appreciating that many exceptions may occur and hamper functional interpretations. Caterpillars are typically chewing herbivores, but we have already met (in the small ant-blue butterfly, p. 73) a notable exception that eats ant larvae. The large conspicuous colonies of scale insects on wattle trees (Fig. 7.3b, p. 101) are also attacked by caterpillars that have become specialist predators on these insects and hide underneath the colonies. Both these categories have close plant-feeding relatives. Also for caterpillars, most species of Lepidoptera are terrestrial but a number of pyralid moths have become aquatic, and their caterpillars feed on water plants in ponds or streams. Likewise, the ecological variety of some of the larger beetle families ensures that a variety of feeding habits may occur – even the presumed quintessentially predatory ladybirds (Coccinellidae) include many plant-feeding taxa. Another concern is that our knowledge of many insects does not allow us to consider this variety properly, in describing or comparing communities. Many workers on insect diversity, quite understandably because it is reasonably easy and reliable, have analysed insect samples only to the level of family, and have allocated all members of that family the same guild role in their functional interpretations.
Richness and Variety
117
As noted repeatedly above, many large families are very varied, and some important perceptions may be masked by this approach. Only identification to lower levels, preferably genus level or, in many taxa, even to species or equivalent fine detail, can reveal these subtleties properly. Thus the abundant small rove beetles (Staphylinidae, one of the largest beetle families in Australia) common in ground litter are often categorised wholly as predators. Many, including the familiar red and black ‘devil’s coachhorse’ (Creophilus erythrocephalus, at about 17–20 mm long one of the largest common rove beetle species) are indeed predators, but many other staphylinids are variously detritivores, fungus-feeders, carrion feeders, with a few even parasites or restricted to living in the nests of ants or termites; a few extend to shoreline habitats and feed on algae there. Fine level taxonomic interpretation allows use of at least some of this ecological background. Bulking taxa together uncritically can easily obsure it. Recognition of this functional variety in insects has led to designation of ‘functional groups’ within assemblages, to aid ecological interpretation, perhaps nowhere more successfully and importantly than for Australian ants. All ants belong to a single family of Hymenoptera (Formicidae) which can be divided into an array of subfamilies, ten of which occur in Australia. One includes only the primitive ‘dinosaur ant’ (Nothomyrmecia macrops), currently known from only a small area of inland South Australia, and the more widespread bullants, collectively regarded as amongst the most ancestral ant forms. Most other, more advanced, subfamilies are more diverse and widely distributed. Collectively, they include just over 100 genera in Australia. Ants are incredibly varied in their ecology and responses to environmental changes and (reflecting their social existence, p. 104) are present in enormous numbers. They can also be captured easily by simple methods, with many surveys relying predominantly on pitfall traps, so comparable samples can be obtained with little effort. Single-site surveys of ants by pitfall traps in the Northern Territory savannas, the Victorian mallee, eucalypt woodlands in the southeast or any of a substantial array of different ecosystems yield many species – with published figures of 40–60 species of ants in small plots of a hectare or so not unusual – together with considerable ecological variety and distributional differences across the continent. Thus, many of the 1,500 or so ant species living in the tropical monsoon regions are not found in the Bassian region, and many Bassian species are absent from the north. But a unifying problem across these regions is that a high proportion of the species have not been named, even though they are referable to clearly recognisable named genera. Many of the genera, and major species groups within some larger ones, exhibit consistent ecological trends and responses, so that some ‘functional analysis’ is valid at this level of taxonomic interpretation, at which a number of functional groups can be recognised (Table 8.1). An intriguing analogy has been to illustrate possible parallels between ant groups and plant life forms – based on the analogy of ant nests providing foci much as do plant roots, and from which foraging can occur in various ways, with numerous and varied interactions with neighbours in the surrounding environment. The relative representation of the various functional groups helps to indicate the ecological ‘condition’ of the site sampled, and the composition can change markedly as site
118
8 Insect Communities
Table 8.1 Functional groups of ants designated in Australia. Groups reflect the immense ecological diversity of ants, and changes in their balance and richness (at tribe, genus or species-group levels) help to indicate responses to environmental change, stress and disturbance without need to analyse samples to species level. Since its original formulation by Greenslade (1978), the approach has been refined and modified considerably (e.g. Andersen 1990) Functional group Characteristics and example taxa Dominant Dolichoderinae Competitively superior, at peak of dominance hierarchies in productive environments. Most common on ground in arid regions: Iridomyrmex, Anonychomyrma, Froggattella. Generalised Myrmicinae Broad distribution in relation to disturbance and environmental stress, often common: Crematogaster, Pheidole, Monomorium. Opportunists Largely unspecialised, with wide habitat spectra; abundant in disturbed habitats where other more dominant groups are limited: Rhytidoponera. Subordinate Camponotini Co-occur with Dominant Dolochoderinae but are competitively inferior: attempt to avoid conflict with that group, many by becoming nocturnal: Camponotus, Polyrhachis. Hot climate specialists Ants adapted to hottest environments: common in arid zones: Melophorus, Meranoplus. Cold climate specialists Ants adapted to cool temperate climates, most common in wet/ cool forests: Monomorium, Notoncus. Tropical climate specialists Many of the genera found in tropical rainforest, some behaviourally dominant: Oecophylla, Mayriella, Tetraponera. Cryptic species Largely soil/deep litter nesters, mostly in low abundance – small and little interaction with other ants: Amblyopone, Hypoponera, Ponera, Stigmacros. Specialist predators Little interaction with other ant groups, specialised, sometimes solitary predatory behaviour: Cerapachys, Myrmecia, Odontomachus.
conditions are altered. Many ecologists have attempted to use ants in this way as indicators of environmental conditions and changes, and to monitor the progress of exercises such as rehabilitation of sites after mining. However, poor knowledge of some important ecological groups amongst Australia’s estimated 6,500 ant species cautions against complacency. Surprises remain – recent surveys in northern Australia have revealed a diverse subterranean ant fauna, to complement the above-ground ants on which most of the inferences on functional groups have been based. Many of the species involved are strongly adapted to underground life, being small, shortlegged and lacking eyes. Several had never been found in above-ground ant surveys in the same areas near Darwin over some 20 years, and others were far more abundant underground than elsewhere. Investigation of this seemingly widespread and specialised ant fauna may lead to further increase in Australia’s ant numbers. Sound knowledge and understanding of the insects clearly underpins any such applications, just as recognition of the ‘core’ insect species of any community helps us to define that community, but – as the above example shows – it may be possible to ‘miss’ whole parts of assemblages during samples or study. Many naturalists are
Evolutionary Radiations
119
able to deduce (or infer) information of much value from their knowledge of particular species or species groups, where they occur, and the species with which they associate or commonly co-occur. At one functional level, species associations may constitute ‘modules’ that are in some way mutually interdependent, or constitute key resources. Thus, a specific plant may not thrive without its specific insect pollinator, and the converse; and many lycaenid butterflies (blues, coppers) need both a specific host ant (which also needs the butterfly) and larval foodplant, as well as adult nectar sources. With increased variety of resources, increased variety of insects may be supported, but the full variety of interactions between insects in any given environment is very difficult to determine. Simply because we do not see that any given insect is not participating in interactions with other species there does not mean that it does not interact with them: in theory, of course, every species present could interact with every other one to give an enormous possible total. Many of the theoretically possible interactions seem highly improbable and unlikely to occur, but it can be just as hard to assert this confidently as to prove an interaction or regular association.
Evolutionary Radiations Consider, for example the enormous number of herbivorous insects that occur on Eucalyptus or Acacia – but only rarely on both of these predominant plant genera. Together, eucalypts and wattles comprise well over 13% of Australia’s vascular plant species, each with several hundred species and Acacia with close to 1,000 species named. In parallel with this plant diversification, some groups of plant-feeding insects have also proliferated, with their current richness the outcome of long periods of progressive co-evolution and interactions with their plant hosts. In some taxa, we can trace the insect transitions with host diversification, as ‘radiations’ to progressively specialised and host-specific taxa. The total insect communities of most of these trees and shrubs have not been characterised fully, but the insects present commonly comprise suites of very closely related species, often very similar in appearance and difficult to differentiate simply or consistently, on different hosts. Collectively, they are major components of Australia’s insect diversity; with their associated (and, in some case, specific) insect predators and parasitoids, the communities have become both species-rich, and functionally intricate, as well as largely host-plant specific. Much of the focus on these insects has been on assemblages, sometimes accompanied by studies on individual species of economic significance, most commonly on the members of plant-feeding insect families or orders. Many published records exist of the form ‘x species of beetles [read: thrips, scale insects, moths, etc] found on Eucalyptus species A’, but the interactions between different groups of herbivores, or even between different co-occurring species within any group, are largely undocumented. Some of the insect groups involved in these radiations have been studied extensively because they are (or have potential to become) pests on Eucalyptus trees used
120
8 Insect Communities
as forestry crops; still others have been accorded the converse role as ‘beneficial’ because they are valued as biological control agents able to suppress pest weeds – such as some bipinnate acacias in South Africa. Those acacias were imported from Australia as the basis of a tan bark industry, but some have spread extensively into natural environments, where their competitive aggression is of concern. And, third, many of the assemblages have massive evolutionary interest both intrinsically and as important components of Australia’s endemic insect diversity. Taxonomic studies and parallel ecological studies have combined in some instances to elucidate the diversity they represent, as in the gall-forming coccids noted on p. 50. Another group of plant bugs, the jumping plant lice and lerp insects (Homoptera, Psylloidea), have proliferated on both these plant genera, but within different subfamilies on each. The better-known group are the lerp insects (subfamily Spondyliaspinae) on Eucalyptus and its close allies. They gain their common name from the aboriginal term ‘lerp’, which refers to the sugary scales formed on gum leaves by the nymphs of many species, and which have been traditionally valued as food. Their elaborate form can be very characteristic for recognition of some genera and species (Fig. 8.2). Several species may occur together on a tree species, ranging from polyphagous psyllids to others that are much more specific. They can occur in very large numbers, sometimes undergo outbreaks, and heavy infestations can lead to senescence, defoliation and dieback of trees. Their ecology and control is thus a major concern in forestry. Less well-known, but perhaps at least equally diverse – with around 100 known species yet to be described, a genus of Psyllinae (Acizzia) has proliferated on Acacia. These Psyllinae do not form lerps, but can also occur in enormous numbers on young foliage and developing seed pods (Fig. 8.3). Most other insect groups do not demonstrate such spectacular parallels on both these host genera, but many have developed intricate relationships with one or other of them, and these can manifest as intimate physiologically based associations such as gall-forming. Galls are outgrowths of the plant induced by insect attack (in this context – some other invertebrates and fungi also cause galls) to constitute very characteristic and site-specific structures, whose appearance is often diagnostic for the insect causative agent (p. 122). Most species of chalcidoid wasps are parasitoids, but some have moved away from this habit to become plant-feeders – possibly initially though parasitising plant-feeding hosts and subsequently feeding also on plant material. Some are now obligate gall-formers. The large family Pteromalidae includes 27 subfamilies in Australia, and several of these are wholly or largely associated with plant galls. Most genera of Ormocerinae, for example, have this life style and are noted here because one of these (Trichilogaster) forms stem and flower galls on Acacia. In high infestations almost all flower buds on a tree can be attacked, effectively preventing any seed recruitment (Fig. 8.4). Several species of this wasp have been deployed to help control pest acacias, as above, in South Africa. Trichilogaster is a rather small genus. Eight of the nine described species are Australian endemics, and one is known from Arabia; the host range documented so far comprises 12 species of Acacia and one of the related genus Paraserianthes. However, the extent of several parallel insect radiations is only gradually becoming clear. A more general scenario is that focused investigation of any such assemblage reveals far greater taxonomic and ecological diversity than supposed initially.
Evolutionary Radiations
121
Fig. 8.2 Psyllidae, sap-sucking bugs, gain the name ‘lerp insects’ from the sugary covering excreted by nymphs and under which each develops. The appearance is characteristic for different taxa: (a) this shell-like form is characteristic of the genus Spondyliaspis, on eucalypts: (b) Glycaspis lerps, also on eucalypts, are often tended by small ants, mainly Iridomyrmex and related taxa
Thrips (Thysanoptera) are common gall-formers and inhabitants on Acacia foliage, with around 250 species of the subfamily Phlaeothripinae now known to do so. In contrast, no species of this group is associated with eucalypt foliage. Intriguingly,
122
8 Insect Communities
Fig. 8.3 Psyllidae can also occur in vast numbers on Acacia: many species of Acizzia exploit soft flush foliage. This single immature seed pod on Acacia dealbata illustrates the density that can occur, with eggs (yellow, darkening as they near hatching), adults and emerging nymphs all present
Fig. 8.4 Flower buds of some Acacia species are attacked by wasps (Trichilogaster, Pteromalidae) to produce small spherical galls, often in very large numbers so that flowering is effectively precluded. The galls themselves are invaded by an array of other insects including parasitoids of Trichilogaster
Evolutionary Radiations
123
molecular studies reveal that all these thrips represent a single lineage; in other words, thrips have successfully invaded Acacia only once during their evolutionary history. Their biology is remarkably diverse, with four major ecological categories recognized: these thrips have been claimed to ‘represent a microcosm or model for the diversity found among all phytophagous insects’, so a little more discussion is warranted to enlarge on this, and to help understand how diversity within this group incorporates variety in how critical resources are used. They mirror some of the patterns of host plant exploitation that have generated equivalent richness in many other groups of insects on plants. As background, the galls are all formed on phyllodes, the broadened leaf petioles that constitute the apparent foliage of most species of Acacia, and many of the thrips are colonial so that a single gall can harbour numerous individuals after foundation by a female or a pair of adults; in many, this has developed into a true social existence (p. 102) with evolution of functionally differentiated ‘soldier’ castes with massively enlarged fore legs used in fighting. Most of the Acacia species affected by thrips occur in the dryer regions of Australia, where it could be argued that living enclosed in the foliage (rather than fully exposed) may confer some protection from desiccation and extreme temperatures. Gall formation is stimulated by feeding activity, with the rapid proliferation of phyllode tissues around a feeding site forming a pouch that envelops the insect. But ‘direct gall-formers’ are only one of the four ecological groupings mentioned above. Members of the second category are ‘phyllodegluers’, that also create an enclosed domicile – but by sticking two or more phyllodes together with a sticky secretion from their rear end. Both these categories produce structures that can be highly diagnostic in appearances for the taxa involved. The third group are called ‘opportunists’ that inhabit galls and phyllode chambers as well as a variety of other retreats (such as bark crevices) on the acacias. Their collective host range thereby extends to Acacia species that do not host the first two groupings, but some are most commonly found in association with them. This opportunism contrasts with the last category, ‘exploiters’, that co-habit with gall-formers or phyllode-gluers and range from species that kill or displace these original inhabitants to those that apparently coexist peacefully with them. Each of these four categories includes numerous variations on the general theme, but illustrate the complexity that can occur within a single insect group exploiting a single plant genus, and fostered by a considerable variety of behavioural (p. 103) and ecological specialisations. This diversity apparently arose from a single ecological opportunity – an ancestral thrips colonising Acacia – but some other radiations appear to be more diverse in origins. In the last chapter we noted ‘mallee moths’ as a spectacular diversification within a moth family (Oecophoridae) linked strongly with change from the most common Lepidoptera feeding habit of caterpillars eating living foliage, to capability to feed on dead foliage, particularly of Myrtaceae. At least 5,000 species of Oecophoridae, perhaps several hundred more, have evolved in Australia to represent around a quarter of the Lepidoptera in the country and, to emphasise the diversity even more, this number is around a quarter of all insect species found in the United Kingdom. Dead eucalypt leaves are regarded as poor food with little nutritive value, and are exploited by rather few groups of insects. Caterpillars of all oecophorid species live in some kind of shelter, most commonly formed from leaves sewn together with
124
8 Insect Communities
Table 8.2 Examples of some major radiations of phytophagous insects on Acacia and Eucalyptus in Australia Taxon Comments HEMIPTERA Psyllidae See text: separate radiations on Eucalyptus (ca 250 species) and Acacia (at least 100 species: one of few insect groups that demonstrates equivalent parallels on these two predominant plant genera). Most feed on foliage, some on flush growth. Eurymelidae A largely endemic family of, often, strikingly coloured hoppers, comprising around 100 species on Eucalyptus; in need of further study. Membracidae Uncertainty over taxonomic complexity of species on Acacia: ‘Sextius virescens’ is common and very variable. Eriococcidae Gall-forming coccoids have diversified on Eucalyptus. THYSANOPTERA Major radiation of gall-forming thrips and associated forms on Acacia (see text). COLEOPTERA Chrysomelidae Scarabaeidae Curculionidae LEPIDOPTERA Oecophoridae HYMENOPTERA Pteromalidae Colletidae
Substantial radiations of paropsine leaf beetles on both plant genera; defoliators as larvae and adults. Many species of Liparetretus and others (such as Christmas beetles, Anoplogathus) found on Eucalyptus. Radiation of seed-feeding weevils (Melanterius) on Acacia. Unusually diverse radiation including, perhaps, several thousand species that mainly feed on dead eucalypt foliage. Gall-forming species of Trichilogaster sometimes abundant on Acacia. Substantial radiation of these native bee pollinators on Eucalyptus.
silk, and are probably buffered from desiccation in the leaf litter. The life cycle of many species starts from eggs laid on the tree canopy and young caterpillars eating living foliage. Later, by cutting leaves the larvae fall to the ground. In contrast to the acacia-galling thrips, feeding on dead leaves seems to have arisen independently in each of the main groups of genera in the family, so the habit has multiple origins. Examples of diverse assemblages and radiations of insects on Acacia or Eucalyptus could be multiplied – some are noted in Table 8.2 – and the variety transcends both taxonomic groups and feeding habits. None exists in isolation and, other than in some localised environments, neither do the host tree genera. Each assemblage must coexist with others, and with the many less well-defined insect associations present on the same plants and in the same sites. Consider also that we have so far discussed only the primary herbivorous insects and those immediately derived from them. The psyllids, for example, are attacked by many small parasitoid wasps – for the time being mostly placed in the genus Psyllaephagus (‘psyllaeating’, with more than 100 described species) in the family Encyrtidae. Many of these tiny wasps are thought to be quite host-specific, but their full host ranges are unknown. Likewise, if we bring samples of well-developed Trichilogaster galls from Acacia inside and rear out their inhabitants, a considerable variety of
Assessing Richness
125
small wasps and other insects are likely to emerge from them – as well as some Trichilogaster. Samples of leaf litter containing oecophorid caterpillars, likewise, yield numerous other beasts – and so on. Each of the herbivore assemblages is part of a complex food web, with each constituent species likely to be attacked by parasitoids and predators with varying capability to find them and exploit them, and with the different developmental and seasonal patterns and resource needs of each herbivore species exposing it to a different array of such natural enemies. If we expand the scope of each herbivore assemblage to be the foundation of these additional associated insects, the richness of this larger module increases substantially. The more generalised natural enemies present may include a considerable variety of different insects amongst their prey, perhaps with different usage at different seasons or in different parts of the environment. Some lacewing larvae occur on or under bark, but not on the more mobile foliage of gum trees, for example. The insects found on a species of Acacia or Eucalyptus will also include a component of tourists, present casually, as well as the many that have some more regular and predictable association, so that the number can become dauntingly high. Up to 200–300 species of arthropods, many of them insects, have been found on or under the loose bark of one Eucalyptus species at one site. Broader surveys of canopy invertebrates on single Eucalyptus species in Western Australia yielded more than 440 species on each of Jarrah (E. marginata) and Marri (E. calophylla). Parallel partitioning across different resources on a plant and taxonomic array (with different but partially overlapping participants) occurs on many other tree species, many of which will support specialist or near-specialist species. Many of the entities envisaged as ecological communities in terrestrial environments comprise complex mosaics of species distributed in different ways across the environment. The total insect richness on a site reflects the richness of plant species and their structure; clearly, a grass or small herb does not provide the same structural components as a large gum tree, and is considerably less apparent (p. 78). Nevertheless, within species of any broad vegetation class from grasses to trees, the same principles of variety apply, with use of the resources by insects influenced by accessibility, competition, and the physical environment.
Assessing Richness Careful sampling and study can help convey some idea of how many insect species occur within an assemblage or community, and the ways in which they may interact. Naturalists in Britain, but not yet in Australia, have collected insects from the same sites – such as their home garden – continually for up to several decades and continued to accumulate previously unseen insect species over such long periods. The term ‘accumulate’ is deliberate – usually, we cannot tell if such discoveries have simply been missed in earlier collecting, or have arrived more recently, but plotting ‘species accumulation curves’ (with some additional statistical appraisal) helps to determine whether our sampling has been sufficient to define the fauna present, so
126
8 Insect Communities
Fig. 8.5 Examples of species accumulation curves for insect surveys. For each successive sample (trap, visit, site, other unit of sampling effort) the number of previously uncaptured species is plotted to indicate a cumulative total. The total species number may reach an asymptote, as in the lower line, indicating that the inventory is complete/near-complete, and that sampling has been sufficient to reveal the species present. If no asymptote is found (upper line), further sampling is needed to achieve that certainty
that the data are valid for characterising the site (community, assemblage) or comparing it with others. Two lines are shown in Fig. 8.5, each representing the accumulated number of species (that is the number of species not found on any previous occasion) with successive or additional samples – either as sample number, across time, or some other defined units of sampling effort. The top line continues to rise steeply – additional species are found even in the latest samples – and suggests that sampling has still been insufficient to estimate the number of species present. The lower line differs. It has gradually decreased in slope and (at least apparently) has reached a level where new species are infrequently encountered with the last few samples taken not yielding any. The level of this asymptote may, with care, be taken as a reasonably reliable estimate of the richness present, and useful in both characterisation and comparison of communities. Various emphases can be put on these numbers, depending on the context of interest. We may wish to compare the composition and richness of communities across different sites or biotopes – perhaps to help rank them for conservation importance or other management – or to help define what the community typical of a given area or representative of a given biotope may be. Or in conservation studies, we may concentrate on particular species of interest, perhaps suspected to be threatened. The variety of uses is almost limitless, but information on community structure and diversity is of far more than academic interest alone. The substantial variety of insects in any freshwater or terrestrial environment will assuredly include those that respond to changes in that environment in some way by changes in their incidence, abundance or distribution, and the sort of data outlined above are thus worthy tools in assessing those influences.
Further Reading
127
Further Reading (The following are key references to enlarge on some of the insect groups used as examples in this chapter) Andersen AN (1990) The use of ant communities to evaluate changes in Australian terrestrial ecosystems: a review and a recipe. Proc Ecol Soc Aust 16:347–357 Crespi BJ, Morris DC, Mound LA (2004) Evolution of ecological and behavioural diversity. Australian Acacia thrips as model organisms. Australian Biological Resources Study, Canberra Greenslade PJM (1978) Ants. In: Low WA (ed) The physical and biological features of Kunnoth Paddock in Central Australia. Technical Paper no. 4. CSIRO Division of Land Resources, Canberra, pp 109–113 Greenslade P (2006) The invertebrates of Macquarie Island (with Insecta by Rieks Dekker van Klinken and Penelope Greenslade). Australian Antarctic Division, Hobart Hollis D (2004) Australian Psylloidea: jumping plant lice and lerp insects. Australian Biological Resources Study, Canberra Majer JD, Shattuck SO, Andersen AN, Beattie AJ (2004) Australian ant research: fabulous fauna, functional groups, pharmaceuticals and the Fatherhood. Aust J Entomol 43:235–247 (background to ant studies, including functional groups definition and applications)
Chapter 9
Insect Populations
Introduction: Population Size and Structure Insects, whether solitary or social, are distributed in many ways across landscapes. Some are virtually ubiquitous across broad ecological panoramas, others are much more localised. And abundance may differ markedly across the distributional range of a species. Restrictions may be geographical – creating patterns of regional or more local endemism – or ecological, depending on the patchiness of critical resources within any given biotope or place, or reflect both of these. The term ‘population’ is central to assessing these conditions and differences, and can be defined in various ways, reflecting these scales – thus ‘the total number of individuals of Species X’ is much more general than ‘the number of individuals of Species X in colony A’, or ‘in forest B’, or ‘at site C’. However, the term is restricted almost universally to single species, rather than to multispecies arrays – but may occasionally be used somewhat more loosely to entities such as ‘the population of predators at site A’, or similar, substituting broader ecological role or guild for individual species. The meaning and scope may need to be defined clearly in each context, simply so that discussants are on common ground. We thus normally consider insect populations as each representing the number of individuals of a focal species in some defined space or ecological arena, ideally one in which the size and distribution of that population can be interpreted or addressed meaningfully. Population size, and its trend across generations, has considerable relevance in conservation and pest management and may even dictate the need for this – as we shall see – but even simple estimates of population size or density of individuals (such as, realistic, estimates of termites over much of Australia at 1,000 or more per square metre) provide a comparative index to help estimate site health or compare with others. Such ‘census population size’ simply enumerates individuals, with no heed to what those individuals do – but is usually qualified by growth stage – such as number of adult butterflies or number of caterpillars. However, the different concept of ‘effective population size’ reduces this number considerably. This is the number of individuals that contribute to the next generation – those that reproduce – and
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_9, © Springer Science+Business Media B.V. 2011
129
130
9 Insect Populations
so excludes at any time those that are post-reproductive, pre-reproductive or otherwise noncontributory (for example, by being excluded from territories). Effective population size may be at least an order of magnitude fewer individuals than a census population, and the difference has considerable functional relevance in considering management issues related to genetic variety, for instance. And all individuals may participate in ecological processes, and eat and be eaten, whether or not they reproduce. Both population size and population structure are important considerations in exploring ‘how insects work’, and much of the information available has been derived directly from studies on economically important pest species or those of conservation interest, for which planning management to respectively suppress or increase numbers draws on awareness of population dynamics. They have additional relevance in understanding insects, because many factors affecting wellbeing and behaviour operate at this level. Changes in population demography (births, deaths, movements) can differ markedly across a wider distributional range in response – for example – to local impacts of natural enemies with distributions differing from that of the species they attack, different food supplies, and utilities such as topography and microclimate. Collectively, the changes within a population of numbers and distributions comprise ‘population dynamics’ with understanding of the various factors affecting and causing these changes telling us much about the individual species and its community context. For a long time, entomologists divided insect populations into two main categories: (1) closed populations, often small and circumscribed in extent, and defined as those in which changes in numbers are due wholly (or very largely) to internal processes – births and deaths – and in which additional changes due to migrations are absent or unimportant; and (2) open populations subject also to immigration and emigration, so that internal demographic processes alone do not fully determine numerical changes. Sudden changes in numbers in the open populations could thus largely reflect these outside influences rather than just those of the local environment. Internal numerical changes through births and deaths lead to continuous changes in numbers and age structure and both short term changes (within a generation) and long term trends in abundance (increase or decrease over sequences of generations or seasons), so ‘dynamics’ is a very suitable description. These changes have applied interest and values. We may, for example, wish to know if a particular species of moth or beetle is likely to become a pest through becoming abundant and/or increasing its distribution, or threatened with extinction as numbers decline to very low levels. For insects which are predictable pests, buildup of numbers may need to be monitored in order to undertake control measures to protect crops and other commodities from their depredations – and economic considerations dictate that this must be undertaken at the most effective and cost-effective time. For threatened species the maxim of ‘safety in numbers’ equates to larger or non-declining numbers. Understanding the causes of numerical change may then become vital, in order to address threats and abate declines to dangerous levels. Population dynamics are easiest to predict in a closed population, simply because external influences can be ignored in the study – although it may be shortsighted to exclude these from the start, without really knowing whether they occur! The normal demographic path is then one of decline in numbers from eggs to adults over a
Population Fluctuations
131
generation, with varying levels and causes of loss at each succeeding stage: if all eggs are present at the same time, as may commonly occur in species with discrete generations and well-defined seasonal development so that each stage is present for a period in which it does not significantly overlap with others, the number of larvae cannot be greater than that of the eggs (other than in the special case of polyembryony, p. 26). We may then, for various purposes, want to know three things about this trajectory of decline, namely (1) its extent, (2) at what stage(s) of the life cycle it occurs or is most significant, and (3) its causes. Each may be complex, especially if we wish to know the precise causes of losses at each stage, but even the overall mortality may be very hard to estimate, and have both inherent and environmental components. Eggs may be fertile or sterile, and the number produced reflects adult nutrition and size; fertile eggs may or may not hatch because of erratic embryonic development, attack by predators, parasitoids or fungi, or being dislodged or lost. Individuals reaching each successive larval instar may in turn survive or not, with a succession of different natural enemies over any extended developmental period, and the growth rate and survival also depending on food quality and local temperature. Pupae, if present, are also susceptible to mortality, with additional predators or parasites likely to participate. Adults may fail to mate, their fecundity influenced by larval food quality, and so on. The list of possible contributory factors to overall mortality is itself very difficult to specify – and will almost certainly differ in relative effects in different parts of the species’ range, and at different times of the year. Particular predator or parasitoid species, for example, may occur in only part of the prey or host range. Clarifying their influences needs careful sampling of the population at intervals sufficient to separately evaluate each growth stage, an exercise that is invariably laborious but most straightforward for insects with a single generation each year, so that the timing of each stage is discrete and, to some extent, predictable. Essentially the exercise needs a defined, discrete, closed population and a series of counts that reveal progressive mortality from the initial number of eggs produced – based on samples, because it is never possible to count every individual present in any reasonably large insect population – and if the various causes of loss are to be determined as well, additional inspections, perhaps assessing natural enemy populations as well, are needed. Very broadly, ‘primary events’ within the population (mainly births, deaths, movements) are influenced by a combination of factors that (1) act directly on individuals, as above, and (2) change the availability and supply of resources for individuals. The latter group link with ecological processes that can either add or subtract individuals by respectively promoting survival and reproduction or inducing premature loss and limiting reproduction.
Population Fluctuations Our most comprehensive understanding of how insect populations fluctuate comes from studies on species that can become abundant – such information is extremely difficult to obtain from species that are rare and occur only in small numbers.
132
9 Insect Populations
Even at their greatest abundance very few individuals may be present or be detected during surveys – and most of the more informative studies are those on pest species evaluated over sequences of generations in order to be able to anticipate their changes and suppress their damaging effects, and for which the costs of the work are easily justified. Once a sound framework of knowledge has been accumulated it is often possible to use the numbers of one particular stage (such as adult moths, fully grown caterpillars, or other which is predictably present and accessible at a particular time [or times] of the year) as a single intergenerational marker to track long-term population changes reasonably reliably, and far more economically than looking at every stage. This tactic has been used constructively to study insects that undergo ‘outbreaks’, periodical or sporadic mass increases in abundance that can be highly damaging – such as locusts on crops, stick insects in forests, and other insects that occur typically in low, non-damaging populations but can respond to environmental changes and rapidly surge in numbers. Any effective warning of this may facilitate management, if needed. However, before looking further at such outbreaks, another form of population structure must be introduced, because it has affected greatly the ways in which insect populations may be interpreted. In particular, it questions whether many populations traditionally regarded as ‘closed’ really are so, and what the consequences and importance of local extinctions may be. If the populations are truly closed, isolated, and have no contact with others – as supposed widely in the many insect species that have become the foci of local conservation concerns – loss may be calamitous. Studies on some butterflies, and more recently on a variety of other insects, have introduced the rather different concept of ‘metapopulations’, whereby the species exists across a landscape as a suite of more-or-less demographically independent population units in different habitat patches, but at any given time may be extinct in some patches and thriving in others. These population units are subject to interchange of individuals through dispersal, perhaps not commonly, and the various habitat patches are subject to sequences of ‘colonisation – extinction – recolonisation’ cycles, perhaps at very irregular or infrequent intervals and reflecting changes in the patches, with the consequence that any local extinction may be part of the species’ natural dynamics. This dispersed structure, the metapopulation, is proving to be quite common amongst insects as more examples are studied. Several patterns have been distinguished (two are shown in Fig. 9.1), but the common feature is the occurrence of a patchily distributed population across a mosaic of more-or-less separated habitat patches within the landscape, each of which may function independently for much of the time. It is often very difficult to decide whether an insect population is in fact a fully closed population or a unit of a metapopulation in this way – but the wellbeing of a metapopulation depends on the various units being within dispersing distance, and without apparent barriers between them. Imposition of any barriers by human activity may cause a previously functional dispersal pattern to collapse and enforce isolation of the various units. This appears to have happened, for example, with the Eltham copper butterfly (Paralucia pyrodiscus lucida) near Melbourne. This subspecies now occurs on a few small sites in suburban outer Melbourne, separated from each other
Population Fluctuations
133
Fig. 9.1 Two patterns of metapopulation structure and principles. (a) The ‘sourcesink’ model, in which a large source area/habitat provides individuals that colonise, regularly or sporadically, small outlying habitat patches which, at any time, may or may not be inhabited by a resident population. (b) The ‘archipelago’ model, in which interchanges of individuals occur across habitat patches, so that the accessibility of each within the landscape as well as its individual condition influences likelihood of successful colonisation
by housing developments through which, it seems, the butterflies cannot or do not disperse. The distance itself is not limiting – in a more open environment in western Victoria, the butterflies seem to easily traverse equivalent distances between clumps of caterpillar food plants. ‘Outbreaks’, the massive increases in insect numbers within populations that lead to names such as ‘swarms’ or ‘plagues’ in the popular press, may be combined with mass movements whereby the insects invade areas that they do not usually inhabit permanently. Every few years the Australian plague locust (Chortoicetes terminifera) ‘erupts’ in this way, and moves southward from inland areas of eastern Australia into the major cropping areas of the south east, sometimes causing massive economic damage. Understanding causes of such outbreaks enables some predictive measures to control them, by forecasting where they are likely to go, when they will be there, and some estimates of numbers involved, together with planning possible control measures needed. One component of this for locusts is to plot the distribution and extent of the overwintering ‘egg beds’ that source the active population once they hatch in spring, and are a good indicator of the potential severity of that later attack. Whilst these locusts can migrate over large distances, many other insects simply increase within the same area, or spread to only more limited extents as numbers build up. Stick insect outbreaks in forests are one such case, where local effects can be much more important than those in wide landscapes. Many such insects are normally present in only low numbers, but may increase rapidly in response to drought, unusual weather or some other form of stress – and, in some instances, may reach a new ‘equilibrium’ at these higher levels of abundance.
134
9 Insect Populations
Fig. 9.2 Caterpillars of the gum-leaf skeletoniser moth, Uraba lugens (Noctuidae, Nolinae), are sometimes abundant and conspicuous, their feeding leading to marked unsightliness of affected trees. Their urticating hairs are presumed deterrent to vertebrate predators, in particular, and they are recognised easily by the unusual habit of retaining the cast head capsules of earlier moults in a ‘turret’ on the thorax
In Western Australia’s Jarrah (Eucalyptus marginata) forests, the leaf-mining moth Perthida glyphopa (Jarrah leaf-miner) became an economic pest only from the 1950s, with some 750,000 ha of forest infested by 1987. A second native moth species, the gum-leaf skeletoniser (Uraba lugens, Fig. 9.2) erupted there in 1982 and had infested about 300,000 ha by 1985. Both feed on senescing foliage of Jarrah and, although they do not kill the trees directly, declines in canopy quality from extensive loss of foliage are related to decreased increments of timber for harvest. Similar outcomes occur from depredations of the autumn gum moth (Mnesampela privata, Geometridae: Fig. 9.3) in south eastern Australia. As in most aspects of insect biology, there is no single or simple explanation of why the outbreaks of these, and other species, occur. Two main groups have been identified as ‘eruptive’ and ‘gradient’ outbreaks, and the above species exemplify these. Eruptive outbreaks spread from the source to cover, sometimes, very large areas, so that they are characterised by invasion of areas where the species does not normally occur – except, perhaps, in very low numbers or otherwise infrequently - in response to particular environmental factors such as food supply and weather sequences, for example linked to winter rainfall. The eruption may be short – lived, if the insects have impacts such as destroying their food supply and attracting abundant natural enemies that cause reduction to low numbers. In contrast, eruptions can last for several years or even longer if not opposed. Gradient outbreaks generally stay close to where they originate, often restricted to particular sites or environments. They may occur in response to a superabundant food supply, such as a specific uniform-aged monoculture forestry crop and persist for as long as that resource is available.
Population Fluctuations
135
Fig. 9.3 The autumn gum moth, Mnesampela privata (Geometridae) is a sporadic outbreak pest on eucalypts, and sometimes important in forestry. The adult moth is cryptic and difficult to detect among foliage when at rest during the day
Perhaps more typically, some form of cyclic outbreak occurs, whereby every decade or so conditions become particularly favourable but in which high numbers of insects are very temporary and the population ‘crashes’ rapidly – often because of mass starvation or massively increased impacts of natural enemies on such large targets. Eruptive outbreaks occur in three species of locusts in Australia, most commonly in C. terminifera (as above) but also (much less frequently) in the spur-throated locust (Austracris guttulosa) and the migratory locust (Locusta migratoria). Their importance is signaled by these being the only pest insects in the country to have a dedicated federal agency (the Australian Plague Locust Commission) for monitoring and forecasting outbreaks, and controlling them. The first documented outbreak of C. terminifera occurred in 1844 and over the twentieth century a pattern emerged of high density populations in some locations in most years, with the less frequent ‘plague populations’ predominantly in eastern Australia but some also in Western Australia. Populations can build up very rapidly, within a year, and persist for up to several years. In years with good rainfall, 3–5 annual generations can develop, with increased densities of locusts developing gregarious behaviour and forming large marching ‘bands’ of immature hoppers and subsequent swarms of flying adults – so that both immature stages and adults participate in expanding the range. Adults can migrate for up to 500–1,000 km over several nights. Outbreaks arise in the semi-arid interior of eastern Australia, following exceptionally wet summers fostering this rapid breeding, so that this weather pattern is a major driver of outbreak development, leading to movement into other areas, including intensive cropping areas that commonly also benefit from higher rainfalls. Outbreaks can collapse rapidly if drought conditions develop, but natural enemies (particularly some parasitoids) may also be important contributors to this.
136
9 Insect Populations
The major species involved in forest stick insect outbreaks in the east is the spurlegged phasmid (Didymuria violescens), which is widespread, particularly in montane forests in south eastern Australia. Outbreaks tend to occur mainly at higher elevations, and result in extensive defoliation of stands of eucalypts, sometimes leading to widespread deaths of trees. Several surveys in Victoria, for example, demonstrate that mountain ash (Eucalyptus regnans) trees died 1–2 years after severe defoliation. Because the stick insects are cryptic and usually live high in the canopy, they are not easy to count directly, so that two rather different approaches to monitoring population sizes have been adopted. First, the eggs, several millimeters long and resembling plant seeds, are dropped to the forest floor and can be sieved from standard areas of litter, and counted. Second, aerial surveys may help to detect ‘hotspots’ of defoliation. In 1963, for example, D. violescens defoliated approximately 650 square miles of Eucalyptus forests, with the damage being described as ‘spectacular’! Summer temperatures affect the regimes of embryonic diapause within the eggs and help to explain local variations in life history pattern. Outbreaks are compounded by the high numbers of eggs every second year, with some spread of stick insects to increase the areas affected. In due course, severe defoliation leads to population declines. Under non-outbreak conditions, it has been proposed that stick insect numbers are controlled by birds (such as the pied currawong, Strepera graculina, and several other large insectivores), for which examination of gizzard contents has revealed very high intake of stick insects. However, whilst the birds are effective predators in ‘normal conditions’ they have little effect during outbreaks, simply because the birds cannot respond sufficiently rapidly to the massive food supply then suddenly available. Outbreaks of psyllids and some other herbivores may be stimulated by water stress (drought), and its correlation with favourable food by stimulating increased nitrogen content in plant sap and foliage. Survival of early stages may then be enhanced, leading to larger adult populations. In conjunction with lack of pressure from natural enemies at higher numbers – as above – food quality and quantity promote continued population growth. As a following effect, severe defoliation of eucalypts or other trees may in some way weaken them and increase their susceptibility to attack by other insects, such as wood-boring beetles, which can then also build up in numbers. There is, perhaps, no single unifying cause of insect outbreaks but the above scenarios indicate two contrasting contexts, and exemplify a common consequence of ‘ecological release’, whereby factors that normally restrict or oppose population increase fail, usually temporarily, to do so and others change to favour growth. Most insects, however, do not exhibit outbreaks. Even though their numbers may fluctuate substantially between successive generations, many are sufficiently consistent to be regarded as ‘common’ or ‘rare’ in terms of higher or lower numbers respectively. Interactions between the various species in communities, and the limitations of the physical environment, usually prevent massive increase in abundance – so that the early sensationalistic literature proclamations along the lines of ‘the progeny of a single aphid (fly, beetle, moth) would cover the world to a depth of X metres within Y months’ are simplistic in not allowing for these wider countering influences.
Further Reading
137
Further Reading Clark LR, Geier PW, Hughes RD, Morris RF (1967) The ecology of insect populations in theory and practice. Methuen, London (a classic early account of how insect populations ‘work’, largely drawing on Australian examples and of persistent value) Elliott HJ, Ohmart CP, Wylie FR (1998) Insect pests of Australian forests. Ecology and management. Inkata Press, Melbourne (includes much background on population fluctuations and causes of outbreaks, as well as basic ecology and control) Floater GJ, Zalucki MP (1999) Life tables of the processionary caterpillar Ochrogaster lunifer Herrich-Schaffer (Lepidoptera: Thaumatopaeidae) at local and regional scales. Aust J Entomol 38:330–339 (example of population analysis to reveal the major mortality factors for this insect, noted in chapter 7 for its unusual behaviour) Readshaw JL (1965) A theory of phasmatid outbreak release. Aust J Zool 13:475–490 (attempt to explain sporadic outbreak phenomena of forest stick insects) (Rentz et al., cited on p. 155, contains an account of the biology of Australian locusts and the work of the Australian Plague Locust Commission)
Chapter 10
Insects in Inland Water Environments
Introduction: Inland Aquatic Habitats for Insects Up to now in this book, aquatic insects have been referred to as ‘freshwater’, to distinguish them from ‘marine’. However, the descriptor ‘freshwater’, strictly, is not accurate, because many insects occur in saline aquifers, salt lakes and other nonfreshwater bodies in Australia. Two examples illustrate this. Near-coastal evaporation pools for manufacture of salt from sea water sometimes teem with small flies, and the aquatic larvae of some Ephydridae there initially appear to contradict suggestions of low abundance of ‘marine’ insects! They thrive in these hypersaline environments, and some other members of this family are renowned for occupying extreme habitats elsewhere in the world. Species occur, for examples, in the hot springs of New Zealand, and one in the New World even breeds in pools of crude petroleum and waste oil. These flies, all related to more conventional ephydrids, have each gained a milieu unreached by other insects, and in which they are largely alone. The southern Australian Ephydrella marshalli has been proclaimed as the ‘most salt-tolerant’ of all insects. Its larvae can crawl around on wet salt crystals at one extreme but are equally at home in distilled water, with the transition (verified in laboratory transfers) from salt concentrations several times those of sea water to zero a formidable physiological feat for any insect. Second, and exemplifying an aquatic system often overlooked as a habitat for insects, many groundwaters, aquifers, are also highly saline or may have strong salinity gradients with depth. Their unique obligate invertebrate fauna (collectively ‘stygobites’) include numerous species of predatory water beetles, Dytiscidae. These underground waters occur as karst, calcrete and alluvial groundwater systems. Although Dytiscidae are the only group of insects yet found amongst the invertebrates of calcrete aquifers, recent surveys in Western Australia have led to reference to these as ‘the world’s most diverse collection’ of these stygobites. Surveys so far have been very incomplete (incorporating only about 10% of the approximately 210 major calcrete aquifers in the region), but reasoned extrapolations suggest, based on implication that each beetle species so far discovered is unique to an individual
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_10, © Springer Science+Business Media B.V. 2011
139
140
10 Insects in Inland Water Environments
aquifer, that numbers may increase massively from those already known. Following the analogy introduced earlier, the aquifers have been described as ‘subterranean islands’ with independently evolved taxa, probably from at least three different terrestrial invasions. For completeness in noting saline systems, salt lakes are found in many parts of Australia, and range from deep persistent water bodies to smaller, shallow and temporary ones. The general term ‘inland waters’ ignores any implications of varying salt content. However, salinity is only one of the influential features of the distribution and incidence of aquatic insects. Parameters such as water flow, depth, and permanence are all correlated with insect habits, much as the mosaic of environmental factors in terrestrial environments affects the spectrum and wellbeing of insects present. And, as on land, such factors can vary widely and render simple categorisations of habitat unsatisfactory. ‘Permanent’ fresh water lakes, for example are characteristic mainly of Tasmania and western Victoria, with lakes elsewhere more irregular. ‘Permanent’ fast flowing rivers and streams are mainly found in the east and south east, whereas elsewhere normally flowing waters can be disrupted by rivers breaking to series of static pools in times of drought. ‘Billabongs’ are a peculiarly Australian name for lengths of water courses isolated from flow, and further confuse designations of still and running waters because they may be isolated for long periods, as more-or-less permanently still, or seasonally or more sporadically linked with the rivers from which they were earlier separated. In general, though, running waters comprise rivers and streams, and still waters encompass lakes, reservoirs, farm dams and a variety of ponds and pools. As with vegetation types on land, each of these can be subdivided almost ad nauseam, and it is vital to acknowledge the gradient nature of their features. This is accepted explicitly in the ‘river continuum ‘ concept, based on the idea that a river is continually changing along its length, through changes in the physical environment – such as depth, width, temperature, bank features and nutrient inputs, and so on. Broad categories within this are headwaters, midreaches and lower reaches, differentiated formally by stream order and major features of organic inputs and transfer downstream. Thus ‘shredders’ are often the predominant arthropods in headwaters, and other groups noted on p. 78 gain greater importance further downstream. The example shown in Fig. 10.1 demonstrates this gradation. Narrow headwaters are often heavily shaded and receive much fallen foliage, so that shredders are important in breakng large units down into smaller particles. Further along a river this, now finer, particle organic matter is available to grazers and collectors and the latter become predominant in the lower reaches. The other commonly used functional descriptor, for insects or other organisms, also draws on water flow – as ‘lentic’ (associated with calm or still waters) versus ‘lotic’ (associated with flowing waters). Lotic and lentic environments pose very different conditions for insects to exploit: strong currents require adaptations to resist being swept away, so that a flattened , streamlined body form, small size and living on or in the substrate are all common features. In lentic environments, the risk of physical removal by water is low, and insects can be larger and more exposed, such as by living on vegetation. However categorising the insects present
Insect Variety
141
Fig. 10.1 Feeding habits of aquatic insects in relation to the River Continuum Concept; the proportions of major trophic groups at different localities along a river from headwaters (a) through mid reaches (b) to wider lower reaches (c) in this generalised example the proportion of predators is not changed, but the balance among other groups differs markedly (see text) (Based on Wiggins 1994)
(other than taxonomically) can invoke their habits – ‘benthic’ insects live on or in the substrate, whereas others live on vegetation or are free-swimming (‘pelagic’) – but a major focus for examining their variety is to separate the orders of insects for which aquatic life is universal, from those in which it has arisen more sporadically amongst terrestrial relatives. Both are very varied, but whichever of these levels of dependence we consider, all are derived from terrestrial forebears.
Insect Variety The variety of insect orders found in freshwater environments is summarised in Table 10.1. Three orders of insects with incomplete metamorphosis are universally (or almost universally) aquatic, and so depend wholly on freshwater environments. All species of mayflies and stoneflies have aquatic larvae, and the adults of these orders (Ephemeroptera, Plecoptera) rarely move far from water. Odonata are also aquatic, with larvae of all but a very few species aquatic – however, adults of many dragonflies are strong flyers and can venture far from water. Indeed, a maturation period for females of many species in which they move up to several tens of kilometres afield, is a regular feature of their life history. Some species are strong, habitual migrants. The strong dispersal ability, even of many delicate-looking damselflies (Fig. 10.2), enables them to colonise seasonal or temporary water bodies effectively, so that their collective habitat range is extensive. All three of these orders have larvae well-adapted to aquatic life, and so are very characteristic in appearance. All, for example, have diagnostic structures for aquatic respiration, and the form of these reflects the kind of water they inhabit, and the supply of oxygen available. Fast flowing waters are typically highly oxygenated, so that mayfly larvae there commonly have their lateral abdominal paired gills broad
142
10 Insects in Inland Water Environments
Table 10.1 Major groups of insects represented in inland waters in Australia PALAEOPTERA Ephemeroptera: all species have aquatic larvae, and adults mostly found close to water bodies; many species in east and south east Odonata: almost all species with aquatic larvae, but adults strong flyers and can move far from water; very few species have larvae in wet forest litter NEOPTERA Plecoptera: all species with aquatic larvae, and adults mostly found close to water bodies; some species flightless, and many very localised in distribution Orthoptera: very few grasshopper species swim. One Australian species (the diving grasshopper, Bermiella acuta) can stay submerged for up to about 10 min, but the order is characteristically terrestrial Hemiptera: aquatic forms arisen in several lineages of Heteroptera, mostly as predators or surface film scavengers Megaloptera: all species with aquatic larvae, all predators and some long-lived; most species appear to have limited distributions Neuroptera: predominantly terrestrial, but aquatic larvae in three families. Sisyridae (spongeflies) are specialised predators of freshwater sponges; Nevrorthidae are benthic predators, mainly in gravel substrates; Osmylidae also include terrestrial forms, but some are littoral/shallow water predators. All contain few species Coleoptera: numerous aquatic taxa, in most major lineages, some wholly so, and most trophic groups represented. Dytiscidae are aquatic predators as larvae and adults; some others are more characteristically littoral Mecoptera: mainly terrestrial, but Nannochorista has aquatic predatory larvae Diptera: numerous forms with aquatic larvae. Most are Nematocera, but some groups of Brachycera (such as Ephydridae) also represented; often in large numbers and many of health importance as disease vectors or with biting habits; mosquitoes, blackflies, sandflies, and others may demand control Lepidoptera: few aquatic forms. Caterpillars of some Pyralidae (Nymphulinae) tunnel or mine within aquatic plants Hymenoptera: very few parasitoid wasps enter water to exploit aquatic hosts such as eggs of water beetles
and flaplike, as a large surface area is not critical. Conversely, mayflies (Fig. 10.3) living in still or turbid water with lower oxygen tensions have the gills filamentous or ‘feathery’ in form to provide large respiratory surfaces. Odonata differ in their gill arrangement between the two main groups. The damselflies have three posterior gills, as long ovoid extensions from the rear end; these ‘caudal lamellae’ are important also as signalling devices – the damselfly equivalent of semaphore flags – and are sometimes banded or strongly marked, rendering them conspicuous. They can be raised and moved around, and larvae of some species studied have a substantial repertoire of different postures helping them to maintain territory and affecting interactions with other individuals. Larvae of many species are active on vegetation and quite conspicuous. Conversely, larvae of true dragonflies – the ‘mudeyes’ of Australian anglers – tend to be more cryptic and live on the substrate. Some characteristically burrow in the substrate. They lack external gills, but undertake respiratory exchange across the walls of the rectum, so water is pumped in and out of the hind gut for this purpose. Gills of stonefly larvae are much more varied, although the
Insect Variety
143
Fig. 10.2 Odonata: a pair of Aurora bluetail damselflies, Ischnura aurora, common around ponds and other still water bodies: the brightly coloured male has here grasped the female behind the head by his posterior claspers during the mating sequence
Fig. 10.3 Ephemeroptera: characteristic appearance of an adult mayfly, with large eyes, three long ‘tails’ and wings held vertically when at rest: the taxon depicted, Marmeneura, is a recentlydescribed genus endemic to Tasmania
pattern is consistent within any given taxon. Gills are typically small filaments, and can occur (sometimes in well-defined localised ‘tufts’, in other species scattered more generally) on almost any part of the body, although never likely to cause confusion in differentiating larvae from those of other orders. Obligate aquatic life occurs universally also in two holometabolous insect orders, the Megaloptera (dobsonflies, alderflies) and Trichoptera (caddisflies), and adults of
144
10 Insects in Inland Water Environments
both these are associated strongly with water and rarely found far from waterside vegetation. Megaloptera, a small order closely related to Neuroptera and regarded as amongst the most primitive insects with complete metamorphosis, include some of the largest Australian aquatic insects – with some species having a wingspan of around 10–11 cm. Their larvae (living for up to several years as active predators) have paired lateral abdominal gills, with the pattern differing between the two families, and – unlike the mayflies – present on almost all segments. Caddisfly larvae are very diverse in forms and habits, and are best-known for the habit in many taxa of constructing protective cases of vegetation, debris or mineral matter in which the larvae dwell. Many species have single or branched filamentous gills, mainly in rows on the abdomen. As with other aquatic groups, differences in structure and habits are sometimes correlated clearly with features of lotic and lentic environments and availability of food and oxygen. As one such correlation, caddis larvae that spin silken capture nets that filter food from the water column are mostly found in running water in which food is swept toward them. The transition from lotic to lentic waters commonly includes changes to lower elevations, decreased current speed and lower dissolved oxygen levels. Lotic systems in eastern Australia have somewhat higher caddisfly richness than many other regions, and many lentic habitats support only relatively few caddisflies, representing a smaller variety of genera or families. Collectively, caddis larvae have numerous feeding roles, as predators, herbivores or omnivores. We then come to the very diverse array of insect orders in which the aquatic habit is more sporadic and, in some, represents independent developments in distantly related lineages. As with solving respiratory problems by developing gills, as above, these various forms show considerable convergence in evolution as adaptations to adopt aquatic existence from a terrestrial origin. However, only Hemiptera amongst hemimetabolous insects have done so, and even this large order remains predominantly terrestrial. The 15 or so families of Hemiptera with aquatic members in Australia have been allocated amongst three groups with rather different associations with water: (1) a few families are really only ‘marginally aquatic’ in that they are shore dwellers that may occasionally enter water, but found only around water margins and in exposed sediments; (2) five families occur only on the surface film, as below; and (3) six families are completely aquatic. However, only two of these (backswimmers and water boatmen, technically Corixidae, Notonectidae) are strongly modified for swimming, whilst the others predominantly clamber around but have some swimming capability. All may fly to reach other habitats. Aquatic life has thus developed independently in several groups of Heteroptera, and almost all of the waterdwelling forms (‘waterbugs’) are predators or scavengers, rather than true herbivores. The group Gerromorpha is perhaps more properly regarded as ‘semiaquatic’, as they have adopted the very specialised and restricted habitat of the water surface film, on which they ‘skate’ as the most diverse group of animals in this arena. Many (such as ‘water striders’) have very long legs, and are able to detect and track vibrations in the surface film, such as those caused by insects falling into water, and on which the stri ders can feed. Aggregative and courtship behaviour also incorporates such vibratory signals, but most species do not venture underwater to live. Many have coatings of
Insect Variety
145
fine water-resistant hairs, amongst which air can be trapped. The more completely aquatic bugs must visit the surface for breathing, and have not developed gills equivalent to the above-noted orders. Two mechanisms are involved. Many of the largest predators, water scorpions and water stick-insects, respire through a posterior filamentous ‘siphon’ which they poke through the surface film from underneath to take in air to the tracheal system. Most smaller bugs, in common with diving beetle adults, either carry an air bubble from the surface as they dive, or utilise some form of ‘plastron respiration’ in which air is carried amongst layers of hairs, and gives the whole insect (or, in some, the more restricted plastron area) a silvery appearance. The term ‘plastron’ is used for modified areas of the body, such as by being densely haired or the cuticle corrugated, and on which a surface film of air can be held, resisting wetting, and allowing respiration under water. It is thus functionally equivalent to a gill. Many species of aquatic bugs have conspicuously raptorial forelegs – resembling those of the terrestrial mantids – for prey capture and handling. Their food seems to be limited simply by physical accessibility, and several of the giant water bugs (Belostomatidae, themselves popular as human food in south-east Asia) can be pests of hatchling fish in aquaculture operations in northern Australia. Actively swimming predators, such as water boatmen and ‘backswimmers’, commonly have very large eyes. All have streamlined bodies, and their legs can be flattened and have dense fringes of long hairs, increasing their effectiveness as oars in rapid swimming. With the exception of Coleoptera, aquatic representatives of the remaining holometabolous orders are restricted to the early stages – other than for occasional visits, for example by the few parasitoid wasps that swim to find aquatic hosts under water. Many independent invasions seem to be involved. Thus, the three families of Neuroptera associated with fresh water each have very distinctive larvae and habits. Sisyridae (sponge flies) are amongst very few specialised predators of fresh water sponges – and, perhaps, also bryozoans – piercing with their long, slender mouthparts; spongefly larvae have ventral abdominal gills. The elusive Nevrorthidae have active aquatic larvae living in gravelly substrates, and feed more widely on small animals, and larvae of some Osmylidae (those of other Australian species occur in terrestrial environments such as under loose bark of gum trees) occur in sand and mud around the edges of waterbodies, and have been suggested to ‘probe’ for prey such as small arthropods, using their elongate straight mouthparts. The aquatic Neuroptera belong to ancient lineages, with the predominant osmylid group involved clearly Gondwanan. The same is probably the case for the rare species of the scorpionfly family Nannochoristidae (p. 243), a very small group with elongate larvae living in silt and thought to feed on fly larvae there. Aquatic Coleoptera, Lepidoptera, and Diptera are also diverse and taxonomically varied. Aquatic caterpillars (with most of the Australian species belonging to the Nymphulinae, a subfamily of the large moth family Pyralidae) often parallel caddis larvae in constructing shelters of bits of foliage, and in the later instars having filamentous gills. However, only about six groups of Lepidoptera have aquatic representatives, considerably fewer than in the other two orders in this suite. In all three, independent evolutions of an aquatic life style have transcended major taxonomic divisions.
146
10 Insects in Inland Water Environments
The variety in Diptera is illustrated by only two large families (and a few smaller ones, such as the Blephariceridae, the net-winged midges) having wholly aquatic larvae, whilst all others have a mixture of aquatic and more terrestrial members. The mosquitoes (Culicidae) and blackflies (Simuliidae) are fully aquatic, and both are important because of the nuisance and medical problems that occur from the adults biting people and animals. Much of the control of these pests may devolve on the aquatic stages. As with the beetles, the other groups represented range from families that are predominantly aquatic (such as the biting midges, Ceratopogonidae) to others with few species involved in aquatic habitats. Larvae of most groups have very characteristic form, reflecting their habitat features. Larvae of Blephariceridae and Simuliidae, for example, are found only in fast-flowing water and both anchor to the substrate. Blepharicerid larvae have a series of ‘suction pads’ enabling them to cling on to rocks, and simuliid larvae have a single posterior sucker for the same purpose – they can move around by ‘looping’ this in a motion reminiscent of that of a tiny leech. Some larvae construct tent-like shelters. Mosquito and many other larvae are more typically pelagic, whilst many members of some families are benthic and burrow in the substrate. As we might by now anticipate, numerous specialisations occur for feeding, respiration or other adaptive roles. Some midge larvae have antennae modified for grasping prey, and a few (members of the widespread genus Chaoborus) have two pairs of internal airsacs that function as hydrostatic organs; some mosquito larvae have respiratory siphons at the posterior end, and these take in air through the water surface much in the manner noted earlier for some bugs. A number of chironomid midge larvae are known as ‘bloodworms’ because, unlike most insects, they have bright red blood containing erythrocruorin, an agent helping respiration in stagnant water with very low oxygen content. Pupae of many aquatic flies also swim actively, contrary to the more usual resting condition of the pupal phase. Adult mosquitoes and their relatives emerge at the water surface. Beetles show equivalent variety and highly parallel adaptations – emphasising the large extent of convergent evolution in insects adopting almost any innovative shift in habitat, feeding or behaviour. In the Coleoptera, we see a replicated array of almost all the aquatic adaptations in larval and adult insects that have developed in other groups, and varying degrees of dependence on the aquatic habitat. About 15 families of Australian beetles include fully aquatic species, with a few others semiaquatic, as in the bugs. Plastron respiration is very well developed, whilst some other beetles carry an air bubble underwater. Aquatic larvae of many beetles have lateral or posterior gills, or obtain oxygen by simple diffusion across the body surface. Gyrinidae, the ‘whirligig beetles’ are the most specialised Coleoptera of the water surface film where, as their common name implies, they skate around rapidly and are often gregarious. They are particularly unusual in that the adult compound eyes are divided horizontally, so that the beetles appear four-eyed, enabling them to see both above and below the water surface. Gyrinids have very long fore legs, with the other legs short and paddle-like, and their larvae somewhat resemble those of Megaloptera in having paired abdominal lateral gills. Many other water beetles have the hind legs elongated for swimming, but most have a strongly ‘streamlined’ body shape. Dytiscidae are predators as adults and larvae and, even neglecting the rich
Further Reading
147
stygobitic fauna, is perhaps the largest family of water beetles in Australia. It is followed closely by Hydrophilidae, in which the adults are herbivores or detritivores. Many groups of aquatic insects in Australia are still very incompletely known. Perhaps only for Odonata and Megaloptera are few further surprises in store, and the fauna ‘well-known’. Not surprisingly, in view of their wider importance, some groups of biting flies have received considerable attention. More innocuous or inconspicuous groups – the mayflies, stoneflies, moths, caddisflies and many beetles and flies – have indeed received increased attention over recent decades, and our knowledge has increased dramatically for many groups, although it is still very incomplete. Unusually amongst insects, ability to recognise the larvae of some orders (and appreciation of their biological variety) is perhaps more advanced than for the corresponding adults, but some geographical biases remain. Mayflies and stoneflies, for example, are cited widely as ‘south eastern groups’. This opinion, notwithstanding that both orders contain well-defined Gondwanan groups, also reflects high intensity of collecting within this area. The full variety of most of these orders in Australia is yet to be revealed.
Further Reading Corbet PS (1999) Dragonflies. Behaviour and ecology of Odonata. Harley Books, Colchester (global synopsis of dragonfly evolution and biology, with much information on behaviour and adaptations to different aquatic environments) Gooderham J, Tsyrlin E (2002) The waterbug book: a guide to the freshwater macroinvertebrates of temperate Australia. CSIRO Publishing, Melbourne (well-illustrated and informative recognition guide) Hawking JH, Smith FJ (1997) Colour guide to invertebrates of Australian inland waters. MurrayDarling Freshwater Research Centre, Albury, Identification Guide no. 8 (photographic guide to aquatic stages of insects and other invertebrates, with notes on diagnostic features, biology and distribution) Wiggins GB (1994) Caddisflies: the underwater architects. University of Toronto Press Inc., Toronto Williams DD, Feltmate BW (1992) Aquatic insects. CAB International, Wallngford (broad survey of biology and variety of aquatic insects)
Chapter 11
Australia’s Alpine Insects
Introduction: Environmental Extremes Australia’s small alpine and subalpine zones, confined to the south east of the country, represent an extreme and harsh environment – and one that is also vulnerable to climate change, with forecasts of severe reduction in snow cover likely to occur over the next few decades as conditions warm. In addition to these projected changes, considerable human pressures exist with development of resorts and facilities, including improved access, for winter sports and, increasingly, for summer recreational tourism. Whilst some ambiguity may remain over the effects of climate change, commercial fears that it may indeed render the areas increasingly unsui table for winter recreation are themselves pressure for more intensive short term developments. The region is thus regarded as one of the most vulnerable in Australia, but supports a considerable variety of characteristic animals and plants that are not found at lower elevations, and that in some way have adapted to thrive in the extreme cold environment with winter snow cover. Some of these species are of major conservation concern as a result of perceived increased vulnerability to warming, and habitat loss through development. Debate will continue over effects of future climate change, but three possible trends are amongst major concerns inferring vulnerability of specialised alpine insects and other localised residents. First, distributions may change: if species are indeed restricted by need for low temperature, warming may render their current environments increasingly untenable, and ‘force’ them to move if they are to survive. Movement must be to cooler regimes, and none may exist. Evidence for this trend has come from studies on British butterflies and dragonflies, in particular, some of which have moved progressively northward in Britain as conditions have warmed over the last few decades, and which have emphasised the importance of maintaining capability for movement in the landscapes (as environmental gradients of elevation and latitude) well beyond currently occupied areas. Under such circumstances, some alpine insects literally have ‘nowhere to go’. Second, insects and their consumable resources may be affected differently so that long-honed patterns of synchronisation and availability may be lost – simplistically,
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_11, © Springer Science+Business Media B.V. 2011
149
150
11 Australia’s Alpine Insects
if a butterfly is forced to emerge, feed and lay eggs earlier in the season, and food is not then available for either the adult or the caterpillars, survival is improbable unless it can move elsewhere. And, third, the structure and composition of the local communities may change, as other, currently non-resident, species of animals and plants are driven upward to counter changes at lower elevations, so that competitive and other interactions may be modified dramatically and at the expense of specialised resident species. Living in the extreme alpine environment necessitates a suite of specialised adaptations. In addition to cool temperatures, insects must contend with prolonged snow cover restricting activity and development for several months each year, coupled with shorter clement periods in which reproduction must be focused, perhaps with an overall reduced variety of resources than is available at lower elevations. In addition to permanent residents, many other insects are vagrants or temporary (perhaps, seasonal) residents in alpine areas, but the restricted resources clearly restrict the richness of residents. The diversity of eucalypts on which so many insects focus at lower elevations, for example, is severely reduced. The few alpine species include snow gum (Eucalyptus pauciflora) on the mainland, and E. coccifera and E. gunnii in Tasmania, but above the treeline only herbfields and heathlands occur. These ecosystems, however, can be botanically quite diverse. The considerable variation in the snowline level, and the generally low elevation even of Australia’s highest country (the highest mountain, Mt Kosciuszko, is only 2,228 m high) renders the designations of ‘alpine’ and ‘subalpine’ difficult to compare with the much higher European or North American parallels. Some Australian ecologists prefer to use the term ‘snow country’ to refer to all areas where regular prolonged snow cover influences the wellbeing of animals living there.
Alpine Insects Categorising the insects found in the snow country is not always easy. In particular, ‘residency’ may reflect different ways of life. One notable visitor, the Bogong moth (Agrotis infusa), has been mentioned earlier (p. 33) as a seasonal migrant to aestivate in highland areas (Fig. 11.1), but it is only one of around a thousand insect species reported from the region. Many of these are indeed casual visitors – some are carried upward from lower elevations on wind currents, for example, over distances that may in reality be only a few hundred metres from much more complex biotopes at lower elevations. Their incidence can be regular, every year, and confounded by specialised behaviour such as hill-topping, by which they fly upward to seek prominent assembly points (p. 98), and the numerous individuals present can give an illusion of residency. The Bogong moth is resident in this sense of regular presence as an intrinsic phase of its life cycle, but does not breed in the Alps. Other insects may breed during warmer parts of the year, after migrating upward in spring, but do not live there permanently, and die off (or migrate) at the onset of cold weather. Others are true residents, with adaptations to enable them to persist and cope with
Alpine Insects
151
Fig. 11.1 The annual migration pattern of the Bogong moth, Agrotis infusa (Noctuidae), in south eastern Australia. Dotted areas are the self-mulching soils that represent the most important breeding areas of the moth, from which adults move in early summer (solid arrowed lines) to aestivate in the restricted alpine areas in the far south east (black), with reverse migrations (broken lines) in late summer (Information after Common 1954; McCormick 2005)
strong seasonal variations and, in many cases, to be restricted to the region. Whilst the richness of permanently resident species is somewhat lower than at many lower elevations, their populations can become very large – a symptom of ‘numerical compensation’ in environments of low interspecific competition and a limited range of resources. Notable examples of true alpine fauna occur in Lepidoptera, Orthoptera and Plecoptera, whilst some other orders (including large groups such as Hemiptera, Diptera and Hymenoptera) – although represented – have rather few truly characte ristic species in the region. Some Hemiptera and the sawfly Pseudoperga lewisii feed on snow gum and so do not extend beyond the treeline, but collectively some 15 orders are represented by permanent residents. Several morphological features of insects are associated with life at high elevations, but are not restricted to species there. In comparison with their lower elevation
152
11 Australia’s Alpine Insects
Fig. 11.2 The grasshopper Phaulacridium vittatum is a very variable species with Bassian distribution, and occurs from sea level to high on alpine areas. It can become very common, and eats a variety of forbs, rather than grasses
relatives, many resident species have one or more of decreased size, long legs (for walking on snow without losing heat), dense hairiness (as in some moths and the ‘hairy cicadas’, with the dual roles of increasing insulation and perhaps also reducing absorption of ultraviolet radiation), dark colour (often black, increasing heat absorption), and many are wingless or flightless. This last feature is paralleled in many insects on isolated islands, and is believed to be associated with the high risk of being blown away from a limited suitable habitat. Different patterns of wing development can occur. In the alpine region, fully winged individuals of the early summer population of the ‘wingless grasshopper’ (Phaulacridium vittatum) (Fig. 11.2) fly from lower levels, but many of their alpine progeny are wingless, so that subsequent dispersal is very limited in extent. Overwintering strategies of alpine insects include three major components – movement to lower elevations to avoid extreme cold; overwintering in an immobile non-feeding stage (such as the diapausing eggs of some grasshoppers); and overwintering in a more active stage. In the last of these strategies, insects may either remain active beneath the snow (as are some beetles and cockroaches), or become immobilised over the most extreme winter weather, resuming activity sporadically as warmer conditions occur. Any insect remaining in the region over winter must, of course, tolerate the cold and resist freezing to death. Three main strategies are involved in avoiding freezing. Physiologically, the easiest is to shelter, so that some insects merely dig beneath litter or into soil where they are insulated and protected from ice, to hibernate. If insects remain in more exposed situations they may survive very cold temperatures by ‘supercooling’. Essentially, the insects manufacture their own antifreeze (‘cryoprotectant’) such as glycol, which prevents them from freezing. Still others are cold-hardy: rather than supercooling, they have chemicals
Alpine Insects
153
Fig. 11.3 Female of Acripeza reticulata – held to induce defensive display of raised fore wings and abdomen flexed and expanded to show rings of red, blue and black (Photo: P.T. Green)
such as sorbitol that protect them against freezing, in many cases aided by behaviour by which they seek sun exposure and bask whenever they are able. Some form of temperature adaptation is a critical aspect of survival in this region, with the conditions of becoming either too cold or too hot problematical at different times of the year. During the course of a summer day, an individual cockroach or grasshopper may seek or avoid heat, and its behaviour change in response to this. Sheltering under stones, or moving between sun and shade are both tactics used to exploit varied topography for climatic needs. Some insects ‘shuttle’ regularly from sun to shade and back again in order to maintain optimal temperature, and this habit seems to be particularly common amongst dark insects such as some apterous metallic cockroaches (Polyzosteria spp.) and grasshoppers. However, one group of grasshoppers have a rather different adaptation: they can change colour from darker to paler during the day, and revert to dark towards evening. They thus change their rate of heat absorption, and obviate need for movement to achieve this. The best-known species that does this is Kosciuscola tristis, the Chameleon grasshopper, a truly alpine species found only rarely below about 1,680 m. Its ability – especially in males – to change colour rapidly in response to temperature has led to it being referred to as ‘one of the world’s most remarkable grasshoppers’ in a recent monograph. It and its relatives have a 2-year life cycle. One of the most conspicuous black Orthoptera in the alpine region, where it can be quite common, is the mountain grasshopper (Acripeza reticulata, actually a katydid, despite its common name) which is known also from a number of lowland sites but is most typical of upland areas where, at times, it can be very common. The shortwinged stumpy females (Fig. 11.3) differ considerably in appearance from the slender long-winged males, but both sexes bask in open areas during the day. Males sometimes lie on their side, presumably leading to maximum possible absorption of heat.
154
11 Australia’s Alpine Insects
Fig. 11.4 Oreixenica latialis theddora, one of several alpine/subalpine endemic satyrine butterflies in the south east. The subspecies is endemic to the Mt Buffalo plateau of Victoria, where it can be locally abundant
When disturbed, the katydids expose their abdomen, strongly banded with bands of blue and orange/red, as a ‘warning’ to possible attackers of distastefulness or danger. Day-flying moths, most of them representatives of families whose members are predominantly nocturnal elsewhere, are also a feature of alpine environments. Small hairy species of Arctiidae (tiger moths) and colourful Geometridae, many with their closest relatives in New Zealand montane environments, are particularly characteristic and some are abundant in grasslands and heathlands. Daytime exposure allows heat absorption but may also render the insects susceptible to predators. Many have brightly coloured hind wings, exposed as a startle display, or are highly cryptic at rest. Although their patterns of variation have not been explored fully, many seem to occur in isolated populations on different highland areas, possibly fostering variety. This scenario is better-documented for butterflies. In Victoria, for example, the isolated Mt Buffalo plateau harbours a subspecies of an alpine brown butterfly (Oreixenica latialis theddora, the Alpine silver xenica: Fig. 11.4) known nowhere else. The plateau is separated from any similar alpine area by lowlands, so that the butterfly is separated effectively from any other population of O. latialis. The variation in the related O. ptunarra (Ptunarra brown), which is endemic to highland regions of Tasmania, has led to several subspecies being named, and some of the variation appears, rather, to be clinal in nature, with the appearance of the butterflies changing subtly along environmental gradients. As noted for some other alpine insects, these butterflies are highly localised, but often abundant where they occur – their caterpillars feed on tussock grasses that are predominant in alpine vegetation. A third insect interest of alpine regions is the restricted freshwater fauna, most notably some highly localised endemic stoneflies that are also amongst the largest
Further Reading
155
Fig. 11.5 Thaumatoperla alpina, a large (5 cm) endemic stonefly (Plecoptera) found only in a small part of the alpine region, where it is associated with small permanent streams, resting on a silky daisy (Celmisia sericophylla), itself a plant of National Significance (Photo: David Bryant)
members of the order, and some of which are designated as threatened and of conservation significance. Some are flightless, unusually amongst Plecoptera. The flightless Riekoperla darlingtoni is known only from near the summit of Mt Donna Buang in Victoria, where its habitat includes small streams bordering a car park for tourist vehicles. Larvae can apparently resist desiccation by burrowing into the substrate. The most characteristic alpine stonefly species are the four large brightly coloured species of Thaumatoperla (Fig. 11.5), a Gondwanan relic lineage and each with a separate, highly localised distribution in the mainland alps. Many of the above examples have considerable conservation interest, simply because they are seen as highly vulnerable to climate changes and other threats to this sensitive region.
Further Reading Common IFB (1954) A study of the ecology of the adult Bogong moth, Agrotis infusa (Boisd.) (Lepidoptera: Noctuidae), with special reference to its behaviour during migration and aestivation. Aust J Zool 2:223–263 Green K, Osborne W (1994) Wildlife of the Australian snow-country. Reed Books, Chatswood (includes much background on alpine insects and their environments and adaptations) McCormick B (2005) Bogong moths and Parliament House. Research Brief no 5, Department of Parliamentary Services, Canberra New TR (2010) Butterfly conservation in south-eastern Australia: progress and prospects. Springer, Dordrecht (includes discussion of biology and conservation needs of alpine butterflies) Rentz DCF, Lewis RC, Su YN, Upton MS (2003) A guide to Australian grasshoppers and locusts. Natural History Publications, Kota Kinabalu (excellent aid to identification, with substantial information on alpine species biology)
Chapter 12
Lowland Insects and Their Environments: Non-forest Habitats
Introduction: Terrestrial Open Habitats In the last chapter we noted the importance of alpine grasslands and herbfields as insect habitats, with a substantial number of localised high elevation insects depen ding on these resources, and sometimes being seasonally abundant in what may generally be low-competition environments. Considerably greater variety of grasslands and allied ‘open vegetation’ habitats occurs at lower elevations, and these – ranging from grasslands and herbfields and others with little or limited tree cover, are considered here. These open systems each have ecological peculiarities and insects that are restricted to them, or predominantly found there. Clearly, they also intergrade with more woody systems, so that biotopes such as mallee and open grassy woodlands are in many ways intermediates between grassland and forest – and their characteristic insects also encompass that breadth of variety. Many of these vegetation types are geographically circumscribed. But, as the major entomological contrasts are with the true forest-dependent insects, it is perhaps sensible to consider them together here and to exemplify some of the features they have in common. Forest insects are discussed in the next chapter.
Grasslands Lowland grasslands – together with other low-growing vegetation such as heathlands and low shrubs – are also important insect habitats, largely for different species than those in the alpine areas, and many are now severely threatened through clea rance in much of the country. Only about 150 years ago, native grasslands were widespread across lowland Australia. They were changed rapidly in the south east following European settlement. The combination of flattish land and few trees, with soil, rendered grasslands and grassy woodlands attractive for conversion and they were the first major vegetation types to be transformed extensively for large scale agriculture (both cropping and pastoral) with resulting severe degradation. T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_12, © Springer Science+Business Media B.V. 2011
157
158
12 Lowland Insects and Their Environments: Non-forest Habitats
Pasture development was accompanied by introductions of alien grasses for ‘improvement’. These changes have led to recent appraisals that lowland native grasslands in the south east are the most threatened ecosystems in Australia, and many formerly very extensive tracts of grassland there have been lost or reduced to tiny patches within their former range; insects depending on those grasslands have suffered in parallel, and some are now of major conservation concern. The grassland milieu is also subject to invasion by a number of aggressive alien weeds and insects, so that the direct effects of habitat loss are compounded by impacts of alien species. In general, though, many grassland insects are poorly known. Widespread lack of information has been attributed in part to grasslands not being ‘charismatic’ habitats, particularly in relation to old growth forests that are much more appealing to people. This is associated with records of even the best-known insect groups (such as butterflies) being rather sparse. Losses of grasslands have helped to emphasise our ignorance of associated losses of insects and their relatives, and the few species highlighted are likely to only indicate much wider actual losses. Morabine grasshoppers, a group of flightless Australian species treated as a subfamily of the monkey grasshoppers (Eumastacidae), comprise around 250 species, many of them associated with grasses. A number of species have been lost as grazing has changed the composition of vegetation. Studies on the New England tablelands of New South Wales, for example, revealed that replacing the original Themeda grassland by shorter-growing Stipa and Danthonia led to complete change of the grasshopper fauna and loss of some species. One notable morabine, Keyacris scurra, is now largely restricted to small protected remnant native grassland patches – these include pioneer cemeteries, long fenced and excluding grazing stock, and so becoming reservoirs of the native biodiversity long since eliminated from surrounding areas. Many morabines seem to have very low reproduction, and produce only around 20 eggs over a lifetime: populations can thus be naturally small, as well as circumscribed. Native grassland continues to be threatened, particularly nowadays close to cities as urban and industrial expansion accelerates, and – whilst grasslands are complex and varied systems – most are extremely difficult to protect from these influences. Tiny protected patches, such as the cemeteries noted above and road or rail reserves are important reservoirs for plants and insects alike, but are both accident-prone and susceptible to edge effects, and can easily be mismanaged. They are vulnerable to accidents or well-intended management steps, such as fire, heavy machinery impacts, mowing or over-sanitation, in addition to their small size rendering them susceptible to alien species invasions. The patches that remain are progressively fragmented and isolated, so that insects living there are faced increasingly with problems of moving elsewhere should this need arise, and their populations become increasingly functionally independent. Genetic survey of one flagship species for south eastern native lowland grasslands, the golden sun-moth (Synemon plana, one of about 40 species of the family Castniidae in Australia, many of them narrowly endemic), showed that populations differed considerably across the moth’s geographical range, most likely reflecting the consequences of isolation and reduced population size and variability. It is even considered possible that the various population clusters revealed
Grasslands
159
Fig. 12.1 Male (upper) and female of the golden sun-moth, Synemon plana (Castniidae), illustrating strong sexual dimorphism. Males fly actively in fine, calm, warm weather but only for a few hours around the middle of the day. Females fly little, but attract overflying males to land and mate by exposing their bright hind wings. Adults do not feed, and live for only a few days. The moth, one of several members of an endemic group of Castniidae of major conservation concern, is an important flagship for native grasslands in south eastern Australia
by this analysis could comprise distinct subspecies: the Victorian populations exa mined differed considerably from those in New South Wales and the Australian Capital Territory. Insects such as flightless grasshoppers and the sun-moth are particularly susceptible to the effects of habitat loss and isolation, simply because their dispersal powers are limited and, especially with surrounding inhospitable terrain, they are not likely to encounter new habitat patches. Even if they could physically disperse, the surrounds are essentially hostile. The golden sun-moth has a number of unusual features that endorse this restriction. First, the mouthparts are poorly developed, and adult moths cannot feed. Second, and associated with this, they live for only a few days, certainly for less than a week, with 3–6 days apparently usual. And, third, females seem reluctant to fly much (some perhaps do not fly at all), and males fly only under rather limited conditions. The two sexes differ greatly in appearance (Fig. 12.1) and behaviour.
160
12 Lowland Insects and Their Environments: Non-forest Habitats
As the common name implies, males fly in bright sunlight, normally for only up to a few hours in the middle of the day, and in calm and dry conditions. If the wind rises to more than a gentle breeze, cloud cover increases markedly, or temperatures are below the low 20s (°C), they are inactive or if conditions decline whilst they are flying, they settle. Under good conditions, males fly fast and about a metre above the ground, roaming over the grassland. Female Synemon fly relatively little. They rest on the ground, exposed between grass tussocks and display their brightly coloured hind wings as males fly overhead. Males respond by settling, and mating ensues. Females oviposit in the bases of grass tussocks and caterpillars feed underground – perhaps for 2 or even 3 years before they complete development. It is very difficult to estimate population size of golden sun-moths. At any site, but starting somewhat earlier in the season inland than further south, the moths may be visible over a continuous period of approximately 6–8 weeks, but each individual is present for only a few days so that a continuing rapid turnover of moths occurs: those we see now will be dead before any inspection a week ahead, for example, and have already replaced any present a week or so ago. Over any extensive site (and despite the caveats above, a few remaining grasslands are of several hundred hectares) microclimate differences (such as of temperature regimes from different aspects, vegetation cover, or slope) may correlate with moths emerging earlier or later in the season, so that even the spatial pattern detected in any single visit may be an artifact. And, with the length of the life cycle being unknown, even the best possible counts from repeated visits may represent only 1 year cohort of the total population present. The moth has received considerable attention recently as a flagship species for native grassland conservation. Whereas Synemon is of major conservation interest, some other grassland moths are very abundant and widespread. The hairy caterpillars of some anthelid moths (Fig. 12.2) can occur in enormous numbers, for example, and are conspicuous as they climb grasses during the day. However, some of the insects discussed above are amongst those that are scarce and localised on grasslands, but wholly or largely dependent on it and are of both evolutionary and conservation interest. Their future, as for that of many others in the same environments, depends on grassland protection from numerous outside influences, such as development, hard-hooved stock compacting soils, and floristic change – particularly from alien weeds. However, a few cases have occurred in which native insects have benefited markedly from pasture improvement, with this leading to increased abundance and, sometimes, further conservation concerns elsewhere. Christmas beetles (Anoplognathus spp.: Scarabaeidae, Rutelinae) are large iridescent scarabs sometimes abundant in mid-summer. They have a complex life cycle which involves alternating between grasslands (as larval habitat) and Eucalyptus trees (for adult food). Female beetles lay eggs in grassland, and the grubs feed underground on grass roots. Emergent adults fly to eucalypts, where for some weeks they feed on foliage, mature, mate and eventually return to grasslands to complete the characteristic annual cycle. With increased pastoralisation, such as on the tablelands of central New South Wales, many eucalypts have been felled to clear land for more pasture, leaving simply scattered older shelter trees for stock.
Grasslands
161
Fig. 12.2 Anthelid moths (Anthelidae), a notable radiation in Australia, include a number of species (many of the poorly-understood genus Pterolocera, but also some of Anthela, here A. ocellata) with hairy caterpillars sometimes abundant and conspicuous on grasslands
Pasture improvement and fertiliser applications were changes that (1) provided larvae with a much greater area of habitat with richly nutritious food and (2) lessened the supply of adult food. The consequence has been that large numbers of healthy larvae could complete development, and those augmented numbers of adults were concentrated on lower numbers of trees, producing intense defoliation as a major contribution to rural dieback, in conjunction with other leaf-eating insects. Christmas beetles do not act alone, and the pattern of seasonal attack leading to dieback has three phases: (1) new foliage in spring is attacked by larvae and adults of leaf beetles (Chryso melidae, Paropsini), as well as by other defoliators such as Lepidoptera; (2) mature foliage in early summer is then eaten by Christmas beetles; and (3) later in the summer regrowing juvenile foliage on previously defoliated trees is eaten by new generations of leaf beetles, with later flush growth susceptible to frost as the season advances. Other agents are also involved in dieback: fungi (particularly root-rots caused by Phytophthora), soil salinity and drought are cited commonly, as examples. Many Australian insects, in addition to an array of introduced species, have been able to capitalise on improved grasslands, and some are regarded as serious pests either as sporadic or seasonal migrants (e.g. locusts, some moths) or more permanent residents. Before European anthropogenic changes, it is presumed that the native species were present in grasslands, but of course without their later notoriety, which has arisen largely within the later improved pastures of eastern and southern Australia. Thus, in Tasmania, the five major pests of improved pastures comprise two species of ‘corbie’ (swift moths: Oncopera intricata and O. rufobrunnea) and three scarabaeid beetles (Aphodius tasmaniae, A. pseudotasmaniae, Adoryphorus couloni). All are Australian endemics, and the first four occur only in Tasmania – and all pose serious economic problems at times.
162
12 Lowland Insects and Their Environments: Non-forest Habitats
Arid Environments The more arid regions of Australia include vast areas of ‘unimproved’ pasture – varied rangelands vital for stock grazing, but subject also to considerable selective floristic changes as this occurs. Whilst elevation, and its accompanying climatic extremes in the snow country is one of the more predictable and challenging environmental gradients with which insects must contend through adaptations to cold, aridity is also a severe challenge to insects, so that those dwelling in the interior of Australia – the most extensive realm in the country – must also cope with physiologically challenging conditions. Well over 70% of Australia has been classified as arid or semiarid, reflecting the extent of the Eyrean region discussed earlier (p. 41), variously providing insects with hot, dry environments with little and highly unpredictable precipitation and long periods of drought. However, it is far from true desert as the term is applied, for example, in North Africa, and contains many distinctive vegetation associations. Some grasshoppers have been categorised as either ‘drought-evading’ or ‘drought-enduring’ species. The first simply live only in wetter areas, such as close to water bodies, and may be very localised. The latter can withstand long inclement periods by adaptations such as dormancy. Most species have not been studied individually, but some have eggs that can remain in soil for several years and hatch only once moisture is present, after which the insects grow and mature rapidly. In some species, even all the eggs in a single pod may not hatch together, so that some are ‘kept in reserve’ as a further counter to unpredictability. Ecology of arid zone grasshoppers has received considerable attention because of locust outbreaks, but the dynamics of many other insects are more poorly known. In parts of northern Australia a well-defined rainy season provides strong seasonal contrast in aridity, and imposes strong influences on insect seasonality. One notable species of the ‘Top End’ of the Northern Territory is Leichardt’s grasshopper (Petasida ephippigera), distinctively patterned in red, navy blue, and black and feeding on Pityrodia on rocky escarpments in sandy areas: surprisingly, in view of its bright dramatic coloration, which has accorded the grasshopper considerable cultural significance and led to its adoption as a tourist icon – it has also appeared on two stamps – it seems not to be poisonous. Also surprisingly, it was not recorded for at least 70 years following early records in the mid-nineteenth century (the first specimens were captured on surveys by the Beagle, most probably in 1839). Whether it genuinely undergoes very long-term population fluctuations, with implications of range shifts through a long-term metapopulation effect, as has been suggested both for this and some other grasshoppers in the region, is unknown. Its scarcity over this long period has been attributed also to the general remoteness of the areas in which it occurs, and lack of focused searching. Adults are present during the wet season, when access is difficult or impossible due to flooding, and the then hot and humid conditions render searching arduous.
Mallee Environments
163
Mallee Environments In contrast to grasslands and heathlands, the widespread ‘mallee’ systems concentrated toward the southern parts of Australia but extending far northward as well have been described as ‘a sea of scrub’, a phrase derived from early explorers, and are in many ways transitional between the southern open savanna systems and more arid-adapted flora to the north. Mallee supports both arid-adapted and more mesic biota, reflecting the erratic rainfall patterns of the area. Much of it occurs on sandy soils, although a considerable variety of soil types occurs, ranging from deep sands to shallow soils overlying calcrete. The term ‘mallee’ refers to the multistemmed growth form of the eucalypts found in this vegetation, fostered through their usual regrowth pattern of regeneration from basal lignotubers, following fires. They become sparser to the north, but the major vegetation subcommunities of mallee are defined largely on the complement of eucalypts present, in addition to understorey. The rich diversity of shrubs is associated with parallel richness of some insect groups, such as jewel beetles (p. 94). Although the eucalypts do set seed, sometimes abundantly, seedling trees are often very sparse in mallee, and one contributor to this is that many seeds are carried away by ants. Indeed, ants are one of the most diverse and important insect groups in these systems and, more widely, in arid and semi-arid ecosystems in Australia. Mallee probably constitutes an environment very favourable to rich ant diversity, by combining bare ground (unimpeded foraging, insolation, sites for nest construction), litter (shelter, food resource), and plants from herbs to small trees (for foods such as seeds, honeydew, and invertebrate prey for various ant consumers). With rather few specialised arboreal forms, most mallee ants seem to be predominantly ground foragers, some of which move sporadically onto trees. The faunas have several characteristic features, namely (1) many ant species are present; (2) they include large numbers of Iridomyrmex (Fig. 12.3), Camponotus and Melophorus; (3) they show functional groups based on activity patterns, with main categories being nocturnal, day active at high temperatures (Melophorus, in particular), or day active at low temperatures; (4) most species nest in soil, and forage on the ground and, often, also on vegetation; and (5) the predominance of Iridomyrmex, assuring that interspecific competition is an important structuring influence in the ant assemblages. Because ants have been reasonably well surveyed in many parts of Australia, their biogeographical patterns and affinities can also be determined, so that the genera found in mallee environments can be attributed as having centres of distribution characterised as ‘Eyrean’, ‘Torresian’, ‘Bassian’ or simply ‘widespread’. More arid mallee systems tend to have greater proportions of widespread genera and lower representation of Bassian forms, but both vegetation and climate influence ant distributions. Based on the abundance of individuals, ants are the predominant insect group foraging on eucalypts in mallee systems. They, and termites, have considerable ecological influence there. Whilst many ant genera are indeed abundant and widespread, as above, others are much more restricted and elusive. Paramount amongst these is the ‘dinosaur ant’
164
12 Lowland Insects and Their Environments: Non-forest Habitats
Fig. 12.3 Nest site of an Iridomyrmex species in semiarid country: a large area has been entirely cleared by the ants to produce a characteristic stony open nest site
(Nothomyrmecia macrops), accepted generally as among the most primitive living ants and so of interest to biologists in interpreting ant relationships and the origins of their social behaviour. Nothomyrmecia had for long been known from only two individuals collected in Western Australia, with repeated searches in the area over more than 40 years following the initial discovery in 1931 failing to rediscover it. By then it had become the stated ‘holy grail’ of ant specialists! When found in mallee near Poochera, South Australia, in 1977 it aroused considerable international interest but – despite progressively complete understanding of its biology based on extensive field and laboratory study – Nothomyrmecia has not yet been rediscovered in Western Australia. Unusually amongst ants, it forages nocturnally and is thus not susceptible to daytime searches, and ants return to their underground nests around dawn. Nothomyrmecia are predominantly nectar-feeders. As with the termites we noted earlier, daily and seasonal activity patterns may be honed by temperature and other climatic needs and these restrictions are perhaps nowhere more evident than amongst insects living in either very hot or very cold climates. However, those conditions can also be approached by living inside plants, and the numerous insects that feed on dead wood are one prime example of this. They are a significant component of insects found in forest environments.
Further Reading
165
Further Reading Clarke GM (2000) Inferring demography from genetics: a case study of the endangered golden sun moth, Synemon plana. In: Young AG, Clarke GM (eds) Genetics, demography and viability of fragmented populations. Cambridge University Press, Cambridge, pp 213–225 (case study of genetic variation related to habitat fragmentation in grassland) Gibson L, New TR (2007) Problems in studying populations of the golden sun-moth Synemon plana (Lepidoptera: Castniidae) in south eastern Australia. J Insect Conserv 11:309–313 Heatwole H, Lowman M (1986) Dieback. Death of an Australian landscape. Reed Books, French’s Forest (discussion of causes and significance of rural and forest dieback of eucalypts in Australia) Taylor RW (1978) Nothomyrmecia macrops: a living-fossil ant rediscovered. Science 201:979–985 (the rediscovery and significance of this notable ant)
Chapter 13
Forest Insects
Introduction: Forest Habitats Forests provide some of the most stable and enduring terrestrial habitats for insects. Botanically and structurally complex plant communities dominated by trees provide a strikingly three-dimensional environment that can be shared amongst numerous species, many of which are restricted to the higher canopy and are rarely encountered by entomologists working on the ground. Indeed, awakening awareness of tropical insect diversity came with exploration of canopy beetles in the neotropics during the 1980s, when insecticide applications by mist-blowers hoisted on ropes to the treetops revealed that many of the insects retrieved as they fell to the ground were not encountered at lower levels and so were apparently restricted to these upper levels of the trees. The northern tropical rainforests are of major interest in being the last vestiges in Australia of an ancient system (designated by some as the oldest type of vegetation in Australia) that once covered up to a third of the continent, and now harbour remnant enclaves of biodiversity long absent from the wider landscape. Most remaining rainforests are within about 100 km of the east coast, where extensive clearing has occurred to leave many as small fragments. Collectively, however, around 70% of the rainforested area estimated to be present around the time of European settlement still remains, with the major concentrations in Queensland (tropical) and Tasmania/ Victoria (temperate). In general, tropical rainforests support some of the most diverse of all insect assemblages. Those of northern Australia are no exception, and are suggested to parallel those of south eastern Asia and elsewhere in their complexity and richness. The complexity of their environments is clearly conducive to insect diversity – as one example, around 60% of Australia’s butterflies occur in north east Queensland forests. However, this total is equivalent to only about a third of the number of butterfly species in Papua New Guinea forests and, together with parallel estimates for some other insects, suggests that Australian tropical rainforests are in fact far less rich in insects than many of those of south east Asia.
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_13, © Springer Science+Business Media B.V. 2011
167
168
13 Forest Insects
Nevertheless, measures of insect richness in the Wet Tropics area of northern Queensland are still very impressive, and a number of surveys at particular sites and along altitudinal transects using various techniques have revealed high richness of arthropods on the forest floor, and also that many taxa have very narrow ranges within this greater areas, so appear to be true narrow range endemics. Thus, a survey at Bellenden Ker yielded 4,029 species of insects (amongst a total of 4,385 arthropods), including 1,514 species of beetles and 1,345 Lepidoptera. Stream communities in the Wet Tropics are also diverse, and dominated by mayflies, caddisflies and non-biting midges. To the south, these forests give way gradually to cooler temperate forests and more open woodlands, all with local specialisations and supporting endemic plant and animal species. Rainforests are unusual in extending from the tropics to cool temperate zones and also in not being dominated by Eucalyptus. Tree species richness tends to decline from north to south, with key species in southern rainforests including Nothofagus cunninghamii (myrtle beach, p. 59) and southern sassafras (Atherospermum moschatum), whilst many of the other southern forests are dominated by one or more characteristic species of Eucalyptus – the mountain ash (Eucalyptus regnans) of southeastern Australia is amongst the world’s tallest and most long-lived trees. As a contrast with northern forests, numerous moths are associated with eucalypts, but Eucalyptus is almost wholly neglected as a food plant by butterflies (so that their northern diversity depends on other plants). Nevertheless some depend indirectly on eucalypts because these are the hosts of the mistletoes on which caterpillars feed, or provide pupation sites under loose bark. Thus, in Tasmania caterpillars of the Australian hairstreak (Pseudalmenus chlorinda) feed on Acacia foliage but pupate under bark of nearby eucalypts, to which they move when fullygrown. In contrast, and as noted previously, Eucalyptus supports substantial radiations of some other insect groups such as paropsine leaf beetles (Figs. 13.1 and 13.2) and pergid sawflies (Fig. 13.3). Much of Australia is not forested, and substantial parts of the forests over the approximately 20% of the country with this vegetation have been cleared or disturbed severely. The Gondwanan elements of southern forests help to characterise the local environment well – the presence of Nothofagus denotes cool wet forests, usually densely shaded and with abundant moist litter, for example.
Forest Insects The insects can be equally characteristic. Moss bugs, Peloridiidae, a family of small plant bugs, have considerable evolutionary interest in long being thought to be a very basal ancestral group of the more advanced plant bugs, and historically have been regarded as intermediate between the two major suborders of Hemiptera. They occur in wet forests of eastern Australia as well as in New Zealand and southern South America, and are most typically found in wet mosses growing on Nothofagus trees. Except for one South American species, moss bugs are flightless, and each
Forest Insects
169
Fig. 13.1 Adult leaf beetles, this one Paropsis atomaria (Chrysomelidae), are a major radiation on eucalypts, and some can occur in enormous numbers, as defoliators
Fig. 13.2 Paropsine beetles lay eggs in groups, and the gregarious larvae are chewing defoliators on eucalypts
genus is highly localised within the family ranges, so that different groups of genera occur in the southeast, southern Queensland and north Queensland, and one further genus is endemic to Lord Howe Island.
170
13 Forest Insects
Fig. 13.3 Adult eucalyptus sawfly, Perga sp.. Pergidae are by far the predominant family of Hymenoptera: Symphyta in Australia, and are believed to be Gondwanan in origin
The butterflies and bugs mentioned above are associated with epiphytic plants or resources provided indirectly by the forests, rather than with the trees themselves. They exemplify the many insects of ‘fringe habitats’ that could not exist without the trees but which we do not think of directly as forest habitats. Another example is the initially unprepossessing complex of ‘tree holes’, in which aquatic insects occupy small waterbodies on and in trees, and in the axils of epiphytic plants on trees. Holes in trees comprise two main categories, reflecting their origin – many are rot holes, due to breakages of branches or trunks, leaving open cavities that may be – or become – waterproof. Others remain covered in bark and are ‘pans’ – depressions in buttress roots or other structures, and sometimes formed when different plant parts grow together. Gradual accumulation of litter in these, sometimes tiny, holes provides the foundation for progressively complex assemblages of organisms to live. A few Odonata are even thought to be wholly dependent on this habitat, and many flies and beetles are not uncommon there. The variety of environments in forests includes the litter and understorey, but many insects are associated primarily with trees themselves. ‘A tree’ is a complex environment which can be shared by many insect species feeding in different ways and in different places. A single longlived eucalypt can provide the living environment for insects over up to hundreds of generations, providing both a stable basic environment with assured supply of wood and foliage, and reliably predictable seasonal resources such as flush growth, flowers and seeds. Complex interactions and co-evolutions are associated through this stability, and are the basis of the major radiations of plant-feeding insects that occur (p. 119). Small variations in foliage chemistry, for example, may determine host plant acceptability or preference to an insect, and seasonal variations in this, in leaf toughness, or nitrogen content as an index of food ‘quality’, may each influence insect development and seasonality. The more seasonal resources are also critical,
Forest Insects
171
in part for the long term opportunities they provide. However, a tree’s utility and value to insects persists long after its death. Saproxylic insects – those feeding on wood – are a major category of interest. Dead standing or fallen trees decay over, perhaps, many decades. Dead wood, surprisingly to many people, is an immensely complicated resource for insects. Parameters such as trunk or stem diameter, standing or fallen, whether with or without bark, humidity and exposure to sunlight, extent and kind of fungus infestations and rot, and degree of degradation all determine its suitability for particular insects – so that woodfeeding beetles can form long-term successional communities within dead trees. Many depend for food not on the wood alone but on particular fungal infections that occur at definable stages of decay, and may feed only on either phloem or xylem tissues at earlier stages of wood decay. The primary insect groups involved are beetles and flies, but a large array of their predators and parasitoids also participate in the communities. Saproxylic insects are very susceptible to forest management practices – the usual primary objective of forestry is removal of wood, first as a harvestable commodity but also involving removal of waste from the forest floor to facilitate access, reduce severity of wildfires, and to prepare sites for plantation plantings. ‘Old wood’ is a critically important resource, with some northern hemisphere studies showing that some beetles depend wholly on this, so that ancient or ‘veteran’ trees are almost irreplaceable. The importance of huge ancient eucalypts has been expressed in Australia mainly as providers of natural rot holes and similar nesting cavities needed by marsupials and parrots, including icon species such as Leadbeater’s possum in Victorian highland forests, so that their loss may have direct impacts on such high profile animals. Their importance for insects is likely to prove just as great. Commercial forestry cycles, even those rotations projected up to 80 years or more, do not allow time to replace trees of this kind, and cannot cater for many saproxylic life forms needing older trees, whilst removal (by burning, chipping, or directly) of dead trees and fallen wood and wastes may be a threat to many other species. Assuring the successional supply of dead wood, as well as providing living trees, is a major insect conservation imperative in forests. Wood-feeding insects vary enormously in the length and complexity of their life cycles, but the persistence of older wood allows some species to become large and develop slowly. Many saproxylic beetles and moths have life cycles lasting several years. The ‘goat moths’ (Cossidae) include some of Australia’s largest moths, for example, and some of the long-horned timber beetles (Cerambycidae) also have larvae reaching up to 6–10 cm long. The spectacular appearance of these insects, together with the stag beetles (Lucanidae) has rendered them highly desirable to collectors, who may be tempted to obtain their specimens through breaking up dead wood and destroying the habitat on which the insects depend. Some of the species are rare, and becoming more so as their habitat declines, and the incentive of high black market prices is a further conservation problem – a single (non-Australian) stag beetle was reputedly sold for almost $US 100,000 a few years ago! Many stag beetle species are individually very variable, so that larger or more spectacularly ornamented ‘trophy’ specimens are more desirable than ‘ordinary’ individuals to collectors. The genetic influences on natural populations of taking such individuals to leave smaller ones are unknown.
172
13 Forest Insects
About 30 species of Lucanidae occur in Tasmania, most of them endemic to the state, and considerable conservation concerns have arisen over their susceptibility to forest clearing and conversion of native forests to plantations of monoculture eucalypts or alien softwood pines. The extent of suitable habitat there continues to decline. The moist eucalypt forests needed by the broad-toothed stag beetle (Lissotes latidens) are one example – both larvae and the flightless adults occur in upper soil layers beneath rotting logs, and removal of these has contributed to beetle declines. Most of the remaining populations, other than on Maria Island, largely a National Park, are not in protected areas. The beetle is associated particularly with large wellrotted logs, and extensive searches (using an excavator to roll logs, and subsequently replace them to conserve the sensitive microclimate and community present) revealed very few individuals, and suggested that it is genuinely scarce. Several species of another genus (Hoplogonus) are associated with mature forests and believed to be very susceptible to clearing and later burning of remnant debris. Forest insects in Australia thus arouse needs for conservation of many rare and localised taxa and assemblages, but also needs to understand and manage a variety of taxa that cause economic damage to trees harvested or planted for forestry, all within the context of daunting ecological complexity. Much of the foundation knowledge of forest insects has come from studies on pest species, so that interim compilations on forest pest insects and how to control them were made a century and more ago, with State and Commonwealth governments appointing specialist forest entomologists to investigate the biology of some ‘key’ species and advise on their management. Although the foundations for this were early, major expansion occurred only from around 1960 onward. Somewhat ironically, much of the initial impetus for this expansion arose from an alien pest on an introduced tree species! The Sirex woodwasp (Sirex noctilio) causes serious damage and losses to plantations of Monterey pine (Pinus radiata), introduced widely as Australia’s predominant softwood crop. Although known from New Zealand from around the beginning of the twentieth century, S. noctilio was found first in Australia, in a pine plantation near Hobart (Tasmania) only in 1952 and is believed to have been introduced in wood from New Zealand. In its native Europe, this wasp is unique amongst woodwasps in being able to kill trees, and is largely specific to Pinus. In general, though, healthy trees can resist attack, and physiologically stressed trees (including those recently felled) are chemically attractive to female wasps, which drill holes through bark to lay eggs into the underlying sapwood. Larvae feed within the wood, particularly on a symbiotic fungus (Amylostereum areolatum) that is also introduced with a toxic mucus during oviposition. Small pine trees can be killed by attack from only one female wasp, and the high values of the softwood industry – involving now around 700,000 ha of P. radiata – led to a campaign to seek biological control agents for S. noctilio and to understand its biology in considerable detail. Surveys for parasitoids of woodwasps were undertaken over much of Europe, from a research unit based in England, and both the relationships between the tree and the symbiotic fungus, and reasons for tree susceptibility received considerable attention within Australia. Healthy pine trees can resist attack by Sirex, and two mechanisms of resistance were clarified in the 1960s. First, strong vigorous trees often had the capability to ‘flood’ wasp oviposition holes
Forest Insects
173
Fig. 13.4 The two-phase life cycle of the nematode worm, Deladenus siricidicola, a parasite and biological control agent of Sirex noctilio (Based on Taylor 1981)
with resin, drowning eggs and small larvae and with the resin also inhibiting or killing Amylostereum. Second, some trees could isolate the fungus by producing a chemical barrier of polyphenols. With water stress, in particular, both these mechanisms broke down and trees became susceptible. Following early release in Tasmania of two wasp parasitoids already present in New Zealand, 19 further species were introduced during the course of the control programme, with control anticipated through use of parasitoids in two complementary ‘guilds’. Two species of Ibalia (Cynipoidea) attack first and second instar Sirex larvae, so can have impacts soon after the eggs hatch. The more diverse guild of Ichneumonidae (Rhyssa, Megarhyssa) and Stephanidae (Schlettererius) attack older larvae, using their very long ovipositor to probe wood and lay eggs on the outside of host larvae, as ectoparasitoids. However, the most innovative control methods deve loped include parasitic nematode worms (Deladenus siricidicola). These nematodes have both a free-living fungus-feeding cycle and a parasitic cycle in their complex life history (Fig. 13.4), with juveniles of the first able to follow either pathway, depending on conditions. Once inside the host wasp, nematodes migrate to the reproductive organs developing within the pupa. In female hosts, they sterilise the insect so that, although oviposition occurs apparently normally, each eggshell contains
174
13 Forest Insects
the fungus and many nematodes, rather than any wasp offspring. The intriguing complexities of this association have been of major interest and importance in thwarting Sirex in Australia, and protecting this major forestry crop. The many native insects affecting forest trees fall into several major ecological guilds, as defoliators, sap feeders, and wood/bark/shoot feeders, and each contains an array of different insects, within which sap-feeders are predominantly Hemiptera, and most wood-feeders are Coleoptera or Lepidoptera. Smaller guilds include seedeaters and gall-formers. Members of any of these groupings can be pests should their activities come into conflict with the aims of forestry, but collectively they also have massive importance in sustaining the ecological integrity of forest environments. From that viewpoint, measures to manage native insect pests in forests must be undertaken within the context of diverse natural communities that, ideally, should not be harmed. This context differs greatly from much crop pest control, in which the target is most commonly an alien species on an alien plant host growing as a monoculture in an artificially simplified environment. Effective pest management for tree pests has in general not progressed to the same extent as for many field cropping systems. Many forest pest control measures have been initiated simply in response to large numbers or outbreaks of key insect species. The possibilities are exemplified by the integrated control programme developed for a native foliage-eating leaf beetle, Chrysophtharta bimaculata (Chrysomelidae) in eucalypt plantations (mainly of Eucalyptus nitens) in Tasmania. Most damage is caused by late instar larvae which, as in most other insect herbivores, eat many times more than young larvae as they grow. Any control of the eggs or young larvae may thereby help to prevent later economic damage, and these stages are indeed amenable to predation by native beetles. The two most important species are a soldier beetle (Chauliognathus lugubris) and a ladybird (Cleobora mellyi), with the latter eating leaf beetle eggs and larvae as both larvae and adults. Together, they can make substantial inroads into the Chrysophtharta population and, if populations of larger larvae remain unduly high, they can then be targeted by applying a synthetic pyrethoid insecticide at that later time. Despite the notoriety of some forest insects as pests, most insects living in Australia’s forests and woodlands remain very poorly known. Outbreaks (p. 134) of some species create sporadic wider attention, but it is also clear that many forest insects occur only in low numbers, rarely (if ever) increase conspicuously in abundance, and are rarely seen even by entomologists. As in other key terrestrial habitats in Australia, trapping programmes and surveys for insects continue to reveal taxonomic novelties, some difficult to interpret and with their background biology obscure.
Further Reading Bedding RA (2009) Controlling the pine-killing woodwasp, Sirex noctilio, with nematodes. Prog Biol Control 6:213–235, (development of long term studies on uses of nematodes in control of Sirex)
Further Reading
175
Elliott HJ, Ohmart CP, Wylie FR (1998) Insect pests of Australian forests: ecology and management. Inkata Press, Melbourne Phillips C (1996) Insects, diseases and deficiencies associated with eucalypts in South Australia. Primary Industries, Mt Gambier (useful illustrated synopsis of insects present) Taylor KL (1981) The Sirex woodwasp: ecology and control of an introduced forest insect. In: Kitching RL, Jones RE (eds.) The ecology of pests. Some Australian case histories. CSIRO, Melbourne, pp 231–248
Chapter 14
Insects and People in Australia
Introduction: Interest and Involvement Australia’s insects are no less unusual than our native marsupials, parrots and other more charismatic animals that are major attractions to international visitors. They are, however, much less noticed and acknowledged. Indeed, most of Australia’s native insects intrude little on human consciousness. People may casually admire a Ulysses butterfly (and even pay to see living native butterflies as exhibits in Butterfly Houses), be fascinated by large termite mounds or trails of foraging ants, buy dragonfly larvae for fishing bait, counter bushflies with the ‘Aussie salute’, swat mosquitoes, and fear encounters with bull ants. However, the few species that come to general notice, positively or negatively, are by far outnumbered by those that are not encountered, observed, or even known. Disparaging epithets such as ‘the cockroach’ or the unspecified ‘bug’ mask diversity of form and ecology that is rarely considered by people other than biologists. The strongest and most widespread public reaction to most insects and their relatives remains hostility, based on (often vague) fears, dislike, ignorance of what they may contribute, and media conditioning to remove, kill or otherwise suppress them. Yet, some are widely and readily admitted as worthy or beneficial. Papilio ulysses, above, is a tourist icon for tropical north Queensland, and stencilled illustrations of the dinosaur ant (Nothomyrmecia macrops, p. 164) adorn the township of Poochera, South Australia. Many people admit to liking butterflies (although the term ‘moths’ for very similar insects may elicit the opposite reaction) and some insects have even been imported to Australia to fulfill human wants. Honeybees, for example, were amongst the first deliberately imported insects, as the basis of a long-thriving apiculture industry, initially predominantly for honey but nowadays also as a major contributor to crop pollination systems. Silkworms were introduced in the middle of the nineteenth century, and many more recent introductions have been as agents for classical biological control. A few classic cases, such as use of the South American moth Cactoblastis cactorum to control prickly pear cactus (Opuntia) in the early 1930s, have entered Australia’s rural history. Prickly pear cactus was brought with the First Fleet to establish a cochineal industry
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_14, © Springer Science+Business Media B.V. 2011
177
178
14 Insects and People in Australia
from the insects feeding on it: the British Government was keen to establish a source of supply of this bright red dye (used for military uniforms, amongst others) independent of the monopoly then held by Spain and Portugal. The cactus was cultivated for fodder from the early 1800s but in the 1920s ‘exploded’ to spread rapidly and occupy vast areas of land. The dramatic effects of Cactoblastis in reducing the cactus within only a few years are well-documented. This, and the more recent use of the South American weevil Cyrtobagous salviniae against salvinia (Salvinia molesta) (and with the weevil detected as a previously undescribed species only during that study), are amongst the all-time success stories of alien weed biological control by imported insects. African and European dung beetles (p. 74) and a considerable variety of predators and parasitoids of arthropod pests have also been released, with varying outcomes, but illustrating the wide needs for such ‘natural enemies’ from elsewhere in the world to feed on alien species that our native herbivores, predators or parasitoids do not attack because they have not evolved in association with them and lack the flexibility to attack them. The principle of ‘classical biological control’ is to seek to restore this relationship by importing species to which the target pest may naturally succumb. Any such introduced agent is likely to become a permanent addition to Australia’s fauna, so that conservation concerns may arise over its wider effects on native species. Europeans were by no means the first to appreciate Australia’s insects. Aboriginal people exploited many for food – the presumed well-known ‘witchetty grubs’ are, in fact, difficult to identify accurately, because the term has been applied seemingly to various beetle larvae (mainly timber beetles, Cerambycidae, p. 31) and moth caterpillars (both of swift moths and goat moths, respectively Hepialidae and Cossidae, almost wholly root-feeding or wood-feeding species), so that any specific record of the grub may not be identifiable even to order level. Some entomologists consider that the term (as ‘witjuti grub’) is most properly applied to large cossid moth caterpillars. The allied term ‘bardi grub’ was first applied to large beetle larvae from grass trees (Xanthorrhoea) but now also has much wider casual applications. Aggregations of Bogong moths in the high country were an important seasonal source of rich fatty food, and many others, such as honeypot ants (Melophorus), ‘sugar bag’ bees (Trigona), the sugary lerp coverings of psyllids, and some scale insects contribute at various levels to food supply. Only in 1951, a notable authority started a book chapter on insects as food in Australia with the statement ‘Entomophagy is common all over Australia’. This seems somewhat of an overstatement, and insect–eating is thought to have declined since European settlement. However it has recently started to expand, with revival a novel facet of tourism and associated with increasing use of native foods (‘bush-tucker’) for restaurants. Many aboriginal food insects were also local tribal totems, with their community importance emphasised in ceremony. Traditionally, the didgeridoo is made from lengths of mallee eucalypts hollowed out by termites. Our varying perceptions of insects are an important component of how we live with them and learn to understand them. Most of the more detailed studies of ‘how insects work’, for example, have come from species that influence human wellbeing, either negatively (as ‘pests’) or positively, and which we wish to manipulate to
Pest Insects
179
either reduce adverse impacts – for instance on crops or stored commodities or as carriers of disease – or to increase benefits, such as crop pollination. In both contexts, accurate identification, or consistent recognition to species level may be critical in determining the outcome. Crop pests may need expensive and complex methods to suppress them, and beneficial insects such as predators and parasitoids, likewise, to increase their impacts. In either case, a misidentification may lead to an unsatisfactory outcome, whereas accurate identification helps us to retrieve any published or recorded information on the species as background to our efforts. Economic impacts of insects are a key driver, and often the major facilitator, of being able to study and understand them, but much of the knowledge that accrues has much wider application. Insect conservation studies – commonly on scarce species with very restricted distribution so in marked contrast with abundant and widespread pests – sometimes draw heavily on the latter for methods and interpretations of changes. In either context, ecological knowledge is fundamental in attempting to impose changes on their abundance. A relatively few native insects are major economic pests, and their study has illuminated understanding of insect ecology and population dynamics.
Pest Insects ‘Pests’ is a term applied very widely to insects, but in a considerable variety of contexts. An individual cockroach or fly may seriously affront sensitivity if found within a kitchen or food storage area, but may be no more than a casual entry attracted to light or finding an open door or window. Many such cases are of native species entering a highly unnatural habitat and unlikely to cause any harm beyond their simple transient presence. The few species of cockroaches habitually found in domestic environments are cosmopolitan species not originally native to Australia. An individual beetle or caterpillar found in imported goods and intercepted by quarantine authority may be more serious as representing a possibly alien or damaging addition to Australia’s fauna. But both these differ markedly from the devastating impacts of many insects feeding on crops or being vectors of diseases affecting people or domestic stock, and through which massive economic and welfare losses may be caused. Crop yield and quality may be lost, affecting both returns to the grower and the markets available; a year 2000 estimate of annual costs of crop damage from the Queensland fruit fly (Bactrocera tryoni, p. 46) was $Au 28.5 million ($Au 25.7–49.9), with this expected to increase substantially. Lesser damage by other insects, such as scarring of fruit surfaces, even minor blemishes from insect feeding or presence of scale insects, may reduce the quality of produce from ‘prime’ (or individual sale quality) to ‘juicing’ quality, simply on aesthetic grounds as causing rejection by purchasers, be they individuals or major supermarket chains. Chemical residues from pest control measures might not be tolerated in export markets. Economic impacts may not be defined in such straightforward ways – in cases of human or stock health from insect-borne diseases, for instance, but illustrate that
180
14 Insects and People in Australia
insect pests can be categorised on any form of their interference with human welfare and economic wellbeing. This may extend even to settlement patterns – aquatic biting flies, including mosquitoes, predominant in parts of coastal Queensland or along major watercourses, have influenced urban expansion in those areas; patterns of stock grazing may be modified in response to disease vectors or nuisance flies; and some crops by climatic suitability to insect pests such as temperature effects on aphid vectors of plant viruses. Insect pests, however defined, can thus demand substantial and sustained attention to manage them – influenced on the one hand by media hyperbole (‘The only good cockroach is a dead cockroach’, irrespective of whether it is a native or naturalised species) but on the other hand on the security of Australia’s primary industries, export markets and natural environment, collectively affecting the livelihoods of many people and local communities. Rather few insect orders contain no pest species, if we include the widespread nuisance values that are very difficult to appraise, but the major orders that include acknowledged pests are noted in Table 14.1, with the most common pest contexts and impacts summarised. The term ‘domestic pest’ generally reflects household nuisance, and the other terms are largely self-explanatory. The orders not included in this table include several of the aquatic orders, many of the smaller exopterygote orders and some endopterygotes with very specialised ways of life, such as Mecoptera and Strepsiptera, or which are recognized more widely as beneficial. A parallel table (Table 14.2) of beneficial insects reflects the ecological variety in some of the larger orders, and the variety of contexts in which human interests interact with insects. Human health issues are exemplified by the impacts of several species of mosquitoes that breed in coastal saltmarshes, mainly in the free water pools fed by tides or fresh water inflows. Their larvae are aquatic, and several species of Aedes are the most common mosquitoes involved. Females lay eggs singly on damp ground or plants, and hatching is stimulated by flooding after a period of drying out, in contrast to the floating eggs of Anopheles (including the malaria-carrying mosquito) and Culicinae (in which the eggs are laid in ‘rafts’, rather than singly). Very high densities of Aedes can develop rapidly, as inundation by tides may be sufficient interval to promote hatch. The generation time for A. vigilax (the predominant species in the Australian tropics) is only about 5 days in warm conditions and up to about 3 weeks under cooler winter climates. It and related species (mainly A. camptorhynchus in cooler regions) transmit some highly debilitating arbovirus diseases, such as Ross River virus and Barmah Forest virus. Ross River virus is important in all states and territories, but the highest incidence of cases (commonly 2,000 a year) are in Queensland, of a total annual Australian case number in the range of 3,000–5,000 with lowest state incidence in Tasmania. As with many other mosquito-borne diseases, the virus is transmitted through the blood meal needed by females to enable eggs to mature, and females are infested by biting some intermediate host mammal. Once the virus has replicated in the insect for a week or so it is then transmitted with the next feed. Costs of human health care and loss of productivity from Ross River virus have led to mandatory control measures of mosquitoes near human settlements in Queensland, with those measures focusing on the aquatic larval stages. Control programmes have three main components: habitat modification, use
Pest Insects
181
Table 14.1 Insect orders with pest species in Australia, and major pest features of each Order Comment Thysanura Few silverfish are minor domestic pests: little economic effect. Odonata Sporadic reports of larvae affecting aquaculture crops by feeding on hatchling/young fish stocks in pondage; occasional economic impact. Blattodea Few domestic species, but general image leads to others categorised falsely as domestic or household pests; contaminants of foodstuffs. Isoptera Range of termite pests of timber in service; some of major economic importance; variety of associated other effects such as disrupting power services through destroying cable sheathing. Dermaptera Few species can be minor pests through entering houses (annoyance) and feeding on plant seedlings (gardens, horticulture, crops). Orthoptera Include major economic pests such as locusts, with massive economic impacts on agriculture and human welfare. Phasmatodea Mostly innocuous, but occasional severe outbreaks on eucalypt forests, then serious forestry pests. Hemiptera Many of the most serious plant pests, particularly amongst Homoptera: aphids, scale insects, leafhoppers and others can occur in large numbers and cause substantial crop losses through twin effects of (1) direct feeding causing weakness, scarring, loss of yield and (2) as vectors of plant viruses in crops and ornamentals. Thysanoptera Similar damage to Homoptera, and for similar reasons. Few are sporadic aesthetic pests as appearing in vast numbers and causing annoyance. Phthiraptera Some lice are major disease vectors, affecting stock and human health; subsidiary effects as ‘non-hygenic’ (e.g. ‘nits’ in hair of schoolchildren). Psocoptera Few species are minor domestic pests in stored products and foods. Coleoptera Numerous, diverse impacts involving herbivores (plant pests by eating foliage, wood, seeds, vectors of diseases such as Dutch elm disease), damage to stored products, or predators. Some of massive economic importance in forestry and crop production. Diptera Many effects, exemplified by ‘biting flies’ such as mosquitoes, blackflies and marchflies causing direct harm to stock and human welfare, augmented strongly by some being vectors of widespread and severe disease: a number of viruses with severe effect are transmitted by flies in Australia. Siphonaptera Similar impacts to those of Phthiraptera, resulting in serious health issues. Lepidoptera Many plant-feeding pests, with most crops subject to depredations from caterpillars of this group. Some are stored products pests (food, cloth); few species cause annoyance as adults through attraction to lights and entering buildings. Hymenoptera Some plant-feeding sawflies and woodwasps are economic pests, some (Sirex) with major economic impacts in forestry. Invasive ants, bees, wasps with major economic and ecological impacts. Social, stinging taxa may occur in large numbers, and may lead to substantial health and welfare effects, such as influencing recreational and settlement activities.
182
14 Insects and People in Australia
Table 14.2 Insect orders with beneficial species in Australia. Note that these ‘values’ are those given from human perception; they do not include the massive ecological benefits from activities of many insects, that largely pass unheeded. For example, Isoptera, viewed widely as pests, are key agents of decomposition in many Australian ecosystems Odonata Blattodea Orthoptera Phasmatodea Hemiptera
Neuroptera
Coleoptera
Lepidoptera
Hymenoptera
A group that receives cultural and aesthetic appreciation; larvae popular as bait for anglers. Countering the generally unpopular image of cockroaches, some large species (Macropanesthia) are popular as low maintenance pets. Mainly elsewhere, some are important human foods. Some stick insects (such as Extatosoma) are popular pets. Few Homoptera of positive economic value (cochineal insects). Predatory Heteroptera occasionally valuable biological control agents in pest management. Elsewhere, some water bugs (Belostomatidae) are popular food items. Some species (brown lacewings, Hemerobiidae; green lacewings, Chrysopidae) manipulable predators valued in biological control programmes. Many are aesthetically and culturally significant, and popular to collectors. Ecological impacts of economic value and human interest, including dung beetle (Scarabaeidae) and carrion beetle (Silphidae) activity; and ladybird (Coccinellidae) predation on aphids and other crop pests. Aesthetically appreciated and popular ‘collectables’; more tangible benefits from silk production and as human foods (in Australia mainly Bogong moths and ‘witchetty grubs’). Apiary: honey, wax, and pollination of crops. Many parasitoid wasps are important as biological control agents.
of chemicals, and use of biological control agents. The last is still in its infancy for saltmarsh species, and the structure of these habitats does not lend itself well to holding permanent populations of predators such as fish or arthropods. Chemical methods include larvicides that can have very high local kill rates, but costs of their application in very patchy and restricted habitats can become high. Nevertheless, both a growth regulator and a formulation of the bacterial pathogen Bacillus thuringiensis can have precise effects so that non-target concerns are minimal. Habitat modification is a long-term endeavour, commonly expensive to institute and may change the fundamental characteristics of the saltmarsh. In Australia, the practice of ‘runnelling’ involves connecting isolated pools to the tidal source to increase frequency of flushing, and facilitating access by predators. Countering the impacts of pest insects is major component of crop and commodity protection, human and stock health issues, ensuring market access for primary produce, and facilitating peace-of-mind. Methods for doing this have evolved considerably since the days of simply spreading toxic pesticides with little regard to any side-effects, such as killing non-target organisms and of spread into the wider environment. Refinements have occurred largely as a consequence of concerns raised by environmentalists over such issues, including the accumulation of toxins in food webs. The predominant chemical ‘insecticides’, after all, were designed to
Pest Insects
183
kill insects, and little thought was given in those early days to the precise array that would die. Since Rachel Carson’s ‘Silent Spring’, as one of the most influential environmental books of the last century, brought these problems forcefully to wide notice in the early 1960s, the concept of ‘integrated pest management’ has deve loped progressively. Pesticides still have place in many pest management programmes but, where needed, are now orchestrated carefully to be in concert with other tactics, and applied in limited quantities and only at particular times, and precisely targeted. Australian initiatives have been important in this transition, and in developing strategies for effective biological control and ‘cultural control’ as major facets of many of the most effective programmes. These are normally conducted in combination with wider environmental issues, including considerations for conservation of native biodiversity. A central theme has been to replace the notion of ‘eradication’ of the pest insect in many contexts by that of ‘suppression’, most commonly trying to reduce numbers of the pest to below those that cause damage or concern. In economic terms, this level can be defined as a threshold (of insect numbers or density) above which economic loss occurs at an ‘economic injury level’ through loss or taint of the produce. It is thus commonly vital to monitor insect numbers to determine when the population reaches this level, and such surveillance is routine in many pest management programmes. Superimposed on this, patterns of seasonal development of the pest insect may be very predictable, and direct us to periods of susceptibility or stages amenable to reduction or avoidance. Practical pest management is thus influenced and constrained by the principle of cost/benefit analysis, to determine a maximum cost that can rationally be spent on management – but is also commonly accompanied by considerable uncertainty over what the sale price for the protected produce may be – agricultural futures markets, even short term, are notoriously labile. Modern ‘integrated pest management’ seeks to understand and exploit any available vulnerability in the insect’s structure, physiology, behaviour and ecology – in short, any way in which it can be suppressed reliably and with the least possible side-effects in the wider environment. Increasingly, the ‘safety’ aspects of pest management for human health, environmental conservation (including avoidance of non-target effects), and access to export and domestic markets are central considerations. It orchestrates all available tactics into a management strategy that may have several purposes, either singly or in combination. Quarantine measures are not confined to detecting alien insects – the prohibition of transfer of fruit on aircraft or along roads within parts of eastern Australia is a significant contribution to restricting spread of economically damaging fruit flies. The ‘Fruit Fly Exclusion Zone’ incorporates some highly productive areas in the south east inland, and is managed by a Tri-State Fruit Fly Committee suported by three state governments (New South Wales, South Australia, Victoria) and the Commonwealth, with the mandate to keep the region fruit fly free. As with many other highly dispersive pests (and other insects), its distribution is likely to change considerably with changing climate – so that the range of B. tryoni shown in Fig. 4.5 (p. 46) could expand southward, providing problems such as increased vulnerability
184
14 Insects and People in Australia
of the exclusion zone, increased severity of attack in southerly parts of its range, increased probability of attack to crops of small scale growers (such as home gardens) and major cities in the south. Such trends have been modeled against a variety of climate change scenarios, and the outcomes caution against any complacency that we understand the focal insects sufficiently to be able to manipulate them predictably. Techniques such as release of sterile males to hamper reproduction in wild populations by providing competition with ‘normal males’ for mates has been attempted for fruit flies, for example, and specific chemical lures are an important approach to both monitoring their incidence and trapping the flies, based on their response to specific chemical stimuli. Almost any biological feature of an insect may be exploited for its management, for example in killing it, preventing reproduction, influencing movement or behaviours such as mate-finding and development, and augmenting attack by predators, diseases and parasitoids. Chemicals called ‘insect growth regulators’, for instance, are part of the armory of modern pesticides, nontoxic and very specific in action. These are hormonal analogues that when applied to early stages (such as caterpillars) prevent them from moulting or reaching adulthood. Many of the arenas in which pest management must be undertaken are, of course, highly artificial – constructs such as broad scale monoculture crops replacing formerly much more diverse natural systems and providing very conspicuous and nutritious targets for pest attack. Agricultural ecosystems (often ‘agroecosystems’) are thus often portrayed as harbouring low biodiversity and inhospitable to native insects, and much of modern pest management, undertaken on a wider landscape perspective, seeks to redress this. Agroecosystems are important, often predominant, components of many Australian landscapes and provide the livelihoods of many people, but have aroused considerable conservation concerns by their extent and uniformity, and by fragmenting natural vegetation and other biotopes to leave only tiny isolated fragments of previously extensive habitats supporting high numbers of native species. The tiny pockets of native vegetation remaining throughout the Western Australian wheatbelt have been preserved somewhat fortuitously – many of them were simply avoided as non-amenable to cultivation through being rocky or steep – but are now a highly significant component of the reserves system there. Much attention is now being given, by pest managers and conservationists alike, to ways in which the highly contrasted conditions of agricultural and natural ecosystems can be reduced, and how environmentally sensitive cultural controls can be promoted widely as a component of pest management. It is not always necessary for large numbers to be present for an insect to be considered a pest. A single codling moth caterpillar in an apple, for example, is ample to condemn that fruit from prime market quality to juice or waste. On the other hand, hundreds to thousands of cabbage aphids on an individual brassica plant may cause little harm and at harvest can be removed easily by routine trimming of loose outer leaves, and washing. Consumer demand is a strong influence on the intensity of insect pest management. In conjunction with increasing demands for high quality and standardisation through plant variety selection and uniform husbandry, competition between supermarket chains continues to demand the highest standards of uniformity and presentation, with grower returns from lesser produce with even minor blemishes sometimes punitively lower.
Further Reading
185
Insects and people meet on many levels, only some of which have been mentioned here. Scientists have long valued insects as ‘tools’, experimental animals that have illuminated wider aspects of biological understanding. Vinegar flies (Drosophila) remain amongst the most significant agents in demonstrating principles of heredity, through the rapidity and ease with which they can be maintained and bred in the laboratory – and that interest led directly to revealing a diverse array of native species of this fly family in Australia. The principles of many ecological interactions, and of population dynamics, were established largely through insect studies, to which those in Australia have been amongst the most penetrating and influential. Insects can be culturally valued, tangible commodities for human consumption or use, beneficial tools for enhancing ecosystem services (such as pollination, and as biological control agents) and for evaluating environmental quality and change, as well as pests inimicable to human priorities on many scales and in many contexts. The alien component of Australia’s insect fauna is of particular concern in the last contexts, by no means least for its real and suggested effects on native insect diversity, together with the changes to habitats engendered by agroecosystems and other human changes. These themes, both vast, are considered in the next two chapters.
Further Reading Carson R (1962) Silent spring. Houghton Mifflin, Boston (an historical classic of lasting value, as a major impetus for interests in the effects of pesticides and a driver for seeking alternatives or insect management) Corey S, Dall D, Milne W (eds.) (1993) Pest control and sustainable agriculture. CSIRO, East Melbourne (includes much classical work on insect pest management) Hunter DM (2004) Advances in the control of locusts (Orthoptera: Acrididae) in eastern Australia: from crop protection to preventive loss. Aust J Entomol 43:293–303 (development of locust management and forecasting exercises, through the work of the Australian Plague Locust Commission) New TR (2002) Insects and pest management in Australian agriculture. Oxford University Press, Melbourne (broad introductory text) Sutherst RW, Collyer BS, Yonow T (2000) The vulnerability of Australian horticulture to the Queensland fruit fly, Bactrocera (Dacus) tryoni, under climate change. Aust J Agric Res 51:467–480 (illustration of how climate modeling may be incorporated into forecasts of pest range changes and needs for control) Waterhouse DF, Sands DPA (2001) Classical biological control of arthropods in Australia. ACIAR Monograph no 77, Canberra. (Includes detailed and well-referenced dossiers on wide variety of imported control agents)
Chapter 15
Australia’s Alien Insects
Introduction: Variety and Impacts The recent ‘alien’ component of Australia’s insect fauna is of particular interest to many people, and in many contexts the species involved are often far better-known than much of our remarkable native fauna, because their interest and impacts extend far beyond the focus of biologists alone. In part, this is because this element (often referred to also as ‘exotic species’) includes a high proportion of the most serious agricultural and domestic pest species, themselves affecting crops, livestock and commodities that are also mainly introduced as the foundations of primary industry, but is also because a number are beneficial. Most of the alien species are confined largely to human-shaped environments, and depend on the facilitating resources provided since European settlement. As one example, Australia has around 170 species of aphids (greenfly and their allies), an ancient group of small sap-sucking bugs that apparently originated on conifers but subsequently proliferated as angiosperms evolved from the Cretaceous onward. A high proportion of these are alien species restricted, with few exceptions, to introduced host plants without which they cannot survive. Some are major crop pests for their dual roles of causing direct yield losses through feeding and as vectors of important plant virus diseases, such as those of cereal crops. They can thereby have massive economic influence and such harmful alien species are by far the best known species amongst this faunal component. Although tiny, aphids can occur in vast numbers and increase populations very rapidly. Aphids are amongst the many small insects that have probably been arriving in Australia for many centuries by being transported on wind currents as part of the ‘aerial plankton’, but many such arrivals could not become established until suitable food plants were present. Detection of ‘an alien insect’ is not always easy – some have been here for so long that they are essentially naturalised, as widespread, well-established, accepted and predictable members of the fauna. Many others are of poorly-documented groups with numerous undiagnosed native species, or difficult to recognise, and individual new arrivals may not be detectable against this background complexity and ignorance. As in many similar contexts, a novelty can be detected only by knowledge of the norm.
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_15, © Springer Science+Business Media B.V. 2011
187
188
15 Australia’s Alien Insects
Some larger insects, of course, are much stronger flyers and some may arrive in Australia by much more active dispersal. Distance dictates that this is unlikely to occur directly (other than by rare accident) for most species, but ‘island-hopping’ (for example across Torres Strait from New Guinea to northern Australia) allows use of various ‘stepping stones’ to break the journey. Invasion of pests such as screwworm fly (Chrysomyia bezziana, a major wound-causing pest of livestock) by this route has long been anticipated, and feared, and has led in this case to surveillance for the fly, for many years using ‘sentinel’ cattle deployed on Cape York and inspected regularly for signs of wounds caused by screwworm. The deep wounds to livestock cause serious debilitation, and even death. The fly is endemic in New Guinea as part of an extensive tropical and subtropical range, and the 150 km separation of Torres Strait contains numerous islands. So far, the only record in Australia has been of several flies trapped on a ship in Darwin in 1988, and which had just returned from offloading cattle in Brunei. Should the fly arrive in Australia, large areas suitable for livestock rearing as far south as central New South Wales seem climatically suitable for it to exploit. Massive economic damage to the industry could follow. Some butterflies found on the Torres Strait islands may not breed there, and are regarded as sporadic vagrants from New Guinea; they have potential to reach Australia and establish if resources become available. The familiar large orange and black wanderer (or, elsewhere, monarch) butterfly (Danaus plexippus) is well-known in eastern Australia, but only arrived in Australia in the nineteenth century from its native North America. There, it undergoes long migratory flights – northward in spring and returning southward in autumn, and its spectacular mass overwintering aggregations in Mexico and California have been designated as an ‘endangered phenomenon’ as the hosting areas become threatened by forest clearing or other development. It has occasionally been reported from Europe as a rare visitor, presumably having crossed the Atlantic, but its passage across the Pacific by islandhopping has been reasonably well documented, and follows introductions of its milkweed (Asclepias spp.) larval food plants as ornamentals; as with the aphids mentioned above, it could not have established until suitable food was available, even if it arrived in Australia. Thus, D. plexippus was reported from Hawaii about 1840, Tonga in 1863, New Zealand (with early records not wholly clear) probably shortly after this, and was found first in Australia in Brisbane (1870) followed closely by Melbourne (1872). The wanderer is naturalised in Australia and is appreciated as an aesthetically pleasing species without obvious economic or other adverse impacts. It is, for example, one of the major species reared for release at weddings and other celebrations, in a practice that can engender conservation concerns over principles of artificially distributing alien species or distorting normal distribution patterns and mixing local gene pools of native species.
Importation and Establishment Many more alien insects have arrived by human agency, commonly as undetected stowaways in ships or aircraft, in trade goods or packing materials. Recurrent urges for more effective quarantine measures and inspections (nowadays, major facets of
Importation and Establishment
189
‘biosecurity’) are difficult to effect – small insects can easily elude detection, and many are discovered first only after they are in the open Australian environment. Some, such as some household insects and ectoparasites of domestic pets and livestock probably arrived with the First Fleet or shortly thereafter, at times when no quarantine was considered. Such passive transport continues, with the benefit that rapid air transport now enables insects and others to reach any part of the world rapidly and buffered against filtering changes such as passing through the tropics. Not all passive transport is human-aided. Timber-infesting beetles may travel by ‘rafting’, being washed up on beaches in driftwood, in which their long-lived larvae are protected. More serious, and potentially controllable by replacement of wood by metals and plastics, is their presence in wooden packing material. They can occur also in wooden artefacts such as tourist souvenirs – hence the attention given to these at customs entry points. In such items, beetles may be imported as small larvae, and emerge only several years later, often undetected except when emergence holes are noticed at some future time. Quarantine measures at both export and import stations do much to control this, through selective fumigation and impoundments, but some risks of undetected importations are inevitable. However, accumulated records of insects intercepted by quarantine indicate the extent of arrivals. Examination of a sample of 3,000 cargo containers in Queensland in 1996 yielded representatives of five families of timber-frequenting beetles and 11 other families of Coleoptera, including several potentially serious agricultural pests, for example. About 770,000 such containers entered Australia that year! Across the Tasman Sea, analysis of one group of bark beetles (Scolytinae) in New Zealand intercepted in the second half of the twentieth century disclosed beetles originating in 59 countries – the undetected waifs of trade are a major contributor to homogenising the global fauna of such groups. Many, in addition, do not travel alone, but are vectors for important plant diseases (Dutch Elm Disease, so devastating to elms in the northern hemisphere, is carried by species of Scolytus, above), as well as mites and other small animals carried on their bodies. Risks to Australia’s natural environments may be substantial, the scale of arrivals is formidable and, although the potential impacts of many of the species involved may be small, this is usually not wholly predictable: wise precaution (‘If in doubt, keep it out!’) is assuredly the safest principle to adopt. Preemptive plans can be prepared for some particularly feared arrivals (such as screwworm, above), but it is impossible to anticipate, and counter in advance, every possible alien. An example of a serious recent concern is esta blishment of the Asian honey bee (Apis cerana) in northern Queensland, as a possible threat to existing apiary. The impressively large numbers of arrivals implied from the above scenarios have additional relevance in that each could lead to establishment, as below. But for an insect species to become established in a new area, individual numbers need not be large. A single gravid female beetle or fly, or a parthenogenetic female aphid arriving alone, may each be the foundress of a newly resident population in Australia. Should favourable conditions persist, that population may grow and establish firmly, aided by the rapid development of individuals and high fecundity that contribute so centrally to the success of many colonising insects. From then on, the population
190
15 Australia’s Alien Insects
may remain largely in and around its areas of introduction, or spread and perhaps invade natural environments. Two modes of range increase are common. First, a species may expand around its initial site, so that the occupied area increases gradually by ‘incremental creep’, perhaps with each subsequent generation moving slightly further. It may or may not enter new environments or stay within its initial favourable milieu – thus, not all species make the transition from urban to rural or artificial to native ecosystems. Second, and not necessarily excluding the above, a species may ‘suddenly’ appear in a locality far from its known distribution or arrival sites. This results commonly from transport by people. European Vespula wasps (below), for example, have been found in a number of country towns in Victoria, to where they have probably been transported as quiescent hibernating queens hiding in firewood or household goods, and subsequently proven capable of founding colonies as warmer weather follows. The European elm leaf betle (Pyrrhalta luteola) can be transported in similar style – and advice issued to the public in Victoria includes the sentence ‘Check cars, caravans and trailers carefully to ensure you are not transporting beetles’. ‘Pest status’ of new arrivals extends beyond direct economic damage to fears that they may become threats to native animals and plants, should insects move into natural environments by straying beyond the highly modified arenas they normally frequent. Globally, alien insects attract massive attention as threats to native biota – and for Australia this is a two-way interaction. For insects arriving in Australia, several basic practical themes form a sequence for attention. These are (1) how do they arrive and how can they be detected – for example by good quarantine process and field monitoring for key species; (2) which species are likely to arrive and become established; (3) how common will they become once established; (4) how far and how rapidly will/can they spread, and what are the possible consequences of this – for example, will they become economically important pests or have ecological impacts. The converse scenario is that many native Australian insects have ventured to other parts of the world. Whilst some have attracted little attention at home, they have become serious concerns (as pests) elsewhere and overseas studies on their biology in attempts to eradicate or suppress them often by far exceed knowledge available within their natural environments.
Consequences In either direction, the basic processes, as noted above, involve arrival, initial colonisation, establishment, and invasion, with a relatively small proportion of species progressing to each subsequent stage along this sequence – the number of arrivals is far greater than those that colonise, and so on. The insects of greatest concern, obviously, are those that establish or invade. Establishment alone may restrict insects to anthropogenic, largely agricultural or urban, environments, so that their impacts – even when severe – are within already largely non-natural areas. Those insects that literally go further and invade more natural ecosystems pose different problems in
Consequences
191
relation to their impacts on native animals and plants. Some are regarded as serious threats by conservationists. Spread of alien insects, by whatever means, is thereby a major concern and considerable efforts have been made to predict and anticipate how this may occur. Climate-matching exercises have been used to help predict spread of pests, and also to anticipate where particular introduced natural enemies imported as biological control agents may be most effective. Thus, the areas of greatest climatic suitability (mainly reflecting aspects of temperature and humidity) for various African and European dung beetles introduced to aid breakdown of cattle dung (p. 74) differ greatly across different species, so that matching a given species with its optimal climate may determine how successfully it establishes and ‘performs’. Planned releases of classical biological control agents in places compatible with their climatic needs is becoming routine practice in attempts to increase their success rates, and some such releases can be ‘tailored’ for given environments. Expensive releases into areas where the agent cannot thrive or be effective are clearly wasteful and, by the same modelling approaches, areas unsuitable for particular pest insects may be detected, and this information help to predict their impacts. As noted earlier, one of the implications of future climate change is that distributions of many alien insects (and other species) will be altered considerably and lead us to re-think their impacts and how the systems they infest will also change their dynamics. Alien social Hymenoptera are of particular concerns in regard to uncontrolled spread and invasions of natural habitats, with the adaptations to social life often closely paralleling those that more generally may make insects successful invaders. Feral honeybees (the invasive outcome of a species introduced delibe rately for benefit) are limited perhaps only by availability of water, and occur throughout much of Australia. Conservation concerns have arisen for several reasons. Honeybees nest in tree hollows that are used also by native birds and marsupials and, in areas where suitable holes are in short supply, may out-compete vertebrates for these and deprive them of breeding opportunity. Very difficult to test, and with studies giving contradictory implications, are accusations that honeybees displace the numerous native bees and other pollinators (such as jewel beetles in Western Australia) and deprive them of adequate nectar sources. The debate is politically charged, with apiarists pressing claims to be allowed to place hives in national parks and other areas in which nectar-rich native trees are still abundant, following extensive land clearing elsewhere. Such areas may have a primary role as sanctuaries for native biodiversity, so deliberate introduction of any species likely to constitute a threat is widely opposed. Many apiarists claim that no such threat from honeybees has been proved. Closer to urban areas, Apis mellifera is one of the insects that contribute to entomophobia and human safety through allergy to stings. European and English wasps, Vespula germanica and V. vulgaris, as accidental introductions, cause similar concerns. The former, in particular, has spread extensively in southern Australia and also has potential to expand much further north. These wasps lack the beneficial features of honeybees as pollinators and honeyproviders and, because they do not die off in winter as is usual in their colder northern temperate range, can produce much larger colonies than in Europe. Early records from Tasmania revealed nests with more than a million cells, each contributing an
192
15 Australia’s Alien Insects
Table 15.1 Major tramp ant species in Australia and their status (After Commonwealth of Australia 2006) Species Common name Status Dolochiderinae Linepithima humile Argentine ant Widely established Tapinoma melanocephalum Ghost ant Widely established Technomyrmex albipes White-footed ant Widely established Formicinae Anoplolepis gracilipes Paratrechina longicornis
Yellow crazy ant Crazy ant
Incursions Widely established
Myrmicinae Monomorium destructor M. pharaonis Pheidole megacephala Solenopsis geminata S. invicta Wasmannia auropunctata
Singapore ant Pharaoh ant African big-headed ant Tropical fire ant Red imported fire ant Little fire ant
Widely established Widely established Widely established Incursions Incursions Incursions
individual wasp as it develops. Vespula wasps can interfere with apiary by robbing bee hives, but much concern over their presence results from stings by wasps attracted to fruit or sugary drinks – sometimes even entering open cans of soft drink unnoticed – so they can be abundant in recreation areas, and interfere with fruitpicking operations. They commonly nest under or near houses. Poison bait applications in towns and country areas frequented by people can become costly, but many pest-control operatives are involved in destroying nests. Widespread predatory habits of the wasps can be a threat to native invertebrates. The most pervasive of invasive social insects are the ants, with several widespread species of ‘tramp ants’ having global notoriety through their effects. Eleven of these have become established, having arrived by various means, in Australia. Five of them are listed among the 100 worst invasive species of all kinds by the World Conservation Union, so acknowledging their enormous potential to threaten native taxa and cause other harm. The impacts documented, and more widely inferred in Australia include displacements of native species, and competition and predation affecting important ecological processes such as seed dispersal and pollination. All these species can form enormous populations, spread rapidly and are highly adapted to colonising natural environments. The major species involved so far are listed in Table 15.1. Countering their impacts is complex but it is sobering to reflect that these are amongst more than 200 alien ant species reported from various parts of the world, with at least 105 species intercepted at Australia’s borders from 1986 to 2002. Ants are transported easily, and inconspicuously, in trade and are easily able to evade detection. Impacts of some invasive ants rival scenarios more usually found in science fiction. The yellow crazy ant (Anoplolepis gracilipes) and the red imported fire ant (Solenopsis invicta) illustrate this variety, in different ways and in different parts of Australia.
Consequences
193
Christmas Island, an outlying Australian Territory to the south of Java, Indonesia, is best-known to many Australians as a refugee-holding centre and for its predominant ground-dwelling invertebrate, the red land crab (Geocarcoidea natalis) which can carpet large areas of the island and occurs at densities of around 2 per square metre (extrapolating to around 1,500 kg per hectare!). A. gracilipes reached Christmas Island about 50 years before its numbers expanded dramatically from 1989 on to produce vast ‘supercolonies’ of ants which contain many queens and which covered about a quarter of the island’s rain forest by 2001. The ants kill land crabs, with a published estimate of 10–15 million crabs (a quarter to a third of the whole population) killed over several years, by large numbers of ants spraying formic acid over their eyes and mouthparts, leading to death within about 2 days, and affecting crabs both during their normal life and during migrations as they traverse supercolony areas. The ecological impacts were much wider. Loss of crabs led to significant changes in litter removal and seedling recruitment as processes previously influenced markedly by crabs – litter cover was doubled, seedling density about 30-fold higher and richness also higher in ant-invaded areas, so that this specific interaction had far-reaching implications for forest structure and dynamics on Christmas Island. Strenuous campaigns to suppress the ant are in progress. The impacts of Solenopsis invicta in Queensland since it was discovered in Brisbane in 2001 are much more varied. As the common name implies, ‘fire ants’ cause severe pain from their stings – which have been likened to holding a burning match against the skin, and produce raised pustules. Should an ant nest in the ground be disturbed, mass attack is rapid, and facilitated through alarm pheromones, so that hundreds to thousands of workers may attack together. Agonising pain and anaphylactic shock are common outcomes, and deaths have been reported in North America. Human health is only one of the concerns over this species, but links strongly with social and amenity activities, so that sports grounds, parks and other recreational areas infested by ants are avoided, to the detriment of tourism. Widespread environmental effects occur through predatory impacts on almost any ground-active animals from nestling birds and small mammals to reptiles and invertebrates, direct attack on seedlings, reduced crop yields by seed predation and direct feeding, together with less obvious outcomes such as by their mounds clogging agricultural machinery, damage to electrical devices and other infrastructure, and undermining roads and pavements. The discovery of S. invicta triggered a massive emergency response as what has been described as probably ‘the most ambitious and important effort ever undertaken’ against an insect pest in Australia, with an initial budget of Au$120 million over 5 years. At its peak, the eradication project employed close to 650 people, and considerable success has been achieved. More than 65,000 nests found in 2001 (indicating that the ant had been present long before it was actually discovered!) contrasted with fewer than 200 by 2006–2007, and many of the latter were small, each with only a single queen and so recently founded. The major ingredient in reducing ant numbers is repeated distributions of a granular corn-based bait containing insect growth regulators. These hormones do not poison the ants, so do not have toxic non-target effects, but sterilise queens and prevent larval development, so that colonies die out through not recruiting further individuals. However, despite
194
15 Australia’s Alien Insects
this enormous effort, and continued vigilance, new outlying infestations continue to be discovered. Any lessening of attention could lead to rapid resurgence of fire ants in the region. Work in China has confirmed that drug-detector dogs can be retrained to discover fire ant nests, as a novel but potentially valuable monitoring aid for low density populations. Other important tramp ants include Argentine ant (Linepithema humile) and the big-headed ant (Pheidole megacephala), both species spreading in and around urban environments where they can displace native ants. Yet another alien social insect ‘issue’ emphasises how controversy over a species’ role in Australia can flow from conflicting priorities and interests. The large earth bumblebee (Bombus terrestris) was introduced to New Zealand from its native Europe in the mid nineteenth century as a crop pollinator. It appeared in Tasmania in 1992 and has since then spread throughout the island state, where it is thought widely to be disadvantaging native pollinators and interfering with natural pollination systems – in part through the bee’s habit of chewing holes in flowers to gain easy access to nectar. However, bumblebees are also ‘buzz pollinators’, using sonifaction from vibrating wing muscles to increase release of pollen from many flowers, and this habit is beneficial for some glasshouse crops, particularly for tomatoes. Strenuous efforts have thus been made to permit importation of bumblebees for pollination of glasshouse tomatoes in New South Wales, whilst in contrast introduction of bumblebees is a listed threatening process in Victoria – where any colonies found would be destroyed. The bee can thrive out-of-doors in Australia, and has a broad climatic tolerance, and different points-of-view of its impacts will assuredly persist.
Australian Insects Abroad Impacts of alien insects vary considerably and, as noted above, may be difficult to evaluate objectively. Our understanding has been derived largely from those that are regarded as either harmful or beneficial, with studies prompted by designing mani pulations to either suppress or enhance those effects and collectively giving much information on how species ‘work’ in new environments. Many more ‘neutral impact’ species are not as well known. But just as ecologists in Australia have sought additional information on insects of interest in their native ranges, those seeking to manipulate alien Australian insects in other parts of the world must draw on whatever background may be available in Australia. In either direction, these insects ‘at home’ are likely to be in greater harmony within natural communities within which they have evolved, and their impacts tempered by interaction with other species, such as natural enemies. Major changes may arise, often unexpectedly, once the insect is freed from this balance by entering a novel environment. Scientists from overseas, in seeking possible ways to suppress Australian insects elsewhere, have made major contributions to understanding ecology of some insect groups – particularly of those containing potential biological control agents.
Australian Insects Abroad
195
Several functional categories of Australian insects abroad are involved, paralleling the reverse interpretations for alien insects in Australia. Some are simply ‘there’, noted as unusual – perhaps transient – species but with little impact and causing no concerns. Several Australian butterflies commonly reach New Zealand on prevailing weather systems, for example; some are established and widespread there. Others, such as the tramp ants noted above and some other serious pests, may be on ‘the most wanted’ lists of serious pests, whose incidence triggers urgent counter-measures. The remaining categories encompass (1) insects introduced as biological control agents for Australian plants that have become alien weeds; (2) insects that have become pests on Australian plants grown as crops or amenities; and (3) other insect pests from Australia. Insects in the last two categories may lead to further introductions of natural enemies (predators, parasitoids) as part of a management programme. Some examples help to illustrate this variety. From the mid nineteenth century onward, several species of Australian bipinnate wattles were planted extensively in southern Africa as the foundation of a tan bark industry. In particular, black wattle (Acacia mearnsii) was introduced to many parts of Africa and Asia as a plantation crop for this purpose. The decline in the industry’s importance was paralleled by spread of the acacia into natural areas, as an invasive weed with massive seed set, long-lived seeds, and soil-stored seed banks and its suppression became desirable. In another context, the phyllodinous Acacia longifolia was also introduced to South Africa for stabilising sand dunes, and became highly invasive, to rank as the second most important invasive plant in the rich fynbos vegetation of the Cape Floristic Kingdom. Biological control necessitated finding suitable candidate agents from Australia, but it was critical that these, as well as being effective, should be sufficiently specific to not attack native African acacias and thereby themselves become pests. Examples of two important radiations of insects noted in Chap. 8 appeared suitable for the various contexts explored, and after extensive screening for specificity, were released to the wild in South Africa. The tiny gall wasp Trichilogaster acaciaelongifoliae lays eggs into the flower buds or vegetative buds of A. longifolia, which then develop into small spherical galls instead of inflorescences or branches, so that both reproductive potential and vegetative growth are markedly reduced (p. 120). Each female wasp, which is parthenogenetic, can lay up to around 400 eggs, and high levels of infestation lead to substantial reduction of the acacia. Almost 90 described species of Melanterius seed weevils are known in Australia. They develop in ripening pods and seeds of Acacia species (p. 124), and five species have been released in South Africa against various acacias and a closely related plant, Paraserianthes lophantha. In several acacias, damage has been reported as up to 90–100% of seeds destroyed, and the weevils are important contributors to slowing recruitment. Fast-growing eucalypts have been used extensively for amenity plantings (such as windbreaks along motorways) and plantation forests for timber production in many parts of the world. Features such as their very rapid growth (so that harvesting rotations can be as short as about a decade), growth on poor soils, and high planting density render them very attractive crops, as well as for use as ornamentals or amenity plantings (such as roadside windbreaks). In California, eucalypts were planted from
196
15 Australia’s Alien Insects
about 1850 onward and, until about 1983 they remained almost pest-free – other than for a single seed-galling eulophid wasp (Quadristichodella nova) reported in 1957. Since then, however, pests have proliferated, with a further 16 Australian insect species becoming established, and some causing severe concern. Eucalypts in many parts of the world have been infested by a considerable variety of Australian insects, for which biological control agents have been sought within Australia. As one example, the timber-boring longhorn beetle Phoracantha semipunctata is one of an array of similar species in Australia but in recent decades has spread to various countries in which eucalypts are now available, and causes significant tree mortality. In California, the beetle is active during warmer parts of the year, when trees may be especially susceptible through water stress. It has been controlled there by use of a small wasp, Avetianella longoi, which attacks the beetle eggs. Several species of Australian braconid parasitoids were also imported to California, and all these species were collected from Victoria during intensive surveys which greatly improved knowledge of these associations in Australia A similar strategy applied to the eucalyptmining sawfly, Phylacteophaga froggatti, which caused concerns over aesthetic and possibly more functional effects by conspicuously blemishing foliage of roadside and ornamental plantings of Eucalyptus in New Zealand. Again, braconid wasp parasitoids were sought in south east Australia and imported successfully – in this case, however, the sawfly damage was confined largely to younger trees, and older trees became largely unaffected. The main need was thus to prevent undue weakening and unsightliness of the trees during a rather short susceptible phase. As a rather broader example involving Acacia insects, and to illustrate the variety of concerns involving insects associated with our predominant flora, a small Australian psyllid bug caused considerable damage to native Hawai’ian acacias (predominantly Acacia koaia) from the early 1970s (following its initial discovery at Honolulu International Airport in 1966), prompting searches for predators and parasitoids as biological control agents. This bug, Psylla (now Acizzia) uncatoides, was originally described from New Zealand and revealed as a native Australian only by these surveys!. It can occur in enormous numbers feeding on flush growth of acacias, and the Hawaiian endemic A. koaia suffered large-scale dieback and death from its attacks. A. uncatoides is polyphagous, and has become a concern elsewhere in the world as well – in California, for example, it has been reported from almost 70 of more than 100 alien Acacia species surveyed, but was common on few of these. Control in Hawai’i and elsewhere has been aided by introducing the Australian ladybird beetle predator Harmonia conformis, suggested to be an obligate predator of Acacia psyllids in Australia. Dramatic reductions of psyllid numbers followed releases of this beetle. Deployment of an Australian ladybird against a bug pest provides a sense of deja vue in classical biological control, as one of the all-time pioneering successes in the discipline is for the vedalia beetle (Rodolia cardinalis, another Australian ladybird) to control the cottony cushion scale (Icerya purchasi) devastating citrus in California in the 1880s. That exercise has been described as ‘unparalleled in the annals of entomology for its drama, human interest, political ramifications, and continuing significance’, as a foundation case in classical biological control. The scale insect became established around Los Angeles in the mid 1880s and had destroyed
Further Reading
197
thousands of citrus trees by 1886 but – from initial releases of around 500 beetles – complete control was realised within 2 years. Both protagonists can increase in numbers rapidly – in California I. purchasi underwent three generations a year, with each female producing around 325–450 eggs. Each female Rodolia lays ‘only’ about 150–190 eggs, but the beetle can have up to 12 generations a year, so that a superabundant food supply fosters rapid development of enormous populations. Despite dramatic successes such as this, modern fears over the impacts of such polyphagous natural enemies on native organisms would be likely to preclude this scenario being repeated today. Even initial screening, as is routine practice in modern cases, would reveal non-specificity and raise debate over Rodolia feeding on nontarget native insects. The importance of Australia’s increasingly effective quarantine requirements is largely to prevent such damage to our native biota from harmful and unwelcome arrivals. Deliberate introductions of insects will continue, but only with the best obtainable evidence that they are likely to be ‘safe’: the reality of the cane toad (p. 205) lingers as a reminder of what may result unintentionally from introductions made without such diligence. But, notwithstanding this, the practices of classical biological control continue to be refined and the approach will continue within integrated pest management, both in using alien insects in Australia and Australian insects abroad.
Further Reading Commonwealth of Australia (2006) Background document for the threat abatement plan to reduce the impacts of tramp ants on biodiversity in Australia and its territories. Department of Heritage and the Environment, Canberra Holway DA, Lack L, Suarez AV, Tsutsui ND, Case TJ (2002) The causes and consequences of ant invasions. Annu Rev Ecol Syst 33:181–233 Maynard GV, Hamilton JG, Grimshaw JF (2004) Quarantine – phytosanitary, sanitary and incursion management: an Australian entomological perspective. Aust J Entomol 43:318–328 Moloney S, Vanderwoude C (2002) Red imported fire Ants: a threat to eastern Australia’s wildlife. Ecol Manag Restor 3:167–175 New TR (1994) Exotic insects in Australia. Gleneagles, Adelaide (overview of Australia’s alien insect component and its significance) O’Dowd DJ, Green PT, Lake PS (2003) Invasional ‘meltdown’on an oceanic island. Ecol Lett 6:812–817 (discussion of the impacts of crazy ants on Christmas Island) Paine TD, Steinbauer MJ, Lawson SA (2011) Native and exotic pests of Eucalyptus: a worldwide perspective. Annu Rev Entomol 56:181–201 (much background information, and many useful references cited) Shoemaker DD, Ross KG, Keller L, Vargo El, Werren JH (2000) Wohlbachia infections in native and introduced populations of fire ants (Solenopsis spp.). Insect Mol Biol 9:661–673 (raises possibility of using these widespread debilitating bacteria as biological conrol agents for fire ants) Tyndale-Biscoe M (1990) Common dung beetles in pastures of south-eastern Australia. CSIRO, Melbourne (illustrated account of introduced dung beetle species, with climatic model-based distribution maps for each species to suggest areas of establishment and value as biological control agents) Withers TM (2001) Colonisation of eucalypts in New Zealand by Australian insects. Austral Ecol 26:467–476
Chapter 16
Conservation
Introduction: Rationale and Needs for Insect Conservation A recurring theme throughout this book is that our knowledge of Australia’s insects is still very incomplete, and that enumerating and interpreting the fauna properly still has far to go. Much of the work needed is very basic, and includes much more field survey for general collecting and documentation to aid plotting of distributions and clarify evolutionary pathways (particularly amongst the less-studied orders of terrestrial insects). It also needs the assurance that the specimens obtained can be preserved and valued as an historical component of Australia’s unique heritage – just as present-day entomologists in Australia must still rely on specimens long preserved in northern hemisphere museums to answer some critical questions of identity, future workers will assuredly need to assess the specimens on which current inferences are based. Admittedly, exchange of information on those early historical specimens has become far easier in these days of first-class digital photography, remote microscopy (whereby images can be transmitted anywhere in ‘real time’ for examination and discussion, an application that is particularly important in identifying quarantine interceptions and in other contexts in which delay may be expensive) and computer communication and databasing systems, so helping to counter both the time delays and the high travel costs for specialists to visit overseas institutions for lengthy periods of study, and the risks of committing fragile specimens to the vagaries of international post. Despite these advantages, the specimens themselves are still of critical importance as the ultimate reference sources. However, we still need to augment our knowledge of ‘what is there’ and to agree standards (such as the optimal ‘boundaries’ between species within major taxonomic groups) by which information on our insects can be communicated effectively to those non-entomologists who may influence their future. Paradoxically, as our fundamental knowledge increases the practicalities of continuing to augment it are becoming more difficult, for two major and complex reasons. The first is that collecting, describing and classifying insects, even using sophisticated ‘modern techniques’ is not an area of growth in Australia, despite
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_16, © Springer Science+Business Media B.V. 2011
199
200
16 Conservation
continuing calls from politicians to ‘document biodiversity’. Many young people, indeed, are dissuaded from specimen collecting and study of natural history as a hobby – often with the best of intentions - by parents and others who see the activity as a threat to the organisms, but also by lack of leadership from school, and the widely competing pastimes of sports and computer games seen widely as more pertinent to modern lifestyles and more acceptable amongst peer groups. Recruitment to the entomological fraternity is small, formerly more widespread ‘mentoring’ from knowledgable enthusiasts has declined, and the number of jobs – particularly longterm career opportunities – in the field are very limited. Many of the taxonomic specialists who have retired from our major museums over recent decades have not been replaced by people with equivalent expertise, or given opportunity to acquire that expertise. ‘Generic skills’ such as statistical modelling and analyses and ability to prepare and interpret molecular information increasingly replace the unique morphological scholarship that may generate a ‘world authority’ on an important or diverse family of moths or flies. Curatorial and research capability have suffered accordingly, and with global effects. Insects (together with numerous other invertebrate groups such as molluscs, earthworms, spiders and other arthropods) are not sufficiently regarded to appeal to employers and most research funding agencies for what may be seen as ‘pedestrian’ documentation of their diversity and distributions. Basic entomological training, the foundation for such interest and appreciation, has declined in Australia’s undergraduate courses: the former Departments of Entomology of several of our leading universities have all lost that status through amalgamations or changes of focus. Should the discipline be taught at all, the basic systematic and evolutionary curriculum is almost always subservient to applied ecology and pest management components as, understandable, needs to help protect Australia’s primary industries. Most insects do not have the notoriety or perceived ‘applied values’ to enter this portfolio. The second reason for concern reflects the changes made, and continuing to be made, to the natural environments in which our insects live and have evolved. Much natural vegetation has been cleared, water bodies drained, regulated or changed in character, soils eroded, alien species introduced or enabled to persist and become invasive, and short term landscape planning for human needs for living, provisions and recreation become prevalent. Looming issues in the longer term include ramifications of climate change. Despite difficulties in evaluating these processes fully, responsible conservation demands some precautionary measures to counter such uncertainty. In short, many Australian insects (and other organisms) including many that depend on specialised environments and resources are threatened by habitat loss or other change to varying extents. Examples discussed earlier include declines of alpine zones, native grasslands, forests and coastal wetlands, and many ecosystems have been extensively fragmented and reduced to tiny proportions of their former extent. Such losses are associated widely with declines of specialised native organisms characteristic of those habitats. For insects, most of those declines are undocumented. We do not know how many species may have been rendered extinct: one of the practical problems of a largely undocumented fauna is that extinctions (as opposed to more local losses, extirpations,
Species Conservation
201
which are much more commonly reported) are extremely difficult to detect. Even for the best-documented group of insects, butterflies, no full species is known to have become extinct in Australia since information started to accumulate – but the distribution of many implies that their ranges have become severely fragmented and reduced in extent. For the conservationist, this poorly documented trend is a salutary signal for caution in further despoliation of natural biotopes. For the evolutionary biologist, the trend assures that many aspects of evolutionary pattern may now never be based on any complete picture and – just as palaeontologists have to cope with many gaps in the fossil record in interpreting evolutionary transitions – critical chapters in the story of Australia’s insect evolution may never become definitive. The need for conservation of insects encompasses such evolutionary considerations, for it is about their future potential as well as their current wellbeing. Conserving Australia’s insect heritage is of far more than parochial interest, but – reflecting the much that we do not yet know of the fauna – it is difficult to estimate the impacts of losses on our perspective or wellbeing in the future. Much of the current effort devolves on single species conservation, necessarily addressing only an alarmingly small proportion of the total, and with increasing realisation that this alone is not sufficient and that wider attention to habitat/biotope conservation might give greater benefit to larger numbers of susceptible taxa.
Species Conservation However, whilst many people accept readily that conservation of marsupials, rare birds, eucalypts or orchids is needed and merits strenuous effort and investment, equivalent support for most insects is difficult to garner. Invertebrates (together with other enormously diverse groups such as fungi and non-vascular plants) are the Cinderellas of conservation. People tend to think first of conservation need in terms of individual species, as providing a tangible focus. Calls to conserve the platypus, koala, Wollemi pine, red-tailed black cockatoo or growling grass frog are likely to gain support and sympathy; parallel calls to conserve an alpine stonefly, torrent midge, Leichardt’s grasshopper or golden-rayed blue butterfly, may be much less rewarded. Image, however, is only one influential factor. Another limitation of focusing on insect species is the sheer number of deserving species, many of which may become of concern. Whereas the number of conservation candidates amongst the vertebrates of any region may be no more than a few tens (however daunting even that total may appear to a manager with very limited resources!), that total might be exceeded even within a single family of beetles or moths depending on a particular threatened habitat. Lists of formally designated ‘threatened species’ or ‘protected species’ of insects under both Commonwealth and State/Territory legislations are very short, and cannot realistically be presumed to reflect the true scale of conservation need. Assuring credibility of the few designated species can itself be problematical, not least because of assurance of proper identification – and, of course, most officers
202
16 Conservation
and land managers responsible for applying protective legislation are unlikely to be entomologists and may have to rely on further specialist advice for insect identifi cation. A particular species of bullant (p. 108) or jewel beetle (p. 94) listed as protected amongst an array of very similar species of lesser conservation concern may be very difficult to identify. Any list of insects believed to be threatened can become very long, and simply impracticable to deal with – the needs far outstrip the resources and expertise available to support conservation management. It then becomes necessary to select the ‘most deserving cases’ for treatment, and doing this rationally and convincingly is difficult, not least because selecting some species may deprive others, equally deserving, of the attention needed to save them from loss simply because support resources are not then available. Following its original medical meaning, such triage may effectively condemn non-selected species to extinction – so that deciding what not to select is also ethically problematical. Conventionally, priorities have often been designated on some estimate of ‘risk of extinction’, most commonly following a hierarchy of threat categories developed through the World Conservation Union and with the ‘most threatened’ species accorded priority. The ranking criteria incorporate aspects of population size, number of populations, extent of loss and decline in abundance and distribution – basically as indices of the rapidity with which the species may face extinction and the urgency of conservation needed to prevent this. For long-lived, large, conspicuous kangaroos and other vertebrates whose biology is reasonably well understood and reproductive output small and predictable, many of these parameters may be assessable easily. Not so for most of our insects – even if we know what they are, early stages may not have been associated reliably with adults, the seasonal pattern of development may not be clear, population sizes are difficult to measure and it may be entirely normal for abundance to vary several-fold between successive generations, so that reliable detection of change in abundance may be a long-term task. The mountain grasshopper (p. 153), for example, is abundant in some years and hardly seen in the same places in other years. The reasons for this variation are unknown. And biological background knowledge for many parallel cases may be highly incomplete. The combination of perception barrier and poor knowledge may make progress very difficult. Basic research on which to found management is integral to most insect species conservation planning in Australia. The major exceptions are the very few groups of insects that people ‘like’ and accept as worthy of conservation, and which are sometimes among the bestdocumented taxa. Butterflies are the paramount popular group, particularly from a long history of collector interest stemming from Britain, and other parts of the northern hemisphere. They are the only insect group in Australia for which relatively comprehensive appraisal of conservation need has been attempted on a species-level basis through a national conservation ‘Action Plan’ that reviewed the status of all species and subspecies of any concern, and drew heavily on advice from the hobbyist fraternity to help justify the status allocated to each taxon. Dragonflies are perhaps the next most popular group – also large, colourful, day-flying insects with consi derable appeal. In Japan, widespread cultural sympathy for dragonfly wellbeing has led to designation of numerous reserves especially for them. However, many other
Species Conservation
203
insects appear more as curiosities than welcome additions to lists of threatened species, and then as isolated members of, sometimes, very diverse lineages with most of their relatives not ranked or assessed for conservation need – they are thereby simply the visible ‘tip of the iceberg’ of those possibly threatened. The conservation status of most Australian insects is simply unknown, other than for the opinions ventured by the few informed workers on any individual group and who have practical and up-to-date knowledge and experience of their distribution and resource needs, together with knowledge of any declines and the possible reasons for these. Emphasis on ‘threat’ as a major criterion for conservation need raises a matter that has led to much ambiguity – the possible confusion between ‘rarity’ and ‘vulnerabi lity’ of an insect. Many insects are encountered very infrequently and are thought of as ‘rare’, because of their elusiveness or very small numbers in samples. Many of the singletons in community samples (p. 66) are referred to in this way. Ecologists have pointed out that rarity has several rather distinct connotations – it can refer to low abundance, to narrow distribution, and to ecological specialisation, so that an insect species with low numbers, very restricted distribution and/or resource needs (such as feeding habits) is ‘rarer’ than one with the converse characteristics. From this perspective, the rarest insects are those that show all three features, which can occur in any combination, and as many insects of greatest conservation concern really do. However, rarity may not represent threat, and be a wholly natural and stable condition: many insects that have long been rare have not changed obviously to become more threatened or vulnerable. Nevertheless, should conditions change, any of these dimensions of rarity may then predispose the insect to vulnerability – and the rarer it is, the greater is the probability that this may eventuate. Australian plants are categorised widely as ‘rare or threatened’ to incorporate this duality. A similar principle applies to the major and potentially universal threat to insects, the loss of habitat and critical resources. By definition, a small patch of suitable habitat may be able to support only a correspondingly small population of an insect compared to the capacity of a large patch, and is thus considered widely to be the more vulnerable. Another general principle is also very pertinent – that ecologically specialised insects in patchy habitats may need conditions that can be sustained only by being protected from disturbance or outside influences. So-called ‘edge effects’ become increasingly important as habitat area declines. This means that outside influences (such as weed invasions, incursions by alien animals, climate changes from increased insolation, and others) penetrate across the edge into the habitat patch and the smaller the patch, the less pristine ‘core’ remains free of their influences. If we sample insects such as ground beetles at various distances across a habitat edge – say, the border between forest and adjacent cleared ground – we may find that many of the most characteristic forest species do not occur near or beyond the edge and that species from ‘outside’ that thrive in more open conditions have penetrated the forest, perhaps competing with the more specialised resident species where they overlap. Even when a small habitat patch continues to support popu lations of these residents, they may be ‘pushed’ increasingly toward the central, less disturbed, region. And in smaller patches pristine interior habitat may disappear completely over time. Conservation of any insect species involves protection of
204
16 Conservation
Fig. 16.1 Lord Howe Island stick insect, Dryococelus australis, an endangered species rescued from extinction by captive breeding (Photo: Patrick Honan)
habitat from both loss and lessening quality, and detect and ameliorate threats that engender conservation concern. Each case needs independent evaluation, but studies on forest beetles have indicated that edge effects on species composition can extend for at least a kilometre into the forest! Such lessons are salutary in designing reserves for insects and other organisms. Alien species are, after habitat destruction, the most general threat to many insects and are frequent considerations in conservation planning. One, involving the toxic nature of an introduced vine to a rare butterfly, was noted on p. 90. Alien predators may be particularly destructive. Predation by rats, following their introduction in 1918 was believed to be the major cause of loss of the spectacular Lord Howe Island stick insect (Dryococelus australis, sometimes termed the ‘land lobster’, Fig. 16.1) from the island early last century. The insect was feared extinct until a tiny population was found on the nearby rugged Balls Pyramid in 2001. It was restricted to a single small clump of the food plant (Melaleuca howeana, itself threatened by being smothered by the alien morning glory, Ipomoea indica). Other than Lepidoptera, this is one of the most ‘charismatic’ Australian insects that need conservation, and the successful captive breeding programme founded in 2003 from a single wild-caught pair has been instrumental in rescuing the species: a 2010 report noted the captive population as then comprising more than 700 individuals and 14,000 eggs. Dryococelus has not been alone in succumbing to rats on Lord Howe Island: around 15 species or subspecies of vertebrates and probably many invertebrates
Species Conservation
205
have also been lost, although some survive on rat-free islets in the region. Rodent eradication programmes at present in progress should allow rat-free conditions for release of captive-bred stick insects to a threat–free environment. Rats and related rodents are almost invariably accidental introductions, notoriously easy to convey in ships’ holds. A few vertebrates of major conservation concern, however, have been imported deliberately. In Australia, the most worrisome is the cane toad (Bufo marinus) imported in 1935 as a biological control agent for a native scarab beetle (Dermolepida albohirtum) in sugar cane fields, and which has continued to spread and have substantial adverse effects on native animals over much of northern Australia, whilst not achieving its original purpose. Many insects are known only from single populations or single sites, many of them appearing vulnerable as they become increasingly isolated and fragmented from human pressures, and difficult to protect effectively against competing inte rests and priorities. The wingless Mount Donna Buang stonefly (Riekoperla darlingtoni) has been called ‘the most relict stonefly in Australia’, and is known from only a few small cold water trickles near the summit of its namesake mountain in Victoria, in an area popular for winter recreation and for summer tourism and sight-seeing. Some of the trickles in montane forest, in which the stoneflies tend to occur amongst strips of fallen eucalypt bark, are below roads and carparks, and subject to run-off from these, and any modifications or expansion as tourist facilities could overrun most of the known habitat of this notable animal. Without effective habitat protection for any such highly localised insect, more detailed management may be redundant. The sites needed may not be large but, following the inferences on fragmentation discussed above, bigger is better and any ways of buffering the site from edge effects should be considered carefully as a facet of protection. The presence of notable insects in larger areas, such as national parks, provides opportunities for habitat protection that are increasingly difficult in other wider landscapes as surrounds become increasingly alienated and pressures for land use multiply. Small urban sites pose particular problems for insect conservation, as they can be surrounded by housing and roads that bear no resemblance to the parental habitat, and which can form very effective barriers to insect dispersal. Because of their isolation and size, each such site may need individually-tailored and continuing management – the focal insects become essentially ‘conservation-dependent’, and expensive to deal with. The Eltham copper butterfly (Paralucia pyrodiscus lucida, Fig. 16.2) is named for the outer northern suburb of Melbourne where it was found on several small sites in 1987, having been believed by many to have become extinct in the region due to urbanisation. It became a local flagship species for insect conservation, and the sites of only 1–2 ha have received continued attention since the late 1980s, when they were designated as butterfly reserves. There is no evidence that adults can move between sites separated by only a few hundred metres of housing developments (p. 133), so that each site is presumed to support an independent population, subject to the vagaries of highly localised influences from external pressures and individualistic succession of vegetation. As for any such insect, knowledge of these impacts can be evaluated only from understanding its basic biological needs. Paralucia pyrodiscus is amongst many Australian Lycaenidae that participate in
206
16 Conservation
Fig. 16.2 The Eltham copper butterfly (Paralucia pyrodiscus lucida) is the target of one of Australia’s longest-running insect conservation campaigns, focused on small populations on remnant urban sites near Melbourne: it is an important ‘flagship species’ for conservation advocacy (Photo: A.L. Yen)
complex relationships with ants, so that its critical resource needs encompass nectar plants for adult butterflies, a specific larval foodplant (sweet bursaria, Bursaria spinosa) and specific mutualistic ants of the genus Notoncus. Eltham copper caterpillars are sheltered by day in the underground nest chambers of Notoncus around the base of individual Bursaria plants, and are escorted by ants up the plants to feed at night. As in many similar associations between ants and caterpillars, it is believed that the ants protect caterpillars from attacks by natural enemies, whilst feeding on sugar–rich secretions from specialised glands on the caterpillars’ body. But this tripartite association of ‘butterfly-plant-ant’ adds a further conservation dimension to the more usual ‘butterfly-plant’ alone one. A rather different conservation need for insects on small urban sites is to garner support from the local residents, and for this to be linked with responsible management and administration from local councils or other supervising body.
Habitat Conservation The above example demonstrates the importance of even small sites for focal species. Habitat patch size is, of course, also important for conserving insect diversity – with smaller sites commonly supporting fewer species than larger ones can do. Whilst any precise general relationship between patch size and number of resident species is perhaps impossible to define, a ‘working maxim’ has flowed from themes of island biogeography in relating island size with species richness. Originally formulated for the reptiles and amphibians of several West Indies islands, but subsequently
Habitat Conservation
207
extended widely as a generality, this principle is that reducing the area of an island to a tenth of its extent reduces the species richness by half. Analyses for insects, initially with ants over much of the western Pacific support this as a broad working formula, and the idea has enormous implications for equivalent habitat islands of many kinds. With the enormous losses of natural habitats in Australia, it is scarcely credible that enormous insect species richness has not been lost as well, and that many surviving populations are now much more isolated and vulnerable than previously. However, most of what we know about insects in habitat patches is based on studies of particular groups rather than all those present, and we may need to presume that their representation on each of a series of patches indeed represents fragments of the original conjoint species pool extending across all these – but, because of the extensive natural patchiness in insect assemblages, this presumption may not always be valid in indicating possible losses. Just because a species is not found, we rarely know whether it has been lost, or has never been there. But the simplistic management implication in regard to habitat area is that size reduction may become critical: higher insect diversity needs larger space in which to live and, even when we do not know how many species live there and what they are, there may be ‘safety’ in size. Conserving remnant habitats for insects may thus have benefits far beyond simply conserving a few focal species that live there, by providing at least a partial buffer against declines of numerous others. However, many insect habitats are not recent fragments or remnants, but small, specialised, and distributed patchily in the landscape. Caves, for example, support an array of insects and other invertebrates not found elsewhere, and which have developed from lineages found in more open habitats. Whereas we may be able to broadly characterise the insect fauna of widespread grasslands or forests, many caves may each have their own independently developed fauna as narrow endemics not found anywhere else. Thus, two ground beetles from Tasmania are found only in the Ida Bay karst area of the south east. Both the Ida Bay cave beetle (Idacarabus troglodytes) and the blind cave beetle (Goedetrechus mendumae: loss of eyes is a widespread trait in cave insects living in the dark) appear to have small populations distributed over only a few square kilometres, with parts of their cave system vulne rable to irregular water flow from surface mining and forestry activities. One conservation measure has been (in part voluntary) regulation of access and activity by recreational cavers. A second example, also of beetles, is of the water beetles found in subterranean aquifers in Western Australia (p. 139). Predatory Dytiscidae have radiated extensively. Many of the 80 or so species described up to 2008 are small (2–4 mm body length) and so can live in the spaces between stones or sand grains in the substrate, and do not need much food. They have been called ‘the world’s most diverse assembly of subterranean diving beetles’ and, because many aquifers have yet to be surveyed, it is likely that many further species may be found. Protection of such habitats is still somewhat fortuitous, aided by their remoteness or inaccessibility, but the wider aim of securing a fully representative series of Australia’s natural habitats in effective reserves has massive implications for insects but is still not a guarantee against extinction. However, the converse scenario of allowing many habitats to decline or disappear will assuredly lead to loss of many species.
208
16 Conservation
Ideally, a planned reserve system must occur at landscape levels to incorporate considerations of connectivity and also incorporate sympathetic management of urban and agricultural systems, as well as those ecosystems that we deem more natural. Agriculture is the single greatest modifier of land use, and many agricultural landscapes bear little or no resemblance to those they have replaced. Considerable effort is now being made to de-intensify some aspects of agricultural production, for example by more enlightened pest management and cultural methods, and to harmonise these systems more with natural landscape and conservation values.
Wider Views In addition to ethical commitment, two other major grounds for insect conservation provide important messages for people involved with the care of Australia’s natural environments, and reflect the sensitivity of many insects to environmental changes and their central ecological roles. The first of these involves the values of insects as ‘indicators’ – invertebrate equivalents of the ‘miner’s canary’ in providing early signs of danger or the impacts of changes. The second in some ways correlates with this, in emphasising the importance of insects in sustaining ecological processes. The term ‘indicators’ is used in several ways, but all incorporate the principle that particular insect species or groups may change in abundance, richness or assemblage composition to mirror environmental changes in some way. Earlier, we noted the functional groups of Australian ants (p. 118), with changes in the balance of these easily-sampled insects providing ‘indicator value’; the ‘EPT Index’ incorporating larvae of mayflies, stoneflies and caddisflies (p. 27) is an aquatic parallel. Indeed the concept of indicators has been validated most effectively amongst aquatic insects, and many ambiguities remain for most terrestrial taxa, but in short many aspects of the impacts of stress imposed by human activities may be revealed by study of changes in insects. Chemicals may induce deformities, so that diminished water purity can lead, over time, to increased frequency of mouthpart deformities in aquatic midge larvae. This, or early losses or declines of monitored insects signal needs for remedial actions or, in a more positive context, changes for worse or better allow us to track the progress of remedial measures such as habitat restoration. Ants have been used to follow the rehabilitation of bauxite mine sites in Western Australia, for example, with considerable changes in their assemblages along the trajectory from bare ground to vegetation assemblies a decade or so later. However, whilst there is reasonable consensus that ants can be used in this way, many insects have been promoted for indicator values on far less experimental evidence of their responses being equally predictable – simply reflecting ecological variety so that ‘everything indicates something’! Even when they are reliable, the practical matters of sampling, identifying and interpreting the changes commonly remain formidable – simply ‘receiving a signal’ is far different from understanding the message it conveys.
Wider Views
209
Major categories of indicators are distinguished by many ecologists, adding to possible confusion in interpretation but revealing the variety of contexts in which insects may be useful. The three main categories are (1) environmental indicators (one or more species that respond predictably to an environmental disturbance or change); (2) ecological indicators (one or more species that demonstrate effects of environmental changes such as habitat or climate change); and (3) biodiversity indicators (a group of related or unrelated species, or functional group, that reflects some measure of diversity – such as richness – of other higher taxa sharing the habitat). These insects can have important roles in helping us to assess and monitor environmental changes but not, necessarily, the functional changes that result. The ecological services roles of insects are even more fundamental to consider, as affecting directly many of the ways in which ecosystems function. Considerable concern has been expressed globally over declines of pollinating insects, for example, with the causes of loss including widespread use of agricultural chemicals and alien species interactions. Likewise, one of the strategies for improving management of crop pests is to promote uses of native predators and parasitoids already living in the relevant areas, rather than incur risks from introducing alien biological control agents. In order for such native species to be effective, the resources they need must usually be augmented by provision of native vegetation to support ‘natural’ prey and hosts and so to function as reservoir habitats over periods when crop pests are not present. It may also be necessary to facilitate their access to the crop in some way, so that conservation occurs with some redesign of agricultural landscapes. This ‘conservation biological control’ allows us to take advantage of the flexibility of native natural enemies to move to a novel food supply, rather than emphasise the specificity so critical for classical biological control agents. Considering ecological roles of insects also takes us beyond the relatively tangible ‘species approach’ to assessing conservation, to emphasise the importance of ecological associations, often incorporating unknown or poorly documented taxa and interactions. The great variety of feeding relationships outlined earlier results in highly complex food webs in which many insects have roles that remain largely unclear – it is still unknown, for example whether many small wasps are parasitoids or hyperparasitoids, and the host range of most has not been defined fully – if detected at all. Conserving ecological interactions, based in conserving the integrity of ecological processes, helps to emphasise the central importance of insects well beyond what we can at present quantify or explain fully, and involves many insects that we do not notice as individual taxa. Preservation of vegetation remnants, wetlands, and other major habitats of insects thus becomes critical for conserving insect diversity harboured there. Loss or decline of key plant species or other resources may lead to cascade effects: loss of resources for a particular butterfly, for example may cause declines of specific predators, parasitoids, mutualistic ants, and so on – but also to losses of co-occurring and coexisting species well beyond this immediate module of interactions. The ‘individual species approach’ simply helps to convey these important messages in some tangible form. In essence the representative species that gain individual treatment are the flagships on which insect conservation
210
16 Conservation
endeavour must sail. Participants in ecological interactions, and contributions to ecological integrity span the whole gamut of insects (and their relatives) present; consequences of their declines or loss may be immense.
Further Reading Honan P (2008) Notes on the biology, captive management and conservation status of the Lord Howe Island stick insect (Dryococelus australis) (Phasmatodea). J Insect Conserv 12:399–413 (account of one of the most innovative and important insect species conservation cases in Australia) New TR (2010) Butterfly conservation in south-eastern Australia: progress and prospects. Springer, Dordrecht (survey of work on butterfly conservation in the region) New TR, Sands DPA (2004) Management of threatened insect species in Australia, with particular reference to butterflies. Aust J Entomol 43:258–270 (broad overview of approaches and examples)
Chapter 17
Diversity and Its Implications for Understanding Australia’s Insects
Introduction: Relevance of Basic Documentation In the preface, mention was made of the rather daunting figure of, possibly, 200,000 or more species of insects occurring within Australia, and much of the intervening text has been about some aspects of this diversity – how it may have arisen, where it is found, what it is, and whether any such bland number can at present represent reality. This chapter provides more background towards being able to analyse this, and to examining its relevance and importance more effectively. It is obvious that any enumeration of insect ‘species’ incorporates various meanings of this term, and that inflation of any suggested number may be achieved through incorporating genetic or molecular information – so that the intergrades between ‘species’ and ‘populations’ are increasingly ambiguous, with refuge sometimes taken in terms such as ‘evolutionarily significant units’ to express variety at levels below formal species. For most insects we still have only minimal knowledge of the ecological variations within a species – in a few better-known groups, such as butterflies, proliferation of named subspecies reflects aspects of that variety, whether trivial or biologically significant – but the ‘proper’ status of local populations, or even overlapping populations, of particular moths, aphids or other insects each restricted to a particular foodplant but allocated a single collective species name indicate the existence of different ‘ecomorphs’ (or ‘races’) within a single recognised taxon. Such biologically discrete forms may be the precursors of distinct species as their isolation persists. The tendency to either inflate or reduce estimates of insect species numbers can become very idiosyncratic, and specialists in any large insect group are likely to differ in their opinions over the proper status of many entities and how ‘a species’ may best be defined. In part, such differences reflect the importance each such person may give to ‘typological’ or ‘biological’ species ideas, mainly in the absence of any information on reproductive compatibility or isolation. We must remember that the great majority of insect species have been described solely from dead specimens, often stored in museums for decades before they are examined, and often from few individuals and only one sex, with little (most commonly, no)
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0_17, © Springer Science+Business Media B.V. 2011
211
212
17 Diversity and Its Implications for Understanding Australia’s Insects
information on their variation or biology. Incidence of polymorphisms, whereby interbreeding insects clearly of the same species may differ markedly in appearance – such as different colour morphs of mimetic butterflies, or winged and wingless individual grasshoppers, perhaps also of different colours – add to the possible confusions. Within the insects, the number of species, however these are defined, is only one index of diversity, albeit an important one, and is sometimes referred to as ‘alphadiversity’ or simply ‘species richness’. It is politically useful, simply because high numbers can be impressive to politicians in conservation and other arguments, but there is a tendency to equate ‘counting species’ with conserving them. Counting species can become an obsession, perhaps understandably so as a fundamental aspect of documenting the complexity of the natural world, an exercise described recently as ‘one of the great intellectual challenges facing humanity today’. But, as a leading tropical region biologist, Dan Janzen, has pointedly asked – does it really matter if a particular tropical forest (in his example within his primary area of interest, Costa Rica) has several thousand insect species or several thousand plus a few more? The resultant knowledge of the forest would be marginally improved, but the potential to conserve it and its inhabitants would be unlikely to change – in either case we would know that there are a lot of insect species, many of them apparently scarce and some unknown elsewhere, and that by any reasonable threshold estimate of diversity, they should be conserved. Increasing numbers of insect surveys in Australia have indeed yielded up to several thousand species in very restricted areas or habitats, particularly in the tropics. The greater concern, whether in a Costa Rican forest, the rainforests of Queensland, or the grasslands of Victoria, is that we now have probably only a fraction of the insect species present that bred there only two centuries ago, before commencement of post-European settlement despoliation, and which continues. Yet, and notwithstanding the comments above, counts of insect richness can be important in practice – high richness may be associated with high ecological complexity and integrity, the ability of ecosystems to function well and to resist disturbances, and may help justify priority for conservation of some areas over others. Loss of richness, or unexpectedly low richness in relation to some reference point may cause concerns: widespread feeling persists that ‘high diversity is good’ and ‘loss of diversity is bad’. Unexpectedly low diversity may also set off warning bells about the condition of the environment. Interpreting that richness further to demonstrate levels of endemism, the occurrence of unusual species, and so on, can strengthen argument for reservation or management. Most such discussions are based on comparative analyses of particular groups of insects, rather than ‘all-taxon inventories’.
Surveying Diversity Surveying any local insect fauna in its entirety is formidably difficult and almost impossible without at least several years of work using a variety of techniques, and any comparative assessments across sites or habitats should be based in comparable
Surveying Diversity
213
levels of sampling in each. Such approaches lead to a second category of ‘diversity’ used by ecologists: ‘beta-diversity’ is a measure of the similarity of sites or habitats in terms of the species present – analysed as the proportion of shared species between pairs of sites, or the turnover between sites along a gradient. Any such comparison obviously depends on interpreting ‘species’ in the same away on all contributive surveys, which may have been undertaken by different people over different periods and using different methods. A universal problem with surveys of this kind is ensuring the consistency needed, by ‘quality control’ – not only in field collection, but also in the later sorting and categorisation of insects in the samples. It is increasingly common, and necessary, for parts of this intensive and laborious work to be undertaken by non-specialists, but uncritical participation by volunteers or technicians with limited expertise and training in how to differentiate the species present can lead to problems. Aquatic insect samples in Australia have been used to test this, by looking at the various sources of errors that arise in sorting by these ‘biodiversity technicians’, and as components of so-called Rapid Biodiversity Assessment exercises undertaken to evaluate species richness efficiently as a tool in assessing environmental condition or change – essentially using the diversity represented in samples of selected taxonomic groups to reflect wider biodiversity, and increasing sample processing as expediently and economically as possible. It is often not appreciated by non-entomologists that in insect diversity studies, actually getting the specimens is usually only a small part of the total costs and effort of interpreting the material: processing costs can be manyfold higher than those of the preceding fieldwork, and reflect the large numbers, numerous species and taxonomic complexities in the samples. Tiny wasps or flies, as examples, must be prepared and handled very carefully and much information is lost by failures to do this. Sorting problems in enumerating species arise from several of the ambiguities over ‘morphs’ mentioned above – that larvae and adults can be very different in appearance and almost impossible to associate unambiguously (in practice this is overcome for most terrestrial insect surveys by pragmatically ignoring larvae, but they are far more important in freshwater surveys), confusion between different instars of aquatic larvae as they grow, differences between males and females, wing polymorphisms (such as long-winged, short-winged and wingless individuals), seasonal colour forms, and so on. In addition, there may be substantial differences between individual sorters – such as in visual acuity, perhaps colourblindness, and extent of deviation in standards with tiredness or boredom increasing over prolonged periods of sorting. Several hours of sitting at a microscope sorting small insects can become very tedious. Quality control is a serious consideration in any exercises of this sort, and validation of categories by more experienced workers is a routine need, as is the eventual checking of species identifications by a specialist in the group(s) concerned. However accurately we can assess the insects present in any given environment or context, we then face the difficulties of explaining why the richness or composition differs between sites or habitats. Thus, we usually do not know if absence of species (even if real; many insects are notoriously difficult to find, despite intensive longterm searches) found elsewhere means that these have been lost, perhaps evincing conservation concern, or that the differences are part of the fine grain spatial variations in distribution that characterise so many insect species. Natural patchiness and
214
17 Diversity and Its Implications for Understanding Australia’s Insects
local endemism are both common amongst Australian insects and it is presumed widely, largely without proof, that destruction of – say – a patch of forest will take with it some insects that occur only there, and so impoverish the total forest fauna. The wider scenario is that sometimes referred to as ‘Centinelan extinctions’, named for an Andean ridge in Ecuador that supported many undescribed forest plants but was cleared before these could be recorded so those species can never be known – these are also called anonymous extinctions, and Australia is assuredly not alone in having lost numerous insect species before they were known; now, they can never be known, and the magnitude of loss will remain undocumented. However, there are some clues from collections made since European settlement, notably that many of the insects described from 1775 onwards have not been re-collected on any later occasion, and many modern taxonomic revisions include new species based on single individuals or very few individuals. Vagaries of collecting, and of retrieval of species in even targeted surveys, are important considerations in evaluating diversity and possible losses, and the serendipitous rediscovery of some species not seen for decades or more counsels caution over presuming that they may have become extinct. One hopes that the example of rediscovery of the Lord Howe Island stick insect (p. 204) may, fortuitously, be multiplied many times and extend to many other taxa at present elusive and, possibly, even extinct. Proving that a naturally rare and elusive insect has indeed been lost is very difficult. Impacts of habitat loss are essentially to decrease habitat extent, often to decrease quality, and to increase fragmentation and often, in consequence, isolation of populations – but isolation itself may foster diversity, as in development of some putative subspecies of butterflies on differentiated habitat fragments that each support an independent population in which variety can be generated. However, the above discussion is mostly about ‘typological species’, those designated from visual, structural differences, and most designations have paid little attention to molecular or genetic differentiation, or biological differences between members of sibling species groups. Surveys of this restricted nature, however valuable, are thus likely to underestimate the variety present in favour of that easily detected, so that many of the features revealing cryptic species indeed remain cryptic. A true enumeration of Australia’s insect species can probably never be finalised, but we can progressively enhance what we do know, and understand better what we do not know, and what it may be important to discover – with the latter helping greatly to focus attention for the future.
Increasing Understanding ‘Taxonomic species’, here those insects designated by names and consensus, are a first framework for understanding, and the total will always be open to refinement as new material and ideas come to hand. A practical sequence advanced by R.W. Taylor in the early 1980s is the transition from ‘species in nature’ to ‘species taxonomically understood’, through the intermediary ‘species represented in collections’.
Increasing Understanding
215
As Taylor (1983) remarked, with Fabricius’ publication, 1775 is the only historical point where all known Australian insects had names!. Since then, the accumulation of species in collections has outpaced the rate of taxonomic description, and it is perhaps unlikely that such a mark will ever be reached again – not least because of the greater awareness of the subtleties of insect variety. Several commentators have traced the changing nature of insect exploration and documentation in Australia, from the initial ‘Period of Exploration’ based on overseas visitations, through the increasing participation of resident naturalists (sometimes termed the Macleayan period, after the dynasty of influential naturalists, from about 1860 on) and thence the progressive passing from mainly amateur interests to domination by professionals (from about 1928 to the present). Throughout this history, collecting and preservation of insect specimens has been critical. Diagnoses and definitions of species can be prepared only from material available for appraisal, and the importance of continuing to collect assiduously throughout Australia cannot be overstated as we seek to improve our national perspective. Whether or not insects in collections receive names is to a large extent at the whim or fortuitous interest from a specialist in that particular group. There is some tendency for members of ‘popular groups’ (such as butterflies) and members of economically important groups, including many pests, to be described expediently, because interest or applied need is evident. Other possible biases are being large (but, again, with the caveat on stick insects noted in the preface) because many small insects are more complicated to differentiate, and insects found near human settlements. A recent survey of description history of Australian scarab beetles suggested that wide-ranging species were generally described earlier than many highly localised taxa, irrespective of their size, and simply reflecting greater likelihood that they would be collected. Proximity to settlement is in part reflected in that the south eastern fauna has generally been described considerably earlier than insects from many remote northern regions of Australia. However, many insect groups have been studied by only a handful of interested taxonomists over the years, and many are in practice largely ‘orphaned’ in that no-one is currently championing them and working actively on the Australian representatives. This unfortunate situation is recognised widely by scientists, but may convey the political message that our insects are simply too unimportant to study: a message misleading in the extreme. Many of our insects are still not yet even represented in collections. The numbers of insects not yet collected are very difficult to assess, but opinions of specialists suggest that for some of the larger terrestrial orders, in particular, they might be considerable. Collections from remote areas of Australia continue to yield novelties, and often to raise queries on the validity of material identified earlier. Many geographical gaps are likely to persist – the intensive surveys noted as revealing the variation of the grasshopper Caledia (p. 47) are unlikely to be replicated widely other than in the rare context of individual doctoral studies and, despite increasing calls to understand ‘biodiversity’, the sheer scale of field effort needed to understand Australia’s insect diversity remains formidable. Add the need for long term surveys in any area to reveal seasonal variations in insect apparency, and the logistic complications of field work in remote areas at some times of the year, and what has
216
17 Diversity and Its Implications for Understanding Australia’s Insects
so far been achieved becomes itself remarkable. Many of the most extensive entomological surveys have been undertaken through State museums or the national CSIRO-based Australian National Insect Collection, and have added enormously to perspective of the main focal groups, with the material archived responsibly for future taxonomic and faunistic study. Surveys have encompassed two major approaches, both valuable and their discoveries complementary. These approaches can be thought of as ‘intensive’, concentrating on particular places such as ‘hotspots’ or particular habitats to determine as fully as possible what lives there, and ‘extensive’, covering large areas or distances with ‘whistlestop’ collecting at intervals, perhaps predetermined by regular distance intervals (or, increasingly, by predetermined GPS points) or in relation to habitats. In trying to enumerate insect species, an intriguing philosophical dilemma arises – simply, because insects and their critical resources are highly dynamic systems, continually responding in subtle ways as their coevolutionary milieu changes, we are dealing with entities that are inherently difficult, perhaps even impossible, to categorise consistently. Any stated number of species, however well justified on typological or biological grounds, is simply a working hypothesis, just as each individual species also is (p. 52). The boundaries between species are unlikely ever to become wholly precise whilst species coexist and interact, but the ‘best possible’ framework of formal taxonomy based on comprehensive revisionary study is the major path to greater understanding of the insects, based on named species. A name confers respectability, whilst any less formal epithet (despite the widespread use of ‘codes’ as an intermediate step in study) may be open to greater ambiguity. The context of ‘revision’, above, is also important. In the past, many insects have been described in short papers with titles along the lines of ‘A new species of genus x, order y’, or similar limited scope. Many are invaluable in providing names for species of economic importance and many also describe the only members of that lineage or genus in Australia; and many are prepared by recognised experts who may have studied that group for many years, so the soundness of their scholarship is undoubted. However, problems to users can arise in two main ways. First, such papers, by their very nature, do not reflect all the possible ambiguities in a large genus or larger insect group – most are written predominantly for other specialists to whom the limitations will be clear. However, a non-expert relying on the paper to identify an ant, beetle, fly (etc.) covered by the title may not appreciate those limits and errors can be perpetuated if uncritical identification is made by ‘going it alone’. Geographical variation, for example, may not be appreciated. Second, and related to this, the reader needs to be able to diagnose that species in the far wider context of the variety in the higher taxon – if the paper describes a new species of weevil or ground beetle, that beetle must be assessed firmly in the context of the genus and tribe within Australia. Leading ground beetle experts in New Zealand recommended recently that isolated descriptions of new taxa are misguided and that beetle taxonomy should be pursued in the context of revisionary studies, rather than piecemeal. Such sentiments have been advanced repeatedly over recent decades and for many diverse groups of insects. They sometimes result from frustrations over the taxonomic ‘mess’ caused by poorly informed individual descriptions. ‘Names’ for insect species
Increasing Understanding
217
are invaluable if accurately applied, but can be very misleading if mistakes are made in identifications – in essence, the values of names parallel Alice’s response to the Gnat in ‘Alice through the looking glass’: the Gnat asked ‘What’s the use of their having names if they won’t answer to them?’, to which Alice replied ‘No use to them, but it’s useful to the people that name them, I suppose’. The nature of a revisionary study was noted earlier in relation to thrips (p. 52) and, as the term implies, will bring together all material available for study and discuss the applicability of all available species names, based wherever possible in examination of the specimens to which those names were first applied: the ‘type specimens’ as the ultimate reference point for each name used. Two outcomes are common (1) previously undetected and undescribed species are diagnosed and named, most commonly with the aid of measurements, illustrations and use of a dichotomous key to characterise them in relation to their close relatives, and (2) some previously used names, designating species recognised to that point, are found not to apply to real species, but to redescriptions of species described beforehand but with patterns of variation not then appreciated. These are reduced to synonyms of the name of the first-described species. A good revision therefore points out the strengths and weaknesses of knowledge, but brings perspective of the group – usually a genus or family but with no formally ordained coverage – as far as possible up-todate as a foundation for future work. Within Australia, many recent revisions have increased the number of included species considerably. As examples, a revision of three genera of parasitoid wasps increased the number of described species from 64 to 266, of a single genus of beeflies (Comptosia) from 26 to 138, and of the beetle subfamily Pselaphinae from 400 to 1,600! Such higher numbers were, of course, suspected in general terms, but revisions add precision and evidence. A comprehensive revision may not always lead to increased numbers of species. Spider ants (Leptomyrmex, Dolichoderinae) are a characteristic group in Australia and some nearby areas, and their characteristic appearance led to more than 40 species and subspecies names and considerable practical difficulties in telling taxa apart. A recent revision has clarified this confusion, with nine subspecies raised to full species status and 19 names synonymised to leave a group of 21 definable species in the larger species group, but with no entirely new species named. This satisfactory outcome is often not possible, and even such definitive revisions may leave precise details elusive, again reflecting lack of collecting – a leading specialist in weevils, in a magisterial study projected as eight volumes to introduce the scope of the Australian fauna, noted that some of the keys he produced were simply to separate the specimens before him, rather than to the full species they might represent. There is some danger that any revision might be taken as the ‘last word’ on the group involved. It is not. It facilitates further investigation and may stimulate others to study the group further, to augment the material in collections, and to study ‘problem groups’ more closely: that work and material becomes the basis for the next revision of that group, progressively working toward even greater elucidation of its richness and variety than was possible from the ‘first revision’ alone. Very high levels of novelty are likely to be encountered for some time to come in the larger endopterygote orders, in particular, but many entomologists believe that a very high proportion of
218
17 Diversity and Its Implications for Understanding Australia’s Insects
species within some of the smaller orders in Australia have now been described or, at least, recognised, and that fewer surprises remain. Thus, whereas a recent estimate for numbers of Megaloptera in Australia is of 24 described from a projected total fauna of 26 species, parallel estimates for Diptera are 6,432 of 30,000, and Coleoptera 22,903 of 80,000–100,000 species! However, simply delimiting species, however it is done, is only a part – although an important part – of understanding insect diversity. Consider the most common information that may be available or inferred from documented samples of insects taken from habitats as representatives of the wider environment in which each might also live. Labels associated with the specimens or samples are likely to tell us (1) place of capture, perhaps quite precisely from a grid reference or latitude/longitude combination; (2) date of capture, as reflection of season of availability; (3) often some idea of habitat or major biotope; (4) sometimes method of capture and/or more habitat details such as ‘on flower of x’, or ‘reared from host y’. Most such information is very general, and ecological information is relatively rare on museum specimen labels. In short, we know nothing about the detailed biology of most insect species that have been collected and described, except what we might infer- often very misleadingly – from any knowledge of related species. We may not know even whether the insects captured are characteristic of the habitat where they were found, because many flying insects are simply ‘tourists’ passing through any area and captured fortuitously by an intercept trap or attracted to light. An insect species is simply a named specimen or series of specimens that happens to have been described. Structured sampling to clarify aspects of beta-diversity (above) can help interpretation by revealing the extent of distributions of some species, but a third ‘level’ of diversity, called ‘gamma-diversity’ is valuable in expressing the extent of spatial patchiness or uniformity of insects within any particular habitat. Rather than comparing across different habitats, as with beta-diversity, gamma-diversity is used in comparing the species turnover between similar habitats in different regions. Thus, different patches of rather similar forest in Queensland may each have locally endemic insects, so high gamma-diversity. One implication is that each such patch is individually important to conserve for our insect heritage, a situation that may contrast markedly with habitats that support a more uniform suite of insects and in which such local endemism is not evident, or high. In this case, some patches of habitat may indeed be considered expendable, as their loss would not diminish the overall species pool of insects depending on that habitat type. However, such ‘death by a thousand cuts’ leading to gradual loss of biotopes already severely restricted in extent can rarely be condoned. Australia is by no means alone in lacking full documentation and understanding of its insect diversity; significant gaps are present also in many other parts of the world. Perhaps only in western Europe and for parts of North America can largely definitive enumeration – including realistic assessment of what is still not known – be made, but even in the United Kingdom, some insects remain difficult to identify consistently to species level. The largest order in the British Isles, Hymenoptera (with slightly more than 7,000 species of a total insect fauna exceeding 22,500 species) still contains many problematical groups, in which variations induced by
Increasing Understanding
219
environmental factors or related to different hosts are not yet clear, and amongst which many closely related and similar species occur. Identifications to beyond genus level, even if that level is possible, are effectively impossible without specialist advice. With such problems persisting in the best studied of all regional insect faunas, one documented over centuries of study and by many resident entomologists, both professional and hobbyist, the major gaps in knowledge elsewhere become more understandable. Consider, for example, the strong tradition of interest in butterflies and moths, perhaps the most important foundation for popularity of insects, but founded predominantly on only one small subset of the Lepidoptera, butterflies. The larger ‘macromoths’ are also quite well-known, but only those of the northern temperate regions can be considered definitively appraised – recent commentators have noted, for example, that the large Indian fauna has not received serious comprehensive treatment since the several volumes in the ‘Fauna of British India’ series produced from 1892. A major exception, of far more than the local interest implied from the series title, is the multi-volume ‘Moths of Borneo’, now nearing completion as the first real introduction to this rich regional tropical fauna. However, the ‘black hole’ of the Lepidoptera is the so-called ‘Microlepidoptera’, most of them tiny nocturnal moths incorporating the primitive lineages of the order, and many of them with caterpillars feeding inside plants, rather than being exposed. Again for south-east Asia, a review undertaken in the early 1990s suggested that the 6,000 or so species then described from the region were likely to be exceeded by those then still unknown. Only for parts of Europe can most microlepidoptera be identified from readily available manuals, in part as a flow-on of study from enthusiasts resident in the region having earlier documented the more accessible Lepidoptera groups quite comprehensively. Altogether, about 160,000 species of Lepidoptera have been described, from a recently estimated global total of ‘around half a million’: the uncertainty was assessed by Kristensen and his colleagues in a 2007 appraisal that considered a number of estimates made by various methods as ‘There are considerably more than a quarter-million Lepidoptera species, pro bably in the order of magnitude of half a million, but there are not a million – let alone several millions’. Similar uncertainties occur in all the other ‘hyperdiverse’ insect groups, with the suspicion that increased taxonomic resources, particularly if devoted to molecular investigations, will be likely to proliferate massively the numbers of ‘academically convincing discrete entities’ that contribute to overall variety. Whether these are termed ‘species’ or retained in some other recognised form is then somewhat of a subjective or individual decision, reminiscent of the traditional separation of morphological taxonomists as ‘splitters’ or ’lumpers’ depending on their individual propensity for inflating the importance of small variations or ignoring them. As noted commonly from the proliferation of names given to local forms of butterflies (‘a formidable subspecific nomenclature’: Kristensen et al. 2007), many indeed are recognised consistently as ‘diagnosable clusters’. But, recapitulating a theme noted in chap. 4, ‘species’ are difficult to define: as Wilkins (2011) put it ‘… there are n + 1 definitions of “species” in a room of n biologists.’
220
17 Diversity and Its Implications for Understanding Australia’s Insects
Estimating global species richness in almost any insect group, especially a large one, has been described as ‘impossible’ – in part because of this elusiveness of consistently applied categories (both within and across major groups), but also because of lack of knowledge of synonymy. It is not uncommon in taxonomic revisions for large numbers of new species descriptions to be counterbalanced by new synonyms, whereby previously-accepted species are ‘sunk’ to inclusion with others as intraspecific variation is reassessed or simple mistakes are noticed. Increasingly impressive statistical treatments purport to depict aspects of ‘uncertainty’ in our estimates of species numbers; some seem to confuse further rather than elucidate, and may even introduce other sources of uncertainty. The ‘informed guesstimates’ to which we alluded in the preface may in some cases be progressing toward ‘better guesstimates’ but lack of definitive information will for long persist. With most insect faunas, particular groups (orders, families) with few species naturally tend to be more completely known than those that are much richer, and levels of uncertainty both less, and less significant. From a previously-noted example, the uncertainty over whether Australia has 24 or 26 species of Megaloptera is clearly far less significant for basic enumeration and documentation than whether 10,000 or 80,000 beetles remain to be described and evaluated in our fauna. The latter continues to defy effective quantification. As just one example, for global evaluation of the largest family, Curculionidae (or superfamily, Curculionoidea) of beetles in Australia, the weevils, the pithy appraisal that ‘Estimates of the numbers of described species are as variable as those making the estimates’ introducing the multi-volume series on the Australian fauna commencing in 1994 has far wider application. In discussing just one genus of weevils from New Guinea (but extending also into northern Queensland as part of its tropical and subtropical range), Riedel (2010) has noted the implications from recent field studies that this genus, Trigonopterus, is ‘hyperdiverse’. Although only 40 or so species have been named, he suggested that ‘an estimate of more than a thousand for the total number of species … can be considered relatively conservative’. Although Coleoptera have long been presumed to be the richest of all insect orders, increasing numbers of entomologists now consider Hymenoptera to be a leading contendor for this role, with Lepidoptera increasingly assured of third place. As with micromoths within Lepidoptera, taxonomic knowledge in different groups of Hymenoptera is very uneven, with species within many of the families of tiny parasitoid wasps, in particular, the major components of undescribed faunas. The enormous tropical faunas are particularly poorly understood, and prospects for redressing this seem remote – not least because of the small number of interested resident entomologists, combined with the pressing needs of burgeoning human populations taking precedence over such ‘academic’ pursuits; however, it is likely that most species of tropical Lepidoptera and Coleoptera, together with most other insects, will be affected by one or more parasitoid wasps, many of them relatively host-specific. In parallel to their biological variety, parasitoid wasps differ enormously in size and appearance. Some species of Megarhyssa, a genus of ichneumon wasps that includes parasitoids of woodwasps, are up to about 20 cm long, with the ovipositor a prodigious 13 cm of this and enabling the female to drill deeply into wood to reach hosts. At the other size extreme, males of Dicopomorpha, a North American parasitoid of psocopteran eggs, are only 0.13 mm long, and are wingless,
References
221
eyeless, and lack also mouthparts and tarsi. Many groups of parasitoids are extraordinarily complex and their variation difficult to study because of both their small size, and the intricacy of the structural differences and how these are affected by hosts and foods. A comment by a leading authority in the early 1990s that ‘It is probable that at least 75% of parasitic Hymenoptera species have yet to be described, and that many of those currently described are not recognisable’ is almost certainly still valid. One recent estimate of richness of Hymenoptera suggested a range of 300,000 to three million species, and another recent contributor writes ‘my guess is about 1,000,000’, with approximately 115,000 species described. Just for completeness in assessing the ‘big four’ holometabolous orders, an estimated 150,000 species of Diptera have been described, with a recent review simply stating that ‘the actual total number of extant fly species is many times that number’. Many entomologists accept easily that the total global number of existing insect species will prove to be within the range of five to ten million species – far fewer than some of the estimates that flowed from studies made in the 1980s. From initial studies on the vast tropical forest canopy fauna and calculations based on the highly localised distributions and host tree specificity of many insects retrieved, some extrapolations of insect numbers extended beyond 50 million! Enumerating insect species, with varying assumptions in intensity of practical sampling within sites or comparing different sites, has generated numerous scientific papers that collectively convey that (1) there are indeed a lot of insect species; and (2) numerous insect groups show highly characteristic ecological and geographical patterns, including ‘hotspots’ of richness reflecting local endemism and radiations. However, we are in reality still far from knowing whether Earth supports five million, ten million or far more insect species – but, as Adler and Foottit (2009) wrote in introducing a recent book on ‘Insect Biodiversity’: ‘Despite disagreements … biologists can agree on four major points, (1) The world supports a great number of insects. (2) We do not know how many species of insects occupy our planet. (3) The value of insects to humanity is enormous. (4) Too few specialists exist to inventory the world’s entomofauna.’ These global pointers are applicable equally for Australia. In that same book, other commentators (Sperling and Roe 2009) provided a succinct summary comment in ‘For many insects, particularly in temperate zones, a good first draft of the species is available based on standard morphological methods. The undescribed remainder is a combination of rare species and species groups with poor morphological delimitations or legitimately messy species boundaries’. For much of Australia, our first draft still needs considerable augmentation and revision to render it an acceptable representation of our ‘considerable variety’ of insect life. Clarifying the implications from the second sentence quoted still requires vast effort to reach any realistic consensus.
References Adler PH, Foottit RG (2009) Introduction. In: Foottit RG, Adler PH (eds.) Insect biodiversity. Science and society. Wiley-Blackwell, Oxford, pp 1–6 Kristensen NP, Scoble MJ, Karscholt O (2007) Lepidoptera phylogeny and systematics: the state of inventorying moth and butterfly diversity. Zootaxa 1668:699–747
222
17 Diversity and Its Implications for Understanding Australia’s Insects
Riedel A (2010) One of a thousand – a new species of Trigonopterus (Coleoptera, Curculionidae, Cryptorhynchinae) from New Guinea. Zootaxa 2403:59–68 Sperling FAH, Roe AD (2009) Molecular dimensions of insect taxonomy. In: Foottit RG, Adler PH (eds.) Insect biodiversity. Science and society. Wiley-Blackwell, Oxford, pp 397–415 Taylor RW (1983) Descriptive taxonomy: past, present and future. In: Highley E, Taylor RW (eds.) Australian systematic entomology: a bicentenary perspective. CSIRO Publishing, Melbourne, pp 93–134 Wilkins JS (2011) Philosophically speaking, how many species concepts are there? Zootaxa 2765:58–60
Further Reading Upton MS (1997) A rich and diverse fauna. The history of the Australian National Insect Collection 1926–1991. CSIRO Publishing, Collingwood (much background on development of interests in Australian insects, and efforts to collect and interpret the fauna)
Appendix
Australia’s Insects: The Players This Appendix is a very brief introduction to, and characterisation of, the orders of insects that occur in Australia, mainly intended to introduce these major groups to non-entomological readers. It does not in any way replace the much greater information present in any standard entomology text, with the two volume ‘Insects of Australia’ (as the most recent overview, but with much additional progress and documentation since it was revised for publication in 1991) containing massively more background, but is simply to point out some features of their recognition, biology, relationships and diversity, to help consolidate some of the earlier text – in which such formalised information would be fragmenting. The sequence of orders adopted (Table A.1) follows the general evolutionary scheme from ‘more primitive’ to ‘more advanced’ noted in Chap. 2. This sequence thus commences with the Apterygota, and moves through the winged insect orders from the ancient Palaeoptera and the various groups of Neoptera to culminate in the insects with a complete metamorphosis, themselves ranging from the more primitive Neuropteroidea to the most advanced insects, the Hymenoptera. The notes here simply cover the generalised features of each order, and how it may be recognized and separated from others: they do not help to identify insects to levels lower than order – but recognition of orders is the most important, initial, step for any student to overcome, as a prelude to work in to the more detailed literature and richness of an order or group that may especially appeal. The hierarchy of taxonomic levels within an order (Table A.2) illustrates the way in which each lower level is a subset of each higher one. A large order is often separable into several ‘suborders’, each characterised by a unifying set of characters but differentiated from members of other suborders, although all clearly share the features that define the parent order. Diagnostic features may become more detailed and complex as we move toward species level separations – thus, whereas genera of moths or beetles within a unifying family may differ in features of wing venation or body sculpture or pattern, characters separating species (and diagnosing each) may at times be less
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0, © Springer Science+Business Media B.V. 2011
223
224
Appendix
Table A.1 The orders of insects found in Australia, indicating the major groupings Group Order Common name Apterygota Archaeognatha Bristletails Zygentoma Silverfish Pterygota Palaeoptera Neoptera Polyneoptera
Paraneoptera
Ephemeroptera Odonata
Mayflies Dragonflies, damselflies
Blattodea Isoptera Mantodea Dermaptera Orthoptera Phasmatodea Embioptera [Zoraptera]* Plecoptera Hemiptera Thysanoptera Phthiraptera Psocoptera
Cockroaches Termites Praying mantids Earwigs Grasshoppers, crickets, katydids (etc) Stick and leaf insects Webspinners Zorapterans Stoneflies Bugs, scale insects, cicadas, Leafhoppers, psyllids (etc) Thrips Parasitic lice Barklice, booklice
Megaloptera Neuroptera Coleoptera Strepsiptera
Alderflies, dobsonflies Lacewings, antlions, mantisflies (etc) Beetles Strepsipterans
Mecoptera Diptera Siphonaptera Trichoptera Lepidoptera
Scorpionflies Flies Fleas Caddisflies Moths, butterflies
Hymenoptera
Ants, bees, wasps
Oligoneoptera Neuropteroidea
Panorpoidea
Hymenopteroidea *See text
obvious – such as small details of genitalic structures visible only after careful dissection and microscopical examination. There is no universal rule for which characters may be important, and different orders may be separated into their categories on different arrays of structural features. With practice, most orders will become identifiable quickly, and with the naked eye – although a low power stereoscopic microscope is an invaluable aid in checking features such as mouthpart form (sucking or chewing, for example) and numbers of tarsal segments of the legs. However, brief the text, it is almost inevitable that some technical vocabulary – the short hand of science – will creep in. Some of these terms, almost all descriptors of structural states, are explained in Chap. 1, but others are explained in simple language in parentheses as they are used.
Appendix
225
Table A.2 Classification hierarchy of insects: the major categories that may apply in formal taxonomy. Not all may be needed in any particular case, and different specialists may query the formal names or status for some rankings between order and family (Division, Series, Superfamily, Tribe) and boundaries between subfamilies may also be uncertain. The example given, for the Australian bushfly, indicates the principle involved, whereby each lower category incorporates all features used to define those larger groups higher in the sequence Order Suborder
Division Series
Superfamily
Family Subfamily Genus Species
Diptera (all the true flies, including mosquitoes, craneflies, marchflies, hoverflies, robberflies, and many others). Brachycera (flies with short antennae, excluding the other suborder Nematocera that have long antennae and include mosquitoes, craneflies and their allies). Cyclorrhapha (a group of Brachycera in which the pupa is formed within the last larval skin, rather than this being discarded). Schizophora (those Cyclorrhapha with a ptilinal suture – a line on the face marking the position of an eversible ‘bladder’ used to open the pupal case for the adult to emerge). Muscoidea (Schizophora that have a second lobe, the calypter, near the posterior base of the wing; one of about 12 superfamilies in this series in Australia). Muscidae (one of about 12 families within the Australian Muscoidea). Muscinae (one of about seven subfamilies of Muscidae). Musca (the genus containing houseflies and their close allies). vetustissima (the individual, unique, name applied to the Australian bushfly).
The various orders fall into two categories in a treatment such as this, reflecting their richness and ecological variety. Most of the ‘smaller’ orders are treated especially briefly, reflecting relative uniformity of their structure and biology, whereby more detailed appraisal is not needed here. In contrast the ‘big five’ – the Hemiptera and the four megadiverse holometabolous orders (Coleoptera, Diptera, Lepidoptera, Hymenoptera) are much more varied and somewhat longer text is needed to convey even a flavour of their structural and ecological variety in coming to dominate so many different biotopes and ecological roles.
Apterygota The primitive wingless insects, closest to the ancestral insect form, are represented by the two living orders which are very similar in general appearance. They are most familiar as the ‘silverfish’, some of which are commonly found in households, whilst others occur in leaf litter, under bark, in rock crevices or within the nests of ants or termites – collectively a range of ‘cryptic habitats’ in which these predominantly nocturnal animals live. The general body form is slender, tapered toward the posterior, and terminating in three long sensory filaments, the lateral cerci and the median dorsal filament. The body of many is covered in flattened scales, and the two orders are separable by their body shape and the form of the compound eyes.
226
Appendix
Order: Archaeognatha (sometimes called Microcoryphia, Bristletails) have a more cylindrical body form than the flattened form of silverfish (Order: Zygentoma, commonly known as Thysanura but that name was applied as a collective for both the included orders, so may cause confusion), and the thorax is strongly ‘arched’. The most conspicuous differentiating feature, however, is that bristletails have large compound eyes, meeting across the top of the head, whereas those of silverfish are small and clearly separated and, in some taxa, absent. Neither order is very diverse. Only about ten Australian species of Archaeognatha have been described, but the fauna is not regarded as well-known and some others may exist. The largest bristletails are up to about 18 mm long, but many others attain scarcely half this size. All belong to one family, Meinertellidae, which has strong southern hemisphere affinities. Silverfish are more diverse, and most biological knowledge is of the few widespread domestic species within the family Lepismatidae, and which are up to about 1 cm long. Other, native, species are placed in the Nicoletiidae. They are renowned as fast runners, as people who have attempted to accost them in houses can attest and, in general, Apterygota have attracted the attention of few specialists over the years. They feed on vegetable and other debris, and there are records of them eating their own cast skins and eggs. Adults can live up to several years, and continue to moult and reproduce throughout that time.
Pterygota All other insects are winged, or have been derived from winged ancestors to become secondarily wingless in response to specialised ways of life. The most primitive of these are the ancient Palaeoptera, which cannot flex their wings, and all others are allotted to the more advanced Neoptera, to which the great majority of insects belong.
Palaeoptera The two orders, very different in appearance, are the remnants of an ancient group that was much more widespread as the earliest winged insects so far found in the fossil record, in the Carboniferous period. Both are associated strongly with freshwater environments and have highly modified aquatic larvae that differ much from the appearance of the corresponding adults and pass through many instars during a long larval life. Adults have very complex wing venation. Order: Ephemeroptera (Mayflies) are the only insects with two fully-winged instars (stages), the ‘subadult’ and true adult stages. Adults, however, are very weak flyers, rarely moving far from their waterside habitats and, as the order name implies, are short-lived (ephemeral). Many live for only a day or so. Antennae are short and
Appendix
227
slender, and the compound eyes often very large and in some taxa (particularly in males) meeting across the top of the head. Mayfly adults do not feed, and the mouthparts are rudimentary. The slender rather soft body ends in three long filaments (the lateral cerci and central dorsal filament) which are sometimes longer than the body itself. The wings are basically triangular in shape, and the hind wing is much smaller than the fore wing – it may even be absent or represented by only a tiny lobe. Wing venation is usually complex, with many cross veins, but becomes simpler in some families. As a feature of the Palaeoptera, the wings cannot be folded along the body, and are held dorsally above the body when not in use. Legs may be long and the basic number of five tarsal segments is sometimes reduced. Males in some species have the joint between the tibia and tarsus of the fore leg reversible, and this aids in grasping a female in a flying swarm as a prelude to mating. Larvae are also diagnosed easily, with the ordinal combination of three apical filaments and several (4–7) pairs of lateral abdominal gills characteristic. Larvae have chewing mouthparts and feed mainly on vegetation and debris, although a few are specialised predators. The number and form of the gills is valuable in classification, and also indicates the kind of habitat, in that still water with low oxygen tensions may necessitate large gills that increase the surface areas for gas exchange, whereas fast-flowing water may have much higher oxygen tensions and also potential to rip off large gills from larvae, or sweep them downstream. Most larvae are free-living and found on vegetation or stones, but some burrow in mud or other bottom sediments. Classification of adult mayflies is largely on wing venation, and of larvae on form of gills and legs. Mayflies can occur in very large numbers, and are noticed especially when adults swarm in large numbers – a behaviour feature rendering them more conspicuous for mating. They otherwise intrude little on humans, but are important in fresh water food webs, for example as fish food; particular mayflies are the models for particular ‘flies’ used as lures in angling. Nine families of Ephemeroptera are recorded from Australia, and many taxa are concentrated in highland waters of the south east, as Gondwanan elements. Some northern elements are also found, and recent research has increased the fauna considerably from the 84 species reported in 1991. Order: Odonata (Damselflies, Dragonflies) differ greatly in appearance from mayflies, and most of the structural specialisations of these robust insects are associated with both larvae and adults being predators. Damselflies and dragonflies represent different suborders, although the common name ‘dragonflies’ is often used to encompass all of the members of the Odonata. The order name refers to the toothed mandibles of the strong chewing mouthparts of most members of the order. The generally slender and delicate-looking damselflies are referred to the suborder Zygoptera, and the stronger-looking dragonflies are in the Epiproctophora, a name now used in preference to the formerly widespread ‘Anisoptera’. Adult Odonata are essentially aerial hunters, feeding on small insects that they capture whilst flying. They are mainly diurnal, and hunt by sight, flying fast with strong wings, and the tapered elongate body aiding aerial manouvreability. The compound eyes are very large, and occupy most of the surface of the head, and
228
Appendix
antennae are reduced to tiny inconspicuous filaments. The short strongly spined legs with hooked claws are oriented forward from the more general vertical position by skewing of the thorax, which enables them to participate in prey capture, with the angled thorax also serving to reduce the chances of wings becoming entangled with actively-resisting prey. Tarsi have only three segments. The enlarged pterothorax also reflects that Odonata are very strong flyers. The large strong wings have very dense and complex venation, difficult to interpret and used extensively in classification. They have a thickened patch, the pterostigma (literally ‘wing-spot’) on the leading edge near the wing apex and this facilitates aerodynamic maintenance. In Zygoptera the two pairs of wings are both slender, similar in form, and at rest are held vertically above the insect’s body. Epiproctophora hold the wings extended horizontally each side of the body, and the hind wing is considerably broader than the fore wing. Odonata lack the long terminal filaments found in mayflies, but may have short ‘claspers’ in that position. Aquatic larvae are also distinctive for the two suborders, and are separated most simply on the form of their respiratory structures. Larval Zygoptera have three ‘caudal lamellae’ (plate-like gills, at the end of the abdomen) which range in different families from being small slender flattened structures to more elaborate globose appendages. They may be conspicuously patterned or banded, possibly aiding crypsis but are also used in territorial display or other elaborate signalling behaviour rituals between larvae, being raised and ‘waved’ to render them conspicuous. True dragonflies (such as ‘mudeye’ larvae) lack external gill structures, and gas exchange takes place internally, across the wall of the rectum. All larvae, however, share a unique adaptation of the mouthparts, wholly diagnostic for Odonata and allowing the insects to be effective ‘ambush predators’ which lie in wait for prey, and detect these largely visually through the large compound eyes. Prey organisms include insects, tadpoles, small fish, or virtually any small animals that venture within capture range. Beneath the head, which is more horizontal in position than in many other insects, bringing the mouthparts forward, the labium is elongated and hinged to form a ‘labial mask’ in front of the head. This structure can be extended rapidly to grasp or impale prey some distance in front of the resting insect, after which the labium is retracted to bring prey to the mandibles. Some dragonflies can fly over long distances, with some being well-known migrants. Many are long-lived, up to at least several months, and it is not unusual even in normal behaviour for them to stray up to many kilometres from water to mature, returning to find mates and breed. Elaborate territorial behaviour is not uncommon, as part of a wide spectrum of behavioural interactions that can be observed easily in these diurnal insects. Odonata have a unique mode of mating, which involves males transferring their sperm to a specialised pouch on the underside of the anterior part of the abdomen, whence it is transferred to the female whilst she is held by the male in a ‘tandem’ or ‘wheel’ position by him grasping the back of her head or prothorax with his posterior claspers. Many Australian Odonata have Gondwanan relationships, whilst others originate from more recent northern lineages. Representatives of 30 families (12 of Zygoptera, 18 of Epiproctophora) occur in the fauna of about 325 described species, and the
Appendix
229
Australian species can be regarded as reasonably well known, due to recent handbooks and keys to both adults and larvae. In size, they range from some tiny Zygoptera (the smallest, Hemiphlebia mirabilis [p. 42], is only about 22 mm in fore wing length) to some very large dragonflies (Petalura can have tip-to-tip wingspan exceeding 16 cm!). Classification of adults to family is largely on wing venation, and of larvae by the form of caudal lamellae (Zygoptera) and general body form.
Neoptera All the remaining orders fall into this vast group, that encompasses an enormous variety of different taxa united by having (or being derived from ancestors that had) wings that can be flexed, folded along the body in some way. The orders fall into several groups which, although not formal taxonomic entities, are useful in helping to reflect relationships and sets of common characters (Table A.1). Thus, the group known as Polyneoptera are the more primitive orders, many of them ancient, closest to the basic neopterous insect type, and have chewing mouthparts, and an incomplete metamorphosis. Several of the ten or so orders included intergrade in various ways, and specialists continue to debate their relationships – the sequence given here is relatively ‘traditional’ in format and divisions. Some of the orders included in this series are small and poorly known. The second grouping contains only four conventionallyrecognised orders. Paraneoptera also have incomplete metamorphosis, but are founded in the development of sucking mouthparts, and dominated by the sap-sucking plant bugs, Hemiptera. The four orders included form a discrete group. The third group is by far the largest of the three. Oligoneoptera are those insects with a complete metamorphosis, and are united mainly by this major feature as they are otherwise very varied in structure and biology. These three groups are regarded widely as reflecting a sequence from the most primitive to the most advanced neopterous insects. Polyneoptera are sometimes referred to as the ‘Blattoid/Orthopteroid insects’ or the wider ‘Blattoid/Plecopteroid insects’, from the names of two orders (Blattodea and Orthoptera or Blattodea and Plecoptera) that delineate major series, and Paraneoptera are termed the ‘hemipteroid insects’, again from the predominant order included in that group.
Polyneoptera Polyneoptera have prominent cerci, and the fore wing is often considerably tougher in texture than the hind wing, and termed a ‘tegmen’. In most groups, the fore wing overlies the hind wing at rest and protects it, and the hind wing is pleated, fanlike, so that when spread it is considerably larger than the fore wing. As implied above, the constituent orders can be grouped into several series, with the aquatic stoneflies (Plecoptera) rather isolated from the remainder, all of which are terrestrial.
230
Appendix
Fig. A.1 Many cockroaches (Blattodea) are apterous and occur in ‘cryptic habitats’ such as under bark or in rotting wood, to which access may be facilitated by the strong pronotal shield and the head being protected beneath this; many species are nocturnal: Polyzosteira sp.
The first three orders in the sequence below illustrate some of the problems of assessing relationships between groups that seem to be very different but share structural features, particularly some intricate anatomical features that assert their close relationship. They are commonly grouped together as Dictyoptera to indicate this affinity and, occasionally, this name is used as a single order name to encompass these. Order: Blattodea (Cockroaches) are regarded as one of the most primitive ‘basal’ orders of Neoptera, as an ancient group which have changed little in appearance from early fossil forms. They are characteristically hard-bodied, often glossy and dorso-ventrally flattened, so very characteristic in shape, and fitting them well to crawl under bark, into crevices in rocks and soil and occupy similar ‘cryptic habitats’, where they hide during the day. Most are nocturnal. Other distinctive features are the long filamentous antennae, and the head being ‘hypognathous’ (tilted forward and scarcely visible from the dorsal surface) and hidden beneath the enlarged pronotum that forms a ‘shield’, perhaps also aiding in moving into confined spaces. Cockroaches have long spined legs with five tarsal segments, and are renowned as fast runners. In common with related orders, wing venation can be very complex, with numerous crossveins and the hind wing pleated. Many cockroaches, however, are secondarily wingless (Fig. A.1) – sometimes only in one sex, most commonly females – and rely on running for movement. Cerci are usually conspicuous, although sometimes shortened. Most cockroaches are omnivorous, although a few specialise on dead wood as food. These insects are most familiar to many people from the few cosmopolitan species found in houses and food stores, but most occur only in natural habitats such as under bark or rotting wood, in leaf litter or in caves. A few species produce live
Appendix
231
larvae, rather than laying eggs, but egg-laying species enclose the eggs within a specially constructed protective case (the ootheca) which is carried around by the parent female, projecting from the tip of her abdomen. Some cockroaches are strongly gregarious. Other than for the tiny Nocticolidae (around 3 mm long; the four Australian species occur in caves or termite nests), most are reasonably large insects ranging from one to several centimetres in length, with the large soil-burrowing Macropanesthia (the largest Australian cockroaches, popular as long-lived and low maintenance pets) exceeding 6 cm. Blattodea is a relatively small order, with around 4,000 species, of which some 440 occur in Australia. Classification is still debated, but two major divisions are generally acknowledged. Blattoidea contains the large family Blattidae, and Blaberoidea, the other four families found in Australia. Families are separated on a combination of wing and genitalic features and, other than for the few cosmopolitan ‘pest’ species, most Australian cockroaches are endemic, perhaps with their strongest affinities with those of the Oriental region. Order: Isoptera (Termites, archaically termed ‘white ants’) are very closely related to cockroaches, and some entomologists believe that they are simply a specialised group of Blattodea related to the wood-feeding cockroaches, and do not merit recognition as a separate order. Termites look very different from typical cockroaches, but many of the differences may be attributed in some way to the universal social existence (p. 104), whereby different morphological castes within a species differ substantially in appearance, and role. Termites feed on cellulose, most commonly as wood or dead grasses and other vegetation, and nest either on or in the ground, or on or in wood. Whether a distinct order or not, they appear to have been derived relatively recently from a cockroach-like ancestor. Termites have a more cylindrical body form than cockroaches, and most are softbodied and pale. Antennae are much shorter than in cockroaches and their segments are beadlike (moniliform); eyes are reduced, absent in non-reproductive castes, and these castes are invariably wingless. Legs are simple, with four tarsal segments and the two pairs of wings of reproductives (which are shed by a basal transverse suture after initial flight) are similar in shape, except in the very primitive Mastotermes (from northern Australia) in which the hind wing is broadened, as in cockroaches. Termites have massive ecological and economic importance, despite there being only a few thousand species, of which around 400 are found in Australia. They are difficult to identify, and the five families are differentiated mainly from characters of the reproductive or soldier castes, as many workers are very similar in appearance. Probably, around 40% of Australian species have not yet been described. Order: Mantodea (Mantids or ‘Praying mantids’), very different in appearance from either of the foregoing orders, nonetheless appear to be very closely related to cockroaches, and are allied with them in some classifications. Mantids are large solitary insects, and ambush predators that feed on a variety of accessible prey captured by the specialised grasping fore legs. The legs are held in front of the insect when it is at rest, and their position has led to the epithet ‘praying’ in the common name. Many mantids are very cryptic, resembling in colour and shape the vegetation on which they rest, and their further predatory specialisations include very large
232
Appendix
compound eyes, often separated widely across the head, strong jaws with the front of the head narrowed – aiding insertion of mouthparts into prey – and wider structural adaptations that result in increased reach, such as the fore legs being at the anterior of an elongated prothorax, and the more posterior legs being long and stiltlike and from which the mantid can lunge forward. Mantids can be fully winged (with the very broad hind wing characteristic of this group of orders), short winged or entirely wingless, sometimes with differences between the sexes. In addition to colour pattern that often resembles vegetation (including some flamboyantly coloured species that hunt whilst resting on flowers), the pronotum may have lateral flanges and the fore wing become elongate or rounded as features that also aid camouflage. The generalised predatory habit includes cannibalism, and some mantids are notorious for so-called ‘sexual cannibalism’ whereby the (usually larger) female devours the male whilst copulation proceeds. Eggs are laid in an ootheca but this, unlike that of cockroaches, is deposited on the substrate and formed from a frothy proteinaceous secretion that hardens on exposure to air. The families of Mantodea are distinguished on features of fore legs, wing venation and general body form. About 2,000 species are known worldwide, and about 200 occur in Australia. By far the richest family in the fauna is Mantidae, containing about three quarters of Australia’s mantids. Two other families also occur, whilst five others are absent from Australia. Amorphoscelidae includes small, grounddwelling or bark-frequenting species, some only about 2 cm long, and the Hymenopteridae are a predominantly tropical family with one species reported from Queensland. In general, rather few mantids occur in the south of the continent, and the order is predominantly a tropical group. Order: Dermaptera (Earwigs). These distinctive insects are unlikely to be confused with any others (except, perhaps and rarely, with some short-winged beetles, from which they are distinguished by their conspicuous cerci, never found amongst beetles), and may be related to the above three orders. They are elongate, slender and hard-bodied, with the head prognathous (so, the jaws oriented forward), and antennae and compound eyes present. Legs are simple, with three tarsal segments, and the pronotum is large. Wings may or may not be present. If so, the fore wing is greatly reduced to a small, veinless, toughened cover (‘elytron’) under which the large fanlike hind wing can be folded and protected when the insect is at rest. Cerci are very conspicuous and often curved or forcepate (‘tweezer-like’); they often differ in development between the sexes. Typical earwigs range from about 0.5 to 5 cm in length. Most earwigs, as cockroaches, are part of the ‘cryptozoa’, those animals not noticed widely because they dwell under bark, in leaf litter and in similar cryptic habitats, and largely nocturnal in habit. They are predominantly omnivorous, but two specialised offshoots (neither found in Australia) have become ectoparasites on bats in Malaysia and rats in West Africa. In general, the earwig most familiar to most Australians is an introduced species, the European earwig (Forficula auricularia), which is often very common and can become a pest by eating seedlings in home gardens. Females lay eggs in batches and may brood these until they hatch, in a form of
Appendix
233
‘parental care’. All typical earwigs belong to the same suborder (Forficulina), with eight families commonly recognized. Seven of these occur in Australia, but only about 65 Australian earwig species have been described, of a world fauna of around 2,000 species. Most Australian species seem to have northern affinities, but most are endemic. Order: Orthoptera (Grasshoppers, Crickets, Katydids and their allies) is by far the largest order of Polyneoptera, and also the most variable with – as the above common names indicate – several rather distinctive major kinds of insect included, and associated with a wide variety of habitats. A few are major pests, as locusts, others are significant cultural icons, and most are not usually noticed. Most of the pest species are so because of their depredations on plants, and most Orthoptera are indeed herbivores, whilst others are scavengers or predators. Orthoptera have strong chewing mouthparts, directed ventrally and with the mandibles often asymmetrical. The body is usually elongated, and the prothorax characteristically enlarged so that the sides extend ventrally and cover the pleural region. Hind legs are usually enlarged, with the femur and tibia used in jumping, and the more anterior legs relatively small and slender. However, the fore legs can become modified – in mole crickets, for example, they are broadened for burrowing into soil. Tarsi have 1–4 segments in different groups. Orthoptera may be fully winged (with the hind wing much broader than the fore wing), short-winged or wingless, and the fore wing of males may be modified for sound production (stridulation) as a major behavioural feature in courtship of these insects (p. 96). Sound receptor organs may be present on the fore tibiae or on the sides of the anterior abdominal segment. Many of these insects are highly cryptic and well-camouflaged to live unnoticed on vegetation, or on the ground. More specialised habitats include caves and burrowing in sand or soil. Two major suborders, both diverse, are recognized. Caelifera includes the typical grasshoppers, locusts and related forms, which are characterised by having short antennae (much shorter than the body and with fewer than 30 segments) and lacking a large ovipositor in the female. Ensifera (katydids – sometimes called long-horned grasshoppers – and crickets) have much longer antennae and females, a long bladelike ovipositor. Both groups have several major subdivisions. More than 3,000 Australian species are known, some of them highly unusual insects (p. 3) and taxonomically very complex to interpret. The largest families are the Acrididae (grasshoppers), Tettigoniidae (katydids) and Gryllidae (crickets), all of which have very high levels of endemism. Order: Phasmatodea (Stick insects, Leaf insects) gain their common names from their highly cryptic body forms that closely resemble the vegetation on which they, as herbivores, pass their lives. Closely related to Orthoptera, phasmatids have sometimes been regarded simply as unusual members of that order. Males are often far smaller than the corresponding females. Wings may be present or absent, and winged forms generally have the fore wing shorter and hardened, and the hind wing large and fanlike. All parts of the body may have protuberances, spines or lobes that increase crypsis, whilst the hind wings are sometimes brightly coloured and used as a startle display.
234
Appendix
Fig. A.2 Phasmatodea: a female of Macleay’s spectre, Extatosoma tiaratum
Almost all the slightly over 100 Australian species are stick insects, some amongst the largest of all insects and placed formally within the suborder Anareolatae. The three leaf insects are attributed to the other suborder, Areolatae, and the two suborders are differentiated by the rather inconspicuous feature of absence or presence of a triangular area of cuticle at the apex of the tibia of the middle and hind legs. Leaf insects are wholly tropical, and most stick insects in Australia, likewise, do not extend to the south, although a few southern species can become abundant and cause damaging outbreaks through devouring foliage of forest trees (p. 136). Many phasmatids are regarded as rare insects, but for many this may more reflect difficulty of detection than actual scarcity. Phasmatid eggs are dropped to the ground individually – they are large and seed-like, and may take months to years to hatch, after which the young larvae climb onto vegetation to feed. Some species, such as MacLeay’s spectre (Extatosoma tiaratum) are popular as pets (Fig. A.2). Order: Embioptera (Webspinners). These small elongate insects (few are more than about 1 cm long) are rarely seen other than by entomologists searching for them, and are not well known in Australia. They most typically occur on the surface of bark or rocks, where they construct silken tunnels and retreats (hence the common name) spun from silk glands in the enlarged basal tarsal segment of the fore legs. Legs themselves are short, with three tarsal segments, but the hind femur is enlarged for backward movement. Unusually amongst the orthopteroids, both wings are slender and the venation quite reduced and simple. Female webspinners are always wingless, and males either winged or wingless. Cerci are reduced to only two segments and the other notable feature of the abdomen is that the male genitalic structures are markedly asymmetrical. Embioptera are almost all herbivores, many grazing on lichens and general debris. Webspinners are most abundant in the tropics and few species occur in the south of Australia. About 65 species are known from Australia but some form ‘species
Appendix
235
complexes’ so that the real richness may prove to be considerably higher. Other than for a very few introduced species, most members of the three families (separable mainly on features of the adult males) are endemic. Order: Zoraptera (‘Zorapterans’) was noted earlier (p. 35) as only ‘marginally Australian’, but might still eventually be found in the tropical north of the Australian mainland. These tiny insects (less than 3 mm long) have rather uncertain relationships and, whereas some entomologists consider them close to the cockroaches, others ally them with the earwigs or even attribute them to the next series, Paraneoptera. Only about 30 species are known worldwide, most of them from the tropics, where they occur mainly under bark and in similar cryptic habitats. Wings, when present, have rather simple venation, and the short cerci are unsegmented. Legs have only two tarsal segments, and the hind femur is often broadened and with strong spines along the inner (ventral) margin. Zoraptera are presumed to be general omnivorous scavengers. Although arthropod remains have occasionally been found in their gut contents, it is not clear whether these originate from predation or from casual ingestion of material on the substrate. Order: Plecoptera (Stoneflies) is very isolated amongst this series of orders, and is often referred to an independent group within the Polyneoptera. Relationships continue to be debated, but there is some consensus that stoneflies may be a sistergroup to all other orders included, and so lack close affinity with any of these. They are the only primarily aquatic members of the series: larvae of all are aquatic, and weak-flying adults most commonly occur only near water. Some are strong runners, and hide on and amongst vegetation and debris. The body, particularly the abdomen, is soft, and many species are short-winged or wingless. Adults graze on algae, lichens, rotting wood and similar substrates, but larvae of different families have a wider dietary range, as detritivores, omnivores or predators. Australian Plecoptera, about 200 species, all belong to southern hemisphere families, and all species appear to be endemic, together with almost all genera. Two genera are shared with New Zealand, and stoneflies seem to be far more diverse in the temperate regions than in the tropics. Classification of adults to family is largely on wing features, and of larvae on the form and position of their gills. More formally, two suborders are recognised (the northern Arctoperlaria and the southern Antarctoperlaria) but, because these are separated on characters of fore leg muscles, they are often ignored in practical identification keys. Only one of the four families in Australia (the Notonemouridae) is referred to the Arctoperlaria.
Paraneoptera Paraneoptera have an incomplete metamorphosis and the number of larval stages is usually far less than in many Polyneoptera. They have the cerci strongly reduced, either to a small sensory tubercle, or wholly absent. The fore and hind wing are both typically membranous, although the fore wing is toughened in some Hemiptera
236
Appendix
(the true plant bugs, Heteroptera). Mouthparts are usually suctorial, more rarely rasping or chewing, but with the major development of the series being through liquid-feeding, initially on plant sap. This group includes four orders, although the separation between two of these is very unclear. The common name of ‘hemipteroid orders’ alludes to the Hemiptera, by far the largest order in the series, to which the Thysanoptera (Thrips) are allied. The other order-pair, the parasitic lice (Phthiraptera) and barklice (Psocoptera) are now often thought of as a single level ‘Psocodea’ to include these, and reflecting that conventional Psocoptera may actually have more than one origin. Order: Hemiptera (Bugs, Scale insects, Leafhoppers, Cicadas and related insects) are a very diverse group, unified by having sucking mouthparts as long piercing stylets, through which liquid is ingested, with most species feeding on plant sap. A few groups (such as male scale insects) do not feed, and their mouthparts are rudimentary. Antennae are relatively short, with no more than ten segments, commonly only four or five. Cerci are absent, and legs have three or fewer tarsal segments. However, within this characterisation, Hemiptera are structurally very diverse, so that each of the major groups – including those listed above – has a very characteristic appearance. Wings may be present or absent, but the hind wing is not larger than the fore wing and, indeed, is often considerably smaller. Wing venation may be very reduced, with few crossveins, and some taxa are characteristically apterous. The order is divided conventionally into two suborders on wing features and in the past these have sometimes been treated as distinct orders. In Homoptera (such as cicadas and leafhoppers), the fore and hind wings are both membranous and differ little, if at all, in texture; the fore wing venation is distinct. Heteroptera (plant bugs) have the fore wing thickened over the basal region, and a smaller membranous apical region in which the veins are distinct. The hind wing is membranous, so there is considerable contrast between the two pairs of wings and within regions of the fore wing. The position of the mouthparts also differs in the two groups, and is thus useful for differentiating wingless forms! Homoptera have mouthparts arising from near the posterior of the head, and those of Heteroptera are more anterior. Homoptera are assumed to be the more primitive, ancestral group and comprise a range of different insects, some of which have been considered as separate suborders. Indeed, some authorities dismiss ‘Homoptera’ in favour of three such groups that encompass its former content as Coleorrhyncha, Sternorrhyncha and Auchenorrhyncha. In some interpretations, Coleorrhyncha, with an unusual combination of structural features, are treated as a sister-group to the Heteroptera and separated clearly from other groups within Homoptera. Whatever arrangement is used is not critical here, other than in reflecting diversity of opinion and the variety within the group. All Homoptera are terrestrial, whereas some Heteroptera, through the avenue of becoming predators, have become aquatic and their body and legs highly modified for life in freshwater environments, as streamlined water boatmen and water bugs. A few others have adopted the ‘half way house’ of living on the water surface, as pondskaters and water striders, and where they feed on drifting corpses of insects and other prey. Some terrestrial Heteroptera are also predators. Some of the assassin bugs have raptorial fore legs resembling those of mantids in
Appendix
237
form and function. A few, such as bedbugs and other assassin bugs, are blood feeders, and some are important vectors of diseases, significant in livestock and human health. Homoptera are herbivores, but some are also important as causing direct damage to plant crops, and as vectors of virus diseases. The three subdivisions of Homoptera mentioned above include the various bugs as follows: Coleorrhyncha contains only the moss bugs (p. 168), as particularly archaic forms; Sternorrhyncha (with two tarsal segments) include plant lice (psyllids, lerp insects), whitefly, aphids, scale insects and mealy bugs, some of which are highly modified wingless forms bearing little resemblance to ‘normal’ winged Hemiptera; and Auchenorrhyncha (with three tarsal segments) incorporates the planthoppers, froghoppers, leafhoppers and cicadas. The Australian Hemiptera, on one recent estimate, includes 11,580 species, of which fewer than 4,500 had then (2003) been described. Of these, most are Homoptera, and some of the radiations of these bugs on native vegetation are indeed complex and diverse. Psyllids were noted earlier (p. 120); the largely endemic Eurymelidae are another striking example, as are the gall-forming coccoids. Almost all groups other than aphids show high levels of endemism. Many different faunal elements are found – some are clearly ancient Gondwanan groups, whilst other lineages have strong Oriental or Pacific affinities. Order: Thysanoptera (Thrips) are small slender insects, related to Hemiptera, but with highly unusual mouthparts. These are asymmetrical, and regarded as ‘raspingpiercing’ with the maxillae and left mandible modified to constitute piercing stylets sited on a basal rostrum. Thrips feed by rasping or piercing the food and sucking up liquids that exude. Many are foliage feeders and some are gall-formers (p. 123); others are predators. Antennae have 4–9 segments. Legs are short, and tarsi have one or two segments and end in an eversible adhesive bladder rather than claws. Wings, when present, are narrow with few veins and commonly fringed (particularly along the hind margin) with long hairs. The two suborders are separated on the form of the abdomen. The apex is conical in Terebrantia and extended into a narrow tube in Tubulifera. The latter includes some of the largest of all thrips, such as Idolothrips found under bark of eucalypts, and which can reach up to about 15 mm long; most thrips are only around 2–3 mm. Thrips can become very abundant on vegetation, in leaf litter or other cryptic habitats such as under bark, and all are terrestrial. They have an unusual developmental pattern that in some ways parallels that of a complete metamorphosis, in that two active larval instars are followed by non-feeding stages, including a ‘prepupal’ and one or two ‘pupal’ instars during which wing buds appear and that in some species occur during inclement parts of the year as resting stages that may incorporate diapause. Recent revisionary studies on Australia’s thrips (p. 123) have markedy increased the number of species known, and the 500 or so species already described may constitute only about a quarter of the total fauna. The two following orders are placed together in a ‘superorder’ Psocodea, in which some relationships are still unclear. They are treated here in the conventional sense, but the formal problems arise because of the possible separate origins of
238
Appendix
some groups of Psocoptera, with the suborder including the booklice probably independently derived from the Phthiraptera. Unusual characters of the hypopharynx (the internal part of the mouthparts) imply close relationships between the notional orders. Order: Phthiraptera (Lice) are highly specialised ectoparasites of birds and mammals, with all stages passed on the hosts, and many of their structural features related to this way of life. They are small (a few mm in length), wingless, the body is broad and dorsoventrally flattened, and often has strong backwardly-directed bristles. Legs are short, with tarsi having one or two segments and are modified for grasping the hairs or feather plumules of their host, some lice having greatly enlarged claws that oppose on an extended tibial process to create a very firm grip. Eyes are reduced or absent, and antennae short with three to five segments. Cerci are absent. Most lice are highly host-specific, and have obligatory associations with those hosts. Relationships within the order are still somewhat confusing, but the two major groups recognised are the Mallophaga (chewing lice, mostly feeding on feathers or body surface debris on the host, and with broad heads and tarsi with two claws) and the Anoplura (sucking lice, blood feeeders that pierce the host with pointed mouthparts, and are characterised by the head being narrow and tarsi having a single claw). All orders of birds and most mammals (except monotremes and bats) host Phthiraptera. As might be expected, the Australian louse fauna is largely endemic, reflecting the endemism of the hosts, but the cosmopolitan lice infesting people and livestock are also significant. Perhaps 650 species of lice occur in Australia, and an estimated one third of these are still undescribed. Eggs of lice, often termed ‘nits’, are glued individually to fur or feathers and individual lice are transferred between hosts only during contacts such as roosting or mating. Host species specificity is augmented by host site specificity so that, for example, a bird may support different species of lice on different parts of the body, and human headlice and pubic lice are site-characteristic and so do not normally occur together. Lice can occur in very large numbers on a host. Some are important vectors of diseases (the spread of typhus by human life is a notorious example), and have considerable importance in human and livestock health. Weakening of hosts by mass attack, even without specific disease involvement, is also of concern. Order: Psocoptera (Barklice, Booklice), despite the ‘louse’ epithet, are not parasites but freeliving, with most of them found on vegetation, where they graze on microflora (such as lichens and encrusting algae) and general organic debris on the surfaces of bark or foliage, some with marked habitat specificity reflecting dietary specialisations. A few species are domestic pests, feeding on flour or grain (or fungal contaminants of these) in storage areas; the common name ‘booklice’ for some of these tiny (2–3 mm) wingless forms arose from their historical association with old books bound with pastes based on flour and later infested with fungi. Outdoor barklice are also small, most commonly in the range of 3–5 mm long. Antennae of Psocoptera are long, legs are generally slender, and tarsi have two or three segments. Compound eyes are large, and, although some species are wingless, most have two pairs of membranous wings. Venation is rather simple, but
Appendix
239
characteristic and valuable in separating families and many genera. The hind wing is usually considerably smaller than the fore wing. Mouthparts are unusual and highly characteristic, with the lacinia of the maxillae modified to form a chisellike ‘pick’. Globally, about 5,000 species have been described, distributed amongst more than 40 Recent families. The order is global in distribution, but rather poorly documented, and the Australian fauna comprises, perhaps, around 300 species. Many are endemic, with both southern and northern relationships evident, and a few species (booklice) are cosmopolitan, having been distributed widely in trade. By far the largest of the three widely recognised suborders is the advanced Psocomorpha (in which antennae never have more than 13 segments and the fore wing usually has a distinct, usually at least slightly sclerotised, pterostigma), which is clearly discrete, whereas relationships and origins within the other two (Trogiomorpha and Troctomorpha, most commonly with more antennal segments and a pterostigma, if present, membranous and never more sclerotised than the remainder of the fore wing) are far less clear, as implied above.
Oligoneoptera Oligoneoptera, by far the majority of recent insects, exhibit a massive variety of life forms and structure amongst the ten orders found in Australia. Although some relationships between orders are reasonably clear, others are still controversial, and the single unifying feature across this diversity is that all have a complete metamorphosis with a pupal stage separating larva and adult. Nevertheless, there is some agreement that the group known as Neuropteroidea includes the most primitive oligoneopterans, and mostly have a pupa with free jaws (decticous pupa) and some limited mobility with the legs not sealed to the body surface (exarate pupa). The obtect pupa (without free jaws or limbs) is more characteristic of other groups, within which a closely-knit group of five orders have traditionally been recognized (sometimes as ‘the Panorpoid orders’) and the most advanced order, Hymenoptera, separated from these and sometimes treated as a third group, Hymenopteroidea. Despite some uncertainties of relationships (for example, of the Strepsiptera, as below), this traditional sequence is adopted here. Neuropteroidea comprises the three orders sometimes placed together in a broad concept of Neuroptera, but most commonly separated as Neuroptera, Megaloptera and (non-Australian) Raphidioptera, together with beetles (Coleoptera) and, possibly, Strepsiptera. The next series, Panorpoidea, gains its name from a northern hemisphere genus (Panorpa) of scorpionflies (Mecoptera), and includes this order together with two major pairs of orders as ‘Diptera plus Siphonaptera’ and ‘Trichoptera plus Lepidoptera’ – again with the Strepsiptera possibly linked, as allied to Diptera within this series. Adults of Megaloptera and Neuroptera both have long multisegmented antennae, large compound eyes and chewing mouthparts. Most have large membranous wings,
240
Appendix
Fig. A.3 Green lacewings (Neuroptera: Chrysopidae) are valued widely as predators of small pest insects. More than 50 Australian species are known, all with a complex open pattern of wing venation, with details having taxonomic importance
with complex venation (hence the common name ‘lacewings’ for many Neuroptera: Fig. A.3), and generally simple legs with five-segmented tarsi, and rather soft bodies. Order: Megaloptera (Alderflies, Dobsonflies) have aquatic larvae, and the weakly-flying adults generally occur only close to water. They are most easily separated from Neuroptera by having the hind wing large and, with an enlarged anal lobe, far the broader. The abdomen is generally very soft, and the insects are sometimes large. Wingspan of the smaller Sialidae (alderflies) is around 2 cm, and of the larger Corydalidae (Dobsonflies, both common names derived for northern hemisphere taxa, but applied globally to the respective families) is up to about 5 cm, with some non-Australian taxa reaching around 17 cm. Adults are short lived and, should they feed at all, are not active predators. In contrast, larvae are predators and some of them long lived, for up to several years. The appearance of the larvae is wholly diagnostic. They have short four-segmented antennae, strong chewing mouthparts, unsegmented tarsi and paired lateral gills on abdominal segments I–VII (Sialidae) or I–VIII (Corydalidae). Sialid larvae also have a long terminal filament, and the abdominal apex of Corydalid larvae has a pair of grasping prolegs. Sialidae, with only about four species in Australia, is considerably less diverse than Corydalidae, but the 30 or so species of both families are all endemic and, many of them Gondwanan. Globally, the Megaloptera is a small order of only some 300 described species.
Appendix
241
Order: Neuroptera (Lacewings, Antlions, Owlflies, Mantisflies and their allies) are much more diverse, typically predatory as both adults and larvae, and containing about 800 Australian species. Hind wings are generally little, if any, broader than the fore wings, and wing venation ranges from very complex to simplified in the tiny ‘dustywings’ (the Coniopterygidae, covered in white or greyish wax and by far the smallest representatives of the order at only 2–3 mm long). The largest species, some of the antlions (Myrmeleontidae, easily the largest family in Australia with perhaps 200 species) span around 12 cm. Most Neuroptera are terrestrial, but three of the 15 families in Australia have aquatic larvae, and two of these are wholly freshwater denizens. Spongeflies (Sisyridae) are specialised larval predators on freshwater sponges; Nevrorthidae and some Osmylidae are substrate-frequenting general predators, with the last family also including terrestrial forms associated with eucalypt bark. Adult lacewings are thought of as predators and, whilst most are so, others feed on nectar or pollen rather than insect prey. Mantisflies (Mantispidae) have raptorial fore legs, resembling those of the mantids from which they gain their names. Larvae are almost wholly predators, with different groups active hunters (as are many green lacewings and brown lacewings valued in pest management) or ambush predators, such as the antlions, with their best known habit being to dig conical pits in sand and lurk at the base of these to eat ants and others that fall in and cannot escape. Most antlion larvae, however, do not do this, and move actively on or under the substrate. The mouthparts of lacewing larvae are unusual and diagnostic, differing markedly from those of Megaloptera. The mandible and maxilla of each side are slender and elongate, and apposed to form a sucking tube that is inserted into the prey and through which body contents are extracted. Body form and mouthpart structure are useful characters in distinguishing families and many lower level taxa. Endemism is high in Australia, and the richness of Myrmeleontidae reflects their radiation in the more arid regions The related ‘owlflies’ (Ascalaphidae) are strongly flying predators and many features (such as very large compound eyes and spiny legs) parallel those of dragonflies, from which they are easily distinguished by having long, clubbed antennae. Both Gondwanan groups (such as Osmylidae and Nymphidae) and northern elements occur. Order: Coleoptera (Beetles) is probably a sister-group to the other neuropteroids, and its members are unified by the combination of chewing mouthparts and the fore wing modified to a hard protective elytron covering the hind wing and, in most, the entire abdomen. They are otherwise very variable in appearance, as befits the largest of all animal groups, and are ecologically diverse and pervasive across virtually all terrestrial and freshwater biomes. More than 350,000 species have been described, but those (about 23,000 species) described from Australia may be only about 20% of the total local fauna. Beetle antennae are basically 11-segmented, sometimes shortened, and very variable in form, and the only major part of the thorax visible from the dorsal surface is the enlarged pronotum. The membranous hind wing, wholly concealed beneath the elytra in most beetles has very reduced venation, with details important in classification. Legs, also, vary considerably with tarsal segment number ranging
242
Appendix
from 3 to 5 in different groups, and sometimes different on different pairs of legs; legs may be modified for a range or activities, such as running, burrowing, swimming or jumping. The abdomen, covered by the elytra other than in some short-winged taxa (such as rove beetles, Staphylinidae), is usually not toughened dorsally. Cerci are absent. Larvae, also, are very variable, with those of many families diagnostic in appearance, and many specialised divergences from the basic oligopod form. Beetles occupy almost all trophic roles, although true parasitic forms are rather unusual. Many are in some way herbivores, with large numbers of foliage eaters, seed eaters, wood feeders and others distributed across many different major lineages and according many beetles severe pest status or positive values depending on the individual context in which each may operate. They may thereby acquire massive economic importance as, for example, pests or biological control agents. Decomposers include many that feed on dung, carrion or dead vegetation. Many others, including numerous water beetles as well as terrestrial groups (such as ground beetles, Carabidae, and many rove beetles), are predators. Within any such broad role, individual species range from extreme specialisation to generalist feeders. Coleoptera are classified conventionally into four suborders, considered natural entities on both adult and larval features. Two of these are particularly diverse, and the others much smaller, although both of the latter, Archostemata and Myxophaga, are represented by a few species in Australia. Adephaga includes the ground beetles and a variety of water beetles, whilst Polyphaga is by far the most diverse beetle group and contains all other families likely to be encountered generally – such as ladybirds, dung beetles, Christmas beetles, timber beetles, leaf beetles, weevils, click beetles and many others. The relativity of these two predominant suborders is demonstrated by the number of families of each in Australia: Adephaga about seven, and Polyphaga, slightly more than a hundred. However, as in other large insect orders the formal boundaries between some families, and relationships between them, continue to be debated. These two suborders are separated by the form of the basal abdominal segment – underneath, it is complete in Polyphaga but divided by the hind coxae in Adephaga. Thorax and hind wing characters are also involved in their differentiation. Order: Strepsiptera (‘Strepsipterans’, Twisted-wing insects) are a small group (perhaps 600 species) of highly specialised parasites, in the past presumed likely to have been derived from Coleoptera (so filling the parasitic role otherwise scarce amongst the beetles), but with some recent analyses suggesting, rather, that they may be more closely allied to Diptera. The major problem of conventional interpretation by comparing morphology is that the Strepsiptera have become extremely specialised so that few structural features resemble those of any other order, so obscuring inferences of their origin. Some authorities suggest that they are not really related to either of the above orders. Strepsiptera thereby remain somewhat enigmatic, and interpretation is hampered further by their very strong sexual dimorphism. Males (most of them only 2–3 mm long) have enlarged branched antennae and prominant bulging eyes; the fore wings are small elytron-like straps (hence, at least superficial parallels with beetles) and the hind wings large and fanlike, with
Appendix
243
venation strongly reduced. They are shortlived and do not feed, and fly to encounter females, which are grublike, wingless, legless, and remain within the host (usually bugs such as leafhoppers, or social wasps), from which they protrude to emit a pheromone mate-attractant. The insects are rarely seen even by most entomologists and possibly 70 species occur in Australia, of which about 40 have so far been named, and exhibiting high endemism. Major taxonomic divisions amongst the eight families are based on male features, such as antennae and wings, whilst females provide few easily accessible characters to enable identification. The so-called ‘Panorpoid orders’, also, include a variety of familiar and less familiar insect orders. The group incorporates five orders, but recent arrangements suggest that applying this single name might be an oversimplification. The flies (Diptera) and fleas (Siphonaptera) are closely related and the scorpionflies (Mecoptera) appear closely allied to these – these three orders are often grouped as the Antliophora. Although debate continues over the proper placement of fleas – whether they are really closer to flies or to a rather unusual group of scorpionflies – the other two orders, caddisflies (Trichoptera) and moths and butterflies (Lepidoptera) are clearly related and grouped confidently as the Amphiesmenoptera. The somewhat archaic ‘Panorpoidea’ reflects the presumed primitive position of the Mecoptera giving rise to two branches as (Diptera plus Siphonaptera) and (Trichoptera plus Lepidoptera); as suggested above, this may not be quite as clearcut as this dichotomy implies. Order: Mecoptera (Scorpionflies) is, however, the smallest and, from the fossil record, most ancient order in this suite, with a global richness of somewhat more than 500 species. Adult scorpionflies are slender, chewing insects, with the head extended to a narrow rostrum and the mouthparts at the tip of this. Eyes are large, and antennae slender. Legs are long, with five-segmented tarsi, and the two pairs of wings are slender, in some taxa with many crossveins present. A few species are shortwinged or wingless. The common name flows from the form of the abdomen which, particularly amongst males, is raised at the tip and expanded to incorporate the genital capsule, and has been likened to the sting of a typical scorpion. Cerci are very short. Larvae are highly unusual in being the only endopterygotes to have compound eyes at this stage, rather than the usual group of simple stemmata, and are somewhat caterpillarlike in appearance, sharing with them the presence of some abdominal prolegs. The most familiar scorpionflies in Australia are the very longlegged ‘hanging flies’ (Bittacidae) with raptorial tarsi, active generalist predators in which males present females with a captured prey insect as a ‘nuptial gift’ during courtship. Adult diets of some other mecopterans are unclear, and the larvae are apparently scavengers on a wide range of animal and vegetable material. The small flightless Apteropanorpidae are found in alpine regions of Tasmania. Nannochorista (the only genus in Australia also known elsewhere, from South America) is associated with waterside vegetation and is also Gondwanan. Other anomalous distributions occur: the family Meropeidae, for example, is known by single species from Western Australia and North America. Only about 30 Australian species have been described.
244
Appendix
Order: Diptera (Flies, including Midges, Mosquitos, Marchflies, Hoverflies, Bushflies and many others) are one of the largest insect orders and – as the order name implies – are characterised by having only two fully developed wings: the fore wings are normal, and the hind wings reduced to small sensory structures, halteres, used to aid orientation during flight. Diptera are otherwise extraordinarily varied in form and biology, and some ectoparasitic forms parallel lice and fleas in having lost their wings. ‘Typical flies’, such as the housefly and closely related bushfly, have a large head and compound eyes. Fly antennae are long and many-segmented in more primitive forms (in which they may be threadlike or plumose), but reduced to short structures with only three segments and a terminal bristle (arista) in more advanced taxa. Mouthparts are also variable, including typical piercing structures (with mandibles and maxillae elongated as stylets, as in female mosquitos) to ‘sponging’ in which mandibles have been lost and the major structure present is an expanded labium (labellum) that is adpressed to the feeding substrate; some ‘biting flies’, such as marchflies (Tabanidae) retain mandibles in conjunction with this, in order to obtain blood. The thorax is large, and fore wing venation generally rather simple, with few crossveins. The abdomen is elongate or short and stout, and genitalic structures can become elaborate and difficult to interpret clearly. Dipteran larvae are also very varied in form, but mostly lack distinct thoracic legs, and the mouthparts of terrestrial taxa may be retracted into the apparent thorax (with true head absent) as highly reduced ‘mouthhooks’, as in typical maggots. Freeliving aquatic larvae, such as those of mosquitos, may be more elaborate, with much more distinct mouthparts including mandibles, and a well-formed head. Collectively, Diptera exploit virtually any organic material available, as predators, scavengers or herbivores, and range from specialists to much more generalist feeders. Some bloodfeeding adults (particularly some mosquitos, midges, marchflies) are vectors of diseases of livestock and people, and/or inflict painful bites. Their varied biology gives Diptera enormous ecological and economic importance with some regarded as highly beneficial (for example hoverflies as predators of small pest insects on crops, and as pollinators) or highly harmful to stock (screwworm, p. 188), crops or other commodities. The two main taxonomic groups of Diptera are the suborders Nematocera (adults with long, many-segmented antennae, and larvae with distinct heads and many of them aquatic) and Brachycera (adult antennae short, as above, and larvae with heads reduced), respectively including the mosquitos, midges, daddy longlegs and their allies, and all the more familiar flies. The extent of larval head reduction is one of the features used to divide Brachycera into two further large groups, the more primitive Orthorrhapha (such as marchflies and robberflies, in which the larval head is partially distinct) and the advanced Cyclorrhapha (most others, such as houseflies, in which the larval head is retracted fully into the thorax). Many of the almost 100 families of flies in Australia await detailed study. Of these, about 20 families are in the Nematocera, slightly fewer in Orthorrhapha and the remaining 50 or so are Cyclorrhapha. The fauna shows strong Gondwanan relationships and may be of the order of 30,000 species, with about two thirds of these undescribed. Species-level endemism is very high.
Appendix
245
Order: Siphonaptera (Fleas), like lice, are all ectoparasites of warm-blooded vertebrates, often with very specific mammal or bird hosts. These two groups show many parallel adaptations for this lifestyle, despite the fundamental differences in their biology. The parallels include loss of wings, and compression of the body – in fleas this is laterally – and backwardly directed combs or strong bristles that facilitate movement on the host. Fleas are small (at most a few mm) and their mouthparts are piercing, enabling all fleas to be blood feeders. Compound eyes are absent, and antennae short. The best-known feature of fleas is their enlarged hind legs that enable strong leaping as major means of locomotion and transfer between hosts. Claws are usually strong, at the end of five-segmented tarsi. Larvae feed on organic debris within the host nest or burrow and are slender, elongate, legless but with well formed heads and chewing mouthparts; however, they lack eyes and have very short antennae with only a single segment. Some fleas are effective vectors of blood-borne diseases – with the ‘plague flea’ (Xenopsylla cheopis) notorious in medical history as the transmitter of bubonic plague. Fleas that infest poultry or domestic pets can occur in very large numbers and attacks can cause anaemia and general debilitation. Whereas some such species associated with those hosts are cosmopolitan, many of the around 100 described Australian fleas are endemic, most of them on native mammals with rather few on bird hosts. The predominant family, Pygiopsyllidae (with about half the Australian species), spans the host range of monotremes, marsupials, rodents and birds. Faunal relationships are largely those of the hosts, as is also the case with Phthiraptera, so that Australian fleas have considerable affinities with those of New Guinea and the Neotropics, in particular. However, limits of some families are still not wholly agreed. Nine families occur in Australia. Amphiesmenoptera, as implied above, are a very clearly defined pair of orders, with the boundaries between the aquatic Trichoptera and the largely terrestrial Lepidoptera in places still unclear. Some primitive moths retain functional mandibles and lack the coiled proboscis so definitive of most Lepidoptera, and the wing hairs of some caddisflies are flattened to constitute scales, for examples. Lepidoptera are presumed to be a relatively recent development amongst insects, with their diversification tracking the radiations of flowering plants from the Cretaceous period onward. Trichoptera appear somewhat earlier in the fossil record. Order: Trichoptera (Caddisflies) gain their order name from having ‘hairy wings’, the membrane in many species being covered with short slender hairs, the fore runner to the flattened scales found in some species and more universally in Lepidoptera. The general appearance of some caddisflies is that of ‘small hairy moths’. Antennae are usually long and threadlike, eyes large, and the head and thorax may have characteristically sited ‘setose warts’; mouthparts, chewing in form, are poorly developed and mandibles generally small. Many are thought not to feed, but some species probably imbibe nectar or take pollen. Palpi, however, are long. The fore wing is almost always more slender than the hind wing, and may be somewhat ‘toughened’ in texture. Legs are long and slender; tarsi have five segments and the tibiae have long apical and/or mid-length spurs, the arrangement of which has taxonomic value. The abdomen is slender and often rather soft.
246
Appendix
Adults usually occur close to water, and larvae of all Trichoptera are aquatic, as the major aquatic group of holometabolous insects. Larvae have a well developed head, chewing mouthparts, strong legs oriented forward, and the soft abdomen most commonly concealed in a protective case constructed of sand or vegetation, with the particular form and materials used being highly characteristic for many families or other taxonomic groups. Many such larvae are herbivores or omnivores, and protrude from the front of the case to feed and move around. Some freeliving forms also occur, some of which are predators. A few are filter feeders, using silken nets to extract small particles from the water current. All lack abdominal prolegs, although many have setose patches or gills on the abdomen, aiding water flow through the case, and gas exchange. Relationships between different groups of caddisflies are still open to debate, with three suborders recognised commonly and separated on features of palpi, venation and genitalic structures for adults, in addition to larval features. Caddisflies range in size from the minute Hydroptilidae (a few mm long, some associated with seepages) to much larger forms up to about 4 cm long, and occur in all kinds of water bodies. The estimated 800 or so species in Australia are divided amongst some 25 families. Endemism is high, and many have strong Gondwanan relationships. Order: Lepidoptera (Moths, Butterflies) are characterised primarily by the broad wings being covered with flattened scales, often brightly coloured and forming intricate patterns, and a coiled suctorial proboscis formed from part of the maxillae. Mandibles are usually absent, and the labial palpi large and conspicuous. Antennae are long and very varied in form. Characteristically the superfamilies Hesperioidea and Papilionioidea (the ‘butterflies’ in Australia) have antennae that end in an expanded club, whilst all others (the moths) have slender or feathery antennae, often sexually dimorphic. However, as with many features used in attempts to ‘pigeonhole’ large groups of insects, exceptions occur – dayflying Castniidae (sun-moths) for example, have clubbed antennae. The most primitive moths, indeed, retain mandibles and lack a coiled proboscis, and feed on pollen rather than nectar. Wings are often large and broad, but with rather simple venation, and in a few species are strongly reduced or absent. Larvae (caterpillars) typically have chewing mouthparts and a varied number of fleshy abdominal prolegs, their form and placement having taxonomic value. Most are herbivores, feeding on or in terrestrial plants, but a few groups have become predators (on scale insects or ant brood, for examples) and some have become aquatic, feeding on water lilies and other plants – some even constructing cases reminiscent of those of Trichoptera. In short, ecological variety is considerable and extends further to some caterpillars eating wool or feathers, and some adult moths piercing fruit and, more rarely, becoming blood feeders. Many Lepidoptera are serious pests, mainly through larval feeding, and may demand complicated control to protect crops and stired commodities, in particular. Lepidoptera are represented by more than 20,000 species in Australia, and at least half of these are not yet named. Several suborders are recognised, but almost all species likely to be encountered belong to the Glossata (with a proboscis), with
Appendix
247
the other three suborders still having mandibles. Within Glossata, one group, Ditrysia, is easily predominant, with about 70 families represented in Australia. For comparison, the global Lepidoptera fauna includes about 120 families, and probably 170,000 described species. Order: Hymenoptera (Sawflies, Bees, Ants, Wasps and their allies) are sometimes regarded as an isolated order containing the most advanced insects; they may be the sister group of ‘Antliophora plus Amphiesmenoptera’ together. Hymenoptera are very varied, with mouthparts ranging from chewing to sucking and chewing, long many-segmented antennae, the compound eyes generally large, the body hardened, and two pairs of membranous wings. The fore wing is usually larger than the hind wing, and wings may be lost completely. Wing venation ranges from complex to extremely simple, with many tiny parasitoid wasps having little trace of venation. Feeding habits are also variable, with numerous herbivores, predators and parasitoids, many of them very restricted in food type or host range. Development of social existence is a major feature of some groups (p. 101), but most species within the order are solitary. The two major divisions of Hymenoptera have been treated as suborders, although that arrangement is probably not wholly natural. Traditionally, however, the Symphyta (sawflies, woodwasps) have been regarded as the more primitive group. They are herbivores, with their caterpillar-like larvae (such as ‘spitfire grubs’ on Eucalyptus) feeding on foliage, and those of woodwasps (Sirex, p. 172) found internally in trees. The adults are characterised by having relatively complex wing venation and the thorax and abdomen joined broadly rather than having a narrow constricted ‘waist’ between them. The latter is characteristic of the other suborder, Apocrita (bees, wasps, ants), but the structure is not as it initially appears, as a division between thorax and abdomen, because the first segment of the abdomen (propodeum) remains broadly attached to the back of the thorax, and the constriction is within the abdomen. For precise morphological definition these regions are, rather, termed ‘mesosoma’ and ‘metasoma’ in Apocrita. The ‘wasp-waist’ so formed is believed to be an adaptation fostered by the parasitoid way of life, in increasing the flexibility of the abdomen so wasps can lay eggs in concealed places or through resisting tissues such as bark or the host body walls. This group contains all the well-known eusocial Hymenoptera and the vast array of solitary parasitoids. Apocrita are regarded commonly as divisible into ‘Aculeata’ (sting bearers, in which the ovipositor has become modified for delivering venom, and eggs are laid from the base of this) and ‘Terebrantia’ (with a ‘proper ovipositor’: a more informal use of the term applied also as a formal name to a group of thrips, p. 123). The Aculeata are generally accepted as a closeknit single group, although relationships within it are still somewhat unclear. More tangibly, they include the Vespoidea (wasps, with the family Formicidae, ants, included within this series), Apoidea (bees and some other wasps, some treated in some systems as a separate group, Sphecoidea) and Chrysidoidea (ruby wasps). Phylogenetic relationships between the various lineages of parasitoids are, likewise, still debated. Knowledge of Australia’s Hymenoptera is very uneven. Around 70 families occur. Some groups have received considerable study – for example, ants, as
248
Appendix
presumed useful tools in environmental assessment and being quite easy to sample – so that generic level frameworks for these are robust, and many species recognisable. Others, predominantly amongst some of the larger groups of parasitoids, remain the province of individual specialists, with little widely intelligible information available for more general application. The fauna has been estimated at 40,000–50,000 species, with perhaps only about 20% so far described. But, to quote from a recent (2004) overview ‘… the Australian fauna is an incredibly rich one containing a very high proportion of endemic species and genera, and numerous unique elements at higher levels’ and ‘… in reality the true size of the Australian fauna is difficult if not impossible to estimate’. Biological knowledge, particularly of many parasitoids, is also very patchy – most of the described species have not been associated clearly with any particular host, and any confirmed record may be only part of a wider range or represent extreme specialisation; the latter is often presumed rather than confirmed. Biological oddities abound: for example, the secondary plant-feeding habit (including gall-forming) of wasps clearly derived from parasitoid lineages is unusually high in Australia, so that wild caught wasps presumed to be parasitoids may, in fact, have very different roles.
Index
A Acacia, 75, 100, 114, 119, 123, 195 A. koaia, 196 A. longifolia, 195 A. mearnsii, 195 Achias, 98 Acizzia uncatoides, 196 Acripeza reticulata, 153 Acrodipsas, 73, 110 A. myrmecophila, 73, 110 Adoryphorus couloni, 161 Aedes, 180 A. camptorhynchus, 180 A. vigilax, 180 Agaonidae, 70 Agathis, 21 Agrotis infusa, 33, 150 Alpine silver xenica, 154 Amber, 20 Amitermes, 105, 109 Amylostereum areolatum, 172 Anaphylaxis, 108 Anopheles, 180 Anthela ocellata, 161 Anthelidae, 161 Anoplognathus, 159 Anoplolepis gracilipes, 192, 193 Antlions, 9, 80 Ants, 45, 73, 100, 110, 163, 192, 194, 206, 208 Aphids, 26, 93, 104, 189 Aphis fabae, 93 Aphodius A. pseudotasmaniae, 161 A. tasmaniae, 161 Apiomorpha, 48, 50 A. conica, 50 A. munita, 48, 50
Apis mellifera, 40, 191 Apterygota, 2, 13, 23, 53, 205 Archaeognatha, 1, 7, 13 Archimantis latistylus, 26 Aretianella longoi, 196 Argentine ant, 105, 194 Aristolochia elegans, 90 Ascalaphidae, 79, 84 Asclepias, 188 Asilidae, 84 Assassin bugs, 79 Assemblages, 115 Atherospermum moschatum, 168 Austracris guttulosa, 135 Australian admiral, 98 Australian hairstreak, 168 Australian plague locust, 3, 133 Austromerope poultoni, 42 Autumn gum moth, 135 B Bacillus thuringiensis, 182 Backswimmers, 144 Bactrocera, 46 B. tryoni, 179, 183 Bark beetles, 189 Bedbugs, 87 Bees, 102, 178, 191 Beetles, 11, 18, 73, 75, 94, 113, 159, 167, 171 Big-headed ant, 194 Biogeographical regions, 40 Biston betularia, 18 Bittacidae, 243 Blattodea, 19, 229 Blephariceridae, 146 Bogong moth, 33, 103, 150
T.R. New, ‘In Considerable Variety’: Introducing the Diversity of Australia’s Insects, DOI 10.1007/978-94-007-1780-0, © Springer Science+Business Media B.V. 2011
249
250 Bombus terrestris, 194 Braconidae, 15 Bristletails, 7 Bufo marinus, 205 Bull ants, 45, 104 Bumblebees, 194 Buprestidae, 45, 94 Bursaria spinosa, 206 Butterflies, 11, 41, 44, 49, 51, 64, 73, 82, 90, 98, 110, 177, 202 C Cactoblastis cactorum, 177, 178 Caddisflies, 20, 78 Caledia captiva, 47, 215 Camouflage, 18, 82 Camponotus, 163 Cane toad, 205 Capital reproducers, 71 Carabidae, 53 Carboniferous fossils, 18 Castes, 103 Castiarina, 46 Castniidae, 158 Cave beetles, 207 Census population, 129 Centinelan extinctions, 214 Cephalotus, 90 Cerambycidae, 171, 178 Chameleon grasshopper, 153 Chaoborus, 146 Chauliognathus lugubris, 174 Chequered swallowtail, 90 Chortoicetes terminifera, 3, 133, 135, Christmas beetles, 159 Chrysomelidae, 161, 169, 174 Chrysomyia bezziana, 188 Chrysophtharta bimaculata, 174 Chrysopidae, 240 Cicadas, 95 Cimicidae, 87 Cinnabar moth, 83 Classification, 224 Cleobora mellyi, 174 Climate matching, 191 Coccinellidae, 116 Cockroaches, 19, 105, 153, 180, 230 Coconut ants, 73 Codling moth, 4, 184 Coleoptera, 18, 53, 97, 146, 189, 241 Collecting, 41, 216 Collembola, 1, 13 Comptosia, 217
Index Cooloola monster, 3 Cooloola propator, 3 Corixidae, 144 Corydalidae, 240 Cossidae, 171, 178 Cottony cushion scale, 196 Crabs, 193 Creophilus erythrocephalus, 117 Cressida cressida, 11 Cretaceous period, 21 Crickets, 3, 15, 23, 24 Critical resources, 61 Crypsis, 82 Cryptocercus, 105 Culicidae, 146 Curculionidae, 220 Cyclochila australasiae, 95 Cydia pomonella, 4 Cynipoidea, 173 Cyrtobagous salviniae, 178 D Danaus plexippus, 188 Dermaptera, 232 Dermolepida albohirtum, 205 Devil’s coach-horse, 117 Diapause, 34 Dicopomorpha, 220 Didymuria violescens, 48, 136 Dinosaur ant, 117, 163, 177 Diplura, 2 Diptera, 2, 4, 53, 79, 80, 84, 98, 139, 146, 244 DNA, 47, 51 Dragonflies, 10, 14, 15, 202 Drosera, 90, 91 Drosophila, 185 Dryococelus australis, 204 Dung beetles, 73, 88, 191 Dutch Elm Disease, 189 Dutchman’s pipe vine, 90 Dytiscidae, 139, 146, 207 E Earwigs, 100 Ecological Vegetation Classes, 57 Ecoregions, 57 Ectognatha, 2 Effective population, 129 Elm leaf beetle, 190 Eltham copper, 132, 206 Embioptera, 234
Index Empididae, 79 Encyrtidae, 124 Endemism, 37 Ephemeroptera, 17, 27, 141, 226 Ephydridae, 139 EPT Index, 27 Eriococcidae, 48 Eucalyptus, 39, 48, 75, 77, 89, 119, 125, 168, 196 E. calophylla, 125 E. camaldulensis, 78 E. coccifera, 150 E. gunnii, 150 E. marginata, 75, 125 E. nitens, 174 E. pauciflora, 150 E. regnans, 136, 168 Eumastacidae, 158 Extatosoma tiaratum, 234 F False head, 83 Ficus, 70 F. macrophylla, 70, 71 Figs, 70 Fig wasps, 70 Fleas, 87 Forficula auricularia, 100, 232 Formicidae, 116, 117 Fruit flies, 46, 179, 183 Functional groups, 78, 118, 140, 208 Fungi, 172 G Gahnia, 44 G. filum, 44 Galls, 100, 107, 114, 120, 123 Gastrimargus musicus, 82 Geocarcoidea natalis, 193 Geometridae, 135, 154 Gigantism, 14 Gills, 15 Glycaspis, 121 Goat moths, 171 Goedetrechus mendumae, 207 Golden sun-moth, 158 Gondwana, 38, 155 Grapholita molesta, 4 Grasshoppers, 15, 47, 82, 152, 153, 158 Grasslands, 59, 157 Greengrocer cicada, 95
251 Green tree ant, 110 Grylloblattodea, 35 Gum emperor moth, 31 Gum-leaf skeletoniser, 134 Gyrinidae, 146 H Hairy cicadas, 96 Harmonia conformis, 196 Harpobittacus, 99 Hecatesia, 96 Helicoverpa, 52 Heliothis, 52 Hemiphlebia mirabilis, 42, 227 Hemiptera, 4, 87, 144, 236 Hepialidae, 72, 178 Hesperilla H. donnysa, 44 H. flavescens, 44 Heteronympha merope, 25, 64 Hexapoda, 1, 13 Hill tops, 98 Homoptera, 120 Honey bees, 40 Honeypot ants, 178 Hoplogonus, 172 Hybridisation, 48 Hymenoptera, 28, 53, 100, 107, 113, 191, 218, 247 I Ibalia, 173 Icerya purchasi, 196 Ichneumonidae, 16, 53, 173 Idacarabus troglodytes, 207 Idiobionts, 86 Imperial blue, 82 Indicators, 208 Insectivorous plants, 90 Interest reproducers, 71 Iridomyrmex, 110, 121, 163, 164 Ischnura aurora, 143 Islands, 115 Isoptera, 106, 231 J Jalmenus evagoras, 82, 83, 110 Jarrah leaf-miner, 75 Jewel beetles, 45, 94 Julodimorpha bakewelli, 94 Jumping jack ant, 108
252 K Kairomones, 99 Kladothrips, 107 Koinobionts, 86 Kosciuscola tristis, 153 L Lacewings, 240 Ladybirds, 116, 174, 196 Leaf beetles, 161, 174 Leaf hoppers, 97 Leaf miners, 75 Learning, 93 Legislation, 201 Leichardt’s grasshopper, 162 Lepidoptera, 4, 53, 60, 113, 115, 219, 246 Lerp insects, 120 Lice, 87, 113 Linepithima humile, 105, 194 Liphyra brassolis, 110 Lissotes latidens, 172 Locust, 3, 69, 135 Locusta migratoria, 135 Longhorn beetles, 196 Lord Howe Island stick insect, 204 Lucanidae, 31, 171, 172 Lycaenidae, 49, 73, 110, 205 M Macleay’s spectre, 234 Magnetic termites, 105 Mantids, 79, 80 Mantispidae, 79 Mantodea, 231 Mantophasmatodea, 3, 35 Marine insects, 19, 139 Marmeneura, 143 Mastotermes darwiniensis, 104, 231 Mastotermitidae, 106 Mayflies, 14, 17 Mecoptera, 243 Megaloptera, 143, 147, 220, 240 Megarhyssa, 173, 220 Melaleuca howeanum, 204 Melanodes anthracitaria, 19 Melanterius, 75, 195 Melophorus, 163, 178 Membracidae, 82 Metamorphosis, 24 Metapopulation, 132 Migration, 33, 62, 150 Migratory locust, 135 Mimicry, 84
Index Mnesampela privata, 135 Mole crickets, 4 Monarch butterfly, 188 Monterey pine, 60, 172 Morabine grasshoppers, 158 Mosquitoes, 180 Moss bugs, 168 Moth butterfly, 110 Moths, 4, 18, 19, 28, 51, 96, 99, 134, 151, 158, 171, 219 Mount Donna Buang stonefly, 205 Mountain katydid, 153 Mudeyes, 10, 142 Muscidae, 88 Mymaridae, 86 Myrmecia, 45, 104, 108 M. gulosa, 45, 108 M. pilosula, 108 M. pyriformis, 108 Myrmeleontidae, 80 Myrtle beech, 56 N Nannochorista, 91, 243 Nasutitermes, 105 Nematodes, 173 Neoptera, 2, 23, 79, 229 Neuroptera, 9, 32, 84, 241 Nevrorthidae, 145 Noctuidae, 134 Nothofagus, 39, 56, 59 N. cunninghamii, 56, 168 Nothomyrmecia macrops, 117, 164, 177 Notoncus, 206 Notonectidae, 114 Nymphalidae, 64 Nymphidae, 79, 81 O Oak-leaf miner, 75 Ochrogaster, 99 Odonata, 17, 27, 42, 79, 141, 147, 227 Oecophoridae, 123 Oecophylla smaragdina, 111 Oenochroma, 51 O. barcodificata, 51 Oligopod larvae, 28 Oncopera O. intricata, 161 O. rufobrunnea, 161 Onitis, 74 Ootheca, 26 Opodiphthera, 32
Index Opuntia, 177 Orchids, 4, 70 Oreixenica latialis, 154 Oriental fruit moth, 4 Ornithoptera richmondia, 90, 95 Orthoptera, 4, 99, 233 Osmylidae, 145 Osmylops sejunctus, 81 Outbreaks, 133 Owlflies, 79, 84 P Palaeoptera, 2, 17, 26 Papilio P. demoleus, 90 P. ulysses, 177 Papyrius, 73 Paralucia pyrodiscus lucida, 132, 205, 206 Paraserianthes, 120 Parasitoid wasps, 15, 73, 85, 124, 172, 217 Paropsis atomaria, 169 Passeromyia, 88 Peloridiidae, 168 Peppered moth, 18 Perga, 170 Pergidae, 170 Perthida glyphopa, 75, 134 Petalura, 42, 229 Petasida ephippigera, 162 Phalaenoides glycine, 28, 29 Phaulacridium vittatum, 152 Pheidole megacephala, 194 Pheromones, 18, 97 Philanisus plebeius, 20 Phoracantha semipunctata, 196 Phthiraptera, 87, 238 Phylacteophaga froggatti, 196 Phyllonorycter messaniella, 75 Pinus radiata, 60, 172 Plant architecture, 59 Plant defences, 76 Platystomatidae, 98 Plecoptera, 25, 27, 141, 235 Pleistodontes, 70 P. froggatti, 70, 71 Polistes, 100 Pollination, 4, 70 Polyembryony, 27 Polypod larvae, 27 Polyzosteira, 230 Population dynamics, 130 Predators, 79 Processionary caterpillars, 99 Protected species, 201
253 Protopod larvae, 27 Pselaphinae, 217 Pseudalmenus chlorinda, 168 Pseudoperla lewisii, 151 Psocodea, 237 Psocoptera, 238 Psyllaephagus, 124 Psyllidae, 196 Psyllids, 97, 120, 121, 136 Psylloidea, 120 Pterolocera, 161 Pteromalidae, 120 Pterygota, 226 Pyralidae, 145 Pyrrhalta luteola, 190 Q Quadristichodella nova, 196 R Radiations, 119 Ragwort, 83 Raphidioptera, 35 Rarity, 66 Red Imported Fire Ant, 192 Reduviidae, 79 Rhyniella praecursor, 13 Rhyniognatha hirsti, 13 Rhyssa, 173 Richmond birdwing, 90, 95 Riekoperla darlingtoni, 155, 205 Robberflies, 84 Rodolia cardinalis, 196, 197 S Sampling units, 65 Saproxylic insects, 171 Sawflies, 28, 30, 89, 151, 170, 196 Scarabaeidae, 159, 161, 205 Schlettererius, 173 Scolytidae, 189 Scolytus, 189 Scorpionflies, 42, 99 Screwworm fly, 188 Sialidae, 240 Silverfish, 7, 23 Siphonaptera, 87 Sirex noctilio, 172, 173 Sisyphus, 74 Sisyridae, 145 Skipper butterflies, 44 Soldier beetles, 174
254 Solenopsis invicta, 192, 193 Southern sassafras, 168 Species accumulation curves, 125 Species concepts, 43, 211 Spitfire grubs, 28, 84, 99 Spondyliaspis, 121 Spur-legged phasmid, 136 Spur-throated locust, 135 Stag beetles, 171 Stalk-eyed flies, 98 Staphylinidae, 117 Stephanidae, 173 Sterile males, 184 Stick insects, 48 Stoneflies, 154, 205 Strepsiptera, 2, 88, 97, 242 Stridulation, 96 Supercooling, 152 Swift moths, 72, 161 Swordgrass Brown, 44 Synemon plana, 158, 159 Syntonarcha iriastis, 96 T Teleogryllus commodus, 3 Tension zones, 48 Termites, 8, 89, 104, 129, 164, 231 Tettigarcta, 96 Tettigarctidae, 96 Thaumatoperla alpina, 155 Thaumatopoeidae, 99 Threats, 203 Thrips, 52, 100, 107, 121 Thysanoptera, 121, 237 Timber beetles, 171 Tisiphone abeona, 44 Torres Strait, 38 Tree holes, 170 Tree-hoppers, 82 Trichilogaster, 120, 122, 125 T. acaciaelongifoliae, 195 Trichoptera, 27, 143, 245 Trictena, 72 Trigona, 178 Tyria jacobaeae, 83
Index U Ulysses butterfly, 177 Uraba lugens, 184 Uromycladium, 114 V Vanessa itea, 98 Vedalia beetle, 196 Vegetation types, 56 Vespidae, 102 Vespula, 190, 191 V. germanica, 191 V. vulgaris, 191 Vine moth, 28, 29 Viruses, 180 W Wasps, 190 Water beetles, 94, 115, 207 Water boatmen, 144 Water scorpions, 79 Webspinners, 234 Weevils, 75, 178, 195, 220 Whirligig beetle, 146 Whistling moth, 96 Wing flexing, 17 World Conservation Union, 202 X Xenopsylla cheopis, 87, 245 Y Yellow crazy ant, 192 Yellow-winged grasshopper, 82 Z Zizina Z. labradus, 49 Z. otis, 49 Zoraptera, 35, 235 Zygentoma, 7