Advances in Insect Physiology, Volume 10

Advances in Insect Physiology

Volume 10

This Page Intentionally Left Blank

Advances in Insect Physiology edited by

J. E. TREHERNE M. J. BERRIDGE and V. B. WIGGLESWORTH Department of Zoology, The University Cambridge, England

Volume 10

1974

ACADEMIC PRESS LONDON A N D NEW YORK A Subsidiary of Harcourt Brace Jovanovich, Publishers

ACADEMIC PRESS INC. (LONDON) LTD 24-28 Oval Road London NW1

US edition published b y ACADEMIC PRESS INC. 111 Fifth Avenue, New York, New York 10003 Copyright 0 1974 by Academic Press Inc. (London) Ltd

A11 Rights Reserved No part of this book may be reproduced in any form, by photostat, microfilm or any other means, without written permission from the publishers

Library of Congress Catalog Card Number: 63-14039 ISBN: 0-12424210-9

PRINTED INGREAT BRITAIN BY THE WHITEFRIARS PRESS LTD., LONDON AND TONBRIDGE

Contributors John Brady

Department of Zoology and Applied Entomology Imperial College of Science and Technology London, England Bernt Linzen

Zoological Institute, University of Munich, G u m a n y Axel Michelsen

Biological Institute, University of Odense, Denmark Harald Nocke Zoological Institute, University of Cologne, Germany Lynn M. Riddiford

Department of Zoology, University of Washington Seattle, Washington, USA James W. Truman

Department of Zoology, University of Washington Seattle, Washington, USA

V

This Page Intentionally Left Blank

Contents Contributors

. . . . . . . . . . . . . . . . . . . . . . .

The Physiology of Insect Circadian Rhythms JOHNBRADY ............

...........

The Tryptophan + Ommochrome Pathway in Insects BERNT LINZEN . . . . . . . . . . . . . .

. . . . . .

Biophysical Aspects of Sound Communication in Insects AXEL MICHELSEN and HARALD NOCKE . . . . Hormonal Mechanisms Underlying Insect Beh.wiour JAMES W . TRUMAN and LYNN M . RIDDIFORD

v

1

117

. . . . . . 247

. . . . . . . 297

Author Index

. . . . . . . . . . . . . . . . . . . . . .

353

Subject Index

. . . . . . . . . . . . . . . . . . . . . .

369

Cumulative List of Authors

. . . . . . . . . . . . . . . .

399

. . . . . . . . . . . . . .

401

Cumulative List of Chapter Titles

vii

This Page Intentionally Left Blank

The Physiology of Insect Circadian Rhythms John Brady Department of Zoology m d Applied Entomology Imperial College of Science and Technology London, England

1 Introduction . 2 Circadian principles . 3 Types of insect circadian rhythms . 3.1 Behavioural rhythms 3.2 Developmental rhythms . 3.3 Physiological rhythms . 4 Timing processes 4.1 Entrainment 4.2 Control of overt rhythms by driving oscillators 4.3 Temperature effects on insect clocks . 4.4 Genetics of insect clocks 4.5 Truman's two clock types . 4.6 Mechanisms of driving oscillators 5 Conclusions

.

. . .

. . *

. . . . *

.

1 2 6

6 15 22 43 43 52 72 74 76 81 91

1 Introduction

No full-length review devoted t o the physiology of insect circadian rhythms has ever been published. Earlier surveys, having a wider scope (Harker, 1958a, 1961), provided the basis for the view of the control of insect rhythms incorporated into the textbooks of' the period (e.g. Harker, 1964; Wigglesworth, 1965;. Marler and Hamilton, 1966), but since then much new information has accrued which rnust alter many of those earlier assumptions. Not only does the resulting confusion require disentangling, but the 300 or so relevant papers published in the interim demand some kind of distillation. Reviews related to the physiology of insect circadian rhythms have been written since Harker's (e.g. Corbet, 1966; Beck, 4968; Danilevsky et al., 1970), and the present work will not duplicate their coverage, except where necessary for clarity. In particular, the relationship between photoperiodism and circadian rhythms will only be touched upon, since it has 1

2

JOHN BRADY

been amply covered by Lees (1968, 1972), Beck (1968), and, with a more ecological flavour, by Danilevsky et al. (1970; this paper is particularly valuable for its Russian bibliography, but has few English language references beyond 1965). Because of their relevance to the subject, the contents of two briefer reviews (Brady, 1969; Truman, 1972a) will, on the other hand, be extensively repeated. Otherwise, the present work will cover the advances in this subject since the early 1960s. It sets out to do two things: first, to outline those insect rhythms that have been described since Harker’s reviews (1958a, 1961); and secondly, in greater detail, to attempt to synthesize the various aspects of the underlying timing mechanisms that have been revealed by recent publications. It is with the endogenicity of insect circadian rhythms and their coupling to the implied underlying physiological oscillators that it is primarily concerned. Numerous entrained die1 rhythms (see p. 4 ) have been described in the last 10 years, but these will not be discussed unless it appears that they indicate a true circadian rhythm, or are illuminating for some other reason. To begin with, a brief summary of the principles of circadian rhythms is given in order to set the scene and t o clarify the terminology. The word “clock” is used throughout for brevity and convenience t o describe the unknown biochemical systems which comprise the relevant driving oscillators; this does not imply that the author necessarily considers that rhythms are controlled in insects by a single physiological oscillator. Indeed, that the reverse is the case will become apparent in the later sections. Also for convenience, the words “dawn” and “dusk” are used in quotes to indicate the instantaneous transitions from dark to light, and light to dark in artificial light cycles; when used without quotes, they refer to natural sunrise and sunset. Where a number appears after a colon in a text reference as, for example in (Robinson, 1973: 123), this refers t o a page (i.e. p. 123) in that work. Cross-references to pages within the present review come after a semicolon and the word “see”, e.g. in (Robinson, 1973: 123; see p. 456), the last number directs the reader to a relevant passage on p. 456 of this review, as would the interpolation (p. 456) by itself. 2 Circadian principles Circadian rhythms, along with their related circa-tidal and circa-lunar rhythms, have characteristics which distinguish them from all other biological oscillations. The most obvious of these is their link with the environmental cycles of days or tides. Most other bio-rhythms (e.g. spontaneous spike discharge, heartbeat, spiracular opening, and even some life-cycles) have no such temporal relationship with external cycles. This

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

3

link with the environment, while suggesting the adaptive significance of circadian rhythms, has led to the almost certainly erroneous assumption that they are a direct response to, and consequence of, cyclical environmental changes. A cockroach placed in an actograph and kept in a 12-h light: 12-h dark cycle (LD 12 : 12) performs the majority o f its locomotor activity during the first hour or two of darkness. This activity recurs at the same time every day, indefintely (Fig.. 1). As long as the LD cycle is maintained, it is impossible to say whether this activity rhythm is a direct response to the environmental signal provided by the artificial sunset, or whether it is endogenously timed. That the latter may be the case is suggested when the activity starts to change before the light signal is given (as in the records shown by Penplaneta, Harker, 1960b; Acheta, Cymborowski, 1969; Aedes, Nayar and Sauerman, 1971; Glossina, Brady, 1972a). The proof that these pre-signal changes indicate endogenous timing is provided by withdrawing the light cycle, ix. leaving the animal in constant light (LL) or constant darkness (DD). Peak activity is then still found to

Fig. 1. Locomotor activity record of a cockroach, Leupphaea maderue (simplified from original record by Roberts, 1962). Each horizontal line represents 24 h of record; the black blocks indicate periods of activity. Successive days are arranged in order down the page. Upper bars represent the light : dark cycle for the days as indicated.

4

JOHN BRADY

occur at approximately its normal time, in spite of the absence of time cues. Furthermore, the activity continues to be expressed rhythmically for days in the absence of any periodic input from light, temperature or humidity (Fig. 1). The important feature of such LL or DD rhythms is that they typically continue at periods which differ slightly, but consistently, from 24 h. This is the strongest evidence that their timing is endogenous, and the result of a response to an underlying physiological “clock”. If they continued ‘at a period of exactly 24 h it would not be possible to rule out their being a response to any of the other environmental features that vary with the earth’s rotation, such as magnetic field or cosmic radiation, which are not controlled in normal circadian experiments. But in practice such rhythms invariably drift relative to sol& time (Fig. l ) , and are therefore temporally independent of such external signals. This fiee-running drift away from solar time is characteristic of all endogenously timed daily rhythms, and is the origin of the term ci~cadian (from circa d i e m ) . Strictly, this word should therefore only be applied t o free-running rhythms, or rhythms which are known t o free run in constant conditions. It carries with it a clear implication of endogenous physiological timing, but is often incorrectly used to describe any rhythms in a LD cycle, even when there is no evidence for their endogenicity (Wurtman, 1967). Such entrained (see below) LD rhythms are conventionally termed die1 (i.e. daily) to distinguish them from the implications of the words diurnal and nocturnal. kircadian terminology borrows freely from physical theory (a full glossary is provided by Aschoff et al., 1965). Thus a rhythm is said t o oscillate, or to be the overt expression of an underlying self-sustaining oscillator. This oscillation has a natural free-running period (and therefore fieguency) which is characteristic of the individual under the ambient conditions. It has a period of 24 h when it is entrained by (i.e. synchronized to) a 24-h LD cycle, which provides it with its time-cue, or zeitgeber. The term photophase is used to describe the light part of a LD cycle in order to avoid confusion with the specific connotation of the term “photoperiod”, and the dark part of the cycle is therefore the scotophase. Any particular point in thGoscillation, e.g. peak activity. is a given phase of the rhythm, and when the rhythm is entrained there is a characteristic time-lag, or phase angle, between the zeitgeber and the measured phase (one oscillation = 360O). In some rhythms it is also useful to consider the amplitude of the oscillation. The word noise is used in the electronic sense to imply irrelevant or extraneous interference in the recording of the oscillation. Fitting these terms to the upper half of Fig. 1: the zeitgeber is “sunset”

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

5

(i.e. the LD transition), the measured phasr of the rhythm is the onset of peak activity which, since it starts exactly at “sunset”, bears aphase angle of 0’ to the zeitgeber; the period is exactly 24 h and theamplitude might be described as the amount of locomotor movement performed during the activity peak. The lower half of Fig. 1 has the rhythm no longer entrained, but free-running at a period of c. 24.5 h. There is a whole class of circadian rhythins which involve developmental events occurring once only in the life of each individual (e.g. eclosion), and which only appear as overt rhythms in synchronous populations. The term gating is applied to such events, implying that the event can only occur when the circadian clock opens the gate at the appointed time, and that if an individual misses a given day’s gate it must wait 24 h for the next one, or 48 h for the one after that. Circadian rhythms are fairly stable, and will not entrain to environmental cycles differing by more than a few hours from the 24; insects usually entrain only within the range of about 18 t o 30 h, but even well within this range, their survival is reduced by prolongtd entrainment to periods other than 24 h (Pittendrigh and Minis, 1972). Another consequence of this stability is that such rhythms generally do not adjust instantaneously to a new zeitgeber cycle, that is, they can only be phase shifted a few hours each cycle in response to a changed zeiigeber time so that they show transient cycles of less than or more than 24 h until they are re-entrained. Moreover, their ability to phase shift is not equal throughout the 24 h, and the amount of shift they perform depends on the phase difference between some sensitive phase of the animal’s ‘‘clock’’ and the new zeitgeber. This results in a phase-response curve (see €1. 48) relating the amount of phase-shift to the time of stimulus. Phase-response curves are characteristic for the species and conditions, and imply certain important features of the underlying driving oscillator or “clock” (see p. 48). The use of these terms is due to the fact that the formal characteristics of circadian rhythms are closely analagous to those of physical oscillators. It is dangerous to follow such analogies too far, however, and the terms are in general used as convenient shorthand only. One other feature of circadian rhythms that marks them off from other bio-rhythms is their stability at different temperatures. Unlike nearly all other physiological processes, circadian rhythms exhibit Qlos which are typically close to unity. They are thus temperature compensated within normal biological limits (see p. 72). To summarize, circadian rhythms have three prime characteristics: (1) they persist in the absence of external time cues; (2) they persist at a period which is, in principle, never exactly 24 h hut alGays fairly close to it; (3) they have Qlos of approximately 1.0.

6

JOHN BRADY

3 Types of insect circadian rhythms 3.1

BEHAVIOURAL RHYTHMS

3.1.1

Locomotor activity

Locomotor activity has been a favourite parameter for the measurement of insect circadian rhythms, no doubt mainly because it is so simple to .record (e.g. via a thread attaching the insect to a pen writing on a smoked drum, Harker, 1956). But it has a less pragmatic advantage, too, since it is performed as part of a number of behaviour patterns and may therefore be taken as the integrated response t o several different endogenous stimuli modulated by the underlying clock. A whole battery of techniques for recording insect movements have been developed: the archetypal rocking box (Szymanski, 1914), running wheel (Roberts, 1960), photocell/light-beam (Brown and Unwin, 1961), capacitance transducer (Schechter et ul., 1963), sound recording (Jones, 1964), flight mill (Chambers and O’Connell, 1969), temperature differentials (Macaulay, 1972), and their modifications, among many others. Some of these measure whole body movement (e.g. Brown and Unwin) and thereby simplify the recording of the behaviour involved in locomotion to a question of movement or no movement. Others (e.g. Roberts’s running wheel) are more selective and record only walking (or flying) movements. This has the advantage of reducing noise in the record, but the disadvantage of missing parts of the activity, such as feeding, that may be an important component in the rhythmic expression of behaviour. The assumption behind all these studies is that the daily rhythmicity of locomotor activity is the result of a response to the insect’s internal clock. In so far as the activity is periodic and circadian, this is evidently true, but because locomotion is behaviour, it is liable to non-rhythmic interferences from many sources. Insects respond t o the environment provided by their actograph in unpredictable ways. Thus, the cockroach, Penplaneta americana, placed in the small cage of a Brown and Unwin or rocking box recorder may show gradually declining activity peaks over a series of days until the rhythm is apparently lost, even in LD. Yet when transferred to a running wheel such an insect immediately reverts t o showing a very clear rhythm (Brady, 1967b). The cricket, Achetu domesticus, on the other hand, continues to show a clear rhythm for weeks in small rocking boxes (Nowosielski and Patton, 1963). The submergence of the measured phase of the rhythm into background noise has led t o confusion in the interpretation of some experiments (see Brady, 1967b: 159). Single phase, sharp onset rhythms of the cockroach or cricket type are perhaps the most convenient for circadian research, but by no means the only form of expression of insect circadian locomotor activity. Bimodal

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

7

versions commonly occur, though these frequently appear only in entrained die1 rhythms (e.g. Gillett et al., 1962; Roberts, 1962; Jones et a/.. 1967) with the second peak presumably forced by the light-on signal. The relationship between bimodal behaviour, “dawn” and “dusk” signals, duration of light phase, and total activity has been analysed for Anopheles gambiae by Jones et al. (1972b); the seccmd peak in this rhythm is evidently only partly explicable as a kinetic response to light-on. In some cases, both peaks are truly endogenous and free run in constant conditions (Chiba, 1964; Nayar and Sauerman, 1971). In cockroaches, for some reason, this occurs more commonly in LL than in DD (Roberts, 1960; Lohmann, 1967). N o cases have yet been reported in insects of such bimodal rhythms dissociating from each other as has been shown to occur in some mammals (Hoffmann, 1971). A rather different form of circadian activity pattern is shown by some Diptera. The tsetse fly, Glossina morsitans, restricts its locomotion to brief 1-min flights separated by long intervals (Brady, 1970). Throughout a 12-h photophase less than eight of these flights may be performed, and yet they are distributed in a circadian manner in constant darkness (Brady, 1972a). This is only evident from a study of the mean ,activity of several tsetse flies. The frequency of flight falls along a U-shaped course through the light phase of a LD cycle (see p. 14), and in a weakly bimodal pattern in DD. The mean total flight duration of an individual fly in LD is 12.5 min in a 12-h photophase, performed in about 17 buists spread through the day, though mostly occurring before noon. In DD the activity of an individual is barely recognizable as a rhythm, and in LL, not at all (Brady. 1972a). Activity patterns similar to the tsetse type also occur in other species (Parker, 1962; Green, 1964a;Jones et al.. 1967). A variant on the simpler locomotor activity rhythms is provided by the granary weevil, Sitophilus granarius (Callant!ra granaria). This shows a circadian rhythm in positive and negative geotaxis, but only does so in unfavourable culture conditions (Birukow, 1961). This implies some sort of migratory response to environmental stress. A circadian component in what is in effect the migratory activity of the milkweed bug, Oncopeltus fasciatus, has also been demonstrated and shown to be a response to photoperiod-as the days lengthen the bugs fly more and get carried north on the prevailing winds (Caldwell and Dingle, 1967; Dingle, 1968). Activity in that classically migratory species, the locust, Schistocerca pegaria, on the other hand, appears not to be circadian (O(dhiambo, 1966), at least not in long-established laboratory cultures, although in Locusta mipatoria it apparently is (Edney, 1937). The relationship between locomotor rhythms and insect migration has been discussed at length by Johnson (1969) and Dingle (19 7 2). No doubt some insects are not rhythmic in 1 heir locomotor activity, but

8

JOHN BRADY

the monumental studies of Lewis and Taylor (1964) on 500 species of night flying insects in England, and Haddow et al. (1961 ) in Africa, suggest that enormous numbers of species are. 3.1.2 Feeding rhythms Historically, the most obvious of all insect rhythms must h'ive heen the feeding activity of blood-sucking Diptera, and yet this rhythm has never been examined in the laboratory. The precision of the timing of feeding by some mosquitoes in the field, e.g. Tueniorhynchus (Mansonia) fuscopoznuta (Haddow et al., 1961: 319), suggests that this might be a rewarding field for circadian study now that artificial feeding and recording techniques arc available (e.g. Kashin, 1966; Galun and Margalit, 1969). The nearest approach has been through the observations of Gillett et al. (1962) o n populations of Aedes aegypti in which the sugar-feeding rhythm persists for at least three days in DD. The inseparability of spontaneous locomotor activity from feeding activity might have been inferred from the relation between the intensity of activity and the degree of starvation of an insect (e.g. Ellis and Hoyle, 1954; Green, 1964a, 1964b; Brady, 1972a). Only rarely have feeding and locomotion been examined independently, but simultaneously in a single insect. In Periplaneta, however, Lipton and Sutherland (1970) found that, with only one possible exception (their Fig. l o ) , locomotor activity and feeding were effectively synchronous in both LD and DD. Somewhat surprisingly, the feeding rhythm of Oncopeltus only persists for 1 or 2 cycles in LL and not at all in DD, even though the oviposition rhythm continues unabated for at least 6 cycles in constant conditions (Caldwell and Dingle, 1967; though see p. 12). One might have expected feeding to be sufficiently modulated by oviposition activity to make it, too, appear rhythmic. In the mosquito, Anopheles gambiae, which is in a different situation from Oncopeltus because it normally has to fly some distance to oviposit, a feeding rhythm would appear to be inseparable from the oviposition rhythm (Haddow and Ssenkubuge, 1962). In this case, however, it looks as if it is oviposition which is the endogenously timed activity, since it is the oviposition, rather than the feeding, which coincides with the endogenous l&omotor peak (Jones et al., 1967). An unusual form of feeding rhythm has been demonstrated in the ant-lion, Myrmelcon obscurus. This insect digs its pit traps with a clear circadian rhythm phased to dusk, when maximum feeding occurs (Youthed and Moran, 1969a). The unexplained feature of this rhythm is that it is apparently modulated into both lunar-monthly and, possibly, lunar-day rhythms even in constant conditions in DD (Youthed and Moran, 1969b). What advantage the ant-lion gains from this lunar periodicity is not clear, since prey availability apparently shows no obvious lunar periodicity.

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

9

What might be termed a post-feeding rhythm has been demonstrated in the tick, Huemuphysalis leporispalustris, by George (1964). This is a parasite of the rabbit, to which it is attached for 5-8 days each instar while it feeds. Having completed engorgement it drops off its host, but only does so during daylight when the rabbit, which ii nocturnal, is resting in its “form”. The tick thereby ensures that it is likely to find its host again once it has completed moulting. Detecting an endogenous rhythm in a parasite creates obvious difficulties, but by ringing the changes on LD, LL and DD light cycles independently on rabbit and tick George (1971) produced strong circumstantial evidence that the drop-off rhythm is primarily endogenous and circadian, probably entrained by a “dusk” zeitgeber, though modulated under some circumstances by the metabolic cycle of the host. What are probably the only examples of learned feeding rhythms in insects occur in the Hymenoptera. Virtually all the research has been done on the honey bee, Apis mellifera, and related species, though ants, too, may possibly be able to learn feeding times (see Wilson, 1971). The adaptive value for bees in being able to feed at specific sites at specific times of day relates to the fact that flowers are rhythmic in their secretion of nectar and presentation of pollen (see Renner, 1960). Bees have an extremely refined ability to time their foraging, and can be taught (in the field) to come to artificial food sources at up to nine different times per day (Koltermann, 1971). They will even remember to come to the correct scent at the correct time, if trained to two scents at two different tmes. This very precise timing ability is evidently a function of their underlying circadian clock. They ignore local time and continue t o forage at their correct circadian time when translocated ihrough several time zones (Renner, 1957), have a free-running rhythm of feeding in LL (with a period of c. 23.4 h), and cannot be trained to feed at cycles which differ by more than 2 h from the 24 (Bcier, 1968; their rhythm of locomotor activity is also circadian, with a free-running period of c. 22 h in DD, Spangler, 1972). As might be expected, the clock mechanism which underlies this ability to feed at specific times is also involved in the time compensation of their sun-compass orientation (Beier and Lindauer, 1970). How bees manage to couple their foraging behaviour to this clock in a manner which permits them to divide the day into as many as nine segments is completely unknown, but clearly indicates a more sophisticated system than that possessed by any other organism. 3.1.3 Sexuul rhythms Sexual behaviour in insects is frequently observea in the field to be associated with certain times of day. These rhythms seem to have been investigated in the laboratory in the Diptera and Lepidoptera only. The

10

JOHN BRADY

first, and still apparently the best evidence that nematocerous mating behaviour is endogenously timed is Bates’s (1941)demonstration that Anopheles superpictus swarms at approximately 24-h intervals when kept under constant dim light. The fact that Aedes triseriatus start to swarm 2 h before light-out in the constant temperature and unchanging light intensities of a laboratory LD cycle (Wright et al., 1966) also suggests endogenous timing. Although mosquito and midge swarming has been studied on other occasions, it has usually been in relation to the light intensities at which it occurs [e.g. Nielsen and Nielsen, 1962;Chiba, 1967). It is evident from Bates’s work that these light intensities are simply the relevant zeitgebers which entrain a circadian rhythm of swarming. In the Queensland fruit fly, Dacus tryoni, mating is restricted under natural daylight to about 30 min around dusk. In the laboratory, an instantaneous “dusk” from 10 000 lux to 10 lux (the optimum intensity for mating under natural conditions) elicited a higher mating response than occurred in field cages (Tychsen and Fletcher, 1971).On the face of it, this rhythm looks like a direct response t o an environmental signal, but Tychsen. and Fletcher found that underlying it is a relatively smooth-wave, endogenous, circadian rhythm of mating responsiveness. In LD, the percentage of both sexes “ready to mate” rises sharply over the last 4 h before “dusk” from zero during most of the photophase t o a peak of 80 per cent at “dusk”, then falls steadily through the night. In males, this rhythm free runs in constant dim light for at least 4 days, but in both sexes damps out rapidly in constant bright light. This loss of rhythm is probably the result of the individuals becoming aperiodic and not because of asynchrony in the population. The courtship rhythm of male Drosophila fruit flies is also extinquished by constant light (50lux), though it persists in DD (Hardeland and Stange, 1971). The most extensively documented circadian sexual rhythms in insects are those associated with the production of and response t o pheromones (Fig. 2). The males of four species of noctuid moths have been shown to vary their responsiveness to a standard female pheromone stimulus along a marked circadian cycle (Shorey and Gaston, 1965). In the case of the cabbage looper moth, Trichoplusia ni, in LD, this vaned from a near zero response during the day $0 a broad peak with 80 per cent of the males responding during the night; in DD the amplitude of the peak was only slightly less. A similar situation prevails in male flour moths, Anagasta kuhnielln, though in this species the amplitude of the first peak in dim LI, was scarcely significant statistically (Traynier, 1970). Varying the light and dark components in a 24-h cycle suggested that the male T. ni response rhythm is not specifically phased t o either “dawn” or “dusk”. By contrast, the rhythm of the pheromone release by the female is phase set by the DL transition (Sower et al., 1971).It has a much narrower

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

11

Fig. 2. Effect of circadian rhythm and light cycle on the electroantennogram, EAG ).( and behavioural responsiveness (0) of male Trichoplrcsia ni to female sex pheromone. Ordinate, amount of pheromone required to elicit a significant EAG or a 50 per cent behavioural response. Points from last 12 h plotted twice. (Redrawn from Payne et al., 1970.)

peak than the male’s response in a LD cycle with near zero “calling” from dawn to midnight and a sharp rise to 30 per cent of females “calling” between midnight and dawn. In a population of females, it free runs for at least three cycles in DD, with little loss in amplitude (Sower et al., 1970). The only non-lepidopterous species from which a pheromone release rhythm has been reported (Marsh, 1972) seems to be the aphid, Megoura viciae. From the physiological point of view, an interesting aspect of the T. ni male’s response is the way it is affected by light (Fig. 2). The antennal response to female pheromone, as measured by electro-antennogram, is dose dependent. But, whereas the behaviourid response is modulated by light intensity and time of day, the antennal response is quite unaffected (Payne et al., 1970). Evidently, the response rhythm is a consequence of central and not peripheral modulation. Although spontaneous activity was not measured in the T. ni observations, closely similar work on the apple moth, Epiphyas postvittana. shows that locomotor activity and pheromone responsiveness are probably exactly synchronous (Bartell and Shorey, 1969). The only investigation of the physiological control of rhythmic pheromone release has been on female silkmoths (Riddiford and Williams, 1971). In Hyalophora cecropia kept under LD conditions, calling begins shortly before “dawn” and continues for about the first 30 min of the light phase. Since calling starts before the lights go on, this rhythm may be presumed to be endogenous and circadian, as in other moths. Riddiford and Williams found that the percentage of females calling before “dawn” was unaffected by the removal of their corpora allata, whereas removal of the

12

JOHN BRADY

corpora cardiaca reduced calling more than four-fold. Re-implantation of up to three pairs of corpora cardiaca, however, did not restore the calling response to the ligKt cycle, nor did calling occur after the nervous connection had been cut between the brain and corpora cardiaca of otherwise intact moths. It appears, therefore, that a corpus cardiacum hormone may be involved in the release of calling behaviour, and that hormone secretion is under nervous control from the brain. (See Chapter 4 by Truman and Riddiford.) Other rhythmic sexual activities that have been shown t o be endogenously timed are singing in grasshoppers (Dumortier, 1968; Loher, 1972), and mating in Oncopeltus (Caldwell and Dingle, 1967), though this latter rhythm damps out very quickly in LL and DD. It seems likely that related diel sexual rhythms, such as spermatophore production in crickets (Mcfarlane, 1968) and glowing by female Lampyris noctiluca (Dreisig, 1971), will also prove to be endogenous and circadian, though they ‘have not yet been tested for this. Female reproductive behaviour after mating is also endogenously rhythmic. Thus the oviposition rhythm of the pink bollworm, Pectinophora gossypiella (Minis, 1965; see p. 28), the grasshopper, Chorthippus curtipennis (Loher and Chandrashekaran, 1970), and the mosquito, Aedes aegypti (Gillett et al., 1961), and the larviposition rhythm of the viviparous tsetse fly (Phelps and Jackson, 1971), are all clearly circadian. No doubt the diel oviposition rhythms of other species such as Anopheles gambiae (Haddow and Ssenkubuge, 1962), Drosophila (Rensing and Hardeland, 1967), and the spider mite, Tetranychus urticae (though not retained in LL, Polcik et al., 1965), will also prove to be circadian. On the other hand, the oviposition rhythm of Oncopeltus fasciatus which is retained in LL, is apparently not circadian, since it is not temperature compensated (Rankin et al., 1972; see p. 72). A unique case of the extinction of an activity rhythm following mating has been reported by McCluskey and Carter (1969). Virgin females of the ant, Pogonomymex californicus, show a clear diel rhythm of activity in LD. When mated they shed their wings, become photonegative, start to lay eggs and cease to show any detectable periodicity. The switch is apparently the result of mating itielf, and not the loss of their wings or the act of nuptial flight. It is not known whether this loss of behavioural rhythmicity indicates a stopping of the clock itself, or merely an uncoupling of the behaviour from it, but in either case it has interesting implications for circadian control.

3.1.4 Changes in responsiveness The behavioural responses of insects to various stimuli have been studied in relation to endogenous input from the physiological state of the animal

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

13

(age, starvation, etc.) but only rarely in relation to circadian time. Although in recent years behavioural studies have often made allowance for possible diel variations, by restricting the observations to standard times of day, the implication behind this procedure, namely that insects modulate their responsiveness across circadian time, has usually not been examined. A few cases have been reported, however. The pheromone response rhythms of male moths is one example which has already been mentioned (p. 10); it ha:, been shown to persist in constant conditions (Shorey and Gaston, 1965). An earlier example is that of Diirrwachter (1957) who found a persistent daily cycle of phototaxis in adult Drosophila maintained in DD except during test (it appears that this might more correctly be called a photokinetic rhythm). Dilrrwgchter showed maximum responsiveness to occur at noon and around midnight, but the phase relations of the pre-conditionin5 LD cycle are not given, so these times may not represent true circadian time. A diel rhythm in phototactic responsiveness which persists in L1, has also been demonstrated in water boatmen, Corixa and Anticorixa (Retising, 1965a), see Fig. 3. The only case where a free-running rhythm has been demonstrated in a phototactic response is in the water flea, Daphnia nzagna (Ringelberg and Servaas, 1971). Here the period is unusually long, c. 28 h (see also Harris, 1963), so that the drift away from solar time is demonstrable within a few days. The tsetse fly, Clossina morsitans, likewise shows a marked diel rhythm in the threshold of‘ its response to a standard visual stimulus (Brady, 1972b). The responsiveness declines from “dawn” t o noon and then rises from noon to “dusk” (Fig. 3). Since this cycle occurs even though environmental conditions remain constant throughout the 12-h photophase, and since it closely parallels the locomotor rhythm which is known to persist in DD (Brady, 1972a; see Fig. 3), the visual response may, like that of Corixa, be aswmed to be truly circadian. There is a parallel situation to these daily changes in responsiveness to standard stimuli, in the orientation behaviou- of some insects. The pond skater, Veelia currens, for example, has a diel rhythm in the laboratory in its angle of phototaxis towards a light source. It approaches with its left side leading till noon, head-on at noon, and with its right side leading after noon, reversing back to the initial position during the night (Birukow, 1960). This rhythm is apparently not circadian, however, since it does not persist in DD or LL and is entrainable to a LO-h LD cycle (Birukow and Busch, 1957). The cabbage white caterpillar, .Pieris brassicae, has a similar diel phototactic rhythm (Birukow, 1966), and the dung beetle, Geotrupes sylvaticus, keeps an approximately constant compass”orientation across the day when exposed to the natural sky or to polarized light in the laboratory (Birukow, 1960). The relationship between this type of daily

JOHN BRADY

14

/

7

1

1

1

I

1

'1

I

0

Fig. 3. (a) Behavioural changes of male tsetse flies (Glossinn morsitnns) across a 12-h photophase of L D 12 : 12. 0, Kinetic responsiveness to a slowly moving visual target; 0, intensity of orientation to same target (redrawn from Brady, 1972b); half circles, percentage of mean probiffg response (redrawn from Brady, 1973); m, proportion of population flying spontaneously in actographs (redrawn from Brady, 1972a); m, hourly catch of male flies in the field on a bait ox (redrawn from Dean et al., 1969). All curves smoothed with three-point sliding means. (b) Behavioural changes of water boatmen (Anticorixa sahlbergi) across a 12-h photophase of LD 12 : 12. A, Spontaneous locomotor activity (arbitrary units); 4 phototactic (= kinetic?) responsiveness (per cent in light part of choice chamber) (redrawn from Rensing, 1965a).

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

15

modulated orientation behaviour and insect migration is discussed by Johnson (1969: 163). A highly specialized form of orientation rhythm has recently been revealed in bees. Returning foragers show a consistent cycle of error, Missweisung, in the angle of their waggle dance. The Missweisung has been shown to be a response t o the diel rhythm of the earth's magnetic field (Lindauer and Martin, 1972). This is, however, not an endogenous circadian rhythm, since it disappears in a constant magnetic field. In the tsetse fly, the rhythms of visual responsiveness, probing responsiveness, spontaneous activity, and feeding in the field are all synchronous (Brady, 1972a, 1972b; 1973; see Fig. 3(a)). Similarly, in male moths pheromone responsiveness and activity are synchronous (Bartell and Shorey, 1969), so also are the phototactic and activity rhythms of water boatmen (Rensing, 1965a; see Fig. 3(b)), and. the oviposition and activity peaks of the mosquito, Anopheles gambiat (Haddow and Ssenkubuge, 1962; Jones et al.. 1967). These coincidences suggest that coupled to the circadian clock there may be some general arousal system which modulates behaviour on a circadian basis. This may be true for much of insect behaviour (see p. 65)-the possibility has never been examined-but it is not true for all their responses. In the tsetse fly, for example, the intensity of orientation towards a moving visual target appears to be quite unchanging across a 12-h photophase (Brady, 1972b; see Fig. 3(a)), and results with a similar implication have been reported for the blowfly, Lucilia sen'cata, in which Goodman (1960) found no change in the intensity of the landing response during a 5-h test period. Furthermoi~,some behavioural rhythms seem specifically timed to be out of phase with each other so as not to conflict (Dingle, 1972).

3.2

DEVELOPMENTAL RHYTHMS

3.2.1 Hatching rhythms Even before the act of oviposition, the development of eggs within the female is apparently rhythmic (Dutkowski and Cymborowski, 1971), with a diel rhythm of RNA synthesis occurring in the follicular epithelium of crickets. Thereafter, since other developmental events are clearly gated by circadian clocks, it might have been expected that egg-hatch would be also. The possibility seems to have been examined thoroughly only in the pink bollworm, Pectinophora gossypiella. In this species, a circadian rhythm of hatch from a population of eggs occurs in LD, OF in DD after a single 15-min light pulse or a 12-h temperature pulse, but not in LL (Minis and Pittendrigh, 1968). No hatch rhythm is initiable until the twelfth day of embryogenesis, when the first cephalic pigmentation occurs. The inference

16

JOHN BRADY

that rhythm initiation requires this pigment for photoreceptive coupling to the environment is not supported by the fact that a temperature signal can equally initiate the rhythm at this time. It appears more probable that the pigmentation coincides with some essential link-up in the CNS (Minis and Pittendrigh, 1968). In Aedes mosquitoes, hatching occurs as a direct response to environmental amelioration, related only to the effects of temperature on embryogenesis and the presence of water after some sort of delayed developmental period (Gillett, 1955), and unrelated to any rhythm (Corbet, 1966). Pre-conditioned Aedes taeniorhynchus eggs hatch at any time of day within 15 min of emersion in de-oxygenated water (Nayar, 1967b). Rensing (1965b) lists references on the subject of insect hatch rhythms. 3.2.2 Pupation rhythms Larval ecdysis in Aedes tueniorhynchus is not gated by a circadian clock, since, even in a LD 12 : 12 cycle, it occurs at quite different times of day, according to the temperature. Thus, the interval between the first and second ecdyses is c. 15 h at 32°C and c. 32 h at 22’C, with peak second ecdyses occurring at “dusk” and in the forenoon, respectively (Nayar, 1967b, Fig. 6). Pupation in this species, on the other hand, clearly is gated by a circadian clock. On a rich diet, the fourth instar larvae from a single day’s egg hatch all ecdyse to become pupae in a single peak. If the diet is restricted, however, the larvae do not all complete development by this first gate, and the pupation is spread out over several days. Under these conditions, pupation does not take place at random but occurs in a series of broad peaks covering the same times on successive days in either LD (Provost and Lum, 1967) or DD (Nayar, 1967a, 1967b). Evidently, larvae that fail to complete development by the first gate do not pupate immediately they‘ complete development, but wait until the gate in the next circadian cycle. An unexpected feature of this rhythm is that in LD 12 : 12 it has a period of c. 22.3 h, implying that it is not entrained by the light cycle. Furthermore, the later emergence peaks are free of any skewness which might imply that the delayed larvae were taking the earliest opportunity to get through the gate and were thereby causing an apparent shift in its timing. On the other hand, a typical phase-response curve resulted from light pulses given at different times in constant darkness (Nayar, 1968), so that the rhythm is to some extent normally entrainable by light signals. Environmental stress from starvation, crowding, or high salinity (conditions which commonly occur in their natural habitat) not only resulted in an extension of larval development so that up to seven pupation peaks were observed from a single day’s hatch of eggs, but also resulted in LD

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

17

effectively entraining the rhythm so that its period became indistinguishable from 24 h (Nayar, 1967b). The light intensity used in these experiments was not specified, but was provided by two 100 W flourescent tubes (Provost and Lum, 1967). Perhaps the apparent lack of entrainment was because the strain of Aedes tueniorhynchus used was adapted to the light intensities of the Florida salt marshes (whence it apparently came), and these would be several log units brighter than those used in the laboratory. In a comparative study of 15 other species of Florida mosquitoes, Nayar and Sauerman (1970) found clear signs of die1 pupation rhythms in only four: apparently such developmental rhythms are not universal in mosquitoes. 3.2.3 Adult eclosion rhythms Owing primarily to the work of Pittendrigh and his colleagues, the formal characteristics of the Drosophih eclosion rhythm are better understood than those of any other circadian system. Knowledge of its physiological control, however, lags behind the knowledge of its entrainment and mathematical configuration. The field is too wide to have justice done to it here, and for fuller details the reader is referred to earlier accounts, e.g. Brett (1955) Pittendrigh and Minis (1964j, Pittendrigh (1965, 1966); unfortunately, no more up-to-date review exists. In outline, the Drosophilu eclosion story is as follows. In LD 12 : 12, adults emerge from the pupa during the forenoon, starting sharply at dawn, and with a distribution skewed strongly to the left, implying that darkness inhibits eclosion (Pittendrigh, 1954). When reared from the egg in continuous darkness they emerge at random with respect to time of day, and it can be demonstrated that this arrhythmicity is due to the individuals’ clocks being stopped, and not to the members of the population being out of phase with each other (Zimmerman, 1969). If, however, such DD pupae are given a single short light pulse (1 min is enough for D. melunoguster, Brett, 1955), subsequent eclosions occur in a rhythmic manner, in D. pseudoobscura restricted to daily peaks about 6 h wide. Moreover, such a light stimulus can be given as far back in development as the first larval instar and still produce the same rhythmicity (Brett, 1955; Zimmerman and Ives, 1971). A similar rhythm is initiated by transferring arrhythmic DD-reared pupae to LL, but this rhythm rapidly damps out (Chandrashekaran and Loher, 1969). In the converse situation, with Drosophilcr reared from the egg in LL, adult emergence is also aperiodic. In this case, a rhythm may be initiated by transferring the culture to DD, whereupon subsequent eclosions occur through 6-h gates with a circadian rhythm phased to”15 h after the onset of darkness (i.e. with peaks at 15 + 24n h after the LL-DD transition, where n = a whole number of days). A rhythm initiated in this way has been the

18

JOHN BRADY

standard material for assaying the Drosophila clock (e.g. Pittendrigh, 1966; Winfree, 1970a, 1970b). It is, for example, very sensitive to re-entrainment, and shows a typical phase-response curve, with maximum phase shifts of up t o 8 h, t o light signals as brief as 0.5 ms (p. 48). Transferring a population of pupae from light to DD always results in the rhythm’s reappearance with its 15 + 24n phase, regardless of how long the light exposure lasted, provided it was more than 12 h. Pittendrigh (1966) therefore concluded that the clock actually stops after 12 h in light, and is re-started from a standard phase point by the onset of darkness. He produced further support for this contention by showing that the phase-response curve to 15-min light interruptions after transfer to darkness bears exactly the same phase angle to the onset of darkness as it does after a LL-DD transfer (1966, Fig. 5 ; see Fig. 10). There are problems for the LL-stopped clock interpretation, however, raised by the work of Chandrashekaran and Loher (1969), who found it possible to initiate a rhythm in arrhythmic pupae reared in LL (of up to 300 lux) by transferring them to still brighter LL (of up to 3000 lux). Although the gate widths of these rhythms were much wider than the standard 6-h gate occurring in DD, clear peaks did survive for at least three cycles at the higher intensity before damping out. As there was no steady-state persistent rhythm to measure, the effect of the LL low to LL high transition on the underlying driving oscillator is impossible t o assess by the phase-response method. Nevertheless, the initiation of a rhythm during constant light, albeit only noisily and briefly, must suggest that the apparently very well-founded conclusion that the Drosophila clock is stopped by 12 h of light is not the whole truth. The following additional aspects of the Drosophila emergence rhythm are considered under the relevant sections below: its entrainment and phase-responses (p. 48); its photoreception (p. 47); its responses t o temperature (p. 73); its genetic control (p. 75). There has been a controversy over whether it is legitimate to infer the control of gated events in individuals from the behaviour of a population. This has centred round the analysis of Drosophila eclosion rhythms, which, of course, only appear as “rhythms” in populations of pupae. Harker (1964, 1965a, 1965b) has objected to the use of population data to explore this phenomenon on the grounds that individuals do not in fact develop synchronously and that a rhythm appears in a population only as a consequence of coincident emergences by pupae of differing chronological ages. If this were so, it would offer some difficulties for Pittendrigh’s two-oscillator model (p. 49). However, Skopik and Pittendrigh (1967) and Pittendrigh and Skopik (1970) have convincingly demonstrated that Harker’s criticisms are based on a false premise. Within the relevant

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

19

statistical limits, individual flies behave identically to populations as far as the processes leading up to the gating of their eclosion are concerned. At least for Drosophila melanogaster, pscudoobscura, and victoria, population data may therefore legitimately be used to infer the characteristics of the docks of individuals. Whenever it has been critically examined in other species, adult eclosion has frequently, though not always (see below), proved to be a circadian rhythm. It has, for example, been shown tl3 be circadian in tsetse flies (Dean et al., 1968) and in several species of moth (Truman, 1972a). For a full review of such rhythms see Kemmert (1962), and for a summary, Rensing (1965b). A special case of rhythmic eclosion is the circadian-lunar-tidalemergence rhythm of the midge, Clunio marinus, which has been studied extensively by Neumann. This species spends its larval and pupal life among red algae between the tidemarks of certain European beaches. Its adults emerge at low tide and live for only 2 h, during which they must mate and oviposit before the tide advances back over their breeding ground. The species exists in a number of geographical races whose €mergence rhythms are very precisely adapted to their local tidal conditia ns and which are genetically distinct in this respect. Thus the Basque coasi. race has peak emergence (in the laboratory, as in the field) at 1 8 3 0 h (local time), whereas the Normandy race emerges around 1400 h. When these two races are crossed, the F, hybrids have peak emergence at 1700 h , and the back-cross with the Normandy race, around 1600 h, suggesting that the daily emergence time is genetically controlled by the inheritance of 3 few quantitative characters (Neumann. 1967). Some races show partial cross-sterility (Neumann, 1971a). In the southern races this daily periodicity is circadian and free runs in constant light (Pfhiger and Neumann, 1971). In the arctic, where the larvae inhabit the mean intertidal zones, the emergence is strictly tidal (12.4 h); in the laboratory under LD 16 : 8, emergence occurred daily, 10 h after “dawn”; and in LL or DD no rhythm occurred. The authors suggest that this race has lost its circadian coupling and times its emergence instead by an “hourglass” measuring 10 h from the time of first exposure on the previous ebb tide. In many of the European races of Clunio, the situation is complicated by the fact that the larvae exist below mean lowwater; their breeding sites are thus exposed only for a few days at the spring tides. They allow for this by gating their emergence rhythm to the 15-day (actually 14.7-day) semi-lunar tide cycle, and in a LD cycle in the laboratory emerge with a c. 15-day periodicity. This semi-lunar rhythm is entrained differently according to race. Thus the southern European (Spain and France) races will entrain, in a laboratory LD cycle, if every 30 days they are given four consecutive

20

JOHN BRADY

nights containing a dim light pulse simulating moonlight. Further north, however, this zeitgeber is not sufficient to entrain the rhythm, and in order to entrain in the laboratory the Helgoland race requires a specific phase relation between a mechanically simulated tidal cycle and the LD cycle (Neumann, 1968, 1971b). It is supposed that this is because summer nights are sufficiently light in the north to make moonlight an unreliable zeitgeber . Neuniann suggests that these races thus contain two clocks: a circadian one and a semi-lunar one. Whereas the circadian clock seems indisputable, the existence of an endogenous lunar periodicity is not so clear, since the 15-day rhythm does not occur unless the animals are reared in 24-h LD cycles. The gating of emergence to 15-day intervals over at least three semi-lunar cycles could therefore indicate either the existence of a semi-lunar clock, or the ability to count 15 days from one gate to the next, which may not be quite the same thing. In the salt marsh mosquito, Aedes taeniorhynchus, the adults emerge in LD with a die1 rhythm. It is not eclosion that is gated, however, but pupation (see p. 16). Eclosion occurs a fixed period after pupation (Provost and Lum, 1967), the duration of which is directly proportional to temperature between 16°C and 32°C (Nielsen and Evans, 1960). The rate of pupal development has a Qlo of c. 2.5. No alteration in the time of eclosion results from transferring larvae, as they pupate in DD, into LD or DL (Nayar, 1967b). 3.2.4 Daily growth layers A previously unsuspected form of developmental rhythm was first reported by Neville in 1963. This takes the form of daily growth layers laid down in the endocuticle during early adult life. During the night chitin crystallites are deposited in the endocuticle in organized lamellae; during the day the same amount of chitin is deposited, but in non-lamellate form. Examined in section under crossed polaroids the lamellate layers are strongly birefringent, and the daily growth layers thereby detectable as pairs of alternating light and dark bands. In locusts this rhythm of endocuticle organization-in reality a secretory rhythm by the epidermal cells-persists with a circadian rhythm for at least 2 weeks in constant darkness. It is, moreover, very nearly perfectly temperature compensated with a Qlo of 1.04 for the frequency (Neville, 1965). The same rhythm in the milkweed bug, Oncopeltus fasciatus, shows the same properties and the same precise temperature compensation (Dingle et al., 1969). This subject has been reviewed by Neville (1967a, 1970), and it seems that the rhythmic organization of cuticle is almost universal in insects. He lists (1970) the species in which it has been found: Orthoptera (16 spp.), Dictyoptera ( 3 ) , Phasmida (2), Dermaptera ( l ) ,Odonata (2), Hemiptera

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

21

(3), Hymenoptera ( l ) ,Coleoptera (5); and to these may be added Diptera (c. 9 spp.) (Schlein and Gratz, 1972). Only in the Coleoptera are the layers not laid down in a circadian rhythm, nor I S their rhythm temperature compensated, so that it appears that true circadian periodicity, or coupling, has been lost by the epidermal cells of beetles (Zelazny and Neville, 1972). The control of the chitin deposition rhythm has some important implications for the study of insect circadian rhythms. In the locust, though not apparently in Penplaneta, Hydrocyrius (Reville, 1965), or Oncopeltus (Dingle et al., 1969), constant light damps out the rhythm in much of the epidermis, e.g. the hind tibiae, so that after a few days (3-6 days at 1 lux, 1 day at > l o 0 lux) the chitin is laid down without visible lamellae. This effect is not mediated via normal photoreceptors, since it still occurs when the eyes and ocelli are all cauterized, and the head capsule opaqued with black paint. Furthermore, hind tibiae implanted into the abdominal haemocoels of locusts kept in LD continue to produce daily layers, whereas those implanted into locusts in LL do not (Neville, 1967b). Evidently, the epidermis does not require nervous connections, either to maintain its rhythm or to damp it out. Neville also showed (1967b) that if locusts having one hind tibia opaqued with paint are kept in constant dim light (50 lux) for 8 days, the exposed tibiae show no daily layers after the first 48 h, but the opaqued tibia shows eight normar circadian lamellations. He inferred that the cessation of the rhythm was therefore not mediated hormonally via the haemolymph. This may well be true, particularly since pupae of the butterfly, Calpodes ethlius, continue to deposit lamellate cuticle if decapitated prior to ecdysis (Locke et al., 1965). Strictly speaking, however, Neville's experiments do not distinguish between the two possibilities: (1) that, as he suggests, the epidermal cells are themselves endogenously rhythmic and light sensitive, responding directly t o the environmental light conditions by entrainment (LD) or damping-out (LL); or (2) that the cells are not endogenously rhythmic, but that there is a hormonal rhythm in the blood from which the cells are uncoupled by the direct action of constant light. The important question is whether the cells are themselves endogenously rhythmic or not. That they may be is implied by observations of another kind. In Oncopeltus, Dingle e t al. (1969) illustrate one tibia1 section which shows some sectors with 10 bands of lamellate endocuticle and others with only 5 . They note, moreover, that the rings sometimes tend to coalesce and are sometimes incomplete. While by no meims conclusive, this sort of condition would be easier to understand if the epidermal cells had considerable rhythmic autonomy, though interacting locally, rather than if the whole epidermis were driven by a central hormonal clock. That there are blood-borne factors regulating the amount of endo-

JOHN BRADY

22

cuticular growth has been demonstrated several times, e.g. by Schlein (1972a, 1972b) in the blowfly, Surcophaga falculutu. While Schlein’s results have no direct bearing on the question of circadian organization of cuticle deposition, they imply that these growth-regulating hormones are unlikely to be involved, since he found one factor promoting growth that is present in the blood around the time of emergence only, and another factor inhibiting growth that is present continuously in the blood from the completion of tanning onwards. It is difficult to see how such a system could operate rhythmically. 3.2.5 Photoperiodism The role of circadian rhythms in the measurement of seasonal photoperiodic change by insects has been,fully reviewed by Lees (1968, 1972) and is mentioned here only for the sake of completeness. It will suffice to point out that the evidence for widespread involvement of rhythms in insect photoperiodism is not as strong as is usually assumed. In at least one case, morph determination in the aphid, Megouru viciue, circadian rhythms are certainly not involved (Lees, 1971). In another case, diapause induction in Pectinophora gossypiella, the action spectra for photoperiod measurement and circadian rhythm entrainment are different, and photoperiodic induction can occur at a stage in development before the circadian clock appears to be functional (Pittendrigh and Minis, 1971: 238). Nevertheless, other cases, notably that of diapause induction in the parasitic wasp, Nasonzi vitripennis, exhibit striking evidence of being rhythmically determined (Saunders, 1970), as is the case in some plants and birds. As Lees suggests (1972), it seems likely that when more photoperiodic functions have been critically examined in insects they will not all prove t o be measured by the same, circadian, clock mechanism. 3.3

PHYSIOLOGICAL RHYTHMS

More and more physiological functions in insects are being shown to fluctuate with die1 rhythmicity. Although very few of these rhythms have also been shown to free run in constant conditions, it is probable that many of them are coupled t o the underlying circadian control of the physiology of the insect and are therefore likely to exhibit circadian characteristics. Some of these rhythms have been discovered when it has been necessary to produce standard insect material for a particular physiological study, and initial trials have yielded inconsistent results at different times of day. In such cases, they have not been studied for the sake of what they reveal about circadian rhythms, but for some quite other purpose, such as testing the susceptibility to insecticides. Nevertheless, these and other physiological rhythms may yield useful information about insect circadian timing

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

23

and are listed below with this in mind. Their separation into categories is for convenience of presentation but results in sm3me artifical distinctions. 3.3.1 Metabolic rhythms As might be expected, the oxygen consumption of an active insect is frequently in phase with its locomotor rhythm. Thus in LD, the grasshopper, Romalea micropera (Fingerman et al., 1958), and the cockroach, Periplanetu americana Wanda and Mrciak, 1957), have their peak oxygen consumption coinciding with peak locomotion. Similarly, in another cockroach, Blattella germanica, a major peak in consumption occurs during the first hour or two of darkness when the animal normally runs about (Beck, 1963). A later study on Per$laneta americana has shown that the oxygen consumption rhythm is maintained in constant light and is therefore truly circadian (Richards, 19693, though in this instance peak consumption seems to have extended through much of the night. No coincident measurement of locomotion was made by Richards, so it is not possible to derive with certainty the relationship between oxygen consumption and locomotor activity. In view of the fact that the oxygen rhythm disappeared after about six days in LL, it would have been interesting to know what happened to the activity rhythm, since under some circumstances this may persist for weeks in LL (Roberts, 1960). In Blattella also, the relationship between metabolic level and locomotor activity is not simple, since oxygen consumption starts at a low level after “dawn” and builds up steadily till “dusk” (Ileck, 1963), during a period when little o r no locomotion would be expected; Beck (1964) considers this curve to show an 8-h rhythm. Richards’s record for Penplaneta, which covers 13 rhythmic days, on the other hand, gives no indication of this sort of change during the light phase. A more extensive series of oxygen consumption measurements during l a n d , pupal and adult life in Drosophila melanogaster has been made by Rensing (1966a), and the phase relationship between this rhythm and that of adult eclosion (see p. 17) studied by Belcher and Brett (1973). In LD 12 : 12, the late third stage larva and pre-pupa have clear bimodal consumption rhythms with minima around “dawn” and “dusk”, and maxima around noon and midnight. In the adults the rhythm is still sharply bimodal but has the phase relationships reversed so that maxima occur at “dusk” and “dawn”, and minima in the fortmoon and early night, thus coinciding with the bimodal activity rhythm (Hardeland and Stange, 1971). In the only other dipteran in which a larval respiratory rhythm has been recorded, the mosquito, Culex pipiens, maximunf consumption occurs around subjective “dawn” and “dusk”, and persists for at least the first day in LL (Buffington, 1968). AIP--2

24

JOHN BRADY

The respiratory rhythm of adult Drosophila is susceptible to manipulation by photoperiod, so that in L D 4 : 20 the “dusk” peak nearly disappears, and in LD 20 : 4 the “dusk” peak is some 3 h, and the night-time minimum some 9 h later than in L D 1 2 : 12 (Rensing, 1966a). The bimodal adult rhythm persists for many days in DD and, less distinctly, for several days in bright LL (1000 lux, Leclerc et al., 1971). It also appears to persist in DD when the head is ligatured off at the neck,. though the phase relationships in the rhythm are then much less distinct and the level of respiration depressed (Rensing 1966a). Other rhythmic processes, apart from locomotion, must make respiratory demands and in this way get expressed overtly. There is, for example, evidence for high-frequency oxygen consumption rhythms in cockroaches, with periods of c. 0.8 h and c. 3 h (Richards and Halberg, 1964), and in the mealworm, Tenebrio molitor, of c. 5 h (Campbell, 1964). T h e shorter period rhythms (<1 h ) in the cockroach are not temperature compensated (Wilkins, 1960), however, so they, at least, can have little to do with circadian timing. Whether the other high-frequency rhythms have any circadian relevance is unknown. A specific and unexplained trimodal rhythm occurs in the very low oxygen consumption of diapausing lepidoptera hkld under LD. Thus Beck reports (1964) an 8-h rhythm in diapausing larvae of the European corn borer, Ostrinia nubilalis, and a more closely documented account by Hayes et al. (1972) reveals a similar 8-h rhythm in the diapausing larvae of the codling moth, Laspeyresiu pomonella. This latter rhythm was statistically sound at 27”C, though not at lower temperatures, when: however, a 24-h component showed through. Beck reports that the Ostrinia 8-h rhythm persists in DD, but gives no data from which one could deduce the form in which the trimodal character of the rhythm survives in a free-run. The significance of these 8-h peaks in respiration remains obscure, but since practically no development or muscular activity is taking place they presumably reflect some function of the maintenance of life in the larvae at very low metabolic levels. 3.3.2 Rhythms in narcotic sensitivity Insects respond rhythmically to anaesthesia in two ways. First, as with other intoxicants (see below), they show a die1 variation in their susceptibility. Thus Nowosielski et al. (1964) found that the cricket, Acheta (Gryllus) domesticus, and the spider mite, Tetranychus urticae, vaned across the LD cycle in their responses to ether, chloroform and carbon tetrachloride. In the cricket, maximum sensitivity t o all three occurred during the first half of the night, thereby coinciding with the locomotor peak; minimum sensitivity occurred through much of the light phase. In Tetranychus, maximum sensitivity t o all three occurred around

THE PHYSIOLOGY OF INSECT CIRCADIAN, RHYTHMS

25

“dawn”, but the minima varied somewhat between “dusk” and midnight. The locomotor rhythm of Tetranychus is not known, but maximum sensitivity t o narcosis occurred just before the sharp rise in the oviposition rhythm after “dawn” (Polcik et al., 1965). The second effect of anaesthetics is t o phase shift other rhythms. Thus Ralph (1959) found that 1-h nitrogen narcosis given to Periplaneta 3-5 h before their expected locomotor peak had no effect on its timing, but given 11 h before the peak delayed it (and subsequent peaks in DD) by about 1h, or given 4 h after the peak delayed the next by about 2 h. This evidence does not amount t o a full phase-response curve, and the phase shifts were small by comparison with the normal variability of cockroach activity timing, especially under experimental stress (Brady, 1967b). It does, however, suggest that the clock itself rnay be affected by narcosis. This is more strongly implied by the effect of prolonged nitrogen narcosis on Drosophila emergence rhythms (Pittendrigh, 1954): 15-h narcosis delayed the next emergence peak by exactly !!5 h and later peaks by 10 h, so that the Drosophila clock appears to have been stopped by this treatment. This is very similar t o the effects of low temperature on some insect rhythms (Harker, 1964: 51; see p. 73). Still further evidence of the disruptive effects of narcosis on clocks is found in work on bees. Their timing ability, which has a circadian basis (p.9), desynchronizes in some way under the influence of prolonged C02 exposure, depending upon the depth and duration of narcosis (Medugorac, 1967; Medugorac and Lindauer, 1967). If they are subjected to 100 per cent C 0 2 for 2 h during the usuall feeding time, and are then kept in LL (in the laboratory), the next day the individual bees feed twice: once at the correct time and once about 4 h later. Medugorac and Lindauer’s experiments thus suggest the existence of two clocks in bees, one being insensitive to COz and the other phase shifted by it in a manner presumably similar to Drosophila’s. Bees art, however, a complex case because of the sophistication of their timing zibility (p. 9), and their use of social zeitgebers within the colony (Medugorac and Lindauer, 1967). These results taken together imply that the functioning of circadian clocks is susceptible t o interference by factors concerned with oxidative metabolism. If so, this may prove a fruitful field for future research. 3.3.3 Rhythms in insecticide susceptibility The development of insecticide resistance by insects led to the design of susceptibility tests for the monitoring of pest populations for potential resistance problems. For years these were carried out under conditions that were standardized in as many respects as possible, but which took no account of circadian variability. That this was a likely possibility was suggested by work on vertebrates (see Polcik t’t al., 1964). It seems to have

26

JOHN BRADY

been first reported in insects by Beck (1963), who found that Blattella was rhythmic in its susceptibility to various poisons. His fuller report (1968: 89) indicates that the cockroaches were reared in LD and then injected with an intoxicant at different times and observed in LL. The responses were complex, and difficult t o interpret because no indication of statistical variability was given. Dimetilan (a carbamate) and potassium cyanide produced broad, unimodal mortality rhythms about 4 h out of phase with each other; Dichlorvos (an organo-phosphate) produced a sharp bimodal mortality rhythm with one of its peaks coinciding with the Dimetilan peak; the responses t o sodium azide and sodium fluoride DNP and DDT, on the other hand, apparently showed no clear rhythmicity. Since 1963 other reports of insecticide susceptibility rhythms have followed. Cole and Adkission (1964, 1965) described a very marked rhythm (10 per cent t o 80 per cent fluctuation in mortality) in the susceptibility of the boll weevil, Anthonomus grandis, to Parathion (an organo-phosphate). This rhythm appeared to be 6-hourly, phased to “dawn”, but the strange feature of it was that the maxima and minima occurred on alternate observations, these being made every 3 h. No intermediate mortalities were detected in the LD 10 : 14 regime, under which the clearest rhythm occurred. It is therefore difficult to be certain of the precise form of the rhythm. Whatever this may have been, the 3-hourly fluctuation in susceptibility was annihilated by decapitation. (See also p. 31.) The susceptibility of spider mites, Tetranychus urticae, t o the organophosphate fumigant, DDVP, on the other hand, follows a much simpler (highly significant) unimodal course with a steady rise during the night to 45 per cent mortality at “dawn”, and a steady fall during the day to 25 per cent mortality immediately after “dusk” (Polcik et al., 1964). The maximum sensitivity thus coincides with the “dawn” rise in oviposition rate (Polcik et al., 1965; Polcik later, 1968, extended this work to several species of insects). A very different phase relationship in the susceptibility of the same mite to Dicofol (an organo-chloride) was reported by Fisher (1967). In this case mortality was consistently low (10-20 per cent) from dawn t o noon, consistently high (60-70 per cent) from noon to sunset when it fell again to climb steadily during the night. However, these mortalities were measured in mites taken from a single 24-h period of natural midsummer illumination with a 12’C temperature cycle, so, even though exposure was under constant conditions, it is difficult t o compare the results with those of Polcik et al. on mites reared in incubators. The oviposition rhythm of the treated mites reported by Fisher was apparently a direct effect of the intoxication, since no such rhythm occurred in untreated controls. Further studies on insecticide susceptibility rhythms include the following. A photoperiod of LD15 : 9 for some unexplained reason makes

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

27

houseflies, Musca domestica, less susceptible to DDT, Dieldrin and Endrin (organo-chlorides) than other photoperiods, including LD 16 : 8 and 14 : 10, the latter giving maximum susceptibility of all the photoperiods tried (Fernandez and Randolph, 1966); no rhythm as such was looked for. On the other hand, greater susceptibility of the same species to Malathion (an organo-phosphate) was found if the flies were kept in LD1O : 14 than in 24 : 10 (Frudden and Wellso, 1968). Although they looked for it, Frudden and Wellso failed t o detect a die1 rhythm in susceptibility, but it does apparently exist for other organo-phosphates (Reinhardt and Esther, 1971). In general, the phase relationships between the circadian behavioural rhythms of insects and their rhythms of insecticide susceptibility show no obvious physiological connections. Both hou!;eflies and cockroaches, for example, were found to be maximally susceptible to pyrethrum (a plant extract) in the afternoon (Sullivan et ul., 1970), whereas houseflies are behaviourally diurnal (Parker, 1962) and cockroaches nocturnal (Roberts, 1960). Likewise in the beetle, Tenebrio molitor, greatest susceptibility to Parathion (an organo-phosphate) occurred around “dawn” and “dusk”, whereas its maximum locomotor activity extended throughout much of the night (Fondacaro and Butz, 1970). A similar .‘dawn” and “dusk” pattern of peak susceptibility in male pink bollworm moths, Pectinophoru gbssypiellu, to Azinphosmethyl (another organo-phosphate) was found by Ware and McComb (1970). As in Tenebrio, therefore, the significance of the phase relationship with the behavioural rh,ythm is not obvious, except that peak susceptibility roughly coincided with the onset and offset of locomotor activity (Leppla and Spangler, 1971 ). In females, which showed only a single, “dawn”, susceptibility peak and whose behavioural rhythms start sharply at dusk (Fig. 4), the relationship is even less clear (though there is an obvious parallel in their X-ray sensitivity curve). In adult Drosophila, on the other hand, Rothert (1970) found that Parathion susceptibility paralleled the oxygen consumption rhythm (p. 23), and therefore the locomotor rhythm, remarkably closely. None of these studies reveals anything of the biochemistry underlying circadian variations in susceptibility to poisons. The rhythmic responses to organo-phosphates, for example, might have been expected t o be informative, since these chemicals act as cholinesterase inhibitors, but most of the evidence is insufficient to draw useful conclusions (though see Bull and Lindquist, 1968; see p. 31, and Fig. 6). The simple assessment of whether an insect dies or not evidently has too many quite unknown intermediate steps between the penetration of the insecticide and the outcome of its effects to answer any important questions about circidian mechanisms. A potentially more rewarding approach has been that of B d and Lindquist (1965), who injected P,,-labelled Disulfoton (Di-syston, a

28

JOHN BRADY

Fig. 4. Daily changes in the physiological functions of female pink bollworm moths (Pectinophora gossypiella). Abscissa, LD 14 : 10, points from last 8 h plotted twice. (a) Number of eggs laid per hour by 10 moths (redrawn from Minis, 1965: Fig. 3). (b) Mean survival after X-irradiation of 25 kr at different times of day (redrawn from Haverty and Ware, 1970). (c) Susceptibility to Azinphosmethyl intoxication, LD50s (presumably in ppm), after treatment at different times of day (redrawn from Ware and McComb, 1970). Cross-hatching, approximate distribution of maximum locomotor activity (from Leppla and Spangler, 1971).

thio-organo-phosphate) into bollworm larvae, Helio this zea, and then looked for metabolites. Separate samples of larvae, held either in LD 14 : 10 or LL, were taken every 2 h throughout the 24, injected with labelled Disulfoton, held for 4 h, and then killed. The concentration of the hydrolytic metabolites of Disulfoton in whole animal extracts of the LD larvae was found to vary with a unimodal cycle across the 24 h, with peak values occurring at night. In the excreta, the rhythm was bimodal, with one peak in the forenoon and a second coinciding with the night peak of the body extracts. This excreted rhythm appeared not to be due to a cycle of defaecation, since the oxidative metabolites in the excreta showed no cyclical fluctuation. The oxidative metabolites in whole body extracts, however, showed a bimodal rhythm of concentration which was phased inversely to the excreted hydrolytic metabolite rhythm. N o statistically significant rhythms were detected in the LL larvae, though this could have

THE PHYSIOLOGY OF ‘INSECT CIRCADIAN RHYTHMS

29

been because the individuals in the sample population were out of phase with each other. 3.3.4 Rhythms in X-ray sensitivity Although much research has been performed on the rhythmic susceptibility of mammals t o X-rays (e.g. Pizzarello et al.. 1964), only three accounts of such experiments on insects seem t o exist. The first. by Harker (1958a, 1964), has never been fully reported, but it was stated that doses of 100, 1000 and 10 000 r all caused a delay in the subsequent peaks of running activity in Periplaneta americana, and that the amount of delay was dose dependent and persisted in DD. The irradiation apparently lasted a very short time, so the effect was not analogous t o the stopping of the clock by low temperature (p. 73). The second report is by Rensing (1969a, 1969b), who investigated the die1 variation in mortality of Drosophila mehnogaster females after doses of 3500 to 20 000 r in LD. At all doses, mortality was highest at “dawn” and “dusk”. Rensing suggests that the daily niaxima in X-ray susceptibility coincide with the maxima of nuclear volumc in various endocrine tissues that he had reported earlier (Rensing, 1 9 6 6 ~ see ; p. 35), and with the maxima of oxygen consumption (1966a, 1966b, 1969b). This phase relationship is not entirely obvious from inspection of his figures, but he is cautious in his interpretation of its significance. The third report is by Haverty and Ware (1970), who measured the mortality of adult bollworms. Pectinophora gossypiella, exposed to 25 000 r at different times across 24 h of LD 1‘4 : 10. They showed that for males, irradiation at all times of day except “dawn” gave a mean survival time of 5-9 days, but that at “dawn” they were much less susceptible, surviving then for a mean of 1 6 days. Females survived longer, but otherwise showed a very similar pattern. In the females, this X-ray sensitivity rhythm mirrored the rhythm of susceptibility to organophosphate insecticide (Ware and McComb, 1970), susceptibility to the insecticide being greatest at “dawn” when susceptibility to X-rays was least (Fig. 4). Neither rhythm correlates in any obvious way with the only welldocumented behavioural rhythm in this species, that of oviposition (Minis, 1965; see Fig. 4). 3.3.5 Biochemical rhythms Implicit in the concept of a biological clock is the assumption that it runs on some sort of biochemical mechanism. This mechanism has proved very elusive, and attempts to measure its immediate products have generally been unrewarding since it is impossible to know whether the substance examined is an important link in the clockwork itself, or some quite incidental, secondary by-product. There is an extensive literature on such

30

JOHN BRADY

second-order biochemistry in vertebrates, but very few investigations have been performed on insects. This was the object of Nowosielski and Patton (1964), who measured die1 changes in the haernolymph concentration of sugars in crickets, Acheta (Gryllus) domesticus (Fig. 5 ) . They found that the level of reducing sugars remained fairly constant throughout the 24 h, but that trehalose showed a marked peak in concentration in the second half of the night,’falling back to base level by “dawn”. This peak occurred in both sexes, and in each of the four replicate series of measurements they made. It seemed not t o be a direct consequence of feeding, which occurred rather unevenly through most of the afternoon and night. Nor was it a direct consequence of locomotion (which occurs mainly at the beginning of the night in adults, Nowosielski and Patton, 1963), since last instar nymphs showed the same trehalose peak but no clear peak of activity. A much clearer association between haemolymph metabolites and the activity rhythm in Penplaneta americana was reported by Hilliard and Butz (1969), who found a c. 30 per cent drop in uric acid concentration over’the

Fig. 5. Daily changes in physiological functions in the haemolymph of crickets, Acheta domesticus (a), and cockroaches, Periplansta americana (b).0, Total sugar concentration in the haemolymph; 0, concentration of uric acid in the haemolymph; crosshatching, approximate distribution of maximum locomotor activity. Abscissa, LD 12 : 12; points from last 8 h plotted twice. ((a) Redrawn from Nowosielski and Patton, 1963, 1964; (b) redrawn from Hilliard and Butz, 1969;Brady, 1967b,1967c.)

M E PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

31

period of peak locomotion after “dusk” (Fig. 5 ) . They suggest that many functions, including excretion, occur at a higher rate during the period of peak activity. Hilliard and Butz also found a clear cycle in the total sugar concentration of the haemolymph (Fig. 5), which, like the cricket’s, climbed steadily until the last hours of the night and then fell precipitously just before “dawn” (a c. 50 per cent drop in females). As in the cricket, therefore, there appears to be little correlaticin between blood-sugar and activity. Nayar (1969) found a trehalose rhythm in the fourth instar larvae of the mosquito, Aedes taeniorhynchus, but this rhythm is much less clear than the cricket’s or cockroach’s. Nayar measured blood glycogen levels as well, but the fluctuations were not obviously rhythmic. Turner and Acree (1967) performed a similar study on the haemolymph hydrocarbons of Periplaneta americana kept in LD or LL. They reported that heptacosadiene increased during the forenoon and that pentacosane and methyl pentacosane showed a corresponding decrease. The fluctuations were rather small, the greatest being an 8 per cent increase in the initial heptacosadiene level in the LD males, and 4 per cent in the LL males; changes in the females were less than half this and appeared scarcely significant statistically. The changes occurred almost entirely in the first 3 h after “dawn”, thereafter remaining fairly constant until “dusk”, presumably returning to the “dawn” level durin;; the night, though this was not measured. The LL animals showed precisely the same phase relationship to “dawn” as the LD animals, their lesser changes seeming not to be due to partial desynchronization of the individuals after 10 days in LL, since their variances were in general less than rhose of the LD animals. N o indication of the actual concentration of the hydrocarbons is given (see also.Acree et al., 1969), nor of their function; the role of this rhythm is therefore difficult t o assess. A brief report by Rounds (1968) suggests that there might be a rhythm in the sensitivity of cockroach mid-gut cells t o extracts of the mid-guts from other individuals. Proteinase release from! the mid-gut was stimulated by injection of extracts into the haemocoel of intact Periplaneta at different times of day. The effects of mid-gut extract, muscle extract, or saline injections were very variable, but statistically indistinguishable at all times except “dusk” when the mid-gut extract caused the release of significantly more proteinase than the other two. Bull and Lindquist (1968), observing that there are rhythms in susceptibility to cholinesterase-inhibiting insecticides, inferred that insects might therefore exhibit fluctuations in cholinesterase activity. In an examination of this possibility in the boll weevil, Anthonomus grandis, they found that the activity of acetylcholinesterase (AChE) in whole insect extracts of weevils kept in LD 14 : 10 was about 25 per cent greater during the light phase (when the insects are active) than during the night (Fig. 6).

32

JOHN BRADY

016

014

Fig. 6. Daily changes in cholinesterase activity in (a) whole boll weevils (Anthonornus grandis), and (b) the brains of crickets (Acheta domesticus). Abscissae, LD 14 : 10 and 12 : 12, respectively; points for last 6 h plotted twice. Ordinate, amount of acetylcholine hydrolysed as indicated; cross-hatching, approximate distribution ,of maximum locomotor activity. ((a) Redrawn from Bull and Lindquist, 1968; (b) compiled from Cymborowski et al., 1970.)

What relation this rhythm bears to the 6-hourly Parathion susceptibility rhythm found in this species by Cole and Adkisson (1964;see p. 26) is unexplained. Bull and Lindquist also showed that weevils reared in LL had a virtually constant enzyme activity level across the day. Whether this was due t o desynchronization of the individuals within the LL population, or to each individual being aperiodic, is not clear, but the small size of the variances in LL imply the latter. Venkatachari and Muralikrishna Dass (1968) observed a similar phase relation between locomotor or nervous activity and the amount of AChE activity in the nervous system of scorpions. A later study by Cymborowski et ul. (1970),specifically concerned with a search for factors controlling insect locomotor rhythms, indicated a sharp c. 40 per cent decrease in AChE activity of cricket (Achetu domesticus) brains, exactly coincident with the animals’ peak locomotor activity (Fig. 6). Confusingly, this is the converse of the relationship between AChE and locomotor activity in Anthonomus and the scorpion.

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

33

Precisely what these daily AChE activity changes signify is very difficult to determine, not least because of the differing phase relationships. Venkatachari’s hypothesis (1971) is that cephalic hormones control the AChE activity level and that this in its turn produces the changes in nervous activity associated with the locomotor rhythm. Whereas this may be one possible interpretation for the scorpion and boll weevil observations, it cannot account for the relationship in crickets. In any case, whatever the phase relationship, it is not at all clear why the changes in AChE level should be considered the cause of the nervous activity rhythm rather than the result of it. An earlier approach to the discovery of biochemical factors directly involved in circadian locomotor control in arthropods was the search by Fowler and Goodnight (1966a, 1966b) for cycles of 5-hydroxytryptamine (5-HT)production in the opilionid, Leioburium longzpes. They extracted the brain plus gut of pooled pairs of aninials at 4-h intervals across a LD I 4 : 10 cycle. This species is noctuinally active, and the 5-HT accumulation rhythm appeared to phase lag the locomotor rhythm by several hours. Thus, the locomotor activity started to increase before “dusk”, whereas there was little evidence of any change in the 5-HT level until after the onset of darkness; peak 5-HT extraction occurred at 02.00 h, but peak activity was about 4 h earlier. Sincc: neither the location nor the timing of the 5-HT release by the brain (of Leiobunum is known, its relationship with locomotor activity remains uncertain. Contemporary research by Hinks (1967) revealed the exisi:ence of tryptophan (a 5-HT precursor) in the medial neurosecretory cells of noctuid moth brains, but his work did not reveal whether the presence of tryptophan was cyclical or not (see also p. 42). Fowler and Goodnight succeeded in culturing isolated Leiobunum brains for 80 days in LD, at the end of which period there was a distinct suggestion of a residual rhythm of 5-HTaccurnulation. However, the tissues had shrunk during culture and the levels of 5-HT were only about a quarter of those in fresh tissue. Since the number of animals used was small, the variances too uneven to permit a comparison of means (Fowler and Goodnight, 1966b: 189), and the die1 differences rather close to the range of sensitivity of the technique (Vanable, 1963), it is difficult to be sure of the real dimensions of this rhythm after the prolonged culture. As the cultures were maintained in a light cycle, it is in any case not clear whether the 5-HT production was endogenously rhythmic or photokinetically induced. In a later study, Fowler et al. (1972) found 5-HT rhythms in Drosophila melanogaster, extracting this product from whole lfirvae or pupae, and the brains plus guts of adults. There were differences between the stages, and according to whether the rearing cycle was LD 10 :14 (“winter”) or 14 : 10

34

JOHN BRADY

(“summer”). In “summer”, adults exhibited a single broad major peak around “dawn”, to which was added a lesser “dusk” peak in “winter”, though whether this minor peak was statistically valid is not clear. There appears to be a potential link here with Rensing’s bimodal oxygen consumption rhythm (p. 23), and with the locomotor rhythm, but on this basis one would expect the bimodality of the 5-HT production to be more obvious. A similarly oriented, but inconclusive investigation was made of die1 changes in potassium and sodium concentrations in cockroach (Periplaneta) blood by Brady (1967d, 1968). This study was concerned with the means by which locomotor activity might be coupled to the clock. That blood ion changes could be involved in this, at least in herbivorous insects, was suggested by the work of Hoyle (1954) and Ellis and Hoyle (1954) on the effect of blood potassium concentration on locust marching, and by the work of Pichon and Boistel (1963) on the effect of a high potassium diet in suppressing the cockroach locomotor rhythm (though there were counter-indications in the work of Barton-Browne and Evans, 1960, on blowflies). In the event, although a significant fall of at least 2 mM litre-’ (or 10 per cent) did occur in the blood potassium of Periplaneta at the time of onset of their locomotor peak, this could not be detected in all series of animals. It was not a readily repeatable observation, nor was it clear whether the 2 mM litre-’ drop would cause sufficient change in muscle or nerve resting potential to affect the general level of excitability and thereby trigger the circadian activity peak as hypothesized (Brady, 1968). 3.3.6 Cellular rhythms a. Endocrine cells. Ever since Harker first reported that blood-borne factors were involved in the control of the cockroach locomotor rhythm (1954), and that these factors were probably neurosecretory (1955), searches have been made for circadian cycles in insect gland cells. The first such investigation was by Klug (1958) who found a rhythm of nuclear volume in the corpora allata cells, and a rhythm in the relative numbers of two classes of medial brain neurosecretory cells in the beetle, Carabus nernoralis. This lead has been followed by a number of other workers, mainly using the same two techniques : (1) the measurement of nuclear volume; (2) the assessment of cytoplasmic changes. Both approaches are beset with difficulties. The potential errors involved in the estimation of nuclear volume from histological sections have been discussed by Brady (1967a), the main point being that nuclei become compressed into a variety of spheroids during fixation, and volume estimations based on diameters of sections will therefore vary according t o the cube of the errors resulting from the different orientations of the sections. The approach of both

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

35

Rensing and Brady has been to minimize the problem by referring simply to nuclear “size”, calculating this from the products of the longest and shortest diameters of the largest section of nucleus from each cell. In a study of Drosophila kept in LD 12 : 12, Rensing (1964, 1966c; Rensing et al., 1965) found reasonably clear diel changes in the nuclear size of neurosecretory cells in the brain, suboesophageal ganglion, and corpora cardiaca, and in the cells of the corpora allata and prothoracic glands. Although the variances were large, all these cells had apparently significant bimodal nuclear size rhythms with maxima in the late night and afternoon, and minima in the early night and forenoon, both in larvae and in adults (Fig. 7). Brady (1967a), in an investigation of the lateral neurosecretory cells of the cockroach suboesophageal ganglion, identified by Harker (1960a) as responsible for the locomotor rhythm, found much less obvious nuclear size cycles. N o detectable changes occurred in the cytoplasm, and the nuclei changed at the most by about 25 per cent in volume (sensu stn’cto) during the period of maximum locomotion (cf. p. 54). At or before the onset of peak activity, when one might have expected the greatest change, none apparently occurred. These two studies reveal an inherent problem in quantifying rhythmic cytological changes by this means. In the cockroach suboesophageal ganglion, diel nuclear size changes occurred in other cells, including motor neurones, precisely coincidentally with the changes in the lateral neurosecretory cells (Brady, 1967a). More strikingly, Rensing found (1966c, 1969c) in Drosophila that the nuclear size cycles of the brain, corpora allata, prothoracic glands, fat body and salivary glands in the larva (Fig. 7), and the brain, suboesophageal ganglion, corpora cardiaca and corpora allata in the adult, all had coincident maxima. The assumption behind this kind of work has been that cyclical changes in endocrine cells would imply that they were causative factors in the regulation of insect circadian rhythms. The observed close coincidence O F their cycles with each other, and with those of non-endocrine cells, could be interpreted more economically, perhaps, on the basis that the fundamental circadian organization of insects results in synchronous metabolic changes in all, or many, of their cells. At the very least, one should be cautious in distinguishing between the central causes of rhythmicity and its effects. The second technique for detecting circadiian cellular cycles, namely that of measuring cytoplasmic changes, has only been applied to neurosecretory tissue. The common procedure is to divide cdls into arbitrary categories on the basis of their histological colouring, or the distribution of their secretory granules, and then to look for changes in these categories across the day. Errors may be introduced by variations in staining or fixation, as well as in the subjective classification. Quite apart from this, unless

JOHN BRADY

36 +2

0 -I

+2

0

Corpora

-21Y r

/'\

ok /

-10

-O

-0

/-

P

Fa? body

glands

- 30

1

1

1

1

1

Fig. 7. Nuclear volume changes in various tissues of Drosophih melanogaster larvae during their 6th (penultimate) day of development at 20'C. Abscissa, LD 1 2 : 12; night-time data repeated on second night. Ordinate, mean deviation of the product of the longest and shortest nuclear diameters of nuclei from their mean growth curve. (Redrawn from Rensing, 1966c: Fig. 5, and 1969c: Fig. 3.)

individually identifiable cells can be looked at across the day, there is never any certainty that the histologically observable differences follow the real secretory sequence in the cells. Further difficulties arise when comparisons are made between different species, although valiant attempts at homologizing the categories of insect neurosecretory cells have been made (e.g. Delphin, 1965).

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

37

Brady (1967a) could find no change in the cytoplasm of any of the suboesophageal ganglion neurosecretory c d s of the cockroach, Pen‘planeta, at any time. De B e d (1965), however, reports that the “A-cells” in the ventral nerve cord of another cockroach, Leucophaen, contained more neurosecretory material during the night (when the animals were active) than during the day. And Cymborowski and Flisiriska-Bojanowska (1970) report changes in the brain and suboesophageal ganglion of Periplaneta, though in their case there were sometimes greater histological changes in the arrhythmic control animals than in the rhythmic experimental ones, so that the significance of these changes is not clear. A similar rhythm in the T-cells” of the suboesophageal ganglion of the stick insect, Clitumnus extradentatus, has been reported by Raabe (1965,- 1966), though in this case there was evidently considerable variability, and little synchrony between the different groups of cells. A more objective approach has been developed by Rensing (1964, 1966c), who used a form of micro-spectrophotometer to compare the absorption of neurosecretory material in the region around the nucleus with that in the axon “hump” of the brain neurosecretory cells of Drosophila. This measure was taken to reflect the relative degree of accumulation or release of neurosecretory material. Rensing found considerable variation between different batches of flies, but they were consistent in showing maximum accumula1:ion near the nucleus around “dawn” and “dusk”, that is some 3 h after their peak nuclear size. The larvae showed similar, but less marked, changes. Rensing’s model for the Drosophila circadian system, incorporating this work, is discussed on p. 8 5 . There has been one brief investigation o f the ultra-structure of these circadian secretory cycles. Dutkowski et al. (1971) examined the medial brain neurosecretory cells of crickets, Acheta domesticus, killed 30 min after “dawn” and 30 min after “dusk”; that is, at times of minimum and maximum locomotor activity. They report marked differences between the two groups. The cells of the inactive animals contained extensive rough endoplasmic reticulum, secretory vesicles in the Golgi region and in the axon only, and nuclei with smooth membranes. The cells of the active animals, on the other hand, contained only fragmented endoplasmic reticulum, apparently quiescent Golgi, secret:ory vesicles in the perikaryon but not in the axon, and nuclei with undulating membranes. Dutkowski et al. did not indicate how reproducible these differences between the two time samples were, but they concluded that the brain cells of the inactive animals were synthesizing and releasing neurosecretion, whereas the cells of the active animals had ceased (temporarily) t o synthesize actively or to release the secretion, and thus accumulated secretion in the perikaryon. The potential value of the electron microscope in such investigations is

38

JOHN BRADY

revealed for the first time; it will be interesting to see how other cellular circadian secretory cycles unfold ultrastructurally when examined in greater detail. Another advance in this field has been the use of autoradiography to measure the cyclical metabolic activities of endocrine tissue. In a series of papers, Cymborowski and his colleagues have used this technique, principally on Acheta domesticus, to try and relate neurosecretory cell function to the control of locomotor activity. They showed that there is a sharp diel rhythm in RNA synthesis (as indicated by H-uridine incorporation) in the medial neurosecretory cells of the brain and the ventral neurosecretory cells of the suboesophageal ganglion in crickets kept in LD 12 : 12 (Cymborowski and Dutkowski, 1969). In the brain cells, nuclear volume changes more or less followed the autoradiographic cycle, with a clear peak at “dawn”, but no such correlation was apparent in the suboesophageal ganglion. They extended this technique to show (1970a) that there is a diel rhythm in protein synthesis in the same cells, with a sharp peak in the brain in the afternoon and peak accumulation of stainable neurosecretory material a few hours later. Dutkowski and Cymborowski (1971) reported that destruction of the medial brain neurosecretory cells in female crickets resulted in a loss of RNA synthesis in the suboesophageal ganglion cells, but in illl increase in RNA synthesis in the egg follicles and fat body. They suggest that the suboesophageal ganglion is therefore acting as an inhibitory intermediary between the brain and these other tissues. This may be so, though other interpretations appear possible. On the basis of their results, Cymborowski (1970b) and Cymborowski and Dutkowski (1969) set out the following proposal for the control of cricket locomotor rhythm. In response to the “dawn” zeitgeber, the brain cells commence a cycle of RNA and then protein synthesis, followed by accumulation and release of hormone. The hormone is transmitted axonally to the suboesophageal ganglion where it stimulates that ganglion’s RNA synthesis and neurosecretion. The brain hormone inhibits locomotor activity, which it keeps at a low level throughout the 24 h except for the first few hours after “dusk” when the hormone is not released and hence shows peak accumulation in the neurosecretory cells. The role of the suboesophageal ganglion’s hormone is as a supplementary inhibitor. This is a simple and attractive hypothesis, but it must be pointed out that the evidence for it is not as strong as it may at first appear. The role of hormones in the control of insect rhythms is discussed at length below (p. 5 5 ) , but it will be convenient to consider this histological aspect now. There are three main weaknesses. First Cymborowski used locomotorily arrhythmic animals from LL as controls, whereas absence of behavioural rhythmicity does not necessarily guarantee total physiological arrhy-

THE PHYSIOLOGY OF INSECT CIRCADIAN R H Y l H M S

39

thmicity (see Cymborowski and Flisiliska-Bojanowska, 1970, Figs 2B and 3; Harker, 1964: 98; and Lohmann, 1967, Figs 2 and 3), so time samples from a population of individuals which may well be asynchronous in LL are not necessarily valid controls. Secondly, Cyniborowski and Dutkowski (1970b) found a quite different phase relationship between the cytological and behavioural events in females from that in males, so that the same secretory kinetics cannot control the locomotor activity of both sexes (nor, for that matter, can the same kinetics occur in both brain and suboesophageal ganglion since the sequence of their cytological events is quite different). Thirdly, the locomotor rhythm persists in DD, but no observation was made to see if the cytological rhythm does as well, or whether it is merely a direct response to the lighl.. In conclusion, it should be noted that, whereas the results of all workers on this subject clearly show the existence of diel rhythms in endocrine (and other) cells, no one has yet shown whether they persist in DD with the same phase angle to the events they are assumed to control. It is a common failing to confuse circadian cause with circadian effect: any two functions which occur with a diel rhythm in an organism are inevitably temporally correlated; it does not follow that they are also physiologically causally correlated. b. Tissue culture. Although techniques for culturing insect cell lines have been available for some years (e.g. Grace, 1962), and one might therefore have expected that attempts would have been made to use isolated insect cells for the sort of research that has been performed on Protista (e.g. Sweeney, 1972), this has never been tried. Organ culture, on the other hand, occasionally has been. The first example was that of Fowler and Goodnight (1966a) discussed above (p. 3 3 ) , the only other example is Rensing’s study of Drosophila salivary gland cell!;. By taking samples across the 24 h, Rensing found ( 1 9 6 9 ~ )that a diel rhythm occurred in the nuclear and nucleolar volume of the salivary gland cells of intact Drosophila nrelanogaster larvae maintained in LD. There were some differences during the course of development, but in 5 - and 6-day old third instar larvae, highly significant maxima in nuclear size occurred during the second half of the night and during the afternoon; minima occurred around “dawn” and “dusk”. The nucleolar size changes were roughly coincident. Rensing then took salivary glands from last (7 th) day larvae and cultured them in vitro. Making repeated measurements across the 24 h on individual nuclei, he found that they still showed bimodal, diel fluctuations in size (c. + 10 per cent of the daily mean product of their long and short diameters, compared with +20 per cent for the fixed ”nuclei), but that the different cells of a single gland in culture tended to lose synchrony and become out of phase with each other. However, the mean size rhythm of

40

JOHN BRADY

several nuclei in a gland remained bimodal and this rhythm survived for at least 4 days in LL (200 lux). Even after 3-4 days in isolation, the nuclear rhythm of cultured glands, in LD, appeared reasonably synchronous with that of the comparable fixed 7-day-old larvae. Rensing found that he could phase shift the rhythm in cultured glands by a change in the light cycle, or by a pulse of ecdysone in the culture medium (Rensing, 1 9 6 9 ~ ) .He had already shown that the ecdysoneinducible puff (74EF/75B) generally appears in the salivary glands of pre-pupae between “dawn” and noon (Rensing and Hardeland, 1967), and he suggests that these phenomena are part of a common temporal system (Rensing, 1971; see p. 85). There is also some evidence (Rensing, 1970) that a pulse of actinomycin-D, though having no immediate effect on the nuclear rhythm of cultured glands, possibly causes a delay phase shift 24 h later. The evidence for a similar effect of in vivo injections of actinomycinD on the nuclear volume rhythm of larval Galleria salivary glands (Kontopp, 1967) is less than convincing, however. 3.3.7 Pharmacological rhythms As an alternative to measuring cytological events in endocrine tissues in order to infer their physiological activities, the bio-assay of extracts of these organs has occasionally been tried. One problem is to find a physiologically relevant assay, and this is particularly difficult when the control of behaviour is being explored. Ralph (1962) made an extensive study of the pharmacologically active substances which he extracted from cockroach nervous system breis with a variety of solvents. The assay he used was t o apply the extracts to semi-isolated cockroach hearts and t o measure increases or decreases in beat frequency. His results showed the existence of several heart accelerating and decelerating substances in the nervous system, of which some at least may be presumed to be neurohormones. The different effects of extracts taken at different times suggested that the substances were present in varying quantities across the 24 h. This would be consistent with the die1 cycle in the amounts of 5-HT extracted by Fowler et al. (1972) from Drosophila brains. No follow-up to these suggestive results has been published, however, and the connection between these pharmacological cycles and either the control of locomotor rhythms or the functioning of the circadian clock remains obscure. Somewhat similar observations have been made by Rao and Gropalakrishnareddy (1967), who used an approach which appears more closely related to locomotor control. The scorpion, Heterometrus fulvipes, is mainly active in the early part of the night, and electrical activity in its exposed nerve cord follows the same temporal pattern (Rao, 1963; Venkatachari, 19 71 ). Rao and Gropalakrishnareddy therefore used isolated

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

41

nerve cords as an assay for the neuro-stimulating effects of blood or brain extracts (cephalothoracic ganglion) at different times of day. They found that blood or brain extract taken from donor!; around the time of their maximum locomotor activity caused a c. 30 per cent increase in impulse frequency over the control saline-perfused rate, whereas blood taken around the end of maximum locomotor activity caused a c. 30 per cent decrease, and brain extract a c. 60 per cent decrease, all these differences being highly significant (Fig. 8).,

50

J40

lo 0

1

t

......._.

L 6

12

18

24

Local time

Fig. 8. Neuro-stimulating effects of brain extract (A), blood (a),and saline ( 0 ) on the isolated nerve cord of the scorpion, Hete~ometrusfulvil,es.Extracts and blood taken at the times indicated from scorpions apparently maintained in natural daylight (as indicated on abscissa), assayed in the forenoon. Ordinate, mean total spikes per second in isolated test nerve cords; bars, 2 x S.E.; cross-hatching, approximate distribution of maximum locomotor activity. (Compiled from Rao and Cropalakrishnareddy, 1967.)

This is an extremely interesting report, and i!; disappointing only in that so far no one else has attempted to repeat it, or develop its evident possibilities. The neuropharmacological effects of brain extract were to be expected from the observations of Fowler et al. (1972) on Drosophila, as well as from the histological evidence of cyclical neurosecretory activity in insects (e.g. Rensing, 1966c; Cymborowski and Dutkowski, 1970a), but Rao and Gropalakrishnareddy appear to have shown for the first time that such circadian histological changes are in f,act reflected in circadian neuropharmacological changes in the haemolyniph (cf. Mothes, 1960, on chromatophorotropic substances in the blood of Carausius). The significance of this work discussed below (p. 69). A more total pharmacological assay has been used by Cymborowski ( 1 9 7 0 ~ who ) measured changes in the LD locomotor activity of crickets,

42

JOHN BRADY

Acheta domesticus, resulting from the injection of various drugs into the haemocoel. Reserpine, or lysergic acid, at c. 10 ppm resulted in abolition of the locomotor peak for 1 or 2 days, sometimes followed by a disturbed rhythm for a further day. Up to 5 pprn of 5-hydroxytryptamine (5-HT) injected into the haemolymph had virtually no effect on locomotor behaviour. However, the same dose injected directly into the protocerebrum of the brain, while having remarkably little immediate effect on the day of injection, caused extreme hyperactivity for at least the next 5 days; injections of the same quantity of water into the brain apparently had no effect. This reaction to 5-HT injection is rather different from that occurring in moths, in which a haemocoel injection of 10 pprn causes an immediate increase in activity, particularly during the ensuing scotophase (Hinks, 1967, 1968). Interestingly, 5-HT injection re-established the scotophase activity missing from moths with ablated medial brain cells (p. 67). The interpretation of such results should evidently be made with caution, but, at least in LD, none of the drugs affected the phase of the rhythm, so that involvement with the clock itself is not implicated. The same conclusion can be drawn from the work of Amouriq (1969) on the effects of amphetamine or ephedrine on the locomotor rhythm of the woodlouse, Porcellio scaber. 3.3.8 Tumours The most sensational results ever described in the field of insect biological clocks were those indicating a specifically circadian relationship between abnormal rhythmic secretion of hormones and the induction of cancer in. cockroaches (Harker, 1958b). Harker reported that the transplantation of suboesophageal ganglia into host cockroaches from donors whose rhythms were 1 2 h out of phase induced malignant tumours in the recipients, whereas transplants from in-phase donors checked the growth of preexisting tumours. Her conclusions were that tumour induction and control were the result of, respectively, out-of-phase or in-phase secretion by the neurosecretory cells of the transplants. The subject of insect neoplasms has been reviewed by Harshbarger and Taylor (1968), and this putative circadian aspect will therefore not be discussed at length here. However, Harker’s report has such far-reaching medical implications that the existence of counter-indications must be pointed out. Two other groups of workers have totally failed to repeat her results. Nishiitsutsuji-Uwo and Pittendrigh (1967) performed all the relevant suboesophageal ganglion transplant experiments plus various other operations and failed to induce any midgut tumours. Sutherland (unpublished; in Harshbarger and Taylor, 1968: 175) apparently failed similarly. There is thus a complete divergence of results. Two explanations seem possible. First, Harker’s strain of Perzpluneta

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

43

americana may have been unusually susceptible to oncogenic agents; though if this were so it is odd that her brain transplants induced no tumours. Secondly, the rationale behind her experiments was that the transplants were releasing neurosecretion rhythmically, as evinced by their effect on locomotor activity. No one else, however, has managed to induce this behavioural response from endocrine transplants (p. 56), and it might logically be argued that until they can do so, and prove that their transplants are secreting rhythmically, they are not testing the same oncogenic agent. Harker herself points out (1963) that insect neoplasms are difficult t o identify, but the nub of this disagreeinent appears to reside in the long-standing controversy over the role of the suboesophageal ganglion hormone in rhythm control. Harker also states (unpublished, 1961: 136) that “major pathological changes” occur in the gut of the cockroaches if they are kept in LD 2 : 6 for 10 days; no one else has ever examined this possibility.

4 Timing processes The concept of a circadian clock supposes i t to consist of three main elements: ( 1) an endogenous physiological oscillator whose natural period is about 24 h; (2) a coupling mechanism linking this oscillator with the physiological processes it controls; (3) a photoreceptor system via which the oscillator is entrained to the frequency of the environmental light cycle. The overall circadian system of an organism must, of course, consist of more components than this simple trinity. For example, entrainment involves not only photoreception, but also a mechanism to couple the photoreceptor with the oscillator, plus a m’eans to adjust the latter’s frequency. Similarly, the coupling between the oscillator and the processes it controls presumably involves many subsidiary elements, probably often including secondary oscillators. Indeed, it is far from clear whether the circadian organization of complex higher metazoa involves a single, primary oscillator, or several; on balance the latter seems more probable (see p. 93). Nevertheless, the system divides conveniently into the three basic elements: entrainment, control of driven rhythms, and primary driving oscillator. 4.1

ENTRAINMENT

As just outlined, entrainment consists of two parts: perception of light and phase adjustment of the oscillator. Under natural conditions these always operate in unison so that the sun’s cycle daily riisets the circadian oscillator by a small amount to keep the organism on an exact 24-h die1 periodicity. However, phase adjustment can operate independently of photoreception,

44

JOHN BRADY

since it can occur in response to temperature (e.g. Roberts, 1962)and other zeitgeber cycles (e.g. social, Medugorac and Lindauer, 1967). We i l l deal with photoreception first.

4.1.1 Photoreception There is good evidence that insects can perceive light not only via their compound eyes and ocelli, but also directly via the brain cells (Lees, 1964; Truman and Riddiford, 1970). It is also possible that other tissues are photosensitive (e.g. terminal abdominal ganglion, Ball, 1965;head hypodermis, Brousse-Gaury, 1967, 1969). As far as circadian entrainment is concerned it might be thought that as long as light is perceived, the route of perception would be immaterial. In at least one case where it has been examined critically, namely the entrainment of the cockroach activity rhythm, this is not so, however. Early experiments on cockroach entrainment led to some confusion (Cloudsley-Thompson, 1953; Harker, 1956; see also Van Cassel, 1968), perhaps partly because the possibility of direct photoreception by the brain was not then realized and the eye-opaquing techniques were inadequate, but more importantly because the nature of entrainment was not fully appreciated. In order to demonstrate the entrainment of a rhythm it is necessary to show: (a) that the rhythm is endogenous and persists in DD or LL, and (b) that the rhythm phase-shifts in response t o a new timing of the zeitgeber cycle (for example a change from LD to DL). The role of a given photoreceptor in this process can then be tested by seeing whether entrainment to the new light cycle occurs in its absence. This procedure was adopted by Roberts (1965)t o show that cockroaches (Leucophaea maderae and Penplaneta americana) free ran in a LD cycle when their compound eyes were painted over, but remained entrained when their ocelli were surgically ablated. He concluded (contrary to Harker, 1956) that it is not the ocelli, but the compound eyes which are the relevant photoreceptors for cockroach entrainment. This result has been confirmed by a more extensive series of observations by Nishiitsutsuji-Uwo and Pittendrigh (1968a)who found in addition that the animals free ran in LD if their optic nerves were cut (i.e. between the optic lobes of the brain and the compound eyes) but the ocelli left intact. Furthermore, they showed that inserting a glass window in the frons of optic nerve-severed animals did not lead t o entrainment in LD (Ball’s result (1972)does not conflict with this finding because he did not preclude light conduction to the retina). They thus concluded that the direct action of light on the cells of the pars intercerebralis was non-entraining (though these cells are not unaffected by light, Cook and Milligan, 1972). This implies a difference from the photoreceptive role of the brain in photoperiodic induction in aphids (Lees, 1964) and silk moths (Williams

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

45

and Adkisson, 1964) and in entrainment o f the eclosion rhythm of silkmoths (Truman and Riddiford, 1970). However, Nishiitsutsuji-Uwo and Pittendrigh point out that they used a light intensity of only 5-20 lux (though sometimes also 80 lux, see Nishiitsutsuji-Uwo and Pittendigh, 1968b: Fig. l o ) , whereas Lees used 500 lux (Williams and Adkisson, and Truman and Riddiford, used 1200-1900 lux). The possibility that higher light intensities may entrain through extra-optic reception is implied by Dumortier’s work (1972) on the stridulation rhythm of the cricket, Ephippigcr. He found that the rhythm apparently remained entrained to LD or DL when the ocelli were destroyed and the optic tracts cut between the optic lobes and the brain, and he questions the interpretation by Nishiitsutsuji-IJwo and Pittendrigh (1968a) of their experiments on the optic lobes. There is a possible weakness in Dumortier’s own interpretation, however. The entrained LD rhythm in crickets with severed optic tracts had a quite different phase relationship to the light cycle from that of normal animals, maximum stridulation starting abruptly at “dawn” and stopping at “dusk” (his Fig. 12), instead of rising steadily from midnight and dying out around noon (his Fig. 4). Since cockroaches lose circadian rhythmicity when their optic tracts are cut (see p. 61) one wonders whether the LD rhythm in these similarly operated crickets was not simply a direct, non-circadian photokinetic response to the high light intensity used (apparently up to 5000 lux at the level of the animals, his p. 109); they were not tested in DD to see if the post-operative rhythm persisted in the absence of a light cycle. This suggestion is supported by the fact thal when Dumortier used light guides implanted against the brain, entrainment only occurred in animals with intact eyes and ocelli, and not in those with cauterized eyes and ocelli, implying that entrainment did require light conduction to, and the use of retinal elements. Furthermore, experiments by Loher (1972) on the stridulatory rhythm of another cricket (Teleogryllus commodus) have shown that at around 30 lux no entrainment occurs without the compound eyes. Neither the ocelli nor light reaching the brain via the transparent eyeless eye cups were sufficient to prevent the rhythm free running in LD at this intensity. Loher and Chandrashekaran (1970) found that the oviposition rhythm of the grasshopper, Chorthippus curtipennis, could be entrained via either the compound eyes or the ocelli. However, like Dumortier (1972), but unlike Nowosielski and Patton (1963), they also report that the rhythm remained normally entrained when the head capsule was completely opaqued (except for the mouth-parts), and they concluded that extracephalic photoreception can entrain. In view of the high light intensity used (c. 2000 lux), and the fact that the eyes and ocelli were intact (Loher, personal communication, and their p. 1685) and the mouth-parts left

JOHN BRADY

46

unpainted, it is not clear whether this conclusion is valid; the absence of light conduction to normal retinal elements is certainly not ruled out. The same doubt must apply to the eye-opaquing experiments of Nowosielski and Patton (1963), and Ball (1972). Evidently the situation vis-&is the photoreceptive route for the entrainment of insect behavioural rhythms is confused. At least for low light intensities (<100 lux). However, two facts appear reasonably firm. First, when the compound eyes are removed, perception of light by the ocelli is not sufficient to entrain. Secondly, Nishiitsutsuji-Uwo and Pittendrigh’s window experiment, and Loher’s transparent eye-cup experiment, show that for some species direct photoreception by the brain cells does not entrain. Whether very bright LD can entrain true circadian rhythms in insects by extraocular light perception remains to be proved. In Crustacea, on the other hand, there is a good evidence for extra-retinal (as well as extra-caudal) entrainment (Page and Larimer, 1972). In contrast to the entrainment of behavioural rhythms, evidence is accumulating to suggest that, as in photoperiodic measurement, developmental rhythms may typically be entrained by direct brain photoreception. This appears to be true at least for the timing of adult eclosion, but may not be general to all growth rhythms (e.g. Neville, 1967b; Dutkowski and Cymborowski, 197 1).

Energy flux (log erq.s-’.cm-*)

Fig. 9. Phase-response curve of the eclosion rhythm of Drosophilu melunogaster to light signals of differing energy values. Each point represents the delay phase shift of the free-running rhythm of a different population of pupae exposed to a single 15-min monochromatic light signal (of c. 456 nm) 3 h after the last “dusk” at the onset of DD. 0. Larvae reared on normal diet; 0, larvae reared on p-carotene-free diet. (Redrawn from Zimmerman and Goldsmith, 1971.)

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

47

In silkmoths, not only is the source of hormone eliciting eclosion behaviour as well as the clock gating its releast: situated in the brain (see p. 52), but so also is the photoreceptive mechanism entraining this clock. Truman and Riddiford (1970) have demonstrai.ed this to be the case by transplanting the brain from the head to the abdomen of pupal moths, and subjecting the anterior and posterior halves of the pupa simultaneously to light cycles 180” out of phase. Eclosion is then entrained by the cycle “seen” by the abdomen, whereas in control pupae with the brain replaced in the head end, eclosion is entrained by the cycle “seen” by the anterior half. In Drosophila, the eclosion rhythm entrains quite normally in eyeless mutants (Engelmann and Honegger, 1966). Moreover, the action spectra of circadian entrainment and of the compound eyes differ (Zimmerman and Ives, 1971), and 0-carotene-based photoreception (typical of all normal retinal cells) is not involved in rhythm entrainment (Zimmerman and Goldsmith, 1971; see Fig. 9). While such results do not prove direct brain photoreception in Drosophila, they very strongly imply extra-ocular reception, and in view of the evidence from silkmoths and aphids, it therefore seems likely that direct brain photoreception probably is involved. 4.1.2 Phase adjustment The adjustment of phase during entrainment consists of: (a) alteration of the natural period of the oscillator in response ‘to an environmental signal (the zeitgeber), and (b) maintenance by this means of a specific phase angle between the oscillator and the zeitgeber. Under natural conditions the oscillator only requires minor phase adjustments each cycle to keep it entrained to the solar day. Under experimental conditions, however, much geater phase changes can be elicited by dramatic and quite unnatural shifts in the zietgeber cycle. By this means the limits of the oscillator’s phase adjustability may be examined. It is the nature of circadian oscillators to be relatively stable-hence their free-running, self-sustained, temperature-compmsated characteristics. It follows that they will resist attempts to phase shift them, in much the same way that an oscillating pendulum will do. Indeed, one of the intriguing features of circadian rhythms is the extent to which their formal characteristics conform t o those of physical oscillators. Thus, for example, they show different amounts of response to a standard phase-shifting stimulus, just as a pendulum does, according to the time in their cycle when the stimulus is given. This gives rise to a phase-response curve of the kind shown in Fig. 10. The way phase-response curves arise in insects is outlined in more detail by Beck (1968: 68), and their implications for circadian theory discussed, for example, by Aschoff (196511).

JOHN BRADY

48

Hours offer stort of

DD

Fig. 10. (a) Phase-response curves for the adult eclosion rhythm of Drosophilu pseudoobscuru; each point represents the final steady-state phase shift of the free-running rhythm of a different population of pupae exposed to a single 15-min ( 0 ) or 0.5-ms ( 0 ) light signal given at the time indicated on the abscissa after the last "dusky at the onset of DD (redrawn from Pittendrigh, 1965: Fig. 5, and 1960: Fig. 14). (b) Phase-response curves for two behavioural rhythms; A, phase shifts of the free-running flight activity rhythm of Anophelesgumbiue in DD to single 1-h light signals given at the times indicated during the first 24 h in DD (redrawn from Jones e t al., 1972a); heavy line. phase shifts of the free-running oviposition rhythm of Pectinophoru gossypiellu in DD t o single 15-min light signals (redrawn, inverted, from Pittendrigh and Minis, 197 1: Fig. 4).

The physiology of the phase adjustment side of entrainment (as opposed to'the photoreceptive side) is almost totally unknown, and, in spite of the striking formal similarities, it is unlikely that pursuit of physical analogies will provide much help. Nevertheless, the phase-response curve is a potentially powerful tool. It must reflect the physiological responses of the circadian oscillator to entraining stimuli, and, as Pittendrigh (1966) has pointed out, it thereby provides the only means there is for recognizing a particular oscillator. Thus, for example, if two or more overt rhythms in an organism have identical response curves, it may safely be inferred that they are coupled to the same circadian clock. Experimentally, phase shifts are usually induced by one of two means (though Aschoff, 1965b, considers others). Either an entrained rhythm is shifted by changing the entraining light cycle to a new solar time (e.g. from LD to DL), or a free-running rhythm in constant darkness is shifted by a single light interruption. Physiologically, the two procedures are quite different. In the first, the new zeitgeber is repeated daily and includes a full photophase, whereas in the second the stimulus is a discrete light pulse. In the first, there may therefore be photokinetic masking effects due to

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

49

inhibition or excitation of the overt rhythm, which have little to do with what is happening to the unseen driving oscillator. In the second, on the other hand, shifts of the overt rhythm are uncontaminated by such exogenous interference, since they occur in ccnstant darkness. This latter procedure is thus much more informative about the oscillator itself. The most important single point about this second type of phaseresponse is that, in spite of the light stimulus occurring once only, the rhythm goes on shifting for several cycles before settling into a new steady-state oscillation. In other words, the physiological effect of a single light pulse (which may be as brief as 0.5 ms, Pittendrigh, 1960) on a free-running rhythm is not instantaneous: transient cycles result, which are shorter or longer than the natural period of the oscillator, until it stabilizes at its new phase. Pittendrigh (1965) has reviewed this problem in the light of the only rhythm in which the characteristics of phasr-responses are known with sufficient precision t o make meaningful predictions: Drosophila eclosion. He has developed a two-oscillator model which very adequately accounts for the known facts, and the reader is referred to his paper for a succinct exposition of the hypothesis. In essence, the model supposes that one driving oscillator is instantaneously shifted by the light signal to the phase position of the ultimate new steady-state rhythm, and that a second driving oscillator (directly coupled to the overt observable rhythm) is not so adaptable, but is coupled to the first and goes through transient cycles until it gets back in phase with the first. Whatever the true biochemical nature of these oscillators may be, consideration of their prolonged perturbation by single stimuli suggests that something of this sort must occur. There is a quite different kind of phase-response curve derived from giving different intensities of stimuli at standard circadian times. The amount of phase shift is not only dependent on the time in the oscillation’s cycle that the stimulus is given, but also on the strength of the stimulus. This can be seen clearly in Zimmerman and Goldsmith’s measurement (1971) of the photosensitivity of the Drosophila eclosion rhythm in normal and P-carotene-deprived cultures (Fig. 9 ) , or, in a behavioural rhythm, in Wobus’s work (1966a) on the effects of lighl intensity on the speed of re-entrainment of cockroach activity rhythms. Evidently, phase-response curves of different shapes will therefore be produced by using different stimuli. A clear example of this has been demonstrated in the Drosophila eclosion rhythm where light breaks of 12 h, 4 h, 1 5 min, and 0.5 ms each result in a different response curve of the free-running rhythm in DD (Pittendrigh, 1960: Fig. 14; 1965: Fig. 5; see Fig. 10). A minor complication to this phas&response system of Drosophilu has recently been revealed by Winfyee (1972b, 1 9 7 2 ~ )He . has found that the precise dimensions of the phase-response vanes according to

50

JOHN BRADY

whether the light stimulus is given during the first, second or third day in DD, implying some sort of long-term dark adaptation by the photoreceptor system. A more surprising feature of the phase-responses of Drosophila was revealed earlier by Winfree (1970a, 1970b). He started free-running emergence rhythms in mixed aged cultures of pupae by transferring them t’rom LL to DD (see p. 17), and then measured the phase shifting caused by giving different light stimuli (of S seconds) at different times ( T hours) during the first 24 h after the LL-DD transition. The phase-response to such signals saturates at about 10 000 erg cm-’, which is equivalent to about 100 s of the weak blue light used (Winfree, 1970b: 333), so that only very brief light stimuli were needed to induce phase shifting. He found that as stimulus S was increased from 5 to 4 5 s, the maximum phase shift grew to about 4 h, but the response curve continued to describe a smooth wave about the zero line (cf. Fig. lO(a)) (Winfree calls this Type 1 resetting). Increasing the stimulus to 55 s, however, resulted in .a switch to a curve showing a maximum phase shift of 12 h (his Fig. 10, 1970b), thereby making a delay shift in the first cycle of the rhythm run into an advance shift in the next (he calls this Type 0 resetting). In both types of curve, maximum phase shifting occurred when the light stimuli were given at T = c. 6.5 h after the LL-DD transition. Winfree conjectured that at this point there ought to be a critical stimulus of S = c. 50 s separating Type 1 responses from Type 0. He tested this hypothesis and discovered the astonishing fact that after a perturbation of T = 6.5 h, S = 50 s, the subsequent emergences occurred at random; the clocks in the individual pupae had apparently stopped. Winfree’s analysis of this phenomenon is a great deal more extensive and mathematically sophisticated than this brief synopsis suggests, and he goes on to consider several esoteric implications of his results. We will mention just one. It transpired that after perturbation at systematically varying T and S , a three-dimensional plot of the phase of the emergence peaks (with the axes: T, log S, and phase) described a helix up the phase axis. The centre of this helix is at the critical T = 6.5, S = 50, and Winfree calls this point T*, S*, or the “singularity point”, at which all the rhythms are at all phases or at no phase. Because of the increasing photosensitivity which develops during the first 3 days in DD (Winfree, 1972c; see above), it should be noted that these precise values of T*, S* apply only to stimuli given during the first 24 h after the LL-DD transition. Interestingly, T* seems to remain almost unchanged (i.e. 6.5 + 24n h); it is the S* that varies, falling from 50 s to 5 s over the 72 h. It appears that the Drosophila emergence clock is unstable at the phase of 6.5 h after initiation of an oscillation, i.e. the point at which it swings over from small phase shifts to large ones (see Fig. lO(a)), and that only a

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

51

very small quantity of energy is required to stop it at this point (Winfree, 1970a). This is perhaps analogous to stopping the hand of a stop-watch at 6 o’clock: if it is at exactly dead-centre it will riot return to zero when the return button is pressed, but given the slightest disturbance will either return clockwise (Type 0) or anti-clockwise (Type 1). The physiological reality of this phenomenon is a complete mystery, but Winfree’s tests show that after T*, S* perturbation the clock is essentially in the same stopped state that it is in pupae that have been reared since hatching in constant darkness (Winfree, 1970a: Fig. 14; see p. 17). Winfree goes on to suggest that the same phenomenon may occur in the clocks of other organisms, pointing out that published phase-response curves all fall into either Type 1 or Type 0, and that their characteristics are qualitatively similar. It does appear from his analysis, however, that Type 1 curves occur mostly in behavioural rhythms and Type 0 in developmental or plant rhythms. This distinction has considerable implications for Truman’s separation of clocks into two qualitatively different kinds (Truman, 1972a; see p. 76). Related to the general phenomenon of entrainment by light signals is the effect of the intensity of constant light on the frequency of free-running rhythms (“Aschoff’s Rule”; Aschoff, 1960). It seems that in many nocturnal animals frequency is inversely proporrional to intensity, whereas in diurnal ones it is commonly directly proportional. This “rule” applies roughly to the rhythms of several nocturnal insects (e.g. cockroaches and crickets; Hoffmann, 1965), but seems not to have been studied in diurnal ones. Dreisig and Nielsen (1971) have proposed a model to explain the operation of the “rule” in nocturnal cockroaches. Their model is, however, mainly a description of experimental observations, rather than the exposition of tangible physiological mechanisms; it depends upon the oscillation of a hypothetical “sensitization” process. These effects of light on circadian oscillators are presumably acting upon the means by which the phase of the oscillation is adjusted, rather than directly on the oscillator itself, though how sepuable these two functions are is not clear (see p. 76). In Protista, photoreception, entrainment and oscillation evidently all occur in the same cell (I hough this need not mean that their mechanisms are inseparable). In insects, on the other hand, there is room for these functions to be anatomicallly distinct. In behavioural rhythms, as we have discussed (p. 44), at least the photoreceptor and oscillator are separate, since the rhythm persists in blinded insects. In developmental rhythms, however, it looks as if these two factors are coexistent in the brain, probably in the central region of the protocerebrum in the case of silkmoths (Truman, 1971b, 1972d; see ppr 47, 85). Consideration of these two anatomical conditions has led rruman (1972a) to formulate the proposal that circadian clocks exist in t w o basic forms (see p. 76).

JOHN BRADY

52

4.2

CONTROL OF OVERT RHYTHMS BY DRIVING OSCXLLATORS

The inaccessibility of circadian oscillators has led to their being explored inferentially via their formal characteristics. As a result, much is known about the mathematical configuration of their oscillations, but almost nothing about their concrete biochemical mechanisms. On the other hand, the physiology of their coupling to the overt rhythms they control has proved relatively accessible to experimentation and is beginning to be understood in some cases-notably in insects. This is considered below in relation to: (1) clocks controlling gated rhythms; (2) clocks controlling behavioural rhythms; (3) clocks controlling other rhythms. 4.2.1 Control of gated events a. Moth eclosion. Not until research was extended to large species were any significant advances made in determining the physiology of the timing of eclosion. In silkmoths, however, it has been possible to perform the sort of ablation and transplant experiments that are the stock-in-trade of insect endocrinologists. This work has been the almost exclusive preserve of Truman, who has recently succinctly reviewed his work (1972a; and see Chapter 4), the outline of which is as follows. Under LD 17 : 7, Hyalophora cecropia characteristically emerge during the first 6 h of the day, with a peak after dawn, and Antherea pernyi during the last 5 h of the day; pupae transferred to DD show subsequent emergence peaks at 22-h intervals (Truman and Riddiford, 1970; see p. 82). Removal of the brain results in random emergence across the 24 h; its replacement results in emergence peaks close to the above times, and its transfer to the other species (also de-brained) results in the recipient adopting the donor’s gate time (Fig. 11) while retaining its own specific motor patterns. Moreover, injection of brain homogenates from one pharate adult into another induces ecdysis at any time of day, 1-2 h after injection, and de-afferented abdominal ganglia produce impulse patterns characteristic of eclosion behaviour when perfused with pharate brain extract (Truman and Sokolove, 1972). One may conclude that: (a) the clock gating eclosion resides in the brain; (b) it exerts its control hormonally; (c) it is entrained by direct photoreception (p. 47); (d) ecdysial behaviour is programmed outside the brain. Interestingly, it is only the gating which is hormonal, since eclosion itself ultimately occurs more or less normally, but ungated, when the brain is removed (Truman, 1971b, 1 9 7 1 ~ )It. appears that the subsequent release of bursicon may also be gated (Truman, 1972a: Fig. 12); presumably the same will prove to be true in other groups, e.g. Diptera (Schlein, 1972a, 1972b; see p. 22). The other aspects of this ecdysis rhythm elucidated by Truman d o not concern the coupling of the clock t o the gated event so

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS H cecropio

1

I

53 A pernyi

I

Fig. 11. Summary of operations demonstrating the hormonal role of the brain in gating silkmoth (Hyalophora cecropia and Antheraea bernyi) eclosion. Cross-hatching indicates the distribution of adult emergences in LD 17 : 7 (czbscissae), resulting from the operations indicated. Emergence occurred arrhythmically when the brain was extirpated ( Z ) , but the gating was restored by its replacement ( 3 ) ; replacement with the brain of the other species resulted,in the recipient pupae taking lip the gate time of the donor brains (4). (Redrawn from Truman and Riddiford, 1970.)

much as the photokinetics of the clock itself and are discussed in the relevant section below (p. 82). b. Pupation. Truman (1972a) went on to examine the gating of brain hormone (as opposed to eclosion hormone) and ecdysone release in another moth, Munducu sextu. He ligatured last instar larvae behind the head and between the second and third abdominal segmenls at different times of day and observed the degree of subsequent “pupation” of the thoracic and abdominal compartments. He found that thoracic: pupation built up from 0 to 100 per cent over the 12-h day of a LD 12 : 12 cycle. Apparently, brain hormone is released in Manduca during a specific period and enforces a similar timing upon the prothoracic glands (although, contrary to Truman’s interpretation of it (1972a: 126), Rensing’s evidence does not definitely demonstrate a daily rhythm in ecdysone release; see p. 85). No other work on the control of developmental rhythms compares with the strength of this experimental evidence from moths. The best alternative approach to the problem has been an attempt to determine histological correlates between endocrine tissues and developmental events. This research has principally been the concern of Rensing (e.g. 1966c), who has studied this aspect of Drosophila development (p. 8 5 ) . The evidence is

54

JOHN BRADY

necessarily circumstantial, and, in the final analysis, depends for its interpretation upon temporal correlations of physiological events which may or may not be causally connected. Nevertheless, since the role of hormones in development is an established fact, the discovery of die1 histological changes in the relevant endocrine tissues is prima facie evidence for their rhythmic secretion, or perhaps more correctly their gated secretion. Rensing’s histological observations cover the ultimate (1971) and penultimate ( 1 9 6 6 ~ )days of larval life. One would expect changes to occur in the relevant glands during the onset of prepupation and ecdysis, but it is not clear whether the glands themselves are gating clocks or simply hormonal effectors for pupation driven from elsewhere. Suspicions that the latter may be the case are aroused by the close synchrony of the nuclear volume changes of the several different tissues examined (see p. 35, and Fig. 7). This problem is discussed at length below (p. 86) in relation t o the hypotheses that Rensing has elaborated to explain his results. Nayar (1969) has briefly extended Rensing’s approach to his study of the pupation rhythm of Aedes taeniorhynchus. He found a 50 per cent change in the cross-sectional area of brain neurosecretory cell nuclei during the last larval day (this would represent a three-fold increase in nuclear volume). No doubt this is associated with secretion of hormones in relation to the incipient pupation, but again it is not clear where the clock fits into the system. c. Larval ecdysis. On the rare occasions when it has been examined from the circadian point of view, larval ecdysis has proved to be an ungated. event-taking place at any time of day, dependent only upon the rate of development (e.g. Nayar, 1967b; see p. 16). The physiology of this has now been investigated in silkmoths, and though, as in Nayar’s mosquitoes, the actual ecdyses are not gated, paradoxically the release of brain hormone apparently is (Truman, 1 9 7 2 ~ ) . 4.2.2 Control of behauioural rhythms Since insect morphogenesis is hormonally regulated, it is not too surprising to learn that the timing of pupation and emergence is also effected via hormones. These are once-in-a-lifetime, gated events and demanded a specific form of coupling with the underlying circadian oscillator. Duration of development is closely temperature dependent, and circadian control is not continuously applied to it. What happens is that once growth is complete, further progress is prevented until the appointed time of day when the ecdysial switch is thrown (in silkmoths, at least, by the release of a hormone). This is like timing the launching of a ship: it may have taken an indefinite period to build, but once complete it cannot move until the chocks are knocked away, and once started down the slipway it cannot be recalled .

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

55

Behavioural rhythms must be controlled bjr a very different form of coupling to the circadian clock. Circadian control must be continuously applied throughout the life of the insect, but must at the same time be adaptable so that the behaviour can adjust to enkironmental exigencies. The sort of case one has in mind is that of thi: tsetse fly. Under constant laboratory conditions, endogenous control results in a major morning peak of activity, a minor evening peak, and little or no activity at midday (Brady, 1972a). In the field, however, the morning peak is often absent and either the evening peak predominates or there is a peak at midday (e.g. Pilson and Pilson, 1967). This modulation of the circadian programme is apparently a response to temperature: when it is less than c. 20°C little flight occurs, though the fly is physiologicall-{ capable of it (Mellanby, 1936). Evidently, circadian control is adaptable, and if activity is prevented at the programmed time it may nevertheless be performed later if conditions are suitable. It is not, however, performed continuously under constant suitable conditions. This sort of control must be exerted via the central nervous system. Research on the subject has concentrated on the control of locomotor rhythms, particularly in cockroaches. Virtually everything that is known of the coupling between clock and overt behaviour in insects derives from this work, which has recently been reviewed (Brady, 1969); the evidence is re-presented and discussed below. a. The role of the sub-oesophageal ganglion. The search for hormonal mechanisms controlling insect activity rhythms dates from Harker’s report in 1954 that rhythmic activity could be induced in arrhythmic Penplaneta by parabiosing them to rhythmic animals. This original experiment was procedurally unsatisfactory for a number of reasons, as outlined by Cymborowski and Brady (1972), but when repeated with the necessary controls nevertheless yielded similar results. In both crickets (Acheta domesticus) and cockroaches (Penplaneta americana), headless animals take up the rhythm of the intact animal stuck on their backs, but highly significantly more rhythms are induced if the haemocoels are interconnected than if they are not. Some sort of influence affecting locomotor activity rhythmicly apparently z i passed via the haemolymph from the intact donor to the headless recipient (Cymborowski and Brady, 1972). Cymborowski and Brady are, however, cautious of Harker’s (1954) original conclusion: “. . . a secretion, carried either in the blood or tissues, is involved in the production of a diurnal rhythm of activity in the cockroach.” They suggest alternative explanations, of which the following seems particularly likely. The donor animal responds to being pinioned upside down by secreting a neuro-active stress factol: (Cook et al., 1969; Brady, 1967b), doing so in greatest amount when its CNS issues abortive commands at its normal activity time. This would be a hormonal response AIP-3

56

JOHN BRADY

to a stressful situation, but would not indicate that intact animals normally regulate their locomotor activity by such means. Harker followed up her parabiosis experiment with the still more impressive report that the transplantation of the suboesophageal ganglion from a rhythmic donor induced a rhythm in a headless arrhythmic recipient (1956). Further experiment suggested that if the two pairs of lateral neurosecretory cells (Brady, 1967a) were destroyed, no rhythms were induced when the ganglia were implanted (Harker, 1960a). The inference drawn was that just these four cells must therefore act as an autonomous, hormonal clock (Harker, 1960c: 164; 1961: 138; 1964: 64). Subsequent research, however, has failed to confirm the induction of rhythms in cockroaches by any sort of suboesophageal ganglion transplant (Leuthold, 1966; Roberts, 1966; Brady, 1967c; and implicitly by Nishiitsutsuji-Uwo and Pittendrigh, 1967; also in Romalea by Fingerman et al., 1958; and in Blaps by Thomas and Finlayson, 1970; and Fletcher, 1966, and personal communication); the few apparently ‘‘successful” transplants reported are not significantly more frequent than random expectation (Brady, 1967c: 173). Furthermore, the implicated neurosecretory ceils show virtually n o histological signs of a secretory rhythm (see p. 35), and may be removed from intact cockroaches by micro-cautery (as can all the other fuchsinophil cells in the suboesophageal ganglion) without in any way . a similar way, impairing the periodicity of the activity (Brady, 1 9 6 7 ~ )In the great bulk of cell bodies and neuropile can be cut off from the ventral part of the suboesophageal ganglion-leaving little more than the through tracts-without stopping the rhythm, at least in LD (Nishiitsutsuji-Uwo and Pittendrigh, 1968b). Quite clearly, the suboesophageal ganglion neurosecretory cells are in no way essential to the timing or control of cockroach locomotor activity. An alternative is that the suboesophageal ganglion is some sort of second-order and perhaps rather ephemeral clock (Brady, 1967c: 176). This view depends upon acceptance of the basis of all Harker’s experiments, namely that isolated suboesophageal ganglia can induce rhythms when implanted into headless cockroaches if the conditions are exactly right. Scepticism about this central point is widespread, yet it should be remembered that failure to repeat an observation does not prove it wrong. Harker’s work is often difficult to assess because, contrary to general practice in this field, raw activity records are not presented, nor is any indication of statistical variability given. A major weakness is the apparently frequent use of single peaks of post-operative activity (see Harker, 1956: 229) to assess the success-and phase-of a transplanted rhythm. Harker has, however, published nine analysed activity records (albeit without ordinate scales), and these show very marked rhythms in presumably headless Periplaneta, apparently induced by implanted sub-

THE PHYSIOLOGY OF INSECT CIRCADIAN R H Y T H M S

57

oesophageal ganglia (1956: Fig. 5b; 1960b: Figs 4c, 5b, 6c, 7b; 1960c: Figs la, lb, 2b; 1964: 26c). It would be difficult to dismiss these as the random outcome of chance, particularly where several circadian peaks ciccur (as in 1956, Fig. 5b). Equally, it is unlikely that interference from unexcluded environmental cues can be the explanation, since unimplanted headless animals were arrhythmic (as in 1956: Fig. 5a; 1964: Fig. 215b), and in some cases phase changes occurred (as in 1960c: Fig. lb). One is obliged to accept the existence of either: (1) some sort of periodic humoral output by transplanted suboesophageal ganglia, not necetjsarily neurosecretory, or (2) a stimulating effect by the transplant re-ini tiating rhythmicity in noncephalic clocks elsewhere in the recipient. The only alternative explanation is misinterpretation of the raw data. The lack of repeatability of the observations, however (including attempts by the reviewer under Harker’s dire
58

JOHN BRADY

locomotor rhythms. The best evidence for brain hormone involvement in behavioural rhythms would be the induction of rhythmicity by brain transplant into brainless (arrhythmic) animals. This either yields no rhythms (Harker, 1956), or ambiguous results (Cymborowski, 1970a; see p. 67). The evidence of Ball and Chaudhury (1973) adds an unfortunate confusion, since they allege that brain transplants in Blaberus elicit entrained rhythms in LD in “blinded” (i.e. eye-opaqued) hosts, but did not test to see whether the “blinded” hosts-whose own brains were left intact-could themselves perceive the light cycle; see also p. 46. The alternative approach is to test for rhythmicity after removal of the brain neurosecretory cells, though this is fraught with difficulty, not least in interpretation. In a series of micro-cautery operations, Brady found (1967b) that the great majority of the medial brain neurosecretory cells could be removed from Penplaneta without stopping the circadian periodicity of its locomotor behaviour. Six animals remained rhythmic in DD with only 5 per cent or less of their original 300 medial cells remaining histologically detectable. In a comparable experiment, Nishiitsutsuji-Uwo et al. (1967) examined 22 cockroaches remaining rhythmic (in LD) after surgical removal of the pars intercerebralis, and could find no medial neurosecretory cells in 11 of them (see their p. 687). Although eight of the remaining 11 animals had some stainable cells, the proportion of the original cell complement left was not indicated; the final three animals were “questionable”. Two interpretive problems arise. (1) Does the absence of stainable cells indicate that no neurosecretory cells are present, or simply that those remaining after the operation are releasing material so fast that their perikarya are empty? ( 2 ) Does arrhythmicity in the post-operative behaviour mean that the driving oscillator has been knocked out and/or the coupling from it broken, or simply that behavioural dis-coordination has resulted from the inadvertent destruction of central integrating circuits? On the first question, Brady (1967b) and Nishiitsutsuji-Uwo et al. (1967) take opposing views. Nishiitsutsuji-Uwo et al. favour the interpretation that failure to find stainable cells “is not a clear demonstration that they were not present”, and hence conclude that rhythmicity survives only when cells survive. Brady favours the alternative interpretation that the failure to find stainable cells does indicate a lack of neurosecretory tissue, and concludes that rhythmicity therefore survives in the absence of the cells. It is difficult to be certain which view is correct, but the latter one receives support from the evidence discussed below. On the second question, these authors agree (Brady, 1967b: 162; Nishiitsutsuji-Uwo et al., 1967 : 680). Arrhythmicity following ablation of a tissue does not necessarily mean that that tissue is involved in rhythm

.

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

59

control: the complex processes integrating behaviour may be upset by a variety of factors quite unrelated to circadian timing (Brady, 1971). One is on safer ground making the converse interprei.ation, namely that if a tissue can be removed without stopping the rhythm, it is dispensable to the timing of that rhythm (always bearing in mind the possibility of multiple control, see pp. 65, 93). Thirteen of the 19 animals made arrhythmic by Nishiitsutsuji-Uwo et al. (1967) were also hyperactive, and the three they examined histologically, while having no detectable neurosecretory cells, were found to have had their corpora pedunculata extensively damaged. Brady (1967b) notes that extending the ablation of Periplaneta brains beyond the area of their medial neurosecretory cells resulted in hyperactivil y of the kind induced by cutting the circumoesophageal connectives or by brain bisection. It is thus entirely possible that the hyperactive arrhythmia found by NishiitsutsujiUwo ct al. in Lrucophaea was merely a consequence of destroying behavioural integration, rather than specifically the removal of timing control. It is the other six arrhythmic animals to which Nishiitsutsuji-Uwo and Pittendrigh (1968b) subsequently drew attention. They pointed out that these six showed a normal amount of activity, and that it was only their rhythmicity that was disturbed. This is a valid point, but unfortunately no autopsies were performed, so that it is not known whether the animals retained any intact neurosecretory cells. In the absence of post-mortem proof of the removal of all neurosecretory cells, it would be rash to conclude that arrhythmicity is contingent upon their destruction. Nishiitsutsuji-Uwo and Pittendrigh (196EDb) found that mid-sagittal bisection of the protocerebrum through the pars intercerebralis failed to stop rhythmicity (in 12/19 animals in LD). Since this operation must sever the great majority of medial neurosecretory cell axons (where the NCC I cross over), they rule out medial cell participation. Instead, they propose that it is the lateral protocerebral neurosecretory cells which are concerned in rhythm control, and that it is these cells which have to be removed to ‘cause arrhythmicity. As we have just noted, there is no evidencis that the six normally active cockroaches made arrhythmic by pars intercei ebralis ablation had had their lateral cells removed. Moreover, Roberts et al. (197 1, and see Pittendrigh in their discussion) have shown that Lrucophnea are made arrhythmic by having their circumoesophageal connectives severed. If there were a hormonal role played in rhythm control by the lateral brain neurosecretory cells, one would expect it to be exerted via the NCC I1 (Willey, 1961: 234) and thence the corpora cardiaca. Disconnec:tion 6f the brain from the ventral nerve cord should not interrupt this route and would therefore not be expected to cause arrhythmicity. The fact that it does must discourage

60

JOHN BRADY

the view of rhythm control involving release of brain hormone to the blood. Further doubt about a rhythmic role for cephalic hormones is cast by the work of Azaryan and Tyshchenko (1970)on crickets,Acheta (Gryllus) domesticus. They report that ligaturing the neck did not stop the crickets’ locomotor rhythm (supporting evidence comes from the survival of respiratory rhythms in Drosophila with ligatured necks; see p. 24). While undoubtedly suggestive, these results should, however, be treated with caution. It is difficult to be certain that such ligatures are watertight,and really tight ligatures leave cockroaches as locomotorily inactive as if they had been beheaded (Brady, unpublished). Azaryan and Tyschenko (1970)went on to explore their cricket brains by micro-cautery, locating two areas in the pars intercerebralis, one either side of the midline, whose destruction caused loss of locomotor rhythmicity. This would agree with Nishiitsutsuji-Uwo and Pittendrigh’s findings, but Azaryan and Tyshchenko, having ruled out a hormonal role for these areas on the basis of their ligature experiment, found that the pars intercerebralis exhibited a die1 rhythm of spike discharge with maximum electrical activity coinciding with the period of minimum locomotor activity. The significance of this observation is, of course, difficult to determine. An electrical rhythm of some sort in the brain is certainly t o be expected and need not be given a causal interpretation any more than the electrical rhythm that occurs in the leg coxae (Sullivan et al., 1962). Nevertheless, the shape of the rhythm found by Azaryan and Tyshchenko is perhaps simpler t o interpret as an inhibitory electrical coupling rather than as being associated with hormone release. d. The role of the ventral nerve cord. The results of cutting the circumoesophageal connectives are not without their complications. In Periplaneta this operation leads to such intense activity that any underlying periodicity could pass undetected in the actograph records (Brady, 1967c; Roberts et al., 1971: Fig. 2). Hyperactivity is what the Roeder-Huber model of locomotor control (Roeder, 1967; Nuber, 1965) would predict from this operation. It is fortunate that the post-operative activity of Leucophaea is not so intense (Roberts et nl.. 1971), and one can safely conclude that in this case it is indeed arrhythmic (see pp. 58-59).It is interesting that in earthworms cutting the circumoesophageal connectives, or indeed cutting only one connective, seems similarly to lead t o a loss of circadian rhythmicity in locomotor activity and negative phototaxis (Bennett, 1970). Cutting the nerve cord connectives between the suboesophageal and prothoracic ganglia results in very low activity levels (again as the Roeder-Huber system predicts), in the same way that beheading does. So far as one can tell, the residual activity is arrhythmic (Leuthold, 1966;

THE PHYSIOLOGY

OF INSECT CIRCADIAN RHYTHMS

61

Brady, 1967c), though one must admit that the evidence is not conclusive. Brady found one out of 11 animals with a pos t-operative record suggesting a rhythm, and Nishiitsutsuji-Uwo and Pittcndrigh (1968b) say they “suspect the presence of rhythmicity” (eight animals). Two points may be made: (1) Brady’s 1/11 could reasonably be attributed t o chance coincidence of pen marks in the rather bleak records; (2) Nishiitsutsuji-Uwo and Pittendrigh were expecting t o find rhythmicity, since they thought it survived circumoesophageal connective severance; in any case they observed the post-operative activity in a LD cycle, and their two published records reveal an almost total absence of pen marks. It would be unwise to draw firm conclusions from this evidence, and it is certainly not strong enough to conflict with the finding that severing the circumoesophageal connectives does cause arrhythmicity. Cutting the ventral nerve cord connectiwes between the pro- and mesothoracic ganglia does not interrupt rhythmicity (4/6 animals remained rhythmic), nor does cutting the cord any further back (17/17 animals remained rhythmic) (Nishiitsutsuji-Uwo and Pittendrigh, 1968b). The implication is strong: nervous disconnection of the brain from the thorax stops rhythmicity in a way which no other nervous disconnections do. e. The role of the optic lobes.The most interesting of all NishiitsutsujiUwo and Pittendrigh’s observations concern their experiments on the optic lobes of cockroaches (principally Lcucophaea). They found (1968b) that if both optic nerves were cut and the compound eyes thereby severed from the optic lobes, the activity remained normally rhythmic but became uncoupled from the environment. The animals were in effect blinded and their rhythm free ran, unentrained by the 1ight:dark cycle in which they were kept. If, on the other hand, thg optic tracts were cut between the optic lobes and the protocerebrum, the animals became arrhythmic, even in LD.These are most suggestive results, and are 411 the more impressive since they have been confirmed by Roberts (1971), who, using Periplaneta, went further and showed that the lamina of the optic lobe could be dispensed with along with the compound eye; it was cuts proximal to the medulla that caused arrhythmicity. The essential role of the optic lobes in another insect behavioural rhythm, cricket stridulation, has also recently been demonstrated (Loher, 1972). It is quite possible, of course, that removal of the optic lobes upsets bilateral integration within the brain in some way, and that the onset of arrhythmicity reflects that, rather than specifically the destruction of circadian control centres. Three points suggest that the results have greater significance than that. First, the optic lobes have felatively few contralateral connections, most of their fibres connecting to ipsilateral protocerebral structures (see Brady, 1971). Quantitatively at least, their bilateral

62

JOHN BRADY

integrative role therefore appears small. Secondly, rhythmicity survives the much more drastic operations of medial bisection of the protocerebrum, or complete removal of one protocerebral hemisphere (Nishiitsutsuji-Uwo and Pittendrigh, 1968b). Bilateral interconnections within the protocerebrum are therefore not vital to the maintenance of rhythmicity. Thirdly, although the post-operative behaviour of the optic-tract-severed cockroaches was not ethologically analysed, their locomotion was measured in running wheels (by both sets of authors), and these only record reasonably coordinated walking movements. While showing no sign of any periodicity (in LD), the records suggest that roughly the normal amount of locomotion was occurring. There are no obvious signs, therefore, that arrhythmicity following severance of the optic lobes was due to gross interference with behaviour. Bearing in mind the risks in interpreting the essentially negative result of arrhythmicity, one may conclude that the inner regions of the optic lobes of cockroaches contain tissue which is vital to the circadian timing of locomotor activity. They may even, as Nishiitsutsuji-Uwo and Pittendrigh suggest, contain the driving oscillator itself, but confirmation of this exciting possibility awaits experimental proof. The lack of any further publication on this subject since early 1968 (apart from Roberts’s and Loher’s confirmations) is a disappointment. There is, however, one point, the significance of which has escaped general notice but which throws particular emphasis on Nishiitsutsuji-Uwo and Pittendrigh’s findings: decapod crustacea lose their locomotor rhythm when deprived of their eyestalks (e.g. Schallek, 1942;Naylor and Williams, 1968),and these organs contain the exact crustacean homologues of the insect optic lobes, including the lamina, medulla and lobula (see Bullock and Horridge, 1965: 1068; and see p- 68). Because the eyestalks also contain the major endocrine tissues of X-organ and sinus gland, the consequences of their removal have generally been given a hormonal interpretation. As far as rhythm control is concerned, however, the evidence is slender (see p. 67). If the optic lobes are the locus of the driving circadian oscillation, what is the nature of their output? Until recently it was supposed that the lobes contained no endocrine tissue (Brady, 1969). It has since been shown, however, that in Penplaneta the lobes each contain about 120 hitherto unsuspected but quite indisputable neurosecretory cells (Beattie, 1971). The possibility exists, therefore, and is pointed out by Beattie, that these cells might be involved in the circadian processes of the optic lobes (one is reminded again of the crustacea). Two facts suggest this to be unlikely, though not impossible. First, the cells are situated in a peripheral mass between the lamina and medulla; very many of them must therefore be removed by Roberts’s operation of cutting off the lamina, which does not

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

63

affect rhythmicity. Secondly, their output (:an scarcely be a rhythmic secretion of a blood-borne hormone, or this would be detected in animals which have had their optic tracts or circumoesophageal connectives servered. f. The cockroach circadian control system--summary and conclusions. The generally accepted facts are as follows: 1. No rhythms of importance are induced by the implantation of endocrine organs. 2. Rhythmicity is not disrupted by: (i) severing the compound eyes from the optic lobes; (ii) splitting the protocerebrum mid-sagittally into two; (iii) removal of one complete protocerebral hemisphere; (iv) severing the ventral nerve cord connectives anywhere posterior to the prothoracic ganglion; (v) removal of the cardiaca-allata complex; (vi) destruction of much of the medial brain neurosecretory lissue, including cutting the internal NCC I; (vii) destruction of the suboesophageal ganglion neurosecretory cells; (viii) in crickets, by ligature of the neck. 3. Rhythmicity is stopped by: (i) beheading; (ii) severing the optic lobes from the protocerebrum; (iii) splitting both protocerebral hemispheres bilaterally; (iv) removal of a large part of the pars intercerebralis; (v) severing the circumoesophageal connectives; (vi) probably, by severing the connectives between the suboesophageal and prothoracic ganglia. The case for hormonal coupling between the driving oscillator and the effector organs in the thorax rests on three grounds: (i) that parabiosis shows that blood-borne factors are involved; (ii) that suboesophageal ganglion transplants induce rhythms; (iii) thal: arrhythmicity ensues when the protocerebral neurosecretory cells are removed. These points may be answered as follows. (i) Although rhythmicity in parabiosed insects is enhanced by haemocoel interconnection (p. 55), rhythm control by hormones is no more strongly implied by this than are several less consequential alternatives. (ii) The phenomenon of rhythm induction by ganglion transplant is exceedingly elusive, and suboesophageal ganglion neurosecretory cell ablation is without effect (p. 56);'it seems unlikely that hormones from this ganglion can play any major role in rhythm control. (iii) Arrhythmicity does not ensue when the majority of the medial neurosecretory cells are destroyed (p. 58), and the status of the lateral cells after pars intercerebralis ablation has caused arrhythmicity is unknown; the production of arrhythmicity by optic lobe removal or circumoesophageal connective severance does not suggest blood-borne hormonal coupling from the brain by any of the usual endocrine routes. The case for electrical or at any rate axortal coupling is more strongly based, since any operations which disconnect the optic lobes from the

64

JOHN BRADY

thorax stop the rhythm, whereas other disconnections generally have no effect. Thus, cutting the optic tracts, splitting the protocerebrum bilaterally or severing the nerve cord between brain and thorax cause arrhythmicity, whereas cutting the optic nerves, splitting the protocerebrum medially or severing the nerve cord posterior to the prothorax do not. All that is necessary for cockroach locomotor activity to be maintained with circadian periodicity is for one protocerebral hemisphere, plus its ipsilateral lobula and medulla, to remain in nervous contact with the thorax. The medial protocerebral structures, the distal parts of the optic lobe, the deutocerebrum, and all the contralateral parts of the brain can be dispensed with (Fig. 12).

Fig. 12. Schematic representation of the insect brain showing the excitatory (+) and inhibitory (-) pathways that are inferred to control locomotor activity (from Brady, 1 9 7 1 , after Huber, 1965). Cross-hatching indicates areas that can be surgically removed without stopping the circadian locomotor rhythm of cockroaches. mb, Mushroom bodies; cb, central body; sg, suboesophageal ganglion.

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

65

The simplest interpretation of these observations is that a driving circadian oscillator exists in the optic lobes arid is coupled electrically to the leg muscles via the protocerebrum, nerve card connectives and thoracic ganglia. At this juncture we should recall the generalities of what is required of the circadian control of behaviour (p. 55): the control must not be too rigid or it will expose the animal unnecessarily to environmental hazard. There must be central modulating inputs to 1:he system which can take account of relevant sensory information. Accepting the optic lobe clock hypothesis for the moment (and there is no good current alternative), where should one look for central modulation of the coupling between oscillator and driven rhythm? Two suggestive bits of evidence are: (a) that bilateral protocerebral bisection causes arrhythmicity; (b) in Nishiitsutsuji-Uwo and Pittendrigh’s six animals (p. 59) pars intercerebralis ablation resulted in loss of rhythm without causing hyperactivity. The exact location of neither of these two protocerebral lesions is known, but, in view of what is inferred for the central control (non-circadian) of insect activity (Huber, 1965, see Fig. 12), it is tempting to suggest that circuits to or within the mushroom bodies were involved. The hypothesis for cockroach activity rhythm control put forward (Brady, 1971) is thus as follows. A circadian pacemaker resides within each optic lobe with its phase entrained to the light signals perceived by the compound eye. These two oscillators feed a circadian output-perhaps of arousal-to the mushroom bodies. The mushroom bodies integrate this rhythmic arousal with all the sensory in forination they monitor, and modulate their inhibitory output of the suboesophageal ganglion (the major excitatory centre in the Roeder-Huber system, see Brady, 1971), so that the latter’s control of thoracic locomotor activi1.y is converted to a rhythm. Inferential support for this view comes from Azaryan and Tyshchenko’s observation (p. 60) of an electrical rhythm in the pars intercerebralis, with maximum electrical activity occurring during the phase of minimum locomotor activity. One obvious implication of this model is that, in common with many other structures in bilaterally symmetrical arimals, circadian control is duplicated. Presumably, the two clocks normally keep each other in phase via connections across the brain. Since they can, however, function unhindered when such coordination is prevented by medial brain bisection, it is intriguing to speculate what would happen :If the two were subjected to different light cycles. It must be emphasized that none of this work reveals anything of the nature of the circadian clock itself (for which see p. 81): it concerns only the coupling between clock and overt rhythm-the cogs between the escapement and the hands, so to speak. While it appears most likely that this coupling is electrical, it must be admitted that the demonstration of

66

JOHN BRADY

neurosecretory cells in cockroach optic lobes (Beattie, 1971) and the apparent demonstration of neurosecretory axons in the circumoesophageal connectives of locusts (Michel, 1972) re-opens the possibility of hormonal components. In view of all the other evidence, however, it is difficult to see how these could involve any blood-borne factors. g. Hormones and rhythm control in other arthropods. The evidence for hormonal coupling between oscillator and overt rhythm in cockroaches is thus seen to be weak. However, this does not mean, as has been pointed out before (Brady, 1969, 1971), that hormones play no role in regulating cockroach behaviour, nor, of course, does it mean that the same situation must prevail in other taxa. There is indeed much good evidence for hormonal regulation of insect behaviour (see Chapter 4 by Truman and Riddiford). The question is whether this regulation concerns specifically the periodicity of the behaviour, that is its timing, or merely other aspects such as its amount of form. The prima facie evidence for hormonal participation in behavioural rhythms (for species other than cockroaches) comes from the histological detection of die1 cycles of endocrine activity (e.g. Rensing, 1966c; Cymborowski and Dutkowski, 1970a; Fowler et al., 1972; see p. 34). As already emphasized, however (p. 391, without experimental support this evidence cannot indicate whether such cycles are causing the behavioural rhythms or are merely caused b y the overall circadian organization of the animal. Such experimental support is generally sadly lacking. For example, no one has even tried to see if the endocrine cycles persist in the absence of a light stimulus in free-running animals (though see Akchiga, 1973). Where evidence does exist, it is either inadequate or open t o alternative interpretation. There are only three important cases: crickets, tidal crustacea, and scorpions. (g,l) Crickets. Experimental work on crickets (Acheta domesticus) is mainly Cymborowski’s, but unfortunately he recorded all his crickets’ activity in LD and apparently in a 3-4°C temperature cycle (Cymborowski, 1970a), so that it is difficult t o know whether the post-operative rhythms were endogenously driven or simply photo-thermo-kinetically stimulated. The facts are as follows. Crickets were made hyperactive, but superficially arrhythmic, by pars intercerebralis ablation. Autopsy of six such animals revealed no stainable medial brain neurosecretory cells, whereas six other animals whose rhythms had recovered post-operatively (after 2-3 weeks) all possessed stainable neurosecretory cells. Cymborowski therefore concurred with Nishiitsutsuji-Uwo et al. (1967; see p. 58) on their interpretation of the necessity of brain neurosecretion for the maintenance of rhythmicity. However, a closer analysis of the one “arrhythmic” record published by Cymborowski (1970a: Fig. 3) suggests distinct signs of an underlying rhythmicity. When the data are smoothed with sliding means, the 6 h

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

67

around “dusk” all show less activity than the other 18 h. It appears premature, therefore, to conclude that brain neurosecretory cell ablation in crickets necessarily destroys their rhythm-at any rate in LD. In moths (Hinks, 1968), the effect of medial brain cell ablation is quite different: the normal night flight disappears, but is replaced by a burst of activity for an hour or two after dawn. Cymborowski’s other important experiment (1970a) consisted of transplanting isolated brains into crickets which had been made hyperactive by pars intercerebralis ablation. For two complete LD cycles after transplantation the recipients showed a marked depression of their activity for the first few hours after “dusk”. It would be valuable to know if this effect still occurred in constant darkness. The sharp coincidence of the onset of activity depression with light-out, however, suggests that it was a strictly photokinetic effect, and that it would not persist in DD or LL. Nevertheless, this phenomenon, which has never been reported before, is extremely interesting. It must imply a blood-borne influence by the implanted brain. This might possibly be a hormone inhibiting the host’s activity, released by the implant when it is stimulated by the onset of darkness. But, bearing in mind the host’s possession of intact eyes and ocelli, and the lack of evidence for direct brain photoreception (p. 46), it seems more likely that it is the host that is responding to the dark signal, and that its response is enhanced by the increased titre of some hormone provided non-rhythmically by the implant. If the activity depression can be shown to persist in DD or LL, there will then be some grounds for inferring an endogenously timed humoral output by the transplanted brain. Meanwhile, because of the precise coincidence of the effect with light-out, it appears more economical to assume a light-induced response. Cymborowski’s interpretation (1970b) of his histological and autoradiographic evidence for secretory cycles in brain imd suboesophageal ganglion neurosecretory cells of crickets is discussed above (p. 38). As with other evidence of this kind, it is difficult to distinguish between cause and effect. (g,2) Tidal crustacea. The effect of eyestalk removal on the level of locomotor activity by decapod crustacea has been known for a long time: the operation causes loss of locomotor rhythmicity combined with either hypoactivity (e.g. Kalmus, 1938) or hyperactivity (e.g. Schallek, 1942). The existence of the main crustacean endocririe tissues (X-organ and sinus gland) in the eyestalks has generally prompted hormonal interpretations of this effect (see Naylor and Williams, 1968). Exactly how the eyestalks control activity is far from clear. The evidence is confusing, partly because the species used show seasonal behavioural differences (Naylor and Atkinson, 1972: Fig. 9), and partly because eyestalk ablation causes initial hyperactivity, Followed by a period of low activity for a few days and then hyperactivity again once the animals have

68

JOHN BRADY

recovered (Naylor and Williams, 1968). Results therefore depend on what sort of ablated animals are used for hormonal assay. Naylor and Williams (1968) found that injection of eyestalk extract made from donors at their phase of minimum locomotion at low tide caused a reduction in the hyperactivity of recuperated eyestalkless crabs (Carcinus maenas). They inferred the existence of an activity inhibiting substance(s) in the eyestalk; later work (Naylor e t al., 1973) suggested that extracts made at high tide were less inhibiting than those made at low tide. A different conclusion was drawn by Powell (in a half-page abstract, 1965), who indicates that eyestalk extracts were stimulating, and excited Carcinus to greater activity when eyestalk ablation had made them inactive, and did so irrespective. of whether the extract was made from donors at high or low tide. He inferred the continuous production by the eyestalk of an activity-promoting substance. However, when he moved his attention to the sinus gland, the neurohaemal organ which releases eyestalk neurosecretion to the blood, he did detect differences at different tide times. Gland extracts from high-tide donors (i.e. at the time of their maximum locomotion) were activity stimulating, whereas extracts made at low tide were activity suppressing (like Naylor and Williams’s whole eyestalk extract). It certainly appears that eyestalks contain substances which can stimulate unnaturally inactive crabs, and which can depress unnaturally active ones. It is also possible that these substances may be released from the sinus gland with a tidal rhythm (or circadian, see Arechiga, 1973). This does not necessarily mean, as Naylor and Williams point out (1968), that the eyestalks are autonomously rhythmic, but it does nevertheless imply the possibility of some sort of hormonal coupling between clock and behaviour. There are difficulties in unreservedly accepting this view, however. First, the conflicting results produced by the different workers; second, the lack of evidential support of Powell’s conclusions; third, the counter-indications from earlier work on crayfish. Schallek (1942) failed to affect the arrhythmic hyperactivity of eyestalkless crayfish (Cambarus) by either injection of eyestalk extract or implantation of whole eyestalks, and found, moreover, that cutting the eyestalk nerve, without otherwise interfering with the animal, had the same effect as total eyestalk removal. If activity-regulating eyestalk secretions are released to the blood via the sinus gland, one would expect implanted eyestalks to have at least some effect on the recipient’s activity; even more would one expect an effect in nerve-severed animals. It is tempting to draw the obvious parallel with the optic lobes of cockroaches (p. 62). Crustacean eyestalk ganglia are homologous with insect optic lobes, and severing the optic tracts in both groups causes arrhythmicity. Indeed, the parallel is closer still: in the cockroach the

THE PHYSIOLOGY OF INSECT CIRCADIAN R H Y T H M S

69

compound eyes plus laminae can be removed (p. 61), and in crayfish the retinae and laminae (Page and Larimer, 1972), without disturbing the rhythm in either case. All these results would be compatible with the hypothesis that a circadian clock lies within the inner layers of the eyestalk ganglia of crustacea and, as in the cockroach, is coupled electrically (or axonally) to the locomotor effector organs. The role of the eyestalk hormones would then be supposed t o be analogous to the effects of the corpora cardiaca hormones in insects, affecting the level of lcicomotor activity but not its periodicity (p. 57). Two minor points conflict with this hypothesis. First, Bliss (1962) reports the survival of a rather indistinct rhythm in a single crab (Gecarcinus lateralis) following eyestalk removal. If true, this would imply the existence of primary oscillators outside this eyestalks. However, the raw record from this crab (her Fig. 12) is far from being obviously rhythmic, and a close inspection of her Fig. 11 suggests that its supposed residual rhythmicity might be a visual artefact introduced by the method of presentation. Bliss herself is extremely cautious in concluding that this record is truly rhythmic; it is certainly not strong enough evidence alone to prove that decapod rhythms survive eyestalk removal, particularly in view of the much larger number of ablations carried out by other workers without a single failure (e.g. Naylor and Williarns, 1968). The second point appears in the discussion of the same paper (Bliss, 1962), where Bliss reports that Powell succeeded in inducing rhythms in eyestalkless Carcinus by implanting eyestalks (i.e. contrary t o Schallek, 1942). However, these results have never been published, or confirmed, and Powell himself (1965) makes no mention of them. These two potentially damaging points are thus seen to be of little real substance, and the hypothesis remains effectively unscathed-at least until stronger evidence than this is produced to confound it (Bregazzi’s recent model (1972) for the hormonal control of the locomotor rhythm of the beach amphipod, Talitrus saltator, has as yet no direct experimental support, and therefore provides no other conflict). (g,3) Scorpions. Evidence for the hormonal control of scorpion activity is based on physiological rather than behavioural observations, and is restricted to the single short paper by Rao and Gropalakrishnareddy (1967) outlined on p. 41. The real significance of this very interesting piece of work is difficult to assess, although its implications for the present discussion are evident. The authors consider that their results indicate the cyclical release of neurostimulant and neurodepressant factors into the blood (consistent with the effects of neurohormones C, and D, in the cockroach, Strejc‘kovi et al., 1965), and thai: the resultant modulation of the electrical activity in the nerve cord “may form the basis for the

70

JOHN BRADY

regulation of the circadian locomotory rhythm . . . in this scorpion”. This amounts to suggesting that the circadian oscillator is coupled to the locomotor effectors via blood-borne factors. It may be the correct interpretation, but there are points which arouse confusion. First, the cycle of concentration in the blood is in phase with the cycle of amounts extractable from the brain (p. 41). This is at variance with the assumption behind many histological investigations, namely that high accumulation of stainable material in neurosecretory cells signifies, low release of secretion. Secondly, it is surprising that in four out of the five time samples, the extract of one brain should have had exactly the same quantitative effect as a 0.4-0.5ml sample of blood (Fig. 8). However, six pooled pairs of donors were assayed separately to provide each time sample (Rao, personal communication), so it is unlikely that the differing effects at each time were due to artefacts such as contamination. Thirdly, there is the implication of Venkatachari’s account (1971 and personal communication; see also Rao, 1964)of the die1 spike rhythm in exposed, completely deafferented nerve cords. Spontaneous spike output increased, particularly by recruitment of additional large units, when the locomotor activity of the intact scorpion would normally have increased. This increased electrical activity was greatest in the anterior connectives of the cord and diminished progressively, posteriorly. This wggests, as Rao notes, “the possibility of the pacemaker, and perhaps the clock, being located in the sub-oesophageal ganglion or brain”, and, according to Venkatachari, “the trigger for spontaneity is flowing from the central nervous system to the peripheral”. The implication of an electrical coupling from the clock is clear, and there seems no obvious role for blood-borne hormones in such a system (though see p. 33 for Venkatachari’s hypothesis of brain hormones and AChE activity). A speculative interpretation of this work might thus run as follows:

1. The driving circadian oscillator is situated in the brain, and is electrically coupled to the locomotor effectors via the nerve cord. 2. The nerve cord is susceptible to influence by hormones (see StrejEkovi et al., 1965),at least when isolated. 3. The brain cyclically produces hormones with depressant and stimulant characteristics. 4.The cooling and manipulations involved in blood sampling induce release of neuroactive stress factors (Cook et al., 1969) from the brain. 5. About 50 spikes per second is the maximum possible output of the isolated cord under the influence of these hormones. The brain extract curve would then reflect the quantities of these factors accumulated in it, and be, in effect, a pharmacological confirmation of the

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

71

histological observations on scorpion neiirosecretory tissue (see Venkatachari, 1971). The blood curve, on the other hand, would reflect the relative excitability of the scorpion at the time of sacrifice, with sufficient stimulant being released at the phase of peak locomotion (1700-2000 h) to switch the assayed nerve cord t o maximum spike output, and at other times variable quantities of the depressant being released. The role of this system in coupling oscillator to effector in the normal intact scorpion might therefore be very similar to the one suggested above for cockroaches and austacea, that is electrical coupling from oscillator to effector but with behaviour susceptible to hormones when under conditions such as “stress”. It would be instructive to see whether scorpion brain transplants elicit locomotor rhythms or not.

4.2.3 Control of other physiological rhythms Coupling between clock and rhythm has scarcely been looked at for the vast majority of the other circadian rhythms. Some of these are rather general events such as coincident nuclear volume cycles in a wide range of tissues (e.g. Rensing, 1966c), colour change in hypodermal cells (e.g. Mothes, 1960), deposition of endocuticle (e.g. Neville, 1970), susceptibility to cholinesterase inhibitors (e.g. Fondacaro and Butz, 1970), susceptibility to X-irradition (e.g. Rensing, 1969a), or general metabolic level in dormant animals (e.g. Hayes e t al., 1972). With the exception of the epidermal events, for which a little evidence does exist, virtually nothing is known about the control of any of these rhythms. For most of them, however, it is difficult to see how the multitude of cells involved could be kept in phase with each other and coupled to a driving oscillator by any means other than humoral. Over small areas of tissue intercellular chemical coupling of some sort can presumably occur. This could explain the localized differences in cuticle deposition by the epidermis (p. 21) and is consistent with the supposition that all cells are potentially rhythmic. However, this is not enough to explain the coordination of such rhythms from end to end of an animal, and blood-borne communication must surely be invoked. For colour change there is real evidence that this is the case. In the stick insect (Carausius morosus) factors are cyclically present in the nervous system, which, when extracted, will influence pigment migration in the cells of isolated segments of hypodermis (Mothes, 1960); the production of one of these factors has been localized to the tritocerebrum (Raabe, 1963a). In what way these substances are normally released and exert their effect is not clear. Although parabiosis to an i n t x t animal induced a colour change rhythm in an otherwise permanently pale headless recipient, no implants of endocrine organs were similarly inductive (Smith, 1967). There is therefore no direct evidence for the source of rhythm control, although it

72

JOHN BRADY

would appear that the chromactive substance emanates from the tritocerebrum, and Raabe states (1963b) that the axons of the tritocerebral neurosecretory cells leave the brain via the NCC I11 and go to the corpora cardiaca. Smith, however, reported that removal of the cardiaca-allata complex had no effect on the rhythm, whereas cutting the circumoesophageal connectives led to permanent paleness. The evidence is thus confusing, but does suggest hormonal control, without clearly indicating via which route the relevant hormones reach the blood. It is further confusing that other species, which do not show colour changes, should apparently produce the same chromactive substance(s) in their tritocerebra (Raabe, 1963~). 4.3

TEMPERATURE EFFECTS ON INSECT CLOCKS

The feature which marks out circadian oscillations from all other biological periodicities is their “temperature compensation”. Circadian rhythms exhibit Qlos which are close to unity in normal biological ranges of temperature (e.g. 1.02 for Drosophila eclosion, Zimmerman et al., 1968; 1.04 for cockroach locomotion, Roberts, 1960; 1.04 for the lamination of cuticle deposition in locusts, Neville, 1965). This characteristic is diagnostic, and so may be used to reject candidate rhythms that have been erroneously considered as participant in circadian systems. The oviposition rhythm of Oncopeltus fasciatus, for example, which is retained indefinitely in LL, and which therefore might be considered a typical circadian behaviour rhythm, proves not to be so, since it follows egg development which is strictly temperature dependent (Rankin et al., 1972). Similarly, where one might have hoped to find a component for the optic lobe clock controlling behaviour (p. 65) in the spontaneous “clock spikes” which occur in house fly optic lobes, these are ruled out because they have a Qloof c. 2 (Hengstenberg, 1971). The subject of temperature compensation attracted much research up to about 1960, but once its existence and significance had been demonstrated there seemed little else to be done with it, at least until the biochemical nature of the basic oscillator was better understood. Temperature compensation has, however, been incorporated comfortably into recent biochemical (Hastings, 1970), mathematical (Pavlidis et al., 1968), and other hypothetical models (Dreisig and Nielsen 1971), though little research has been camed out on its nature in recent years. The fundamental paradox is that, although the period and frequency of circadian oscillators are protected against differences in the constant levels of ambient temperature, their phase can be shifted by temperature changes and entrained by temperature cycles (e.g. Roberts, 1962; Zimmerman et

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

73

al., 1968). Furthermore, in at least one case, Drosophila pseudoobscura, a rhythm of emergence can be induced in individuals which have been reared in DD and shown t o be arrhythmic, by subjecting them to a 12-h rise of 8°C (Zimmerman, 1969). That is to say, their clocks are started by experiencing a temperature shock in an otherwise uniform (dark) environment. This is like the situation in the beetle Slaps rnucronata in which a 1-h pulse of l o o C increase is highly effective at re-starting damped-out locomotor rhythms (Thomas and Finlayson, 1970). The paradox was explored further by Zimnierman et al. (1968), who found that the Drosophila rhythm in DD was phase advanced by an 8°C step up and phase delayed by an 8°C step down. Twelve-hour pulses of 8°C increase, on the other hand, caused either an advance or a delay according to the circadian time at which they were given, that is, they gave rise to a typical phase-response curve (p. 48) with shifts ranging from c. 0-3.5h (Zimmerman et al., 1968). These authors' interpretation of the paradox incorporates negative temperature feedback in a mathematical model for the oscillator based upon the interaction of infradian oscillators (Pavlidis, 1969; see p. 88). In later work, on a different strain of Drosophila pseudoobscura, Winfree (1972a) found the effect of 8OC higher temperature pulses much simpler than Zimmerman et al. had observed. In Winfree's strain, a 12-h pulse resulted in very large phase shifts of around 12 h, irrespective of the circadian time at which the pulse was given. The reason for this difference between strains is not clear, but the response of Winfree's strain is physiologically simpler in the sense that the higher temperature upset the flies' clocks in such a manner that when returned to normal temperature they all re-started at the same phase, with subsequent emergence peaks recurring at 1 2 + 24n h after the step down. This is slightly different from the phase at which Drosophila clocks re-start aiter an LL-DD transition; in this case emergence peaks recur at 15 + 24n h .after the transition (p. 17). In the locomotor rhythm of the crab, Carcinus maenas, temperature steps of up to f l 5 " C have a quite different effect. The first post-treatment peak is shifted in the opposite direction from the Drosophila rhythm of Zimmerman et al., but subsequent peaks continue at the old phase. The underlying oscillator in the crab is thus apparerdy unaffected by this sort of temperature treatment (Naylor et al., 1971). Interestingly, a single 6-h 4°C chill serves t o initiate a tidal rhythm in crabs which had previously shown only circadian frequencies (Williams and Naylor, 1967). No similar phenomenon has been reported in any other arthropod, but it does occur in plants (Wilkins, 1965). Free-running insect circadian rhythms respond t o this kind of chilling, either by their clocks stopping for the duration of the chill or by various

74

JOHN BRADY

degrees of phase shifting (see Biinning, 1967: 54; Harker, 1964: 51). These low-temperature (i.e. near 0°C) effects seem to have received little attention in recent years, in spite of their potential value as an experimental tool. An exception is the work of Youthed and Moran (1969a) on ant-lion larvae (p. 8 ) but in their case the phase-shifting effects of low temperature bore no clear relation to the duration of the chill. Harker ( 1 9 6 0 ~ )used a more refined method and chilled cockroach suboesophageal ganglia in situ. This was done with a c. 0.5 mm diameter copper probe at 3°C placed against the ganglion (presumably exposed, though this is not stated) for up to 18 h. Various phase shifts in the subsequently transplanted ganglia were then detected. However, since the brain and corpora cardiaca are less than 2 mm away, they must also have been chilled. Any phase shifting cannot therefore be attributed exclusively to the cooling of the suboesophageal ganglion: the optic lobe clock (p. 65) will presumably also have stopped, and return to normal temperature must surely have resulted in the release of stress factors (Cook et al., 1969) which are known to influence the performance of rhythmic locomotor activity (Brady, 1967b). No controls were done with immobilized,’ unchilled animals (see Beament, 1958) nor with ganglia chilled after removal from the animal. In view of this, and the fact that chilling whole cockroaches has widely different phase-shifting effects (from 6-13 h delay for 8 h at 3°C) according to the circadian time at which the chill is given (Biinning, 1958, 1959), the interpretation of this experiment should be made with caution. The assay of rhythmicity in the suboesophageal ganglion by transplantation is considered above (p. 56). The effects of thermal manipulations on circadian rhythms in general were last extensively reviewed by Wilkins (1965). 4.4 GENETICS OF INSECX CLOCKS In the sense that different species of insect have rhythms which differ in form or phase, it is evident that clocks must be genetically defined. One can go further and say that differences between the races of some species may be expressed genetically in different rhythm characteristics, as for example in the geographical races of Clunio marinus (Neumann, 1967, 1971a; see p. 19). Apart from this and the few exceptions mentioned below, the subject has not been extensively studied. Results of Coluzzi’s crossing (1972) of two strains of Anopheles stephensi have essentially similar implications t o Neumann’s work on Clunio. Adults of the homozygous “Standard” karyotype showed a morning peak of emergence, those of the “Karachi” karyotype an afternoon peak and those of the St-Kr heterokaryotype an intermediate peak at midday. Selection for early or late emergence from a field strain

THE PHYSIOLOGY

OF

INSECT CIRCADIAN RHYTHMS

75

containing both karyotypes greatly increased the frequency of St or Kr homokaryotypes respectively. One feels that by inference from genetic characteristics this sort of study should reveal something of the nature of circadian clocks. The results of experiments by Pittendrigh (1967; see Pittendrigh and Minis, 1971) on Drosophilu, however, are less encouraging. He selected early and late emergers from a laboratory stock and succeeded in producing two strains whose peak emergence times were separated by c. 4 h (wild strains are much less variable for this character, presuinably because in the wild non-optimum deviations in emergence time are continuously selected out (Clayton and Paietta, 1972)). The phase angle of the overt rhythm to the zeitgeber was therefore genetically determined, as in Clunio marinus and Anopheles stephensi. However, the inference that this overt rhythm reflected the condition of the underlying driting oscillator proved to be quite wrong. When he studied the phase-response curves of the two strains he found them identical in both amplitude and phase angle to the LL-DD transition (Pittendrigh and Minis, 1971: Fig. 3). Evidently selection did not affect the driving oscillator, only the phase angle of the driven rhythm to it. Very possibly, critical analysis of the Clunio and Anopheles strains will reveal a similar situation. This does not mean that clocks are not genetically defined-of course they must be-but it does mean that the inherited phase relationships of overt rhythms may be a misleading guide to the characteristics of their driving oscillators. A rather more specific genetic link with rhythms is implicit in the work of Rensing et al. (1968). Using mutant male and female Drosophila they showed that the ratio of X-chromosomes to autosomes was correlated with the phase of peak oxygen consumption. The higher ratio strains showed a greater tendency to an evening peak and the lower ratio strains a greater tendency to a morning peak, with intersex!-s and wild-type male and females exhibiting both peaks (see p. 23). Crossing experiments suggested that it was the size of the evening peak lhat was regulated by the X-chromosome, and the authors speculate that the autosomes therefore regulate the morning one. Although, as with Pittendrigh’s different emergence strains, this effect may have little to do with the clock itself, the fact that it is the X-chromosome that is implicated is of interest in view of Konopka and Benzer’s findings discussed below. Using a completely novel approach to the understanding of rhythms, these latter two workers (Konopka and 13enzer, 1971) constructed Drosophila mutants by the use of a mutagenic agent and then scanned the offspring for circadian aberrations. They succeeded in producing three rhythm mutants: (1) arrhythmic, with no dei:ectable emergence rhythm;

76

JOHN BRADY

(2) short-period, having a free-running emergence rhythm with a period of C. 19 h; (3) long-period, having a free-running emergence rhythm with a period of c. 28 h; the normal strain free ran at c. 24 h. Of even greater interest is the fact that the locomotor rhythm of the three mutants showed exactly the same characteristics as their emergence rhythms: arrhythmicity, and periods of 19 h and 28 h respectively. Thus the same driving oscillator seems t o gate emergence as later controls adult daily behaviour. Furthermore, the eclosion clock is entrainable right back in the early larval stages (Brett, 1955; see p. 17), so, as Konopka and Benzer point out, the oscillation apparently persists throughout development, metamorphosis and adult life. This is by no means what would have been expected from the different characteristics of developmental and behavioural clocks in other insects (p. 77). While it is true that the similarity of period in the two rhythms measured in each mutant strongly implies that the same clock controls both, one would be happier about this interpretation if the phase-response curves of both rhythms were also proved to be identical. It is possible to imagine, for example, that there are many different oscillators in an insect (see p. 93), . each running on fundamentally similar biochemical mechanisms. If this were the case, the genetic control of their period could be common to all-without implying anything about their number. It is, therefore, perhaps still slightly too early to be certain that both eclosion ahd behaviour of Drosophila are timed by one and the same clock. Be this as it may, Konopka and Benzer (1971) pursued the matter further and succeeded in locating the site of the relevant mutations. Recombination and complementation tests showed this t o lie within a single narrow band on the X-chromosome (cf. Rensing et al., 1968; see above). The rhythm characteristics of the various heterozygotes suggested that the =rhythmic and long-period genes were simple recessives to the short-period and normal, and that the latter two interacted t o produce an intermediate effect. It rather appears that the three mutations affected the same functional gene, and hence that a prime parameter of the clock(s)the period-is therefore determined by a single gene. Finally, Konopka and Benzer indicate that preliminary construction of genetic mosaics suggested that the rhythm type corresponded to the genotype of the head. 4.5 TRUMAN'S TWO CLOCK TYPES Consideration of the anatomical apposition or separation of oscillators and photoreceptors led Truman (1972a) to formulate the proposal that circadian clocks exist in two forms (see p. 51). His Type I clocks have photoreceptor and oscillator anatomically co-existent, in the same cell in unicellular organisms, and possibly in metazoa also. The prime consequence of this is that light in effect acts directly on the oscillator-which may

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

77

actually involve the photoreceptive pigment-and the organism’s response to constant light is for its circadian clock to stop. This is known to occur in several protozoa (see Truman, 1972a) and in the eclosion rhythms of Drosophilu (Pittendrigh, 1966) and silkmoths (Truman, 1972b); interestingly it probably also occurs in the chitin 1amellogt:nesis rhythms of locust epidermal cells (p. 21). The secondary consequence is that such rhythms typically show a saw-toothed phase-response curve with a maximum shift of eight or more hours, and a sharp swing from phase advance to phase delay (i.e. Winfree’s Type 0 resetting; see p. 50 and Fig. lO(a)). Truman’s Type I1 clocks, on the other hand, have photoreceptor and oscillator anatomically separate. This is typical of the behavioural rhythms whose entrainment we have considered (p. 44). The separation of the photoreceptor from the oscillator protects the latter from the direct action of light, and as a consequence the rhythm free runs in LL. Truman points out that this condition has never been observed in unicellular organisms but is typical of behavioural rhythms in insects and other metazoa. In contrast to Type I clocks, these rhythms show low-amplitude phase-response curves (i.e. Winfree’s Type 1 resetting; see p. 50 and Fig. 10(b)) with maximum phase shifts of around 2 h, resulting in the sort 3f gradual entrainment illustrated, for example, by Brady ( 1 9 6 7 ~ ) . In general, insect clocks seem to fall conveniently either into Truman’s Type I controlling once-In-a-lifetime gated events such as eclosion, or into Truman’s Type I1 controlling daily repeated behavioural events. Winfree’s review of published phase-response curves (1970h: Table C1; see p. 51) implies that this dichotomy may extend beyond the Insecta. So far, sufficient is known about only one insect developmental clock-the silkmoth’s-to fit it on all criteria into Type I, but it seems likely that the Drosophilu clock will also prove to be of this type when more is known of the anatomy of its photoreception (see p. 47). The pupation rhythm of Aedes tueniorhynchus also conforms to Type I, at least with respect to its phase-response curve and annihilation by LL (Nayar, 1967b, 1968). On the other hand, the emergence rhythm of Pectinophoru is a possible exception: its phase-response curve, though incomplete (Pittendrigh and Minis, 197 1: Fig. 4), appears to be a smooth-wave relatively low amplitude one, more typical of Type I1 criteria. It does, however, show shifts of up to c . 4 h. Truman includes two behavioural rhythms in his Type I list, Pectinophoru oviposition and Curuusius locomotion, but i t would be fairer to his proposal to say that not enough is known about either to classify them firmly. The Pectinophoru rhythm shows a smooth-wave, low-amplitude, albeit very noisy Type I1 phase-response curve 1 Pittendrigh and Minis, 1971; see p. 48), and the Curuusius loss of rhythmicity in” LL need not be taken as a good Type I criterion since it is due more to inactivity than to true arrhythmicity (Eidmann, 1956).

78

JOHN BRADY

One consequence of this apparent duality of insect clock systems is that in any individual, a Type I clock gating its development must switch off at eclosion and a Type I1 then take over to control adult behaviour. Thus in Pectinophora gossypiella, Pittendrigh and Minis (1971) note that the oscillators driving the egg-hatch, adult eclosion, and oviposition rhythms are probably different. All three differ in one or more ways with respect to: (a) their free-running periods in DD; (b) their selectability for strains showing early and late phase angles to the zeitgeber; (c) the response of their free-running rhythms to red light. It thus looks as if there are at least three driving oscillators in Pectinophora, including two in the adult. These should fall into Type I (egg-hatch and eclosion) or Type I1 (oviposition). The superficial similarity between the three phase-response curves (Pittendrigh and Minis, 1971: Fig. 4) need not discourage this expectation, since the size of their variances precludes accurate comparison. It would be instructive to know at what point in development the oviposition rhythm is entrainable. The other examples which suggest a similar ontogenetic distinction between developmental and adult behavioural clocks occur in mosquitoes. In Aedes taeniorhynchus, adult eclosion is in effect timed by the clock gating the last larval moult, since pupal development is aperiodic and proceeds independently of circadian time (Nayar, 1967b; see p. 20). The activity rhythm of the adults, however, does not start until 36 h after eclosion, even under LD 12 : 12, and in DD none is initiated spontaneously; a light-dark transition has to be “seen” 36 h or more after eclosion (Nayar and Sauerman, 1971). This suggests that the developmental clock stops at pupation, and that the adult behavioural one does not start until some time after emergence. That these two clocks are different is suggested by the fact that in DD the free-running period of the developmental rhythm is 21.6 h (Nayar, 1967a) and that of the activity rhythm is 23.5 h (Nayar and Sauerman, 1971). A similar inference is possibly to be drawn from Chiba’s work (1966b) on the late-pupal initiation of flight rhythms in Culex pipiens. One would expect these clocks to fall into Type I (developmental) or Type I1 (adult behavioural), but this remains to be determined. Confusingly, the evidence from Drosophila (Konopka and Benzer, 1971; see p. 76 and below) implies that the same Type I clock may gate emergence as controls adult behaviour, surviving through development and metamorphosis. This would not have been anticipated from the observations on Pectinophora and mosquitoes just considered, but different species have presumably adopted different emphases between the available clock systems during the course of their evolution. As far as behavioural rhythms are concerned, none has yet been proved to be entrained by direct brain photoreception and in the one case where the critical experiments were carried out (cockroach) extra-ocular photo-

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

79

reception was shown not to entrain (see p. 44).Moreover, many insect behavioural rhythms survive well in constant light, e.g. cricket locomotion in 150 lux (Nowosielski and Patton, 1963); cockroach locomotion in 250 lux (Roberts, 1960); grain weevil locomotion, geotaxis and phototaxis in 200 lux (Birukow, 1964); ant-lion digging in 1500 lux (Youthed and Moran, 1969a); bee feeding in 1500 lux (Beier, 1968); and grasshopper oviposition in 1750 lux (Loher and Chandrashekaran, 1970). Conformity with Type I1 criteria may therefore be the general rule for behavioural rhythms in insects. The Diptera, however, present difficulties, since the great majority of behavioural rhythms in this order are very rapidly damped out by constant light. In strictly nocturnal species, such as mosquitoes (Chiba, 1964; Jones et al., 1967) whose activity may be inhibited by light, this is possibly to be expected. But in Aedes taeniorhynchus, which is mainly crepuscular, the arrhythmic activity in LL is greater in amount than the rhythmic activity in LD (Nayar and Sauerman, 1971), a condition which also occurs in the crepuscular mating rhythm of the fruitfly, Dmus tryoni (Tychsen and Fletcher, 1971). And in three strictly diurnal flies, Musca domestica (Parker, 1962), Phormia regina (Green, 196421) and Clossina morsitans (Brady, 1972a), constant light (in Musca and Glossina of only c. 100 lux) rapidly damps out the rhythm without materially affecting the amount of activity. Do these Dipterous clocks invalidate Truman’s proposal, or are they exceptions to it? The question cannot be answered until the site of photoreceptor and oscillator is known and it is discovered whether the arrhythmic behaviour results from the insect’s circadian clock stopping or simply from the overt rhythm becoming uncoupled from the clock. This latter can only be done by the sort of experiment performed by Pittendrigh (1966) to show that 12 or more hours of light stops the clock controlling Drosophila eclosion, since after this exposure the rhythm re-starts from exactly the same phase point whenever the culture is returned to constant darkness, no matter at what circadian time the LL-DD transition occurs (though see p. 18). This test has not been performed with its full vigour on any of the insect behavioural rhythms that damp out in LL. However, experiments which amount to a partial test have been performed on a few Diptera. Parker’s observations (1962) on Musca are the clearest examples. His Fig. 11 shows that a 12-h phase shift of the light cycle evoked an immediate 180’ shift in the activity rhythm, which remained stable at it; new phase when the flies were placed in DD (dim LL). Results with a similar implication can be detected in Chiba (1966a: Fig. 3), and in Jones et al. (1967: Fig. 4). Because the phase angle of these Dipterous rhythms to the “dawn” or “dusk” zeitgebers is effectively O ” , it is difficult to be sure that the clock was stopped by 12 hours’ light in the way that the Drosophila eclosion

80

JOHN BRADY

clock seems to be (cf. Pittendrigh, 1966: Fig. 5 ) . There is the alternative possibility that they respond to big zeitgeber time changes with equally big instantaneous phase shifts. This would mean phase-response curves with a slope of 1 and the ability to complete up to 12 h of phase shift within one cycle. However, the only Dipterous behavioural phase-response curve in existence, that ofAnophcles gambiae to 1-h light pulses in DD (Jones e t al., 1972a; see Fig. 10(b), p. 48) has a low amplitude with maximum phase shifts of about 2 h, so that it does not look as if this can be the correct interpretation. Truman’s main reason for distinguishing two clock types was the evidence for the unity of photoreception and oscillation in Type I clocks, and for their separability in Type 11. The other features of his classification are, so to speak, consequential embroideries upon this fundamental anatomical distinction, and should probably be considered as general indicators rather than absolute “taxonomic” characters. The amplitude of phase-response curves, for example, depends partly upon the duration and intensity of the light stimulus (e.g. Pittendrigh, 1960: Fig. 14; see Figs 9 and 10). The form of the curve is thus unlikely to provide absolute means of classification. Indeed, many otherwise typical Type I rhythms show large phase shifts in response to reversed light cycles (e.g. Dumortier, 1968; Youthed and Moran, 1969a). Similarly, the effects of LL are not absolute, even on Drosophila eclosion (p. 18). Thus, though many insect behavioural rhythms are remarkably stable in constant light, like all other circadian rhythms, they, too, damp out at high enough intensities. For example, the locomotor rhythm of the cockroach, Blaberus craniifer, while stable in LL 10 lux, damps out in LL 100-300 lux (Wobus, 1966b). Nothing is known about the site of photoreception for the entrainment of any dipterous behavioural rhythm. If it proves to be performed extra-ocularly at the site of the clock, then the rhythm will by definition be controlled by a Type I system. The effects of LL on dipterous rhythms, and the rapidity of their phase shifting would fit in with this. Furthermore, Konopka and Benzer’s evidence (1971; see p. 76) implies that the eclosion and adult activity rhythms of Drosophila may be coupled to the same clock. Since the phase-response curve and reaction to LL of the eclosion rhythm indicate Type I gating, it is difficult to avoid the inference that the behavioural rhythms may also be driven by a Type I clock. Presumably, the locomotor rhythm, like the male courtship rhythm (Hardeland and Stange, 1971), and the locomotor rhythms of other flies, will cease in LL; it will be interesting to see if it also has a Type I phase-response curve. It begins to look as if the Diptera differ from other insects in employing Type I clocks to regulate their behavioural rhythms. The case for universal Type I control in this order is not absolutely clear, however. For example, the oxygen consumption rhythm of Drosophila melanogaster is not annihilated by LL (of 1000 lux, Leclerc et al., 1971) so that this species

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

81

does not control its circadian functions exclusively by a Type I clock. Also, as noted above, the phase-response curve of Anopheles gambiae (Jones et al., 1972a) has a low amplitude (Fig. l o ) , strikingly different from the typical Type I curve for the pupation rhythm of another mosquito, Aedes taeniorhynchus, in which shifts of up to 14 h occur (Nayar, 1968). Furthermore, there is a single case of a dipterous behavioural rhythm surviving in LL, albeit noisily, namely that of Aedes aegypti (Taylor and Jones, 1969: Fig. 2). Probably, one should conclude that Truman’s two clock types, while providing a valuable framework upon which to hang analyses of circadian mechanisms for comparison, ought not be considered as exclusive (see p. 93).

4.6 MECHANISMS OF DRIVING OSCILLATORS In metazoa, the fundamental difficulties of exploring circadian mechanisms are made more difficult by the lack of knowledge of the precise anatomical whereabouts of all but a very few driving oscillators. Only two have been isolated and shown to oscillate independently; both come from the sea hare, Aplysia californica (Strumwasser, 1965: Jacklet, 1969; see p. 90). Two other putative clocks occur in insects. While they have not yet been isolated or shown to oscillate independently in the sense that the Aplysia clocks have, they d o appear to have been located with sufficient precision to permit experimental exploration. The first of these is in the optic lobes of cockroaches (p. 65), and the second in the protocerebrum of silkworms (p. 5’2).The first is concerned with the timing of a daily repeated behavioural rhythm, and the second with the gating of adult eclosion. Truman’s distinction between the two clock types they exemplify is thus apposite (p. 7 7 ) . 4.6.1 The cockroach optic lobe clock The clock controlling cockroach activity, which is presumed to reside in the optic lobes, is a Type I1 oscillator, apparently not entrainable by the direct action of light, largely immune to the disruptive effects of constant light and relatively stable in the face of phase-shifting stimuli; its coupling to the rhythms it controls is probably neural (p. 63) The mechanism of the oscillator itself, however, seems not to depend upon the interaction of action potentials. Truman has found (personal communication) that injection of tetrodotoxin (which specifically blocks Na+ penetration of axons and hence prevents the generation of action potentials) into cockroaches whose rhythms were free running in DD, stopped all s i F s of-locomotor activity for several days, but when the animals recovered, h e i r rhythm was resumed at exactly the correct phase, as if it had never stopped. That is to say the free-running drift of the rhythm before and after injection fell upon the

82

JOHN BRADY

same straight line. It therefore appears that although all motor activity ceased the clock continued to measure time normally. Assuming that tetrodotoxin penetrates the brain and optic lobes, this means that the clock does not function on the basis of reverberatory circuits or other oscillations involving the propagation of action potentials. Other electrical interactions, such as micropotentials or electrotonic conduction, on the other hand, are not ruled out. Unfortunately, nothing more is known of the mechanism of this clock other than can be inferred from the general characteristics of circadian clocks outlined below. 4.6.2 The silkworm protocerebral clock The silkworm protocerebral clock is a Type I oscillator which includes its own photoreceptor pigment and is therefore susceptible to inhibition by constant light and sensitive to phase-shifting light signals; it is hormonally coupled to the event it gates. Truman (1971a, 1971c, 1972b) has investigated its photokinetics and developed a model to explain bow the clock may work. The relevant observations were based on the phasing of emergence (which reflects the 1%h earlier release of eclosion hormone, see p. 52) of Antherea pernyi to a range of photoperiodic signals. He first found (1971a) that eclosion time does not bear a constant phase relationship to either “dusk” or “dawn” but is a function of them both (Fig. 13). For example, in LD 20 : 4 mean hormone release occurs about 18 h after the last “dusk” and 14 h after the “dawn”, whereas in LD 4 : 20 it occurs about 23 h after the last “dusk” and 4 h after the last “dawn”. In constant darkness, on the other hand, hormone release occurs with a free-running rhythm phased to 22 h from the onset of darkness, and in constant light the rhythm stops. Clearly, the appearance of light during this 22-h free-running cycle accelerates or decelerates it, according to the length of the photophase. The second important observation (1971a) was that the variance of eclosion time is also a function of the photoperiod length. The narrowest range of emergence time spans c. 3.5 h daily, and eclosion is completed within this gate under most photoperiods of less than about 20 h light. But with longer photophases the gate widens out until under LD 23 : 1 emergence spans at least 16 h (Fig. 13). This suggested that maximal accuracy of gating is achieved only if the clock is allowed to run for at least 4 h in darkness. In order to test this inference, Truman released developing pupae into constant darkness and then subjected them t o half-hour light interruptions at different times after the onset of DD (1972b). He found that interruptions given after more than 4 h of darkness, advanced the gate, but re-set the clock in a complex way. Interruptions given during the first c. 4 h after light-out, however, re-set the clock so that emergence was phased to

83

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS 1

l

o

Durotion of scotophose(h)

Fig. 13. Characteristics of the clock gating silkmoth (Antheraea pernyi) eclosion. between the length of the scotophase (abrcissa) and the delay between “dawn” and the release of eclosion hormone (left ordinate). Dotted diagonal, relationship if the progress of the 22-h free-running period started at “dusk” were not altered by the presence of light. 0, Relative accuracy of the clock after completion of various periods in darkness (abscissa), expressed as the percentage of moths emerging through the minimal 3.5-h gate (right ordinate). (Redrawn from Truman, 1971a.) 0 , Relationship

the time of interruption ( i x . 22 h after i t ) and not to the onset of DD. Furthermore, such early light interruptions could be given at hourly intervals almost indefinitely (up to 20 were tried) and eclosion was still phased to 22 h after the end of the last one. This must mean that the processes occurring in the clock are completely photoreversible for the first 4 h of darkness, and that once these processes ;ire complete the clock is relatively insensitive to light (though not unaffectcd by it). Truman’s interpretation is that the clock is not a circadian oscillator in the normal sense, but is an “hourglass” which is turned over every “dusk”, and whose “sand” takes 22 h to run out if light (does not interfere with it. Under constant darkness it re-cycles automatically, since once a cycle is complete, the presence of darkness permits the photoreversible process to start off again. If the cycle is completed in the light phase, however, it stops until the next “dusk” starts it off again; this is like the Drosophila clock stopping after more than 12 h of light (Pittendrigh, 1966; seep. 18). It appears, therefore, that the behaviour of the silkmoth eclosion clock is explicable as the outcome of three photochemica.1 processes (Fig. 14): (1) an initial photoreversible process taking 4 h to complete in darkness and almost instantaneously reversed by light; (2) a dark-decay process lasting

84

JOHN BRADY

1

0

I

I

4

1

1

8

I

I

I

12

I

16

I

I

20

1

i

1

Hours from onset of darkness

Fig. 14. Truman’s model for the silkmoth (Antheraea pernyi) photochemical hourglass gating eclosion. Ordinate, concentration of hypothetical substances involved. Heavy broken line. increase in concentration of the substance built up during the initial photoreversible process begun at “dusk”, reaching threshold level in c. 4 h and then releasing a dark-decay process shown as the heavy solid line. Light arriving during the first 4 h returns the system to zero, but arriving any later causes either an acceleration (early light) or a deceleration (late light) in the dark-decay process, as implied in the light broken lines. When either the dark-decay or the light-decay processes reach the zero threshold, release of eclosion hormone occurs at the experimentally observed times indicated by the arrows; 0 , experimental observations shown in Fig. 13. (Modified from Truman, 1972b.)

about 18 h, during which the substance built up in (1) decomposes spontaneously in darkness; ( 3 ) a light-decay process, taking about 14 h to break down all the substance built up by process (1). Truman has called the free-running, dark-completed cycle (i.e. (1) and ( 2 ) ) the scotonon, and the process initiated by light after the completion of (1) the photonon (i.e. 3). In passing, it may be noted that this model could equally well be drawn with Fig. 14 the other way up, that is to say with (1) being a decrease and (2) an increase in concentration. It might then be slightly easier to visualize the gate being switched open when a threshold concentration was exceeded. This model seems to account for the experimental facts very simply and economically. Clearly, however, it will not be the final story. For example, Truman notes (1972b) that the transition from process ( 1 ) to (2) is not sharp. Also, the phase-response curve to light interruptions (Truman, 1972b: Fig. 2) implies a rather complex effect during process (2); and the fact that light arriving after more than 16 h of dark delays hormone release, whereas light arriving after less than 12 h of dark advances it, similarly implies a complex progression of process ( 3 ) (see Fig. 14). There is also the

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

a5

major question of how it is that release of eclosion hormone gets coupled to the end of the scotonon only on the 19th day of development. The model is nevertheless valuable. It describes the functioning of the only insect circadian clock that has proved surgically manipulatable (pace Harker, see p. 56). The parallel between its (entrainment and that of Drosophila suggests that it may serve as a p a r a d i p for other gating clocks. Anatomically, it is less well defined. It will gate eclosion when transplanted, only if most of the protocere,brum is included. Interestingly, it does function without the optic lobes (cf. the cockroach), but neither the medial nor the lateral parts of pars intercerebralis containing the neurosecretory cells can be dispensed with without disrupting its mechanism (Truman, 3972d). 4.6.3 Rensing's model of the Drosophila circadian clock Rensing has recently elaborated (1971) a model of the clock system in Drosophila larvae. It sets out to explain how the timing of the overall metabolism of the animal may occur and how the timing of pupation may be coupled to this by a temporal sequence of neuroendocrine events, not unlike the sequence suggested by Cymborowski (1970b) for the control of cricket locomotor activity (p. 38). Based on his own observations on Drosophila brain neurosecretory cell changes during the last two days of larval development (Rensing, 1966c, 1971; see p. 35), and on those of Cymborowski and Dutkowski (1969, 1970a; see p. 38) on cricket brains, the sequence runs as follows. Maximum nuclear size which occurs on the last larval night about 3 h after "dusk", indicates inaximum RNA synthesis. This results in synthesis of neurosecretory materi.114-6 h later, followed by its delivery to the corpora cardiaca, and thence the blood, over the next 6-9 h. This release of brain hormone stimulatcs the almost immediate release of ecdysone by the prothoracic glands (as evinced by the induction of puff 74EF/75B in the salivary gland chromosomes), followed 6-8 h later by pupation. Clearly, a sequence of some such sort must occur on the last larval day, since it is well known that brain hormone induces ecdysone release, and that this induces ecdysis. These latter two processes are not rhythmic, however: they take place once only, when larval development is complete. The nuclear volume changes in the brain cells, on the other hand, are bimodally rhythmic in the sense that peak volumes occur in the late night and afternoon of the penultimate day (Rensing, 1966c: Fig. 5a), and in the early night and forenoon of the last larval or prepupal day (Rensing, 1971: Fig. la). During the 48 h before pupation there are thus four distinct neurosecretory cell RNA synthesis peaks from which to" start the proposed sequence. The fact that there are the same four peaks in the prothoracic gland cells does not make the system any easier to understand. Moreover, it

86

JOHN BRADY

is by n o means certain that the observed twice-daily transmission of neurosecretory material from the perikarya to the axons of the brain cells necessarily siLpifiesrelease of this material from the corpora cardiaca to the blood. It should be noted also that the pupation “gate” is very wide indeed. Quite unlike the subsequent adult eclosion which is restricted to discrete 6-h peaks, pupation occurs throughout the 24 h with only a ? 2 per cent deviation across the day (Rensing and Hardeland, 1967: Fig. 3 ; cf. Bakker and Nelissen, 1963). If the brain cells d o time pupation, it is not easy to see why, with their clear daily changes (Rensing, 1966c; see p. 36), they should achieve such a poor result. The developmental link between them and the timing of pupation must surely involve more than merely the direct sequence proposed. Perhaps one should look again at the histological data (Rensing, 1966c: Figs 5 , 7 and 8; 1969c: Fig. 3), and ask why it should be that the cells of the brain neurosecretory tissue, corpora allata, prothoracic glands, salivary glands, and fat body of the larva should all have precisely synchronous rhythms, and why the same synchrony (and phase) should occur in these and other tissues in the adult. Clearly this implies central coordination, and presumably humoral coordination at that, b u t where should one look for its provenance? Why single out the brain cells when their nuclei d o not even show a phase lead that might imply some sort of a driving role? Of course, the demonstrated role of the brain in gating silkmoth eclosion and pupation (Truman, 1972a; see p. 52) does suggest the possibility o f a similar system timing Drosophila pupation, b u t any relationship between this and the repeated peaks of brain cell nuclear size awaits experimental demonstration; understandably, no transplant experiments have been performed on Drosophila to elucidate this problem. However, Rensing’s model is not so much concerned with the timing of pupation as with how the cellular activities of the different tissues are synchronized. He supports the idea of hormonal synchronization with his observation that the nuclear volume cycle of Drosophilu salivary gland cells can be differentially phase shifted in vitro by 1-h pulses of 80 pg ml-’ of ecdysone. A pulse given at the phase of minimum nuclear volume causes a 3-6 h advance shift 18 h later, whereas a pulse at maximum nuclear size is without effect (Rensing, 1971). This is very interesting, and is the first demonstration in insects of the commonly assumed effects of hormones on cellular rhythms. One should note, however, that the dose of ecdysone was apparently large (though see Kambysellis and Williams, 197 1): a concentration approximately 1000 times greater than the peak pre-pupal titre in Calliphora larvae (Shaaya and Karlson, 1965; Kaplanis et al., 1966). Also, the shifted salivary gland cell

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

a7

rhythms were observed for only 30 h after the ecdysone pulse, so that the circadian stability of the new peak times is nDt positively known. Since both the ecdysone pulse and a control pulse of‘ the culture medium set up disturbances in the rhythm which left the phase relationship between the maximum and minimum nuclear sizes quite (different from that of the untreated controls (Rensing, 1969c: Fig. lo), this may b e important. The variances of this rhythm are evidently large with peak times spanning a range of 6 h, overlapping the 6-h range of minima times by as much as 3 h (Rensing, 1971) and this in a bimodal rhythm. It would therefore appear difficult to be certain of distinguishing between non-circadian disturbance and specific phase effects when manipulating this system. The implied influence of a hormone on a cellular rhythm is, nevertheless, suggestive, and Rensing uses it as a basis for a speculative model of how the overall circadian system in Drosophila may work. He proposes a cyclical sequence of gene activation ( A + B + C + A + etc.) which operates in all cells. Each step involves protein synthesis, precursor pool sizes, and feedback, resulting in an independent net self-sustained oscillation in each cell. Thus far the model is essentially a simplified version of the ‘‘chronon” concept of Ehret and Trucco (1967); Rensing’s sophistication is the suggestion that these multiple cellular oscillators are synchronized by a twice-daily pulse of ecdysone produced as a result of the brain-prothoracic-gland sequence (outlined above. It must be admitted that there are difficulties for this model. First, there is no direct evidence for a daily or twice-daily pulsed release of ecdysone; evidence exists only for its presence at a low titre in third-stage fly larvae up to their last day, and then for a sudden ten-fold increase in its titre (Shaaya and Karlson, 1965; Barritt and Birt, (970). Secondly, the same system cannot be involved in synchronizing celluilar rhythms in adults, since they possess no prothoracic glands. Thirdly, it is difficult to see how the brain-neurosecretory-cell-prothoracic-glandaxis could produce twice-daily ecdysone pulses without showing some sort of phase differences from each other, and from the tissues they are supposed to be synchronizing. Perhaps one should not be looking for an endocrine coordination of the cellular rhythms so much as for less specific humoi-a1 changes such as ionic fluctuations (Brady, 1968; though see p. 34), or other weak cellular interactions (Winfree, 1967; see p. 89). 4.6.4 Oscillator mec hanisms-general considerations Advances in the field of cell biology in recent ylears have not been matched by advances in the understanding of the mechanisms of circadian oscillators. A cell biologist looking at this subject for the first time will find himself lost: there are numerous models and hypotheses, but sadly few AIP-4

88

JOHN BRADY

facts. It is impossible to talk in terms of organelles, membranes, gradients, pumps, or even biochemistry in any way except theoretically (see review by Hastings, 1970). There is a hint that messenger RNA is involved, and hence the nucleus, since actinomycin D stops all the cellular oscillators upon which it has been tested (see Sweeney, 1972; and Schweiger. 1972). But there is also a hint that membrane electro-permeabilities are involved rather than nuclear chemistry, since the only manipulations which alter the frequency of circadian oscillators-as opposed to merely stopping their overt expression-are treatment with heavy water (Enright, 1971a) or alcohol (Enright, 1971b) (see Bunning, 1972). In addition, there is a suggestion that oxidative metabolism may be at the heart of the matter (Bryant, 1972; see also p. 25). These are almost the only hard facts that there are t o go on and a little thought shows their “hardness” to be largely superficial. One of the main stumbling blocks to the construction of biochemical models of circadian oscillators has always been the fact that all the known biochemical oscillations are reactions whose cycle time lasts seconds or minutes rather than hours; there is a time gap of about two orders of magnitude. Recent theoretical studies-whose detailed consideration is outside the scope of the present review-have pointed the way to possible solutions of this problem, however. It is often proposed that high-frequency biochemical oscillators could interact to produce a net low-frequency (circadian) oscillation by the production of “beats”. Pavlidis (1969: 420), however, has pointed out the several limitations to this theory, and has proposed, instead, a model involving strong inhibitory coupling between the high-frequency oscillators. His mathematical exploration of this model revealed its potential for reducing biochemical oscillations to circadian frequencies. Using a computer simulation with populations of up to 30 oscillators and a wide range of inhibitory coupling factors, he found that strong negative coupling resulted in the greatest net frequency reduction and that with this the population behaved as if it were a single oscillator, showing a typical circadian-type phase-response curve to disturbing stimuli and a theoretical means of temperature compensation. In a later study, Pavlidis (197 1) used the model further to show that it can also explain better than any of the alternatives how “rhythm splitting” (Hoffmann, 1971) may occur, and how “Aschoff‘s rule” (p. 5 1) can arise in a population of oscillators. At least theoretically, therefore, we can consider high-frequency biochemical oscillators as the origin of circadian oscillations within individual cells. But this tells us nothing of how populations of oscillating cells are synchronized within each organism. Another theoretical study (Winfree, 1967) suggests that the answer may well lie in a weak chemical coupling

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

89

between the cells. Winfree’s mathematical analysis and computer simulation showed that large populations of weakly coupled oscillators (e.g. yeast cells in culture) tend to synchronize each other to a common phase and periodicity, even though starting at random phase with respect to each other. The only important proviso is that the natural frequency of these oscillators should be reasonably similar (’“infree’s example had 100 oscillators with a range of *2!h per cent in period). This model, which concerns the production of a common rhythm by a large population of similarly oscillating cells (or organisms), should not be confused with Pavlidis’s, which concerns the production o f a single net low-frequency rhythm by a small population of widely differing biochemical oscillators presumptively within a single cell (though see p. 90). Winfree’s model shows how populations of cells within an organism may interact to correct the cells’ individual differences and produce a precise, synchronous circadian rhythm overall. This is particularly helpful to understanding how some of the observed ti:;sue rhythms may work; for example, the evidence that patches of insect epidermal cells operate in unison to lay down a circadian structural pt:riodicity in the endocuticle, but do so semi-independently of neighbouring patches (p. 21). On the assumption (widely held) that each cell is primitively capable of operating with a circadian rhythm, this sort of phenomenon might well be expected from Winfree’s model. The epidermal cells, each oscillating at roughly a 24-h periodicity, interact biochemically or biophysically with their neighbours until the local population is in phase. The same sort of local intercellular interactions could serve to explain why different kinds of neurons in a single ganglion have coincident nuclear volume rhythms (Brady, 1967a; see p. 35). And similar intercellular couplings may account for the much wider coincidence between different tissues reported by Rensing (p. 35). This synchrony of different tissues in Drosophila is so wide, however, that some sort of blood-borne coupling must surely also be involved. The important question is whether these kinds of cellular interactions can provide a general explanation of how circadian rhythmicity is controlled in animals. The answer comes in two parts, negative and positive. In those animals where it has been tested (cockroaches, p. 61; silkmoths, p. 52; crustacea, p. 67; and birds, Gaston arid Menaker, 1968; Binkley et al., 1971) overt circadian rhythmicity ceases upon removal of specific pieces of tissue. Thus the innate periodicity of the remaining cells is not enough to overcome this loss. On the other hand, the tissues removed in these cases are quite large, so the putative diiving oscillators they contain may well consist of many cells and may well interact in the way suggested. The only single-celled exception is the “parabolic burster” neuron of

90

JOHN BRADY

Aplysiu (Strumwasser, 1965), but the driving function of this oscillation is quite unknown. A possible key to the whole problem lies in the other clock found in Aplysiu. Jacklet discovered (1969) that it was possible to culture the eye of Aplysiu in isolation and that in constant darkness the isolated eye continued to send a circadian rhythm of trains of compound action potentials down its qptic nerve. The isolated eye is responsive to light cycles and its spike rhythm is entrained by them (Eskin, 1971);,the mechanisms of photoreception, phase adjustment, and oscillation therefore all reside in the same organ. More importantly, for our present consideration, the accuracy of this clock apparently depends upon the number of cells it contains. Jacklet and Geronimo (1971) found that they could slice away up to 80 per cent of the retinal cells of the eyecup with little effect on the periodicity of the output by the optic nerve, but that removal of any more resulted in a collapse of its circadian modulation. They suggest three possible explanations: (a) a population of neurons driven by a master oscillator; (b) a population of circadian oscillator neurons acting together; (c) a population of infradian oscillator neurons acting together. Against the first hypothesis Jacklet and Geronimo point out that the period of the output should stay constant until the master is cut away, whereas in fact the period is reduced slightly as the eye is reduced (until the breakdown at 20 per cent). The possibility that a small area of specialized clock cells exists around the optic nerve is perhaps not completely ruled out, but seems unlikely since the residual oscillating 20 per cent can be derived by cutting the eye either sagittally or transversely. Against their second hypothesis they say that there should still be a circadian component of some sort in the output of eyes reduced to less than 20 per cent, but none in fact occurs. It may also be noted that one might expect an increase in the variance of the output as the cell population was successively reduced, but this did not occur either (Jacklet and Geronimo, 1971: Fig. 2). Jacklet and Geronimo therefore conclude that their third hypothesis is the most plausible and adduce Pavlidis’s model as a mechanism for how it may work. On the basis of the available experimental facts this does, indeed, seem the most reasonable interpretation. Frequency reduction in Pavlidis’s model is a function of the number of oscillators in the system ( n ) and also the degree of inhibitory coupling ( r ) via the expression (n - l ) r (Pavlidis, 1969: 427). Assuming the mechanism and therefore the strength of any such coupling between cells does not change in the Aplysia eye as its retinal population is reduced, one may infer that it is the number of cells (oscillators) that is the most important parameter in the production of a net low-frequency output. The surprise is that, if correct, this means that the eye’s circadian output is not at all dependent upon its constituent cells’

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

91

individual circadian oscillations and that it i:i therefore not the result of a specialized form of the cellular interactions implicit in Winfree’s model, which we have just suggested may occur in other tissues (p. 89). No insect (or any other) clock has yet been analysed in such elegant neurophysiological detail, so that it is difficult to forecast how typical the Aplysia eye may be of other clocks. One is, however, reminded of the existence of a circadian rhythm in the ERG of beetle eyes (Dytiscus fasciventris) found years before the role of the optic lobes in regulating behavioural rhythms in insects had even been thought of (Jahn and Crescitelli, 1940; Jahn and Wulff, 1943), and of the more recent detailed demonstration of a similar optic lobe rhythm in crayfish (ArCchiga and Wiersma, 1969; Ankhiga, 1973).

5 Conclusions Consideration of the foregoing pages, particularly section 3, suggests that insect circadian rhythms fall into four broad categories: 1. Cellular metabolism, covering rhythmic functions of cell physiology such as nuclear volume (p. 34) and protein synthesis cycles (p. 38). 2. General physiology, covering (a) non-spccific physiological rhythms of the whole organism indicated by events such as oxygen consumption (p. 23), narcosis susceptibility (p. 24), or metabolite changes (p. 30); and (b) more specific physiological rhythms such as those implied by some insecticide susceptibility rhythms (p. 27), AChE activity, or hormone synthesis cycles (p. 35). 3. Developmental gating, covering the timing of once-in-a-lifetime events such as hatching or adult eclosion (p. 15). 4. Behaviour, covering all those motor functions which are rhythmically modulated, such as general activity 1 evel, feeding, or oviposition (p. 6), and including the central modulation of response thresholds

(P- 12). Clearly these categories are not separable into physiologically watertight compartments. For example, cellular rhythms presumably enter into the secretory cycles of endocrine glands; the consumption of oxygen or the production of some metabolites must be correlated with activities such as locomotion; and the gating of emergence must involve not only the hormonal control of morphogenesis, but ,also the switching on of the relevant eclosion behaviour. The circadian functioning of the whole animal is the integrated outcome of all these rhythms. There are in principle four ways this complex circadian organization could be achieved: (a) by exogenous, environmental control (Brown, 1970); (b) by a single master clock driving all the rest of the animal’s

92

JOHN BRADY

rhythms; (c) by a few master clocks synchronized by environmental signals; (d) by every cell being its own clock, interacting with its neighbours and synchronized with the environmental cycle by sensory-humoral means. Without going into the long-standing controversy over the possibility of exogenous control of rhythms (as opposed to their entrainment), we may note that this view is very difficult to match with the facts and is not generally accepted (e.g. Marler and Hamilton, 1966: 29; Hastings, 1970; and see p. 4). Equally, the second possibility of one master driving oscillator is almost certainly too simplistic a notion. It conflicts, for example, with the idea of autonomous rhythmicity in cultured tissues (p. 39), the apparent dissociation of rhythms within individual humans (Aschoff, 1965a; Conroy and Mills, 1970: 121), or the existence of different kinds of clocks in insects (p. 76). This leaves us with the last two possibilities. How do they match up with the circadian organization of the four categories of insect rhythm just described? Strictly speaking, the existence of autonomous circadian periodicity in individual cells has never been observed in insects. Quite apart from a priori evolutionary considerations (e.g. Harker, 1958a: 33), however, there is a strong implication from organ culture experiments (p. 39) that such primitive periodicity does survive in at least some cells. The mechanism of protistan clocks, from which these metazoan cellular clocks may be presumed to derive, has yet to be elucidated (see Sweeney, 1972) but seems to involve protein synthesis, and, one may suppose, Pavlidis-type biochemical interactions (p. 88). It would be wise to assume that all metazoan cells contain the necessary equipment to perform such oscillations, even if they do not normally use it independently. Rhythms in general physiology are mainly not of the kind that suggest an origin in specific driving oscillators. It seems more likely that changes in oxygen consumption, narcosis susceptibility, or blood sugar levels, for instance, are the incidental outcome of several cyclical activities, rather than specifically controlled by their own clocks, though hormonal control cannot be ruled out. On the other hand, other physiological rhythms seem clearly to be the overt expression of quite specific clocks. The secretory cycles of some endocrine tissues, for example, must be synchronized by some sort of a clock. Similarly, changes in cholinesterase inhibitor susceptibility and AChE level are presumably the expression of rhythmic nervous activity or perhaps permeability. It seems likely that when other physiological functions are examined for rhythmicity, many will be shown to fluctuate as a result of local driving oscillations. For example, the suggestion of an excretory rhythm (p. 31) makes a self-sustaining oscillation in the malpighian tubules a possibility. Developmental gating requires a very specific form of clock. Essentially all that is needed is an hourglass to link the opening of the gate to a specific

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

93

time after dawn or dusk. The fact that these clocks may oscillate in constant darkness presumably reflects the potential hazard that constant darkness presents for any small animal, especially when non-motile; constant light, of course, can never occur in nature. A simple photochemical hourglass, like that of silkmoth pupae (p. 83), would appear adequate for any such gating system. It is interesting that that rather analagous time function, photoperiod measurement, is effected in at least one insect (the aphid, Megoura viciae) by a similar non-oscillatory photochemical hourglass (Lees, 1972). Moving to the circadian control of behaviour one is faced with a totally different problem in chronometry. Rhythmic behaviour cannot be controlled by an hourglass or similar system since it is an on-going process, largely unaffected by constant light (p. 79). In insects it occurs in two common forms: (1) single peaks of activity with sharp onsets or offsets (e.g. cockroach locomotion, p. 3); (2) smoot'hly modulated peaks, often double, of changing response thresholds (e.g. tsetse fly, p. 14). It is possible, but by n o means necessary, that these two are qualitatively different. Closer analysis of the onset of cockroach locomotion, for example, reveals its suddenness to be more apparent than real (Harker, 1960b: Fig. 1; Sullivan et al., 1962). The sharp onset of this type of rhythm has often been supposed to be due to the throwing of a switch (rather like that occurring in gated phenomena), but it could equally well be explained by either a cycle of intensity of endogenous nervous signals surpassing a fixed response threshold, or by a cycle of threshold levels falling below a fixed intensity of nervous motor signals. Since the latter occurs with respect to experimentally applied exogenous stimuli (p. 13), it would be economical to suggest that the same system Lies at the root of all daily behavioural rhythms. As discussed on p. 5 5 , the rhythmic control of behaviour must be adaptable if it is to be adaptive, the simple throwing of a switch will not do. One is led to conclude that a semi-hierarchical range of clocks exist in each insect and that the different types of rhythm are each driven by a different type of clock. The following four kinds seem to occur. 1. Cellular clocks-in many, or all individual cells, potentially autonomous and independent but normally interacting t.3 keep the components of a tissue in phase. 2. Physiological clocks-lower-order driving oscillators, some presumably endocrine, many probably needing to be driven from above and therefore not themselves autonomous. 3. Gating clocks-switching developmental events on at specific phase angles to the zeitgeber, and in some, possibly many cases consisting of simple photochemical hourglasses.

JOHN BRADY

94

4. Behavioural clocks-probably

a single driving complex in the central nervous system, controlling behaviour via neural modulation of arousal, and thereby secondarily driving many lower-order rhythms.

The main known rhythms that these four clock types control are summarized in Table 1. These may be visualized as existing in the insect in the network shown, highly simplified, in Fig. 15. The linked spheres (ignoring bilateral symmetry) represent real anatomical connections within the nervous system, and are meant to imply neural coupling via the primary driving oscillator (“CNS Clock(s)”) to the environmental light cycle. The gating LIGHT

CELLULAR RHYTHMS

PHYSIOLOGICAL RHYTHMS

LIGHT

BEHAVIOURAL RHYTHMS

DEVELOPMENTAL RHYTHMS

Fig. 15. Schematic network of the inferred circadian clock system of insects. The linked spheres represent interconnected nervous and endocrine centres. The primary CNS clock controlling behavioural, and probably also other rhythms, is presumed to reside in the optic lobes and t o be entrained via the photoreception of the compound eyes. This clock is coupled to and drives the second- and third-order oscillations indicated. T h e developmental hourglass possesses its own independent photoreceptor system. For further explanation see text.

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

95

TABLE 1 Clock

Overt rhythm controlled

Example references

Nuclear volume RNA-protein synthesis

Rensing ( 1 9 6 6 ~ ) Cymborowski and Dutkowski (1969, 1970a) Rensing (1!169a) X-ray sensitivity Rensing (1!)69c) Tissue culture Fowler and Goodnight (1966a. 19856b) Cuticle deposition Neville (19 70) 2. Physiological Cholinesterase inhibition Bull and Lindquist (1965, 1968) Fondacaro and Butz (1970) Blood sugar Hilliard ant1 Butz (1969) Brady (1968) Blood ions Hormone production Fowler et al. (1966a, 1972) Rensing (1!)66c) Rao and Gropalakrishna reddy (1967) Minis and Pittendrigh 3. Gating Hatching (1968) Pupation Nayar (1967a, 1967b, 1968) Truman (1972a) Pittendrigh (1966) Adult eclosion Truman (1972a) Neumann (1971b) Roberts (1!160, 1962) Locomotor activity 4. Behavioural Brady (1972a) Jones et al. (1972b) Feeding Gillett et al. (1962) Renner (1960) Tychsen and Fletcher Mating (1971) Hardeland ,md Stange (1971) Response threshalds Shorey and Gaston (1965) Brady ( 1 97 2b) Birukow (1960) 1. Cellular

0

Page discussed 35 38 29 39 33

20 27.31 27 30 34 33 35 40 15 16 53 17 52 19 6

7 7 8 9 10 10 13 13 13

96

JOHN BRADY

hourglass has its own photoreceptor system, but also a connection to the primary CNS clock in order to accommodate the implication that a single clock controls both behaviour and eclosion in Drosophila. The endocrine organs and sensory-motor centres, though rhythmic in output, are presumed not to be autonomously rhythmic but driven from above. The cell clocks do not have to be hormonally synchronized as shown, but it does seem likely that they are chemically coupled with the rest of the animal in some way. Some physiological rhythms are probably driven directly by endocrine cycles and some by direct nervous control (not shown), but others probably via intermediary third-order oscillators. It is not clear whether behavioural rhythms are the outcome of sensory threshold modulation or excitatory modulation-the Figure implies both. The minimum number of autonomous driving oscillators that this system implies is : one central nervous clock controlling behaviour (typically Truman’s Type 11, p. 77); one photo-biochemical hourglass gating development (typically Truman’s Type I); and potentially one primitive cellular oscillator in each cell. It seems most unlikely that this will prove to be the final score. In particular, one suspects the existence of autonomous endocrine oscillators, and possibly the existence of different clocks controlling physiological rhythms from that controlling behaviour. No one has yet investigated whether any physiological rhythms survive in behaviourally arrhythmic insects, but the persistence of an oxygen consumption rhythm in adult Drosophila in LL (pp. 24, 81) suggests that they may. More certain is the existence of different kinds of gating clock within one insect (p. 7 8 ) , and possibly two kinds of behavioural clocks in bees (P- 2 5 ) There is a great array of experiments waiting to be done to elucidate these and other circadian problems. In this review attention has been drawn to several, but some seem sufficiently fundamental to merit a final closing men tion. First, there is a need to demonstrate the existence of autonomous endocrine clocks, and to examine the action of their output on the relevant effector systems. In spite of the amount of research based on the assumption that some endocrine organs are primary driving oscillators, none has yet been proved to oscillate autonomously @ace Harker, see p. 56). The only exception is the clock gating silkmoth eclosion, and that consists of most of the pupal brain and incorporates its own photoentrainment system. Secondly, much needs t o be done, in a more refined way than hitherto, in the line of cut-and-look experiments to locate the optic lobe and protocerebral components of the behavioural clock. Virtually no neurophysiology has been done on insect behavioural rhythms. and one hopes to see nervous rhythms recorded from de-afferented parts of the CNS,

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYT’HMS

97

ultimately developed to an Aplysiu-type neural clock culture (p. 90). Thirdly, the existence, or otherwise, of multiple primary clocks is waiting to be revealed. This could be done by the demonstration of the persistence of physiological rhythms (e.g. oxygen consumption) in insects made behaviourally arrhythmic (e.g. by optic lobe removal). Alternatively, it could be done by demonstrating the dissociation of two or more rhythms in free-running individuals. As this has apparently been possible in humans, it may also be possible in insects. The existence of different kinds of behavioural clock is already implicit from comparison of the Diptera with other orders (p. 80): the reality of this difference awaits the demonstration of the form and site of dipterous photo-entrainment. Fourthly, the further construction of Drosophilu rhythm mutants and their physiological exploration seems likely to be a particularly fruitful field. Not only is there the apparent potential for anatomical exploration by mozaic construction, but also the exploration of what is missing in arrhythmic mutants. In addition, measurement of the different phaseresponse curves should reveal whether the period differences between the mutants reflects the fact that their different rhythms are under control by a single clock, or merely the genetic control of one parameter of several different sorts of clock. Finally, there are the intriguing implications that oxidative metabolism is intimately involved in insect clocks-both of 1 he gating and behavioural types (see p. 25). This invites further exploration. Possibly, the respiratory rhythms of diapausing pupae (p. 24) will prove convenient for study because of their very low rate of performance of all other functions. Evidently, there is much to be done. But if metazoan circadian mechanisms are t o be understood at all, insects provide material that is in many ways uniquely suitable for the necessary research.

References Acree, F., Jr., Turner, R. B., Smittle, B. J . and Burden, C. S. (1965). Hydrocarbons in haemolymph of cockroaches of different ages. /. Insect Physiol. 11,905-910. Amouriq, L. (1969). Effets de la dextroamphCtamine et de l’tphidrine sur le rythme nycthimtral et I’activiti motrice de Porcellio scaber Law. (Porcellionidne, Isopodes terrestres). C.r. hebd. Stanc. Acad. Sci.. Paris, 268, 136-139. Arichiga, H. (1973). Circadian rhythm of sensory input in the crayfish. In “Neurosciences Research Program” (Ed. F. 0. Schmitt). In press. M.I.T. Press. ArCchiga, H. and Wiersma, C. A. C. (1969). Circadiax rhythm of responsiveness in crayfish visual units. J . Neurobiol. 1, 71-85. Aschoff, J. (1960). Exogenous and endogenous comporients in”circadianrhythms. Cold Spring Harb. S y m p . quant. Biol. 25, 11-28. Aschoff, J. (1965a). Circadian rhythms in man. Scirnce, N. Y . 148, 1427-1432.

98

JOHN BRADY

Aschoff, J. (1965b). Response curves in circadian periodicity. In “Circadian Clocks” (Ed. J. Aschoff), pp. 95-1 11. North-Holland, Amsterdam. Aschoff, J., Klotter, K. and Wever, R. (1965). Circadian vocabulary. In “Circadian Clocks” (Ed. J. Aschoff), pp. x-xix. North-Holland, Amsterdam. Azaryan, A. G . and Tyshchenko, V. P. (1970). Neural regulation of the circadian rhythm of motor activity in the house cricket, Gryllus domesticus L. (Orthoptera, Gryllidae). Entomol. Obozr. 49, 72-82. (As translated from Russian in Ent. Rev. 49, 3845.) Bakker, K. and Nelissen, F. X. (1963). On the relations between the duration of the larval and pupal period, weight and diurnal rhythm in emergence in Drosophila mekanogaster. Entomologiu exp. appl. 6, 37-52. Ball, H. J. (1965). Photosensitivity in the terminal abdominal ganglion of Periplaneta americana (L.). J . Insect Physiol. 11, 1311-1315. Ball, H. J. (1972). Photic entrainment of circadian activity rhythms by direct brain illumination in the cockroach Blaberus craniifer. J. Insect Physiol. 18. 2449-2455. Ball, H. J. and Chaudhury, M. F. B. (1973).Photic entrainment of circadian rhythms by illumination of implanted brain tissues in the cockroach Blaberus craniifer. J. Insect Physiol. 19, 823-830. Barritt, L. C. and Birt, L. M. (1970). Prothoracic gland hormone in the sheep blowfly, Luciliu cuprina. J. Insect Physiol. 16, 671-677. Bartell, R. J. and Shorey, H. H. (1969). A quantitative bioassay for the sex pheromone of Epiphyas postuittana (Lepidoptera) and factors limiting male responsiveness. J. Insect Physiol. 15, 33-40. Barton Browne, L. and Evans, D. R. (1960). Locomotor activity of the blowfly as a function of feeding and starvation. J. Insect Physiol. 4, 27-37. Bates, M. (1941). Laboratory observations on the sexual behavior of anopheline mosquitoes. 1.exp. Zool. 86, 153-173. Beament, J. W. L. (1958). A paralysing agent in the blood of cockroaches. J. Insect Physiol. 2 , 199-214. Beattie, T. M . (1971). Histology, histochemistry, and ultrastructure of neurosecretory cells in the optic lobe of the cockroach, Periplaneta americana. J. Insect Physiol. 1 7 , 1843-1855. Beck, S . D. (1963). Physiology and ecology of photoperiodism. Bull. ent. SOC. A m . 9, 8-16. Beck, S . D. (1964). Time-measurement in insect photoperiodism. A m . Nat.98, 329-346. Beck, S. D. (1968). ‘‘Insect Photoperiodism”. Academic Press, New York and London. Beier, W. (1968). Beeinflussung der inneren Uhr der Bienen durch Phasenverschiebung des Licht-Dunkel-Zeitgebers. Z. Bienenforsch. 9, 356-378. Beier, W. and Lindauer, M. (1970). Der Sonnenstand als Zeitgeber fiir die Biene. Apidologie, 1, 5-28. Belcher, K. S. and Brett, W. J. (1973). Relationship between a metabolic rhythm and emergence rhythm in Drosophilu melanogaster. J. Insect Physiol. 19, 277-286. Bennett, M. F. (1970). Bilateral symmetry in the mediation of circadian differences in earthworms. Z. uergl. Physiol. 69, 1-5. Binkley, S., Kluth, E. and Menaker, M. (1971). Pineal function in sparrows: circadian rhythms and body temperature. Science, N. Y. 174,3 11-314.

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

99

Birukow, G. (1960). Innate types of chronometry in insect orientation. Cold Spring Harb. Symp. quant. Biol. 25.403412. Birukow, G . (1964). Aktivitats- und Orientierungsrhythmik beim Kornkafer (Calandra panaria L.). Z . Tierpsychol. 21, 279-301. Birukow, G. (1966). Orientation behaviour in insects and factors which influence it. In “Insect Behaviour” (Ed. P. T. Haskell), Symp. R. ent. SOC., Lond. 3, 1-12. Birukow, G. and Busch, E. (1957). Lichtkompassorieritierung beim Wasserlaufer V e h cumens F. (Heteroptera) am Tage und zur Nachtzeit. 11.: Orientierungsrhythmik in verschiedenen Lichtbedingungen. Z. Tierpsychol. 14. 184-203. Bliss, D. E. (1962). Neuroendocrine control of locomotor activity in the land crab Gecarcinus lateralis. In “Neurosecretion” (Eds H. H-IlPr and R. B. Clark), Mem. S O C . Endocrinol. No. 12, 391410. Academic Press, New ‘fork and London. Brady. J. (1967a). Histological observations on circadian changes in the neurosecretory cells of cockroach sub-oesophageal ganglia. /. Insect Physiol. 13, 201-213. Brady, J. (1967b). Control of the circadian rhythm of activity in the cockroach. I. The role of the corpora cardiaca, brain and stress. /. exp. Biol. 47, 153-163. Brady, J. ( 1 9 6 7 ~ )Control . of the circadian rhythm of activity in the cockroach. 11. The role of the sub-oesophageal ganglion and ventral newe cord. /. exp. Biol 47,165-1 78. Brady, J. (1967d). The relationship between blood ions and blood-cell density in insects. J. exp. Biol. 47, 313-326. Brady, J. (1968). Control of the circadian rhythm of activity in the cockroach. 111. A possible role of the blood-electrolytes. /. exp. Biol. 49, 39-47. Brady, J. (1969). How are insect circadian rhythms controlled? Nature, Lond. 223, 781-784. Brady, J. (1970). Characteristics of spontaneous actitity in tsetse flies. Nature, Lond. 228, 286-287. Brady, J. (1971). The search for an insect clock. In “Biochronometry” (Ed. M. Menaker), pp. 51 7-526. National Academy of Sciences, Washington. Brady, J. (1972a). Spontaneous, circadian components of tsetse fly activity. J. Insect Physiol. 18,471484. Brady,J. (1972b). The visual responsiveness of the tsetse fly Glossina morsitans Westw. (Glossinidae) t o moving objects: the effects of hunger, sex, host odour and stimulus characteristics. Bull. ent. Res. 62, 257-279. Brady, J. (1973). Changes in the probing responsivene!;s of starving tsetse flies (Glossina morsitans). Bull. ent. Res. 63. In press. Bregazzi, P. K. (1972). The effects of low temperature upon the locomotor activity rhythm of Talitrus saltator (Montagu) (Crustacea: Amphipoda). /. exp. Biol. 57, 393-399. Brett, W. J. (1955). Persistent diurnal rhythmicity in Drosophila emergence. Ann. ent. SOC. A m . 48, 119-131. BrousseGaury, P. (1967). A propos de la sensiblliti photique extraoculaire des Dictyopttres, mise en Cvidence de photoricepteurs associis aux nerfs tigumentaires. C.7. hebd. Sianc. Acad. Sci., Paris, 264, 2481-2482. BrousseGaury, P. (1969). Des stimulus photiques transmis par des nerfs tigumentaires contrblent des cellules neurostcrttrices chez les Dictyopttrk Blaberidae et Blattidae. C.Y. hebd. Sianc. Acad. Sci., Paris, 268, 383-386.

100

JOHN BRADY

Brown, F. A. (1970). Hypothesis of environmental timing of the clock. In “The Biological Clock” (F. A. Brown, J. W. Hastings and J. D. Palmer), pp. 13-59. Academic Press, New York and London. Brown, R. H. J. and Unwin, D. M. (1961). An activity recording system using infra-red detection. /. Insect Physiol. 7, 203-209. Bryant, T. R. (1972). Gas exchange in dry seeds: circadian rhythmicity in the absence of DNA replication, transcription, and translation. Science, N. Y . 178, 634-636. Buffington, J. D. (1968). A circadian pattern in the respiration of larvae of the mosquito Culex pipiens. Mosquito News, 28, 95-98. Bull, D. L. and Lindquist, D. A. (1965). A comparative study of insecticide metabolism in photoperiod-entrained and unentrained bollworm larvae Heliothis zea (Boddie). Comp. Biochem. Physiol. 16, 321-325. Bull, D. L. and Lindquist, D. A. (1968). Cholinesterase in boll weevils, .4nthonomus grandis Boheman-I. Distribution and some properties of the crude enzyme. Comp. Biochem. Physiol. 25, 639-649. Bullock, T. H. and Horridge, C. A. (1965). “Structure and Function in the Nervous Systems of Invertebrates”. W. H. Freeman, San Francisco. Biinning, E. (1958). h e r den Temperatureinfluss auf die endogene Tagesrhythmik besonders bei Periplaneta americana. Biol. Zbl. 77, 141-152. Biinning, E. (1959). Zur Analyse des Zeitsinnes bei Periplaneta americana. Z. Naturf. 14b, 1-4. Biinning, E. (1967). “The Physiological Clock”. Springer, Berlin and New York. Biinning, E. (1972). Symptoms, problems, and common features of circadian rhythms in plants and animals. In “Circadian Rhythmicity”, Proc. int. Symp. Circadian Rhythmicity, Wageningen, 1971, pp. 11-31. Pudoc, Wageningen, Holland. Caldwell, R. L. and Dingle, H. (1967). T h e regulation of cyclic reproductive and feeding activity in the milkweed bug Oncopeltus by temperature and photoperiod. Biol. Bull. mar. biol. Lab., Woods Hole, 133, 510-525. Campbell, B. 0. (1964). Solar and lunar periodicities in oxygen consumption by the mealworm, Tenebrio molitor. Diss. Abstr. 25, 3648-3649. Chambers, D. L. and O’Connell, T. B. (1969). A flight mill for studies with the Mexican fruit fly. Ann. ent. S O C . A m . 62, 917-920. Chandrashekaran, M. K. and Loher, W. (1969). T h e effect of light intensity on the circadian rhythms of eclosion in Drosophila pseudoo bscura. Z. vergl. Physiol. 62, 337-347. Chiba, Y. (1964). The diurnal activity of the mosquito, Culex pipiens pallens, in relation to light condition. I. Activity under the continuous light condition. S c i Rep. T6hoku Uniu., IV (Biol.), 30, 67-75. Chiba, Y. (1966a). The diurnal activity of the mosquito, Culex pipiens pallens, in relation to light condition. 111. The effect of the rearing light condition o n the spontaneous diurnal rhythm. Sci. Rep. T6hoku Uniu.. I V (Biol.), 32. 105-109. Chiba, Y. (1966b). The diurnal activity of the mosquito, Culex pipiens pallens, in relation to the light condition. IV. The developmental stage at which the physiological 24-hr rhythm is initiated. Sci. Rep. T6hoku Uniu., I V (Biol.), 32, 197-200. Chiba, Y. (1967). Activity of mosquitoes, Culex pipiens pallens and Aedes japonicus under a step-wise decrease of light intensity. S c i Rep. Z h o k u Univ. I V (Biol.), 33, 7-13.

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

101

Clayton, D. L. and Paietta, J. V. (1972). Selection for circadian eclosion time in Drosophila melanogaster. Science, N . Y . 178, 994-935. Cloudsley-Thompson, J. L. (1953). Studies in diurnal rhythms. 111. Photoperiodism in the cockroach Periplaneta americana (L.). Ann. M,zg. natl. Hist. (12th). 6 , 705-7 12. Cole C. L. and Adkisson, P. L. (1964). Daily rhythm in the susceptibility of an insect to a toxic agent. Science, N . Y . 144, 1148-1149. Cole, C. L. and Adkisson, P. L. (1965). A circadian rhythm in the susceptibility of an insect to an insecticide. In “Circadian Clocks” (Ed. J. Aschoff), pp. 309-313. North-Holland, Amsterdam. Coluzzi, M. (1972). Inversion polymorphism and adult emergence in Anopheles stephensi. Science, N. Y . 176, 59-60. Conroy, R. T. W. L. and Mills, J. N. (1970). “Human Circadian Rhythms”. Churchill, London. Cook, B. J., de la Cuesta, M. and Pomonis, J. G. (1969). The distribution of factor S in the cockroach, Periplaneta americana. and its role in stress paralysis. /. Insect Physiol. 15. 963-975. Cook D. J. and Milligan, J. V. (1972). Elcctrophysiology and histology of the medial neurosecretory cells in adult male cockroaches, Periplaneta americana. /. Insect Physiol. 18, 1197-1214. Corbet, P. S. (1966). The role of rhythms In insect behaviour. In “Insect Behaviour” (Ed. P. T. Haskell). Symp. R. ent. SOC. Lond. 3, 13-28. Cymborowski, B. (1969). The effect of light on the circadian rhythm of locomotor activity of the house cricket (Acheta domesticus L ). 2001. Poloniae, 19, 85-95. Cymborowski, B. (1370a). Investigations on the neurohormonal factors controlling circadian rhythm of locomotor activity in the house cricket (Acheta domesticus L.). I. The role of the brain and suboesophageal ganglion. Zool. Poloniae, 20, 103-125. Cymborowski, B. (1970b). Investigations on the neurohormonal factors controlling circadian rhythm of locomotor activity in the house cricket Acheta domesticus L. 11. Daily histochemical changes in the neurosecretory cells of the pars intercerebralis and suboesophageal ganglion. Zool. Polonine, 20. 127-149. Cymborowski, B. ( 1 9 7 0 ~ ) .The assumed participation of 5-hydroxytryptamine in regulation of the circadian rhythm of locomotor activity in Acheta domesticus L. Comp. gen. Pharmac. 1, 316-322. Cymborowski, B. and Brady, J. (1972). Insect circadian rhythms transmitted by parabiosis-a re-examination. Nature, Lond. 236, 221-222. Cymborowski, B. and Dutkowski, A. (1969). Circadian changes in RNA synthesis in the neurosecretory cells of the brain and suboesophageal ganglion of the house cricket. /. Insect Physiol. 15, 1187-1197. Cymborowski, B. and Dutkowski, A. (1970a). Circadian changes in protein synthesis in the neurosecretory cells of the central nervous system of Acheta domesticus. 1. Insect Physiol. 16, 341-348. Cymborowski, B. and Dutkowski, A. (1970b). Sex dil’ferences in locomotor activity and RNA synthesis in the central nervous system of Acheta domesticus L. during the circadian rhythm. Z. uergl. Physiol. 7 0 , 29-34. Cymborowski, B. and Flisiliska-Bojanowska, A. (1‘370). f i e effect of light on the locomotor activity and structure of neurosecretory cells of the brain and subesophageal ganglion of Periplaneta arnericana L. Zool. Poloniae, 20,387-399.

152

JOHN BRADY

Cymborowski, B., Skangiel-Kramska, J. and Dutkowski, A. (1970). Circadian changes of acetylcholinesterase activity in the brain of house-crickets (Acheta domesticus L.). Comp. Biochem. Physiol. 32, 367-370. Danilevsky, A. S., Goryshin, N. I. and Tyshchenko, V. P. (1970). Biological rhythms in terrestrial arthropods. A . Rev. Ent. 15, 201-244. Dean, G. J. (W.), Wilson, F. and Wortham, S. (1968). Some factors affecting eclosion of Glossina morsitans Westw. from pupae. Bull. ent. Res. 58, 367-377. Dean, G. J. W.,Paget, J. and Wilson, F. (1969). Observations on the behaviour of tsetse flies (Glossina morsitans orientalis Vanderplank and G . pallidipes Austen) during an attempt to concentrate breeding around cattle. J . appl. Ecol. 6 , 13-26. De BessC, N. (1965). Recherches histophysiologiques sur la neuroskcrbtion dans la chafne nerveuse ventrale d'une blatte, Leucophaea rnaderae (F.). C.r. hebd. Sianc. Acad. Sci., Paris, 260, 7014-7017. Delphin, F. (1965). The histology and possible functions of neurosecretory cells in the ventral ganglia of Schistocerca gregaria Forskil (Orthoptera: Acrididae). Trans. R . ent. S O C .Lond. 117, 167-214. Dingle, H. (1968). Life history and population consequences of density, photo-period, and temperature in a migrant insect, the milkweed bug Oncopeltus. A m . Nut. 102, 149-163. Dingle H. (1972). Migration strategies of insects. Science, N.Y. 175, 1327-1335. Dingle, H., Caldwell, R. L. and Haskell, J. B. (1969). Temperature and circadian control of cuticle growth in the bug, Oncopeltus fasciatus. J. Insect Physiol. 15, 373-378. Dreisig, H. (1971). Control of the glowing of Lampyris noctiluca in the field (Coleoptera: Lampyridae). J. Zool., Lond. 165, 229-244. Dreisig, H. and Nielsen, E. T. (1971). Circadian rhythm of locomotion and its temperature dependence in Blattella germanica. j . exp. Biol. 54, 187-198. Dumortier, B. (1968). Commande nycthCmCrale du rythme d'activtC stridulatoire et r6le des photorCcepteurs chez Ephippiger (lnsectes. Orthopt., Tettigonioiden). Ann. kpiphyties, 19, 147-156. Dumortier, B. (1972). Photoreception in the circadian rhythm of stridulatory activity in Ephippiger (Ins., Orthoptera). Likely existence of two photoreceptive systems. j . comp. Physiol. 77, 80-112. Diirrwachter, G. (1957). Untersuchungen iiber Phototaxis und Geotaxis einiger Drosophila-Mutanten nach Aufzucht in verschiedenen Lichtbedingungen. Z. Tierpsychol. 14, 1-28. Dutkowski, A. B. and Cymborowski, B. (1971). RBle of neurosecretory cells of pars intercerebralis in regulating RNA synthesis in some tissues of Acheta domesticus. /. Insect Physiol. 17, 99-108. Dutkowski, A. B., Cymborowski, B. and Przerqcka, A. (1971). Circadian changes in the ultrastructure of neuroscretory cells of the pars intercerebralis of the house cricket. 1. Insect Physiol. 17, 17 63-1 7 72. Edney, E. B. (1937). A study of spontaneous locomotor activity in Locusta migrutoria migratorioides (R. & F.) by the actograph method. Bull. ent. Res. 28, 243-278. Ehret, C. F. and Trucco, E. (1967). Molecular models for the circadian clock: 1. The chronon concept. J . theor. Biol. 15, 240-262. Eidmann, H. (1956). Uber rhythmische Erscheinungen bei der Stabheuschrecke Carausius morosus Br. Z. uercl. Physiol. 38. 370-390.

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

103

Ellis, P. E. and Hoyle, G. (1954). A physiological interpretation of the marching of hoppers of the African migratory locust (Locusta mi.vatoria migratorioides R. & F.). J. exp. Biol. 31, 271-279. Engelmann, W. and Honegger, H. W. (1966). Tagesperiodischer Schliipfrhythmik einer augenlosen Drosophila melanogaster-Mutante.Naturuiissenschaften, 53. 588. Enright, J. T. (1971a). Heavy water slows biological timing processes. Z. uergl. Physiol. 72, 1-16. Enright, J. T. (1971b). The internal clock of drunker1 isopods. 2. uergl. Physiol. 7 5 , 332-346. Eskin, A. (1971). Properties of the Aplysia visual system: in uitro entrainment of the circadian rhythm and centrifugal regulation of the eye. Z. vergl. Physiol. 74, 3 53-3 7 1. Fernandez, A. T. and Randolph, N. M. (1966). The susceptibility of houseflies reared under various photoperiods to insecticide residues. J. econ. Ent. 59, 37-39. Fingerman, M., Lago, A. D. and Lowe, M. E. (1958). Rhythms of locomotor activity and 02-consumption of the grasshopper Romalea tnicroptera. A m . Midl. Nat. 59, 58-66. Fisher, R. W. (1967). Die1 periodicity in sensitivity of Tetranychus urticae (Acarina: Tetranychidae) to dicofol. Can. E n t . 99, 281-284. Fletcher, B. S. (1966). Cyclical changes in insect physiology and behaviour-studies on the circadian activity rhythm and neurosecretory system of Blaps mucronata Lat. Ph.D. Thesis, University of Birmingham. Fondacaro, J. D. and Butz, A. (1970). Circadian rhythm of locomotor activity and susceptibility to methyl parathion of adult Tenebrio molitor (Coleoptera: Tenebrionidae). Ann. ent. SOC.A m . 63, 952-955. Fowler, D. J. and Goodnight, C. J. (1966a). Neurosecretory cells: daily rhythmicity in Leiobunum longipes. Science, N . Y . 152, 1078-1080. Fowler, D. J. and Goodnight, C. J. (1966b). The cyclic production of 5-hydroxytryptamine in the opilionid. ,4771. 2001. 6 , 187-193 Fowler, D. J., Goodnight, C. J. and Labrie, M. M. (1972). Circadian rhythms of 5-hydroxytyptamine (serotonin) production in larvae, pupae, and adults of Drosophila melanogaster (Diptera: Drosophilidae). Ann. ent. SOC. A m . 65, 138-141. Frudden, L. and Wellso, S. G. (1968). Daily susceptibility of house flies to malathion. /. econ. Ent. 61, 1692-1694. Galun, R. and Margalit, J. (1969). Adenine nucleotides as feeding stimulants of the tsetse fly Clossina austeni Newst. Nature, Lond. 222, 583-584. Gaston, S. and Menaker, M. (1968). Pineal function: the biological clock of the sparrow? Science, N. Y . 160, 1125-1127. George J. E. (1964). The circadian rhythm of “drop-off” of engorged Haemaphysalis leporispalustris from rabbits. Proc. 1st int. C o n p . A carology, Acarologia, 6 (suppl.), 343-349. George, J. E. (1971). Drop-off rhythms of engorged rabbit ticks, Haemaphysalis leporispalustris (Packard, 1896) (Acari: Ixodidae). J. med. Ent. 8,461-479. Gillett, J. D. (1955). Variation in the hatching response of Aedes eggs (Diptera: Culicidae). Bull. ent. Res. 46, 241-254. Gillett, J. D., Corbet, P. S. and Haddow, A. J. (1961). Observations on the ovipositioncycle of Aedes (Stegomyia) aegypti (Linnaeus). VI. Ann. trop. Med. Parasit. 55, 427431.

104

JOHN BRADY

Gillett, J. D., Haddow, A. J. and Corbet, P. S. (1962). The sugar-feeding-cycle in a cage-population of mosquitoes. Entomologia exp. appl. 5, 223-232. Goodman, L. J. (1960). The landing responses of insects. 1. The landing response of the fly, Lucilia sericata, and other Calliphorinae. J. exp. Biol. 37, 854-878. Grace, T. D. C. (1962). Establishment of four strains of cells from insect tissues grown in vitro. Nature, Lond. 195, 788-789. Green, G . W. (1964a). The control of spontaneous locomotor activity in Phormia rcgina Meigen-I. Locomotor activity patterns of intact flies. J. Insect Physiol. 10, 711-726. Green, G. W. (1964b). The control of spontaneous locomotor activity in PhormiQ regina Meigen-11. Experiments to determine the mechanism involved. J. fnsect Physiol. 10, 727-752. Haddow, A. J. and Ssenkubuge, Y. (1962). Laboratory observations on the ovipositioncycle in the mosquito Anopheles (Cellia) gambiac Giles. A n n . trop. Med. Parasit. 56, 352-355. Haddow, A. J., Corbet, P. S. and Gillett, J. D. (1961). Entomological studies from a high tower in Mpanga Forest, Uganda. 7rans. R . ent. SOC.Lond. 113, 249-368. Hardeland, R . and Stange, G . (1971). Einfliisse von Geschlecht und Alter auf die lokomotorische Aktivitat von Drosophila. J. Insect Physiol. 17, 427-434. Harker, J. E. (1954). Diurnal rhythms in Periplaneta americana L. Nature, Lond. 173, 689-690. Harker, J. E. (1955). Control of diurnal rhythms of activity in Periplaneta americana L. Nature, Lond. 175, 733. Harker, J. E. (1956). Factors controlling the diurnal rhythm of activity of Periplaneta americana L. J. e x p . Biol. 33, 224-234. Harker. J. E. (1958a). Diurnal rhythms in the animal kingdom. Biol. Rev. 33, 1-52. Harker, J. E. (1958b). Experimental production of midgut tumours in Periplaneta americana L. J . e x p . Biol. 35. 251-259. Harker, J. E. (1960a). Endocrine and nervous factors in insect circadian rhythms. Cold Spring Harb. S y m p . p a n t . Biol. 25, 279-287. Harker, J. E. (1960b). The effect of perturbations in the environmental cycle of the diurnal rhythm of activity of Periplaneta americana L. J . exp. Biol. 37. 154-163. Harker, J. E. ( 1 9 6 0 ~ ) .Internal factors controlling the sub-oesophageal ganglion neurosecretory cycle in Pcriplaneta americana L. J. cxp. Biol. 37, 164-170. Harker, J. E. (1961). Diurnal rhythms. A . Rev. Ent. 6. 131-146. Harker, J. E. (1963). Tumors. In “Insect Pathology an Advanced Treatise” (Ed. E. A. Steinhaus), Vol. 1, pp. 191-213. Academic Press, New York and London. Harker, J. E. (1964). “The Physiology of Diurnal Rhythms”. Cambridge University Press, London. Harker, J. E. (1965a). The effect of a biological clock on the developmental rate of Drosophila pupae. J. exp. Biol. 42. 323-337. Harker, J. E. (1965b). The effect of photoperiod on the developmental rate of Drosophila pupae. J. e x p . Biol. 43, 4 1 1 4 2 3 . Harris, J. E. (1963). The role of endogenous rhythms in vertical migration. J. mar. biol. ASS. U . K . 43, 153-166. Harshbarger, J. C. and Taylor, R. L. (1968). Neoplasms of insects. A . Rev. Ent. 13, 159-190.

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

105

Hastings, J. W. (1970). Cellular-biochemical clock hypochesis. In “The Biological Clock: Two Views” (F. A. Brown, J. W. Hastings and J. D. Palmer), pp. 61-91. Academic Press, New York and London. Haverty, M. 1. and Ware, G. W. (1970). Circadian sensitivity and dosage-rate response to X-irradiation in the pink bollworm. /. econ. Ent. 63, 1296-1300. Hayes, D. K., Horton, J., Schechter, M. S. and Halberg, F. (1972). Rhythm of oxygen uptake in diapausing larvae of the codling moth at several temperatures. Ann. ent. SOC.A m . 65, 93-96. Hengstenberg, R. (1971). Das Augenmuskelsystem der Stubenfliege Musca domestica. I. Analyse der “clock-spikes” und ihre Quellen. Kyberni-tik, 2, 56-77. Hilliard, S. D. and Butz, A. (1969). Daily fluctuations in the concentrations of total sugar and uric acid in the hemolymph of Periplanetiz americana. Ann. ent. SOC. A m . 62, 71-74. Hinks, C . F. (1967). Relationship between serotonin and the circadian rhythm in some noctuid moths. Nature. Lond. 214, 386-387. Hinks, C. F. (1968). Studies on the neurosecretory cells of the brain in Lepidoptera, with special reference to the circadian rhythm of the adult insects. Ph.D. Thesis, Imperial College, London University. Hoffmann, K. (1965). Overt circadian frequencies and circadian rule. In “Circadian Clocks” (Ed. J. Aschoff), pp. 8 7-94. North-Holland, Amsterdam. Hoffmann, K. (1971). Splitting of the circadian rhythm as a function of light intensity. In “Biochronometry” (Ed. M. Menaker), pp. 134-151. National Academy of Sciences, Washington. Hoyle, G. (1954). Changes in the blood potassium concentration of the African migratory locust (Locusta migratoria migratorioidei R. & F.) during food deprivation, and the effect on neuromuscular activity. 1.e x p . Biol. 31, 260-270. Huber, F. (1965). Neural integration (central nervous system). In “The Physiology of Insecta” (Ed. M. Rockstein), Vol. 2, pp. 333-406. Academic Press, New York and London. Jacklet, J. W. (1969). Circadian rhythm of optic newt: impulses recorded in darkness from isolated eye of AplysM. Science, N. Y . 164, 562-563. Jacklet, J. W. and Geronimo, J. (1971). Circadian rhythm: population of interacting neurons. Science, N. Y.174, 299-302. Jahn, T. L. and Crescitelli, F. (1940). Diurnal changes in the electrical response of the compound eye. Biol. Bull. mar. biol. Lab., Woods Hol‘e, 78,42-52. Jahn, T. L. and Wulff, V. J. (1943). Electrical aspects of a diurnal rhythm in the eye of Dytiscus fasciventris. Physiol. Zool. 16, 101-109. Janda, V. and Mrciak, M. (1957). Gesamtstoffwechsel der Insekten. VI. Die Bewegungsaktivitat der Schabe Periplaneta americana L. w a r e n d des Tages und ihre Beziehung zum Sauerstoffverbrauch. Acta SOC.2001. Bohem. 21, 244-255. Johnson, C. G. (1969). “Migration and Dispersal of Insects by Flight”. Methuen, London. Jones, M. D. R. (1964). The automatic recording of mosquito activity. 1.Insect Physiol. 10, 343-351. Jones, M. D. R., Cubbin, C. M. and Marsh, D. (1972a). The circadian rhythm of flight activity of the mosquito Anopheles gambiae: the light-response rhythm. /. exp. Biol. 57, 337-346.

106

JOHN BRADY

Jones, M. D. R., Cubbin, C. M. and Marsh, D. (1972b). Light-on effects and the question of bimodality in the circadian flight activity of the mosquito Anopheles gambiae. J. exp. Biol. 57, 347-357. Jones, M. D. R., Hill, M. and Hope, A. M. (1967). The circadian flight activity of the mosquito Anopheles gambiae: phase setting by the light rigime. J. exp. Biol. 47, 503-5 11. Kalmus, H. (1938). Das Aktogramm des Flusskrebses und seine Beeinflussung durch Organextrakte. Z. uergl. Physiol. 25, 798-802. Kambysellis, M. P. and Williams, C. M. (1971). In uitro development of insect tislues. 11. The role of ecdysone in the spermatogenesis of silkworms. Biol. Bull. mar. biol. Lab., Woods Hole. 141, 541-552. Kaplanis, J. N., Thompson, M. J., Yamamoto, R. T., Robbins, W. E. and Louloudes, S. J. (1966). Ecdysones from the pupa of the tobacco hornworm. Manduca sexta uohannson). Steroids, 8, 605-623. Kashin, P. (1966). Electronic recording of the mosquito bite. 1.Insect Physiol. 12, 281-286. Klug, H. (1958). Histo-physiologische Untersuchungen iiber die Aktivitatsperiodik bei Carabiden. Wits. Z. Humboldt-Uniu. Berlin, math.-natur. Reihe, 8, 405-434. Koltermann. R. (1971). 24-Std-Periodik in d r r Langzeiterinnerung an Duft- und Farbsignale bei der Honigbiene. Z. uergl. Physiol. 75,49-68. Konopka, R. J. and Benzer, S. (1971). Clock mutants of Drosophila melanogaster. Proc. natn. Acad. Sci. U.S.A. 68, 2112-2116. Kontopp, B. (1967). Lokomotorische Aktivitatsperiodik und Neurosekretion bei der Wachsmotte Galleria mellonella L. Thesis, Gottingen University. (Quoted in Rensing, 1970.) Leclerc, M. M., Schieber, J. P. and Lienhard, J. P. (1971). Le rythme nycthtmiral de la consommation d’oxygkne de la Drosophile. C.T. Sianc. SOC. Biol. Fil. 165,421-425. Lees, A. D. (1964). The location of the photoperiodic receptors in the aphid Megoura uiciae Buckton. J. exp. Biol. 41, 119-133. Lees, A. D. (1968). Photoperiodism in insects. In “Photophysiology”, Vol. 4, pp. 47-137. Academic Press, New York and London. Lees, A. D. (1971). The relevance of action spectra in the study of insect photoperiodism. In “Biochronometry” (Ed. M. Menaker), pp. 372-380. National Academy of Sciences, Washington. Lees, A. D. (1972). The role of circadian rhythmicity in photoperiodic induction in animals. In “Circadian Rhythmicity”. Proc. int. Symp. circadian Rhythmicity, Wageningen, 1971, pp. 87-1 10. Pudoc. Wageningen, Holland. Leppla, N. C. and Spangler, H. G. (1971). A flight-cage actograph for recording circadian periodicity of pink bollworm moths. Ann. ent. SOC. A m . 64, 1431-1434. Leuthold, R. (1966). Die Bewegungsaktivitat der weiblichen Schabe Leucophaea maderae (F.) im Laufe des Fortpflanzungszyklus und ihre experimentelle Beeinflussung.]. Insect Physiol. 12, 1303-1331. Lewis, T. and Taylor, L. R. (1964). Diurnal periodicity of flight by insects. Trans. R . ent. SOC. Lond. 116, 393-476. Lindauer, M. and Martin, H. (1972). Magnetic effect on dancing bees. In “Animal Orientation and Navigation” (Eds S. R. Galler, K. Schmidt-Koenig, G. J. Jacobs and

THE PHYSIOLOGY OF INSECT CIRCADIAN RHVTHMS

107

R. E. Belleville), pp. 559-567. National Aeronautics and Space Administration, Washington. Lipton, G. R. and Sutherland, D. J. (1970). Fecding rhythms in the American cockroach, Periplaneta americana. J. Insect Physiol. 16, 1757-1767. Locke, M., Condoulis, W. V. and Hurshman, L. F. (1!365). Molt and intermolt activities in the epidermal cells of an insect. Science, N . Y. 149, 437-438. Loher, W. (1972). Circadian control of stridulation in the cricket Teleopyllus commodus Walker. J . c o m p . Physiol. 79, 173-190. Loher, W. and Chandrashekaran, M. K. (1970). Circadian rhythmicity in the oviposition of the grasshopper Chorthippus curtipennis. J . Insect Physiol. 16, 1677-1688. Lohmann, M. (1967). Zur Bedeutung der lokomotorischen Aktivitat in circadianen Systemen. 2. uergl. Physiol. 55, 307-332. Macaulay, E. D. M. (1972). A simple insect flight recorder. Entomologia exp. Qppl. 15, 252-254. McCluskey, E. S. and Carter, C. E. (1969). Loss of rhythmic activity in female ants caused by mating. Comp. Biochem. Physiol. 31, 217-226. McFarlane, J. E. (1968). Die1 periodicity in spermatophore formation in the house cricket, Acheta domesticus (L.). Can. J. Zoo!. 46, 695-698. Marler, P. and Hamilton, W. J. (1966). “Mechanisms of Animal Behavior”, Chapter 2. John Wiley, New York. Marsh, D. (1972). Sex pheromone in the aphid Megoura uiciae. Nature, Lond., New Biol. 238, 31-32. Medugorac, I. (1967). Orientierung der Bienen in Raum und Zeit nach Dauernarkose. 2. Bienenforsch. 9, 105-119. Medugorac, I. and Lindauer, M. (1967). Das Zeitgedachtnis der Bienen unter dem Einfluss von Narkose und von sozialen Zeitgebern. Z . uergl. Physiol. 55,450-474. Mellanby, K. (1936). Experimental work with the tsetse-fly, Clossina palpalis, in Uganda. Bull. ent. Res. 27,611-632. Michel, R. (1972). Influence des corpora cardiaca su; les possibilitis de vol soutenu du criquet ptlerin Schistocerca gregariu. J . Insect Physiol. 18. 1811-1827. Minis, D. H. (1965). Parallel peculiarities in the entrainment of a circadian rhythm and photoperiodic induction in the pink boll wonn (Pectinophora gossypiella). In “Circadian Clocks” (Ed. J. Aschoff), pp. 333-343. North-Holland, Amsterdam. Minis, D. H.and Pittendrigh, C. S. (1968). Circadian oscillation controlling hatching: its ontogeny during embryogenesis of a moth. Science, N . Y. 159, 534-536. Mothes, G. (1960). Weitere Untersuchungen iiber den physiologischen Farbwechsel von Carausius morosus (Br.). 2001. Jb. (Physiol.), 69, 133-162. Nayar, J. K. (1967a). Endogenous diurnal rhythm of pupation in a mosquito population. Nature, L o n d 214, 828-829. Nayar, J. K. (1967b). T h e pupation rhythm in A e d e s tneniorhynchus (Diptera: Culicidae). 11. Ontogenetic timing, rate of development, and endogenous diurnal rhythm of pupation. Ann. ent. SOC.A m . 60,946-971’. Nayar, J. K. (1968). The pupation rhythm in Aedes taeniorhynchus. IV. Further studies of the endogenous diurnal (circadian) rhythm of pupation. Ann. ent. SOC.Am. 61, 1408-1417. Nayar, J. K. (1969). The pupation rhythm in Aedes taeniorhynchus. V. Physiology of

108

JOHN BRADY

growth and endogenous diurnal rhythm of pupation. Ann. ent. SOC. Am. 62,

1079-1087. Nayar, J. K. and Sauerman, D. M., Jr. (1970).A comparative study of growth and development in Florida mosquitoes. Part 1: effects of environmental factors on ontogenetic timings, endogenous diurnal rhythm and synchrony of pupation and emergence. J. med. Ent. 7, 163-174. Nayar, J. K. and Sauerman, D. M., Jr. (1971).The effect of light regimes on the circadian rhythm of flight activity in the mosquito Aedes taeniorhynchus. J. exp. Biol. 54, 745-756. Naylor, E. and Atkinson, R. J. A. (1972).Pressure and the rhythmic behaviour of inshore marine animals. Symp., SOC. exp. Biol. 26, 395-415. Naylor, E. and Williams, B. G. (1968). Effects of eyestalk removal on rhythmic locomotor activity in Carcinus. J . exp. Biol. 49, 107-116. Naylor, E., Atkinson, R. J. A. and Williams, B. G. (1971).External factors influencing the tidal rhythm of shore crabs. J . interdiscipl. Cycle Res. 2, 173-180. Naylor. E., Smith, G . and Williams, B. G. (1973).The role of the eyestalk in the tidal activity rhythms of the shore crab Carcinus muenas (L.). In “Neurobiology of Invertebrates, Tihany, 1971” (Ed. J. Salinki), pp. 423-429. Akadtmiai Kiadb, Hungarian Academy of Sciences, Budapest. Neumann, D. (1967).Genetic adaptation in emergence time of Clunio populations to different tidal conditions. Helgolander wiss. Meeresunters. 15, 163-171. Neumann, D. (1968).Die Steuerung einer semilunaren Schliipfperiodik mit Hilfe eines kunstlichen Gezeitenzyklus. Z . vergl. Physiol. 6 0 , 63-78. Neumann, D. (1971a).Eine nicht-reziproke Kreuzungssterilitat zwischen okologischen Rassen der Miicke Clunio marinus. Oecologia, 8, 1-20. Neumann, D. (1971b). The temporal programming of development in the intertidal chironomid Clunio marinus (Diptera: Chironomidae). Can. Ent. 103,3 15-318. Neville, A. C. (1963).Daily growth layers in locust rubber-like cuticle, influenced by an external rhythm. J. Insect Physiol. 9, 177-186. Neville, A. C. (1965).Circadian organization of chitin in some insect skeletons. Q. /. micr. Sci. 106, 315-325. Neville, A. C. (1967a).Chitin orientation in cuticle and its control. In “Advances in Insect Physiology” (Eds J. W. L. Beament, J. E. Treherne and V. B. Wigglesworth), Vol. 4,pp. 213-286.Academic Press, London and New York. Neville, A. C. (1967b).A dermal light sense influencing skeletal structure in locusts. J. Insect Physiol. 13,933-939. Neville, A. C. (1970).Cuticle ultrastructure in relation to the whole insect. In “Insect Ultrastructure” (Ed. A. C. Neville), Symp. R . ent. SOC. Lond. 5, 17-39. Nielsen, E. T. and Evans, D. G. (1960). Duration of the pupal stage of Aedes taeniorhynchus with a discussion of the velocity of development as a function of temperature. Oikos, 11, 200-222. Nielsen, H. T. and Nielsen, E. T. (1962). Swarming of mosquitoes. Laboratory experiments under controlled conditions. Entomologia exp. appl. 5 , 14-32. Nishiitsutsuji-Uwo, J. and Pittendrigh, C. S. (1967).T h e neuroendocrine basis of midgut tumour induction in cockroaches. /. Insect Physiol. 13, 851-859. Nishiitsutsuji-Uwo, J. and Pittendrigh, C. S. (1968a).Central nervous system control of circadian rhythmicity in the cockroach. 11. The pathway of light signals that entrain the rhythm. Z. uergl. Physiol. 58, 1-13.

THE PHYSIOLOGY OF INSECT CIRCADIAN RHY’T’HMS

109

Nishiitsutsuji-Uwo, J. and Pittendrigh, C. S. (1968b).Central nervous system control of circadian rhythmicity in the cockroach. 111. The optic lobes, locus of the driving oscillation? Z . uergl. Physiol. 58, 14-46. Nishiitsutsuji-Uwo, J., Petropulos, S. F. and Pittendrigh, C. S. (1967).Central nervous system control of Circadian rhythmicity in the cockroach. I. Role of the pars intercerebralis. Biol. Bull. mar. biol. Lab., Woods Hole, 133, 679-696. Nowosielski, J. W. and Patton, R. L. (1963).Studies on circadian rhythm of the house cricket, Gryllus domesticus L. J. Insect Physiol. 9,401-410. Nowosielski, J. W. and Patton, R. L. (1964). Daily fluctuation in the blood sugar concentration of the house cricket, Gryllus domesticus L. Science, N.Y. 144,

180-181. Nowosielski, J. W., Patton, R. L. and Naegele, J . A. (1964).Daily rhythm of narcotic sensitivity in the house cricket, Gryllus domesticus L., and the two-spotted spider mite, Tetranychus urticae Koch. J . cell. comp. Physrul. 63,393-398. Odhiambo, T. R. (1966).The metabolic effects of the corpus allatum hormone in the male desert locust. 11. Spontaneous locomotor activity. J. exp. Biol. 45, 51-63. Page, T. L. and Larimer, J. L. (1972).Entrainment of the circadian locomotor activity rhythm in crayfish. The role of the eyes and caudal photoreceptor. J . comp. Physiol.

78, 107-120. Parker, A. H. (1962).Studies on the diurnal rhythms of the housefly, Musca domestica L., in a dry tropical environment. Acta troprca, 19,97-119. Pavlidis, T. (1969). Populations of interacting oscillators and circadian rhythms. J. theor. Biol. 22, 418-436. Pavlidis, T. (1971).Populations of biochemical oscillators as circadian clocks. J. theor. Biol. 3 3 , 319-338. Pavlidis, T., Zimmerman, W. F. and Osborn, J. (1968) A mathematical model for the temperature effects on circadian rhythms. J. theor. Biol. 18,210-221. Payne, T. L., Shorey, €I. H. and Caston, L. K. (1970).Sex pheromones of noctuid moths: factors influencing antennal responsiveness in males of Trichoplusia ni. J. Insect Physiol. 16,1043-1055. Pfliiger, W. and Neumann, D. (1971).Die Steuerung einer gezeitenparallelen Schliipfrhythmik nach dem Sanduhr-Prinzip. Oecologia, 7, 262-266. Phelps, R. J. and Jackson, P. J. (1971).Factors influencing the moment of larviposition and eclosion in Clossina morsitans orientalis Vanderplank (Diptera: Muscidae). J. ent. SOC.sth. Afr. 34, 145-157. Pichon, Y. and Boistel, J. (1963). Modifications of the ionic content of the haemolymph and of the activity of Peripfaneta americana in relation to diet. 1.Insect Physiol. 9,887-891. Pilson, R. D. and Pilson, B. M. (1967). Behaviour studies of Clossina morsitans Westw. in the field. Bull. ent. Res. 57, 227-257. Pittendrigh, C. S. (1954).On temperature independence in the clock system controlling emergence time in Drosophila. Proc. natn. Acad. S c i U.S.A. 40, 1018-1029. Pittendrigh, C. S. (1960). Circadian rhythms and the circaden organization of living systems. Cold Spring Harb. Symp. quant. Biol. 25, 1119-184. Pittendrigh, C. S. (1965).On the mechanism of the entrainment of a circadian rhythm by light cycles. In “Circadian Clocks” (Ed. J. Aschoff), pp. 27 7-297.North-Holland, Amsterdam.

110

JOHN BRADY

Pittendrigh, C. S. (1966). The circadian oscillation in Drosophila pseudoobscura pupae: a model for the photoperiodic clock. Z. Pflanzenphysiol. 54, 275-307. Pittendrigh, C. S. (1967). Circadian systems, I. T h e driving oscillation and its assay in Drosophila pseudoobscura. Proc. natn. Acad. Sci. U.S.A. 58, 1762-1767. Pittendrigh, C. S. and Minis, D. H. (1964). The entrainment of circadian oscillations by light and their role as photoperiodic clocks. Am. Nut. 98, 261-294. Pittendrigh, C. S. and Minis, D. H. (1971). The photoperiodic time measurement in Pectinophora gossypiella and its relation to the circadian system in that species. In “Biochronometry” (Ed. M. Menaker), pp. 212-250. National Academy of Sciences, Washington. Pittendrigh, C. S. and Minis, D. H. (1972). Circadian systems: longevity as a function of circadian resonance in Drosophila melanogaster. Proc. natn. Acad. Sci. U.S.A. 69, 1537-1539. Pittendrigh, C. S. and Skopik, S. D. (1970). Circadian systems, V. The driving oscillation and the temporal sequence of development. Proc. natn. Acad. S c i U.S.A. 65, 500-507. Pizzarello, D. J., Isaak, D., Chua, K. E. and Rhyne, A. L. (1964). Circadian rhythmicity in the sensitivity of two strains of mice to whole-body radiation. Science, N. Y . 145, 286-29 1. Polcik, B. (1968). Daily activity rhythms of some insects with particular consideration of their sensitivity to insecticides. (In Polish, English summary.) Bull. ent. Pologne, 38,497-538. Polcik, B., Nowosielski, J. W. and Naegele, J. A. (1964). Daily sensitivity rhythm of the two-spotted spider mite, Tetranych?rs urticae, to DDVP. Science, N . Y . 145,405-406. Polcik, B., Nowosielski, J. W. and Naegele, J. A. (1965). Daily rhythm of oviposition in the two-spotted spider mite. /. econ. Ent. 58,467-469. Powell, B. L. (1965). The hormonal control of the tidal rhythm of locomotor activ,ity in Carcinus maenas. Cen. comp. Endocr. 5 , 705-706. Provost, M. W. and Lum. P. T. M. (1967). T h e pupation rhythm in Aedes taeniorhynchus (Diptera: Culicidae). I. Introduction. Ann. ent. SOC. A m . 60, 138-149. Raabe, M. (1963a). Mise en tvidence, chez des insectes d’ordres variis, d’iliments neurosicriterurs trit~icCrtbraux. C.r. hebd, Skanc. Acad. S c i , Paris, 257, 11 711173. Raabe. M. (1963b). Existence chez divers insectes d’une innervation tritoctrtbrale des corpora cardiaca. C.r. hebd. Sdanc. Acad. Sci., Paris, 257, 1552-1555. Raabe M. ( 1 9 6 3 ~ ) .Recherches exptrimentales sur la localisation intra-ctrtbrale du facteur chromactif des insectes. C.r. hebd. Skanc. Acad. Sci., Paris, 257, 1804-1806. Raabe, M. (1965). Etude des phinomenes de neurosicrktion au niveau de la chaine nerveuse ventrale des phasmides. Bull. SOC.zool. Fr. 90, 631-654. Raabe, M. (1966). Recherches sur la neurostcrition dans la chaine nerveuse ventrale du Phasme, Clitumnus extradentatus: variations d’activiti des diffirents tltments neurosicriteurs. C.r. hebd. Sdanc. Acad. Sci., Paris, 262, 303-306. Ralph, C. L. (1959!. Modifications of activity rhythms of Periplaneta americana (L.) induced by carbon dioxide and nitrogen. PhysioL Zool. 32, 57-62. Ralph, C. L. (1962). Heart accelerators and decelerators in the nervous system of Periplaneta americam (L.). J. Insect Physiol. 8.431439.

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

111

Rankin, M. A., Caldwell, R. L. and Dingle, H. (1972). An analysis of a circadian rhythm of oviposition in Oncopeltus fasciatus. J. exp. Biol. 515, 353-359. Rao, K. P. (1963). Some aspects of the electrical activity in the central nervous system of the scorpion, Heterometrus swammerdumi. R o c . XVI int. Congr. Zool. 2 , 69. Rao, K. P. (1964). Neurophysiological studies on an arachnid, the scorpion Heterometrus fulvipes. J. Anim. Morph. Physiol. 11, 133-14’2. Rao, K. P. and Gropalakrishnareddy, T. (1967). Blood borne factors in circadian rhythms of activity. Nature, Lond. 213, 1047-1048. Reinhardt, R. and Esther, H. (1971). Untersuchungen zur tageszeitlichen Empfindlichkeit einiger Insekten gegen Organo-Phosphor-Insektizide.Zool. Jb. (Syst.) 98, 51 1-519. Remmert, H. (1962). “Der Schliipfrhythmus der Insekten”. Steiner Verlag, Wiesbaden. Renner, M. (1957). Neue Versuche iiber den Zeitsinn der Honigbiene. Z . vergl. Physiol. 40, 85-118. Renner, M. (1960). The contribution of the honey bee to the study of time-sense and astronomical orientation. Cold Spring Hurb. Symp. qitunt. Biol. 25, 361-367. Rensing, L. (1964). Daily rhythmicity of corpus allatum and neurosecretory cells in Drosophila melanoguster (Meig). Science, N. Y . 144. 1586-1587. Rensing, L. (1965a). Tagesperiodik von Aktivitat und Phototaxis bei Corixa punctutu und Anticorixa sahlbergi. 2. vergl. Physiol. 50. 250-2!i3. Rensing, L. (1965b). Ontogenetic timing and circadian rhythms in insects. In “Circadian Clocks” (Ed. J. Aschoff), pp. 406-41 2. North-Holland, Amsterdam. Rensing, L. (1966a). Zur circadianen Rhythmik ties Sauerstoffverbrauches von Drosophila. Z . vergl. Physiol. 53, 62-83. Rensing, L. (1966b). Aspects of the circadian organisaticm of Drosophilu. Union int. Sci. biol. 4, 37-45. Rensing, L. ( 1 9 6 6 ~ )Zur . circadianen, Rhythmik des Hormonsystems von Drosophilu. 2. Zellforsch. 74, 539-558. Rensing, L. (1969a). Ein circadianer Rhythmus der Empfindlichkeit gegen Rontgenstrahlen bei Drosophilu. Z . vergl. Physiol, 6 2 , 214-220. Rensing, L. (1969b). Tagesperiodik von Zellfunktionen und Strahlenempfindlichkeit in Normal- und Tumorgewebe. Nuturw. Rdsch. Stuttguri:, 22, 390-396. Rensing, L. (1969c). Circadiane Rhythmik von Drosoi’lhilu-Speicheldriisen in vivo, in vitro und nach Ecdysonzugabe. J . Insect Physiol. 15, 2285-2303. Rensing, L. (1970). Zur Rolle der RNS-Synthese bei zellularen circadianen Rhythmen. Verh. d t . 2001. Ces. 1970. Zool. Anz. (Suppl.), 34, 2E4P-287. Rensing, L. (1971). Hormonal control of circadian rhythms in Drosophila. In “Biochronometry” (Ed. M. Menaker), pp. 527-540. National Academy of Sciences, Washington. Rensing, L. and Hardeland, R. (1967). Zur Wirkung deir circadianen Rhythmik auf die Entwicklung von Drosophilu. J. Insect Physiol. 13, 1547-1568. Rensing, L., Brunken, W. and Hardeland, R. (1968). On the genetics of a circadian rhythm in Drosophila. Experientiu. 24, 509-5 10. Rensing. L., Thach, B. and Bruce, V. (1965). Daily rhythms i n the endocrine glands of Drosophila larvae. Experientiu, 21, 103-104. Richards, A. G. (1969). Oxygen consumption of ‘the American cockroach under complete inanition prolonged 10 death. Ann. ent. S O C . A m . 62, 1313-1316.

112

JOHN BRADY

Richards, A. G. and Halberg, F. (1964). Oxygen uptake rhythms in a cockroach gauged by variance spectra. Experientia, 20.40-42. Riddiford, L. M. and Williams, C. M. (1971). Role of the corpora cardiaca in the behavior of saturniid moths. I. Release of sex pheromone. Biol. Bull. mar. biol. Lab., Woods Hole, 140, 1-7. Ringelberg, J. and Servaas, H. (197 1). A circadian rhythm in Daphnia magna. Oecologia, 6, 289-292. Roberts, S. K. de F. (1960). Circadian activity rhythms in cockroaches. I. T h e free-running rhythm in steady-state. J. cell. comp. Physiol. 55, 99-1 10. Roberts, S. K. de F. (1962). Circadian activity rhythms in cockroaches. 11. Entrainment and phase shifting. J . cell. comp. Physiol. 59, 175-186. Roberts, S. K. (de F.) (1965). Photoreception and entrainment of cockroach activity rhythms. Science, N. Y. 148,958-959. Roberts, S. K. de F. (1966). Circadian activity rhythms in cockroaches. 111. T h e role of endocrine and neural factors. J. cell. Physiol. 67,473-486. Roberts, S. K. (de F.) (1971). Comment, in Roberts et al. (1971). In “Biochronornetry” (Ed. M. Menaker), pp. 514-516. National Academy of Sciences, Washington. Roberts, S. K. (de F.), Skopik, S. D. and Driskill, R. J. (1971). Circadian rhythms in cockroaches: does brain hormone mediate the locomotor cycle? In “Biochronometry” (Ed. M. Menaker). pp. 505-516. National Academy of Science, Washington. Roeder, K. D. (1967). “Nerve Cells and Insect Behavior” (2nd ed.). Harvard University Press, Cambridge, Mass. Rothert, H. (1970). Tagesperiodische Schwankungen der Empfindlichkeit von Drosophila melanogaster gegeniiber Parathion. 2. angew. Ent. 65,403-409. Rounds, H. D. (1968). Diurnal variation in the effectiveness of extracts of cockroach midgut in the release of intestinal proteinase activity. Comp. Biochem. Physiol. 25, 1125-1128. Saunders, D. S. (1970). Circadian clock in insect photoperiodism. Science, N.Y. 168, 601-603. Schallek, W. (1942). Some mechanisms controlling locomotor activity in the crayfish. J. exp. Zool. 91, 155-166. Schechter, M. S., Dutky, S. R. and Sullivan, W. N. (1963). Recording circadian rhythms of the cockroach with a capacity-sensing device. J. econ. Ent. 56, 76-79. Schlein, Y. (1972a). Factors that influence the post-emergence growth in Smcophaga falcukzta. J. Insect Physiol. 18, 199-209. Schlein, Y. (1972b). Postemergence growth in the fly Sarcophagafalculata initiated by neurosecretion from the ocellar nerve. Nature, Lond., New Biol. 236, 217-219. Schlein, Y. and Gratz, N. G. (1972). Age determination of some flies and mosquitos by daily growth layers of skeletal apodemes. Bull. Wld Hlth Org. 47, 71-75. Schweiger, H. G. (1972). Circadian rhythms: subcellular and biochemical aspects. In “Circadian Rhythmicity”, Proc. int. Symp. circadiun Rhythmicity, Wageningen, 1971, pp. 157-174. Pudoc, Wageningen, Holland. Shaaya, E. and Karlson, P. (1965). Der Ecdysontitre wahrend der Insektenentwicklung11. Die postembryonale Entwicklung der Schmeissfliege Calliphora erythrocephalu Meig. J. Insect Physiol. 11, 65-69. Shepard, M. and Keeley, L. L. (1972). Circadian rhythmicity and capacity for enforced

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

113

activity in the cockroach, Blaberus discoidalis, after cardiacectomy-allatectomy. /. Insect Physiol. 18, 595-601. Shorey, H. H. and Gaston, L. K. (1965). Sex pheromones of noctuid moths. V. Circadian rhythm of pheromone-responsiveness in males of Autographa californica, Heliothis virescens. Spodoptera exigua, and Trichoplusia ni (Lepidoptera: Noctuidae). Ann. ent. SOC. A m . 58, 597-600. Skopik, S. D. and Pittendrigh, C. S. (1967). Circadian systems, 11. The oscillation in the individual Drosophda pupa; its independence of developmental stage. Proc. natn. Acad. S c i U.S.A. 58, 1862-1869. Smith, U. (1967). The mechanism and control of co1ov.r change in the stick insect, Carausius morosus. Ph.D. Thesis, Cambridge University. Sower, L. L., Shorey, H. H. and Gaston, L. K. (1970). Sex pheromones of noctuid moths. XXI. Light:dark cycle regulation and light inhibition of sex pheromone release by females of Trichoplusia ni. Ann. ent. SOC. A m . 6 3 , 1090-1092. Sower, L. L., Shorey, H. H. and Gaston, L. K. (1971). Sex pheromones of noctuid moths. XXV. Effects of temperature and photoperiod on circadian rhythms of sex pheromone release by females of Trichoplusia ni. Ann. c’nt. SOC.A m . 64, 488-492. Spangler, H. G. (1972). Daily activity rhythms of individual worker and drone honey bees. Ann. ent. SOC. A m . 65, 1073-1076. Strejckovi, A., Servit, Z.and Novik, V. J. A. (1965). Effect of neurohormones C1 and D, on spontaneous electrical activity of the central nervous system of the .cockroach./. Insect Physiol. 11,889-896. Strumwasser, F. (1965). The demonstration and manipulation of a circadian rhythm in a single neuron. In “Circadian Clocks” (Ed. J. Aschoff),pp. 442-462. North-Holland, Amsterdam. Sullivan, W. N., Cawley, B., Hayes, D. K., Rosenthal, J. and Halberg, F. (1970). Circadian rhythm in susceptibility, of house flies and Madeira cockroaches to pyrethrum.]. econ. Ent. 63, 159-163. Sullivan, W. N., Schechter, M. S., Dutky, S. R. and Keller, J. C. (1962). Monitoring electrophysiological responses of cockroaches for space research. /. econ. Ent. 55, 985-989. Sweeney, B. M. (1972). Circadian rhythms in unicellular organisms. In “Circadian Rhythmicity”, Proc. int. Symp. circadian Rhythmicity. Wageningen, 1971, pp. 137-156. Pudoc, Wageningen, Holland. Szymanski, J. S. (1914). Eine Methode zur Untersuchiing der Ruhe und Aktivitatsperioden bei Tieren. Pfliig. Arch. ges. Physiol. 158, 343-385. Taylor, B. and Jones, M. D. R. (1969). The circadian rhythm of flight activity in the mosquito Aedes aegypti (L.): the phase-setting effects of light-on and light-off. /. exp. Biol. 51, 59-70. Thomas, R. and Finlayson, t. H. (1970). Initiation of circadian rhythms in arrhythmic churchyard beetles (Blaps mucronata). Nature, Lond. 228. 577-578. Traynier, R. M. M. (1970). Sexual behaviour of the meditcrranean flour moth, Anagasta kiihniella: some influences of age, photoperiod, and light intensity. Can. Ent. 102, 534-540. Truman, J. W. (1971a). Hour-glass behavior of the circadian clock controlling eclosion of the silkmoth Antheraea pernyi. Proc. natn. Acad. SCI. U.S.A. 6 8 , 595-599.

114

JOHN BRADY

Truman, J . W. (1971b). Physiology of insect ecdysis. I. The eclosion behaviour of saturniid moths and its hormonal release. J. exp. Biol. 54, 805-814. Truman, J. W. ( 1 9 7 1 ~ ) .T h e role of the brain in the ecdysis rhythm of silkmoths: comparison with the photoperiodic termination of diapause. In “Biochronometry” (Ed. M. Menaker), pp. 483-504. National Academy of Sciences, Washington. Truman, J. W. (1972a). Circadian rhythms and physiology with special reference to neuroendocrine processes in insects. In “Circadian Rhythmicity”, Proc. int. Symp. circadian Rhythmicity, Wageningen, 1971, pp. 111-135. Pudoc, Wageningen, Holland. Truman, J. W. (1972b). Involvement of a photoreversible process in the circadian clock controlling silkmoth eclosion. Z. uergl. Physiol. 76, 32-40. Truman, J. W. ( 1 9 7 2 ~ )Physiology . of insect rhythms. I. Circadian organization of the endocrine events underlying the moulting cycle of larval tobacco hornworms. J . exp. Biol. 57, 805-820. Truman, J. W. (1972d). Physiology of insect rhythms. 11. The silkmoth brain as the location of the biological clock controlling eclosion. J. comp. Physiol. 81, 99-114. Truman, J. W. and Riddiford, L. M. (1970). Neuroendocrine control of ecdysis in silkmoths. Science, N . Y . 167, 1624-1626. Truman, J. W. and Sokolove, P. G. (1972). Silk moth eclosion: hormonal triggering of a centrally programmed pattern of behavior. Science, N . Y . 175, 1491-1493. Turner, R. B. and Acree. F., Jr. (1967). The effect of photoperiod on the daily fluctuation of haemolymph hydrocarbons in the American cockroach. J. Insect Physiol. 13, 519-522. Tychsen, P. H. and Fletcher, B. S. (1971). Studies on the rhythm of mating in the Queensland fruit fly, Dacus tryoni. J. Insect Physiol. 17, 2139-2156. Vanable, J. W., Jr. (1963). A ninhydrin reaction giving a sensitive quantitative fluorescence assay for 5-hydroxytryptamine. Analyt. Biochem. 6 , 393-403. Van Cassel, M. (1968). Contribution a I’itude du rBle des yeux et des ocelles dans la prise et le maintien du rythme d’activitt locomotrice circadien chez Periplaneta americana. Ann. Epiphyties, 19, 159-164. Venkatachari, S. A. T. (1971). Analysis of spontaneity in the nervous system of the scorpion, Heterometrus fulvipes. Ind. J. exp. Biol. 9, 338-344. Venkatachari, S. A. T. and Muralikrishna Dass, P. (1968). Choline esterase activity rhythm in the ventral nerve cord of scorpion. Life S c i 7 (Pt. 11), 617-621. Ware, G. W. and McComb, M. (1970). Circadian susceptibility of pink bollworm moths to azinphosmethyl. J . econ. Ent. 63, 1941-1942. Weber, W. and Gaude, H. (1971). Ultrastruktur des Neurohaemalorgans im Nervus corporis allati I1 von Acheta domesticus. 2. Zellforsch. 121, 561-572. Wigglesworth, V. B. (1965). “The Principles of Insect Physiology” (6th ed.), p. 305. Methuen, London. Wilkins, M. B. (1960). A temperature-dependent endogenous rhythm in the rate of carbon dioxide output of Periplaneta americana. Nature, Lond. 185,481-482. Wilkins, M. B. (1965). The influence of temperature and temperature changes on biological clocks. In “Circadian Clocks” (Ed. J. Aschoff), pp. 146-163. NorthHolland, Amsterdam. Willey, R. B. (1961). The morphology of the stomodeal nervous system in Periplaneta americana (L.) and other Blattaria. J. Morph. 108, 219-261.

THE PHYSIOLOGY OF INSECT CIRCADIAN RHYTHMS

115

Williams, B. G. and Naylor, E. (1967). Spontaneously induced rhythm of tidal periodicity in laboratory-reared Carcinus. J. exp. Biol. 47, 229-234. Williams, C. M. and Adkisson, P. L. (1964). Physiology of insect diapause. XIV. An endocrine mechanism for the photoperiodic control of pupal diapause in the oak silkworm, Antheraea pernyi. Biol. Bull. mar. bzol. Lab., Woods Hole, 127, 51 1-525. Wilson, E. 0. (1971). “The Insect Societies”. University Press Harvard, Cambridge, Mass. Winfree, A. T. (1967). Biological rhythms and the behavior of populations of coupled oscillators. J. theor. Biol. 16, 15-42. Winfree, A. T. (1970a). The temporal morphology of a biological clock. In “Lectures on Mathematics in the Life Sciences” (Ed. M. Gerstenhaber), Vol. 2, pp. 109-150. American Mathematical Society, Providence, USA. Winfree, A. T. (1970b). Integrated view of resetting a circadian clock. /. theor. Biol. 28, 327-3 74. Winfree, A. T. (1972a). Acute temperature sensitivity of the circadian rhythm in Drosophila. J. Insect Physiol. 18, 181-185. Winfree, A. T. (1972b). Slow dark-adaptation in Drosophila’s circadian clock. J . comp. Physiol. 77, 418-434. Winfree. A. T. ( 1 9 7 2 ~ ) .On the photosensitivity of the circadian time-sense in Drosophila pseudoobscura. J. theor. Biol. 35, 159-189. Wobus, U. (1966a). Der Einfluss der Lichtintensitat auf die Resynchronisation der circadianen Laufaktivitat drr Schabe Blaberus cranifir Burm. (lnsecta: Blattariae). Z. vergl. Physiol. 52. 276-289. Wobus, U. (1966b). Der Einfluss der Lichtintensitat auf die circadiane Laufaktivitat der Schabe Blaberus craniifer Burm. (Insecta: Blattariae). Biol. Zbl. 85, 305-323. Wright, J. E., Kappus, K. D. and Venard, C. E. (1966). Swirming and mating behavior in laboratory colonies of Aedes txiseriatus (Diptera: Culicidae). Ann. ent. Sac. A m . 59, 1110-1112. Wurtman, R. J. (1967). Ambiguities in the use of the term circadian. Science, N. Y . 156, 104. Youthed, G. J. and Moran, V. C. (1969a). The solar-day activity rhythm of myrmeleontid larvae. J . Insect Physiol. 15. 1103-1116. Youthed, G. J. and Moran, V. C. (1969b). The lunar-day activity rhythm of myrmeleontid larvae. J. Insect Physiol. 15, 1259.1271. Zelazny, B. and Neville, A. C. (1972). Endocuticle layer formation controlled by non-circadian clocks in beetles. J . Insect Physiol. 18, 1967-1979. Zimmerman, W. F. (1969). On the absence of circadian rhythmicity in Drosophila pseudoobscura pupae. Biol. Bull. mar. hiol. Lab., Woods Hole, 136,494-500. Zimmerman, W. F. and Goldsmith, T. H. (1971). Photosensitivity of the circadian rhythm and of visual receptors in carotenoid-depleted Drosophila. Science, N . Y . 171, 11 67-1169. Zimmerman, W. F. and Ives, D. (1971). Some photophysiological aspects of circadian rhythmicity in Drosophila. In “Biochronometry” (Ed. M. Menaker), pp. 381-391. National Academy of Sciences, Washington. Zimmerrnan, W. F., Pittendrigh, C. S. and Pavlidis, T. (1 968). Temperature compensation of the circadian oscillation in Drosophila pseudoobscura and its entrainment by temperature cycles. J. Insect Physiol. 14, 669-684.

This Page Intentionally Left Blank

The Tryptophan- Ommochrome Pathway in Insects' Bernt Linzen Zoological Institute, University of Munich, Germany

1 Introduction . 118 . 120 2 Fluorescent tryptophan metabolites found in insects 2.1.Notes on methodology . 120 . 122 2.2. 'Tryptophan and its fluorescent metabolites . 132 3 The absence of the glutarate pathway . 134 4 Ommochromes . 134 4.1 Notes on nomenclature . 135 4.2 Isolation . . 138 4.3 Properties . 150 4.4 Distribution and tissue localization . . 162 4.5 Deposition of ommochromes: the pigment granule!; . 164 4.6 Binding of ommochromes to proteins . 165 5 Functions of ommochromes . 166 5.1 Ommochromes as screening pigments . 169 5.2 Ommochromes as pattern pigments. Relation to other pigments . 173 5.3 Ommochromes in morphological colour change . 176 5.4 Ommochromes as waste products . . 179 6 Enzymes involved in the kynurenine pathway . 180 6.1 Tryptophan oxygenase (EC 1.13.1.12) . 189 6.2 Kynurenine formamidase (aryl-formylamine amidohydrolase EC 3.5.1.9) . 189 6.3 Kynurenine-3-hydroxylase(EC 1.14.1.2) 6.4 Kynureninase and kynurenine transaminase (EC 3.7.1.3;2.6.1.7) . . 193 7 Ommochrome biosynthesis . 193 8 Tryptophan metabolism in insect development . 197 8.1 Eggs and embryonal development . 197 8.2 Larval development. Hemimetabola . 199 8.3 Accumulation of tryptophan metabolites during metamorphosis of holo. 201 metabolous insects . 212 8.4 Ontogeny of enzyme activities

.

.

I This paper is dedicated to Professor Adolf Butenandl. o n thc occasion of his 70th birthday, in token of my gratitude and admiration for his inspiring leadership, his fervent devotion to the chemistry of life, and his lucid manner of conveying thought and scholarship.

117

118

8.5 Attempts to establish a tryptophan balance 9 Detrimental effects of tryptophan and of tryptophan metabolites 10 Concluding remarks Acknowledgements References

BERNT LINZEN

. . . . .

218 220 223 224 225

1 Introduction

'Tryptophan is an outstanding amino acid. It has the highest molecular weight of all amino acids occurring in proteins, and comprises, in its indolyl moiety, a system capable of donating electrons and, therefore, liable to form complexes with a range of other molecules. Molecular interactions have been observed between tryptophan and nucleic acids, and a variety of other molecules of biological importance (details will be given in section 9). It is obvious that this property is both beneficial (allowing recognition and association of reacting systems, e.g. macromolecules) and detrimental (by interference with such associations). Thus, tryptophan may be viewed as a valuable ingredient of living matter; an ingredient, however, which must be utilized with some precaution. It can accordingly be readily appreciated why tryptophan is less common than other amino acids. 'The average tryptophan content of animal proteins is about 1 per cent on a weight basis (Block and Weiss, 1956), corresponding to little more than ?4 per cent on a molar basis-only one tenth of what could be expected if the amino acids were uniformly distributed. In vertebrate blood plasma the level of free tryptophan is extremely low (0.01 1 mg ml-' in human plasma) and efficient mechanisms exist to prevent it from rising. In insects, which are characterized by extraordinarily high concentrations of free amino acids, the level of tryptophan is not proportionally higher, often being below the level of detection. In insects too, the tryptophan level seldom rises, and if it does, then only for short periods. On the other hand, organisms have taken advantage of the potentialities of this molecule and have transformed it into a wide range of biologically active compounds. In these either the indole ring or its benzene nucleus are retained, the latter also being recast into the pyridine ring. Among these are a plant growth hormone (indole acetic acid), an enzyme cofactor (nicotinic acid), a hormone (melatonin), a neurotransmitter substance (serotonin), and such powerful drugs as the ergot and Rauwolfia alcaloids. Of the amino acids, only phenylalanine is comparable with regard to the highly branched metabolic pathways and great variety of compounds formed. In insects, a major product of tryptophan degradation is a group of 10 to 15 pigments, the ommochromes. Being brownish-yellow, bright red or dark violet-purple, these pigments produce the deep tinge of insect eyes and

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

119

contribute to the brilliant colouration of many species. Pigmentation itself has always attracted investigators. But the principal stimulus for investigations in this field was the discovery of pigment mutants which were widely used for study by geneticists and deve1opment;d physiologists. The search for the principle mediating the pigmentation of insect eyes represented the first important step into the field of biochemical genetics. The metabolic pathway from tryptophan tci ommochromes forms the subject of the present review. Its aim is to assemble data which contribute to a physiological understanding of this pathway in insects, to appraise the role of ommochromes in insect life and the relation of tryptophan metabolism to insect development. It is intended to describe ommochromes in some detail, but not to present ommochrome chemistry proper, as this would require a good deal of phenoxazine chemistry in general. Furthermore, the genetic aspects of ommochromes will not be covered, although, since Ziegler’s (1961) article, a great many new mutants have been described. However, work on mutants will be called upon whenever it contributes to the understanding of the normal function. Finally, the histology of ommochrome pigmentation will be omitted, except for a brief treatment of the pigment granules and a short chapter on ommochromes as pattern pigments, indicating their role in morphological colour change. Chromatophores and their hormonal control will, however, be completely omitted. The earliest phase of ommochrome biochemistry was characterized by the search for the factors linking gene and phenotype. This was initiated by the transplantation experiments on Ephestia (Kiihn, 1932; Caspari, 1933) and on Drosophila (Beadle and Ephrussi, 1935; Ephrussi and Beadle, 1935). These led to the recognition of the diffusible, colour inducing substances (the alleged “gene-hormones”) as pigment precursors (v’ = a’-substance = kynurenine; cn’substance = 3-hydroxy-kynurenine (Butenandt et al., 1940a, 1949)), and to the establishment of a scheme in which the genes were proposed to govern a sequence of biochemical reactions by providing specific enzymes. Details and references of this early period may be found in the reviews of Becker (1938, 1942), Plagge (1939), Kiihn (1941), Ephrussi (1942a, 1942b), Caspari (1949),and Ziegler (1961). Ommochromes themselves were first studied broadly and recognized as a separate and hitherto unknown pigment class by Erich Becker‘ in Kiihn’s laboratory (Becker, 1939, 1941a, 1942). Hi!; work was resumed in Butenandt’s laboratory in the early fifties, and was crowned by the Becker not only initiated the study of ommochromes, but also the search for ecdysone; he took part in the identification of the ac(v+)-substance, and published papers on pteridines and other subjects. He was a most gifted insect biochemist, and his death (August 1941) in the Russian campaign was one of the losses to German science experienced in the Third Reich and Second World War. AIP-5

120

BERNT LINZEN

elucidation and chemical synthesis of xanthommatin, the first natural phenoxazinone pigment to be identified (Butenandt et al., 1954a, 1954b, 1954c, 1954d). Progress in the post-war period is reflected in the “ommochrome series” (see references) of Butenandt’s group and in a series of review papers by Butenandt (1957, 1959, 1960), Cromartie (1959), Stamm and Galarza (1961), Butenandt and Schaefer (1962), Schaefer (1964; chemistry of phenoxazinones), Linzen (1967), and Fuzeau-Braesch (1972; pigments and colour changes). In recent years, the interest in the enzymatic basis of ommochrome formation, and in the physiological function of ommochromes has grown continuously. Proper consideration will be given to these developing areas.

2 Fluorescent tryptophan metabolites found in insects 2.1 NOTES ON METHODOLOGY Since all tryptophan metabolites retaining the aromatic ring fluoresce, they can be easily detected under ultraviolet light. However, as in the case of kynurenine, fluorescence may not appear in solution but only in the dry state. In insect work the amount of material is in any case restricted and paper or thin-layer chromatography must be employed for separation. Fluorescent spots may then be measured by fluorimetry. Several commercial instruments are available for this purpose, but cheap and simple instruments-nevertheless efficient-can be easily constructed (Semm and Fried, 1952; Kiihn, 1955; Egelhaaf, 1963a; Linzen, 1971a). As the quantity of light emitted from a fluorescent spot on a chromatogram is subject to many variables (e.g. spot size, thickness and quality of paper, time after drying the chromatogram) it is advisable to run several standards on the same sheet of paper. Standard curves are not linear; the variation of the data is much greater than in photometry, but can be reduced with some care and experience. Paper chromatography and paper electrophoresis, followed by fluorimetry, often allow the determination of fluorescent metabolites in individual animals or even small portions of tissue. Tryptophan which emits only light of short wavelengths can be measured with high specificity at the nanomole level by a method first described by Egelhaaf (1957) which is based on the formation of “tryptochrome” and “iodotryptochrome” by spraying with potassium iodate (Fearon and Boggust, 1950). The method was modified by Linzen (Biickmann et al., 1966). Interference may arise from overlapping spots; if practicable this can be overcome by twodimensional separation. Suitable solvent systems are given by Hadorn and Kiihn (1953), Egelhaaf (1963a), Pinamonti et al. (1964), Linzen and Ishiguro (1966), and Wessing and Eichelberg (1968), to cite only a few.

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

121

Column chromatogaphy has been used for isolation of trace compounds from large quantities of material and for preparation of pure compounds in quantity. Benassi et al. (1964) studied chroinatography of tryptophan metabolites on Amberlite IR-120 using various buffers and achieved good separation with a pyridine-formic acid system. This has been applied to insect material (Pinamonti ct al., 1964). Chromatography on DOWEX 50, similar to the procedure of Brown and Price (1956), was employed to isolate 3-hydroxy-kynurenine at the preparative scale (Ishiguro and Linzen, 1965) and the glucosides of 3-hydroxy-kynurmine (Linzen and Ishiguro, 1966) and of 3-hydroxy-anthranilic acid (Ishiguro and Linzen, 1966). For purification of 3-hydroxy-kynurenine, particularly for separation from kynurenine, chromatography on Sephadex G-;!5 is very efficient (Hiraga, 1964; Ishiguro and Linzen, 1965). A range of colour reactions is available for the determination of individual compounds in mixtures and even raw extracts. Tryptophan is usually determined by the Spies and Chambers (1948) method; a material to which this method was not applicable tiecause of the reaction of coloured impurities with nitrite was encountered by Linzen (1971b). Kynurenine may be determined by the Brattcm-Marshall (1939) reaction, but this is not entirely specific. Tryptophan yields some colour and the degradation of ommochromes produces Bratton-Marshall positive material (Pinamonti and Petris, 1966). The reaction has been standardized to give highly reproducible results (Schartau and Linzeri, to be published). Another compound suitable for kynurenine determ (nation is Tsudas reagent (Mochizuki, 1953, cited by Koga and Osanai 1967); tryptophan and anthranilic acid will also produce some colour which, however can be removed by washing with butanol. This method was successfully applied to the analysis of silkworm eggs (Koga and Osanai, 1967). 3-Hydroxy-kynurenine may be determined by either of two photometric methods: Inagami (1954a) obtained a yellow compound with maximal absorption at 390 nm by treatment with n trow acid. Linzen (1963) oxidized 3-hydroxy-kynurenine to give xanthommatin and extracted the latter after reduction into butanol. Exactly neutral p h in the oxidation step and sufficient amounts of oxidant and reductant to cope with possible impurities are critical in this method. Both inethods have been used in various laboratories. Inagami’s procedure requires less manipulation but is not entirely specific (3-hydroxy-anthranilic itcid and “products of the tyrosine-tyrosinase reaction” will also react), and the yellow product is photosensitive. Linzeri’s method is somewhat more laborious but very specific; the red colour obtained is stable for a1 least 24 h and less likely to be interfered with by “background absorption”. Recently a fluorimetric method for 3-hydroxy-kynurenine determination was published by Watanabe et al. (1970a, 1970b); the procedure involves a

BERNT LINZEN

122

number of steps, including thin layer chromatography. Insect material has not been examined by this method as yet. 2.2

TRYPTOPHAN AND ITS FLUORESCENT METABOLITES

The scheme on page 123 depicts the metabolic conversions of the kynurenine pathway as far as they have been demonstrated in insects. 3-Hydroxy-kynurenine is a key compound in this scheme, as would be expected from the variety of reactive sites of this molecule. It may be anticipated that a number of other metabolites will be found in the course of time, namely compounds derived from the peripheral metabolites of this scheme, by further substitution. A multitude of such compoundsglucuronides, methylated, or acetylated compounds-have been discovered in mammals (for review, see Henderson et al., 1962). The primary and some of the secondary metabolites of tryptophan are rather easily demonstrated in most insects; a systematic listing of fluorescent metabolites is, therefore, not given. Instead, probable basic concentration levels will be given and examples provided in which particular compounds are of special significance. These findings are in most cases related to insect development in some way or another, yet it is hardly possible to separate the mere observation of a metabolite from its developmental aspect. The following paragraphs should therefore be viewed in conjunction with section 8, in which the developmental relations of tryptophan metabolism are described in detail. 2.2.1 Tryptophan As in other animals, both protein-bound and free tryptophan are at low concentrations. The determination of protein-bound tryptophan can be troublesome, especially if whole tissues or animals are t o be analysed (cf. Friedman and Finley, 1971). Thus, in some of the earlier studies extremely low values are reported which appear doubtful in the light of present day experience (e.g. -0.3 per cent in proteins of Saturnia p y r i and several other species: Stamm and Aguirre, 1955a, 1955b, 1 9 5 5 ~ ) .Green (1949) estimated the tryptophan content of Drosophila melanogaster protein at 0.84 mol per cent; that of D . virilis was slightly higher. Data on other Diptera reported by Jezewska (1926) and Grassmader (1968) are not comparable as they are given on a floating basis (per animal or percentage of a certain stage). Linzen and Schartau (to be published) determined protein tryptophan in Phormia and obtained values between 1.1 and 1.3 per cent. In Bombyx mori rb the average tryptophan content ranged between 0.96 per cent (in larvae with filled spinning glands) and 1.38 per cent (young pupa). In the pharate adult and in moths just after emergence 1.16 per cent were found (Linzen, 1971a). In Ephestia the total tryptophan

THE TRYPTOPHAN

-+

123

OMMOCHROME PATHWAY I N INSECTS

Glutorote pothwoy

&)lfDY dH H

I

Cysteine

t

Methionine

H2N CH

(Glucoside)

I

CH,

5:H2

-

.

COOH

$ H2 & @ H

COOH

COOH (12)

0 - SO3H The metabolism of tryptophan in insects. Thick itmows indicate well-established transformations; thin arrows indicate reactions involving other metabolic pathways or reactions of unknown significance. (1) tryptophan, (2) formyl-kynurenine, (3) kynurenine, (4a) kynurenic acid, (4bj kynurine, (5a) anthranilic acid, (5b) anthranilylglycine, (6) 3-hydroxy-kynurenine, (7a) xanthurenic acid, (7b) 4,8-dihydroxyquinoline, (7c) methyl ester of 8-hydroxy-quinaldic acid, (8a) .3-hydroxy-anthranilic acid, (8b) 3-hydroxy-anthranilyI-glycine, (9) xanthommatin, (10) cinnabarinic acid, (11 ) rhodommatin, (12) ommatin D. All 4-hydroxy derivatives of i;he quinoline ring are written in their tautomeric form. "L", "I" and "F" are unknown derivatives isolated by Inagami (1958) from silkworm pupae. The glucoside of 3-hydroxy-anthranilic acid has been detected only after injection of the parent compound into locusts.

BERNT LINZEN

124

was estimated at 61 (moths) to 105 (larvae) pg mg-’ nitrogen (Butenandt and Albrecht, 1952), protein-bound tryptophan at 1 to 1.5 per cent (w/w) (Egelhaaf, 1963a). Some representative data on free tryptophan content are combined in Table 1. In spite of the uncertainties implied with the term “concentration”, the larval tryptophan level is probably between 0.2 and 0.3 mM, while in mature insects it may fall below 0.1 mM. Some of the higher values (Ephestia, last stage; freshly emerged Diptera) are related to the accumulation of tryptophan during metamorphosis. The generally low concentration of free tryptophan is in sharp contrast to the extremely high levels of other amino acids present in the blood of many insects; it is in fact comparable to the low concentration found in human plasma (-0.05 mM). The true concentration of tryptophan in solution may be much lower than estimated above, since some tissues may be able to concentrate tryptophan. This is true of the Malpighian tubules in Drosophilu (Wessing and Bonse, 1962), TABLE 1 Representative data for concentration of free tryptophan in various insects

Species

Ephestia kiihniella

Experimental data

Eggs: -4 pg mg-’ N Larvae, 3 weeks: -2 pg mg-’ N Larvae, last stage: -18 pg mg-’ N Larvae: -40 pg g-’ fresh weight Bombyx mori rb Moths: <20 pgg-’ fresh weight Cerura uinula Larvae, last stage (haemolymph): 60 pg ml-’ Calliphora erythro- Larvae off food: 2.6 pg per animal cephala Fly, just emerged: -4 pg per animal Phormia terraeLate larvae, feeding: 65 pg g-’ nouae Flies, just emerged: -50 pg g-’ Flies, aged: -10 pg g-’ Dosophila melano. Flies, just emerged: 0.215 mg g-’ dry weight gaster

Approximate “concentration” mM kg-‘

Reference?

0.38 0.19 1.69 0.2
a

0.3

c

0.14

d

0.4 0.32 0.25 0.05 0.35

b

e

f

“Concentration” assumes equal distribution; it is calculated on the basis of average relations of fresh to dry weight, and of fresh weight to total protein. t a. Egelhaaf (1963a); b. Linzen (1971a); c. Biickmann et al. (1966);d. Langer and Grassmader (1965); e. Linzen and Schartau (to be published); f. Green (1949); similar values obtained by Shapard (1960).

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

125

but probably also in other Diptera. Bonse (1969) noticed that the white strain of Drosophila was unable to accumulate tryptophan in the Malpighian tubules, an interesting pleiotropic effect of this mutation. 2.2.2 Formyl-kynurenine Because of the high activity of kynurenine-formamidase and the spontaneous hydrolysis, which occurs even at neutral pH, this compound will probabIy not be observed in extracts. However, it is well established that formyl-kynurenine is the product of the tryptophan-oxygenase catalysed reaction. Green (1952) fed formyl-kynurenine t o vermilion mutants of Drosophila and observed formation of brown eye pigment. With this very labile compound it was not possible t o distinguish between a block in the oxygenase or the formamidase reaction for the same effect would have been obtained with kynurenine. A related compound, a-hydroxytryptophan (isolated from hydrolyzates of phalloidin), also appeared to be effective as pigment precursor in the feeding test (Butenandt et al., 1940b), although some doubt has been subsequently. cast on these results (Kikkawa, 1953). 2.2.3 Kynurenine (the “a+-”= “v+-substance”) Kynurenine was identified as an ommochrome precursor by Butenandt et al. (1940a) shortly after the discovery by Tatum (1939) that certain bacteria produced a substance from tryptophan which induced eye colour formation in the Drosophila mutant vermilion. Tatum and Haagen-Smit (1941) also identified the bacterial product with kynurenine, and Kikkawa (1941) was able t o isolate the compound from eggs of B. mori white-1. The correct structure of kynurenine was given by Butenandt et al. (1942, 1943). -Kynurenine is ubiquitous in insects, despite its extremely low concentration. In Phormia terraenovae the basic level is 12 pg g-’ (50 p M ) (Linzen and Schartau, t o be published); in EphestM (Egelhaaf, 1963a) and B o m b y x prior to spinning (Linzen, unpublished) similar levels have been determined. In the cricket, Gryllus bimaculatus, the concentration is between 5 and 30 pg per animal (50 to 150 p M ) during the last larval stage, the higher values being attained only in the last two days before the final moult (Tiedt, 1971). Extremely low values were reported for individual tissues of Schistocerca (Pinamonti et al., 1964) although the recovery might have been low too, due to the method used. Again, “concentration” is not an exactly appropriate term, as the kynurenine determined in whole animals may be confined to individual tissues. This is certainly true in Drosophilu, where the anterior region of the larval fat body accumulates kynurenine near the time of pupation, and exhibits a light blue fluorescence (Rizki, 1961). This is in accord with the

126

BEANT LINZEN

transplantation experiments of Beadle (1937) which showed the fat body to be a source of the v+-substance. Since the fluorescence of kynurenine in solution is extremely weak (Udenfriend, 1969; this corrected earlier reports of strong fluorescence), it must be in a bound form in the fat body of Drosophilu. Rizki (196 1) described the appearance of fluorescent spherical granules around the nuclei of the fat cells, which later were scattered over the cytoplasm while increasing in size. Rizki and Rizki (1963) explicitly state that there is no diffusion of kynurenine between fat cells, although removal via the haemolymph is not excluded (it may be noted that the results of the transplantation experiments can not ( b e related to this particular problem, since the fat body is subject to histolysis early in metamorphosis). In addition to the fat body, the Malpighian tubules could possibly accumulate kynurenine, although the evidence is not as convincing as in the case of tryptophan and 3-hydroxy-kynurenine storage. Kynurenine can be regularly detected by paper chromatography (Ursprung et d.,1958; Wessing and Eichelberg, 1968; Eichelberg, 1968; Bonse. 1969) and is said to be excreted “at a high rate” (Bonse). Malpighian tubules also contain the enzyme, tryptophan oxygenase (see below), and exhibit the highest known specific activity of this enzyme in Phormia larvae. Finally. kynurenine storage is also observed in eggs of Drosophilu (but missing in v/v and v/+ eggs) (Graf, 1957). Here also, brightly fluorescent inclusions. are observed which are assumed to be yolk granules (Muckenthaler, 1971). Although it was observed that the kynurenine disappeared during embryonal development, no physiological significance could be envisaged, for v/v eggs are fully viable. In Ephestiu kynurenine is absent from the eggs (Egelhaaf 1957) while in Bombyx eggs the concentration is 0.3 mM soon after completion of pigment synthesis (Koga and Osanai, 1967). Kynurenine is a major factor in the pigmentation of the wings of Pupilionid butterflies (complete reference list in Umebachi and Yoshida, 1970). From the yellow scales of these wings a large amount of kynurenine can be extracted (Umebachi and Nakamura, 1954); only a small part of this is originally in a free state. Most of it is bound in the form of a yellow pigment called “papiliochrome”. This can be resolved by paper chromatography into five components; two of these, Y-IIa and Y-IIb, have been examined in detail (Umebachi and Yoshida, 1970). From uv, ir, ord, and cd spectra it was inferred that they have the same chemical structure but are optical isomers. Kynurenine is split off even under extremely mild conditions. The other half of the molecule is a phenolic nitrogen-containing compound, derived from dopamine. The biosynthesis of these pigments occurs in discrete areas of the wing epithelium, which can be beautifully demonstrated by radioautography of wings after administration of labelled precursors (Umebachi, 1959; Umebachi and Yoshida 1970).

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

127

2.2.4 3-Hydroxy-kynurenine (the “cn+-substance”) This compound, comprising five functional groups and an aromatic nucleus, occupies the key position in tryptophan metabolism. It was first isolated from flies and simultaneously synthesized by Butenandt’s group (Butenandt, 1949; Butenandt et a f . 1949; Schlossberger, 1949), but independently obtained by Kikkawa from Bonzbyx eggs (Hirata et a f . , 1949, 1950) and Musajo (chemical synthesis; Musajo et af., 1950). Possibly, Kikkawa had a minute .quantity of the natural compound in his hands as early as 1943 (cf. Hirata et af., 1950). Today, the synthetic (racemic) compound is commercially available (synthesis: Butenandt and Hallmann, 1950; Musajo et af., 1950; Hirata and Nakanishi, 1950; Butenandt e t al., 1957f), as well as the natural L-isomer which can be isolated with good yields from Cafliphora pupae (Ishiguro and Linzen, 1965) and from the Bombyx mutant rb (Iwata and Ogata, 1966). The latter, which accumulates 3-hydroxy-kynurenine to abnormal levels (Makino et a f . , 1954), is the source of the commercial product. Although 3-hydroxy-kynurenine has been idlmtified in many species, there are relatively few data which would allow the definition of a basic concentration level in either the larval or adult stages of insects. Most quantitative determinations are concerned with alterations occurring during metamorphosis. Inagami (1958), in his study of tryptophan metabolism in the rb (red blood, jap. “aka-aka”) mutant of BGmbyx mori, has obtained many data on 3-hydroxy-kynurenine concentration in haemolymph and tissues of various strains (Table 2). Data on larval and imaginal 3-hydroxykynurenine in whole rb animals (0.06 and 0.03 p M g-’ , respectively; Linzen and Ishiguro, 1966) and the level observed in last stage Cerura Vinula larvae (-0.05 mM) agree with Inagami’s values in the order of magnitude. The high concentration in Malpighian tubules in Inagami’s TABLE 2 Concentration of 3-hydroxy-kynureNne in haemolymph and various tissues of Bombyx mori larvae* (Inagami, 1958)

Haemolymph Integument Malpighian tubules Anterior gut Posterior gut

pg ml-’

mM

30 90 230 70 40

0.13 0.4 1 0.3 0.18

* Normal strain, middle of 5th stage; all values have been rounded off by the author.

128

BERNT LINZEN

experiments is striking. It may be related to the high activity of kynurenine-3-hydroxylase in this tissue (see below), but the possibility of an accumulation mechanism cannot be eliminated at this stage. In blowflies the basic concentration varies with species. In Phormia it is about 0.15 mM both in larvae and in adult animals (Linzen and Schartau, to be published), while Calliphora larvae (off the food) contain 20 pg per animal, corresponding roughly to a 1 mM concentration; yet nearly all the substance is bound or precipitated in the Malpighian tubules (Hendrichs-Hertel and Linzen, 1969). Of the Hemimetabola, only Gryllus and Schistocerca have been examined. In G . bimaculatus whole larvae the level is 0.1 to 0.2 mM (Tiedt, 1971), while in Schistocerca the compound is barely detectable, except for the integument (Pinamonti et al., 1964) and eggs (Colombo and Pinamonti, 1965). Beadle had demonstrated (1937) that Malpighian tubules of Drosophila melanogaster are rich in cn+-substance. Since then, 3-hydroxy-kynurenine turned out to be a typical constituent of this tissue in Drosophila, Calliphora, and Bombyx (Inagami, 1958; Wessing and Danneel, 1961; Eichelberg, 1968; Hertel, 1968; Wessing and Eichelberg 1968. Eichelberg and Wessing, 1971). In Calliphora it is restricted to the Malpighian tubules, according to paper chromatographic analysis (Henning, 1957; Hertel, 1968). It is of interest to note that isolated Malpighian tubules of the stick insect, Carausius rnorosus, can be maintained active for a prolonged period of time if 3-hydroxy-kynurenine is added to the Ringer solution (Ramsay, 1956). 3-Hydroxy-kynurenine is concentrated by the Malpighian tubules of Drosophila and forms bright yellow concretions which gradually' fill out ampulla structures of the endoplasmic reticulum. Good pictures of these are published by Wessing and Danneel (1961) and Wessing and Eichelberg ( 1972). The 3-hydroxy-kynurenine concretions appear within minutes after injecting the substance into larvae. They may be viewed as temporary deposits, since in the course of adult development in the pupa-during eye pigmentation-the contents of the ampullae disappear gradually. In the white mutant of Drosophila, the Malpighian tubules have lost the capacity to accumulate (and possibly t o synthesize) 3-hydroxy-kynurenine (Bonse, 1969). While it has been suggested by Wessing that the Malpighian tubules of Drosophila are capable of both storing and synthesizing 3-hydroxykynurenine, no evidence has been advanced to support this hypothesis for any dipterans. Yet the ovary of Bombyx mori provides an example in which both preferential binding and biosynthesis from kynurenine cooperate in the establishment of a 3-hydroxy-kynurenine reserve. The ovaries show kynurenine-3-hydroxylaseactivity throughout their growth (Linzen and Hendrichs-Hertel, 1970), but at the same time they take up the

THE TRYPTOPHAN

-*

OMMOCHROME PATHWAY IN INSECTS

129

compound from the haemolymph and accumulate it in the oocytes. This process of 3-hydroxy-kynurenine accumulation is under the control of the “diapause hormone” secreted by the suboesophageal ganglion (Yamashita and Hasegawa, 1966, and earlier papers). A.s in the case of’ 3-hydroxykynurenine accumulation in eggs of Drosoohila (Muckenthaler, 197 1 ), 3-hydroxy-kynurenine is bound by the yolk spheres of the oocyte. This was beautifully demonstrated by injecting tritium-labelled 3-hydroxykynurenine into white-l pupae (where kynurenine hydroxylation is blocked) and fixing the ovarian follicles 12 h later for preparation of autoradiographs. Tritium label was confined t o the surface of the yolk spheres, but absent from the rest of the oocyte cytoplasm, from the follicle, and nurse cells (Sonobe and Ohnishi, 1970). While the function of accumulated kynurenine in Drosophila e!ggs is obscure, 3-hydroxykynurenine in diapause eggs of Bombyx is consumed during synthesis of ommochromes in the “serosa” (Kikkawa, 1953; Koga and Goda, 1962; Koga and Osanai, 1967). It is said that the viability of pigmented diapause eggs is increased in comparison to unpigmented eggs. With respect to eye pigmentation in Droso,bhzla, it is by no means clear whether the deposits of 3-hydroxy-kynureniiie in the Malpighian tubules are an indispensable requirement or merely a supplementary supply of pigment precursor. Eye pigmentation is essentially autonomous (Danneel, 1941; Horikawa, 1958). It is of special interesf, therefore, that in the honey bee apparently a major proportion of tryptophan transformation occurs in the developing eyes. This is deduced from the fact that eyeless mutants accumulate tryptophan, rather than its metabolites, which is later excreted by the adult bee. There are a number of Apis mutants (cf. Dustmann, 1969, for a listing) which are blocked at diffcrent steps along the tryptophan --* ommochrome pathway. A mutant type which is not known from other laboratory insects is chartreuse. The group o f chartreuse mutants i s easily distinguished by a strong green fluorescence of the eyes, if viewed under ultraviolet light. In most chartreuse mutants, also the general appearance of the eyes is greenish after emergence, but may turn red or brown after some time. These mutants accumulate 3-hydroxy-kynurenine in a granular form in the pigment cells of the compound eyes (maximally 140 pg per animal), so that the pigment precursor actually assumes pigment function. Binding 3-hydroxy-kynurenine in this way is evidently a distinctive step in ommochrome formation; it is blocked in another series of mutants, viz. cream (cr), pearl (p), and brick (bk, partially blocked) which accordingly accumulate 3-hydroxy-kynurenine, but e vcrete it upon emergence (Dustmann, 1968, 1969, and personal commu~iication). The function of a pigment is performed by 3-hydmxy-kynurenine also in butterflies of the subfamilies Heliconiime and Zthomiime. At first the yellow compound isolated from the wings and bodies of these species was

130

BERNT LINZEN

considered to have the structure of 2.3-dihydro-xanthurenic acid, which is analogous to “kynurenine yellow” described by Kotake and Shichiri (1 93 1). A reexamination, supplemented by mass spectral and nmr data, later revealed the identity of the yellow compound with 3-hydroxykynurenine (Brown, 1965; Brown and Becher, 1967; Tokuyama et al., 1967). In this connection, it may be mentioned that in the Papilionids which make use of kynurenine in producing a light yellow pigment, the level of 3-hydroxy-kynurenine remains low at all stages (Umebachi and Katayama, 1966). As a rule, binding of a compound produces a shift of the absorption maximum which is related to binding energy; in the case of ommochrome binding this shift may amount to 60 nm. It is desirable, therefore, to examine the spectra of 3-hydroxy-kynurenine in the different states in which it is present in insects; that is, dissolved in the presence of proteins, in the state of amorphous concretions, in the adsorbed condition in wing scales and yolk spheres and in the granular form of chartreuse bees. Probably, some conclusions may be drawn From such measurements concerning the availability of 3-hydroxy-kynurenine at a given stage of development and concerning the specificity-and thus physiological significance-of 3-hydroxy-kynureninebinding. 3-Hydroxy-kynurenine may be formed in excess of the demand and, being a phenolic compound, is subject to known detoxication mechanisms. Thus, the sulphuric acid ester (Inagami, 1958) and the P-glucoside (Linzen and Ishiguro, 1966) are formed in Bombyx mori rb pupae; the sulphate also occurs in the bug, Rhodnius prolixus (Viscontini and Schmid 1963). Inagami also mentioned a “glucuronide” of 3-hydroxy-kynurenine,but it is surmised that the glucoside (which has been identified more rigorously) was mistaken for the glucuronide. Pryor (1955), in a short communication, has suggested the participation of 3-hydroxy-kynurenine in the cuticular tanning of insects. Since then, to the author’s knowledge, no experiments have appeared to support this conjecture. Neither is there convincing evidence of formation of “mixed melanins” from 3-hydroxy-kynurenine and tyrosine metabolites, which had been proposed by Inagami (1954b). 2.2.5 Quinoline derivatives Kynurenine and 3-hydroxy-kynurenine may be transaminated, the resulting keto acids spontaneously undergo cyclization to form kynurenic acid. and xanthurenic acid respectively. Kynurenic and xanthurenic acids have been identified in a variety of insects: Schistocerca gregaria (Pinamonti et al., 1964), Ephestia kiihniella (xanthurenic acid only after tryptophan injection, kynurenic acid excluded: Egelhaaf, 1963a), Plodia interpunctella (xanthurenic acid in wings; Mohlmann, 1958), Bombyx mori (both acids:

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY I N INSECTS

131

Inagami, 1958), Habrobracon juglandis (both acids: Leibenguth, 1965, 1967a), Drosophila melanogaster (kynurenic acid: Danneel and Zimmermann, 1954; Wessing and Eichelberg, 1968; xanthurenic acid: Umebachi and Tsuchitani, 1955; Wessing and Eichelberg, 1968), Musca domestica (both: Colombo and Pinamonti, 1967; Laudani and Grigolo, 1968). In most species only minute amounts of both compounds are found, and usually kynurenic acid is even less abundant than xanthurenic acid. Xanthurenic acid is the major end product of tryptophan metabolism in the spinning larva of the parasitic wasp, Habrobracon juglandis: it is excreted by the prepupa (Leibenguth, 1965). This explains a paradoxical observation first made by Beadle et al. (1938): Larvae of the Habrobracon mutant o which are blocked in the hydroxylation of kynurenine and therefore devoid of ommochrome pigments should ingest a sufficient supply of 3-hydroxy-kynurenine and become pigmented if feeding on wild type Ephestia. This is, however, not the case, even if the level of 3-hydroxy-kynurenine in the host is elevated by injection. Under these circumstances, the Habrobracon larva will even excrete some 3-hydroxykynurenine. Nevertheless in the prepupa not ,itrace of the substance can be detected. Leibenguth (1967, 1970) explained this convincingly by the fact that all larval 3-hydroxy-kynurenine is removed via transamination. Eye pigmentation in Habrobracon thus depends on pupal production of 3-hydroxy-kynurenine, which of course is not any more influenced by the supply of host material. Three compounds related to kynurenic and xanthurenic acids, respectively, have been isolated from insects. One is kynurin (4-hydroxyquinoline), 605 mg of which were obtained l’rom 200 kg of dried silkworm pupae (Butenandt et al., 1951). Formally, kynurine could originate by decarboxylation of kynurenic acid, yet no experiments have been performed to check this. Correspondingly, 4,8-dihydroxy-quinoline was isolated by Inagami (1958) from silkworm pupae, where it appeared late in the pupal stage and more abundantly in males than in females. Finally the methyl ester of 8-hydroxy-quinaldic acid was identified by Schildknecht et al. (1969) in Ilybius fenestratus. This is a most intriguing finding, since it implies a dehydroxylation reaction, which is rather uncommon in nature. The compound is elaborated by defence glands, the opening of which are located in the prothorax of this water beetle and produces convulsions if the beetle is ingested by a frog or a mouse. 2.2.6 Anthranilic and 3-hydroxy-anthranilic acids The specificity of the enzyme, kynureninaje is low with respect to the benzoyl half of its substrate (Weber and Wiss, 1966), for even 5-hydroxykynurenine (Butenandt et al., 1953) and the ommatins (Butenandt et al., 1954b) will be attacked. One should, therefore, expect that whenever this

132

BERNT LINZEN

enzyme is present in insects, both anthranilic and 3-hydroxy-anthranilic acids are formed. This is true in Bombyx mori, where both acids and their glycine conjugates were indentified (Kikkawa, 1951, 1953). Inagami (1958) found another conjugate in white-1 meconia, which he considered to be a glucuronide (today one would rather expect a glucoside). Inagami concluded that the anthranilic acids and their conjugates were the main end products of tryptophan metabolism during larval development of the silkworm, while in the pupa the formation of 3-hydroxy-kynurenine and of the pigments was favoured. Inagami also described three unknown fluorescent compounds (“F”, “L”, and “I”) which, according to their presence or absence in the normal strain and the white-1 and rb-mutants, must be .considered to be metabolites of 3-hydroxy-anthranilic acid. If injected into locusts, 3-hydroxy-anthranilic acid is rapidly converted into its fl-glucoside (Ishiguro and Linzen, 1966). In flies, the anthranilic acids occur at best in trace amounts. Wessing and Eichelberg, in their list of fluorescent substances in the Malpighian tubules of Drosophila (1968), do not mention either compound. While Laudani and Grigolo (1968) did not detect either compound in Musca dornestica strains, Colombo and Pinamonti (1967) reported “trace amounts”. By injecting trypt~phan-methylene-’~C and searching for labelled alanine after variable lengths of time and in each developmental stage, Linzen and Schartau (to be published) proved the virtual absence of kynureninase activity in the blowfly, Phormia tmaenovae. Egelhaaf (1963a) drew attention to a pair of fluorescent compounds, “r” and “s”, which appeared on two-dimensional paper chromatograms of Ephestia extracts, but were absent in extracts of the a mutant. After administration of tryptophan, labelled in the methylene carbon, these compounds were not labelled. This experiment should be repeated using tryptophan labelled in the benzene nucleus to clarify the biogenetic relation. In passing, it should be mentioned that Calam (1972) isolated indole acetic acid as a major constituent from the salivary glands of Dasyneura species. This compound is probably responsible for the induction of galls in various plants by these insects. 3 The absence of the glutarate pathway

In mammals, the major part of administered tryptophan is metabolized via 3-hydroxy-anthranilic acid, its unstable oxidation product cr-aminofl-carboxy-muconic-e-semialdehyde, and glutaryl-CoA (Henderson et al., 1962). Oxidation of the latter gives rise to CO, and acetate. The open chain intermediate provides a branching point, since reaction between the amino group and the aldehyde group leads to ring closure and formation of

THE TRYPTOPHAN

+

133

OMMOCHROME PATHWAY IN INSECTS

pyridine carboxylic acids, notably nicotinic acld. Thus, the oxidation of 3-hydroxy-anthranilicacid has two important functions: to form a building block for the most important cofactors in oxidative metabolism and to provide a means of breaking down the benzene nucleus of tryptophan and to remove potentially detrimental intermediates. The function of the glutarate pathway may easily be tested by injecting tryptophan labelled with l4 C in the aromatic nucleus and testing for expired * 4 CO, That this pathway is not operative in inslxts was indicated by the accumulation of metabolites prior to the branching point and by the cognition that all insects so far examined require nicotinic acid in their diet (Dadd, 1970). Schultz and Rudkin (1948) have specifically tested for a possible sparing action of tryptophan added to the growth medium of Drosophila larvae and found that there was none. Accordingly, no difference in the nicotinic acid content of normal, white-1, and white-2 strains of Bombyx (150-180 gg g-' dried material) was detected, and injections of tryptophan, 3-hydroxy-kynurenine, or 3-hydroxy-anthranilic acid failed to raise this level (Kikkawa and Kuwma, 1952). The clue to these observations is provided in a most important paper by Lan and Gholson (1965). The authors injected 1),~-tryptophan-5-'~C into a variety of invertebrates, and poikilothermic and homoiothermic vertebrates, and collected the %O, produced during the subsequent 4.5 to 24 h. While all vertebrates produced radioactive CO,, the three insects (a grasshopper, Dissosteira longipennis, a cockroach, Periplaneta americana, and a cricket, Gryllus assimilis) did not. Assays for the three enzymes, 3-hydroxy-anthranilic acid oxidase, picolinic carboxylase, and a-aminomuconic semialdehyde dehydrogenase, yielded negative results in each case. Although the latter tests which were cosducted on whole organisms might not be fully conclusive by themselves (the authors report some contradictory results with snails and earthworms), it is clear that none of the arthropods tested are able to degrade the benzene nucleus of tryptophan. In a later study which included also three hcJometabolous insects (&is mellifera, Tenebrio molitor, Phormia terraenovae) Schartau and Linzen (1969) furnished supplementary isotopic data. With the usual proviso one might state that in insects the degradation of tryptophan is blocked at the step of 3-hydroxy-anthranilic oxidation. This is the key to understanding the functional significance of the accumulation of tryptophan metabolites and of the formation of ommochromes. Without this block, ommochromes would probably 'lot be as characteristic of insects as they are in fact. It is difficult to evaluate advantages and disadvantages of this metabolic block. In the author's opinion this block poses problems (e.g. the absolute need for niootinic acid, or the accumulation of highly reactive compounds) rather than meeting requirements evolved with the arthropod structure. In contrast, Brunet (1965)

.

BERNT LINZEN

134

speculates that ommochrome formation has been of such a high selective value for insects that loss of the ability to oxidize 3-hydroxy-anthranilic acid was nothing less than a logical consequence. This problem will be alluded to in later parts of this review.

4 Ornrnochrornes 4.1

NOTES ON NOMENCLATURE

Erich Becker, in his first papers on ommochromes (1939, 1941a), had distinguished between two groups of pigments: the ommatins (of rather low molecular weight) and the ommins (of a higher degree of polymerization). His major criteria to distinguish between the two were the degree of stability in alkaline media (the ommatins being extremely labile) and the ability to pass dialysis membranes. Only in his posthumous paper (1942) did Becker propose the term “ommochromes” to cover all related pigments. He was aware of the fact that some of these had been given different names which might have deserved priority, such as Chauvin’s (1938) “acridioxanthin” and “acridioerythrin” of the migratory locust. However, it may be argued-as Becker contended-that many of the pigment preparations thus denoted were not satisfactorily defined chemically, and that closer examination could justify a revision of nomenclature. For similar reasons the term “insectorubin” (Goodwin and Srisukh, 1950) will not be used in this paper, since it designates a preparation which by modern methods can be shown to be composed of different pigments, and since it suggests a widespread distribution of the locust ommochromes, which in fact is not the case. Linzen (1967) proposed a subdivision of the ommochromes based on structural criteria. Those ommochromes containing a 1,P-pyridino-3Hphenoxazine moiety were designated t o be ommatins, and those which yielded 3-hydroxy-kynurenine plus “pigment IV” upon acid hydrolysis, were called ommins. For the highly acid- and alkali-resistant eye-pigments of the migratory locust, new names, ommidin and cryptommidin, were introduced (Linzen, 1966) which replaced “acridioxanthin” and “insectorubin”. It may be left open to further discussion whether other pigments containing only a simple 3H-phenoxazine nucleus (such as cinnabarinic acid) should be included in the ommatin group, which then would not be defined by its substituent but rather by its basic structure, or should be grouped separately. Typical ommatins are xanthommatin (and its reduced form, dihydro- or “hydro-”xanthommatin), rhodommatin (dihydro-xanthommatin-0-fl-D glucoside), and ommatin D (dihydro-xanthommatin sulphate); the latter two are excreted by nymphalid butterflies in their meconia. Ommatin C is

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

135

an artifact obtained by Butenandt et al. (1954a). Bouthier (personal communication) has observed recently that one of the two “labile redox pigments” described by Linzen (1966) in locust integument actually decomposes into xanthommatin and an unknown compound. Bouthier proposed to name these two pigments “acridi ommatins” (acridiommatin I and 11). Needham (1970) has described an “oniscoid ommatin” from isopods which awaits further chemical study. Becker’s (1942)“phaeommatin” from Calliphora eyes is undoubtedly xanthommatin. Becker’s best studied ommin, “skotommin” of Ephestia kuhniella, is a mixture of closely related pigments; the name is no longer in use. Butenandt et al. (1959)used letters (in the order of motility in a particular paper chromatographic system) to distinguish between these components; ommin A is the most abundant in most cases. Linzen (1967)suggested an index determined by the wavelength of maximal absorption of visible light in Naz HPO, solution (e.g. ‘‘ommin537”). Kiihn’s “ommochrome I ” and “ommochrome 11” (Kiihn and Egelhaaf, 1959) are ommins in all probability, as are Kawase’s (1955) “+-chrome 11” and “+-chrome 111” (“+-chrome I” is identical to xanthommatin). Lacciferic acid and laccaic acid which are obtained from stick lac, behave in a similar way to ommins (Singh et al., 1966). On the other hand, Becker (1939)had described a yellow water-soluble pigment as “xanthommin”. Although the latter has never been reinvestigated, it is doubtful whether it really belongs to the ommin group. Table 3 lists the ommochromes and some of the features characteristic of each subgroup.

4.2 ISOLATION As ommochromes are generally of low solubility, it is advantageous to extract other low molecular weight compounds first, usually by repeated treatment with ether (or benzene, acetone etc.) and with methyl alcohol. The pigments themselves will then be extracted by methyl alcohol containing 1-5 per cent of concentrated HCI and preferably gassed with SOz (in particular if xanthommatin is to be extracted), or by 0.1 M NazHPO, (if rhodommatin and ommatin D are involved). The best composition of the solvent for extraction is determined by the properties of the starting material and should be checked first. For example, for extraction of rhodommatin and ommatin D from butterfly wings it was found to be advantageous to employ dilute ammonia. Purification may proceed, via repeated precipjtation, from slightly alkaline buffers, via preparative paper or column chromatography. The latter is especially efficient for separation of rhodommatin and ommatin D which always occur in association. On the analytical or micro-preparative

TABLE 3 A classification of ommochromes in the order of (presumably) increasing complexity Compound 1 Ommatins Cinnabarinic acid

Characteristics

Precursors

Distribution in insects

In haemolymph and excreta of Bombyx mori, normal strain (also a pigment in mushrooms) Ubiquitous, mixed with other ommochromes: in Diptera only eye pigment Excretory and wing pigments in Lepidoptera

Decoburized by reduction; autoxidation a t alkaline pH; alkali-labile

3-Hydroxy-anthranilic acid

Xanthommatin

Yellow in oxidized, red in reduced state, autoxidizable, very labile

3-Hydroxy-kynurenine

Rhodommatin, Ommatin D

Red, stable against oxidation in air; alkali-labile

Acridiommatin I Acridiommatin I1

Red, readily oxidized, extremely labile; acridiommatin I1 decomposed to xanthommatin (?)

3-Hydroxy-kynurenine , xanthommatin? Glucose and sulphate respectively Many tissues and excreta of Orthoptera; Tryptophan ria 3-hydroxy-kynurenine Odonata (integument) and xanthommatin?

Yellow; red in acid; very stable against hydrolysis; sulphur-containing; readily dialysed

3-Hydroxy-kynurenine, methionin (proved for ommidin only)

Eye pigments in Orthoptera, occasionally in integument

No autoxidation in alkaline media; violet to purple, yellow after oxidation by nitrite: relatively stable in alkaline media; slowly dialysed; acid hydrolysis yields 3-hydroxy-kynurenine and “pigment IV”

3-Hydroxy-kynurenine, mrthionine via cysteine

Nearly ubiquitous in insect eyes, except for some Diptera and Orthoptera; frequently in hypodermis and around internal organs; not in excreta

2 Ommidins Ommidin Cryp tommidin 3. Ommins Ommin A (skotommin, ommin52,) and at least four related compounds

-id

Data on spectral properties are listed in Table 4; details of distribution among.insects are given in Table 5. The listing of cinnabarinic md of +he Urridiommathm in the ommatin group,u regnrded as proviaional and subject to further diacussion.

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

137

scale, Ecteola is most convenient, since the ommatins are adsorbed in a very narrow zone even from extremely dilute buffer (:pH 7 ) solutions.' Elution is achieved by pyridine-acetate buffers (Butenandt et al., 1960b). Reduced xanthommatin may be isolated by the same method but yields of only about 65 per cent are obtained. On the preparative scale, circular paper chromatography was employed in the earlier work of the Butenandt group (Butenandt et al., 1954a). Later, polyamide powder proved to be a suitable adsorbent for column chromatography (Kiibler, 1960), and conditions have been worked out which allow the isolation of rhodommatin and ommatin D in milligram to gram amounts (Butenandt et al., 1960c; Traub 1962). Xanthommatin is normally a companion of ommins. It may be separated by taking advantage of its solubility-in the oxidized state-in dilute acids (Butenandt et al.. 1954a; Dustmann, 1964), 01- by chromatography on SE-Sephadex (Osanai and Koga, 1966). Wend (1969) has devised a chromatographic method for xanthommatin determination which results in rather broad peaks, but which might be expected to retain ommins. The ommins are still the most refractory group in every respect. Their separation is best achieved by circular paper chromatography. Two solvent systems are available: collidin-water ( 3 : 1) (Butenandt et al., 1959) or formic acid-methyl alcohol-water-concentrated. HCl (68 : 1 5 : 12 : 2) (Linzen, 1968). The latter has been developed lrom similar systems used earlier in Butenandt's laboratory. Success of separation depends on the amounts of pigment detected and on the type and amount of impurities present. Again there are variables in this system which must first be determined. Butenandt et al. (1967) have also succeeded in purifying ommin A by treatment with phenol-water, followed by chromatography on polyamide. In the latter step, rapid elution is critical, as prolonged contact of the pigments with the column material leads 1.0 irreversible adsorption. Apparently the separation of the various ommiris is not very sharp. Even the most painstakingly purified ommin preparations usually contain small amounts of protein (Butenandt and Neubert, 1958; Butenandt et al., 1967) which may amount to 1 or 2 per cent. In the cour:je of biosynthetic studies, Linzen (1970) has examined a variety of methods (treatment with TCA, phenol, proteinases) to remove this residual contamination but has not been successful. Finally, the ommidins may be obtained in a pure state by a small number of steps, comprising circular paper chromatography and ion-exchange chromatography (Linzen, 1966; Bouthier, 1969). A few ommochromes have been obtained in crystalline state: dihydroxanthommatin (Butenandt et al., 1954a), rhodommatin (as a pyridine complex: Butenandt et al., 1963), ommidin (Bouthief, 1969), cryptomI

The properties of Ecteola vary and each batch must first be checked.

1 38

BERNT LINZEN

midin, and acridiommatin I (Bouthier, 1972). In all these cases, clusters of needle-shaped crystals are obtained. Becker (1942) also described crystallized ommatins, but according to Butenandt et al. (1954a) his preparations were of low purity. For details the original literature should be consulted; many helpful suggestions will be found, together with descriptions of unsuccessful approaches, in the dissertations of Beckmann (1956), Neubert (1956), Linzen (1957), Baumann (1957), Traub (1962), and Kiibler (1960). Furthermore, the series o f ommochrome papers by the Butenandt school will provide a broader background on phenoxazinones.

4.3

PROPERTIES

4.3.1 Solubility, aggregation and adsorption Ommochromes have, generally speaking, many adverse properties: limited solubility, instability in both alkaline and acid media, together with a strong tendency towards aggregation and adsorption. These fa.ctors contributed to their late discovery, in spite of their wide distribution, as well as to the slow progress of ommochrome chemistry and biochemistry. In their red, reduced state ommochromes are insoluble in all common neutral organic solvents, water (sometimes colloidal solutions are obtained), and dilute acids. To varying degrees they are soluble in alkaline and strongly acid solutions, in buffers around and above neutrality, in formamide, phenol-water, ethylene glycol, glycerol, methyl cellosolve, acetic anhydride and in solutions of urea. Exact quantitative data on solubility are not given in any case. Estimates for the solubility of ommidin in formic acid and trifluoroacetic acid are from 2 to 5 per cent, in dimethyl sulphoxide, 2 per cent, and in sodium hydroxide, more than 10 per cent Whether a pigment preparation will dissolve or not is highly dependent on its degree of dryness. Preparations which have been rigorously dried and stored for long periods require drastic treatment for dissolution. Ommochromes may decompose relatively rapidly in many of the solvents listed above. Again a systematic study has not been undertaken, but the splitting of xanthommatin in buffer solution of pH 8 (Butenandt et al., 1960b), and of ommatin D in 5 N HCl (Butenandt et al., 1960c) may serve as examples. The insolubility of ommochromes represents one of their important physiological properties (see below); solubilization of xanthommatin by formation of the P-glucoside (rhodommatin) and the sulphate (ommatin D) is of significance in Lepidoptera. A phenomenon no less annoying than the low solubility of ommochromes is their tendency for strong aggregation. This is evident in

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

139

the formation of colloidal solutions and the migration of ommins as “mixed” bands in paper chromatography. Elution of pigment bands and rechromatography yields a main band plus at leist the two adjoining ones. The degree of association of xanthommatiri and of three catalins (xanthommatin-like phenoxazinones employed in ophthalmology) was measured by Nakagaki et al. (1962). From the diffusion constant it was calculated that xanthommatin formed aggregates (micelles) of 2 to 5 molecules (in neutral buffer) depending on temperature (18 to 37” C). It may be noted that the size of the micelles is independent of concentration once a critical concentration has been passed, but depends on the properties of the solvent (e.g. ion concentration) (Scholtan, 1955). Micelle formation is a common phenomenon in the physical chemistry of pigments; it is typical of compounds with hydrophobic ini.eraction and a fairly rigid structure (Worz and Scheibe, 1969, and Brooker et al., cited by them). Both these conditions pertain in ommochromes. Certainly they are not only of theoretical significance, but also of physiological importance (e.g. in relation to ommochromes being excrei ory products deposited permanently in a living cell). The strength of adsorption is related to the tendency to aggregate (Worz and Scheibe, 1969). It is not surprising, therefore, that ommochromes are strongly adsorbed to all kinds of material, the ommins in particular. An outstanding example is the adsorption of ommins to ordinary laboratory glass. This can be shown when a raw extract of ommins in methyl alcohol-HC1 is brought to complete dryness in a rotary evaporator, cooled, and ice-cold buffer (pH 7.5-9) carefully poured over the residue. A light brown, fluffy proteinaceous material is then brought into suspension, while the dark violet ommin mixture remains stuck tci the glass wall. In ommin isolation this step has been routinely employed. 4.3.2 Chromatography The chromatography of ommochromes is hampered by adsorption. For elution, either strong acids, or solvents based on pyridine or its methyl derivatives have t o be employed. A few of these have been mentioned above. Methods for column chromatography are described by Butenandt et al. (1960b, 1960c), Linzen (1966), Butenandt el al. (1967), Wenzl (1969), Bouthier (1969), Osanai and Koga (1966), Umebachi and Uchida (1970); for convenience the list comprises both primary and secondary sources. For paper chromatographic identification extremely powerful eluants must also be employed. The most useful systems for ommatins (Butenandt et al., 1954a) are: (1) collidine-lutidine-water (1 : 1 : 2; upper phase) which gives a neat separation of rhodommatin, ommatin D, and ommidin, while xanthommatin is decomposed; (2) collidine-0.5 M KH2P04 (2 : 3; upper phase) by which the substituted ommatins are again fairly well separated,

140

BERNT LINZEN

while xanthommatin is found at the salt front (this is best seen under ultraviolet light; its position depends on the temperature and age of the system) and ommidin is found at about R F 0.1. There is no migration of ommins in either system, although migration of some ill-defined cherry-red materid is observed. Ommins and ommidins are, in the author’s experience, best separated by circular paper chromatography, using formic acid-methyl alcohol-waterconcentrated HCl in the ratio of 68 : 15 : 12 : 2 (50 : 15 : 20 : 1 was also employed for ommidins) (Linzen, 1966, 1967). A further system is collidin-water ( 3 : 1) (Butenandt et al., 1959) in which ommin A-the slowest moving in all systems-is not removed completely from the starting line (but is in the formic acid systems). Kuhn and Egelhaaf (1959) obtained good separation of the Ephestia ommins by ascending paper chromatography with a formic acid system. Even well isolated ommin bands must be subjected to rechromatography if they are to be considered pure. The only two pigments which can be chromatographed under milder conditions are the “labile ommochromes” (acridiommatins) of locust integument. Butanol-acetic acid-water (4 : 1 : 1) and propanol-formic acid-water (1 : 1 : 4) are suitable solvent systems for these pigments. A number of solvent systems for thin-layer chromatography have been reported by Ajami and Riddiford (1971a). However, only one or two spots are obtained from the ommin fraction, instead of the three to six expected according to experience with other material.

4.3.3 Redox properties; spectra The most conspicuous property of ommochromes is the change from red to yellow colour observed upon oxidation. This is very uncommon, as the oxidized states of pigments are usually more deeply coloured while a lightening is brought about by reduction. This unusual behaviour of ommochromes is very useful for analytical purposes, enabling them to be easily observed in histological sections or on paper chromatograms; oxidation and reduction can be repeated many times. Spontaneous oxidation of dihydro-xanthommatin occurs if it is left in contact with air at pH values around or above neutrality, and also if a diluted solution of ommins in acid methyl alcohol is left to stand for a prolonged time. In contrast, rhodommatin and ommatin D are stable against autoxidation, as the reduced state is protected by the substituents. If, on the other hand, neutral solutions of these two pigments are treated with ferricyanide they turn yellow immediately. Dihydro-xanthommatin is liberated by this reaction, but, surprisingly, after reduction of the solution sizable quantities of rhodommatin and ommatin D can be recovered. These observations have been interpreted in terms of the primary formation

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY I N INSECTS

141

of a phenoxazonium cation which is stabilized by mesomerization and hydrolysed in a secondary, much slower reaction (Fig. 1 ; Traub, 1962). The oxidation-reduction potential of an omvnin preparation from ganglia and optic lobes of Cecropia silkworms was estimated to lie between 123 and 217 mV (Ajami and Riddiford, 1971a). YOOH H2N-C-H

1

- 2 e , -H+

+

COOH H,N - -H

FOOH HZN-C-H

COOH

NH

0+ti*+

Glucose

Fig. 1. Oxidative splitting of rhodommatin. The intbermediary cation is stabilized by mesomerization and hydrolysed in a secondary, s b w reaction. (Proposed by Traub, 1962.)

The explanation for the peculiar spectral behaviour of reduced ommochromes has recently been given by Schafer and Geyer (1972). By comparing the spectra of numerous and. carefully selected model compounds they concluded that upon reduction a new extended resonance system is formed (Fig. 2). This system is characterized by the polar structures involving the phenoxazine nitrogen and the carbonyl groups in its neighbourhood (the 4-hydroxy-pyridine ring being in its tautomeric

BERNT LINZEN

142 COOH

Fig. 2. Structures of xanthommatin and dihydro-xanthommatin. Upon reduction a new extended resonance system is formed which is responsible for the deep red colour obtained. The location of the “central” hydrogen is not known precisely. (From Schafer and Geyer, 1972.)

pyridone form) and by the complete ring system lying in a plane. It is conceded that the hydrogen attached t o the phenoxazine nitrogen might form a bridge to either carbonyl oxygen and thus contribute to the shift of the absorption band, but it is argued that this contribution is relatively small. In fact, Pfleiderer as well as Schafer (both personal communications) discuss the possibility that the hydrogen atom is not attached primarily to the phenoxazine nitrogen but to the carbonyl oxygen of the pyridone ring. Spectral data of ommochromes are presented in Table 4 and a few spectra are shown in Figs 3-8. In the publications from Butenandt’s laboratory frequently an a-value is used as a measure of specific absorbance; this is equal to absorbance of a 0.1 per cent solution in a 1 cm cell. Phenoxazinones usually exhibit four absorption bands in the uv and visible range. These are designated A, By C, and D in the studies of Butenandt et al. (1954d, 1960a) and Schafer and Geyer (1972). Under certain conditions C and D merge, as in the case of xanthommatin in neutral solution. If ommochromes are dissolved in concentrated sulphuric acid, a violet halochromic effect is observed. Generally, the spectra of the ommins are rather poorly defined, and small shifts in wavelength of the maxima are often observed in apparently identical pigments. Linzen

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

143

TABLE 4 Spectral data of ommochromes* Compound Xanthommatin

Dihydro-xanthommatin

E

Amax

0.1%

235 440 243 375 475

31 000 13200 29 400 7 350 11 730

380 495 -

Rhodommatin

Omrnatin D

Acridiommatin I

Acridiommatin I1

308 445 492 210 377 49 7

-

4680? 7 200

8 500 4470 6400** 34 900 3 280 7 100

log E

Solvent

4.49 4.12 4.47 3.87 4.07

pH 7, 0.067 M phosphate 5 N HC1

a

pH 7, 0.067 M phosphate ascorbate added 5 N HC1

C

-

3.67? 3.86 -

3.93 3.65 3.81 4.54 3.52 3.85

-

-

-

(265) 306 43 7

(15 700) 1 4 300 7 200

(4.19) 4.16 3.86

-

-

-

(285) 370 490 215 305 440 230 360 443 234 3 70 47 2 424 484 230 368 455 448 360 465

(15 000) 3 160 7 500 33 300 13 000 6 620

(4.18) 3.50 3.87 4.52 4.11 3.82

Referencet

a,

b

C

d

butanol/HCl pH 7.4, 0.067 M phosphate

a

5 N HCI

a

pH 7.4, 0.067 M phosphate

a, e

5 N HCI

a

0.1 M.Na2HPO4

1

5 N HCI

I

HCOOH

1

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

0.1 M NazHP04

1

-

-

-

5NHCI HCOOA

1

-

I

144

BERNT LINZEN TABLE 4-continued E

Compound

xmax

Ommidin

Cryptommidin

Ommin mixture* (Crangon vulg.) “Pigment IV”

Cinnabarinic acid

,

0.1%

log E pH

230 320 430 403 240 -43 3 48 5 (510) 426 224 438 460 228 408 520 530 245 375 49 7 315 435 440 45 5

Solvent

.,

Referencet

9.5

N NaOH

5 N HCI

f f

0.1 M NazHPOe 5 N HC1

1 N NaOH

8.0 -

-

-

4.35 3.4 3.76 4.12 3.72 4.35 4.33

pH 7.5, 0.067 M phosphate 5 N HCI pH 7.1

i

s h

5 N HCI pyridine

k

Data in brackets refer to inflexions; if an absorption band is expected but has not been recorded, a dash has been set. * See text for further details. ** A value about half as high had been obtained in earlier studies (Butenandt et al., 1960b); correspondingly, all measurements calculated with the old absorbance value must be doubled. t a . Butenandt et al. (1960b); b. Butenandt et al. (1954d);c. Butenandt et al. (1963); ; Linzen (1966);g. Butenandt et aL d. Dustmann (1969); e. Butenandt et al. ( 1 9 6 0 ~ )f. (1958a); h. Butenandt et al. (1967);i. Bouthier (1972);k. Gripenberg et al. (1957).

(unpublished) has measured absorption spectra of several ommins isolated by circular paper chromatography and obtained the following wavelengths of the maxima: 527 nm (ommin A of A p i s ) , 536 nm (ommin B), 516-517 nm (ommin D); 530 nm (ommin A of Ephestiu), 537 nm (ommin B); 525 nm (ommin A of Gryllus birnaculutus), 534 nm (ommin B), 512 nm (ommin D). Ommin A and B are apparently the best defined members of the group, at present.

THE TRYPTOPHAN

-*

OMMOCHROME PATHWAY IN INSECTS

145

nrn

Fig. 3. Ultraviolet-visible spectra of xanthommatin in buffer solution of pH 7.0 to 7.3 -') and 5 N HCI (- - -).

k3.4 Degradation reactions 3mmochromes are, as a rule, unstable in both alkaline and acid media. This is especially true of the ommatins which bleach rapidly. In 0.5 N sodium hydroxide at 90°C, for example, destruction is complete within 2 h, while in bicarbonate solution at room temperature two days are required. From the reaction mixture a number of fluorescent or ninhydrin-positive compounds may be isolated, of which xanthurenic acid (Butenandt et al., 1954b) and 2-amino-3-hydroxy-acetophenone(Butenandt et al., 1957e) have been identified. Ommins yield the same products, albeit under somewhat more drastic conditions (Butenandt ct al., 1958a). By acidifying md extracting into butanol the two compounlds may be readily prepared For paper chromatography.

BERNT LINZEN

146

30 OOC

20 000 u,

10 000

I

I

I

I

I

I

I

I

I

I

I

L

I

I

I

I

I

nrn

Fig. 4. Ultraviolet-visible spectra of rhodommatin in buffer solution of pH 7.0 to 7.3 +) and 5 N HCl (- - -).

These two compounds are not primary degradation products but are formed from the reaction intermediate, 3-hydroxy-kynurenine. 3Hydroxy-kynurenine itself can be isolated only if the decomposition of xanthommatin is performed under very mild conditions (pH 8 at 37” C for 4 to 5 h). Under these circumstances also the “right-hand” part of the xanthommatin molecule is obtained (Butenandt et al., 1960b). 3-Hydroxy-kynurenine will also be split off if ommochromes are heated in a 1 : 1 mixture of formic acid and 6 N HCI for 10 to 24 h (Butenandt et al., 1958a). In the case of the ommins a brick-red pigment is also formed which has been isolated and well characterized (Butenandt and Neubert, 1958; Butenandt et al.. 1967). This has been provisionally designated

THE TRYPTOPHAN

-,OMMOCHROME PATHWAY ird

INSECTS

147

30000 -

200

300

400

500

nm F i g . 5. Llltraviolet-visible spectra of ommatin D in buffer solution of pH 7.0 to 7.3 (-) and 5 N HCI (- - -).

“pigment IV”. It is at present not clear whether “pigment IV” is an integral constituent of ommins, or whether it is an artifact formed during the hydrolysis. Attempts t o reconstitute an ommin from “pigment IV” and 3-hydroxy-kynurenine have failed. Linzen (1967) has shown (by paper chromatographic comparison only) that each of the individual ommins of Gy l l u s bimaculatus yield “pigment IV” and 3. hydroxy-kynurenine. He concluded that the difference between the ommins must be due to a third, still unknown portion of the pigment molecule. This conclusion must be viewed with some reservation, as other interpretations q e also conceivable. Yet a paper of de Almeida (1968) lends support t o this hypothesis. Thus, if a solution of ommin A in formic acid is left to stand for a few days, ommin

148

20 000

10 000

I

2 1

I

I

I

I

300

I

I

I

I

I

400

I

I

I

I

I

I

I

500

nm

Fig. 6. Ultraviolet-visible spectrum of pigment “IV” in buffer solution of pH 7.0 to 7.3. (Courtesy of Dr Schaefer.)

A gradually disappears and “ommochrome I” (probably identical to ommin B) is formed; at the same time a blue fluorescing compound appears.

“Pigment IV” is a rather stable compound, but can be further broken down into another molecule of 3-hydroxy-kynurenine and a sulphurcontaining hydroxy-quinone, once it has been oxidized. Butenandt et al. (1 964) synthesized 4,5,8-trihydroxy-6-mercapto-quinoline-P-carboxylic acid which they conceived could arise from degradation of “pigment IV”. However, the synthetic compound and the degradation product behaved differently on paper chromatograms and had different uv spectra. The exception to the rule of ommochrome lability is provided by ommidin which is stable against both boiling acid and hot sodium hydroxide solution. Oxidative degradation of ommochromes has been attempted but has not met with success (Butenandt et nl., 1957d). A remarkable reaction (in terms of enzyme specificity) is the splitting of

nm

Fig. 7. Ultraviolet-visible spectra of ommidin isolated from the migratory locust. a = E!;-.(From Linzen, 1966.)

nm

Fig. 8. Ultraviolet-visible spectra of ommin in phosphate buffer,’pH 7.5 (-, left), 5 N HCI (- - -) and concentrated sulphuric acid (--, right). a! = A$;$. (From Butenandt et al., 1958a.)

BERNT LINZEN

150

the ommatin amino acid side chain by kynureninase. Both xanthommatin and rhodommatin yield alanine, but other ommochromes have apparently not been tested.

4.4

DISTRIBUTION AND TISSUE LOCALIZATION

Ommochromes are ubiquitous in the Arthropoda and occur in a number of other invertebrates, such as Anthomedusae (Yoshida et al., 1967), Polychaeta (Dales, 1962), Echiurida (Linzen, 1959), and Cephalopoda (Schwinck, 1953; Butenandt et al., 1958b). The data relating to insects are summarized in Table 5. The following principles have been employed in compiling Table 5. 1. No material earlier than Becker’s work has been included. Becker’s findings are classified as a “pre-war study” to indicate lack of chromatographic separation and identification with presently known individual ommochromes. 2. Any implied criticism is meant to stimulate reinvestigation whenever appropriate. This survey revealed that virtually no species has been analysed completely with regard to the qualitative and quantitative occurrence of ommochromes. In particular the mixture of ommins has usually not been separated but has been treated as a single pigment. Certainly, much of the paper chromatographic work was intended only as a quick check on the distribution of a particular pigment and should be supplemented by detailed studies. 3. The listing includes only positive statements about the presence or absence of ommochromes (e.g. if for a particular species the occurrence of ommins in the eyes is documented, it is understood that no other tissue, or stage, has been investigated properly). Quantitative data have been included wherever appropriate. Data for “isolated” ommins are presented to give at least the order of magnitude. However, losses during purification might have been compensated by the impurities still present. 4. Statements about the redox-states of ommochromes have been avoided, as they can be assessed only by microspectrophotometry with the exception of those cases in which a striking colour change is macroscopically visible. 5. Not all references available for a particular species have been cited and often not the earliest ones in cases where later authors have provided identification in modern nomenclature. Frequent reference has been made to Becker’s work t o demonstrate its broad approach and high impact. In many cases work on mutants has merely been hinted at. From this Table it is possible t o derive the following generalizations. 1 . The most widely distributed ommochrome is xanthommatin. It may be expected, at least in trace amounts, whenever other ommochromes are observed (with one exception described under point 2). As ommochrome

TABLE 5 Distribution of ommochromes among insects

Order, species

Details of ommochrome pigmentation

ODONATA Anax imperator LEACH

Eyes: xanthommatin

Aeschnn cyanea Aeschnn juncea L. Aeschna mixta Sympetrum flaueolum L.

“Ommatin in eggs” Ommin mixture in eyes “Ommatin” and “ommin” in eyes Xanthommatin (? ) in eyes

Sympetrum sanguineum 0 . - F . M~~LLER Sympetrum wlgatum

Xanthommatin in eyes

“Pooled Sympetsum species” ORTHOPTERA Gryllus bimaculatus DE GEER

Gryllus domesticus L. Gryllotalpa vulgaris L.

“Ommatin” and “ommin” in eyes; “ommatin” as hypodermal pigment in diaphragm and fat body Acridiommatin I as hypodermal pigment Eyes: ommidin as main pigment, accompanied by xanthommatin and ommins. Integument: ommin mixture, large quantity in abdomen, source for micropreparative isolation. Approximate ratio of the four slowest moving ommins 3 : 5 : 4 : 3 (Linzen, unpublished). Traces of xanthommatin Eyes: wnthommatin Ommin mixture in eyes

Criticism*

PC only, superficial, ommins expected. Integument? Prewar study PC only. Integument? Re-war study PC only; ommins expected. Integument ? PC only; ommins expected. Integument? Re-war study

Reference?

C

b a b C

C

b, d

Preliminary study

PP

Qualitative data well documented; no quantitative data

c, e , f

Preliminary PC. Other pigments? Preliminary PC. Xanthommatin?

C

a

TABLE 5-continued Order, species

Details of ommochrome pigmentation

ORTHOPTERA Rornalea rnicroptera BEAUV.

Eyes: ommidin main pigment, cryptommidin, some xanthommatin. Integument: same pigments Tylotropidius speciosus WALK. Eyes: ommidin, xanthommatin Eyes: ommidin, xanthommatin Acanthacris ruficornis fulva SJOST. Anacridiurn aegyptiurn L. “Insectorubin” (acridiommatins) in integument Eyes: ommidin No mada cris sep te rnfascia ta SERV. Eyes: ommidin Ornithacris cyanea STOLL. Schistocerca gregaria FORSK. Eyes: ommidin (main pigment), xanthommatin. Integument: acridiommatins I and 11. Quantitative data o n total ommochrome (“insectorubin”) content under various conditions given by Goodwin and Srisukh ( 1 9 5 0 b ) Eyes: ommidin (main pigment), cryptommidin, Locusta m e a t o r i a L. xanthommatin (?), acridiommatins I and 11. Integument: acridiommatins I and 11, xanthommatin (artifact?). Faeces: acridiommatins I and 11. Ommochrome quantity reduced in integument of albino mutant, not in eyes Oedipoda coerulescens L. Eyes: ommidin. Whole animals: xanthommatin Eyes: ommin-like pigment, “yellow pigment in Dociostaurus maroccanus great quantity” (ommidin?) “Ommatins” as hypodermal pigments and a t Further species inacrtion of musculature

Giticism*

Referencet

Only qualitative data

f

PC only; superficial PC only; superficial

f f

Pigments not separated PC only; superficial

f

PC only; superficial Not as well documented as for Locusta

nn

f c. f

Qualitative data well documented; quantitative data desirable

Superficially studied N o identification by modem techniques Prewar study

b

PHASMIDA Carausius morosus BRUNNER

Eyes: ommins, xanthommatin, no ommidin. Integument: ommins and xanthommatin; quantity dependent o n environmental conditions. Extremes: 5 pg of ommins, trace of xanthommatin in light animals; >540 pg of ommins, 260 pg of xanthommatin after dark adaptation

Careful quantitative analysis; ommins not separated

Eyes: ommin mixture, 3.2 mg isolated from 3500 animals, including larvae. No ommochrome isolated from rest of body Integument: xanthommatin, ommins

Ommins not separated. Xanthommatin?

a

Ommins not separated. Eye pigments?

00

Lacciferic acid and laccaic acid isolated from whole larvae; assumed to be ommin-like Eyes: ommins, traces of xanthommatin. No ommochromes in body, wings, eggs, or excreta Eyes: ommins

Further identification required

n

PC only

c, m,d

PC only; superficial

a

Eye pigments?

0

Chrysopa vulgaris SCHN.

Whole animals: 3.1 and 6.1 & per animal of xanthommatin in green and brown specimens, respectively “Ommatin” in integument

Pre-war study

b

LEPIDOPTERA cossus cossus L.

Larvae: xanthommatin in hypodermis

Well documented; other stages not investigated

ff

DICI’YOPTERA Blatta orientalis L.

Mantis religwsa L. HEMIPTERA LaccifL.r lacca

Rhodnius prolixus

Gerris lacustris L. NEUROPTERA Chrysopa carnea

TABLE 5’-continued Order, species

Details of ommochrome pigmentation

Criticism*

Referencet

LEPIDOPTERA

Galleria melonella F. Ephestia kiihniella 2.

Ptychopoda seriata SCHRK.

Plodia mterpunctella

Antheraea pernyi Hyalophora cecropia

Bombyx mori L.

Eyes: ommins and xanthommatin. No ommochromes in wings Larvae: ommins in ocelli, ommatin (xanthommatin?) in hypodermis. Moths: 3 ommins and xanthommatin in eyes and testis sheaths; ommochromes associated with nervous tissue; a number of mutants studied Larvae: ommatin (xanthommatin?) in hypodermis. Moths: same ommochromes in eyes as in Ephestia; ommochromes surrounding nervous system; “ommatin” in excreta Eyes: 3 ommins (ommin A, ommochromes I and I1 in Kiihns nomenclature) and xanthommatin. No ommochromes in wing scales. In mutant ra only xanthommatin As in Hyalophora cecropia below Eyes: ommins, xanthommatin. Ganglia and optic lobes: ommins (no xanthommatin?). No ommochromes in red areas of wings, nor in meconium Eggs: xanthommatin (0.22 mgg-’), 3 ommins (1.7 mg g-’ 1. Larval hypodermis: xanthommatin (= + - chrome “I”). Excreta: cinnabarinic acid. Adult eyes: ommin mixture (-4 pg per animal isolated). Several mutants investigated by Kawase (1955) and by Inagami (1958).

PC only

a. c

Qualitatively well documented; quantitative data desirable

a, b, c. d , s

Partly pre-war study, but eye pigments separated by PC

b, mm

Qualitative study. Other stages?

Qualitatively well documented; no separation of ommins Many qualified experiments by various workers; ommins might be better defined

Xanthommatin (protein-bound) in haemolymph of rb mutant Vanessa (Pyrameis) atalanta L. Meconia of imago: rhodommatin and ommatin D (cf. Table 6 ) . Same pigments in wings. Eyes: ommins and xanthommatin Vanessa (Pyrameu) cardui L. As in V. atalanta (cf. Table 6 for quantitative data on excreted ommochromes) Larval and pupal hypodermis: wnthommatin. Aglais (Vanessa) urticae L. Meconium: rhodommatin and ommatin D, no xanthommatin. Wings: rhodommatin, ommatin D (-20 and -5 pg per animal, respectively), no xanthommatin. Eyes: ommins (9 pg per animal isolated), xanthommatin Larval hypodermis: xanthommatin (?). Meconia, Inachis (Vanessa) io L. wings,and eyes of imago: same ommochromes as in Aglais urticae Araschnia laevana L. Cut contents of pupa: rhodommatin and ommatin D Larvae and pupae: xanthommatin in hypodermis Argynnis paphia L. of aii stages. ivieconium of imago: rhodommatm and ommatin D (cf. Table 6). no xanthommatin. Wings of both sexes: rhodommatin and ommatin D, no xanthommatin. Eyes: xanthommatin (3 pg per eye isolated), ommins. In valesina strain only traces of ommatins in wings Xanthommatin in hypodermis of hibernating Hestina japonica larvae; rhodommatin in gut contents at end of &pause Identical results as in Hestina above Sasakia charonda

Well documented, ommins not separated. Larvae?

a, c, m

Well documented, ommins not separated. Larvae? Well documented

c, m a. c, m

Well documented, but no quantitative data

a, c. m

PC only Well documented

m

Quantitative study might reveal further interesting results. NO ommatin D ?

ee

c,

ee

m

TABLE 5-continued Order, species

LEPIDOPTERA Heliconius spec. Aporia crataegi L.

Pieris brassicae L. Colias edusa F. Papilio machaon L.

Papilio xuthus

Parmssius apollo L. Biston cetularia Bupalus piniarius L. Sphinx ligustri L. Sphinx pimstri L. Deilephila (Pergesa) elpenor

Details of ommochrome pigmentation

Wings: xanthommatin (possibly artifact). Eyes: ommin mixture Larvae: xanthommatin in hypodermis. Meconium: rhodommatin and ommatin D (cf. Table 6). Eyes: ommins and xanthommatin Traces of rhodommatin and ommatin D in meconium Eyes: ommin mixture Eyes: ommins and xanthommatin. N o ommochrome in wings Testes: 0.18 pmol of xanthommatin per animal. Eyes: 0.6 pmol of xanthommatin per animal, plus ommin mixture. Wings: yellow pigments derived from kynurenine belong t o different pigment class Eyes: ommin mixture. Wings: red spots not containing ommochromes Eyes: ommins Eyes: ommins and xanthommatin. No ommochromes in wings, whole pupae, meconia Larval hypodermis: xanthommatin. Meconia: rhodommatin and ommatin D. Eyes: ommins Eyes: ommins and xanthommatin. N o ommochromes in wings or meconia Eyes: ommins and xanthommatin. Pink pigmentation of body and wings not based on ommochromes

Criticism*

Referencet

PC only

a, c

PC only. Eye pigment?

C

PC only. Xanthommatin? PC only

a. c

a

Well documented; estimate of xanthommatin in eyes appears high

dd

PC only; superficial

a

PC only; superficial PC only

a c, m

Limited qualitative study PC only PC only

Cerura uinuh L.

Agrotis comes Scoliopteryx libatrix L.

Larvae: xanthommatin in “saddle-patch”. dihydroxanthommatin in neck fold; spectacular colour change due t o ommochrome synthesis a t end of feeding period. Peak quantities: xanthommatin 0.6 mg; rhodommatin 0.5 mg; ommatin D 0.5 mg. Rhodommatin and ommatin D in meconia. (See page 175) Eyes: ommins Eyes: ommins and xanthommatin. No ommochromes in wings or meconia

Very detailed analysis, but adult animals not investigated

U

PC only; superficial PC only

a

c, m

DIPTERA

Tipula oleracea L. Culex pipiens L. Phryne fenestralis

Tabanus spec. (bovinus?) Luphria gibbosa Dwctria atricapilla Syrphus pyrastri L. Tubifera pendula L. Ceratitis capitata WIED. Drosophila rnelanogaster MEIG. Musca domestica..I

Eyes: ommins (9 pg per animal isolated), xanthommatin Eyes: unidentified ommochromes Eyes: ommins, “ommatin”. An ommatin in portions of fat body underlying hypodermis, and in testis sheath, Eyes: ommins Eyes: ommins, “ommatin” Eyes: ommins Eyes: xanthommatin only ommochrome “Ommatin” in abdominal pigment tissue (fat body?) Eyes: xanthommatin only ommochrome (-6 pg per animal; original estimate -13 pg) Eyes: xanthommatin (“brown pigment”). In Malpighian tubules of red mutant xanthommatin and ommin Eyes: xanthommatin only ommochrome. Several mutants studied quantitatively

a, c

Not adequately studied Prewar study

gg

PC only; xanthommatin? Prewar study PC only; xanthommatin? PC only Prewar study

a

b

b a C

b 11 c,

hh

c, ii

TABLE 5-continued Order, species

Details of ommochrome pigmentation

Criticism*

Referencet

DIPTERA

Calliphora erythrocephala MEIGEN Phormin terraenovae 4 related species

b , k k , rr

Eyes: xanthommatin only ommochrome (-50 pg per animal; original estimate -80 pg). “Ommatin” in testes Eyes: xanthommatin only ommochrome (-35 pg per animal) Eyes: xanthommatin still detectable in 50-yearold museum specimens

Well documented

Ommochromes (xanthommatin and “ommochrome 11”) in eyes only Whole pupae (early stage) “filled with dihydroxanthommatin”; in late pupae reduced amount of oxidized xanthommatin and no other ommochromes Eyes: xanthommatin, ommins Eyes: ommin mixture Ommin and xanthommatin content in eyes (pg per animal): 25 and 22 (drones); 14 and 13 (worker bees); 9 and 7 (queens); 11.6 and 27 (worker bees, age 10 days, data of Neese). Ommochrome synthesis continues after ecdysis; many mutants investigated by Dustmann 11968, 1969). Ommin mixture consists of four components, ratio 4 : 17 : 42 : 3 7 (ommin A = 4) (Linzen, unpublished)

Mixture of all ommins more likely than “ommochrome 11” alone Localization of pigment in pupal tissues?

r

PC only Age not given by Dustmann

a a, c, P, q

ss C

HYMENOPTERA

Ha bro bra con jugla nd is ASHMEAD Phobocampe unicincta

A t t a sexdens Apis mellifica L.

c, m

COLEOPTERA Cicindela spec. Melolontha uulgaris L. Tenebrio molitor

Eyes: ommin, “yellow component” Eyes: ommin mixture Pigment of Malpighian tubules resembles ommins

Pre-war study PC only Prewar study

d a

b

* PC, paper chromatography. “Pre-war study” designates lack of identification according to present standards (no chromatography, spectra missing or poor). f a . Butenandt e t a l . (1958b); b. Becker ( 1 9 4 2 ) ; ~Butenandt . etal. (1960b);d. Becker (1939); e. Fuzeau-Braesch (1957);f. Linzen (1966); g. Fuzeau-Braesch (1968); h. Bouthier (1966); i Bouthier (1969); k. Calarza and Stamm (1959); 1. Dustmann (1964); m. Kiibler (1960); n. Singh et al. (1966); 0 . Riidiger and Klose (1970); p. Dustmann (1966); q. Neese (1972); r. Leibenguth (1967a);s. Kiihn and Egelhaaf (1959); t. Mohlmann (1958); u. Linzen and Buckmann (1961); u. Kuhn and de Almeida (1961); w . Neubert (1956); x. Ajami and Riddiford (1971a);y. Kawase (1955); z. Inagami (1958); an. Koga and Osanai (1967); bb. Ishiguro and Nagamura (1971a. 1971b); cc. Butenandt et al. (1959); dd. Umebachi and Uchida (1970); ee. Osanai (1966a);ff. Merlini and Nasini (1968);gg. Dennhofer (1971); hh. Wessing and Bonse (1966); ii. Hiraga (1964); kk. Butenandt and Neubert (1955); 11. Ziegler and Feron (1965); mm. Kuhn (1963);nn. Colombo et al. (1955); 0 0 . Vuilleaume (1968); pp. Bouthier (personal communication); IT. Linzen (1963); 5s. Linzen and Schartau (to be published).

1 60

BERNT LINZEN

synthesis is an oxidation process, and as 3-hydroxy-kynurenine is very reactive, the formation of xanthommatin as a by-product seems t o be almost unavoidable. Xanthommatin is a regular companion of ommins. One might speculate (though there is yet no evidence to support such a speculation) that the varying proportion of xanthommatin t o ommins depends mainly on the amount of the other ommin precursors which are derived from methionine. While in most cases of joint occurrence, ommins (as a group) appear t o be predominant, they are at least equalled b y the amount of xanthommatin in bees. On the other hand, there are a few species, all belonging t o the Diptera, Cyclorrapha, in which xanthommatin appears to be the only ommochrome present in the eyes: Calliphora, Musca, Drosophila, Syrphus. For some time it was thought that these species had abandoned ommins completely, and that this represented a trait useful for chemotaxonomy. In 1966, however, Wessing and Bonse demonstrated ommins in Malpighian tubules of the Drosophila red mutant. Further chemical proof is desirable to strengthen this important finding. Xanthommatin may be the only ommochrome in the hypodermis of Lepidopteran larvae (Ceruru, Ephestia, Vanessa), but is certainly not the only ommochrome in internal pigment tissues. Thus, ommins have been found in ganglia of Hyalophora cecropia (Ajami and Riddiford, 1971a), while the testis sheath of Ephestia contains the whole series of Ephestk ommochromes (Kuhn and Egelhaaf, 1959). Conversely, in Papilio xuthus xanthommatin is the only ommochrome in the testes (Umebachi and Uchida, 1970). 2. Xanthommatin is unlikely to be a native pigment if it is found in conjunction with rhodommatin and ommatin D (as in butterfly wings and meconia), since it is a degradation product of these pigments. Ommatin D liberates sulphuric acid upon hydrolysis, thereby initiating an autocatalytic reaction. Kubler (1960) and Butenandt et ul. (1960b) examined this problem very carefully by paper and column chromatography of freshly deposited Nymphalid meconia and were unable to detect even traces of xanthommatin if the meconia were processed immediately. Similarly, xanthommatin is absent from fresh wing material of Pyrumeis atalunta and Vanessa urticae, although it assumes an increasing proportion of the total ommatins in aged material (see also Linzen and Biickmann 1961). In retrospect it can be seen that it was fortunate that in the early phase of ommochrome chemistry xanthommatin should arise as an artifact by decomposition of the two other ommatins. It was thus isolated in high yield from Vanessa meconia and, being the simplest of the ommatins, the first of which the chemical structure could be established. Rhodommatin and ommatin D apparently always occur jointly, although the instability of ommatin D might present some difficulties in its

THE TRYPTOPHAN

+

OMMOGHROME PATHWAY IN INSECTS

7 61

detection. These two ommatins have only been ohserved in Lepidoptera, an astonishing fact in view of the widespread ability of insects t o make fl-glucosides and sulphuric acid esters. They haw been found only in fat body, haemolymph, Malpighian tubules, gut, excreta and wings. and are in all probability excretory products. Rhodommatin and ommatin D are never in the form of granules, being either diffusely distributed or dissolved in vacuoles. While in meconia rhodommatin and ommatin D have been found in approximately equal amounts (Butenandt et al., 1960b), there is some indication that ommatin D predominates in Rhopaloceran wings. 3. Ommidin and cryptommidin are characteristic of the Orthoptera and have not been described in other insect orders. Ommidin also appears to be present in small amounts in Limulus. The ommidins are associated with xanthommatin (locust eyes), but may be also associated with ommins (Gryllus eyes). They are probably closely related to the latter by their common derivation from both tryptophan and methionin. Ommidins are bound to small granules as are the ommins. They are typical eye pigments, but in one instance, in Romalea microptera, they have been extracted from hypodermis (Linzen, 1966). 4. The ommins never occur singly, but always as a group of at least three, five or six (“at least six” in the case of Sepia; Butenandt et al., 1959). The quantitative relations of these have been determined by the author only in two cases (Gryllus bimaculutus, Apis) and only by making gross simplifications (“amount” =A5mnm x vo ume). Even in these two cases there appeared major quantitative differences which show that the composition of a given mixture of ommins is species specific. Next to xanthommatin, ommins are the most widespread ommochromes, being absent only from part of the Orthoptera and from a number of Diptera. In many, if not in most, insect species they are the dominant ommochromes. They occur in eyes, integument, pigment sheaths of testes, ganglia, and as egg pigments. They have, however, not been demonstrated in Lepidopteran wings, nor excreta; the substii.uted ommatins and the ommins appear to exclude each other. A simple phenoxazinone has recently been identified in the silkworm: cinnabarinic acid (2-amino-phenoxazin-3-one-l,9dicarboxylic acid; Ishiguro and Nagamura, 1971a), which had previously been isolated from tropical mushroom species (Gripenberg et al.. 1957; Gripenberg, 1958). Cinnabarinic acid originates from 3-hydroxy-an1hranilic acid which is a major metabolite of tryptophan in the silkworm. In the mutant rb (which is deficient in kynureninase and accumulates 3-hydroxy-kynurenine) xanthommatin is formed instead. As a whole, ommochrome pigmentation might still be expected to yield a number of interesting results. For example, the xridiommatins formed in the hypodermis of the migratory locust have been little studied. From the

166

BERNT LINZEN

synthesized in insect eyes are yellowish brown in early stages, to mention only a few examples. Biickmann (1964, 1965) has pursued this problem in the case of epidermal xanthommatin in Cerura, and has denied a respiratory function, though he surmised that in periods of anoxia xanthommatin might play a minor function as an electron acceptor. Ommochromes are thus unlikely t o have a primary function as redox system. On the other hand, the evidence for a function as screening and pattern pigments, and for the deposition of ommochromes as metabolic end products is convincing and will be considered in some detail. 5.1

OMMOCHROMES AS SCREENING PIGMENTS

The light receptors of arthropod compound eyes are surrounded by the pigment cells in such a way that the screening function of ommochromes appears self evident. Screening affects light perception in two ways: first, the total light energy impinging on the receptors is reduced (so as to lower the sensitivity) and, secondly, the largely unidirectional light impinging on the receptors results in an increase in acuity. The first of these two’ principles is critical in marine or nocturnal animals and is counteracted by mechanisms which move the screening pigment in response t o light intensity. It is also relevant in the perception of light of long wavelengths, where absorption by visual pigments is low. A comparison of the absorption spectra of ommochromes in solution and of ommochromes in situ reveals some instructive differences and shows the degree of adaptation of these screening pigments at every level of organization. Microspectrophotometry has been applied t o the study of ommochromes in Musca dornestica by Strother (1966). A detailed investigation of Calliphora screening pigments (xanthommatin, pteridines) has been carried out by Langer (1967) and of ommin granules by Hoglund et al. (1970) and by Langer and Struwe (1972). Both the oxidized and the reduced forms of xanthommatin occur in vivo. This was first demonstrated in Calliphora by conventional light microscopy by Hanser (1959). However, Langer’s measurements on individual and small groups of granules suggest that there might be two different states of xanthommatin. In one type of granule it appears to be partially oxidized (to an unknown and certainly variable percentage) and possibly to be in the free state; in the other it is mainly reduced and bound to protein. The comparison of Calliphora mutants has been helpful in this deduction. In the white mutant of Calliphora, bright yellow granules are observed which exhibit absorption spectra very close to the spectrum of (oxidized) xanthommatin in solution (Ama = 435 nm). In the wild type, brownish-yellow granules are found. Their spectra may be imitated by a solution of fully reduced xanthommatin which has been left in contact

THE TRYPTOPHAN

+

OMMOCH ROME PATHWAY IN INSECTS

163

of colour, depending on the state of pigment (see p. 166). The distribution of pigment granules in Drosophila wild-type eyes and in various mutants has been described by Zeutschel (1958) and try Nolte (1950). Nolte also disputes earlier histological literature. Two detailed electron microscopic studies have been published by Shoup (1966) and Fuge (1967) who arrive at similar conclusions, dissenting only in the origin of the granules. Although pigment deposition might conceivably be preceded by the formation of colourless “matrix” granules, it is clear from their observations that both occur simultaneously. Granule formation commences abruptly at 44 t o 48 h of pupal age. Shoup frequently observed small ommochrome granules (“Type I” granules in her nomenclature) adjacent to Golgi regions. Fuge, however, denied this and considered small cisternae of the endoplasmic reticulum to be precursors of the granules. These, it was postulated, were then filled with pigment and thereby increased in size. Fuge, on the other hand, suggested a derivation of drosopterin granules from Golgi vesicles, for which Shoup does not find any evidence. The problem remains unsettled m d justifies a further close investigation of the initial phases of granule biogenesis. Growth of the granules is characterized by a linear increase of the diameter which doubles about every 24 h. Schwabl and Linzen (1972) measured granule growth in basal and approximately mid-ommatidial regions of the secondary pigment cells in Drosophila.They found significant size differences, the basal granules being larger from the initiation of pigmentation. The above work does not provide evidence of any “matrix”, “ground substance”, or “carrier” of the pigment in Drosophila. Although microspectrophotometry might provide an answer by determining the position of maximal pigment absorption, direct positive evidence is provided by Schneider’s electron micrographs of Calliphora wild-type and chalky eyes (Langer, 1967). The chalky mutant which is completely devoid of all pigments, nevertheless, contains granules of appropriate size and shape which are filled by very fine granular, opaque material. In Ephestio Hanser (1946, 1948) demonstrated “carrier granules”, which lightly stained with haematoxilin after removal of the pigment. She also described “pregranules” which subsequently took on reddish pigmentation. These granules which carry ommins are again surrounded by a membrane and are probably derived from vesicles formed by the Golgi apparatus and which actually decrease to reach their final size (0.5 to 0 . 6 p m ) while they are gradually filled with electron-dense material (Horstmann, 1971). Maier’s (1965) earlier findings of “structural complexes” as granule precursors appear to be a misinterpretation. It appears then that the existence of a carrier protein in the pigment granules is beyond doubt, even if this is not apparent in the v and cn

164

BERNT LINZEN

mutants of Drosophila. The absence of “empty” granules in these mutants may be explained by the assumption that ommochrome precursors are required for the formation of the granules. In fact by feeding kynurenine to u larvae, and 3-hydroxy-kynurenine to cn larvae, normal growth of pigment granules is induced (Schwabl and Linzen, 1972). These granules also show a membrane. One might suppose that the growing granules carry the complete pigment synthesizing apparatus which is gradually covered by pigment (and probably inactivated). This would correspond with the development of the melanosome. In Ephestiu, Muth (1967, 1968 1969) has shown that the capacity for ommochrome synthesis in the eye is influenced by the amount of precursor (administered kynurenine or 3-hydroxy-kynurenine). Muth developed a mathematical model which was able to describe pigment synthesis and comprised an induction of the biosynthetic apparatus. Langer and Struwe (1972) attempted to estimate the pigment concentration within the granules from microspectrophotometric measurements. They arrived at a value of about 0.5 M , corresponding to roughly 20 per cent of the fresh weight. It might be added that Fuzeau-Braesch (1971) observed using the scanning electron microscope, “nicely spherical’’ ommochrome and pteridine granules of 2 to 5 pm in diameter, in two Orthoptera. Virtually no biochemistry has been done on ommochrome granules. The author is also extremely sceptical about the few papers published because of his doubts about the purity of the preparations employed. It is, in fact, not easy t o obtain pure pigment granules, and a small contamination by other particles or membrane fractions may simulate metabolic properties actually not present. On the basis of histological stiining the granules were assumed to contain ribonucleic acid (Caspari and Richards, 1947/1948). Bellafny (1958) prepared “insectorubin particles” from locust tissues and measured a number of mitochondrial activities; he cautiously concluded that insectorubin particles were a species of mitochondria. Likewise, Ziegler and Jaenicke (1959) determined a number of mitochondrial enzymes in preparations of Drosophila pigment granules which, according to a re-examination in the author’s laboratory, could not have been more than 50 per cent pure. Currently, the xanthommatin granules of Culliphoru are being studied in the author’s laboratory. They are separated by density gradient centrifugation; their density was estimated at about 1.30. Purity was checked by the ratio of xanthommatin-residual nitrogen and by assay for succinic dehydrogenase. The best preparations were completely devoid of the latter enzyme. 4.6 BINDING OF OMMOCHROMES TO PROTEINS Although it had been generally appreciated, since Becker’s work, that ommochromes were bound to proteins few attempts have been made to

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY I N INSECTS

165

isolate such chromoproteids. Bowness and Violken (1959) reported on a pigment from housefly heads (identification of the chromophore is lacking) which was bleached by strong light and also in the dark at pH 8. Yoshida et al. (1967) compared light-sensitive pigments from Anthomedusan ocelli and Lucilia caesar and stated that a macromo1ec:ular pigment extracted with cetyltrimethylammonium bromide behaved identically in chromatography on Sephadex G-100. The Japanese workers identified xanthommatin as a chromophoric group of this material. More recently, Ishiguro and Nagamura (1971b) and Ajami and Riddiford (1971b) isolated ommochrome-proteids. By extraction with Ringer solution, precipitation with ammonium sulphate, filtration through Sephadex G-200 and chromatography on DEAE-cellulose, the Japanese workers isolated a xanthommatin-carrying protein from the haeinolymph of the rb mutant of Bombyx mori which appeared homogenous in disc electrophoresis. From the xanthommatin content of 15-30 pg mg-’ protein, the molecular weight may be calculated to be above 140 000 Daltons. The Harvard group separated a xanthommatin-proteid from an ommin-proteid by gel filtration. The latter had a molecular weight -of about 24 000 Daltons; the redox potential at pH 7 was found to be 196 f 7 mV. The absorption maxima of both chromoproteids are situated at surprisingly short wavelengths and are not at all comparable to the microspectrophotometric data (Langer 1967; Hoglund et al., 1370) of whole granules :in v i m . The ommin-proteid actually absorbs at shorter wavelengths than purified ommin in solution, so that some doubts may arise concerning denaturation of the chromoproteids during isolation.

5 Functions of ommochromes An understanding of the functions of ommothromes requires a knowledge of the localization of these pigments and their characteristic properties: redox behaviour, absorption of ultraviolet and visible light, and low solubility. These properties could enable omrnochromes to act as electron accepting or donating systems, as authentic functional pigments and as metabolic end products. Of these possible roles, the participation in electron transfer has intrigued many worker!;, and continues to do so. A respiratory function of ommochromes was suggested by Horowitz (1940), Horowitz and Baumberger (1941), Bellamy (1958), and Harano and Chino (1971) and has been implied by other workers. Such a function might appear most likely, for a change from the oxidized to the reduced state is in fact observed in some species. The eggs of Ure(zhiscaupo, for example. turn from yellow to red during ripening (Horowitz, ” 1940; Horowitz and Baumberger, 1941), while the larva of Ceruru vinula displays a spectacular colour change prior to pupation. This starts by a reduction of xanthommatin (Biickmann, papers to be cited below). The ommochromes

166

BERNT LINZEN

synthesized in insect eyes are yellowish brown in early stages, to mention only a few examples. Biickmann (1964, 1965) has pursued this problem in the case of epidermal xanthommatin in Cerura, and has denied a respiratory function, though he surmised that in periods of anoxia xanthommatin might play a minor function as an electron acceptor. Ommochromes are thus unlikely t o have a primary function as redox system. On the other hand, the evidence for a function as screening and pattern pigments, and for the deposition of ommochromes as metabolic end products is convincing and will be considered in some detail. 5.1

OMMOCHROMES AS SCREENING PIGMENTS

The light receptors of arthropod compound eyes are surrounded by the pigment cells in such a way that the screening function of ommochromes appears self evident. Screening affects light perception in two ways: first, the total light energy impinging on the receptors is reduced (so as to lower the sensitivity) and, secondly, the largely unidirectional light impinging on the receptors results in an increase in acuity. The first of these two’ principles is critical in marine or nocturnal animals and is counteracted by mechanisms which move the screening pigment in response t o light intensity. It is also relevant in the perception of light of long wavelengths, where absorption by visual pigments is low. A comparison of the absorption spectra of ommochromes in solution and of ommochromes in situ reveals some instructive differences and shows the degree of adaptation of these screening pigments at every level of organization. Microspectrophotometry has been applied t o the study of ommochromes in Musca dornestica by Strother (1966). A detailed investigation of Calliphora screening pigments (xanthommatin, pteridines) has been carried out by Langer (1967) and of ommin granules by Hoglund et al. (1970) and by Langer and Struwe (1972). Both the oxidized and the reduced forms of xanthommatin occur in vivo. This was first demonstrated in Calliphora by conventional light microscopy by Hanser (1959). However, Langer’s measurements on individual and small groups of granules suggest that there might be two different states of xanthommatin. In one type of granule it appears to be partially oxidized (to an unknown and certainly variable percentage) and possibly to be in the free state; in the other it is mainly reduced and bound to protein. The comparison of Calliphora mutants has been helpful in this deduction. In the white mutant of Calliphora, bright yellow granules are observed which exhibit absorption spectra very close to the spectrum of (oxidized) xanthommatin in solution (Ama = 435 nm). In the wild type, brownish-yellow granules are found. Their spectra may be imitated by a solution of fully reduced xanthommatin which has been left in contact

THE TRYPTOPHAN

+

OMMOCH ROME PATHWAY ird INSECTS

167

with air for a prolonged period (Amax = 490 nm I. Granules which appear bright red in the light microscope (the most numerous in the wild type) absorb maximally between 520 and 560 nm. If these are extracted with glycerol, so as to remove unbound ommochrome, maximal absorption occurs at 560 nm. The large bathochromic shift. from 492 nm (dihydroxanthommatin in solution) to 560 nm is due t o the binding of the pigment to protein. A bathochromic shift was also measured by reflection spectrometry of the integument of Hestina japonica (Osanai. 1966b). To appreciate the natural situation it is necessary to superimpose the various spectra. Making some reasonable assumptions, and taking into consideration the absorption of blue light by the pteridines also present, a more or less even level of absorption is obtained which extends from 300 to 590 nm. This corresponds to the situation revealed by microspectrophotometry in a radially illuminated eye in which a bleached rhabdome is used as reference spot (Fig. 9). The total absorbance can be estimated t o exceed 6. The absorbance of single granules lies tietween 0.25 (Langer and Struwe, 1972) and 1.0 (Langer, personal communication). Thus, the ommochrome screen constitutes a homogenous gTey filter with extremely low transmission and a sharp cut-off at 600 to 640 nm. When ommins constitute the major screening pigment the situation is relatively simple, as the oxidized state is not involved. The absorption maximum of pigment granules of the moth, Cderio euphorbiae, lies at

nm Fig. 9. Microspectrophotometry of Culliphoru pigment cells, demonstrating neutral grey filter function and cut-off at 600 nm in wild-type individuals (a) and effect of mutation to white (b). (Traced from an original recording in Langer, 1967.)

168

BERNT LINZEN

547 nm (Hoglund et al., 1970); the bathochromic shift caused by binding of the ommins to protein is therefore only 20 to 30 nm. In Heliconius the shift is a little greater (Langer and Struwe, 1972). Measurement of individual granules has revealed variations in light absorption between 350 and 540 nm. Difference spectra suggest that this might be due t o a variable admixture of oxidized xanthommatin. Again, the general shape of the spectrum is that of a neutral grey filter with a cut-off just below 600 nrn. The high absorbance values of this filter results in a substantial decrease in sensitivity in comparison to mutants devoid of screening pigment. Autrum (1955) compared electroretinograms of Calliphora wild-type and white eyes and found that the sensitivity of the mutant eyes was 100 times higher for red light, but l o 3 to l o 4 times higher for light of shorter wavelengths. If the amplitude of the “on-effect” (in mV) is plotted against wavelength a single maximum near 510 nm is obtained in the mutant, whereas a second maximum around 640 nm appears in the wild-type, following elevation of the quantum level. This means that stray light becomes effective (by exciting a greater number of receptors) in the region of high transmission of the pigment screen in spite of the low sensitivity of the receptors at this wavelength. The higher proportion of stray light was also demonstrated by measuring single cell potentials (Washizu et al., 1964). The mutant chalky, which is not only devoid of ommochromes but also of pteridines, shows still higher “on-effects” in the short wavelength region. The carbohydrate utilization, measured in isolated irradiated heads, was also found to be highest in the eyes of the chalky mutant (Langer and Hoffmann, 1966). The decrease in visual acuity associated with the absence of ommochromes has been studied by several workers (Drosophila: Kalmus, 1943; Fingerman, 1952; Gotz, 1964; Hengstenberg and Gotz, 1967; Burnet etaf., 1968; Wehner et al., 1969. Calliphora: Autrum, 1961. A p i s : Neese, 1968). In Calliphora the optomotor responses of wild-type and white flies were identical as long as bright illumination prevailed. At low light intensities, however, acuity was reduced in the mutant (Autrum, 1961). Similarly contrast perception is lowered in Drosophila mutants wa and w , due to the higher proportion of stray light relative to total light intensity. The relative light intensities reaching the photoreceptors of +, se, wa , and w eyes were estimated at 1 : 1 : 7 : 19 (Gotz, 1964; Hengstenberg and G6tz, 1967), similar proportions also being obtained by independent experiments (Wehner et al., 1969). In Drosophila u b w the visual acuity increased if the larvae were fed increasing amounts of kynurenine, causing biosynthesis of ommochromes (Burnet et al., 1968). In the chartreuse mutant of the honey bee, which lacks all n m m o chromes at the time of emergence and which forms only xanthommatin and traces of ommins during imaginal life, orientation is severely impeded.

THE TRYPTOPHAN

-+

OMMOCHROME PATHWAY IN INSECTS

169

Chartreuse worker bees are inferior collectors in comparison to the wild-type individuals. They show lowered contrast perception in training experiments, and reduced flight speed in field experiments, and this becomes more marked in areas which offer few orientation marks (Neese, 1968). In conclusion, it would appear that visual acuity of the compound eye is reduced by the lack of ommochromes. This does not result from an increase in the visual field of a’ single ommatidium, but from the increased “background illumination” of the receptors, which makes it more difficult to distinguish between areas of different light intensity. For demonstration at the level of single receptor cells and relevant discussion, see Washizu et al. (1964) and Streck (1972). In a number of insects, ommochromes form pigment sheaths around internal organs, such as testes (Ephestia, Papilio) and ganglia (Ptychopoda seriata, Fig. 10, Hyalophora cecropia). Since no experimental data concerning the function of these pigment sheaths are available we are presently left with blind speculation; for a few hints see Ajami and Riddiford (1971b).

Fig. 10. Young caterpillar of Ptychopodo seriuta made transparent to demonstrate pigmentation of ganglia. (From Kuhn, 1940.)

5.2

OMMOCHROMES AS PATTERN PIGMENTS. RELATb3N TO OTHER PIGMENTS

In no other phylum of animals is pigmentation ;is varied and as complex as in the Arthropoda, notably in insects. Ommochxomes represent one of half a dozen kinds of pigments, each comprising a variety of compounds occurring in different tissues and patterns and characterized by its own ontogeny. Ommochromes-besides their universal distribution in arthropod eyes-are frequently found in the hypodermis OF inseits (cf. Table 5 ) , and will, if they are not occluded by cuticular melanin, contribute to the outward appearance of the animals. Thus the brownish or reddish

170

BERNT LINZEN

colouration of many Orthoptera and Odonata (cf. also Becker, 1941; Krieger, 1954) are due to hypodermal ommochrome granules. The striped pattern of Schistocerca eyes is produced by ommochrome granules filling the tips of secondary pigment cells in distinct areas (Nolte, 1965). In the Neuroptera, hibernating Chrysopa females attain their light brown tint by formation of xanthommatin (Becker, 1942; Riidiger and Klose 1970). In the Lepidoptera many larvae have ommochrome granules in their hypodermis. These participate both in diffuse patterns and in those rich in contrast (Cerura vinula, Sphinx ligustri). The most brilliant appearance of ommochromes is in the wings of many Rhopaloceran butterflies (e.g. Argynnis paphia, Vanessa species, Heliconius species). The function of insect pigment patterns will not be discussed in detail. A particularly good example of the importance of ommochromes in this respect is provided by the wings of a butterfly, Argynnis paphia (Magnus, 1958). The primary stimulus for mating behaviour of the male is provided by the rhythmical flashing of orange-yellow in the fluttering flight of the female. This induces the male to aim directly at the female and. in the presence of a specific odour, courtship will ensue. Magnus analysed the parameters of the innate optical releaser, which arouses the male to fly into target by placing dummies on a merry-go-round set up in a wood clearing and observing the males’ behaviour. He discovered that the quality of the natural colouration is optimal. However, neither the pattern of black specks nor the natural contour of the wings form part of the releasing signal. On the other hand, increasing the stimulus quantity (by increasing the coloured area and the frequency of its appearance) rendered dummy insects more attractive. This occurred whether the dummy was a rotating cylinder or a more natural device with flapping wings. Thus, “superoptimal releasers” of the first stage in mating behaviour can be easily constructed. In natural populations a mutant strain, A . paphia, forma valesina, is found in low percentage (0.3-0.4 per cent in the Stuttgart area) which lacks the ommochrome pigmentation of the wings of the female. These greyish, inconspicuous females usually remain unnoticed by the males (except for direct olfactory stimuli at short distance) and would have little chance to mate. On the other hand, they are less likely to fall prey t o birds. The advantage gained by cryptic colouration is thus a counterweight maintaining a delicate balance between the two strains. The brightness of reduced ommochromes may also serve in warning colouration. In the larva of Cerura vinula most of the hypodermal xanthommatin appears to be in the oxidized state. However, the intersegmental skin between the head capsule and prothorax is brilliantly red, due to its content of dihydro-xanthommatin. This part is deeply infolded and can suddenly be displayed if the animal is disturbed. At the same time 30 per cent formic acid is ejected from a gland in the ventral part

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

171

of the neck (Schildknecht and Schmidt, 1963), while at the hind end of the animal two filaments can be extruded and weved vigorously. These again are coloured red by reduced xanthommatin (Linzen and Biickmann, 1961; Biickmann, 1965, and other papers). Cerura is also a good example to demonstrate the concealing effect which ommochromes may produce against a suitable background. The feeding larva exhibits a disruptive pattern, wich a countershading in both components of this pattern (lateral and kentral green areas, dorsal “saddle-patch”). At the time when the fully grown larva is about to leave its food and t o crawl down the trunk of the food tree, the green colour (due to a bilin in the haemolymph) is covered by new ommochrome synthesized in the hypodermis. The animal attains a dark red shade which. fits well into the new background met with. In Orthoptera, ommochromes are at least in part responsible for adaptive colouration (see below). Cott, in his book on adaptive colouration (1940-1966), presents several examples (Mantis religiosa, Lepidopterous pupae), which document the survival value of this principle. other examples may be found in the more recent literature (e.g. de Ruiter, 1955). Very little is known about the relation o f ommochromes t o other pigments. It appears that hypodermal ominochromes are frequently associated with melanin, localized in the overlying cuticle. Dustmann (1964) nicely demonstrated this correlation in the stick insect (Fig. l l ) , but stated that ommochrome was found also in areas where the overlying cuticle was devoid of melanin. Biickmann (1965) concluded that in Cerura all areas of the integument covered by cornpletely black cuticle also contained ommochrome granules, but that these never contribute t o external colouration. In other areas ommoclirome pigmentation might occur per se, the cuticle being devoid of melanin. A linkage between cyticular melanin and hypodermai ommochrorne is also observed in other species (Nickerson, 1956; also, cf. Fig. 3 in Kiihn, 1940). The causal relation of this is not clear, for as both pigments are derived from essential

Fig. 11. Correspondence between cuticular melanin spots (Me) and epidermal ommochrome in the integument of Carmrsius morosus. (From Dustmann, 1964.)

172

BERNT LINZEN

amino acids which might be expected t o accumulate simultaneously. A loose metabolic coupling of their biosynthesis is plausible, but could not explain the graduated differences observed. Ommochrome pigmentation may supplement melanin pigmentation to make a species look entirely dark. This is the case in Gryllus bimaculatus. In the abdomen the ventral and lateral parts of the integument are intensely coloured by ommin granules, so that the obscure aspect is maintained even if the deeply melanized sclerites move apart during expansion of the abdomen (as in mature females full of eggs). In Gryllus, in contrast to Cerura and Carausius, melanin and ommochrome production do not appear to be metabolically interdependent. There are several mutants in which both ommochrome and pteridine biosynthesis are affected. This is the case in the white mutants of Drosophila and in the mutant chalky of Calliphora. Possibly, in these mutants the structural protein or “matrix” of the pigment granules is blocked. However, the relation must be highly complex, for in Drosophila basic differences of structure and origin of the two types of granules have been shown (Shoup, 1966; Fuge, 1967). Both granule types are separated from the cytoplasm by a membrane, and one may speculate about a possible interference of the mutant genes with the proper and timely synthesis of this structural element. Another type of metabolic link between ommochromes and pteridines seems to exist in Ephestia. The mutation to a (which in biochemical terms results in an inactivation of tryptophan oxygenase) brings about not only the accumulation of free tryptophan and the absence of ommochromes, but also a striking change in the pattern of fluorescent compounds. This is shown in both paper chromatograms and in histological sections (Hadorn and Kiihn, 1953; Kiihn and Egelhaaf, 1955; Egelhaaf, 1956a, 1956b, 1963a; Reisener-Glasewald, 1956). The increase in fluorescence is due t o dihydro-ekapterin (which during isolation is oxidized to ekapterin) (Viscontini and Stierlin, 1961, 1963). Injection of kynurenine will not only result in normal ommochrome synthesis, but also restore the normal pattern of fluorescent pteridines (Kiihn and Egelhaaf, 1955). No plausible biochemical explanation is available t o the author. Ghosh and Forrest (1967a, 1967b) who discussed the involvement of pteridines in the tryptophan + ommochrome pathway, do not discuss this problem. Kiihn (1956) advanced the hypothesis that ommochromes and pteridines (and possibly their precursors) might compete for reactive sites on the pigment granules. In view of the specificity of most enzymes and, in Drosophila, of the deposition of ommochromes and pteridines on distinct types of granules, the hypothesis has lost its attractiveness. Moreover, Reisener-Glasewald (1956) found the most striking increase of fluorescence in the a eye in the crystalline cones. Ziegler and Harmsen (1969) consider cone fluorescence t o be an artifact

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

173

resulting from secondary adsorption of dissolved pteridines during fixation of the tissue. It is not clear to the author whether there is experimental evidence to support this argument. Since tetrahydro-pteridines may act as cofactors in hydroxylation, their involvement in ommochrome biosynthesis has been considered (Ghosh and Forrest, 1967a, 1967b). From recent work it appears unlikely that the enzyme, kynurenine-3-hydroxylase(see p. 191), has a pteridine cofactor. The possible inhibition of tryptophan oxygenase by preridines should, however, be considered. With regard to pigment function, pteridines and ommochromes may, of course, supplement each other both in patterns o external colouration and in the screening of the ommatidia in the compound eye (see pp. 166-167).

5.3 OMMOCHROMES IN

MORPHOLOGICAL COLOUR CHANGE

Morphological colour change has been dealt with in recent reviews by Rowel1 (1971) and Fuzeau-Braesch (1972), with valuable bibliographies. Morphological colour change is related to the aspects of pattern, of adaptation, and of development. The latter aspect gains much in importance when pigmentation is essentially a one-way process (i.e. when the pigment once deposited is not further meta.bolized). This is the case both with melanins and with ommochromes. But, while melanins may be shed along with the cuticle at each moult, ommochromes, being hypodermal, may merely be diluted by further growth and cell multiplication. The role of ommochromes was quantitatively analysed in two species, namely Carausius morosus (Biickmann and Dustmann, 1962; Dustmann, 1964) and Cerura vinula (Linzen and Biickmann, 1961; Biickmann et al., 1966). In the author’s view, the most completi: biochemical analysis of morphological colour change in any species is that carried out in the stick insect, Carausius, by Buckmann and his students. The stick insect is able to adopt a great variety of colour shades. While the amount of pigments present sets the range of environmental stimuli to which the animal is able to respond, fine control is achieved by means of thromatophores which are controlled by neurohormones. Together with cuticular melanin, ommochromes are chiefly responsible for the light or sombre appearance. Dustmann (1964) identified xanthommatin and the usual mixture of ommins. If the animals are under normal daylight conditions, the hypodermal ommochrome content remains low (4.7 pg per animal). However, if the lower halves of the eyes are covered by black lacquer, a dramatic increase of nearly a hundredfold is observed within a few moults. Ommochrome synthesis is also enhanced at high temperature. Under these conditions it is much less dependent on the pattern and intensity of illumination. The total ommochrome content may amount to 0.8 mg per

174

BERNT LINZEN

animal which is equivalent to 0.6 per cent of the dry weight. The percentage of xanthommatin is increased under conditions of induced ommochrome synthesis. I t is important to notice that ommochrome pigmentation is irreversible. For example, if during the 4th larval stage the conditions were changed to induce light colouration, there was no change in ommochrome content up to the adult stage. Similarly, if conditions causing light pigmentation prevail throughout from the day of hatching, a constancy of ommochrome content is demonstrated. Approximately 5 pg of ommochromes are synthesized during embryonal development and no ommochrome at all thereafter. According to Berthold (1971) the formation of ommochromes is correlated with a decrease in excretion of kynurenic acid which appears to be the normal end product of tryptophan metabolism in the stick insect. The induction of ommochrome synthesis may become effective within any given larval stage; but conditions must be changed immediately after a moult. This may reflect either a critical stage in the responsiveness of the integument, or a periodical sensitivity of the brain-corpora allata system which mediates between environmental change and integumental reaction. Berthold (1971) has critically examined the role of the corpora allata in induced ommochrome synthesis and found it to be at best indirect. Willig (1969) analysed the carotenoids and bile pigment (biliverdin) of this species and found the quantitative changes to be much less spectacular. While the ommochrome content can vary by two orders of magnitude, the changes in these two other pigment groups do not exceed a factor of two. Interestingly, the carotenoid changes also involve pigment transport from the fat body to the hypodermis. In a recent paper (Berthold and Henze, 1971), the participation of pteridines in the colour change of Carausius is also shown, dark animals containing less leukopterin and isoxanthopterin than light green ones. Thus, at least 16 pigments are involved in colour adaptation of Carausius: melanin, xanthommatin, the ommin mixture (at least four pigments), biliverdin, five carotenoids, four pteridines (other fluorescent compounds are present in trace amounts). All of these, with the exception of ommochromes and melanin, play a minor role. From the observations of Krieger (1954), it may be surmised that similar principles may be responsible for adaptive colour change in the larvae of Odonata. Krieger assigned the major role to melanin without, however, having performed quantitative measurements. In the case of the praying mantis and of the migratory locust there is some controversy in the literature (cf. Rowell, 1971, and Fuzeau-Braesch, 1972) resulting from the conflicting findings of Vuilleaume (1968) and of other workers in this field. Vuilleaume contends that the major factor responsible ‘for the colour varieties in these species is a bile pigment which may be oxidized to various degrees under irradiation. Although Vuilleaume had identified ommo-

THE TRYPTOPHAN -+ OMMOCHROME PATHWAY IN INSECTS

175

chromes in the integument of Mantis, she denied the functional significance of these pigments. Susef-Michieli (1965), on the contrary, found a “much greater quantity of ommochromes” in brown individuals, while finding no evidence for the presence of bile pigment. From Goodwin’s (1952) discussion of locust pigmentation it is evident that the importance to be assigned to each pigment critically depends 011 the developmental stages. While “insectorubin” (ommochromes) is present in different amounts in gregariu and solitaria hoppers, these differences are not crucial in the outward appearance of the two phases, w’iich is, on the contrary, determined by the amounts of bile pigment a i d of melanin. However, in the immature and mature adult animals the ommochromes are at least partly responsible for the external aspect. As the methods available for quantitative pigment determination are now much refined, chemical analysis should be able to solve these current problems. A quite different case of colour change is observed in Lepidopterous caterpillars (e.g. Cerura vinufa) in which it represents an obligate step in development. This type of colour change, although brought about by hormone action, is not mediated by neurohuinoral response to changing environmental stimuli. In Ceruru the synthesis of large quantities of ommochromes appears to be a secondary, metabolic effect, which, however, is precisely correlated with the abandonment of the leaves and the move to the trunk of the food tree. Buckmann (main papers: 1953, 1959a, 1959b, 1963, 1965; Karlson and Buckmann, 1956) has analysed the morphological and hormonal basis of this striking phenomenon and has found a number of interesting effects of temperature and atmospheric composition. The biochemical basis has been studied in collaboration with the present author (Linzen and Buckmann, 1961; Biickmann et ul., 1966). The larva of Ceruru is marked by a dark brown rhomboid pattern, the “saddle-patch” which contains xanthommatin. At the end of feeding, about 10 days prior to pupation, the pigment is reduced to its red state. At the same time synthesis of dihydro-xanthomniatin starts all over the integument. This causes the larva to turn almost black (because the haemolymph is dark green). Shortly afterwards the fat-body turns red, due to formation of rhodommatin and ommatin D, while the green haemolymph pigment gradually disappears. While the larva is contained within its cocoon it is bright red for a short period. The dihydro-xanthommatin then begins to disappear from the hypodermis, causing the anterior part of the larva to turn greenish again. Finally all of the ommochrome is found in the fat body and in the excretory organs, while the hypodermis retains only small amounts of red pigment. Colour photographs of these dramatic changes, with others demonstrating the effects of ”ligatures, have been published by Buckmann (1959). In Ceruru, rhcidommatin and ommatin D are found in many parts of the body: integument, fat body, gut, Malpighian

BERNT LINZEN

176

tubules (where they are formed at a very early stage), haemolymph (transiently). It is a tempting thought that rhodommatin and ommatin D in the haemolymph should represent the transport form of dihydroxanthommatin removed from the hypodermis. This suggestion has, however, not been proved. The total amount of ommochromes per animal is about 1 mg equivalent to 0.2 per cent of the total dry weight. Chemical analysis of the haemolymph during the period of colour change (Biickmann et al., 1966) has revealed significant changes in the contents of protein, carbohydrate, and individual amino acids, notably, proline, phenylalanine. tyrosine, and the aliphatic amino acids. Tryptophan and 3-hydroxy-kynurenine undergo a dramatic and transient increase. The chemical results are in accord with the histological findings. Thus immediately after termination of larval growth a profound rearrangement and partial degradation of tissues is initiated which leads to the liberation of proteins. Simultaneously the contents of the spinning gland are built up. It may be assumed that this secretion is poor in tryptophan, as are most silk proteins, so that there is an excess of tryptophan which is metabolized to yield ommochromes. Colour change thus appears to be secondary to the metabolic processes which are primarily governed by the necessity of preparing the organism for pupation and of providing a specialized secretion for protection. Similar but less spectacular colour changes have been observed in many Lepidoptera, but biochemical investigations are scarce. In Hestina japonica and Sasakia charonda there is a larval diapause. When entering into diapause these larvae turn brown by formation of xanthommatin in the hypodermis. Again, at the end of hibernation xanthommatin has disappeared from the integument, while rhodommatin can be detected in the gut contents (Osanai and Arai, 1962a, 1962b; Osanai, 1966a). In the Neuropteran, Chrysopa, the xanthommatin content doubles at the onset of hibernation, while biliverdin drops by about 70 per cent (Riidiger and Klose, 1970). 5.4 OMMOCHROMES

AS WASTE PRODUCTS

In a number of Lepidoptera, the meconia produced at pupal emergence have a vivid carmine colour. This is due to the presence of rhodommatin and ommatin D. Large-scale isolation of ommatins was in fact first carried out with meconia of Vanessa urticae both by Becker (1942) and Butenandt et al. (1954a). Some quantitative data are given in Table 6. These ommochromes are formed early in metamorphosis. They make their first appearance at the time when the larvae leave the food and crawl about to find a suitable place for pupation. In Vanessa urticae, orange pigment appears in the midgut wall at the time when the larva is spinning the small web to hook itself up. Twelve hours later, at the time of

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

177

TABLE 6

The quantity of ommatins in meconia of some Lepidoptera Gug per animal) Species Aporia crataegi Pyrameis atalanta Pyrameis cardui Argynnis paphia

Rhodommatin

Ommatin D

244 130 27 38

225 115 52 36

Taken from Butenandt et al. (1960b).

pupation, the gut is filled with red fluid containing about 50 pg each of rhodommatin and ommatin D (Kiibler, 1960). The histology of this process is described in detail for Ptychopoda serzata (Wolfram, 1949) and for Cerura vinula (Linzen and Buckmann, 1961). In both species, ommochromes appear first in the Malpighian tubules; they are then excreted and thus contribute to the red colour of the last faeces of the larva. Some time afterwards (but prior to any change in the hypodermis) the epithelium of the midgut is replaced. Ommochromes are formed in vacuoles of the new cells and are later released into the gut lumen. From the histological description of this process it appears that the ornmochromes are really synthesized locally within the gut cells and not taken up from the haemolymph. Yet, it is possible that precursors are absorbed from some other tissues. In Cerura, for example, pigment production is linked to profound histological changes occurring in other tissues: notably rejuvenation of the midgut and histolysis of the fat body. It is not surprising, therefore, that ommochromes are also formed when animals or tissues are maltreated SO as to disrupt normal metabolic function. In fourth-stage larvae of Cerura, the formation of ommochromes in the Malpighian tubules and gut can be induced by ligation or simply by starving the animals (Biickmann, 1959a). It is assumed that under these circumstances the animals draw on their protein reserves as an energy source, thus leading to an excess of tryptophan. This assumption is also supported by the observation that locusts produce brick-red faeces in the periods of moulting (Chauvin, 1939). Chauvin related the red colour to “acridioxanthin”. Bouthier (19 72), using chromatographic methods, identified the “acridionimatins” I and 11, and succeeded in purifying some pigment from this starting material. Locust droppings are also red, when the animals are starved. It has been noticed by several authors that dama!ge to insect tissues may result in ommochrome formation. In Drosophila, I he Malpighian tubules

178

BERNT LINZEN

are coloured red after non-specific experimental injury such as uv irradiation and osmotic shocks (Ursprung et al., 1958; Hertweck, 1960; cf. also Wessing and Bonse, 1962). In all probability this pigment is an ommochrome. Berthold (1971) noted that regenerated legs of Carausius morosus are more reddish than normal legs. She also observed more ommochrome granules in histological sections. Furthermore, if allatectomy, or a corresponding sham operation, is performed on the stick insect, a reddening of the integument anterior to the wound (in the head, and the antennae) is observed. Changes in at least three types of pigment are responsible for this effect and an increase in both xanthommatin and ommin was demonstrated by quantitative chemical analysis. In this connection, Berthold recalls a general statement by Giersberg (1928), namely that exposure of the stick insect to extreme environmental conditions, whatever these are, results in preferential synthesis of dark pigments, whereas specimens kept under optimal rearing conditions remain green. Berthold relates the incidence of ommochrome formation in the stick insect, whenever it is not under neurohormonal control, to a failure in the removal of kynurenine (e.g. by an impairment of circulation). She assumes that normally kynurenine is transported to and transaminated in the fat body. The typical excretory metabolite of tryptophan in the stick insect is kynurenic acid. In the light of the foregoing discussion, the effects of high temperature on the ommochrome content of Cerura larvae (Biickmann, 1963, 1965) are quite unexpected. Raising the temperature to 35" C, which severely harms the larvae at prolonged exposure, does not result in an increase but in a decline in hypodermal ommochrome content. The most conspicuous event is the turning red of previously brown regions of the integument. Thir effect can also be produced topically by applying small heating elements. This colour change is slowly reversed, if the animals are returned to a temperature of 20°C. In the course of further growth such larvae synthesize much more xanthommatin in their integument than control animals continuously kept at the lower temperature. Evidently, we are concerned here with both short-term and long-term effects of high temperature, which are brought about by different mechanisms. A change in the oxidation state of the hypodermal xanthommatin can also be brought about by placing Cerura larvae intermittently into an atmosphere of nitrogen or by plugging part of the spiracles. This treatment causes a slight, though not statistically significant, increase in the amount of xanthommatin (Buckmann, 1965). Changes in temperature and atmospheric composition have been known for a long time to affect development of insect pigments (cf. Biedermann, 1914), but very little is known about the physiological and biochemical mechanisms involved. In conclusion, ommochrome formation is frequently a consequence of

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

179

cellular injury or impairment of normal metabolism. It is without doubt a means of “metabolic excretion” of tryptophan and is comparable in this respect to the production of alkaloids in plant metabolism. One might ask whether there is any selective value in forming an end product without really removing it along with the excreta. Harmsen (1966), in his discussion of the excretory role of pteridines in insects, requires that two criteria (among others) should be fulfilled for a pigment t o be regarded as an excretory or “dry storage” product: ( 1 ) synthesis of the pigment should be continual during development and not take place shortly before it functions as a colouring matter; ( 2 ) the pigment should be produced in excess of the demand (if colouration alone is considered). The first requirement does not appear to be justified in insects as development itself is not continual (i.e. not a gradual change of size and structure). The excretory role of pigment synthesis may appear orily in periods of altered protein metabolism, as in the larval moults or during pupal metamorphosis. This condition would certainly apply to ommochrorne synthesis. Harmsen’s second requirement is met by the observation that hypodermal ommochromes are frequently occluded by melanin in the overlying cuticle, and thus never required for pigmentation. A topic not considered in Harmsen’s paper is the possible advantage of “local excretion”. Two aspects are pertinent to ihis problem. First, the absolute amount of tryptophan to be excreted is usually low, as a result of the low percentage of tryptophan in proteins. Secondly, as reported above, most ommochromes are insoluble at physiological pH and may form concretions of high density. In contrast, all precursors of ommochromes are more or less readily soluble. In addition, the function of tryptophan both as an essential amino acid (which i s conserved for protein synthesis within the cell) and as a possibly harmful substance, must be contemplated. It can thus be conceived that “excretion in situ” of tryptophan, by transforming it into an insoluble metabolite, could be more efficient than removal by diffusion. This would be especially the case if circulation of the haemolymph and its clearance by Malpighian tubules Gr other excretory tissues were slow. This reasoning could also apply to cells not in direct contact with streaming haemolymph because they are tucked away in tissue folds or because of other interposing tissues.

6 Enzymes involved in the kynurenine pathway The study of enzymes involved in the de,gradation of tryptophan has lagged behind the work on corresponding mammalian or microbial enzymes by five to ten years. The first to be detected in insect h6mogenates were kynurenine formamidase (Glassman, 1956) and lrynureninase (Inagami, 1955, 1958). While these have not been studied in detail in insects, more

BERNT LINZEN

180

attention has been paid to tryptophan oxygenase because of the distinction between suppressible and unsuppressible vermilion mutants in Drosophila melanogaster. 6.1

TRYPTOPHAN OXYCENASE (EC 1.13.1.12)

This enzyme, already well known from vertebrate and microbial sources, was first detected by Egelhaaf (1958) in Ephestia kiihniella and by Baglioni (1959) in Drosophila. It is a soluble enzyme which may have a Fe porphyrin prosthetic group. While Marzluf (1965a) could not restore the activity of dialysed tryptophan oxygenase nor increase the activity at any stage of purification by adding haematin, Baillie and Chovnick (1971) and Schartau (unpublished) were able to stimulate enzyme activity several-fold by the addition of methemoglobin. Nawa (1970) observed dependence of the activity of the purified enzyme on added methylene blue. In insect extracts, tryptophan oxygenase activity is usually measured by means of the Bratton-Marshall reaction of the product, kynurenine (Bratton and Marshall, 1939). It is tacitly assumed that the formylkynurenine formed during the incubation is hydrolysed by excess kynurenine formamidase or spontaneously during the incubation or after addition of trichloroacetic acid. The Bratton-Marshall reaction, however, is not entirely specific. Pinamonti and Petris (1966) pointed out that the degradation of ommochromes in the weakly alkaline incubation medium (Butenandt et al., 1960b) leads to Bratton-Marshall positive products. This is a serious interference in the assay of tryptophan oxygenase in many arthropod tissues, but can be overcome by suitable controls. In addition, other reactions might contribute to errors. This has been observed in Bombyx tissues by using two different assays (Linzen, 1971b). In such cases the kynurenine formed may be separated by paper chromatography or high-voltage electrophoresis and measured by paper fluorimetry (Egelhaaf, 1958, 1963a; Linzen, 1971b). By incubating directly on chromatography paper, Egelhaaf (1963a) could demonstrate the activity of the enzyme in single fat body lobules or ovarioles. Partial purification of the enzyme has been achieved by a number of workers. Hiraga (1964) separated tryptophan oxygenase from kynurenine formamidase by gradient elution from DEAE-cellulose. Chromatography on DEAE-cellulose was also the most efficient step in the work of Marzluf (1965a), Baillie and Chovnick (1971), Nawa (1970), and Schartau (unpublished), who purified tryptophan oxygenase about 16-, 65-, 80- and 200-fold, respectively. In Baillie’s and Chovnick’s study, the activity of the purified enzyme became almost entirely dependent on added methemoglobin. The molecular weight of the Drosophila enzyme was estimated at

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IIV INSECTS

181

150 000 Daltons (Baillie and Chovnick, 1971) and that of the Phormia enzyme at 120 000 Daltons (Schartau, unpublished). This is probably an oligomer composed of several subunits and which would be expected to exhibit allosteric properties. Baillie and Chovnick indeed observed a sigmoidal dependence of reaction velocity on substrate concentration which was abolished by preincubation with cu-methyl-tryptophan. The same phenomenon had been described for microbial tryptophan oxygenase (Feigelson and Maeno, 1967). Probably the “superadditivity effect”, observed in Ephestia a’/a by Egelhaaf and Caspari (1960) and clearly brought into focus by Tartof‘s (1969) investigation of the vermilion suppressor i n Drosophila, is related to subunit interaction. If wild-type and mutant extracts are mixed, the reaction rate is higher than the sum of the rates in each single extract. In the vermilion mutants superadditivity is correlated with suppressibility. Furthermore, the material responsible for the effect is thermolabile; this is precipitated by ammonium sulphate and, when chromatographed;eluted at the same position as the wild-type enzyme (Baillie and Chovnick, 1971). Tartof‘s conclusion is that the subunit of the mutant enzyme contains an intact catalytic site but is unable to associate, association supposedly being a prerequisite for activity. Formation of hybrid oligomers, however, which would occur after mixing wild-type and mutant extracts, should result in activation of the catalytic sites of both wild-type and mutant enzyme subunits. Some data on tryptophan oxygenase are summarized in Table 7. The pH-optimum is usually sharp and on the alkaline side. Substrate inhibition has been observed several times but is not mentioned when higher substrate concentrations have not been tested. The enzyme is inhibited by a number of agents which are listed in Table 8. Of particular interest is inhibition by some naturally occurring pteridines (Ghosh and Forrest, 1967b). Though only two compounds are listed in the Table, it may be assumed that other pteridines (folic acid, biopterin, isosepiapterin and isoxanthopterin) will also affect the insect enzyme, since they exert strong inhibition on the rat liver enzyme. It was argued that in a group of white mutants of Drosophila, failure to reduce a pteridine precursor could lead to a double effect on tryptophan degradation: accumulation of a pteridine inhibitory on tryptophan oxygenase, and lack of a reduced cofactor supposedly required for kynurenine hydroxylation. Although up to now thLere is not much evidence to prove this hypothesis, some sort of interaction between tryptophan and pteridine metabolism clearly exists as shown by the observations of the Rizkis in Drosophila (Rizki, 1964; Rizki and Rizlki, 1964: altered pattern of kynurenine formation in larval fat bodies of the rosy i n d sepia mutants) and of Kiihn and his collaborators in Ephestia (Kiihn, 1956). The specific activity of the enzyme in total hoinogenates of insects is of

BERNT LINZEN

182

TABLE Tryptophan oxygenase in insects-selected data on

Species

Drosophila melanogaster

Strain and stage

Oregon-R, flies Oregon-R, all stages Oregon-R, larvae Oregon-R, flies Oregon-R, flies

Substrate inhibition

pH optimum in uitro

Kln mM litre-'

-

-

-

7.4 7.4

1 1.5

-

(3) * *

-

-

8.0*

Wild-type, flies

-

Ephestia kiihniella

Various stages

Sch isto cerca gregaria Habro bracon jugla nd is Bombyx mori

Adults

7.4

0.6

Wild-type 33, various stages rb mutant

-

2.6

8.5

-

Phormia terraenouae

All stages

8.25

Gryllus bimaculatus

Ultimate larval

8.5-9 .O

2.6

Yes

3.5

Yes

* For suppressed uk.

* * Sigmoidal velocity curve, value given for half-maximal velocity; after preincubation with a-methyl-tryptophan the K , is about 1.5 x lo-' M litre-'. 05 It has to be assumed that optimal assay conditions were frequently not attained. Only a small part of the available data are included in this column; most values have been rounded or are given as order of magnitude. Activities are based o n different entities.

THE TRYPTOPHAN

+

183

OMMOCIiROME PATHWAY I N INSECTS

7 properties and occurrence of the enzyme Optimal mM litre-'

Stimulated by hematin?

Activity in whole homogeniites or tissues5 5

Inducibility Reference?

-

-

133 nmol g-' h-' ,whole flies

-

a

-

-

600 nmol g-' h-' ,whole flies

-

b

-

no

2.3 x lo-* nmol per larva per hour (arbitrary units given) 300 nmol g-' h-", whole animals

yes Yes

d

-

e

-

Yes

10 nmol mg-' protein h-' , whole animals

-

f

-

-

30 pmol mg-' protein min-' , whole animal 140 pmol mg-' protein min-' , larva fat body 40 pmol mg-' protein min-' , spin gland 8 nmol mg-' protein h-' , fat body 4 nmol mg-' protein h-', whole animal

no

g

-

-

6 2.4

8 .O

2.5-5

yes

-

60 nmol mg-' protein h-' , fat body$ 3 nmol mg-' protein h-' ,gut 12 nmol mg-' protein h-' , ovaries 12 nmol mg-' protein h-' , testes 30 nmol mg-' protein h-' ,developing wing high: fat body, testes, Malpighian tubes medium: gut, flight muscle low: ovaries, larval "residual body" 1 nmol mg-' protein min-' , fat body

C

-

h

no

I

-

i

no

k

-

I

$ Much depending on stage, maximal activities are given. In almost all studies the incubation temperature was 37OC. f a . Baglioni (1960); 6. Kaufman (1962); c. Rizki and Rizki (1963); d. Marzluf (1965a. 1965b); e. Tartof (1969); f. Baillie and Chovnick (1971); g. Egelhaaf (1958, 1963a); h. Pinamonti and Petris (1966); i Leibenguth (1967a)pj. Linzen (1971b); k. Schartau (unpublished); I. Tiedt (1971). AIP-7

BERNT LINZEN

184 TABLE 8

Effects of various inhibitors on insect tryptophan oxygenase

Inhibitor cu++ Cyanide Fluoride Azide Hydroxylamine Phenylthiourea Sodium diethyldithiocarbaminate D-tryptophan 5-Methyl-D &tryptophan 2-Amino-4-hydroxy-6-hydroxy methyl-pteridine 2-Amino-4-hydroxy-6-carboxypt eridine

Concentration Inhibition M litre-' % 2.5 4 5 2 5 4 3 7.5 2 5 5 6.5 6

x 10-~ x 10-5 x 10-~

x ~ o x10-4

x ~ o x 10-~ x 10-~ x x 10-~

x 10-~ x 10-~

68 * 47 21 ~ 32 49 ~ 70 25 24 50 39 51 73

x ~ O - ~

6

Species

Reference?

Drosophila Ephestio

b a a a a b

Drosophila Ephestia Drosophila

a a b b b c c

? a . Egelhaaf (1963a); b. Marzluf (1965a, 1965b); c. Ghosh and Forrest (1967b). (From the data provided by these authors inhibitor concentrations were selected for which the reported inhibition was nearest to 50 per cent. The species are Drosophila melanogaster and Ephestia kiihniella.) * At a protein concentration of 6.6 mg ml-' .

the same order of magnitude as the activity in vertebrate liver. A precise comparison is impeded by variations of incubation conditions and of calculating the data. Within a given insect species, considerable differences exist between various strains: in Drosophilu melanoguster the Sevelen strain has higher activity than the Oregon-R and the Pavia strains; the ratios are about 100 : 77 : 48 (Baglioni, 1960; Kaufman, 1962). If tissues are dissected and assayed separately, the specific activity in some of them is distinctly higher than in vertebrate liver. A striking difference of localization is provided by the occurrence of tryptophan oxygenase in a variety of insect tissues, while it is restricted to liver in vertebrates. While Rizki (1964) working with Drosophilu and Pinamonti and Petris (1966) studying Schistocercu, both believed that tryptophan oxygenase was restricted to fat bodies, it has been reported by Kaufman (1962) that Malpighian tubules of Drosophilu also contained the enzyme. Egelhaaf's meticulous studies on Ephestia (1958, 196Sa) extended the domain of the enzyme to gonads, spinning gland, gut, and epidermis.

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

185

Work in the author’s laboratory (Linzen, 1971b; Tiedt, 1971; Linzen and Schartau, in preparation) leads to the conclusion that the distribution of the enzyme among tissues is almost random. It may be found in almost any tissue, and there is no evident rule as to the specific activity characteristic of any particular tissue, or of specific activity proportions. For example, in Phormia the specific activity is highest in testes, and there is high activity in both larval and imaginal Malpighian tubules, w’hile in Bombyx mori rb it is at best “medium” in the testes and absent fr3m the Malpighian tubules. The ontogenetic pattern of tryptophan oxygenase in Bombyx will be discussed below. Evidence on inducibility of tryptophan oxygimase in insects is scarce and conflicting. The Rizkis (Rizki, 1963, 1964; Rizki and Rizki, 1963, 1964) have shown that upon addition of tryptophan to the growth medium of Drosophila larvae a large amount of kynurenine is formed in the fat body. Kynurenine is normally localized in the anterior lobes, but extends throughout the entire fat body after tryptophan feeding. The activity of tryptophan oxygenase was markedly raised iii extracts of those larvae. These results were confirmed by Marzluf (19155b) for a wild-type and a suppressed vermilion strain. On the other hand, in Ephestia (Egelhaaf, 1963), in the parasitic wasp, Habrobracon (Leibenguth, 1967), and in Phormia flies (Schartau, unpublished), tryptophan oxygenase is not inducible by feeding or by injecting tryptophan In this connection one should also focus attention on the relation of enzyme activity to gene dosage. In Drosophila the gene v + responsible for tryptophan oxygenase is located on the X chrornosome so that in males one would expect half the activity in females, if there were no dosage compensation. However, in males the activity equals the activity in females (Kaufman, 1962). If gene dosage is increased, the response of males is about twice the response of females (Baillie and Chovnick, 1971), suggesting a rather complicated regulatory mechanism (for discussion compare Seecof et al., 1969). In v+/v heterozygotes the activity is always higher than 50 per cent of v+/v+ activity (depending on the particular v mutant) but does not exceed 80 per cent. This, may be interpreted as an in vivo superadditivity effect. In Ephestia it was demonstrated both by direct enzyme assay (Egalhaaf and Caspari, 1960) and in vivo by tryptophan loading (Egelhaaf, 1963a) that tryptophan oxygenase activity depends on the number of a + alleles present. In vitro there is a marked superadditivity effect; however, no data are available to relate in vivo and in vitro activities, as in Drosophila. Investigations of Drosophila tryptophan oxygenase were usually stimulated by a search for the mechanism cif suppression of vermilion mutants. Vermilion mutants are practically indistinguishable by their phenotypes but fall into two classes when two sets of experiments are

186

BERNT LINZEN

employed. In the presence of a suppressor mutant part of the u mutants ( u s ) are suppressed, while the others are unaffected (u”) (Green, 1952). Furthermore, if u s larvae are subjected to partial starvation the adults will have pigmented eyes (‘I’atum and Beadle, 1939; Green, 1954). Suppressible to u s “ . mutants are u ( u l ), u’, u k , unsuppressible mutants u36f, u 4 & , There is a series of suppressor mutations affecting vermilion, abbreviated su(s), the mutant most widely used for study being su(s)’. For details see Lindsley and Grell (1968) and references cited therein, and Baillie and Chovnick (1971). ’Tryptophan oxygenase activity in various mutants is compared in Table 9. Activity is only partially restored upon introduction of su(s) but will suffice ,for normal production of pigment precursor. The data also indicate that the “unsuppressible” mutant u36f will produce a small but detectable amount of additional activity if combined with su(s)’. This is substantiated by Kizki (1 964) who observed the appearance of kynurenine fluorescence in fat body regions 1 and 2 of u36f .w(s)’ larvae after feeding tryptophan. In u 1 su(s)’ the fluorescence extends further back t o regions 3 and 4. Rizki states that “the difference between the unsuppressible u36f allele and the suppressible vermilion alleles is one of a quantitative nature rather than absolute”. Early hypotheses implied that suppressible mutants produced functional tryptophan oxygenase which, however, was blocked by an inhibitor present

TABLE 9

Relative activities of tryptophan oxygenase in various combinations of vermilion and s u ( s ) alleles

Strain Ore-R bw u’bw u36fb~ u‘

x

U36f

u ‘su(s)

=

u Isu (s) 3 u ISU (S)* -v-pr

u 36 f su (s) 2

Percentage enzyme activity as estimated by Baglioni (1960) Kaufman (1962) Tartof (1969) 100 100 11 7 7 20 17 14 6

100 66,180* 2 1

20

1.5

100 1 1

9 10 4

It is emphasized that enzymatic activity is highly dependent on genetic background and therefore varies in different strains. * Two different strains.

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

I a7

in the mutant, or else had to be activated in some way. Marzluf (1965a, 1965b) compared the properties of tryptophan oxygenase from wild type and suppressed vermilion flies, but did not find any difference in K,, pH dependence, energy of activation, percentage of inhibition by varying Cu++ concentration, and rate of inactivation. Neithcr did he find evidence for any easily removable inhibitor, and suggested that su(s) was a regulatory gene which induced the us mutants to produce tryptophan oxygenase at a low rate. Tartof (1969) later discovered that there was a difference between u+ and suppressed u’ tryptophan oxygenase on the one hand, and oxygenase of uk on the other (the pH of maximal activity being shifted from 7.4 to 8.0 in the latter). He also found a strong correlation between suppressibility of the u alleles and the superadditivity effect exerted by their products on normal tryptophan oxygenase. Thus, us mutants are assumed to produce a potentially functional enzyme which, in a given cellular environment, does not attain the conformation necessary for function. Tartof concluded that su(s) is not an informational suppressor, bu: an indirect or metabolic suppressor (see Gorini and Beckwith, 1966). su(S) should be responsible for an altered cellular environment which would allow the us enzyme to become active (e.g. by dimerization). This is consistent with the recessive character of su(s). Similar conclusions have been drawn by Baillie and Chovnick (1971). Direct evidence for such a type of mechanism was recently brought forward in a most exciting paper (Jacobson, 15171). Jacobson showed that if a u homogenate was treated with ribonuckase T I , a high increase in tryptophan oxygenase activity was observed. In wild-type homogenates the activity was not affected. Conversely, if ‘‘activated” (i.e. RNAase T , digested) tryptophan oxygenase was mixed with tRNA prepared from wild-type flies, strong inhibition resulted. The tRNA was further fractionated and tested against activated u 3xygenase. The inhibitory fraction was identified as a specific iso-accepting form of tyrosine tRNA (Fig. 12). Jacobson assumed that tryptophan oxygenase could associate with tRNATyr, but that in the case of the vermilion enzyme such association led t o inhibition. Logically su(s) is expected to affect the inhibitory tRNA. This has been shown in recent work by Twardzik et al. (1971). The tRNA’Y‘ of wild-type Drosophila could be resolved into three (two major and one minor) peaks; if homozygous su(s)’ flies were worked up in the same manner the second of these was missing. Three other species of tRNA were also examined and found unaffected by JU(S)’. The authors showed that both su(s)’ and the gene responsible for the altexd elution profile are located at the same position on the left end of the X chromosome. The fact that su(s)’ is recessive, and that it does not alter the total amdunt of

BERNT LINZEN

188

I

-

NaCl concentration ( M I

0.4 8

4

p"

( b)

0.3 6-

0.2 4-

2-

0.2

0.3

0.5

0.4

NaH, PO, concentrotion

(M)

Fig. 12. (a) Fractionation of Drorophilu tRNh by reverse-phase chromatography and effect of fractions on tryptophan oxygenase activity (0-0-0). (b) Resolution of the active tRNA peaks by chromatography on hydroxylapatite. Three peaks are obtained, two of which can be charged with tyrosine ( 0 . . . . . . 0 ) . The tRNA eluted last strongly inhibits tryptophan oxygenase. (From Jacobson, 1 9 7 1 . )

tRNATY', but only alters the quantitative distribution among the two major iso-accepting forms, is a strong argument for the view that the suppressor mutation does not affect the structural gene of rRNATY' (of which a dozen copies might be present) but the gene coding for an enzyme (e.g. a methylase) by which tRNATY' is altered secondarily. Genetically modified tRNA has been shown in many cases to be the cause of suppression in bacteria, acting by altered coding specificity. It is intriguing, therefore, t o find tRNA as a mediator of suppression in Urosophifa. However, the mechanism is entirely different for the effect is indirect, and occurs through the removal of a species of molecules (tRNATy') which would otherwise react with a cytoplasmic enzyme (try oxygenase). The activity of this enzyme is not affected in the wild type, but it is impaired by the interaction if it is itself altered by t h e us mutation.

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

189

A number of puzzling observations still await clarification. Tartof (1969), for example, found that su(s)’ influenced the measurable tryptophan oxygenase activity in v’/v heterozygotes. In two of these crosses the activity was rendered less than additive (only 35 per cent) by the presence of the suppressor. This indicates that interaction with tRNATY‘ is not restricted to the suppressible vermilion enzyme, but that it is a normal event determining the physical state of tryptophan oxygenase.

6.2 KYNURENINE

FORMAM~DASE(ARYL-FORMYLAMINE

AMIDOHYDROLASE EC 3.5.1.9)

The earliest report on this enzyme in insects appears to be the one by. Glassman (1956). Later, Hiraga (1964) showed that the enzyme could be chromatographed on DEAE-cellulose. Recently some additional data have been gathered in the author’s laboratory. The hydrolysis of formylkynurenine is conveniently measured by recording absorbance at 360 nm (Mehler and Knox, 1950). Table 10 presents some of the results. In addition, it must be mentioned that the enzyme fiom Drosophila does not hydrolyse formylanthranilic acid (Glassman, 1956). There are two characteristic observations: the activity is found in almost any tissue and exceeds the activity of tryptophan oxygenase by one or two orders of magnitude. The latter fact is also known for vertebrate liver, and one wonders whether the enzyme might not perform some other function in addition to splitting formyl-kynurenine. In passing, it should be noted that kynurenine formamidase extracted from rat liver or rat skin is strongly inhibited by organophosphorus compounds (JansCri et al., 1969). While the enzyme has recently been purified 1800-fold from rat liver and obtained in an apparently homogenous state (Shinohara and Ishiguro, 1970), no corresponding attempt has been made with insect material. To the author’s knowledge no mutant exists which lacks kynurenine formamidase activity. Such mutants are, however, riot likely to be detected by phenotype, because formylkynurenine undergoes spontaneous hydrolysis. 6.3 KYNURENINE-3-HYDROXYLASE (EC 1.14.1.2) Attempts t o demonstrate insect kynurenine hydroxylase in uitro were for a long time unsuccessful, in spite of a marked accuinulation of 3-hydroxykynurenine in certain developmental stages. In retrospect it appears that the main reason for failure was the fact that hynurenine hydroxylase is inactivated by light, a phenomenon first detected in Staudhger’s laboratory (Mayer et al., 1968) and probably due to the participation of FAD in the reaction.

TABLE 10 Kynurenine formamidase in insects: data on properties and occurrence of the enzyme*

Species

Stage Optimal pH investigated

Km M litre-'

Inhibitors

D. melanogaster

Flies

7.3

3.1 x 1 0 - ~

NaHS03

D. uwilis Cry Ilus b ima cula tu s

Flies Last larva1

7.3 9.0

3.1 x 1 0 - ~ 1.5 x 1 0 - ~

NaHSO 3 Substrate

Phormia terraenovae

All stages

7.3

1.3 x 1 0 - ~

Bombyx mori

All stages

9.25

Substrate L-kynurenine Substrate

Activity and localization** Arbitrary units, no striking differences between strains

Reference+ a

a

Fat body: 4 nmol min-' mg-' protein Malpighian tubes: 15 nmol min-' mg-' protein Gut: 1-2 nmol min-' mg-' protein Testes: 4 nmol min-' mg-' protein Ovaries: no activity protein Larvae: 30 nmol min-' Flies: 4 nmol min-' mgYgprotein High: testes Medium: ovaries, Malpighian tubes, eyes, gut. fat body Low: silk gland, developing wings All stage dependent

-'

* See also Hiraga (19 64).

** In homogenates of whole animals or isolated tissues. t a. Glassman (1956); b. Tiedt (1971); c. Linzen and Schartau (unpublished); d.Linzen

(unpublished).

b

c

d

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

191

Some indirect observations on thc activity of the enzyme were made by Egelhaaf (1963a, 1963b), by injecting kynurenine into Ephestia moths and measuring the appearance of the product. A linear increase of about 3 pg h-’mg-’N was observed during at least 90 min. Ovaries (mainly the upper parts) and fat body were considered to be especially active. In 1967 Ghosh and Forrest and Linzen and Hertel simultaneously published in uitro assays of kynurenine hydroxylase in Drosophila and Calliphora respectively. Later investigations were concerned with Bombyx mori rb (Linzen and Hendrichs-Hertel, 1970), Schistocerca gregaria (Pinamonti et al., 1970-71), and Apis mellifica (Dustmann, personal communication). 3-Hydroxy-kynurenine was measured either by Inagami’s nitrous acid method (Ghosh and Forrest), by oxidation to xanthommatin (author’s laboratory) or by two-dimensional chi-omatographic isolation. Hendrichs-Hertel and Linzen (1969) proved enzymatic formation of 3-hydroxy-kynurenine independently by incubating tritiated kynurenine and isolating the labelled reaction product. Evidently the optimal conditions for assay of the enzyme have not yet been awertained. Both in the author’s laboratory and that of the Italian group a preparation given by Ginoulhiac et al. ( 962) has been adopted with slight modifications, while Ghosh and Forrest (1967a) developed a medium of their own. NADPH is required for normal activity, but there is also a slight effect when NADH is added. Ghosh and Forrest (1967) also speculate on the participation of a pteridine cofactor. In the author’s view this is doubtful; in mammalian preparations a pteridine has not been found as part of the system, while the participaticn of FAD is now a well-established fact (Okamoto and Hayaishi, 1967, Horn et al.. 1971). The optimal substrate concentration was 4 mM in Ghosh and Forrest’s work, while in Calliphora preparations there appeared strong substrate inhibition at this level (Hendrichs-Hertel and Linzen, 1969). The K , has been determined only for the enzyme of Schistocerca (Pinamonti et al., 1970-71) and comes to 4 x lo-’ mol litre-’. The optimal pH is slightly above neutrality. Cysteine and azide ions appear to be stimulatory, except for the bee enzyme which is not stimulated by azide, and even inhibited by cysteine (Dustmann, personal communication). While cyanide inhibits oxidation of NADPH by the fraction of “light” mitochondria (Mayer and Staudinger, 1967), and might therefore increase the observed activity of kynurenine hydroxylase indirectly, it was shown to be inhibitory in recent experiments by Dustmann (personal communication) and Linzen (unpublished). Rat liver kynurenine hydroxylase is inhibited b y cyanide only in a purified state (Horn et al., 1971). Dustmann reported strong inhibition of the bee enzyme by xanthommatin, so that wild-type eye extracts become completely inactive. Evidently compartmeritation is an important factor in pigment biosynthesis.

192

BERNT LINZEN

In mammals (Okamoto ~t al., 1967) and in Neurosporu (Cassady and Wagner, 1971) kynurenine hydroxylase is clearly localized in the outer mitochondrial membrane and for this reason is a useful marker enzyme of this fraction. It is probably also restricted to mitochondria in insects (Ghosh and Forrest, 1967, and Pinamonti et al., 1971, traced the activity to the mitochondrial membrane fraction), despite the contradictory findings of Hendrichs-Hertel and Linzen (1969) who consistently detected part of the activity in the soluble fraction. It may be argued that in insects even very careful homogenization (with a teflon pestle) might damage mitochondria because of the grinding action of cuticular particles. It might be further supposed that in some species the outer mitochondrial membrane could be extremely labile. Such a situation has been encountered by Graszynski (1970) in crayfish. Attention should also be drawn to electron microscopic studies by Wessing (1962, 1963) on the fine structure of Drosophila Malipighian tubules. The endoplasmic reticulum of these cells forms a system of channels, which occasionally widen to form “storage ampoules”. These ampoules, in certain stages, are crammed with solid 3-hydroxy-kynurenine. Wessing proposed that these structures, which are typically found in the neighbourhood of Golgi apparatus, might also be a site of kynurenine hydroxylation. Direct evidence to support this hypothesis is, however, still lacking. As in the case of tryptophan oxygenase, kynurenine hydraxylase cannot be assigned to a particular organ or tissue. The enzyme has been found in many types of tissue (nervous tissue has not been examined). In Rombyx the Malpighian tubules show a high peak of activity two days prior to spinning, but fat body, ovaries, and eyes are also able to hydroxylate kynurenine (see also p. 214 for more detailed treatment). In Schistocerca the eyes are the main source of the enzyme besides some activity in the integument (Pinamonti et al.. 1970-71) in bees the eyes are the only tissue active (Dustmann, personal Eommunication). In Calliphora the Malpighian tubules are particularly active, while the data obtained for fat body are at the limit of detection. In eye discs, kynurenine hydroxylase should certainly be present, since Danneel had demonstrated in 1941 that explanted heads of Drosophila u would readily form ommochrome if kept in a solution containing kynurenine. Horikawa (1958) has made similar, more elaborate experiments with eye-antenna1 discs in culture. HendrichsHertel and Linzen (1969) failed to detect the enzyme in Calliphora eye tissue, possibly due to the inhibition of the enzyme by xanthommatin mentioned above. Both in Ephestia and in Drosophila, kynurenine hydroxylase is a constitutive enzyme, as evident from the experiments of Egelhaaf, Danneel, and Horikawa (all cited above), with mutants blocked at the tryptophan

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY

IN INSECTS

193

oxygenase step. The same holds for the bee mutant snow (Dustmann, personal communication). 6.4

KYNURENINASE AND KYNURENINE TRANSAMINASE (EC 3.7.1.3; 2.6.1.7)

Kynureninase (EC 3.7.1.3) is situated in the main pathway of tryptophan degradation in vertebrates, while kynurenine transaminase (EC 2.6.1.7) catalyses a side reaction. To the knowledge of the author, the only study relating t o the former enzyme in insect material is Inagami's (1958). In Bombyx mori, the major metabolites of tryptophan are anthranilic acid and its conjugates. These are lacking in the mutant rb where 3-hydroxykynurenine is accumulated. Inagami proved that the mutation to rb is associated with a loss of kynureninase activity. With wild-type homogenates he could demonstrate the enzyme in gut wall (larvae) and in whole pupae where both kynurenine and 3-hydroxy-kynurenine were cleaved. The enzyme precipitated between 35 and 5 5 per cent saturation with ammonium sulphate and remained active after prolonged dialysis even if pyridoxal phosphate was omitted from the redction mixture. Inagami concluded that the cofactor is tightly bound to the enzyme. Kynurenine transaminasr has been studied during development of the parasitic wasp, Habrobracon juglandis (Leibenguth, 1967a), and more closely in Schistocerca gregaria (Pinamonti et al., 1970). In both cases distinct enzymes, specific for either kynurenine or 3-hydroxy-kynurenine, were not found. For the locust enzyme the K, for the two substrates is 5.4 x and 1.5 x l o d 3 mol litre-' respectively. Pyruvate and oxaloacetate are better amino group acceptors than a-kstoglutarate. The K, for pyridoxal phosphate was found to be 1.6 x mol litre-'. Optimal activity occurs at pH 8. The only tissue active is fat body.

7 Ommochrome biosynthesis The oxidation of ortho-aminophenols can be rasily performed in the laboratory. Nevertheless, phenoxazinones will not result in every case; extensive studies of model reactions (Butenandt et al.. 1954d, 1957a, 1957b, 1957c) have shown that different ring systems can be formed from depending 3-hydroxy-anthranilic acid or 2-amino-3-hydroxy-acetophenone, on the conditions of oxidation. Enzymes catalysing the formation of phenoxazinones should, therefore, be specific not only for a particular substrate, but also for the type of reaction. Such enzymes were first demonstrated in microbial and plant material (Weissbach and Katz, 1961; Katz and Weissbach, 1962; Nair and Vaidyanathan, 1964; Nair and Vining,

194

BERNT LINZEN

1964-to cite a small selection of papers). FMN was reported to act as a cofactor in some of these systems. Systems oxidizing 3-hydroxy-anthranilic acid to cinnabarinic acid were also obtained from vertebrate liver (Joshi and Brown, 1959; Gutmann and Nagasawa, 1959; Morgan and Weimorts, 1964; Subba Rao et al., 1965), but there is no evidence that these systems have a biological role. Since ortho-aminophenols are so easily oxidized, reports on enzymatic formation of xanthommatin should be viewed with some caution. The first of these studies (Butenandt et al., 1956) was started as a re-examination of Inagami’s (1954b) finding of a “red melanin” which arose during the incubation of a mixture of DOPA and 3-hydroxy-kynurenine with tyrosinase. The formation of a “mixed” red melanin was not confirmed. However, it turned out that Calliphora tyrosinase does not oxidize 3-hydroxy-kynurenine at neutral pH, but brings about the formation of xanthommatin, in the presence of DOPA in the incubation medium. Thus, DOPA-quinone as the primary oxidation product serves in lieu of a cofactor of phenoxazinone synthesis. Although the frequent association of cuticular , melanin with hypodermal ommochromes (Dustmann, 1964) is suggestive of a similar mechanism in vivo, there are many arguments against this possibility. First, while ommochromes and melanins are often found in close proximity, they do not really occur jointly; neither are they intermixed. Secondly, a correlation between tyrosinase activity and ommochrome synthesis could not be established in Hestina larvae or in silkworm diapause eggs (Osanai, 1968a, 1968b). Thirdly, mutants devoid of xanthommatin still have an active tyrosinase (Phillips et al., 1970). The oxidation of 3-hydroxy-kynurenine by tyrosinase from three sources other than insects has recently been employed for micropreparative synthesis of xanthommatin (de Antoni et al., 1970). These enzymes acted directly on the substrate, without mediation by DOPA. Similar objections may be raised against the direct participation of the cytochrome system in xanthommatin synthesis. The formation of phenoxazinones, as observed by Joshi and Brown (1959), and Gutmann and Nagasawa (1959), must again be regarded (as in the case of the tyrosinase-DOPA system) as an unspecific oxidation by a system of high oxidation potential. It was also reasoned by Osanai (1967) that diapausing eggs of the silkworm are practically devoid of cytochrome c. A model system of a different type, by which phenoxazinones may be generated, was recently described by Ishiguro et al. (1971a). They showed that 3-hydroxy-anthranilic acid is oxidized to cinnabarinic acid by haemoglobin in the presence of Mg++.Interestingly, the magnesium-stimulated oxidation is fairly specific, as 3-hydroxy-kynurenine and DOPA are not oxidized at all, and o-aminophenol is oxidized at a reduced rate. Mention must also be made of the work of Rohner and Wolsky (Rohner, 1959; Wolsky, 1960)

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY I N INSECTS

195

who observed inhibition of pigment synthesis after exposing developing insects t o an oxygen-carbon monoxide atmosphere. Phillips and Forrest (1970) have recently reported the formation of phenoxazinones by an extract of Drosophila pupae. 3-Hydroxy-anthranilic acid was converted three times as fast as 3-hydroxy-kynurenine. The enzyme could be separated from tyrosinase activity and activated by heat treatment and passage over Sephadex G-50. It would be interesting to know whether this enzyme is localized in the eye anlagen. Surprisingly, the active principle is also present in mutants unablc to convert 3-hydroxykynurenine t o xanthommatin. Rhodommatin and ommatin D might arise from xanthommatin by direct glycosylation and esterification, respectively. In vitro evidence of these transformations is still lacking. Kubler (1960) injected labelled xanthommatin (prepared by biosynthesis in Callzphora) into late larvae o f Argynnis paphia and isolated the two substituted ommatins by repeated precipitation and two-dimensional paper chromatography. Ilhodommatin was clearly labelled, but an unexpected result was the complete absence of radioactivity from ommatin D. Actually, this supports the assumption that xanthommatin is glucosylated directly, for if the radioactivity in rhodommatin originated from 3-hydroxy-kynurenine (formed by decomposition of xanthommatin) one would also have expected radioactive labelling of ommatin D. It is curious, however, that neither at the onset of excretory pigment formation in Malpighian tubules and gut wall, nor later, can xanthommatin be detected among the ommatins present (Kubler, 1960). Very little is known about the biosynthesis of ommins and ommidins. Since both contain a sulphur atom, the origin of this posed the first problem. Attempts t o clarify this by incorporation of labelled sulphur are complicated by the strong tendency of ommins to retain small amounts of protein which might label more strongly. 1,in:cen ( 1 970) undertook the isolation and purification of “pigment IV” which still contains the sulphur atom. He demonstrated the incorporation of 35S from methionine and cysteine, sulphate and sulphide giving negative results. As the elementary analysis of “pigment IV” and dihydro-xanthonimatin differ by one atom each of carbon, nitrogen, and sulphur (in favour of the former) the incorporation of thiocyanate was tested recently. In these experiments no radioactivity was recovered in “pigment IV”. Unless it is established whether “pigment IV” is an artifact produced during the vigorous hydrolysis of ommins, or part of the original ommin molecule, all incorporation studies will be subject to some uncertainty. In developing insect eyes, ommochromes appear in sequence. In Ephestia and Ptychopoda a striking observation has been made-(Kuhn, 1960, 1963; see also Muth, 1967): xanthommatin is the first ommochrome t o be detected, followed by “ommochrome 11” and “ommochrome I” and,

BERNT LINZEN

196

finally, by ommin A (Fig. 13). This is exactly the series of decreasing mobility of the ommochromes in paper chromatography which is thought to correspond to the increasing complexity (in former terms, increasing degree of polymerization) of the ommochromes. The finding of this order is highly suggestive of a precursor role played by the “simpler” ommochromes in the biosynthesis of the more complex ones. De Almeida (1968), on the basis of extracting ommochromes for different lengths of time, speculated that ommin A was deposited later on the matrix of the pigment granule than the other ommochromes. Direct evidence of the conversion of one ommochrome into another is nevertheless still not available. .

. ,.

.

(1 :

.

:

..

i.(. ) J

:

.., ; (i,)-Br~ck

. .

.

.. :.

red

.I

Carmine

Bluish violet

(a)

(b)

( C )

(d)

Fig. 13. Sequence of appearance of ommochromes during eye pigmentation in Ptychopodn serinta. X, xanthommatin; 0, ommin (probably ommin A); I and I1 are Kiihn’s ommochromes “I” and “11”-probably ommins. (From Kiihn, 1963.)

Finally, in relation to ommochrome biosynthesis, we may ask whether there is any turnover of ommochromes. According to orthodox opinion ommochromes are regarded as end products of tryptophan metabolism which are either excreted or stored until the death of the insect. A drastic decrease of ommatin content in the gut of diapausing Ceruru pupae (Linzen and Biickmann, 1 9 6 l ) , originally viewed as metabolic breakdown, is in all probability due to spontaneous decomposition. Dustmann’s (1964) examination of the chemical basis for morphological colour change in the stick insect, Curuusius morosus, has revealed that during the life of the insect the ommochrome content only increases. Thus, most results do not favour the hypothesis of a metabolic degradation of ommochromes. Recently Hehl and Linzen (to be published) measured the incorporation of radioactivity from labelled tryptophan or 3-hydroxy-kynurenine into xanthommatin in the compound eye of the blowfly, Calliphoru. They found that even in the aged fly, when the amount of xanthommatin is at a

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

197

constant level, there is still some incorporation. This might be indicative of a very slow turnover. There is some histologica.1 evidence of degradation of pigment granules. In the eyes of the cricket, Pteronemobius heydeni, autophagous vacuoles have been observed, which enclose material of diverse origin, including membrane fragments and pigment granules (Wachmann, 1969). “Granulolysis” is also documented by electron micrographs of the stomatopod crustacean, Squilla mantis (Perrelet et al., 1971). Still, the turnover of ommochromes remains a matter for debate.

8 Tryptophan metabolism in insect development The most important fact with regard t o tryptophan metabolism in insects is the failure t o degrade 3-hydroxy-anthranilic acid t o water, carbon dioxide and ammonia (i.e. t o easily permeable products). As a consequence, metabolites produced prior t o this metabolic block will accumulate whenever the exchange of chemical compounds between the insect and its environment is impaired or rendered impossi‘ble. This is the case during embryonal development in the egg in dormant and pupal stages and during moults. Accumulated metabolites may be subsequently excreted or used for some secondary function. Examples have been given in section 5 .

8.1 EGGS A N D EMBRYONAL DEVELOPMENT Accumulation of tryptophan metabolites in the eggs is known in a number of species. In Ephestia it causes the pigmentation of the ocelli in larvae of the a/a genotype. It might be expected that such larvae would be devoid of pigment, but if they originate from heterozygous (a +/a) mothers they can draw on a supply of 3-hydroxy-kynurenine laid down during growth of the oocyte (Kiihn and Plagge, 1937; Egelhaaf, 1963a). If kynurenine is injected into Ephestia females it is transiently accumulated in the ovarioles, its concentration subsequently decreasing slowly while 3-hydroxy-kynurenine increases. It is assumed, therefore, that the enzyme, kynurenine-3hydroxylase, is present in this tissue. In ovaries of the silkworm the activity of this enzyme is relatively high. In addition, there is active uptake of 3-hydroxy-kynurenine from the haemolymph. The selective absorption of this metabolite is under control of the “diapiiuse hormone”. This can be best demonstrated by use of the white-l mutant which is unable to hydroxylate kynurenine (Yamashita and Hasegawa, 1966, and earlier papers; Sonobe and Ohnishi, 1970). The level of 3-hydroxy-kynurenine in diapause eggs continues t o rise during the lirst 1 2 h after oviposition, suggesting that the hydroxylase is located wi :hin the egg cytoplasm. The disappearance of 3-hydroxy-kynurenine in silkworm eggs and its conversion into ommochromes (Fig. 14) has been measured several times. Kikkawa

198

BERNT LINZEN

I

2

3

4

5

6

7

Days after oriposition

Fig. 14. Concentrations of free tryptophan, kynurenine, 3-hydroxy-kynurenine, xanthommatin and ommins in eggs of the silkworm, Eombyx mori The eggs are entering diapause. (From Koga and Osanai, 1967.)

(1941j first measured disappearance of the “+-chromogen”, its identity being established later (Kikkawa, 1953; Inagami, 1958; Koga and Goda, 1962; Koga and Osanai, 1967; further work cited by Inagami). Inagami compared a number of mutants and found that the level of 3-hydroxykynurenine remains constant if ommochrome synthesis is blocked genetically. This shows that 3-hydroxy-kynurenine is neither transaminated nor cleaved by kynureninase at this stage of the life cycle. If the eggs are not fertilized and pigment synthesis does not ensue, the level of J-hydroxykynurenine, as measured by Ehrlich’s diazo reaction (not specific) remains

THE TRYPTOPHAN

+

OMMOCH ROME PATHWAY I N INSECTS

199

constant even 40 days after oviposition (Kikkawa, 1941). In fertilized diapause eggs about 95 per cent of the 3-hydroxy-kynurenine content is utilized for ommochrome synthesis by the serosa (Fig. 14), the remainder being conserved through the months of diapause. During the period of embryonic development there is also no decrease in level, the final stage examined being one day before hatching (Inagami, 1958). The function of the ommochromes in the serosa is not clear. !screening against light could be a plausible role. If white-1 eggs are supplied with S-hydroxykynurenine, they will synthesize ommochromes in the normal manner; such eggs will hatch at a higher percentage than unsupplemented white-1 eggs, and embryonal development will be more rapid (Kikkawa, 1948, cited by Kikkawa, 1953). However, a re-examination of the causal relations is desirable, since secondary metabolic effects might also be implied. It is most intriguing, therefore, that in the Noi.odontid moth, Cerura vinula, a copious amount of an ommochrome (omniatin D) is produced by the endothelium of the ovariole stalk, a proces:; which later extends through the efferent ducts. Prior to oviposition, this pigment is dissolved and precipitated on the surface of the eggs, the exocliorion. The eggs appear reddish-brown. Thus the same postulated effect of protection from light could be achieved as in the silkworm, but by completely different means (Geiger, personal communication). According to Geiger, the viability of 'the eggs is not influenced by the pigmentation. In spite of the examples just cited, the accumulation of tryptophan or of its metabolites within eggs does not appear to be the rule. In eggs of the migratory locust, Schistocerca gregaria. only traces clf kynurenine could be detected immediately after oviposition, other metabolites being completely absent. 3-Hydroxy-kynurenine appeared after ten days of incubation and rose gradually to about 0.1 pmol g-' fresh weight at thk time of hatching. During the last third of incubation, small quantities of kynurenic and xanthurenic acids were also found (Colombo and Pinamonti, 1965). Apparently, embryonic development itself does not lead to an excess of tryptophan or of its metabolites. This can be understood if it is assumed that the pattern of embryonic metabolism is such that the composition of the yolk proteins is optimal. In Schistocerca, the appearance of S-hydroxykynurenine is clearly linked to the onset of ommochrome synthesis in the developing eyes. An account of the accumulation of kynurenine in eggs of Drosophila melanogaster is provided on page 126.

8.2 LARVAL DEVELOPMENT.HEMIMETABOLA Most studies devoted to the developmental aspects of metabolism in insects were stimulated by the dramatic changes observed or expected during

200

BERNT LINZEN

metamorphosis. Relatively little is known, therefore, about the regulation of tryptophan metabolism during larval development. However, since the initiation of each moult requires some time (of the order of one day in the large holometabolous insects), and since further time elapses until feeding is resumed after a moult, one may anticipate that similar though quantitatively less important alterations in the pattern of tryptophan metabolism occur during larval moults. Buckmann (1959a) noticed that ommochromcs appeared in the excreta in the early larval stages of Cerzira vinula, i f the animals were subjected to starvation or ligation. Evidently the potential to synthesize ommochromes (and to perform the prior degradative steps) is always present in the excretory system. In the green areas of the integument of the larvae, ommochromes are not formed after experimental injury o f this kind. In contrast, large quantities of xanthommatin are synthesized in the integument during the preparation for t.he larval-pupal moult. Possibly the juvenile hormone, in conjunction with other factors, controls pigment synthesis in the hypodermis, but not in the excretory organs. Wolfram (1949) observed ommochrome excretion in young larvae of Ptychopoda only in 2 out of 40 specimens; these two were examined “immediately prior t o moulting”. The foregoing observations are indicative of a stimulation of tryptophan breakdown under abnormal conditions, such as injury o r impairment of food supply. Further insight can be gained from some data on tryptophan metabolism in Hemimetabola. Pinamonti et ul. (1964) determined several metabolites in tissues of the last larval stage and in the young imago of Schistocerca greguria. In most samples the levels were too low’ for quantitative estimation, but the level of kynurenic acid was found to rise in the integument and in the Malpighian tubules, as the larvae approached the final moult. 3-Hydroxy-kynurenine is chiefly found in the integument and is not subject to major variations. In the cricket, Gryllus bimaculutus, the levels of free tryptophan, kynurenine, and 3-hydroxy-kynurenine have been determined in whole last stage larvae (Tiedt, 1971). Only the tryptophan and kynurenine levels change significantly (Fig. 15). 3-Hydroxy-kynurenine is strongly incorporated into ommins one day after the last larval moult (and possibly earlier ones), and subsequently at a decreasing rate (Linzen, 1968). Finally, the activity of the enzyme, tryptophan oxygenase, is elevated during the periods of moulting. On the basis of these data and the observation that locusts excrete ommochromes during each moult, it can be tentatively concluded that during the periods of moulting tryptophan is likely t o be liberated from proteins (which are broken down in the course of structural rearrangement or as a source of energy) and to rise to levels at which immediate degradation becomes critical. In the intermoult periods it is probably

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IF1 INSECTS

201

0

.-E 0 c L

0

20

i

0

50

100

Per cent of stage

Fig. 15. Concentrations of tryptophan (- - -) and kynurenine (-) bimuculatus during the final larval stage. (Adapted from Tiedt, 1971.)

in Cryllus

consumed in protein synthesis. Similar considerations should apply to the larval development of the Holometabola. An observation, not mentioned before, t o support this has been made by Inagami (1958): in the haemolymph of silkworm larvae there is a doubling of 3-hydroxykynurenine concentration (from 0.25 t o 0.5 c(M m1-I) at the time of moul ting.

8.3 ACCUMULATION OF

TRYPTOPHAN METABOLITES DUWNG METAMORPHOSIS OF

HOLOMETABOLOUS INSECTS

The metamorphosis of holometabolous insects presents serious metabolic problems. The morphological transformation is so radical that the animal must withdraw for a while from its activities and from its hitherto prevailing environment, discontinuing excretion and thus changing into a nearly closed system. Furthermore, all Holometabola attain their maximal weight prior to metamorphosis; the subsequent reduction of biomass, which accompanies development towards the adult animal, may amount to 80 per cent. This figure presumably marks an extreme, resulting, if part of the accumulated protein of the larva is consumed, in the construction of cocoons or in the generation of other devices which protect the metamorphosing insect. As reported above, ommochromes are formed in many species of Lepidoptera at the onset of metamorphosis and are excreted upon eclosion from the pupa. From what we know about tryptophan degradation, this burst of pigment production indicates a massive Iireakdbwn of proteins. Such a breakdown should also be reflected by alterations in the levels of other metabolites. There is ample information to confirm this.

202

BERNT LINZEN

8.3.1 Bombyx mori The silkworm loses about 80 per cent of its fresh weight during metamorphosis: Yamafuji (1937) gives 78 per cent and Linzen (1971a) 8 5 per cent for the mutant 7-6. The varying water content of different developmental stages makes such data difficult t o interpret, but measurement of total DNA has produced very similar figures (Linzen, 1971a; Chinzei and Tojo, 1972). The most spectacular decline is seen during the two days immediately preceding and following the onset .of cocoon spinning. In the rb mutant more than 40 per cent o f the total DNA loss occurred during this period. This recalls the statement of Kellner et al. (1884) that more than half of the silkworm’s total body protein is consumed during cocoon spinning. Since silk is poor in tryptophan (Lucas et al., 1958) but must be synthesized from protein of average composition, an excess of tryptophan should ensue. A steep rise and a peak of free tryptophan concentration is in fact observed at this stage (cf. upper diagram of Fig. 20), but the absolute amount is only a fraction of what must be expected (Linzen, 1971a; see also Fukuda et nl.. 1955, for tryptophan concentration in haemolymph). The remainder must be sought in the last larval excreta and in the form of various metabolites. Two-dimensional fractionation of extracts on paper (Lirizen and Ishiguro, 1966) has indeed revealed, in the mutant rb, about ten fluorescent compounds which either make their first appearance at this stage or increase significantly in quantity. Some of these might belong to other metabolic pathways, and some might be accumulated only in this particular mutant because of the metabolic block at the kynureninase step. Kynurenine and 3-hydroxy-kynurenine are both accumulated, and both the glucoside and the sulphate (Inagami, 1958) of 3-hydroxy-kynurenine can be identified. In the normal strain a group of metabolites which are derived from anthranilic and 3-hydroxy-anthranilic acids are found. Some of these, such as Inagami’s compounds “F”, “I”, and “L”, still await identification. An appreciable, though not measured amount of tryptophan, is converted into pigment, thus causing the larvae to change colour. Quantitative data on metabolite concentrations have been obtained by Kikkawa (1953), Inagami (1958), Linzen and Ishiguro (1966), Ishiguro et al. (1971b), Ishiguro and Nagamura (1971), and Linzen (unpublished). Starting on the first day of spinning, the level of kynurenine rises to 0.3-0.5 mg g-’ in white-l pupae, and t o 0.28 mg g-’ in male pupae of the mutant rb. In the mutant rb the kynurenine concentration was found to diverge in males and females, beginning on day 5 after pupation (Fig. 16). Just before eclosion there is only half as much kynurenine in the female as in the male. Ishiguro et nl. (1971) described a reciprocal divergence of the 3-hydroxy-kynurenine level at the same time of development. The pronounced conversion of kynurenine to 3-hydroxy-kynurenine in the

THE TRYPTOPHAN

+

203

OMMOCHROME PATHWAY IN INSECTS

*

. *

. .

.

0

. 0

O

0

O

. 0

1 . 4

r

I

I

I

1

I

1

1

I

1

I

1

1

I

female is without doubt due to the activity of kynurenine-3-hydroxylasein the ovaries. The level of 3-hydroxy-kynurenine rises immediately upon cessation of feeding, reaches 0.4 mg g-' in young pupae and a peak of 1 mg g-' two days prior to eclosion. The compound is not evenly distributed. It is found dissolved in the haemolymph and occurs at greater concentrations in hypodermis and in Malpighian tubules (0.15, 0.4, and 1 PM ml-' or g, respectively, in the middle of the 5th larval stage). This distribution may change. During the period of cocoon spinning the capacity of the Malpighian tubules to store 3-hydroxy-kynurenine is reduced (Hertel, 1968) At the same time, more 3-hydroxy-kynurenine goes into fat body and gut. In the late pupa, part of the compound is oxidized to ommochromes. Another portion is conserved within oocytes in the female: if the 3-hydroxy-kynurenine content of the ovaries is subtracted from the total, identical quantities are obtained in females and males (Ishiguro et al., 1971b). In Bombyx mori, ommochromes are synthesized in the fat body, in the gut, and in Malpighian tubules mainly during the very e a l y and the late stages of metamorphosi:. This is deduced from visual inspection. A quantitative analysis of the xanthommatin content in pupae of the mutant

L

204

BERNT LINZEN

rb (Ishiguro et al., 1971b) revealed only a single burst of xanthommatin synthesis which occurred shortly before eclosion. If the appearance of red pigment and the gross morphological changes are compared from day to day, there seems to exist an overall parallelism in the intensity of both. In fat body, the stages of intensive pigment synthesis coincide with periods of declining DNA-RNA ratio, each time following a peak of this ratio (cf. Chinzei and Tojo, 1972).

8.3.2 Crrura vinula Pupae o f this Notodontid moth undergo a diapause, which separates the two periods of interest with respect t o tryptophan breakdown and ommochromc synthesis (i.e. the periods of greatest morphogenetic activity). The early phase, from cessation of feeding until pupation, was investigated by Linzen and Biickmann (1961) and by Biickmann et al. (1966). This period is more extended than in Bonzbyx morz (9.5 versus 3-4 days) and can be easily divided into successive stages on the basis of the spectacular colour change described above (p. 175). In the course of one week, the larva loses 50 per cent in weight (Fig. 17(a)). Much of this represents water loss although a considerable amount of protein is spent in construction of the cocoon. Total haemolymph proteins show an absolute decrease in quantity. The protein concentration fluctuates strongly, however, as a consequence of fluctuations in haemolymph volume. In contrast, the concentration of ninhydrin-positive material shows a surprising constancy, while individual amino acids are subject to profound changes. A peak in tryptophan and 3-hydroxy-kynurenine concentration coincides with the stage of maximal dark-red colouration (Buckmann’s stage 11) (Fig. 17(b)); by this time the larva is actively building its cocoon. The level of kynurenine has not been determined, but, according to Geiger (personal communication), “large quantities” are excreted at the last defecation. Only at the beginning of the metamorphosis of the Cerura larva are ommochromes synthesized in the integument. From stage I1 onwards they are deposited in fat body and gut. By injecting tryptophan, J-hydroxykynurenine, and ecdysone (either alone or in combination) into intact feeding animals or into isolated abdomina, it was demonstrated (Biickmann, personal communication) that the larvae are capable of degrading tryptophan to 3-hydroxy-kynurenine at any time. Integumental xanthommatin is, however, formed only under the influence of ecdysone. Further experiments are necessary t o discover whether ecdysone directly induces individual enzymes along the ommochrome biosynthetic pathway or whether the synthesis of ommochromes is a more remote effect. At the termination of pupal diapause there is a further shift of the ommochromes (Geiger, personal communication). The total ommochrome

1.6 -

50

I4-

1.0 -

08-

10

0.2q'j.o.,

5

15

10

5

Days

U I I I I I O20,III

I

m

Stages (a)

I

I

r n P

.dI

I I

,

I I I 10 Days

I I I 15

I

P2 Stages

(b)

Fig. 1 7 (a and b). Total body weight, haemolymph volume, haemolymph as per cent of body weight, and concentrations of free tryptophan and 3-hydroxy-kynurenine in the haemolymph of Cerura vinula L. at the onset of metamorphosis. Abscissa in days and in stages: 0 2 and 0 4 are feeding larvae, I to I11 stages of morphological colour change, IV prepupa, P pupa, U marks beginning of integumental colour change. (From Biickmann e t al., 1966.)

206

BERNT LINZEN

in fat body decreases by 70 per cent, while at the same time there is a strong increase in the gut. It is highly probable that the pigment is transported via the haemolymph but in contrast to the stage prior to diapause (when ommatins appear in solution) no ommatins could be demonstrated in the haemolymph of the pharate adult. 8.3.3 Phormia terraenovae Less complicated metabolic problems than those dealt with in the above two species can be expected in insects which do not spin cocoons but rely on other mechanisms for protection. As an example, the blowfly, Phormia terraenovae, was analysed with respect to the number and quantitative changes of tryptophan metabolites (Linzen and Schartau, to be published). It was found that the inventory of metabolites is restricted to the pathway leading directly to the ommochromes; traces of kynurenic and xanthurenic acids but neither of the anthranilic acids were found. The concentration levels of tryptophan and of two of its metabolites are shown in Fig. 18. Again, the primary event at the onset of metamorphosis is the liberation of tryptophan prior to pupation. It is puzzling that its degradation is not noticeable three days later, although, as will be reported below; tryptophan oxygenase is always present. Spatial separation of enzyme and substrate, or temporary inhibition of the enzyme must be postulated. The rise of the kynurenine level coincides with the beginning of eye pigmentation. Although the rate of ommochrome synthesis is maximal during the day before eclosion, synthesis continues until the third day of the ffy’s life. In Phormia also the course of total protein-bound tryptophan was determined. Although the method applied (Roth, 1939) is debatable (cf. Linzen, 1971a), the values obtained appear to be meaningful if compared with other data: from the day of pupation to eclosion, the fresh weight decreases by 20 mg, the “extracted dry weight” by 6 mg, and proteinbound tryptophan by 30-35 pg (all data, per animal). A balance will be presented below, which shows that practically all tryptophan liberated is converted into xanthommatin, which is used as screening pigment in the large compound eyes. 8.3.4 Other species During studies dealing with particular aspects of tryptophan metabolism in insects, many results have been obtained which fit into the frame defined by the above three examples. In Ephestia (Egelhaaf, 1957, 1963a) the tryptophan level begins to rise in the prepupa and reaches a peak of about 120 pg g-’ fresh weight on day 3 (20 per cent) of pupal development. In the strain BK 14 (wild type) the rise of the kynurenine level is roughly parallel. After the 4th day in the

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

207

Fig. 18. Levels of tryptophan, kynurenine and 3-hydroxy-kynurenine during development of Phormia terraenovae. Abscissa in per cent of larval development and in days. Figures along curves give amount in pg per animal. Figures at right margin give amount on 11th and 15th day after eclosion. P,, and F, are days of puparium formation and eclosion, respectively.

pupa there is a sharp drop. This could be a consequence of continued kynurenine hydroxylation at a moment when liberation of tryptophan from protein slows down. The metabolite levels dif6-r with strains and an influence of rearing conditions must be expected also, but has not been so far examined. The level of 3-hydroxy-kynurenine reaches its maximal value (-1 mM) in the pupal stage. In the mutant a which lacks kynurenine and all subsequent metabolites, the level of free tryptophan is elevated correspondingly (0.75 mM at the time of pupation, -1.5 mM at peak concentration) and remains high after emergence of the moth, although a

208

BEANT LINZEN

fraction of the tryptophan content is taken up by the eggs. In Plodia, the course of tryptophan concentration has not been followed but kynurenine and 3-hydroxy-kynurenine rise 10-20-fold, and about 4-fold, respectively (Mohlmann, 1958; de Almeida, 1961). The data obtained in Calliphora erythrocephala compare well with those of Phormia (free tryptophan: Langer and Grassmader, 1965; Grassmader, 1968; 3-hydroxy-kynurenine and xanthommatin: Linzen, 1963). After a sharp rise, the tryptophan level declines slowly. A comparison of the wild type with the mutant chalky (pigmentless) reveals that about two thirds of the tryptophan stored must be excreted upon emergence and may therefore reside within the Malpighian tubules or hindgut. Analysis of the meconia for their tryptophan content has demonstrated this. 3-Hydroxy-kynurenine rises from the egg through the larval stages until the middle of the pupal stage (-35pg per animal, -2 mM) and is later consumed during eye pigment synthesis. The amount of xanthommatin in wild-type eyes given by Linzen (1963) must be reduced by a half, as it is based on an extinction coefficient which has now been shown to be incorrect (EY.fZ = 7.32; true value, 15.1 or higher). It is clearly seen that 3-hydroxy-kynurenine is quantitatively converted into xanthommatin, further 3-hydroxykynurenine (15-18 pg per animal) being required for completion of eye pigmentation. This is synthesized during the second half of pupal development. Again, this is seen in the 3-hydroxy-kynurenine level of the mutant chalky. In the mutant, however, the excess of 3-hydroxykynurenine is not accumulated within the eyes but is excreted. The meconia contain 16.4 pg per animal (Grassmader, 1968), but it is almost certain that excretion of 3-hydroxy-kynurenine continues beyond this time. In Drosophila, which has been the subject of so many early investigations, the metabolite and enzyme levels have not been as well studied. Danneel and Zimmermann (1954) compared various mutant strains with respect to the presence or absence of kynurenine and reported that kynurenine disappeared from head extracts 60 h after pupation, but persisted in the remaining body throughout the adult stage. The role played in tryptophan metabolism by the Malpighian tubules was analysed in detail by measuring the fluorescence of tryptophan and the two kynurenines after paper chromatographic separation. Wessing and Bonse (1962) discovered that the Malpighian tubules of Drosophila accumulate free tryptophan (Fig. 19). This is most obvious in the vermilion mutant where 0.1 pg per animal is found after hatching from the pupa. This corresponds to some 25 per cent of the total free tryptophan, if Green’s (1949) figures are converted t o a per animal basis by assuming 1 mg fresh weight per fly and 30 per cent dry matter. The storage function of the Malpighian tubules is even more pronounced in the adult fly (Eichelberg, 1968);the fluorescence

THE TRYPTOPHAN

+

II

OMMOCHROME PATHWAY

'\

I \ I \

I

I

9-

;

I I I

0 -

.-

5

209

4

-

IO-

-

IN INSECTS

I ' I

7-

I

&

\ \

\ \ \

\ \

\ \

I I

I

, I I I I

/

I

I-2d

5-6d

10--14d

Fig. 19. Free tryptophan content of the Malpighian tubules in Drosophila melanogaster during metamorphosis. (From Wessing and Bonse, 1962.) Wild type (-), vermilion mutant (- - -).

intensities obtained are much higher (by more than an order of magnitude in the case of kynurenine) than in the larval and pupal stages. Only 3-hydroxy-kynurenine is highest in the larval stages. Unfortunately, only relative values were given. In the parasitic wasp, Habrobrucon juglandis, there is a switch in tryptophan catabolism from transamination of the kynurenines during the preparatory phase of metamorphosis to ommochrome synthesis during the pupal stage (Leibenguth, 1965, 1967a, 1967b, 1970). Transamination in the spinning larva is so efficient that at the prepupal stage no J-hydroxykynurenine is left, even if the endogenous supply iir supplemented (cf. p. 131). The products, kynurenic and xanthurenic acids, are excreted. Kynurenine is always at a low level, while 3-hydroxy-k ynurenine reaches a maximum a day before eclosion. It is not excreted i n the meconia, but presumably consumed in continuing eye pigment synthcsis. A selection of the quantitative data relating to tryptophan and metabolite concentrations is presented in Table 11. It is beyond doubt that

TABLE 11 Accumulation of tryptophan and tryptophan metabolites during metamorphosis of holometabolous insects (selected data, converted to approximate concentration) ~

Species

Ephestia kiihniella (mutant a ) (wild type) Bombyx mori, mutant rb

(wild type)

Papilio xuthus

Compound*

Concentration (mMg-' fresh weight) Larvae At peak

TRY 0.5 KYN 0.05 3-HO-KYN 0.25 TRY 0.15 KY N 0.02 1.o 3-HO-KYN 3-HO-KYNglucoside 0.06 3-HO-KYN 0.1 TRY
1.8 0.2 1.o 0.75 1.3 3.O 0.3 3.0 1.6 1.7

Stage of peak**

80% PD 50% PD 75% PD

One day in cocoon 80% PD 80% PD 70% PD 80% PD 60% PD 90% PD

constant at 0.1 mM

~~~~~

Remarks

Depends on strain

Reference

a a

a b Males, less in females Females, less in males

d

Females, less in males

d d

C

e e e

Cerura vinula Calliphora ery throcephala

Phormia terraenovae Habrobracon juglandis (wild type)

(mutant

0)

TRY 3-HO-KYN TRY

0.3 0.1 0.1

1.8 1.o 0.5

Larva in cocoon

Haemolymph concentration

20% PD

3-HO-KYN

0.8

2.5

30% PD

Crude estimate from original per animal values Concentrated in Malpighian tubes

See Fig. 19 0.12

0.6

20% PD

3.0 0.06 0.8

70% PD Spinning larva

KYN 3-HO-KYN KA XA KY N

0.75 0.0 0.1 0.3

f f g h

I

4.6

* TRY, tryptophan; KYN, kynurenine; 3-HO-KYN, 3-hydroxy-kynurenine; KA, kynurenic acid; XA,xanthurenic acid. ** PD, pupal and pharate adult development. t a. Egelhaaf (1963a); b. Linzen (1971a); c. Linzen (unpublished); d. lshiguro et al. (1971b); e. Umebachi and Katayama f. Biickmann et al. (1966);g. Langer and Grassmader (1965); h. Linzen (1963); i. Leibenguth (1967a);j. Leibenguth (1965).

(1966);

21 2

BERNT LINZEN

liberation and accumulation of free tryptophan at the onset of metamorphosis (usually evident when the larva ceases t o feed) and its conversion into fluorescent metabolites or ommochromes is a phenomenon common to holometabolous insects. In each case, tryptophan accumulation is transitory. It appears that the metabolite most likely to persist at elevated levels is 3-hydroxy-kynurenine. In the majority of cases, kynurenine is held at very low concentration. It may be argued that the transitory rise of metabolite levels is a consequence of limited capacity of degradation. Alternatively, it might be assumed that different steps of the degradative pathway are performed in different tissues (e.g. tryptophan oxygenation in fat body, and kynurenine hydroxylation in Malpighian tubules), so that diffusion of the intermediates could become the limiting factor. In several instances the measurement of metabolite levels and of enzyme activities in vitro have led to conflicting results. Spatial separation is but one explanation, while others are conceivable. Certainly a point to be considered is the physical state of the metabolites. They may be accumulated in certain tissues, either in the form of solid concrements (as 3-hydroxy-kynurenine in Malpighian tubules) or adsorbed (as in oocyte yolk spheres). Even if a metabolite is found dissolved in the haemolymph, it is by no means established whether it is in true solution or whether it is partly or wholly adsorbed by proteins. In the case of mammalian plasma it has been shown that tryptophan is the only amino acid which is bound by protein to a significant extent (McMenamy and Oncley 1958;McArthur and Dawkins, 1969). No comparable data are available for kynurenine or 3-hydroxy-kynurenine, nor has the binding of tryptophan to proteins been studied in insect blood.

8.4 ONTOGENY OF ENZYME ACTIVITIES It is obvious that the activities of the enzymes degrading tryptophan are not exempt from the profound reorganization of the holometabolous organism and its functions. Kaufman (1962) found that in Drosophila melanogaster the activity of tryptophan oxygenase increases in larvae, remains at an elevated level in pupae, and doubles its activity in adult flies. The first increase could serve in the synthesis of eye pigments (which starts on the second day after puparium formation) by providing the required precursor. No particular physiological role could be assigned to the enzyme in the adult stage. From data obtained by Wessing and Bonse (1962)it is evident that the presence of the enzyme counteracts an accumulation of free tryptophan. There is practically no rise during the pupal stage in the Oregon wild-type strain, while there is a tremendous increase in the mutant vermilion. This excess is gradually excreted after eclosion.

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

21 3

While Drosophila kynurenine formamidase was not studied during development (according to Glassman, 1956. the activity is constant in flies), marked changes have been observed in the activity of kynurenine-3hydroxylase (Ghosh and Forrest, 1967a). The American authors did not find any activity in larvae, but they reported pronounced activity in pupae (150 m p m o l g-' h-' at 37'C) and low actkity in flies. Today, a re-examination taking the precautions realized in the meantime might possibly yield higher values. Quite different results were obtained during development of blowflies (Hendrichs-Hertel and Linzen, 1968; Linzen and Schartau, to be published). Both in Calliphora erythrocephala and in Phormia terraenovae the specific activjty of tryptophan oxygenase is hiF,h in larvae and in flies, so that a U-curve is obtained. This is quite unexpected if compared to the metabolite levels (Fig. 18). These would have suggested a rise in tryptophan oxygenase activity in the middle of adult development. However, the measurement of the enzyme activity in crude homogenates is subject to errors, as homogenization is an artifact per se. Inhibitors as well as activators could be liberated during this process and could combine with the enzyme. In Phormia too, the significance of high oxygenase activity in the adult fly is obscure. Most activity resides in fat body, in Malpighian tubules, and in the testes, there being only medium activity in the ovaries. In blowflies, kynurenine formamidase and kynurenine-3-hydroxylase exhibit activity curves which are quite different from each other and from the curve of tryptophan oxygenase. The former enzyme is especially active in larval extracts; the activity declines during the pupal stage and reaches a low but constant level in the adult fly. Even at this stage it surpasses the activity of tryptophan oxygenase by a factor o f ten. Kynurenine-3hydroxylase (Calliphora) increases during larval development, culminates the time of pupation, and declines to zero at the time of eclosion. This would accord with the observed 3-hydroxy-kynurenine level, which rises after pupation and declines in the second half of the pupal stage. From the data available for Phormia, a similar ontogeny of specific activity is apparent. The low activity of kynurenine-3-hydroxylasein pharate adult and adult flies might be due, in part at least, to the inhibitory action of xanthommatin liberated in the course of homogenizing the animals. Although it is possible that each of the enzymes under consideration might be activated or inhibited independently (by factors present in the homogenates but not present in vivo at the site of the particular enzyme), the mere recognition of very different activity curves is striking. An argument to support the notion of true differential actimtion in vivo is Ghosh and Forrest's (1967a) finding that the course of kynurenine-3hydroxylase activity is almost idcntical in crude extracts and in a

21 4

BERNT LINZEN

(presumably mitochondrial) particle fraction. It is important to recognize that the three enzymes are not in the least coordinated, in the sense of a programmed, concerted pattern of activity. Nor is there much evidence for a relation between substrate levels and enzyme activity. Only in Drosophilu has it been observed that tryptophan oxygenase in fat body cells can be induced by feeding tryptophan (cf. p. 185). Not only in blowflies, but also in other species, conflicting results were obtained when metabolite levels and enzyme activities were compared. In the parasitic wasp, Hubrobrucon juglandis, Leibenguth ( 1 967a, 196713) observed a sharp rise of kynurenine transaminase activity during the time from the spinning larva until eclosion of the imago, there being no significant formation of kynurenic and xanthurenic acids in vivo. Leibenguth suggested that pyridoxalphosphate and a-ketoglutarate concentrations might become limiting in the pupa. In Habrobracon also, tryptophan oxygenase activity is low in the pupa. Evidence was obtained, using mixed homogenates, that pupal extracts contain an inhibitor, which reduced the activity of imaginal extracts by 50 per cent. The inhibitor is inactivated by heat. In mosquito pupae, the activity of tryptophan oxygenase is only 40 per cent of the activity in larvae or in adult animals (Prasad and French, 197 1). More detailed information came out of studies with Ephestiu kri’hniellu and Bombyx mori. Egelhaaf (1963a) discovered that tryptophan oxygenase activity is located within a number of tissues and that the specific activity within any tissue differs with developmental stage. Thus, in the fat body there is a sharp drop from the feeding larva to the prepupa, while in the hypodermis the activity rises. Activity present in larval and pupal testes disappeared completely after eclosion. Egelhaaf stated that the activity which he had measured in whole animals was subject to much less variation than the activity in individual tissues, and emphasized the importance of a separate assay. In the silkworm, tryptophan oxygenase and kynurenine-3-hydroxylase were measured in a number of tissues in daily intervals (Linzen and Hendrichs-Hertel, 1970; Linzen, 1971b). The results of these studies, which are presented in idealized form in Fig. 20, lead to the following conclusions. 1. The in vivo transformations of the respective substrates can be easily accounted for even when the in uitro activities of both enzymes are low (i.e. in the pU range). Fig. 20. Specfic activities in vitro (in picomoles product per min per mg protein) of tryptophan oxygenase (-) and kynurenine-3-hydroxylase(- - -) in various tiggues of Bombyx mori (mutant r b ) during metamorphosis. The inset in the upper diagram shows concentration of free tryptophan. S, first day inside cocoon (larva still visible); P, pupation; E, eclosion.

THE TRYPTOPHAN + OMMOCHROME PATHWAY IN INSECTS

21 5

100

1000

AIP-8

21 6

BERNT LINZEN

2. The enzymes of tryptophan catabolism are not characteristic of any particular tissue. While both of the enzymes studied occur in fat body (i.e. in cells specialized for intermediary metabolism) either one or both are also found in excretory, reproductive, or organs of locomotion. This is in sharp contrast to the restricted localization observed in the vertebrate. 3. The activities of the two enzymes are not linked, in spite of their rather closely related function. Thus, either both may be detected in a particular tissue or may occur singly. 4. The ontogenetic pattern of enzyme activity differs with each tissue and is at the same time a characteristic of each of the two enzymes. Thus, in fat body, both enzymes fit into the common U-pattern. In the Malpighian tubules a most striking feature is the sharp peak of the hydroxylase just before the onset of spinning. Testes and ovaries both demonstrate gradually decreasing oxygenase activity, while on the other hand there is a transitory appearance of the hydroxylase in ovaries. The notion of differential enzyme activity patterns in the development of insects has thus been amply confirmed. While it cannot be ruled out that extraneous factors (such as contamination by other tissues) may have influenced the results, it appears extremely unlikely that the differences observed could have exclusively originated from such artifacts. The changing levels of activity must to some extent be linked t o other events in the tissue under study. In other metabolic pathways, researchers have encountered similar situations (e.g. Rechsteiner, 1970; Barnes and Goodfellow, 1971), although in no case have the separate tissues been studied in so much detail. The question thus arises as to the means by which such enzyme activities are regulated. The central problem is then whether the observed changes are related t o alterations in the concentration of enzyme protein, or to changes in the state of activation of existent enzyme. It is difficult t o give an unequivocal answer, since sequential changes in enzyme protein in insects have not been measured. However, if the state of activation of a particular enzyme should change with development, the causes of such differences would be partially offset by the incubating conditions which provide constancy of pH, ionic strength, substrate, and cofactor concentrations. One is on somewhat safer ground, if, as has been done in the case of fructose diphosphate aldolase (Bauer and Levenbook, 1969), the enzyme is partially purified by electrophoresis or other procedures. It would appear, therefore, that it is most likely that it is the levels of apoenzyme which change. Several principles can be envisaged which might cause sequential changes of enzyme concentration. 1. Enzyme induction by substrate. According t o our present knowledge,

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY I N INSECTS

21 7

the metabolite and enzyme levels are rather poorly related, and give no indication of probable induction. In several insect species there is evidence to show that tryptophan oxygenase and kynurenine-3-hydroxylase are constitutive enzymes with the exception of larval Drosophila tryptophan oxygenase. 2 . Differential enzyme synthesis in response to hormones. At present, no evidence is at hand. Biickmann (1959a) and Karlson and Biickmann (1956) have shown that colour change in the larva of Cerura is induced by ecdysone. Yet, it must be stressed that the kynurenine pathway is of secondary importance in metabolism and, therefore, not likely to be under direct control of morphogenetic hormones. 3. Cell multiplication, turnover and cell death. Some of the changes are evidently related to the formation or breakdown of tissues: tryptophan oxygenase activity in the wings of Bombyx increases during cell multiplication and growth, but drops when differentiation comes to completion and when the hypodermal cells die away. Also, the appearance of kynurenine3-hydroxylase in silkworm ovaries is clearly correlated with growth. In contrast, the rise in kynurenine-3-hydroxylase activity in the Malpighian tubules occurs during a period of cell constancy, and the drop of both enzymes in fat body during a period in which rxpid growth of fat body masses is manifest and total soluble protein is still increasing. 4. Differential activation of transcription and/or translation, according to a developmental program. This would most easily explain the changing levels of specific enzyme activity. However, in view of the results obtained in Bombyx, an explanation must be found for the rather accidental appearance of tryptophan oxygenase and kynurenine-3-hydroxylase activities. A hypothesis proposed by Linzen (1971b) implies that the corresponding genes are not linked (neither in the usual meaning nor in any other functional way), and are transcribed soldy according to tissue specific developmental programs. Reference is made to the theory of Britten and Davidson (1969), and it is recalled that in Drosophila the vermilion and cinnabar genes are located on (different chromosomes. Random (though programmed) gene expression ir different tissues would be compatible with normal function and development if the metabolite controlled were of minor importance (both by its quantity and by its position on the metabolic map) and if diffusion and transport as possibly limiting factors could be accounted for. The only requirement in such cases would be that the metabolite should be supplied at a minimum rate or be kept below a certain concentration. Such conditions might hold for the kynurenine pathway. 5 . Intracellular protein turnover with chunging ratios of enzyme synthesis and degradation. This would shift the problem of regulation as

21 8

BERNT LINZEN

outlined under (4) t o an extrinsic function. While some work has been published on the turnover of haemolymph proteins (cf. Wyatt, 1968), our knowledge of intracellular protein and enzyme turnover in insects is apparently nil. At present, it is impossible to decide which alternative is of greater importance. It is anticipated that each one will eventually be found to contribute to the accomplishment of ordered and responsive transformation of the metabolic intermediates. Finally, the relation of enzyme activity determined in vitro to the rate of product appearance in vzvo might be examined. Most attempts of this kind have shown that the measured enzyme activity is greatly in excess of the demands of the organism. However, in vitro activity is by definition measured under optimal conditions of substrate concentration, pH etc., which may not necessarily prevail within the cell. For the very reason of the adaptive significance of the K, value, most enzymes must operate in vivo at substrate concentrations which are far from optimal (cf. Atkinson, 1969). These considerations can also be applied to the enzymes of the kynurenine pathway in insects. In most instances the in vitro activity (at optimal conditions) is far in excess of the in vivo reaction rates. Only at the onset of metamorphosis it appears that saturation of the enzymes could result from rapid liberation of tryptophan. In Bombyx mori it was estimated (Linzen, 1971b) that the rate of free tryptophan decrease at the time of pupation and in the young pupa is equal to and therefore limited by the total activity of tryptophan oxygenase in the organism This is low at this stage as the larval enzyme (mainly in fat body) has gone down by this time, while the enzyme of the pharate adult (mainly in ovaries and wings) has not yet been synthesized. This situation is indicated in the top diagram of Fig. 20.

8.5

AlTEMPTS TO ESTABLISH A TRYPTOPHAN BALANCE

While it is generally understood that in vertebrates all dietary tryptophan is eventually broken down via the kynurenine and glutarate pathways, ii is nevertheless difficult to substantiate this statement by straightforward experiments (cf. LeMem, 1971). Yet, since in vertebrates only two pathways of tryptophan degradation areLknown, and, as the amount of 5-hydroxy-indoleacetic acid excreted .per day accounts for less than 1 per cent of the total amount of tryptophan administered (under conditions of tryptophan load), the only question which remains to be answered is which fraction of the remaining 99 per cent is metabolized via the kynurenine pathway and which by intestinal microorganisms. (The latter factor, incidentally, indicates the necessity of quantitating intestinal tryptophan

THE TRYPTOPHAN

+

OMMOCH ROME PATHWAY IN INSECTS

21 9

resorption and of tryptophan loss caused by secretion of digestive enzymes and by turnover of the intestinal lining.) The metamorphosis of the Holometabola provides an opportunity to assess the scope of the kynurenine pathway in insects since with respect to tryptophan metabolism one is dealing with a closed system (cf. pp. 133 and 197). It is feasible to set up a balance of tryptophan metabolism by determining total protein-bound tryptophan and the sum of free tryptophan plus all detectable metabolites at two points of development. This has been tried in the Bombyx mori mutant rb (Linzen, 1971a), and in the blowfly, Phormia terraenovae (Linzen and Schartau, to be published). In Bombyx, the period studied extends from the end of spinning (S + 2 days) to 48 h before eclosion (S + 11 days). The total decrease in protein-bound tryptophan during this period is about 5 pmol (if, in Fig. 4 of Linzen, 1971a, a line is drawn from S + 1 to S + 12 days). Free tryptophan decreases by 0.7 pmol. During the same period about 1 pmol of kynurenine, 2.7 pmol of 3-hydroxy-kynurenine 0.5 pmol of xanthommatin (corresponding to 1 pmol of tryptophan), and about 0.3 pmol of 3-hydroxy-kynurenine glucoside are formed (Linzen and Ishiguro, 1966; Ishiguro et a/., 1971b; Linzen, unpublished). Taken together, these metabolites make up for 5 pmol of tryptophan. The synthesis of small amounts of 3-hydroxy-kynurenine sulphate, of xanthurenic acid, 4,8dihydroxy-quinoline, ommins, and possibly other compounds of minor importance must also be considered. In Phormia the situation is simpler, since the tryptophan metabolism is less diverse. Between the 3rd and 5th day after puparium formation there is a decrease of 28 nmol of free and 46 nmol of bound tryptophan per animal. These are almost stoichiometrically accounted for by the metabolites: 20 nmol of kynurenine, 9.6 nmol of 3-hydroxy-kynurenine, and about 23 nmol of xanthommatin (to be multiplied by 2). There is also fair agreement between the decline in protein-bound t,ryptophan from the day of puparium formation t o the day of eclosion (0.15-0.1 7 pmol), and the amount of xanthommatin synthesized in the eye!; (0.066-0.071 pmol per animal, multiply by 2). Difficulties arise, however, if estimates are made on a day-to-day basis due t o sampling error and to the fact that the analyses of the different compounds were partly performed on different batches of animals. fn view of the relatively crude method uscd for the determination of total protein-bound tryptophan (the xanthoprotein reaction) it would be highly desirable that the results were checked by independent methods. In spite of such reservation it is tentatively concluded, at least in these two species, that most, if not all, tryptophan is degraded via the kynurenine pathway. It may be argued that protein breakdown achiives an extreme at the beginning of metamorphosis, and that in o h r developmental stages different routes of tryptophan catabolism might b'e favoured. However, at

220

BERNT LINZEN

present there is no indication that other metabolites of tryptophan than those listed above are formed to a greater extent in any stage of insect development. 9 Detrimental effects of tryptophan and tryptophan metabolites

Many of the conclusions and interpretations in this review have emphasized that removal of free tryptophan from the cellular environment is of vital importance. The crucial question (the Cretchenfrage) is whether an elevated level of tryptophan can, in fact, produce toxic or harmful effects. Such a possibility was actually implied as early as 1952 by Nolte in a discussion of the role of the eye colour genes in Drosophila. Such effects might be difficult to detect, as they should be of a quantitative rather than of a qualitative nature. However, a survey of the literature (Grober and Linzen, unpublished) has revealed a wide range of negative effects cauacd by tryptophan or some of its metabolites. It appears that many of thew are related to developmental processes. Normal growth may be retartied or deflected to a pathological condition, such as the induction of bladder tumours and of leukemias by ortho-aminophenols (3-hydroxy-kynurenine, 3-hydroxy-anthranilic acid). In insects, there are two lines of evidence relevant to this problem: retarded development in various species and increased penetrance of tumour promoting genes in Drosophila. A retardation of development by about 2 per cent was reported in the first paper about the a mutant of Ephestia (Kuhn and Henke, 1930).' In Drosophila, Wilson (1945) studied the effects of tryptophan added t o sterile culture media at increasing concentration. She observed a significant prolongation of the pupal stage at concentrations from 10 mM down to 0.4 mM. At concentrations above 20 mM also the size of the animais was reduced. Hinton et al. (1951) reported a retardation of larval development in a chemically defined medium by L-tryptophan at concentrations above 15 mM, with an approximate doubling of the time necessary until pupation at 45 mM. Interestingly, addition of ribonucleic acid at least partly abolished this effect. Parsons and Green (1959) studied the effect of the v mutations on fitness. Under conditions of low competition there was no difference in the fitness of males (obtained by crossing homozygous 6w males to 6w females homozygous for various vermilion and vermilion suppressor genes), but in crowded cultures the suppressed vermilion males were fitter than unsuppressed (the sex ratio was used as a criterion). Since unsuppressed vermilion flies contain a higher level of free tryptophan, this would be in line with the hypothesis of tryptophan toxicity, but surprisingly the same

'

According to personal communication from L. Caspari to A. Egelhaaf this effect has not been reproduced.

THE TRYPTOPHAN

+

OMMOCH ROME PATHWAY IN INSECTS

221

improvement of fitness as brought about by the suppressor can also be achieved by feeding kynurenine. This result is pexplexing: the authors state that “competition was almost completely larval”, competition between adults being negligible. As ommochrome synthesis induced by vermilion suppression or kynurenine supplementation is not a larval but a pupal process, the effect of kynurenine on fitness mighi be mediated by another, possibly roundabout effect. The grain beetle, Oryraephilus surinamensis, grows best on a diet containing 0.2 per cent tryptophan, while increasing the concentration to 0.5 and 1 per cent results in a reduced rate of development (Davis, 1968). The formation of tumours represents quitc a different aspect of tryptophan toxicity. In their attempt to establish a chemically defined medium for raising Drosophila larvae, Hinton e t a ! . (1951) observed that at concentrations of 30 mM tryptophan “all the flier showed abnormalities of various types-for example, tumours throughout the body, deformed heads, wavy bristles” and more. These observations were made in a wild-type strain. In Drosophila, there exist strains with a hi:gh tumour incidence (e.g. tu and v t A ). The tumours formed in thcse strains are often of the benign type and arise from transformation and accumulation of cells rather than from uncontrolled cell division (cf. Rizki, 1957, and papers cited by him; a possibly important paper, but not obtained by the author, is that by Ghelelovich, 1969). Plaine and Glass (1955) studied the effects of various agents on the formation of melanotic tumours and “erupt” eyes, and found that tryptophan, along with indole, was most effective (see also Mittler, 1952). Deleterious concentrations of tryptophan and of tryptophan metabolites may apparently arise under natural conditions, tumour incidence being raised significantly if tu is crossed with v, cn, or st (Kanehisa, 1956a). Feeding supplementary tryptophan t o such crosses enhances the effect, while kynurenine promotes tumour incidence to a lesser extent (Kanehisa, 1956b; see also Kanehisa and Fujii, 1967). In a search for a metabolite which might be madc directly responsible for promoting melanotic tumours, Burnet and Sang (1968) found that anthranilic acid had a strong effect, while kynurlsnine was not effective. This could agree with Kanehisa’s (1956b) observations. It is noteworthy that L-methionine counteracted the tumorigenic effect. In contrast to the results just described, which are related t o the tu and tu-er strains, the penetrance of tu-h (tumorous head) is significantly lowered if the larvae are raised on a medium containing 0.5 per cent L-tryptophan (Simmons and Gardner, 1958). With regard to all of these studies it must be emphasized strongly that the active agent (a particular constituent of the diet) and its effect (a malfunction of a particular type of cell in .3 complex organism) are as loosely linked as possible, so that speculation on possible biochemical mechanisms is without any basis.

222

BERNT LINZEN

From the physicochemical viewpoint tryptophan can be shown to have properties which make it prone to both desirable and undesirable interactions. In short, these properties reside in the indole nucleus which is capable of donating negative electrical charge to suitable acceptor molecules. Charge-transfer complexes of tryptophan are observed most readily when solutions of the two reaction partners are frozen and when by the freezing process the solutes are concentrated and forced to form aggregates. In dilute aqueous solution, solvation will prevent complex formation to varying degrees. A number of interactions have, however, been observed even under these unfavourable conditions. It must be noted that the cellular milieu cannot be looked at as a simple solution but that the solvent capacity of a cell is very limited and that the space available for diffusion may often present dimensions which are not much larger than the diameter of a single hydrated molecule (cf. Sols and Marco, 1971). It has also occasionally been suggested that water within a living cell may locally be in the state of an ordered lattice and therefore not participate in solvation. Thus, the probability of in uivo interaction between tryptophan and other molecules might correspond more closely to the condition of the “frozen solution” than expected by the orthodox view. ’The association of tryptophan with other molecules has been demonstrated by a variety of methods. Isenberg and Szent-Gyorgyi (1 958) examined the red product obtained by mixing 1 mM tryptophan and riboflavin-5-phosphate solutions and concluded that by transfer of one electron from tryptophan, riboflavin is reduced to its semi-quinoid state which in turn is stabilized by complex formation. Serotonin complexes seven times more strongly than tryptophan. Similar complexes are formed between tryptophan and pteridines, thereby shifting the absorption peak of the pteridines from about 350-390 nm to about 400-430 nm (Fujimori, 1959). Protein-bound tryptophan may react as well as free tryptophan. Particularly interesting is the association of tryptophan with nucleic acids and their constituents. If DNA (1 mM) is heated in the presence of L-tryptophan (1 mM) the normal hyperchromic effect is abolished (Pieber et al.. 1969). This must be caused by association of the amino acid with the base moiety of the nucleotides, as revealed by nmr spectroscopy of appropriate mixtures. The solubility of ribonucleotides is also increased in these mixtures as a linear function of tryptophan concentration, the purines responding more strongly than the pyrimidines. The uv spectra did not change, but the cd spectra were clearly different from the calculated mean spectra of each pair of compounds. The viscosity of DNA-tryptophan mixtures did not change. The interpretation of these results (Arcaya et al., 1971) is that tryptophan interacts with DNA by intercalation between the nucleotide bases without changing much the tertiary structure of the macromolecule. This is supported by Raszka and Mandel (1971) who

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

223

demonstrated interaction between aromatic amino acids and polyadenylic acid by nmr spectroscopy and alluded to the possibility of “local melting” of DNA during replication and transcription, by this principle. HCline and collaborators (Montenay-Garestier and Helene, 197 1;Heline, 19 7 1; Dimicoli and Hiline, 19 7 1) save conducted extensive physicochemical studies on tryptophan-nucleic acid interactions. They demonstrate a stoichiometric interaction between the amino acid and the bases and give a detailed account of the spectroscopic changes. It is emphasized by them that stacking of the two-ring systems (i.e. close sidewise orientation of planar structures) is necessary for complex formation. They also contend that in the ground state of the molecules charge transfer plays a minor role in stabilizing the. complex, in comparison to van der Waals forces, although charge transfer occurs upon excitation by light. Heline et al. (1971) emphasize that tryptophan-nucleotide interaction may be of great importance in the binding of proteins to nucleic acids, implying that recognition of specific regions of nucleic acids could be made possible by this type of interaction. We might now envisage the potential hazards which may arise if tryptophan accumulates to levels high enough to provoke frequent interaction with nucleic acids. Probably nucleic acids which are bound in nucleoproteins would be protected, but this would concern only part of the total nucleic aci+ of a cell. The juxtaposition of detrimental effects in growth and development caused by tryptophan, and of the molecular interaction of tryptophan with nucleic acids is highly suggestive, but a warning must be expressed against premature conclusions. These two lines of observation are still too distant to be connected, a variety of experimental approaches being necessary to test this hypothesis. It must, for example, be determined whether tryptophan interferes with nucleic acid function and whether it could ever reach effective concentrations within cells. It must also be demonstrated that the observed inhibition of growth is directly related to impaired protein synthesis. Growth is subject t o so many variables that none but the most careful and circumspect analysis will reveal the mechanism of tryptophan action.

10 Concludingremarks It has been the intention of the author t o show that the transformation of tryptophan into ommochromes, a biochemical pathway of seemingly minor importance, is implicated in a variety of physiological functions. In particular the necessity has been emphasized for the organism to limit the concentration of free tryptophan to a level compatible with the requirements of the complex biochemical machinery and its manifold and delicate interactions of macromolecules. In insects this necessity is correlated with the inability to cleave the aromatic nucleus of 3-hydroxy-anthranilic acid.

224

BERNT LINZEN

In spite of the survey made by Lan and Gholson (1965)it cannot be stated whether this failure represents an “oversight of evolution” or whether it is a feature acquired secondarily (as Brunet, 1965,has speculated). Whatever it is, the outlet of ommochrome synthesis has endowed insects (and the rest of the arthropods) with a most valuable and variable means of handling light, by screening and by selective ,reflexion in pigment patterns. At the same time ommochromes have properties well suited to the function of storage excretion. It is opportune to emphasize some outstanding problems. While most of the intermediates in tryptophan catabolism might have been identified, the chemical structure of only the minority of the ommochromes is currently known. It is expected that the structure of the ommidins and of the acridiommatins will become known within a few years, but the ommins might continue to pose problems to the chemist for some time. On the other hand, the enzymology of tryptophan degradation in insects, including the terminal steps of pigment synthesis, must be investigated further. Finally, the mechanisms governing each single reaction at different sites and at different developmental stages of the insect organism should become the subject of future research. Throughout this review little attention has been paid to historical aspects: the great findings of the decade between 1930 and 1940 have been passed over. While the research along the tryptophan + ommochrome pathway in insects arose from the basic problem of the mechanism of gene expression, it has since contributed results which have gained importance in various fields of interest far beyond the domain of a single animal phylum. Filling the gaps in our knowledge was, however, not the only outcome of this research. While considerable insight has been gained it has also been realized that many new and basic questions have emerged. Without doubt, the discovery of facts has been paralleled by the discovery of problems.

Acknowledgements

I should like to express my gratitude to M. Alain Bouthier, Professor D. Biickmann, and Drs J. Dustmann, R. Geiger, and W. Schaefer for freely making available unpublished results or pending manuscripts, to Mr Nobuo Kita for his help with Japanese papers, t o Professor A. Egelhaaf, Dr H. Kress, and Mrs M. Linzen for reading parts or the whole of the manuscript, and to those colleagues who provided me with reprints and copies of research papers. Most of the research in the author’s laboratory, including work reported here for the first time, was supported generously by the Deutsche Forschungsgemeinschaft.

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

225

References Ajami, A. M. and Riddiford, L. M. (1971a). Identification of an ommochrome in the eyes and nervous systems of saturniid moths. Biochemistry (Washington), 10, 1451-1455. Ajami, A. M. and Riddiford, L. M. (1971b). Purification and characterization of an ommochrome-protein from the eyes of saturniid moths. Biochemistry (Washington), 10, 1455-1460. Almeida, F. F. de (1961). h e r die Konzentration von Kynurenin und HydroxyKynurenin wahrend der Entwicklung bei den Genolypen ra’ und ra von Plodia interpunctella. Naturwissenschaften, 48, 27-28. Almeida, F. F. de (1968). Das Ommochrom I als wahrscheinliche Vorstufe des Ommins in der Ommochromsynthese von Plodia interpunctella L.nd Ephestia kiihniella. Molec. gem Genetics, 102, 307-315. Arcaya, G., Pantoja, M. E.. Pieber, M., Romero, C. and Tohi, J. C. (1971). Molecular interaction of L-tryptophan with bases, ribonucleosides and DNA. 2. Naturf: 26b, 1026-1030. Atkinson, D. E. (1969). Limitation of metabolite concentrations and the conservation of solvent capacity in the living cell. In “Current Topic:i in Cellular Regulation” (Eds B. L. Horecker and E. R. Stadtman), Vol. 1; pp. 29-43. Academic Press, New York and London. Autrum, H. (1955). Die spektrale Empfindlichkeit der Augenmutation white-apricot von Calliphora erythrocephala. Biol. Zbl. 74, 51 5-524. Autrum, H. (1961). Die Sehschtirfe pigmentfreier Fazettenaugen von Calliphora erythrocephala Biol. Zbl. 80, 1-4. Baglioni, C. (1959). Genetic control of tryptophan peroxidase-oxidase in Drosophila melanogaster. Nature, Lond. 184, 1084-1085. Baglioni, C. (1960). Genetic control of tryptophan pyrrolase in Drosophila melanogaster and D. uirilis. Heredity, 15, 87-96. Baillie, D. L. and Chovnick, A. (1971). Studies on the genetic control of tryptophan pyrrolase in Drosophila melanogaster. Molec. gen. C e n e t c s , 112, 341-353. Barnes, F. J. and Goodfellow, R. D. (1971). Mevalonate kinase: localization and variation in activity during the development of Sarcophaga bullata. J. Insect PhysioL 17, 1415-1427. Bauer, A. C . and Levenbook, L. (1969). Fructose diphosphate aldolase during growth and development of the blowfly, Phormia regina (Meigen). Comp. Biochem. Physiol. 28, 619-632. Baumann, U. (1957). Untersuchungen iiber Ommochrome. h e r Bombyx-Ommin und iiber den oxydativen Abbau substituierter Phenoxazone. Doctoral Thesis, University of Tiibingen. Beadle, G. W. (1937). Development of eye colors in Drosophila: fat bodies and Malpighian tubes as sources of diffusible substances. Proc. natn. Acad. S c i U.S.A. 23. 146-152. Beadle, G. W. and Ephrussi, B. (1935). Diffkrenciation de la couleur de I’oeil cinnabar chez la Drosophile (Drosophila melanogaster). C.7. h e b d Sianc. Acad. Sci, Paris, 201. 620-622.

226

BERNT LINZEN

Beadle, G. W., Anderson, R. L. and Maxwell, J. (1938). A comparison of the diffusible substances concerned with eye color development in Drosophila, Ephestia and Habrobracon. Proc. natn. Acad. Sci U.S.A. 24,80-85. Becker, E. (1938). Die Gen-Wirkstoff-Systeme der Augenausfirbung bei Insekten. Naturwissenschaften, 26,433441. Becker, E. (1939). d e r die Natur des Augenpigments von Ephestia kiihniella und seinen Vergleich mit den Augenpigmenten anderer Insekten. Biol. ZbL 59, 597-627. Becker, E. (1941a). Die Pigmente der Ommin- und Ommatingruppe, eine neue Klasse von Naturfarbstoffen. Natunuissenschaften, 29,237-238. Becker, E. (1941b). Ein Beitrag zur Kenntnis der Libellenpigmente. Biol. Zbl. 61, 588-602. Becker, E. (1942). Uber Eigenschaften. Verbreitung und die genetisch-entwicklungsphysiologische Bedeutung der Pigmente der Ommatin- und Ommingruppe (Ommochrome) bei den Arthropoden. Z. VererbLehre, 80, 157-204. Beckmann, R. (1956). h e r Rhodommatin, ein natiirlich vorkommendes Ommochrom. Doctoral Thesis, University of Tiibingen. Bellamy, D. (1958). The structure and metabolic properties of tissue preparations from Schistocerca gregaria (Desert Locust). Biochem. J. 70, 580-589. Benassi, C. A., Veronese, F. M. and Scoffone, E. (1964). Separazione cromatografica di derivati biologici del triptofano su resine a scambio ionico. Ann. Chim. 54, 627-638. Berthold, G. (1971). h e r die hormonale Steuerung von Wachstum und Pigmentierung der Indischen Stabheuschrecke Carausius morosus Br. Doctoral Thesis. University of Giessen. Berthold, G. und Henze, M. (1971). h e r die Beteiligung der Pterine an den Farbmodifikationen der Stabheuschrecke, Carausius morosus. J. Insect Physiol. 17, 2375-238 1. Biedermann, W. (1914). Farbe und Zeichnung der Insekten. In “HandbuCh der vergleichenden Physiologie” (Ed. H. Winterstein), Vol. 3/1/11, pp. 1657-1994. Fischer-Verlag, Jena. Block, R. J. and Weiss, K. W. (1956). “Amino Acid Handbook”. Charles C. Thomas, Springfield. Bonse, A. (1969). Uber d a s Auftreten von Pterinen, Tryptophan und dessen Derivate in verschiedenen Organen der Mutante white von Drosophila melanogaster. Z . Nature 24b, 128-131. Bouthier, A. (1966). Modifications des pigments (ommochromes et pterines) en relation avec la mutation albinos chez Locusta mipatoria cinerascens Fabr. (Orthoptkres, Aaididae). C.r. hebd. Sianc. Acad. Sci., Paris, Sir. D. 262, 1480-1483. Bouthier, A. (1969). Isolement d’ommidine i 1’Ctat cristallisk partir d’yeux de Locusta migratoria L. (Orthopdres, Acrididae). C.r. hebd. S*anc. Acad. S c i . Paris, 268, 1410-14 12. Bouthier, A. (1972). The ommochromes of the Migratory Locust. NATO A d a Stud fnst. Chemistry of Insects, Varenna (Unpublished.) Bowness, J. M. and Wolken, J. J. (1959). A light-sensitive yellow pigment from the house-fly. J. gen. Physiol. 42, 779-792. Bratton, A. C. and Marshall, E. K. (1939). A new coupling component for sulfanil-amide determination. J. biol. C h e m 128,537-550.

THE TRYPTOPHAN

-+

OMMOCHROME PATHWAY IN INSECTS

227

Britten, R. J. and Davidson, E. H. (1969). Gene regulation for higher cells: a theory. Science (WashinKton), 165, 349-357. Brown, K. S. (1965). A new L-a-amino acid from Lepidoptera. J. A m . chem. SOC. 87, 4202-4203. Brown, K. S. and Becher, D. (1967). The mass spectra of’the kynurenines. Tetrahedron Lett. 1967, 1721-1726. Brown, R. R. and Price, J. M. (1956). Quantitative studies. on metabolites of tryptophan in the urine of the dog, cat, rat, and man. J biol. Chem. 219, 985-997. Brunet, P. C. J. (1965). The metabolism of aromatic conipounds. In “Aspects of Insect Biochemistry” (Ed. T. W. Goodwin), pp. 49-77. Biochem. SOC. Symp. No. 25. Academic Press, London and New York. Biickmann, D. (1953). Uber den Verlauf und die Auslosiing von Verhaltensanderungen und Urnfirbungen erwachsener Schmetterlingsraupen. IlioL Zbl. 72, 276-3 1 1. Biickmann, D. (1959a). Die Auslosung der Umfarbung durch das Hautungshormon bei Cerura uinula L. (Lepidoptera, Notodontidae). J. Insect Physiol. 3, 159-189. Biickmann, D. (1959b). Farbwechsel als Teilvorgang tler Metamorphose und seine Hemmung in jungen Schmetterlingsraupen. Zool. A m . SuppL 22, 137-144. Biickmann, D. (1963). Der Einfluss der Temperatur auf das Epidermispigment der Raupen von Cerura uinula L. Z . Naturf. 18b, 255-264. Biickmann. D. (1964). Zur Frage einer respiratorischen Funktion von Ommochromen. Nuturwissenschaften, 51, 344. Biickmann, D. (1965). Das Redoxverhalten von Ommochrom in uiuo. Untersuchungen an Cerura uinula L. J. Insect Physiol. 11, 1427-1462. Biickmann, D. and Dustmann, J. H. (1962). Biochemische Untersuchungen iiber den morphologischen Farbwechsel von Carausius morosus. Natunuissenschaften, 49,379. Biickmann. D., Willig, A. and Linzen, B. (1966). Verindelungen der Hamolymphe vor der Verpuppung von Cerura uinuia L.: Der Gehalt an Eiweiss, Aminosauren, Ommochrom-Vorstufen und Ommochromen. Z. Nuturf. 21b, 1184-1195. Burnet, B. and Sang, J. H. (1968). Physiological genetics of melanotic tumoun in Drosophila melanogaster. V. Amino acid metabolism and tumour formation in the tu b w ; s t su-tu strain. Genetics, 59, 211-235. Burnet, B., Connolly, K. and Beck, J. (1968). Phenogenetic studies on visual acuity in Drosophila rnelanogaster. J. Inscct Physiol. 14, 855-866. Butenandt, A. (1949). Biochemische Untersuchungen zur Wirkungsweise der Erbfaktoren. Angew. C h e m 61, 262-263. Butenandt. A. (1957). Uber Ommochrome, eine Klasse natiirlicher PhenoxazonFarbstoffe. Angew. C h e m 69, 16-23. Butenandt, A. (1959). Wirkstoffe des Insektenreiches. Natunuissenschaften, 46, 461-47 1. Butenandt, A. (1960). Uber neue Naturfarbstoffe. ihre Biogenese und physiologische Bedeutung. XVII. Int. Congr. Pure Appl. Chem. Munich, ‘Vol. 11, 11-31. Butenandt, A. and Albrecht, W. (1952). Bestimmungen des Tryptophangehaltes verschiedener Rassen der Mehlmotte Ephestia kiihniella a l s Beitrag zur Analyse der Genwirkungen. Z . Nuturf. 7b, 287-290. Butenandt, A. and Hdlmann, G. (1950). Neue Synthesen des d,l-Kynurenins und d,l-3-Oxy-Kynurenins. 2. Nuturf 5b. 444-446.

228

BERNT LINZEN

Butenandt, A. and Neubert, G. (1955). Uber Ommochrome, V. Mitteilung. Xanthommatin, ein Augenfarbstoff der Schmeissfliege Calliphora erythrocephala. HoppeSeyler’s Z . physiol. C h e m 301, 109-114. Butenandt, A. and Neubert, G. (1958). Uber Ommochrome, XVII. Zur Konstitution der Ommine I. Liebig’s Ann. C h e m 618, 167-173. Butenandt, A. and Schafer, W. (1962). Ommochromes. In “Recent Progress in the Chemistry of Natural and Synthetic Colouring Matters and Related Fields” (Eds T. S. Gore, B. S. Joshi, S. V. Sunthankar and B. D. Tilak), pp. 13-33. Academic Press, New York and London. Butenandt, A., Weidel, W. and Becker, E. (1940a) Kynurenin als Augenpigmentbildung ausliisendes Agens bei Insekten. Naturzuissenschaften, 28, 63-64. Butenandt, A., Weidel, W. and Becker, E. (1940b). a-Oxytryptophan als “Prokynurenin” in der zur Augenpigmentbildung fiihrenden Reaktionskette bei Insekten. Natunuissenschaften, 28,447-448. Butenandt, A., Weidel, W. and Derjugin, W. von (1942). Zur Konstitution des Kynurenins. Naturwissenschaften, 30, 51. Butenandt, A., Weidel, W., Weichert, R. and Derjugin, W. von (1943). Uber Kynurenin. Physiologie, Kostitutionsermittlung und Synthese. Hoppe-Seyler’s Z . physiol. Chem. 279, 27-43. Butenandt, A., Weidel, W. and Schlossberger, H. (1949). 3-Oxy-Kynurenin als cn+-Gen-abhingiges Glied im intermediiren Tryptophan-Stoffwechsel.Z. Naturf. 4b, 242-244. Butenandt, A., Karlson, P. and Zillig, W. (1951). Uber das Vorkommen von Kynurin in Seidenspinnerpuppen. Hoppe-Seyler’s Z. physiol. Chem. 288, 125-129. Butenandt, A., Schulz, G. and Hanser, G. (1953). Uber 5-Oxy-kynurenin, seine Synthese und physiologische Bedeutung. Hoppe-Seyler’s Z. physiol. Chem. 295, 404-41 0 . Butenandt, A., Schiedt, U., Biekert, E. and Kornmann, P. (1954a). Uber Ommochrome, I. Mitteilung: Isolierung von Xanthommatin, Rhodommatin und Ommatin C aus den Schlupfsekreten von Vanessa urticae. Liebig’s Ann. Chem. 586, 21 7-228. Butenandt, A., Schiedt, U. and Biekert, E. (1954b). Uber Ommochrome, 11. Mitteilung: Alkalischer und fermentativer Abbau von Xanthommatin und Rhodommatin. Alkalischer Abbau der Kynurenin-Seitenkette. Liebigs Ann. Chem. 586, 229-239. Butenandt, A.. Schiedt, U. and Biekert. E. ( 1 9 5 4 ~ )h. e r Ommochrome, HI. Mitteilung: Synthese des Xanthommatins. Liebig’s Ann. Chem. 588, 106-116. Butenandt, A., Schiedt, U.. Biekert, E. and Cromartie, R. J. T. (1954d). Uber Ommochrome, IV. Mitteilung: Konstitution des Xanthommatins. Liebig’s A n n C h e m 590. 75-90; Butenandt, A., Biekert, E. and Linzen, B. (1956). Uber Ommochrome, VII. Mitteilung. Modellversuche zur Bildung des Xanthommatins in uiuo. Hoppe-Seyler’s Z . physioL C h e m 305, 284-289. Butenandt, A., Keck, J. and Neubert, G. (1957a). h e r Ommochrome VIII. ModellVersuche zur Konstitution der Ommochrome: Uber Oxydationsprodukte der 3-Hydroxy-anthranilsiure. Liebigs Ann. Chem. 602, 61-72. Butenandt, A., Biekert, E. and Neubert, C. (1957b). h e r Ommochrome IX. Modell-Versuche zur Konstitution der Ommochrome: h e r 3-Hydroxy-5-acetylphenoxazon-(2) und 1.6-Diacetyl-triphendioxazin. Ein neuer Weg zur Darstellung von Phenoxazonen. Liebigs A n n Chem. 602, 72-80.

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

229

Butenandt, A., Biekert, E. and Neubert, G. ( 1 9 5 7 ~ ) Uber . Ommochrome X. Modellversuche zur Konstitution der Ommochrome: 5-Dibi:nz(b.f)azepinchinonirnine als Reaktionsprodukte der 5-Acetyl-phenoxazone mit Alkali und als Oxydationsprodukte des 2-Amino-3-hydroxy-acetophenons. Liebig’s Ann. Chem. 603, 200-206. Butenandt, A., Biekert, E. and Baumann, U. (1957d). 13mmochrome. XI. Modellversuche zur Konstitution der Ommochrome: Oxydativer Abbau des 3-Amino-4,5diacetylphenoxazons-(2).Arch. Biochem Biophys. 69, 100-105. Butenandt, A., Biekert, E. and Beckmann, R. (1957’:). Uber Ommochrome, XII. Reindarstellung von Rhodommatin und Ommatin D. Zur Struktur des Rhodommatins. Liebig’s Ann. C h e m 607, 207-215. Butenandt, A., Hallmann, G. and Beckmann, R. (1957:‘). Synthesen des 3-Hydroxykynurenins. Chem. Ber. 90, 1120-1124. Butenandt, A., Biekert, E. and Linzen, B. (1958a). Gber Ommochrome, XIII. Isolierung und Charakterisierung von Omminen. Ifoppe-Seyler’s Z. physiol. Chem. 31 2, 227-236. Butenandt, A., Biekert, E. and Linzen, B. (1958b). Uber Ommochrome, XIV. Zur Verbreitung der Ommine im Tierreich. Hoppe-Seylttr’s Z. physiol. Chem. 313, 251-258. Butenandt, A., Neubert, G. and Baumann, U. (1959). h e r Ommochrome, XVI. Versuche zur Reindarstellung eines Ommins. Hoppe-StylerS Z. physiol. Chem. 3 14, 15-21. Butenandt, A., Biekert, E. and Schafer, W. (1960a). Uber Ommochrome, XVIII. Modellversuche zur Konstitution der Ommochronie: Die Kondensation von Hydroxy-chinonen mit o-Amino-phenolen. Liebig’s Ann. Chem. 6 3 2 , 134-143. Butenandt, A., Biekert, E., Kiibler, H. and Linzen, B. (19611b). Uber Ommochrome, XX. Zur Verbreitung der Ommatine im Tierreich. Neue Methoden zu ihrer Identifizierung und quantitativen Bestimmung. Hoppe-Seyler’s Z. physisl. Chem. 3 19, 238-256. Butenandt, A., Biekert, E., Koga, N. and Traub, P. (19601:). Uber Ommochrome, XXI. Konstitution und Synthese des Ommatins D. Hoppe-Seyler’s Z. physiol. Chem. 321. 258-275. Butenandt, A., Biekert, E., Kubler, H., Linzen, B. and Traub, P. (1963). Uber Ommochrome, XXII. Konstitution und Synthese des Rhodommatins. Hoppe-Seyler’s Z. physiol. Chem. 334, 71-83. Butenandt, A., Biekert, E. and Hark, E. (1964). Darstellung und Umwandlungen der 4,5,8-Trihydroxy-6-mercapto-chinolincarbonsaure-(2). Chem. Ber. 97, 285-294. Butenandt, A., Schafer, W. and Neubert, G . (1967). u b e r Ommochrome, 24. Mitt. Zur Konstitution der Ommine 11. Monatsh. Chem. 98, 946-935. Calam, D. H. (1972). Indolylacetic acid in the salivary glands of Dasyneura larvae. N A T O Adu. Stud. Inst. Chemistry of Insects, Varenna, 1972. (Unpublished.) Caspari, E. (1933). Uber die Wirkung eines pleiotropen Gens bei der Mehlmotte Ephestia kiihniella Zeller. Roux’ Arch. EntwMech. Org. :L30, 353-381. Caspari, E. (1949). Physiological action of eye color mucants in the moths Ephestia kiihniella and Ptychopoda seriata. Q. Rev. Biol. 24, 185-199. Caspari, E. and Richards, J. (1947/1948). Genic action. Carnegie fnst. Washington Year Book, 47, 183-189. Cassady, W. E. and Wagner, R. P. (1971). Separation of mitochondria1 membranes of Neurospora crassa I. Localization of L-kynurenine-3-hydroxylase.J. Cell B i d . 49, 536-541.

230

BERNT LINZEN

Chauvin, R. (1938).Sur le rougissement du criquet pelerin. C.r. hebd. Sianc. Acad. S c i , Paris, 207, 1018-1020. Chauvin, R. (1 939). Sur le pigment rose rejete dans les excrkmknts par le Criquet pelerin au moment de la mue. C.r. Sianc. SOC.Biol. 130. 1194-1195. Chinzei, Y. and Tojo, S. (1972).Nucleic acid changes in the whole body and several organs of the silkworm, Bombyx mori, during metamorphosis. J. Insect Physiol. 18,

1683-1698. Colombo, G. and Pinamonti, S. (1965).Metaboliti del triptofano “via chinurenina” in Schistocerca gregaria Forsk durante lo sviluppo embrionale. Rend. Classe Sci fis. mat. nut. Acc. Naz. Lincei, Ser. VIII. 38, 728-731. Colombo, G. and Pinamonti, S. (1967).Analisi biochimica di alcuni mutanti di Musca domestica L. A t t i A s s . genet. Ital. 12, 341-344. Colombo, G., Franco, P. and Moccellin, E. (1955).Sulla pigmentazione degli Acridoidei. Ricerche sulla pigmentazione delle larve di Anacridium aegyptium L. Boll. Zool.

Turin,22. 309-322. Cott, H. B. (1940/66).“Adaptive Coloration in Animals”. Methuen, London. Cromartie, R. J. T. (1959).Insect pigments. A . Rev. Ent. 4, 59-76. Dadd, R. H. (1970).Arthropod nutrition. In “Chemical Zoology” (Eds M. Florkin and B. T. Scheer), Vol. V/A, pp. 35-95.Academic Press, New York and London. Dales, R. P. (1962).T h e nature of the pigments in the crowns of sabellid and serpulid polychaetes. J. mar. biol. Ass. U.K. 42. 259-274. Danneel, R. (1941).Die Ausfarbung uberlebender v- und cn-Drosophila-Augen mit Produkten des Tryptophanstoffwechsels. Biol. Zbl. 61,388-399. Danneel, R. and Zimmermann, B. (1954).Uber das Vorkommen und Schicksal des Kynurenins bei verschiedenen Drosophilarassen. Z. Naturf. 9b, 788-792. Davis, G. R. F. (1968).Quantitative requirements of the saw-toothed grain beetle, Oryzaephilus surinumensis, for dietary L-tryptophan. J. Insect Physiol. 14,

1693-1696. De Antoni, A., Allegri, G. and Costa, C. (1970).Xantommatina per via enzimatica. Gazz. c h i m Ital. 100. 1050-1055. Dennhofer, U. (1971). Augenfarbmutationen bei der Stechmiicke Culex pipiens L.-ihre chemische und genetische Ursache. Z. Nuturf. 26b, 599-603. Dimicoli, J.-L. and Htlkne, C1. (1971). Interactions entre acides a m i n b et acides nuclkiques. 111-Etude par absorption et resonance magnetique nuclkaire de la formation de complexes entre le tryptophane et les constituants des acides nuclkiques. Biochimie, 53, 331-345. Dustmann, J. H. (1964). Die Redoxpigmente von Carausius morosus und ihre Bedeutung fur den morphologischen Farbwechsel. Z. uergl. Physiol. 49. 28-57. Dustmann. J. H. (1966).Uber Pigmentuntersuchungen an den Augen der Honigbiene Apis mellifica. Naturwissenschaften, 53, 208. Dustmann, J. H. (1968).Pigment studies on several eye-colour mutants of the honey bee, Apis mellifera. Nature, Lond. 219, 950-952. Dustmann, J. H. (1969). Eine chemische Analyse der Augenfarbmutanten von Apis mellifica. J. Insect Physiol. 15, 2225-2238. Egelhaaf, A. (1 956a). Quantitativer Vergleich der Fluoreszenzmuster der Eier dreier Genotypen von Ephestin kiihniella Natunuirsenschaften, 43, 165-166. Egelhaaf, A. (1956b).Die fluoreszierenden Stoffe in den Organen dreier Genotypen von

THE TRYPTOPHAN

--*

OMMOCHROME PATHWAY IN INSECTS

231

Ephestia kiihniella. Beitrag zur Analyse von Genwirkungen in Pterin- und Tryptophanstoffwechsel. Z . VererbLehre, 8 7, 769-783. Egelhaaf, A. (1957). Der Gehalt an freiem Tryptophan und Kynurenin bei den Genotypen a+ und a von Ephestia kiihniella w a r e n d der Entwicklung. Z. Naturf.

12b. 465-472. Egelhaaf, A. (1958).Nachweis eines genabhkgigen L-Tryptophanoxydase-Systemsbei Ephestia kiihniella. Z . Naturf. 13b, 275-279. Egelhaaf, A. (1963a).Uber das Wirkungsmuster des a-Locus von Ephestia kiihniella. Z . VererbLehre, 94,348-384. Egelhaaf, A. (1963b).Biochemische Wirkungen des a-Locus von Ephestia kiihniella in der Ei- und Larvenentwicklung. Zool. Anr. Suppl. 26. 79-84. Egelhaaf, A. and Caspari, E. (1960).h e r die Wirkungswe:.se und genetische Kontrolle des Tryptophanoxydasesystems bei Ephestia kiihniella. 2 . VererbLehre, 91, 373-379. Eichelberg, D. (1968). Quantitative Untersuchungen uber die Ommochromvorstufen in den Malpighischen Gefassen verschiedener Drosophila-Mutanten. Z Naturf. 23b,

547-554. Eichelberg, D. and Wessing, A. (1971). Elektronenoptische Untersuchungen an den Nierentubuli (Malpighische Gefasse) von Drosophila melanogaster XI. Transzelluliire membrangebundene Stofftransportmechanisnien. Z . i’ellforsch. mikrosk. Anat.

121, 127-152. Ephrussi, B. (1942a).Analysis of eye color differentiation in Drosophila. Cold Spring Harb. Symp. quant. Biol. 10,40-48. Ephrussi, B. (1942b).Chemistry of “eye color hormones” of Drosophila. Q. Rev. Biol.

17, 327-353. Ephrussi, B. and Beadle, G. W. (1935).La transplantation dcs disques imaginaux chez la Drosophile. C.r. hebd. Sdanc. Acad. Sci.. Paris, 201.98-100. Fearon, W. R and Boggust, W. A. (1950). Pigments derived from tryptophan: (1) urorosein, (2)tryptochrome. Biochem. J. 46, 62-67. Feigelson, P. and Maeno, H. (1967).Studies on entyme-substrate interactions in the regulation of tryptophan oxygenase activity. Biochem. biophys. Res. Commun 28.

289-293. Fingerman, M. (1952).The role of the eye pigments of L)rosophila melanogaster in photic orientation. J. exp. Zool. 120, 131-164. Friedman, M. and Finley, J. W. (1971).Methods of tryptophan analysis. J. agr. Fd C h e m 19.626-631. Fuge, H. (1967).Die Pigmentbildung im Auge von Drosopkila melanogaster und ihre Beeinflussung durch den white+-Locus. Z . Zellforsch. mikrosk. Anat. 83,468-507. Fujimori, E. (1959).Interaction between pteridines and tryptophan. Proc. natn. Acad. Sci. U.S.A. 45, 133-136. Fukuda, T.,Kirimura, J., Matuda, M. and Suzuki, T. (1955).Biochemical studies on the formation of the silk protein-I. ?he kinds of free amino acids concerned in the biosynthesis of the silk protein. J. Biochem., Tokyo. 42, 341-346. Fuzeau-Braesch, S. (1957). Fractionnement des pigments o:cydort$ucteurs de I’hypoderme et des yeux d’un grillon. C.r. hebd. Sdanc. Acad. S 8 i , Paris, 245,2401-2404. Fuzeau-Braesch, S. (1968).Recherche de la xanthommatine chez deux insectes: Locusta migratoria L. et. Oedipoda coerulesens L. C.Y. hebd. Sdanc. Acad. S c i , Paris, 267,

2030-2033.

232

BERNT LINZEN

Fuzeau-Braesch, S. (1971). Observation de granules pigmentaires (ommochromes et ptkidines) du ttgument d'insectes au microscope klectronique i balayage. C.r. hebd SPanc. Acad. Sci., Paris, Sir. D. 272, 3322-3324. Fuzeau-Braesch, S. (1972). Pigments and color changes. A . Rev. Ent. 17, 403-424. Galarza, A. M. and Stamm, M. D. (1959). Estudios sobre omocromes: 11.-datos sobre omocromes en insectos. Act. V. Reunwn An. SOC. Espan. Cienc. Fisiol. Madrid, 1959. Ghelelovich, S . (1969). Melanotic tumors in Drosophila melanogaster. Natl. Cancer Inst. Monogr. No. 31, p. 263. Ghosh, D. and Forrest, H. S. (1967a). Enzymatic studies on the hydroxylation of kynurenine in Drosophila melanogaster. Genetics, Princeton, 55, 4 2 3 4 3 1. Ghosh, D. and Forrest, H. S. (1967b). Inhibition of tryptophan pyrrolase by some naturally occurring pteridines. Arch. Biochem. Biophys. 120, 578-582. Giersberg, H. ( 1928). h r den morphologischen und physiologischen Farbwechsel der Stabheuschrecke Dixippus morosus. Z. vergl. Physiol. 7, 65 7-695. Ginoulhiac, E., Tenconi, L. T. and Fabro, S. (1962). Ricerche sulla chinureninaidrossilasi: (1) Metodo di dosaggio e determinazione in diversi organi del ratto. Boll. SOC.ital. Biol. sper. 38. 1803-1807. Glassman, E. (1956). Kynurenine formamidase in mutants of Drosophila Genetics, Princeton, 41, 566-574. Goldsmith, T. H. and Philpott, D. E. (1957). The microstructure of the compound eyes of insects. J. biophys. biochem Cytol. 3, 429-440. Goodwin, T. W. (1952). The biochemistry of locust pigmentation. Biol. Rev. (Cambridge), 27, 439-460. Goodwin, T. W. and Srisukh, S. (1950). Biochemistry of locusts. 3. Insectorubin: the redox pigment present in the integument and eyes of the Desert Locust (Schistocerca gregaria Forsk.), the African Migratory Locust (Locusta mipatoria migratorioides R. and F.) and other insects. Biochem. 1.47, 549-554. Gorini, L. and Beckwith, J. R. (1966). Suppression. A . Rev. Microbiol. 20,401-422. Gotz, K. G . (1964). Optomotorische Untersuchung des visuellen Systems einiger Augenmutanten der Fruchtfliege Drosophila Kybemetik, 2, 77-92. Graf, G . E. (1957). Biochemical predetermination in Drosophila. Experientia, 13, 404-405. Grassmader, H.-H. ( 1968). Untersuchungen iiber die Ommochromsynthese in Wildform und Mutante c von Calliphora erythrocephala. Ph.D. Thesis, University of Wiirzburg. Graszynski. K. (1 970). Intrazellulire Lokalisation von Enzymen des Flusskrebses. Enzyme der Fettsaureoxydation und des Ketonkorperstoffwechsels, Dehydrogenasen des energieliefernden Stoffwechsels und Proteasen der Mitteldanndriise. Z. vergl. Physiol. 66, 107-122. Green, M. M. (1949). A study of tryptophane in eye color mutants of Drosophila Genetics 34, 564-572. Green, M. M. (1952). Mutant isoalleles at the vermilion locus in Drosophila melanogaster. Proc. natn. Acad. Sci. U.S.A. 38, 300-305. Green, M. M. (1954). Pseudo-allelism at the vermilion locus in Drosophila melanogaster. Proc. natn. Acad. Sci. U.S.A. 40, 92-99. Gripenberg, J. (1958). Fungus pigments VIII. The structure of cinnabarin and cinnabarinic acid. Acta c h e m scand. 12, 603-610.

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

233

Gripenberg, J., Honkanen, E. and Patoharju, 0. (1957). Fungus pigments V. Degradations of cinnabarin. A c t a chem. scand. 11, 1485-1492. Gutmann, H. R. and Nagasawa, H. T. (1959). The oxidation of o-aminophenols by cytochrome c and cytochrome oxidase. J. biol. Chem. :!34, 1593-1599. Hadorn, E. and Kiihn, A. (1953). Chromatographische und fluorometrische Untersuchungen zur biochemischen Polyphanie von Ac.genfarb-Genen bei Ephesticr kiihniella. Z. Naturf. 8b, 582-589. Hanser, G. (1946). Tragergranula und Ommochroinpigmente in den Augen verschiedener Rassen und Transplantat-Trager von Ephestia kiihniella Z. und Ptychopoda seriata Schrk. Z . Naturf. 1, 396-399. Hanser. G. (1948). h e r die Histogenese der Augenpigmentgranula bei verschiedenen Rassen von Ephestia kiihniella Z. und Ptychopoda seriata Schrk. 2. VererbLehre. 82, 74-97. Hanser, G. (1959). Genphysiologische Untersuchungen an tier w-Mutante von Calliphora erythrocephala. 2. Naturf. 14b, 194-201. Harano, T. and Chino, H. (1971). A new diaphorase from Bombyx silkworm eggs-cytochrome c reductase activity mediated with xanthommatin. Archs Biochem. Biophys. 146, 467-476. Harmsen, R. (1966). The excretory role of pteridines in insects. J . exp. Biol. 45, 1-13. HtlPne, C1. (1971). Role of aromatic amino-acid residues in the binding of enzymes and proteins to nucleic acids. Nature, N e w Biol. 234, 120-1211. Htlkne, Cl., Dimicoli, J.-L. and Brun, P. (1971). Binding of tryptamine and 5-hydroxytryptamine (serotonin) to nucleic acids. Fluorescence and proton magnetic resonance studies. Biochemistry. N. Y . 10, 3802-3809. Henderson, L. M., Gholson, R. K. and Dalgliesh, C. E. (1962). Metabolism of aromatic amino acids. In “Comparative Biochemistry” (Eds M. Florkin and H. S. Mason), Vol. 4, pp. 245-342. Academic Press, New York and London. Hendrichs-Hertel, U. and Linzen, B. (1969). Uber die Kynurenin-3-Hydroxylase bei Calliphora erythrocephala. Z. vergl. Physiol. 64,411-431. Hengstenberg, R. and Gotz, K. G. (1967). Der Einfluss des Schirmpigmentgehalts auf die Helligkeits- und Kontrastwahrnehmung bei Drosophila-Augenmutanten. Kybernetik, 3,276-285. Henning, H. D. (1957). Unpublished experiments at the Max Planck lnstitute of Biochemistry, Munich. Hertel, U. (1968). Untersuchungen iiber die Kynurenin-3-Hydroxylase bei Insekten. Ph.D. Thesis, University of Munich. Hertweck, H. (1960). Notes on the formation of a red eye pigment in the Malpighian tubules and fat body in Drosophila melanogaster. Folia. biol., Praha, 6 , 293-296. Hinton, T., Noyes, D. T. and Ellis, J. (1951). Amino acids and growth factors in a chemically defined medium for Drosophila. Physiol. 2001. 24, 335-353. Hiraga, S. (1964). Tryptophan metabolism in eye-color muta.nts of the housefly. Jap. J. Genet. 39,240-253. Hirata, Y. and Nakanishi, K. (1950). The synthesis of 3-hydroxykynurenine (+ chromogen). Bull. chem. SOC.,Japan, 23, 134-137. Hirata, Y.,Nakanishi, K. and Kikkawa, H. (1949). Crystals.of + chromogen obtained in silkworms. Jap. J. Genet. 24, 190. (In Japanese.)

234

BERNT LINZEN

Hirata, Y., Nakanishi, K. and Kikkawa, H. (1950). Crystalline + chromogen obtained from Bombyx mori. Science, 112, 307-308. Hoglund, G., Langer, H., Struwe, G. and Thorell, B. (1970). Spectral absorption by screening pigment granules in the compound eyes of a moth and a wasp. Z. uergl. Physiol. 67. 238-242. Horikawa, M. (1958). Developmental-genetic studies of tissue cultured eye discs of D. nielanogaster I. Growth, differentiation and tryptophan metabolism. Cytologia ( T o k y o ) , 23,468-477. Horn, U., Ullrich, V. and Staudinger, H. (1971). Purification and characterization of L-kynurenine 3-hydroxylase (EC 1.14.1.2) from rat liver. Hoppe-Seyler’s Z. physiol. C h e m 352, 837-842. Horowitz, N. H. (1940). A respiratory pigment from the eggs of a marine worm. R o c . natn. Acad. Sci. U.S.A. 26, 161-163. Horowitz. N. H. and Baumberger, J. P. (1941). Studies on the respiratory pigment of Urechis eggs.]. biol. C h e m 141,407-415. Horstmann, G . (1971). Uber die Pigmentgranulogenese im Auge von Ephestin kiihniella Z. Z . Naturf. 26b, 484-485. Inagami, K. (1954a). Chemical and genetical studies on the formation bf the pigment in the silkworm-111. On the microanalysis of 3-hydroxy-kynurenine. J. Sericult. Sci Japan, 23,299-303. (In Japanese.) (1955). Chem. Abstr. 4 9 , 1 0 5 3 6 ~ . Inagami, K. (1954b). Mechanism of the formation of red melanin in the silkworm. Nature, Lond. 174, 1105. Inagami, K (1955). Chemical and genetical studies o n the formation of pigment and metabolism of tryptophan in the silkworm-X. Biochemical studies on the “Red Blood-rb”gene in the silkworm.]. agric. chem. SOC.Japan, 29, 12. (In Japanese.) Inagami, K. (1958). The mechanism of formation of tryptophan pigments and the metabolism of tryptophan in silkworm. Bull. Kumamoto sericult. Exp. Stat. 6 , 5-58. (In Japanese.) Isenberg, I. and Szent-Gyorgyi, A. (1958). Free radical formation in riboflavin complexes. R o c . natn. Acad. Sci U.S.A. 44, 857-862. Ishiguro, I. and Linzen. B. (1965). Praparative Isolierung und Tritium-Markierung von 3-Hydroxy-L-kynurenin. Hoppe-Seyler’s Z. physiol. Chem. 340. 286-288. Ishiguro, I. and Linzen, B. (1966). Bildung von Glucosiden bei zwei Insektenarten, Locusta migratoria und Bombyx mori J. Insect Physiol. 12, 267-273. Ishiguro, I. and Nagamura, Y. (1971a). The red pigments in feces urine of the silkworm, Bombyx m o r i Medicine and Biology, 82, 149-153. Ishiguro, I. and Nagamura, Y. (1971b). Identification of xanthommatin and isolation of red-protein in the body-fluid of rb-mutant of the silkworm. Medicine and Biology, 82, 123-126. Ishiguro, I., Nagamura, Y. and Hara, A. (1971a). Studies on the formation of phenoxazine pigment from o-aminophenol derivatives by hemoglobin. 1. Conversion of 3-OH-anthranilic acid into cinnabarinic acid in the presence of MnZ+. Yukugaku Zasshi, 91,760-765. Ishiguro, I., Nagamura, Y. and Yamamoto, T. (1971b). Formation of phenoxazinepigment and conversion of 3-hydroxy-kynurenine in the silkworm with metamorphosis. Seikagaku, 43, 401-407.

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

235

Iwata, S. and Ogata, K. (1966). Large scale isolation of L-3-hydroxy-kynurenine from mutant strain (aka-aka) of silkworm larva. Yakugaku Zasshi, 86,249-250. Jacobson, K. B. (1971). Role of an isoacceptor transfer ribonucleic acid as an enzyme inhibitor: effect on tryptophan pyrrolase of Drosophila. Nature, New Biol. 23 1, 17-19. JansCn, C. T., Santti, R. S. and Hopsu-Havu, V. K. (1969). Kynurenine formamidase in rat skin. J. invest. Derm. 52, 430-436. Jezewska, M. (1926). Les changements de la teneur en tryptophane au cours de dkveloppement des chrysalides des mouches. C.r. Siunc SOC. Biol. 95, 910-912. Joshi, S. and Brown, R. R. (1959). Pigment formation from hydroxy-kynurenine by a rat liver system. Proc. Fedn. A m . SOCSexp. Biol. 18,255. Kalmus, H. (1943). The optomotor responses of some eye mutants of Drosophila. J. Genet. 45, 206-213. Kanehisa, T. (1956a). Eye-colour genes and tumour incidence. Jap. J. Genet. 31, 144-146. Kanehisa, T. (1956b). Relation between the formation of melanotic tumours and tryptophane metabolism involving eye-colour in Drosophila. Annotnes Z O O L jap. 29, 97-100. Kanehisa, T. and Fujii, S. (1967). Further biochemical studies on “melanotic tumour” and “freckled-type melanotic tumor” in D. melanogaster. Jap. J. Genet. 42, 299. Karlson, P. and Buckmann, D. (1956). Experimentelle Auslosung der Umfikbung bei Ceruru-Raupen durch Prothorakaldriisenhormon. Nuturulissenschaften, 4 3 , 4 4 4 5 . Katz, E. and Weissbach, H. (1962). Biosynthesis of the actinomycin chromophore; enzymatic conversion of 4-methyl-3-hydroxyanthranilic acid to actinocin. J. biol. Chem. 237, 882-886. Kaufman, S. (1962). Studies on tryptophan pyrrolase in Drosophila melunogaster. Genetics, Princeton, 47, 807-817. Kawase, S. (1955). Studies on the pigments found in the integument of Bombyx mori L. I. Red pigment and some other epidermal pigments. /. sericult. S c i /upan, 24, 69-76. Kellner, O., Sako, T. and Sawano, J. (1884). Chemische Untersuchungen uber die Entwicklung und ErnZhrung des Seidenspinners (Bombyx mori). Lundw. Vers.-Stat. 30. 59-86. Kikkawa, H. (1941). Mechanism of pigment formation in Bombyx and Drosophila. Genetics, Princeton, 28, 587-607. Kikkawa, H. (1951). On fluorescent substances involved in silkworm eggs. J. sericult. S c i Japan, 20,66. (In Japanese, cited after Kikkawa, 1953.) Kikkawa, H. (1953). Biochemical genetics of Bombyx mori (silkworm). Adu. Genet. 5, 107-140. Kikkawa, H. and Kuwana, H. (1952). Origin of nicotinic acid in Eombyx mori (silkworms). Annotncs zool. /up. 25, 30-33. Koga, N. and Coda, Y. (1962). 3-Hydroxy-kynwenin in frisch gelegten Bombyx-Eiern. Hoppe-Scyler’s Z . physiol. Chem. 328 272-274. Koga, N. and Osanai, M. (1967). Der Gehalt an Tryptophan, Kynurcnin, 3-Hydroxykynurenin und Ommochromen bci den iiberwinternden Eiern des Seidenspinners Eombyx mori wiihrend der Entwicklung. Hoppe-Scyler’s 2. physwl. Chem. 348, 979-982.

236

BERNT LINZEN

Kolb, G. and Autrum, H. (1972).Die Feinstruktur im Auge der Biene bei Hell- und Dunkeladaptation. J. comp. Physiol. 77, 113-125. Kotake, Y. and Shichiri, G. (1931). h e r die Spaltung des Kynurenins mittels Natriumbicarbonatlosung.Hoppe-Seyler's Z . physiol. Chem. 195,152-158. Krieger, F. (1954).Untersuchungen iiber den Farbwechsel der Libellenlarven Z. vergl. Physiol. 36, 352-366. Kiibler, H. ( 1 960). Untersuchungen zur Biogenese und Verbreitung der Ommochrome. Doctoral Thesis, University of Munich. Kiihn, A. (1932). Entwicklungsphysiologische Wirkung einiger Gene bei Ephestia kiihniella. Naturwissenschaften, 20, 974-977. Kiihn, A. (1940). Eine Mutation der Augen- und Korperpigmentierung (dec) bei Ptychopoda seriata Schrk. Z. VerebLehre, 78, 1-13. Kiihn, A. (1941). Uber eine Gen-Wirkkette der Pigmentbildung bei Insekten. Nachr. Akad. Wiss. Cattingen, Math.Phys. Kl. 1941, 231-261. Kiihn, A. (1955).Ein Fluorometer fiir Papierchromatogramme. Natunuissenschaften 42,

529-531. Kiihn, A. (1956). Versuche zur Entwicklung eines Modells der Genwirkung. Naturwissenschaften, 43,25-28. Kiihn, A. (1960). h e r den Ausfirbungsverlauf der Imaginalaugen von EphestiaGenotypen in der Puppe. Biol. Zbl. 79, 385-392. Kiihn, A. (1963). Die zeitliche Folge der Bildung der Ommochrome in den Imaginalaugen von Ephestia und Ptychopada. Z . Naturf. 18b,252-254. Kiihn, A. and Almeida, F. F. de (1961).Fluoreszierende Stoffe und Ommochrome bei Genotypen von Plodia interpunctella. Z. VererbLehre 92,126-132'. Kiihn, A. and Egelhaaf, A. (1955).Zur chemischen Polyphanie bei Ephestia kiihniella. Naturwissenschaften, 14,634-635. Kiihn, A. and Egelhaaf, A. (1959).h e r die Mutation br (braunaugig) bei Ephestia kuhniella. Z. VererbLehre, 90,244-250. Kiihn, A. and Henke, K. (1930). Eine Mutation der Augenfarbe und Entwicklungsgeschwindigkeit bei der Mehlmotte Ephestia kiihniella. Wilhelm Roux Arch. EntwMech. Org. 122,204-212. Kiihn, A. and Plagge, E. (1937). Radetermination der Raupenpigmentierung bei Ephestia kiihniella durch den Genotypus der Mutter und durch arteigene und artfremde Implantate. Biol. Zbl. 57, 113-126. Lan, S. J. and Gholson, R. K. (1965).A comparative study of tryptophan catabolism. J. biol. Chem. 240,3934-3937. Langer, H . (1967). Uber die Pigmentgranula im Facettenaugc von Calliphora erythrocephala. Z . vergl. Physiol. 55, 354377. Langer, H. and Crassmader, H. (1965). Freies Tryptophan in Wildform und Augenfarbmutanten von Calliphora erythrocephula wiihrend der Imaginalentwicklung. Z. Naturf. 20b, 172-177. Langer, H. and Hoffmann, Ch. (1966). Elektro- und stoffwechselphysiologische Untersuchungen iiber den Einfluss von Ommochromen und Pteridinen auf die Funktion des Facettenauges von Calliphora erythrocephah J. Insect Physiol. 12,

357-387. Langer, H. and Struwe, C. (1972).Spectral absorption by screening pigment granules in the compound eye of butterflies (Heliconius). J. comp. Physiol. 79 203-212.

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

237

Laudani, U. and Grigolo, A. (1968). Sulla presenza di alcuni metaboliti del triptofano, pterine ed altre sostanze U.V. fluorescenti in individui di Musca domestica L., normali e mutanti peril colore degli occhi. A t t i A s s . genet. Ital. 13, 180-182. ’ Leibenguth, F. (1965). Kynurensaure und Xanthurensaure bei Habrobracon juglandis Ashmead. Z . Naturf. 20b, 315-317. Leibenguth, F. (1967a). Entwicklungsphysiologisch-genetische Untersuchungen am Tryptophanstoffwechsel zweier Genotypen von Habrobracon juglandis Ashmead. Wilhelm Roux A r c h EntwMech. Org. 158, 246-300. Leibenguth, F. (1967b). Regulation of tryptophan metabolism in the parasitic wasp, Habro bracon juglandis. Experien tia, 23, 1069-107 1. Leibenguth, F. (1970). Concerning non-darkening of mutant Habrobracon (Bracon hebetor) eyes as consequence of new chromogen-reducing mechanism in insect larvae. Experientia, 26, 659. Leklem, J. E. (1971). Quantitative aspects of tryptophan metabolism in humans and other species: a review. A m . J . clin. N u b . 24, 659-672. Lindsley, D. L. and Grell, E. H. (1968). Genetic variations of Drosophila melanogaster. Carnegie Institute, Washington, Public. No. 627. Linzen, B. (1957). h e r Ommine und iiber das Auftreten von Ommochromen im Tierreich. Doctoral Thesis, University of Tubingen. Linzen, B. (1959). h e r Ommochrome, XV. a e r die Identifizierung des “Urechochroms” als Xanthommatin. Hoppe-Seyler’s Z, physiol. Chem. 314, 12-14. Linzen, B. (1963). Eine spezifische quantitative Bestimmung des 3-Hydroxy-kynurenins. 3-Hydroxy-kynurenin und Xanthommatin in der Imaginalentwicklung von Call:phora. Hoppe-Seyler’s 2. physiol. Chem. 333, 145-148. e r Ommochrome, XXIII. Ommidine-ein neuer Typus von Linzen, B. (1966). a Ommochromen aus Orthopteren. Z. Naturf. 21b, 1038-1047. Linzen, B. (1967). Zur Biochemie der Ommochrome. Unterteilung, Vorkommen, Biosynthese und physiologische Zusammenhkge. Natunuissenschaften, 54. 259-267. Linzen, B. (1968). Versuche zur Biosynthese der Ommochrome. ZooL Anz. Suppl. 31, 148-153. Linzen, B. (1970). Zur Biosynthese von Ommochromen, I. Einbau ”S-markierter Vorstufen in Ommine. Hoppe-Seyler’s Z. physiol. Chem. 351, 622-628. Linzen, B. (1971a). Tryptophan catabolism in silkworm metamorphosis: changes in total matter, DNA, free, and protein-bound tryptophan. J. Insect Physiol. 17. 2005-2016. Linzen, B. (1971b). Autonomes Verhalten Tryptophan-abbauender Enzyme wahrend der Metamorphose von Bombyx mori rb: Lokalisation und zeitlicher Verlauf der Tryptophanpyrrolase. Wilhelm Roux Arch. EntwMech. Osg. 168, 320-331. Lmzen, B. and Buckmann, D. (1961). Biochemische und histologische Untersuchungen zur Umfirbung der Raupe von Cerura Dinula L. Z. Nuturf. 16b, 6-18. Linzen, B. and Ishiguro, I. (1966). 3-Hydroxy-kynurenin bei Bombyx mori rb.-Ein neuer Tryptophanmetabolit: 3-Hydroxy-kynurenin-glucosid. Z. Naturf. 21b, 132-137. Linzen, B. and Hendrichs-Hertel, U. (1970). Kynurenin-3-Hydroxylase in der Metamorphose von Bombyx mori rb: Organ- und stadienspezifische Wechsel der Enzymaktivitat. Wilhelm Roux Arch. EntwMech. Org. 165, 26-34.

238

BERNT LINZEN

Linzen, B. and Hertel, U. (1967). In uitro assay of insect kynurenine-3-hydroxylase. Natunuissenschaften, 54,21. Lucas, F., Shaw, J. T. B. and Smith, S. G . (1958). The silk fibroins. Adv. Protein Chem. 13. 108-242. Magnus, D. (1958). Experimentelle Untersuchungen zur Bionomie und Ethologie des Kaisermantels Argynnis paphin L. (Lep. Nymph.). I. Uber optische Ausloser von Anfliegereaktionen und ihre Bedeutung fiir das Sichfiiden der Geschlechter. Z . Tierpsychol. 15, 397-426. Maier, W. (1965). Elektronenmikroskopischer Vergleich von Retinula-Pigmentzellen’der Wildform wa+ und der Mutante wa von Ephestia kiihniella Z. Z . Naturf. 20b, 31 2-3 14. Makino, K., Takahashi, H., Satoh, K. and Inagami, K. (1954). Abnormal accumulation of 3-hydroxy-kynurenine in the mutant “aka-aka” of silkworm. Nature, Lond. 179, 586-587. Marzluf, G . A. (1965a). Tryptophan pyrrolase of Drosophila: partial purification and properties. Z VererbLehre, 97, 10-17. Marzluf, G. A. (1965b). Enzymatic studies with the suppressor of vermilion of Drosophiln melanogaster. Genetics, Princeton, 52. 503-512. Mayer, G. and Staudinger, H. (1967). Zur Frage der Aktivierung von L-Kynurenin-3Hydroxylase in Rattenlebermitochondrien durch Cyanid. Hoppe-Seyler ‘s Z . physiol. Chem. 348, 733-734. Mayer, G., Ullrich, V. and Staudinger, H. (1968). Possible involvement of cytochrorne b5 in L-kynurenine-3-hydroxylaseof rat liver mitochondria. Hoppe-Seyler’s Z . physiol. Chem. 349, 459-464. McArthur, J. N. and Dawkins, P. D. (1969). The effect of sodium salicylate on the binding of L-tryptophan to serum proteins. J. Pharm. Pharmac. 21, 744750. McMenamy, R. H. and Oncley, J. L. (1958). The specific binding of L-tryptophaq to serum albumin. J. biol. Chem. 233, 1436-1447. Mehler, A. H. and Knox, W. E. (1950). The conversion of tryptophan to kynurenine in liver. 11. The. enzymatic hydrolysis of formylkynurenine. 1. biol. Chem. 187, 431-439. Merlini, L. and Nasini, G. (1968). Pigmenti degli insetti V.-Xantommatina nel tegument0 di Cossus cossus L. (Lepidoptera Cossidae). Ist. Lombardo, Rendic. Cl. Sci. (A), 102,890-892. Mittler, S. (1952). Influence of amino acids upon incidence of tumors in TuS0jstock of D. melanogaster. Science, 116, 657-659. Mohlmann, E. (1958). Chromatographische Untersuchungen an verschiedenen Cenotypen von Plodin interpunctella im Vergleich mit anderen Schmetterlingsarten. Z . VererbLehre, 89, 651-674. Montenay-Garestier, Th. and Htlkne, C1. (1971). Reflectance and luminiscence studies of molecular complex formation between tryptophan and nucleic acid components in frozen aqueous solutions. Biochemistry (Washington), 10, 300-306. Morgan, L. R. and Weimorts, D. M. (1964). The conversion of 3-hydroxyanthranilic acid to 2-aminophenoxazin-3-one-l,8-dicarboxylic acid by extracts of acetone powders of poikilothermic animal livers. Biochim. biophys. Acta, 82, 645-647. Mnckenthaler, F. A. (1971). Kynurenine localization in the egg of Drosophila rnelanogaster Experientia, 27,828-830.

THE TRYPTOPHAN

--*

OMMOCHROME PATHWAY itv INSECTS

239

Musajo, L., Spada, A. and Casini, E. (1950). Sulla sintesi della d,l-3-ossichinurenina. G m z . chim. Ital. 80, 171-176. Muth, F. W. (1967). Quantitative Untersuchungen zum Pigmentierungsgeschehen in den Augen der Mehlmotte Ephestia kiihniella. I. Grundlagen. Wilhelm Roux Arch. EntwMech. Org. 159, 379-411. Muth, F. W. (1968). Quantitative Untersuchungen zum Pigmentierungsgeschehen in den Augen der Mehlmotte Ephestiu kiihniella. 11. Die Pigmentbildung im a-Stamm nach Injektion kleiner Mengen von 3-Hydroxykynurenin. Wilhelm Roux Arch. EntwMech Org. 161, 336-350. Muth, F. W. (1969). Quantitative Untersuchungen zum Pigmentierungsgeschehen in den Augen der Mehlmotte Ephestiu kiihniella 111. Die Pigmentbildung im a-Stamm nach Injektion hoher Dosen von Kynurenin und der Versuch einer Deutung. Wilhelm Roux Arch. EntwMech. Org. 162, 56-77. Nair, P. M. and Vaidyanathan, C. S. (1964). Isophenoxazine synthase. Biochim. biophys. Acta, 81, 507-516. Nair, P. M. and Vining, L. C. (1964). An isophenoxazine synthase from Pycnoporus coccineus (Fr.) Bond. and Sing. Can. J. Biochem. 42, 1515-1526. Nakagaki, M., Koga, N. and Iwata, S. (1962). Bio-physicochemical studies on phenoxazone compounds. I. Diffusion constant and degree of association. Yakugaku Zasshi, 82, 1134-1138. Nawa, S. (1970). Purification and properties of tryptophan pyrrolase in Ephestia. A . Rep. Nut. Inst. Genetics (Japan), 21, 24-25. Needham, A. E. (1970). The integumental pigments of some isopod crustacea. Comp. Biochem. Physiol. 35, 509-534. Neese, V. (1968). Zur optischen Orientierung der Augenmutante “Chartreuse” von Apis mellifica L. Z . uergl. Physiol. 60, 41-62. Neese, V. (1972). Die Altersabhangigkeit des Ommochromgehalts im Komplexauge der Bienen. Ein quantitativer Vergleich zwischen Wildtyp und der Augenmutante chartreuse. J. Insect Physiol. 18, 229-236. Neubert, G. (1956). Modellversuche zur Bildung der Ommochrome. Doctoral Thesis, University of Tiibingen. Nickerson, B. (1956). Pigmentation of hoppers of the Desert Locust (Schistocerca gregaria Forskal) in relation to phase coloration. Anti-Locust Bull No. 24. Nolte, D. J. (1950). The eye-pigmentary system of Drosophila: the pigment cells. J. Genet. 50, 79-99. Nolte, D. J. (1952). The eye-pigmentary system of Drosophila 111. The action of eye-colour genes. J. Genet. 51, 142-186. Nolte, D. J. (1961). The pigment granules in the compound eye of Drosophila Heredity, Lond. 16, 25-38. Nolte, D. J. (1965). The pigmentation of Locusts. S.Afr. J. S i i 61, 173-178. Okamoto, H. and Hayaishi, 0. (1967). Flavin adenine dinucleotide requirement for kynurenine hydroxylase of rat liver mitochondria. Biocheri. biophys. Res. Commun. 29, 394399. Okamoto, H., Yamamoto, S., Nozaki, M. and Hayaishi, 0. ”(1967). On the submitochondrial localization of L-kynurenine-3-hydrox) lase. Biochem. Biophys. Res. Commun. 26, 309-314. Osanai, M. (1966a). a e r die Umfirbung der Raupen von Hestina japonica zu Beginn

240

BERNT LINZEN

der Uberwinterung, 11. Pigmente der Raupen von Hestina japonica und Sasakia charonda. Hoppe-Seyler’s Z. physiol. Chem. 347, 145-155. Osanai, M. (1966b). h e r die Umfiirbung der Raupen von Hestina japonica zu Beginn der Uberwinterung, 111. Ein Ommochromoprotein mit Redoxverhalten in der Haut der Hestina-Raupe. J. Insect Physiol. 12, 1295-1301. Osanai, M. (1967). Mange1 an Cytochrom c im iiberwinternden Ei des Seidenspinners, Bom byx m o r i Hoppe-Seyler’s Z . physiol. Chem. 348, 469-470. Osanai, M. (1968a). Tyrosinaseaktivitat bei der Hestinn-Raupe zu Beginn der h e r winterung. J. Insect Physiol. 14. 261-267. Osanai, M. (1968b). Tyrosinaseaktivitat in den Eiern von Ommochrom-Mangelmutanten des Seidenspinners, Bombyx mori J. Insect Physiol. 14, 403-408. Osanai, M. and Arai, Y. (1962a). Effect of ligation at different levels on the change of body colour in. the larvae of the nymphalid butterfly, Hestinn japonica. at the beginning of hibernation. Zool. Mag. (Dobotsugaku Zasshi), 71, 202-205. (In Japanese with English summary and colour plates.) Osanai, M. and Arai, Y.(1962b). Uber die Umfiirbung der Raupen von Hestinajaponica zu Beginn der Uberwinterung I. Durchschniirungsversuche an der Umfiirbung der Hestina-Raupe. Cen. comp. Endocr. 2, 31 1-316. Osanai, M. and Koga, N. (1966). Isolierung des Xanthommatins und der Ommine aus Eiem des Seidenspinners mit SE-Sephadex. Eine neue Methode zur Abtrennung und quantitativen Bestimmung der Ommochrome. Hoppe-Seyler’s Z. physiol. Chem. 347, 156-160. Parsons, P. A. and Green, M. M. (1959). Pleiotropy and competition at the vermilion locus in Drosophila melanogaster. Proc. nntn. Acad. Sci U.S.A. 45,993-996. Perrelet, A., Orci, L. and Baumann, F. (1971). Evidence for granulolysis in the retinula cells of a stomatopod crustacean, Squilla mantis. J. Cell Biol. 48, 684-688. Phillips, J. P. and Forrest, H. S. (1970). Terminal synthesis of xanthommatin in Drosophila melanogaster. 11. Enzymatic formation of the phenoxazinone nucleus. Biochem. Genetics 4, 489-498. Phillips, J. P., Simmons, J. R. and Bowman, J. T. (1970). Terminal synthesis of xanthommatin in Drosophila melanogaster. I. Roles of phenol oxidase and substrate availability. Biochem. Genetics, 4, 481487. Pieber, M., Romero, C., Tirapegui, C. and Tohi, J. (1969). Absence of hyperchromic effect in DNA heated above Tm. in the presence of tryptophan. Z. Naturf. 24b, 508-5 10. Pinamonti, S . and Petris, A. (1966). Studies on tryptophan pyrrolase of Schistocerca gregaria Forsk. (Orthoptera). Comp. Biochem. Physiol. 17, 1079-1087. Pinamonti, S., D’Orazio, G . and Veronese, F. M. (1964). Ricerche s u l metabolismo del triptofano via chinurenina in Schistocerca gregaria Forsk. (Ortotteri). Boll. Zool. 31, 567-57 1. Pinamonti, S . , Petris, A. and Miliani, M. (1970). Studies on the kynurenine transaminare of Schistocerca gregaria Forsk. (Orthoptera, Acrididae). Comp. Biochem. Physiol. 37, 311-320. Pinamonti, S., Petris, A. and Ottaviani, F. (1970/71). Studio sulla chinurenina-3-idrossilasiin Schistocerca gregurie Forsk. (Orthoptera, Arcrididae). A t t i A c c a d . S c i Ferrara, 48, 1-12. Plagge, E. (1939). Genabhangige Wirkstoffe bei Tieren. Ergebn. Biol. 17. 105-150.

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

241

Plaine, H. L. and Glass, B. (1955). Influence of tryptophan and related compounds upon the action of a specific gene and the induction of melanotic tumors in Drosophih melanogaster. J. Genet. 53, 244261. Prasad, C. and French, W. L. (1971). Tryptophan pyrrolase activity during development of the mosquito (Culex pipiens). Comp. Biochem. Physiol. 38b 627-629. Ramsay, J. A. (1956). Excretion by the Malpighian tubules of the stick insect, Dixippus morosus (Orthoptera, Phasmidae) : calcium, magnesium, chloride, phosphate and hydrogen ions. J. exp. Biol. 33, 697-708. Raszka, M. and Mandel, M. (1971). Interaction of aromatic amino acids with neutral polyadenylic acid. Proc. natn. A c a d S c i U.S.A. 68, 1190-1191. Rechsteiner, M. C. (1970). Drosophila lactate dehydrogenase and a-glycerolphosphate dehydrogenase: distribution and change in activity during development. J. Insect Physiol. 16, 1179-1192. Reisener-Glasewald, E. (1956). Uber die Entwicklung des Bestandes an fluoreszierenden Stoffen in den Kopfen von Ephestia kiihniella in Abhangigkeit von verschiedenen Augenfarbgenen. Z. VererbLehre, 87,668-693. Rizki, M. T. M. (1957). Tumor formation in relation to metamorphosis in Drosophila melanogaster. J. Morph. 100, 459-472. Rizki, M. T. M. (1961). Intracellular localization of kynurenine in the fat body of Drosophila J. biophys. biochem. Cytol. 9, 567-572. Rizki, T. M. (1963). Genetic control of cytodifferentiation. .I. CellBiol. 16, 513-520. Rizki, T. M. (1964). Mutant genes regulating the inducibility of kynurenine synthesis. J. CellBiol. 21, 203-211. Rizki, T. M. and Rizki, R. hi. (1963). An inducible enzyme system in the larval cells of Drosophila melanogaster. J. Cell Biol. 17, 87-92. Rizki, T. M. and Rizki, R. M. (1964). Factors affecting the intracellular synthesis of kynurenine. J. Cell Biol. 21, 27-33. Rohner, Sr. M. C. (1959). The influence of depression of respiratory metabolism by carbon monoxide o n the pigment development in the insect eye. Z. VererbLehre, 90, 257-262. Roth, H. (1939). Die Bestimmung des freien und gebundenen Tryptophans in Pflanzen. Angew. Chem. 52.149-151. Rowell, C. H. F. (1971). The variable coloration of the acridoid grasshoppers. In “Advances in Insect Physiology” (Eds J. W. L. Beament, J. E. Treherne and V. B. Wigglesworth), Vol. 8, pp. 145-198. Academic Press, London and New York. Riidiger, W. and Klose, W. (1970). h e r die Pigmente der E’lorfliege Chrysopa carnea Experientia, 26,498. Ruiter, L. de (1955). Countershading in caterpillars. Archs n k i d Zool. 11, 1-57. Schiifer, W. (1964). The chemistry of the phenoxazones. Prog. Org. Chem. 6, 135-163. Schafer, W. and Geyer, I. (1972). fher das Redoxverhalten der Ommochrome. UV/S-Spektren von 3H-Phenoxazinonen-(3) und Phenoxazinen. Tetrahedron, 28, 526 1-5279. Schartau, W . and Linzen, B. (1969). Tryptophan wird von Insekten nicht veratmet. Naturw issenschaften, 56, 329. Schildknecht, H. and Schmidt, H. (1963). Die chemische Zusammensetzung des Wehrsekretes von Dicranura uinula. XVII. Mitteilung iiber Insektenabwehrstoffe. Z. Naturf. 18b, 585-587.

242

BERNT LINZEN

Schildknecht, H., Birringer, H. and Krauss. D. (1969). Uber Arthropoden-Abwehrstoffe XXXVI. Aufklirung des gelben Prothorakalwehrdriisen-Farbstoffes von Ilybius fenestratus. Z . Naturf. 24b, 38-47. Schlossberger, H. G. (1949). Synthese und physiologische Bedeutung des 3-Oxykynurenin, eines Zwischenproduktes bei der Bildung der Ommochrome aus Tryptophan im Insektenorganismus. Ph.D. Thesis, University of Tubingen. Scholtan, W. (1956). Kolloidchemische Eigenschaften von Phenothiazinderivaten, insbesondere von Megaphen. Medizin und Chemie (reports of Farbenfabriken Bayer A.G.) 5. 295-313. Verlag Chemie, Weinheim. Schultz, J. and Rudkin, G. T. (1948). Absence of a sparing action of tryptophane on nicotinamide requirements of the fly, Drosophila melanogaster. Fedn Proc. Fedn A m . Socs exp. Biol. 7, 185. Schwabl, G. and Linzen, B. (1972). ijber die Bildung der Augenpigmentgranula bei Drosophila v und cn nach Verfiitterung von Ommochromvorstufen. Wilhelm Roux Arch. EntwMech. Org. 171, 223-227. Schwinck, I. (1953). iiber den Nachweis eines Redox-Pigmentes (Ommochrom) in der Haut von Sepia officinalis. Naturwissenschaften, 40, 365. Seecof, R. L., Kaplan, W. D. and Futch, D. G. (1969). Dosage compensation for enzyme activities in Drosophila melanogaster. Proc. natn. Acad. Sci. U.S.A. 62. 528-535. Semm, K. and Fried, R. (1952). Einfache Apparatur zur Fluoreszenzmessung auf Filtrierpapierstreifen. Naturwissenschaften, 39, 326-327. Shapard. P. B. (1960). A physiological study of the vermilion eye color mutants of Drosophila melanogaster. Genetics, Princeton, 45, 359-3 7 6. Shinohara, R. and Ishiguro, I. (1970). The purification and properties of formamidase from rat liver. Biochim. biophys. Acta, 198. 324-331. Shoup, J. R. (1966). The development of pigment granules in the eyes of wild type and mutant Drosophila melanogaster. f. Cell Biol. 29. 223-249. Simmons, J. R. and Gardner, E. J. (1958). The effect of tryptophan on the effect of tumorous head in crosses among selected stocks of Drosophila melanogaster. Genetics, Princeton, 43, 164-171. Singh, H.. Seshadri, T. R. and Subramanian, G. B. V. (1966). A note on the colouring matter of lac larvae. Tetrahedron lett. 1966, 1101-1108. Sols, A. and Marco, R. (1970). Concentrations of metabolites and binding sites. Implications in metabolic regulation. In “Current topics in cellular regulation” (Eds B. L. Horecker and E. R. Stadtman), Vol. 2, pp. 227-273. Academic Press, New York and London. Sonobe, H. and Ohnishi, E. (1970). Accumulation of 3-hydroxy-kynurenine in ovarian follicles in relation to diapause in the silkworm, Bombyx mori L. Devl Growth Different. 12,41-52. Spies, J. R. and Chambers, D. C. (1948). Tryptophan determination. Color-forming reactions with p-dimethyl-aminobenzaldehyde and sodium nitrite in sulfuric acid. Analyt. Chem. 20, 30-39. Stamm, M. D. and Aguirre, L. (1955a). Estudios sobre bioquimica de insectos I. Aminoacidos aromiticos y triptbfano en la metamorfosis de Saturnia P y r i Rev. Espan. Fisiol. 11, 63-68. Stamm, M. D. and Aguirre, L. (1955b). Estudios sobre bioquimica de insectos 11.

THE TRYPTOPHAN

+

OMMOCHROME PATHWAY IN INSECTS

243

Aminoicidos aromiticos y tript6fano en la metamorfosis de Pieris Brassicae y Ocnogyna Baetica Rev. Espan. Fisiol. 11, 69-74. Stamm, M. D. and Aguirre, L. ( 1 9 5 5 ~ ) .Estudios sobrc bioquimica de insectos 111. Aminoicidos aromiticos y tript6fano en dos cole6pteros. un hemiptero y un ortbptero. Rev. Espan. Fisiol. 11, 75-81. Stamm, M. D. and Galarza, A. M. (1961). Bioquimica de 10s omocromos. An. Real. Acad. Farm. 1961, 115-140. Streck, P. (1972). Der Einfluss des Schirmpigmentes auf dz s Sehfeld einzelner Sehzellen der Fliege Calliphora erythrocephala Meig. Z. uergl. Physiol. 76, 372-402. Strother, G. K. (1966). Absorption of Musca domestictr screening pigment. J. gen. Physiol. 49, 1087-1088. Subba Rao, P. V., Jegannathan, N. S. and Vaidyanathan, C. S. (1965). Enzymic conversion of 3-hydroxyanthranilic acid into cinnabarinic acid by the nuclear fraction of rat liver. Biochem. J. 95, 628-632. Suiec-Michieli, S. (1965). Biochemical investigations of the morphological colour changes in Mantis religiosa (Dictyoptera). PTOC. XII. Int. Congr. Ent. London, 1964, Sect. 3, pp. 208-209. Tartof, K. D. (1969). Interacting gene systems: I. Thc regulation of tryptophan pyrrolase by the vermilion-suppressor of vermilion system in Drosophila Genetics, Princeton, 62, 781-795. Tatum, E. L. (1939). Development of eye-colors in Drosophila: Bacterial synthesis of u+ hormone. Proc. nutn. Acad. Scr U.S.A. 25, 486-490. Tatum, E. L. and Beadle, G. W. (1939). Effect of diet O’L eye-color development in Drosophila melanogaster. Biol. Bull. mm. biol Lab.. Woods Hole, 77, 415-422. Tatum, E. L. and Haagen-Smit, A. J. (1941). Identification of Drosophila u’ hormone of bacterial origin. J. biol. Chem. 140, 575-580. Tiedt, S. (1971). Untersuchungen zum Tryptophanabbaii in der Entwicklung von Cryllus bimaculatus de Geer. Dipl. Thesis, University of Munich. Tokuyama, T., Senoh, S., Sakan, T., Brown, K. S. and Witkop, B. (1967). The photoreduction of kynurenic acid t o kynurenine yellow and the occurrence of 3-hydroxy-L-kynurenine in butterflies. J. Amer. chem. S O C . 89, 1017-1021. Traub, P. (1962). Die Struktur und Synthese des Rhodommatins und Ommatins D. Ein Beitrag zur Chemie der Ommochrome. Doctoral Thesis, University of Munich. Twardzik, D. R., Grell, E. H. and Jacobson, K. B. (1971). Mechanism of suppression in Drosophila: a change in tyrosine transfer RNA. J. molec Lliol. 57, 231-245. Udenfriend, S. (1969). “Fluorescence Assay in Biology and Medicine”, Vol. 11, pp. 235-237. Academic Press, London and New York. Umebachi. Y. (1959). Yellow pigments in the wings of the papilionid butterflies. 111. The radioautographs of the wings of five species of Papiliu injected with CI4-labelled tryptophan. Annotnes zool. jap. 32, 112-116. Umebachi, Y. and Katayama, M. (1966). Tryptophan and tyrosine metabolism in the pupa of papilionid butterflies. 11. The general pattern of tryptophan metabolism during the pupal stage of Papilio xuthus. J. Insect Physwl. 12, 1539-1547. Umebachi, Y. and Nakamura, A. (1954). The presence of kynurenine in the wings of the papilionid butterflies. Zool. Mag. Tokyo, 63, 57-61. (In Japanese with English summary.)

244

BERNT LINZEN

Umebachi, Y. and Tsuchitani, K. (1955). The presence of xanthurenic acid in the fruit-fly, Drosophila melanogaster. J. Biochem., Tokyo, 42, 81 7-824. Umebachi, Y. and Uchida, T. (1970). Ommochromes of the testis and eye of Papilio xuthus. /. Insect Physiol. 16, 1797-1812. Umebachi, Y. and Yoshida, K. (1970). Some chemical and physical properties of papiliochrome I1 in the wings of Papdio xuthus. J. Insect Physiol. 16, 1203-1 228. Ursprung, H., Graf, C. E. and Anders, G. (1958). Experimentell ausgeloste Bildungvon rotem Pigment in den Malpighischen Gefassen von Drosophila melanogaster. Rev. Suisse Zool. 65, 449-460. Viscontini, M. and Schmid, H. (1963). Fluoreszierende Stoffe aus Rhodnius prolixus Stal (Hemiptera, Heteroptera). Ein neues Kynureninderivat: 3-Hydroxy-kynureninschwefelsaure-ester = Rhodnitin. Helv. chim. Acta, 46, 2509-2516. Viscontini, M. and Stierlin, E. (1961). Isolierung, Struktur und Synthese von Pterinen aus Ephestia kiihniella ZELLER. Helv. chim. Acta, 44, 1783-1785. Viscontini, M. and Stierlin, E. (1963). Synthese von Erythropterin, Ekapterin und Lepidopterin. Helv. chim. Acta, 46, 51-56. Vuilleaume, M. (1968). Pigmentations et variations pigmentaires de trois insectes: Mantis religiosa (L.), Sphodromantis viridis (F.) et Locusta mipatoria (L.). Bull. biol. Fr. Belg. 102, 147-232. Wachmann, E. (1969). Multivesikulare und andere Einschlusskorper in den Retinulazellen der Sumpfgrille Pteronemobius heydeni (Fischer). Z. Zellforsch. 99, 263-276. Washizu, Y., Burkhardt, D. and Streck, P. (1964). Visual field of single retinula cells and interommatidial inclination in the compound eye of the blowfly Calliphora erythrocephala. Z. vergl. Physiol. 48, 413-418. Watanabe, M., Watanabe, Y. and Okada, M. (1970a). A new fluorimetric method for the determination of 3-hydroxykynurenine. Clin. chim. Acta, 27, 461466. Watanabe, M., Tamura, Z. and Okada, M. (1970b). Structure of a product of a fluorescence reaction of 3-hydroxykynurenine. Chem. Phann. Bull. 18, 285-290. Weber, F. and Wiss, 0. (1966). Am Tryptophan-Stoffwechsel beteiligte Enzyme. In “Handbuch der physiologisch- und pathologisch-chemischen Analyse” (Eds K. Lang and E. Lehnartz), 10th ed., Vol. VI/B, pp. 806-860. Springer-Verlag, Berlin, Heidelberg and New York. Wehner, R., Cartenmann, C. and Jungi, Th. (1969). Contrast perception in eye colour mutants of Drosophila melanogaster and Drosophila subobscura. J. Insect Physiol. 15, 815-823. Weissbach, H. and Katz, E. (1961). Studies on the biosynthesis of actinomycin: enzymic synthesis of the phenoxazone chromophore. /. bioL Chem. 236, PC 17-18. Wenzl, H. (1969). Saulenchromatographische Abtrennung von Xanthommatin aus biologischen Extrakten. Mikromethode. Hoppe-Seyler’s Z. physioL Chem. 350, 446-448. Wessing, A. (1962). Elektronenmikroskopische Studien zur Funktion der Malpighischen Gefasse von Drosophila melanogaster. Protoplasma. 55, 264-293. Wessing, A. (1963). Elektronenmikroskopische Studien zur Funktion der Malpighischen Gefasse von Drosophila melanogaster. 11. Die Gefasse der Puppe. Protoplasma, 56, 433-465.

THE TRYPTOPHAN + OMMOCHROME PATHWAY IF1 INSECTS

245

Wessing, A. and Bonse, A. (1962). Untersuchungen iiber die Speicherung und Ausscheidung von freiem Tryptophan durch die Malpighischen Gefasse von Drosophila melanogaster. Z. Naturf. 17b, 620-622. Wessing, A. and Bonse, A. (1966). Natur und Bildung lies roten Farbstoffes in den Nierentubuli der Mutante “red” von Drosophila mtdanogaster. Z. Naturf. 21b, 1219-1223. Wessing, A. and Danneel, R. (1961). Die Speicherung von Oxykynurenin in den Malpighischen Gefiissen verschiedener Augenfarbenmutanten von Drosophila melanogaster. Z . Naturf. 16b, 388-390. Wessing, A. and Eichelberg, D. (1968). Die fluoresziereiiden Stoffe aus den Malpighischen Gefassen der Wildform und verschiedener Augenfarbenmutanten von Drosophila melanogaster. Z . Naturf. 23b. 376-386. Wessing, A. and Eichelberg, D. ( 1 9 72). Elektronenmikrosltopische Untersuchungen an den Nierentubuli (Malpighische Gefasse) von Drosophila melanogaster. 111. Die intrazellulire Speicherung der Aminosaure, 3-Hydroxykynurenin. Z. Zellforsch. 125, 132-142. Willig, A. (1969). Die Carotinoide und der Gallenfarbstoff der Stabheuschrecke, Carausius morosus, und ihre Beteiligung an der Entstehung der Farbmodifikationen. J. Insect PhysioL 15, 1907-1927. Wilson, L. P. (1945). Tolerance of larvae of Drosophila for amino acids: tryptophane. Growth, 9, 145-154. Wolfram, R. (1949). a e r die Histogenese der Exkretpigmerte bei verschiedenen Rassen von Ptychopoda serMta Schrk. und Ephestia kiihniella Z. Z . VererbLehre, 83, 254-298. Wolsky, A. (1960). Oxidative metabolism and ommochrome synthesis in insects. XIth Int. Congr. Entomol. Vienna, 3, 189-193. Worz, 0. and Scheibe, G. (1969). b e r die Ursacheri der Alggregation von Farbstoffen, das Spektrum der photographischen Sensibilisierung und die biologische Aktivitat. Z . Naturf. 24b, 381-390. Wyatt, G. R. (1968). Biochemistry of insect metamorphosis. In “Metamorphosis, a Problem in Developmental Biology” (Eds W. Etkin and L. I. Gilbert), pp. 143-184. North Holland, Amsterdam, and Appleton-Century-Crofts. New York. Yamafuji, K. (1937). Biologie der Seide. In “Tabulae Biologicae” (Eds W. Junk, C. Oppenheimer and W. Weisbach), Vol. 14, pp. 35-50. Uitgeverij Dr W. Junk, Den Haag. Yamashita, 0. and Hasegawa, K. (1966). Further studies on the mode of action of the diapause hormone in the silkworm, Bombyx mori L. J. InsectPhysiol. 12, 957-962. Yasuzumi, G. and Deguchi, N. (1958). Submicroscopic structure of the compound eye as revealed by electron microscopy. J . Ultrastruct. Res. 1, :?59-270. Yoshida, M., Ohtsuki, H. and Suguri, S. (1967). Ommochrome from anthomedusan ocelli and its photoreduction. Photochem. PhotobioL 6, 875-884. Zeutzschel, B. (1958). Entwicklung und Lage der Augenpigmente bei verschiedenen Drosophilamutanten. Z. VererbLehre. 89, 508-520. Ziegler, I. (1960). Zur Feinstruktur der Augengranula bei Dt-osophifa melanogaster. Z . VererbLehre, 91,206-209.

246

BERNT LINZEN

Ziegler, I. (1961). Genetic aspects of ommochrome and pterine pigments. Adv. Genet. 10,349-403. Ziegler, I. and Feron, M. (1965). Quantitative Bestimmung der hydrierten Pterine und des Xanthommatins in den Augen von Cerutitis cupituta Wild. (Dipt., Trypetidae). 2. Nuturf. 20b, 318-322. Ziegler, I. and Harmsen, R. (1969). The biology of pteridines in insects. Adv. Insect Ph_ysiol. 6 , 139-203. Ziegler, I. and Jaenicke, L. (1959). Zur Wirkungsweise des white-Allels bei Drosophilu rnelunogaster. Z. VererbLehre, 90, 53-61.

Biophysical Aspects of Sound Communication in Insects Axel Michelsen Biological Institute, Universiv of Odense, Denmark

and Harald Nocke Zoological Institute, University of Cologne, Germany 1 Introduction 2 Some properties of sound . 2.1 Vibrations, impedances, and sound radiation 2.2 Types of sound fields . 3 Sound production in insects . 3.1 General aspects . 3.2 Mechanisms of frequency multiplication 3.3 The sound radiatpr , 3.4 The driving vibration and radiated sound 3.5 Thesoundguide . 3.6 Efficiency of singing . 4 Propagation of insect sounds . . . 5 Insect ears as sound receivers . 5.1 The effect‘lve parameters of sound 5.2 The forces acting on ears , 5.3 Influence from the surroundings . 5.4 The tympana1 vibrations 5.5 The behaviour of the receptor organ 5.6 Some atypical insect ears 6 Conclusions References

. .

. . . ’

. . . . . . . . . . .

. . .

. . .

247 248 248 251 253 253 254 258 262 264 267 268 271 271 274 275 279 285 288 291 293

1 Introduction The last decade has seen an increasing interest in the sound communication of insects. Much information has been accumulated in “classical” fields such as the anatomy of sound-producing and sound-receiving structures, the analysis of sounds, acoustic behaviour, and hexing. In addition, new areas of investigation have been opened up: the neural mFchanisms responsible for sound production and auditory processing in the CNS, and the biophysics of sound production and hearing. In a few insects, most aspects AIP-9

247

248

AXEL MICHELSEN AND HARALD NOCKE

of acoustic communication have been studied, and we are now beginning to understand some of the principles of bioengineering in this field. A system analysis of the entire field of acoustic communication may not be far away in a few cases. Several reviews have been published on sound communication and limited areas such as hearing, sound production, or acoustic behaviour (e.g. Busnel, 1963; Bennet-Clark, 1971; Markl, 1972; Huber and Elsner, 1973; Michelsen, 1973a). In this paper we will concentrate on the biophysical aspects of sound communication. This relatively “new” area has not been reviewed before, and we felt that the time had come to discuss the main results. Acoustics is a difficult field for non-specialists. In 1940 Pumphrey wrote: “. . . in earlier work on hearing little attempt to control the intensity or harmonic content of sound stimuli could be made. It is consequently impossible in most cases for a reviewer to assign the proper weight to pieces of conflicting evidence which have been brought forward from time to time.” The situation has improved, but still too many investigators ignore the acoustical aspects of their experiments. There are many textbooks on acoustics available, but only a few of them deal with the problems in a practical manner (e.g. Morse, 1948; Beranek, 1954; Olson, 1957). The majority of physics texts choose to discuss two simple situations: either the obstacles are very small compared with the wavelength of sound, or they are very large. The newcomer soon discovers that his particular problem is somewhere between these extremes. Throughout this paper we have included some theory and practical hints, which should be helpful to those planning experiments. We should like to thank Mr 0. Juhl Pedersen, M.Sc. for comments’on the acoustical sections. 2 Some properties of sound The physics of sound waves is described in many textbooks. Here, we will concentrate on a few aspects which are of particular importance for the understanding of sound communication in insects. Sound waves are longitudinal, i.e. the air particles move in the direction of propagation of the wave. Sound waves may be described as alternate compressions and rarefactions, but also as movements of air particles. It is important t o realize that sound pressure is a scalar quantity, although sound waves have directional properties. In the past, this has caused some confusion in discussions on insect hearing (see below). 2.1

VIBRATIONS, IMPEDANCES, AND SOUND RADIATION

Impedances are well known to those familiar with electrical circuits. There is a direct analogy between electrical and mechanical systems, and it is

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION IN INSECTS

249

often useful to consider the electrical analogy of mechanical systems. In mechanics it is practical to define a mechanical impedance Z as the complex ratio of velocity u t o force I.‘ Z=F/v

or

v=F/Z

(1)

In the “impedance-type analogy” used hew, the m a s m is equivalent to an inductance, the compliance C of the springiness to a capacitance, and the (viscous) mechanical resistance R to an ohmic resistance. Further, the force is equivalent to a voltage and the velclcity to a current. With this analogy the behaviour of mechanical systems may be analysed in the same way as circuits of alternating current. Other types of analogy may be used for special problems (see Beranek, 1954). The impedance was defined as the complex ratio of Flu. Just as an inductance o r capacitance in electrical circuits, the introduction o f a mass or a compliance in mechanical systems causes phase shifts. In the simple driven oscillator (Fig. 1 ) one finds for steady-state vibration:

Z = R + j w m + l/@C)

(2)

where w is 2n times the frequencyj. The symbol j(Jr1) indicates the phase. R and F are in phase. wm and 1 IwC mxy be drawn as vectors which are totally out of phase and 90” out o f phase with R . The magnitude and phase of Z can then be found by adding the vectors. The influence o f m and C depends on f, so (as in electrical circuits) a change of frequency will cause a phase shift of u relative to F. At a certain frequency w m will be equal to L/wC, and these terms cancel out. Thus, we are left with Z .= R . This is the resonance frequency, where the velocity is a maximum and in phase with the force. Measurements of phase shifts are very useful in studies of insect cars (see below). It should be remembered that the displacement d and the acceleration a are in opposite phase and 90” out of phase with the velocity: u = d / ( j o ) =jwa

(3)

Fig. 1. Analogous mechanical and electrical systems. Both systems are resonant and driven by a harmonic force.

250

AXEL MICHELSEN AND HARALD NOCKE

Equation (2) is only valid for simple driven oscillators where all parts of the mass vibrate with a uniform amplitude of motion. This is not the case in membranes, so more sophisticated equations must be used (see Michelsen, 1971b). However, at low frequencies (the fundamental mode of vibration) the “average” behaviour of a membrane is almost in accordance with that of a simple oscillator. In equation (2) we only considered the mass, compliance and resistance of the oscillator itself. The medium (air) will, however, react against the movement and thus load the oscillator. The contribution of this load may be represented by a radiation impedance 2, which must be added to the impedance of the oscillator itself. This is also a complex quantity, and it consists of a reactive part (90” out of phase with F) and a resistive part (in phase). At low frequencies one may imagine the’ reactive part j X as representing the load of a thin layer of air which is moved with the oscillator. The resistive part R, represents the load of the radiation of sound energy from the oscillator. We now see that sound-radiating insects are faced with a problem of impedance matching: assuming constant driving force, maximum sound power radiation is obtained if the radiation resistance equals the (internal) resistance of the oscillator. This problem is really serious for most insects. The magnitude of the radiation resistance R , depends very much upon the frequency of oscillation, upon the size of the oscillator, and upon the surroundings. At low frequencies the wavelengths of sound waves are much larger than the diameter of the oscillator, and the sound radiation is very inefficient. This can be understood by considering a large and a small sphere pulsating with the same amplitude of motion at a certain low frequency: during the outward pulsation lasting one quarter of a period, the local compression caused by the sphere will spread h / 4 away (where h is the wavelength of the sound). During this time the small and large spheres will cause a compression wave in the same volume of air, but the magnitude of the local compression depends on the increase in volume of the sphere, i.e. on the third power of the diameter. Therefore, for the sound emission to be effective the diameter of the sound source should be reasonably large compared with the wavelength of sound (Fig. 2). The efficiency of sound radiation also depends upon the surroundings (Fig. 2). Let us consider a plane, circular disc (piston) vibrating in free space. At low frequencies the sound source is small compared with the wavelength; therefore the sound radiation will be inefficient, because the compression and rarefaction on the two sides will cancel each other (short-circuiting). At high frequencies, however, the diameter of the disc will be large relative to the wavelength of sound, and the sound radiation will be quite effective. If the same disc is placed in an infinite flat baffle, the radiation resistance at low frequencies increases considerably (Fig. 2).

BIOPHYSICAL ASPECTS

OF SOUND COMMUNICATION IN INSECTS

251

Fig. 2. The radiation impedance R , of a free disc and a disc mounted in an infinite baffle. r is the radius of the disc, and k is proportional to frequency (cf. the text). (Redrawn from Beranek, 1954.)

In the free disc it increases with the fourth power of frequency, but only with the second power in the case with an infinite baffle. Most insect sound producing organs are somewhere between these two extremes. It is not surprising that modifications of the structures surrounding the sound emitter may cause considerable changes in the efficiency of sound emission. Finally, it should be mentioned that R , is proportional to the density and sound velocity of the surrounding medium. This is important for some insects, because a change in the surrounding medium may enable the insect to use other frequencies for communication (see below). 2.2

TYPES OF SOUND FIELDS

Several types of sound field exist, and each has special properties. It is, therefore, important for the experimenter to know the kind of field he is working with. It is practical to distinguish bctween” two types of sound fields: the spherical near fields, and the plane or diffuse far fields. But the limits between them are not sharp.

252

AXEL MICHELSEN AND HARALD NOCKE

2.2.1 Near fields Close to a sound source the sound waves will not be plane, but more or less spherical. In a spherical sound field the pressure gradient will not be a simple function of the sound pressure. Figure 3 shows the additional output obtained from pressure gradient receivers (ribbon microphones or Some insect ears) when placed at five different distances from a small sound source. From a biological point of view, the Figure demonstrates why insect ears sensitive to pressure gradient may hear faint sounds close t o a sound source. From an experimental point of view, the Figure is a warning against placing preparations of insect ears too close to the sound source; it should be remembered that many insect ears are mixed pressure and pressure gradient receivers of unknown properties, so in most experiments it is not possible to correct for the near field conditions.

m

U

HZ

kHz

Fig. 3. The surplus pressure gradient (in dB relative to that in a free far field of the same sound pressure level) for various distances from the sound source.

2.2.2 Far fields There is no sharp limit between near and far fields. Figure 3 demonstrates that in practice the far field starts about one wavelength away, if the sound receiver is not too large. In most biological experiments one should work with a “free” sound field, i.e. the sound waves should be almost plane and travel in one direction only. The creation of a good approximation to a free field is not very difficult within the frequency range used by most insects

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION IN INSECTS

253

(see Michelsen, 1971a). At lower frequencies much larger experimental rooms and sound absorbers are necessary; in practice the distance between the sound absorbing walls should be at least one wavelength of the lowest frequency used. If the walls of the room are reflecting most of the sound, the sound waves will travel in all directions within the room, and the sound field will be more or less diffuse. However, the creation of a well-defined diffuse field is a very difficult task, so in most experiments it is wise to avoid reflected sound waves (“echoes”).

3 Sound production in insects 3.1

GENERAL ASPECTS

The design of a biologically efficient sound transmitter in insects is limited by the physical properties of the sound-propagating media (air, water, solids) and by the small size of insects, the nature of their ectoskeleton, and the properties of their muscles. The relationship between size and efficiency of sound emission has been discussed in a qualitative way in section 2.1. If we reduce the sound radiating structure of the insect to a rigid disc vibrating in an infinite flat baffle (Fig. 2 ) , the active acoustic power P emitted per unit area for kr Q 1 is given by (Beranek, 1954):

P = R,uZ = nfck2r4u2/ 2 Watt

(4)

where R, is the radiation resistance, u is the velocity of vibration, 5 is the density of the medium, c is the velocity of sound in the medium, k = w / c = 21rf/c is the wave number, and r,is the radius of the disc. Thus, the acoustic power depends upon the radiation resistance R, and upon the velocity of the oscillating disc. The radiation resistance varies with the characteristic impedance c c of the medium, the radius of the disc, and the frequency f of the emitted sound. The frequency range of efficient sound emission may be found by simple calculation using Fig. 2. For constant u, a disc will transfer a maximum of acoustic power t o the surrounding air if R , / c c = 1 (* in Fig. 2). This occurs at vibration frequencies where kr > about 2 (Fig. 2 ) , i.e. 21rfi.l~> 2. With r = 3 mm (about the size of the harp in crickets) anti c = 340 m - I , one finds f > 33 kHz. It is apparent from Fig. 2 that the sound powlrr radiation of the harp in a baffle will be 10 times less efficient at kr “0.6 (f”10 kHz). If the harp is like a free disc, one finds this efficiency at kr 21 1.3 (f 20 kHz). From this example we see that insects can only obtain an effective sound emission in the kHz range, and that small insects have to use ultrasonic frequencies. However, several factors set an upper limit to the use of very high sound

254

AXEL MICHELSEN AND HARALD NOCKE

frequencies, which at first glance would appear ideal for sound communication (see below). In these calculations the insect was supposed to use air as the medium of sound propagation. One possibility of compensating for the small dimensions of the sound radiator is to use solids instead of air as sound propagating medium. Some stridulating ants (Attu) make use of this interesting possibility. In air these small insects emit most of the sound between 20 and 60 kHz. However, if they become buried in the earth, they emit a “distress call” with a maximum intensity at 1-3 kHz. This call is propagated through the ground and causes their companions to come to their rescue (Markl, 1965, 1967). This frequency shift is partly explained by the fact that the characteristic impedance {c is much higher for solids than for air. The use of denser media causes an increase in the radiation resistance R,, so that an efficient sound emission is possible at much lower frequencies. On the other hand, the sound signals are attenuated about 6 dB cm-’ of travel through the ground (Markl, 1968). Thus, the improved conditions of sound emission gained by the use of solids as a medium must be paid for by a short range of the sound. Markl observed a behavioural reaction towards a buried soldier up to a depth of 5 cm. Still, this range corresponds to about 10 m, if compared o n a human scale (Markl, 1971).

3.2

MECHANISMS OF FREQUENCY MULTIPLICATION

All active movements in insects are based upon muscle contractions. The contraction frequency does not exceed 1 kHz, even in myogenic muscles. Since insects can only achieve an effective sound emission in the kHz range, the muscle contractions are not suited for driving the sound radiating structures directly. The carrier frequencies in the kHz range are generated by means of a frequency multiplier which converts each muscle twitch into a vibration of higher frequency.

3.2.1 Stridulation Stridulation has been defined as “sound emission by friction of differentiated regions” (Dumortier, 1963). However, in most cases the sound is not produced by the frictional process itself, so it is more appropriate to define stridulation as “frequency multiplication by friction of differentiated regions of the ectoskeleton” (Leston and Pringle, 1963; Nocke, 1972). Numerous examples of stridulation apparatus have been described (see Dumortier, 1963). Despite their different anatomical position they are surprisingly similar: they normally consist of a “file” (pars stridens) and a “scraper” (plectrum). Crickets, for example, use the elytra-elytral method of stridulation (see Dumortier, 1963). In contrast to tettigoniids, the Gryllids have almost symmetric elytra. The right elytrum covers the left.

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION

IN INSECTS

255

Fig. 4. The morphology of the sound-producing apparatus in crickets. (a) The right wing with the harp (H) and the mirror (M). (b) Three teeth from the file. (c) The scraper.

The stridulatory apparatus consists of a file (Fig. 4(b)), located on the ventral side of vein v7 (Fig. 4(a)), and a scraper (Fig. 4(c)), situated on the inner edge of the elytrum. The sound is usually generated only during the closing stroke of the wings (Pierce, 1948; Davis, 1968), when the left-hand scraper is moving over the right-hand file “against the grain”. Closely associated with the stridulatory apparatus is the harp (see Fig. 4(a)). The chitin lamina in the harp region has a total diameter of only 2.5 pm, as may be seen from cross-sections of the tcgmen. The surface of this lamina is corrugated’ and the vein vl-v4 (diameter c. 16 pm) subdivides the lamina into sections of different sizes. (The functional significance of the harp will be discussed in section 3.3.) The physics of stridulation, i.e. the nature of’ the vibration generated by the movement of the scraper over the teeth of the file, has been discussed by Mark1 (1968). His results may well apply t 3 all stridulating insects. It must be stressed, however, that so far no experimental study has been performed. We therefore do not know how well the- actual vibrations fit with theory. When the scraper is moving over a tooth of the file, different forces can be expected between the scraper and tooth (see Fig. 5 and

256

AXEL MICHELSEN AND HARALD NOCKE

f

Fig. 5. Expected events during stridulation. (a-c) Shows the expected forces acting during the movement of the scraper over a tooth (shown in d). (e) The expected sawtooth vibration and its Fourier spectrum. (Redrawn from Markl, 1968.)

Markl, 1968, for details). These forces cause the file and its immediate vicinity to be bent downwards. After the tooth peak has passed, the elastic counter-force will drive the file back. From this elementary process of stridulation one may expect a vibration of the cuticle in the vicinity of the file with a waveform similar to a sawtooth (Markl, 1968). By Fourier analysis it can be shown that a periodical sawtooth function contains the fundamental frequency as well as all harmonics. The amplitude of the harmonics decreases inversely in proportion to their ordinal number. The sawtooth spectrum is therefore

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION I N INSECTS

257

essentially determined by the fundamental frequency, i.e. the tooth impact rate of the stridulation. In addition t o the periodic sawtooth vibration, a series of periodic shocks contribute to the elementary process of stridulattion (Markl, 1968). These shocks occur each time the scraper hits the slope of a tooth after having passed over the preceding tooth. The Fourier analysis of these periodic impacts gives a constant and frequency-independent spectrum as long as the impact duration A t is much smaller than the period T of the tooth impact rate (in practice this is the case when A t < T/lO; see Skudrzyk, 1954). The periodical impacts generated during the counter movement of scraper and file may explain the origin and the predominant role of the ultrasonic frequency components in many insect sounds (e.g. the frequency components of the airborne sound emitted by leaf-cutting ants-Markl, 1968).

3.2.2 The click mechanism of cicadas The sound transmitter of cicadas is the tymbal. This doubly curved chitin plate is a monostable system. After having been dented inward (which is achieved in the living animal by the tymbal muscle) it jumps back into the outposition (see Pringle, 1954). In both positions the tymbal receives a shock which causes it to make a series of dampsed vibrations at its natural frequency (see Fig. 6). The vibration frequency i!; far superior to the rate of the in-out clicks of the tymbal (i.e. the muscle contraction frequency), so that a frequency multiplication is achieved. The oscillating tymbal functions not only as a frequency multiplier, but it also emits the sound.

Y

Fig. 6. The vibrations produced by a single in-out click in a cicada. The time scale is in ms. (From Pringle, 1954.)

The tymbals of Magacicada, for instance, are fitted with 12 strengthening ribs which click in 12 sreps t o the in-position. Each tymbal muscle contraction therefore generates a series of 12 sound" pulses (Moore and Sawyer, 1966; Reid, 1971). A similar mechanism has recently been described for the Australian cicada Abricta curvicosta (Young, 1972).

258

3.3

AXEL MICHELSEN AND HARALD NOCKE THE SOUND RADIATOR

3.3.1 Vibrational properties The vibrational properties of the sound-radiating structure have so far only been studied experimentally in crickets (Nocke, 1971). The following section will therefore be restricted to the sound radiator of Cryllus campestris, i.e. the harp. The vibrational properties of the mirror of crickets and the tegminal resonators have been discussed by Nocke (1971). The efficiency of sound emission depends greatly on the vibrational amplitude of the tegminal sound radiator (see equation 4). The relatively small amplitude of vibration generated during stridulation may be considerably increased by the resonance of the radiator. The resonance phenomenon has been discussed in section 2.1. At the resonance frequency the mechanical impedance is a minimum and equal to the damping (equation 2). The amplitude of vibration of the sound radiator will therefore be a maximum if the energy is fed in at a rate corresponding to the resonance frequency. At resonance all the energy is used to overcome the frictional losses in the vibrating structure and in the surrounding medium. The latter loss is equal to the sound radiation. At steady-state the energy fed in balances the energy used, but at the beginning and end of the vibration there will be a period of build-up and a period of decay. The time needed for build-up depends upon the magnitude of the resistive loss. A small resistive loss (small damping) means a long-lasting build-up and decay and a sharp tuning of the resonance curve (high Q value). The acoustic power P emitted depends upon the 2nd power of the vibration amplitude of the sound radiator, cf. equation 4. The acoustic power can consequently be boosted by the use of a resonating sound radiator. I t is therefore of interest to determine the resonating properties of the radiator, in c a w the harp. The harp is the radiator of the predominant 4 kHz component of the calling song (see below). It is of triangular shape (Fig. 4(a)) and, physically speaking, a plate. If it were circular, its resonance frequency would be expected to be near 5 kHz (Nocke, 1971). In some experiments the harp was stimulated by airborne sound (Nocke, 1971). It must be kept in mind, however, that in the living animal the harp is excited from the file. This difference in the excitation is of little importance for the fundamental mode of the harp vibrations, but not for harmonics. The study of the oscillation properties of the harp, therefore, had to be restricted to the fundamental mode of the harp vibrations. This shortcoming has been accepted because the principal sound energy is determined by the fundamental mode (near 4 kHz for the harp). If the harp is stimulated with sound of varying frequencies its vibrational

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION IN INSECTS

259

amplitude will be a maximum near 4 kHz (Fig. 7(b)). This is close to its calculated resonance frequency and to the predominant frequency component of the calling song. When the harp is dusted with fine cork powder its vibrations near 4 kHz can be observed as a vigorous oscillation of the powder on the harp surface (see Fig. 7(a)). The cork dust outside the harp area remains at rest.

Fig. 7. The vibration of the cricket harp when driven by sound waves. (a) Vigorous movement of cork powder over the harp (sound level 125 dB). (b) The resonance near 4 kHz. (c) The gradual decay of the harp vibration at the end of sound stimulation.

The harp may also be caused to vibrate by 1.he calling song of another Gryllus male, even at distances up to 1.5 m (Nocke, 1971). The harp could therefore also be used for the reception of sound. However, because of the problem of impedance matching (section 21.1) this would be very inefficient. In these experiments the relative amplitude and phase of vibration can be measured by means of a capacitance electrode. This reveals a typical resonance curve (Fig. 7(b)). The Q value can be found from the curve or from the build-up or decay of the harp vibrations (see Fig. 7(c) and Nocke, 1971). In Fig. 7(b) the Q value is about 28. This m a n s that the relative vibrational amplitude is 28 times greater at resonance than at much lower frequencies. This is to say that because of the resonance the acoustic power

260

AXEL MICHELSEN AND HARALD NOCKE

emitted by the harp will be increased by a factor (28)’ (cf. equation 4). It is interesting t o note that the Q value of the sound radiating tymbal of some cicadas is of the same order of magnitude as that of the harp in crickets. The sound pulse (Fig. 6 ) emitted by one tymbal action of the cicada, Platypleura capitata, for example, is an almost pure sinusoidal oscillation with an exponential decay (Pringle, 1954). From the decay shown in Fig. 6 one finds a Q of about 24. In comparison, the average value of the Q determined from the decay in Gryllus campestris is 19 (Nocke, 1971). The resonance frequency and Q factor of the harp are independent of the location on the harp surface (Fig. 8), i.e. although the harp is divided into sections by the veins, it vibrates as a unit. The uniformity of the harp vibration throws some light on the functional significance of the harp veins (vl-v4, see Fig. 4(a)). These veins and the corrugated structure of the chitin lamina between them increase the flexural stiffness and, therefore, the natural frequency of the relatively large harp, without increasing the harp mass (“studding principle”). Thus, the stiffness of the harp is mainly determined by the veins. The resonance frequency is determined by the stiffness and the effective mass, which may not be quite identical to the real mass. The effective mass can be found if the harp is treated as a “simple oscillator” (Morse, 1948), the resonance frequency of which is given by

where fo is the resonance frequency, K is the stiffness constant, and m is the mass. With this equation the effective mass may be found by fastening a known additional mass on the harp surface and measuring the change in resonance frequency (Nocke, 1971). In such an experiment the resonance frequency of a harp was 4.85 kHz before and 4.45 kHz after the addition of 3 x lo-’ g. From this observation we can calculate the effective mass of the resonator t o be 1.6 x 10-4g. This value may be compared with the measured harp mass of 1.9 x 10-4g. The difference between the effective and the real mass is very moderate. 3.3.2 Sound radiating properties So far we have been examining the vibration properties of the sound radiator (harp). Its role in sound emission has still to be discussed. The distribution of sound level over the tegmen (Fig. 9 ) of a Gryllus male shows that the harp emits most of the predominant 4 kHz component of the calling and rivalry songs. After removal of the harps the sound pressure level decreases about

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION IN INSECTS

261

Sound frequency (kHz1

Fig. 8. Resonance curves obtained from various parts of the harp in the cricket.

46 dB (200 times). The decrease is gradual and independent of the order of removal of the harps. Thus, the two harps radiate sound t o about the same extent. The frequency spectrum of the calling and rivalry songs is a maximum at 4 kHz, but in the courtship song it is. at 14 kHz. It is not known which part of the tegmen radiates the 14 kHz component.

262

AXEL MICHELSEN AND HARALD NOCKE

Fig. 9. The distribution of sound level over the cricket wing.

3.4

THE DRIVING VIBRATION AND RADIATED SOUND

Cicadas use the same structure (the tymbal) for both frequency multiplication and sound emission. The properties of the emitted sound signal are thus determined by the vibrational properties of the tymbal (Pringle, 1954) and the general laws of sound emission. In stridulating insects, on the other hand, frequency multiplication and sound radiation are linked to different structures. Here, the properties of the emitted sound also depend upon the interaction between the stridulatory apparatus and the sound radiating structure. The following possibilities exist.

3.4.1 “Non-resonant ” sound radiator: ant If none of the predominant frequency components of the driving vibration correspond to the resonance frequency or frequencies of the sound radiator, and/or if the sound radiator is heavily damped, the frequency spectrum will be broad and without a clear maximum. The stridulating ants (Attu) studied by Mark1 (1968) are a typical example of this. Their stridulatory apparatus forces the cuticle of the whole body into vibrations, which then emit sounds. There is no high-Q resonance system, so the whole spectrum (Fig. 5(e)) of the driving vibration is emitted.

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION IN INSECTS

263

The frequency spectrum of the driving vibration in Atta extends from frequencies below 1 kHz to ultrasonic frequencies. Because of their small size the ants emit most of the sound at 20-60 kHz when stridulating in air. When buried in the ground, Atta emits sounds with a maximum intensity at 1-3kHz (see section 3.1), but one would also expect some ultrasonic sound in this situation. The reason for the absence of jound above 5 kHz appears to be the heavy damping of higher frequencies in the substratum (e.g. the substratum acts as a low pass filter). 3.4.2 Resonant sound radiators: crickets If the resonance frequency or frequencies of the lightly damped sound radiator coincides with one of the predominant frequency components of the driving vibration, the emitted sound will show a clear maximum at a certain frequency or frequencies. The cricket, Gryllus campestris, is a typical example of this: the 4 kHz carrier frequency of two of the songs is equal to the tooth impact rate (Nocke, 1972). Furthermore, the tooth impact rate (the fundamental frequency of‘ the sawtooth vibration generated during stridulation) is equal to the resonance frequency of the sound radiator (harp). The courtship song, on the other hand, has a carrier frequency of 14 kHz. Here again, this agrees well with the tooth impact rate of this song. However, nothing is known so far about the 14 kHz radiator. 3.4.3 Tet t igo niids The physics of sound emission in tettigoniids is riot quite understood but it is clearly different from that of crickets. In tettigoniids, the tooth impact rate is much lower than the frequency of the emitted sound. In the tettigoniid, Phlugis, for example, the tooth impact rate is only 800 Hz (Fig. lO(a)). The carrier frequency of the song is around 55 kHz (see Fig. 10(b) and Suga, 1966). Figure lO(a) shows that a series of six heavily damped oscillations lasting about 0.11 ms are emitted by the sound radiator at every tooth impact. We do not know whether the 55 kHz vibration is equal to the resonance frequency of the sound radiator. One possibility is that the frequency spectrum. of the sawtooth vibration contains the 55 kHz as a component. However, with a basic frequency of 800 Hz such an ultrasonic component is likely t o be rather weak (see above). An alternative possibility is that another structure could be set into a damped series of 5 5 kHz vibrations by the “kick” from the tooth impact. This mechanism would be similar t o the ‘‘click’’ action of the tymbal in cicadas (see above). In addition to the sound measurements of Bailey (1971) and Bailey and Broughton (1971) we very much need a direct analysis of the vibrational properties of the different tegminal structures of tettigoniids.

264

AXEL MICHELSEN AND HARALD NOCKE

Fig. 10. Oscillogram (a) and sonagram (b) of the song of the tettigoniid Phlugis. (From Suga, 1966.)

3.5 THE SOUND GUIDE At ‘Yow’’ frequencies, the radiation resistance (R,) and thus the sound radiated from a disc depends very much upon the surroundings (Fig. 2). A free disc is not an efficient sound radiator, because the compression and rarefaction waves of the two sides will interact (cf. section 2.1). At high frequencies (above * in Fig. 2) there is not much difference between a free disc and a disc in an “infinite” baffle. However, very few insects use very high frequencies, so most insects can increase their acoustic power considerably by means of a sound guide (a baffle or a horn). 3.5.1 The ‘ffiee disc”: Drosophila Drosophila males produce a series of sound pulses during courtship by vibrating their wings (Spieth, 1952; Shorey, 1962). The wings are small (area 1.2 mm’), and they are moved very slowly (about 200 Hz), so very little sound is produced. Thus, the function of the wing beat is not t o produce real sound waves (pressure waves), but t o create local particle movements (“air flows”), which can be detected by the movement-sensitive antennae of the female (see section 5.6.3). However, it is interesting as a

BIOPHYSICAL ASPECTS

OF SOUND COMMUNICATION IN INSECTS

265

reference because it demonstrates the “starting point” in the evolution of sound production in insects. The acoustic power emitted by both sides of the wing amounts to only W (Bennet-Clark, 1971). This value may be compared with the 3x acoustic power of the cricket, which is about 6 x W, i.e. eleven decades larger. If the Drosophila wing were placed in a baffle, the power would be about l o 6 times larger. By increasing the frequency o f vibration of the “baffled wing” (by means of frequency multiplication) up to say 10 kHz, the acoustic power would be increased (10 000/200)2 = 2500 times (cf. equation 4).The Drosophila would thus approach the situation of the cricket. These calculations are, of course, pure speculation, but they do indicate the enormous gain in acoustic output obtained by means of baffles and frequency multiplication. 3.5.2 Baffled sound radiators Instead of using an “infinite” baffle the insects could back their sound radiator with a closed box (as in a domestic loudspeaker cabinet). This has been done by the cicadas. The reason why this efficient method is not more commonly used is probably to be found irk the evolution of soundproducing mechanisms: The stridulatory mechanisms involving the wings or legs (in Orthoptera) make use of some muscles also used in locomotion, and they probably evolved from incidental sound production during flight or walking (see Huber, 1962). In other insects, e.g. beetles, the use of sound for communication may have evolved from tactile signals (Michelsen, 1966b). A baffle does not need to be “infinite”. In practice, it merely has t o be at least one quarter of a wavelength of the smallest wavelength to be emitted in order to act approximately as a closed box (Reichardt, 1968). The cricket, Gryllus campestris, has developed something between a finite baffle and a closed box. The harps are situated in the dorsal field of a “box” formed by the dorsal and lateral fields of both tegmina. The baffle function of the tegminal areas forming this “box” of Gryllus campestris has been established (Nocke, 1971). An acoustic s h x t circuit is, however, still possible at the open end of the “box”. As the above rule indicates, the distance between the harp and the distal anal field should be at least one quarter of a wavelength, i.e. approximately 2 crn for the calling song. It is, however, only about 0.7 cm (see Fig. 4). This seems to impair the efficiency of sound emission. 3.5.3 The horn of molecrickets Molecrickets (e.g. Gryllotalpa vinae) are the only insests known to modify their surroundings for acoustic purposes (Bennet-Clark, 1970). They build their burrow as a double exponential horn (Fig. 11). This horn acts as an

266

AXEL MICHELSEN AND HARALD NOCKE

,Constriction

Fig. 1 1 . The burrow of mole crickets. (From Bennet-Clark, 1970.)

acoustic transformer considerably increasing the efficiency of sound emission by the harps vibrating at the horn throat. A long horn placed in front of even a very small, vibrating disc boosts the emitted acoustic power because a unit area of the vibrating disc is loaded with a radiation resistance of the order of {c (i.e. R , / { c = 1, see Fig. 2). The horn also directs the sound, so that it is transmitted into a smaller spherical angle (Fig. 12). In molecrickets the sound level at ground level is 10 dB higher in the longitudinal axis of the animal than perpendicular to this axis. The radiation resistance (and the emitted acoustic power) of an exponential horn drops to zero below the so-called “cut-off frequency” (lowest frequency which can be emitted effectively by the horn). For the “singing burrow” of Gryllotalpa vinae a cut-off frequency was calculated of less than 1 kHz (Bennet-Clark, 1970). This is well below the carrier frequency of the calling song (3.5 kHz at 16°C). In addition, a bulb (see Fig. 11) probably tunes the horn in such a way that it is an almost purely resistive load t o the harps (vibrating at the horn throat). The efficiency of sound emission depends greatly on the acoustic

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION IN INSECTS

267

0" I

2m

!m

-900

I

180"

Fig. 12. The directional distribution of sound around the burrow of a mole cricket. Radial distances to 80 dB isobars are plotted for various latitudes. Vertical is Oo and ground level is 90°. (From Bennet-Clark, 1970.)

properties of the horn-shaped burrow. This becomes evident if Gryllotalpa vinae sings (the rivalry song) outside the burrow. The acoustic power is then only approximately 4 per cent of that produced in the burrow (Bennet-Clark, 19 70).

3.6

EFFICIENCY OF SINGING

Gryllotalpa produces a calling song which consists of a continuous series of sound pulses. The song can last for more than 1 h. This calling song is very intense and can be heard by humans over distances up to 600 m. The peak sound pressure level 1 m above the burrow is 92 dB (Bennet-Clark, 1970). From the surface area of the 80 dB isobair (Fig. 1 2 ) a total acoustic power of about 1.2 mW was calculated for the song of Gryllotalpa (Bennet-Clark, 1970). In comparison, the human" voice generates an acoustic power of only 2.5 x lo-' mW during normal conversation. The estimated power of the muscles used for singing is only 3.5 mW. Thus, the

268

AXEL MICHELSEN AND HARALD NOCKE

efficiency of the molecricket sound transmitter is about 30 per cent. This far exceeds that of a domestic loudspeaker (Bennet-Clark, 1970). In Gryllus campestris, on the other hand, the muscle power was estimated as 1.2 mW. The mean sound power of Gryllus is about 6.10-* mW (at 3OoC), so the efficiency of converting muscle power into acoustic power is 5 per cent (Bennet-Clark, 1970). This corresponds roughly t o the efficiency of a domestic loudspeaker.

4 Propagation of insect sounds The pressure of sound waves travelling from a spherical sound source will decrease 6 dB (2 times) per doubling of the distance to the origin. This is a simple consequence of the geometry of the surface of the air space (sphere) occupied by the energy of the sound wave. However, in addition there will be a gain due to reflection from the ground and a loss due to absorption in the medium and to scattering and absorption by the ground and vegetation. In recent years much work has been done on these problems, but most investigators have been interested in the reduction of noise from cars or aeroplanes. The frequency range and the position in space of these sound sources are rather different from those of singing insects, so in most cases we can only guess about the parameters important for insect communication. Experimental work in this area is badly needed. Absorption in air is caused both by viscosity and heat conduction in the air and by the collision of molecules, where translational and vibrational energy is exchanged. The attenuation of sound caused by both processes and indicated as a number of dB per metre increases with a little more than the first power of frequency. The absorption is a fairly complicated function of temperature, relative humidity and sound frequency. Nomograms for determining the theoretical attenuation caused by these factors can be found in several textbooks (e.g. Beranek, 1954). However, according to recent experimental work the measured attenuation is often far from the expected. Attenuation has been studied as a function of temperature, humidity and frequency within the frequency range 1-12.5 kHz (Evans and Bazley, 1956; Harris, 1967). The reader is referred to these papers for detailed information. Here, we will concentrate on the order of magnitude of the absorption at various frequencies. In Fig. 1 3 some degrees of additional attenuation (in dB m - I ) have been plotted together with the attenuation due to distance (curve A). Their sum will then give the total attenuation. In evaluating the effect of the total attenuation it should be remembered that for most insects the threshold of hearing is about 40-60 dB below their sound level measured 1 m away. Apparently, an additional attentuation of

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION IN INSECTS

8o 70

-m -u ._ c

t

5

2

I

I I

I I

I

60 -

I

269

I dBm-1

I I

1

50-

0 ZJ

k

c

t

40-

30

-

20 -

I

2

3 4 5

7

0

- 15-20 30

5 0 75 100

Distance (m)

Fig. 13. The attenuation caused by various factors (see the text) compared with that due to distance (curve A).

1 dB m-' will limit the sound communication to about 20-30 m for most insects; 0.5 dB m-' gives 50-100 m, and so on. Below 1 kHz the absorption is very small. At higher frequencies and within the range of temperature and relative humidity 1O-3O0Cand 5-100 per cent, respectively, the attenuation varies within the following ranges: at 4 kHz it is 0.01-0.1 dB m-I; at 8 kHz, 0.05-0.25 dB m-': at 12.5 kHz, 0.1-0.4 dB m-'. Above 12 kHz the experimental studies are scarce, but at 26.5"C and 37 per cent relative humidity this attentuation has been measured up to 500 kHz (Fig. 14; Sivian, 1947). This figure clearly demonstrates why most insects prefer rather moderate frequencies. At 50 kHz, for example, the attenuation will limit the range of communication to about 10-20 m. This is illustrated in tettigoniids, which have their dominant frequencies in the range 8-40 kHz. A comparative study shows that species with low and high population densities use the 8-15 kHz and 20-40 kHz range, respectively (Dubrovin and Zhantiev, 1970). The absorption caused by vegetation and by the ground is of particular interest for the propagation of insect songs. Scattering and ground attenuation are the principal factors in sound attenuation by vegetation (Aylor, 1971). The ground attenuation depends upon the distance between

270

AXEL MICHELSEN AND HARALD NOCKE

'V

2

1

I02

10

I02

5

1

Frequency (kHz)

Fig. 14. The attenuation in air at ultrasonic frequencies. The broken line indicates the measured attenuation at 26'C and 37 per cent relative humidity (from Sivian, 1947). The full line was calculated by Evans and Bazley, 1956.

the sound source and ground. If the sound source (insect) is situated about 1 m from the ground, then the attenuation is a maximum between some hundred Hz and about 1-2 kHz; it can be neglected since very few insects communicate in this frequency range. However, many insects sing on or near to the ground; no experimental data are available for this case, but ground attenuation may be expected to be important also at higher frequencies. Trees and corn are very effective in reducing the transmission of sound in the frequency range used by insects. Corn crops have an excess attenuation of about 6 dB per 100 ft for each doubling of frequency between 500 and 4000 Hz, and the attenuation at 4 kHz is about 1 dB m-l. At 4 kHz the excess attenuation for brush is about 0.5 dB m - l , and that for pine about 0.2 dB m - l . In all cases the attenuation increases with frequency, but no exact data seem to be available for the ultrasonic range. The effect of these attenuations may be evaluated from Fig. 13. The velocity of sound depends upon the temperature of the medium.

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION I N INSECTS

271

This is likely to be important for the propagation of insect sounds, because singing insects are often situated in a temperature gradient. In the daytime the temperature at the surface of the ground may be considerably higher than that of the air some metres from the ground (see Lowry, 1970). One may expect that the song is refracted upwards in the air during the daytime, because the sound wave will travel faster near to the ground than in the air above. Thus, less sound will reach a listener sitting on the ground some metres away. In contrast, at night or after rain the temperature at the ground surface may be less than that in the air .lbove. In this situation the song will be refracted downwards from the warmer air above the ground, i.e. there will be a gain in sound level for listeners sitting on the ground. Similar effects are caused by a gradient in the speed of the wind, but in this case the effect will, of course, depend on the direction of the wind. Nothing is known about any behavioral adaptations of insects to these physical phenomena.

5 Insect ears as sound receivers The anatomy of insect hearing organs differs widely (see Autrum, 1963; Michelsen, 1973a). The most common types are hair receptors and tympanal organs, but other structures may be used for sound reception (e.g. the antennae of mosquitoes and the palps of hawkmoths). Hair receptors and tympanal organs are obviously very different structures, and tympanal organs evolved independently in sevcral groups of insects. So there is no reason to suppose that insect ears should all work in the same way. Many of the past disputes about the function of “insect ears” have been caused by attempts t o generalize the findings on one type of ear t o cover all hearing organs in insects.

5.1 THE EFFECTIVE PARAMETERS OF

SOUND

The idea that insect ears are “displacement receptors” (Pumphrey, 1940) has been accepted by many biologists. In the following we shall argue that some “ears” (e.g. hair receptors) are movement detectors; most insect ears are, however, pressure gradient receivers (Aurrum, 1940) or pressure receivers-or something in between. Pumphrey’s idea is probably correct for sensory hairs, but it is misleading for most insect ears. The reason is that, although the sound waves and particle movements have a direction, the sound pressure is a scalar quantity. The directivity of insect ears is only comprehensible, if one realizes the difference between the particle displacement and the pressure gradient of sound waves: So the problem is not only a question of terminology. Pumphrey pointed out that sound waves may be described both as

27 2

AXEL MICHELSEN AND HARALD NOCKE

movements of air particles in the direction of propagation of the wave and as alternating changes of pressure. Further, he wrote: “. . . such propagated disturbances can be detected by an instrument of appropriate sensitivity and speed which is responsive either to pressure changes OT to displacements of the medium. An instrument of the former type is comparable in principle to an aneroid barometer; one of the latter type is comparable in principle to a wind-gauge.’’ And: “. . . the diaphragm (of insect ears) or moving vane (sensory hair) should ideally be without mass, and the hinge or suspension completely pliant so that the moving parts follow the movements of the air exactly. ” The effect of the movements of air particles can be observed in a microscope, if one observes the movements of oil particles of about 1 pm in diameter floating in the air. Such small particles have so small a mass that they will follow the undulatory motion of the air particles rather well. Probably, the same will be true for some insect hairs, which are fastened to the body so that they are able to move rather freely; this problem has, however, never been studied from a physical point of view. The sclerotized ring encircling the tympanal membrane of the locust may be removed from the animal and mounted on a small platform (see Fig. 15(b) and Michelsen, 1971a). Such an “isolated” ear is similar t o a ribbon microphone, and the driving forces can be calculated from the pressure gradient, i.e. the sound pressures acting o n the front and back of the membrane (see below). If the membrane is turned in the sound field, the driving force will decrease. It reaches zero when the membrane is parallel to the direction of the sound waves, since the sound pressure is now the same on both sides. This result is also evident if one uses the particle movements (displacements) for the argument. Now, let us place the membrane in the wall of an open “box” as shown in Fig. 15(d). Here, the membrane is perpendicular to the longitudinal axis of the open “box”, and it is parallel to the direction of the sound waves. In this position, which resembles that found in situ in the locust, the force acting to move the membrane will be a maximum. This is obvious, since the difference in sound pressure acting on the front and back of the membrane will be determined by the orientation of the ‘‘box” and not by the orientation of the membrane. However, it is not t o be expected from a consideration of the direction of the particle movements. This difference between the acoustics of hair receptors and tympanal organs was realized by Autrum (1936, 1940), who, in developing the theory that some auditory organs in arthropods are “velocity” (movement) receptors, says: “Die entwickelte Theorie gilt fur allen Arthropoden, soweit sie keine Trommelfelle besitzen.” Unfortunately, this view did not influence the authors of textbooks. In order to avoid confusion, we would suggest that the physical

BIOPHYSICAL ASPECTS OF SOUND COMMUNlCATlClN IN INSECTS

(a)

273

(b)

Fig. 15. A pressure receiver (a), a symmetrical pressure fFadient receiver (b), and an asymmetrical sound receiver (c). In (d) ,the approximate position of the tympanum in some locust species is indicated (cf. the text).

parameters of sound acting to move the ears of insects should be referred to as movement reciivers (hairs), pressure or pressure gradient receivers (membranes). The terms “displacement receptor”, “velocity receptor”, and acceleration receptor”, on the other hand, should be used for the particular physical parameter of the resulting vibra.tion which is capable of exciting the receptor cells. The sound receiving properties of insect ears may be tested in a standing wave, but exact experiments are difficult to perform, especially at high frequencies. The method was first used by Autrum (1936)who placed ants at different positions in a standing wave and olmrved the behavioural reactions. The ants moved their antennae when they were in the pressure nodes (where sound pressure is a minimum, but particle movement and pressure gradient are a maximum), but not in the pressure antinodes. From this observation Autrum concluded that the non-tympafial hearing organs of insects respond to the movement of air particles. This is, however, not true for tympana1 organs (Autrum, 1936). In theory it should be possible 66

274

AXEL MICHELSEN AND HARALD NOCKE

to determine the relative amount of pressure and pressure gradient properties by measuring the sensitivity in the pressure antinodes and nodes, respectively. The possible sources of error in such experiments are, however, s o large that it is doubtful whether they can be controlled. 5.2

THE FORCES ACTING ON EARS

The force acting to move a tympanal membrane depends upon the amount of sound reaching the back of the membrane. If the membrane is backed by a closed chamber (Fig. 15(a)),the ear is a pressure receiver; the force is the membrane area A times the sound pressure p . The locust ear acts as a pressure receiver to frequencies above 10 kHz (Michelsen, 19 7 1, 197 1c). From their anatomy and frequency range one may also suspect the ears of lacewings (see section 5.6.1 and Miller, 1971) and the slit sense organs of spiders (Barth, 1967) to be pressure receivers. However, it is difficult to use anatomy to guess about the type of sound reception, since sound waves may be quite effective in penetrating the cuticle (Michelsen, 1 9 7 1 ~ ) . Therefore, ears with only one tympanal membrane are not necessarily pressure receivers. The other extreme is the pure pressure gradient receiver (Fig. 15(b)).The force F is given by F = j A p (7rAllh) cos @. (6) Here again we have the area A and the sound pressure p . But we also find the “effective distance” A1 from the front to the back of the membrane, and the angle of incidence (@in Fig. 15(b)). As discussed above, the force is zero when @ is 90” or 270”. j is -and indicates a 90” phase difference between F a‘nd p . The term (27rAl cos @)/A indicates the phase difference between the sound acting on the front and back of the membrane respectively. If A1 is small in comparison with A, equation 6 may be written

F

=

- A ( d p / d x ) A l cos @

(7) where dpldx indicates the pressure gradient: hence the name “pressure gradient receiver”. The phase differences between the sound waves acting on the front and back of the membrane are most important, because the force is determined by drawing the pressures as vectors and finding their difference. It should be noted that the force may be twice that of a similar pressure receiver; this occurs when the sound waves are of opposite phase (180’ difference). The effective distance of true pressure gradient receivers is often small compared with the wavelength of the sound, and consequently the driving force will only be a fraction of that acting on a similar pressure receiver. For example, in the “isolated” locust ear (Fig. 16), A1 is about 0.8 mm and

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION I N INSECTS

275

....--.Thin membrane

- .. Miillerk argon ..

Thick membrrinc

E Fig. 16. The isolated locust ear preparation mounted in a sound field.

the driving force at 3.5 kHz (A = 10 cm) is only about 0.05 PA (Michelsen, 1 9 7 1 ~ )The . forces acting on the intact locust ear Ere, however, of the same order of magnitude as those of pressure receivers. 5.3 INFLUENCE FROM THE SURROUNDINGS

Most insect ears are probably something between a pressure and a pressure gradient receiver (Fig. 15(c)). The force can still be calculated from the sound acting on the front and back of the membrane, but the situation may be quite complex (see Michelsen, 1 9 7 1 ~ )One . has to consider the effect of diffraction which causes a surplus sound pressure on the surface of the body facing the sound source. Also, the tisvues behind the ear may cause attenuation and additional phase shifts for the sound reaching the back of the membrane. 5.3.1 Diffraction

Equations 6 and 7 have been derived on the assumption that the ear and its surroundings are so small relative to the wavelength of sound that they d o not disturb the sound field. In practice, this means that the diameter of the sound receiver must be less than about one-tenth of the wavelength. In the case of the locust (Michelsen, 1971c) the body ha:; the shape of a cylinder which is about 0.7 cm in diameter, i.e. the locust body is likely t o disturb the sound field at frequencies above 5 kHz (A = 7 cm). The effect of diffraction at moderate frequencies is mainly that of producing a surplus sound pressure at the surface lacing the sound source. At higher frequencies the distribution of sound pressure on the surface of the object may be very complicated. It is possible to calculate the effect of diffraction for obstacles of very simple shape (see Morse, 1948; Beranek,

276

AXEL MICHELSEN AND HARALD NOCKE

1954), but the diffraction pattern depends very much upon the shape and acoustical properties of the obstacle. So in practice the effect of diffraction should he studied experimentally. In the locust this has been done by placing a small microphone in the body of the animal with its membrane in the plane of the body surface (see Fig. 17(d); Michelsen, 1971c; Lee Miller, in preparation). Unfortunately, in many cases this kind of approach is not possible. The changes in sound pressure caused by diffraction are of particular interest in studies of directionality. In pressure receivers (like the locust ear above 1 0 kHz, see below) the diffraction is the only source of information about the direction of the sound wave (since insect ears are presumably unable to signal the phase of the sound reaching each ear). In ears working as pressure gradient receivers or mixed pressure and pressure gradient receivers the direction of the gradient affects the force acting to move the tympanal membrane (equation 7). So in this case the effect of the sound direction is composed of two independent factors.

5.3.2 Sound conduction through insect bodies Some tympanal organs have two tympanal membranes separated by air sacs (tettigoniids, noctuid moths). In other cases the two ears are connected by a series of air sacs (locusts). The pressure gradient properties of such ears are normally explained by assuming that sound can travel from one membrane to the rear of the other. In such models the rest of the animal is assumed to behave like a rigid box. However, the bodies of insects are not rigid. The sound will also cause the cuticle to vibrate and some sound will be transmitted through the insect by other routes than via the tympanal membranes. The extra-tympana1 transmission may be fairly efficient at some frequencies and less at others. Sound travelling through the internal tissues and air spaces of insects will also be attenuated. Thus one may expect both the other tympanal membrane, the cuticle, and the internal tissues to act as acoustic filters, which in total determine the amplitude and phase of the sound reaching the rear of the tympanum. a. Locusts. These filtering properties have been studied in the locust (Michelsen, 1 9 7 1 ~ )In . one type of experiment a microphone was placed in the body behind the tympanum (Fig. 17(g)). Below 3 kHz the attentuation was less than 6 dB ( 2 times); above 3 kHz it increased t o 30 dB (at 10-20 kHz) and then gradually decreased. Blocking the tympanum with wax (Fig. 17(h)) had very little effect on the transmission of low-frequency sound, but at higher frequencies the sound level inside the animal was significantly less. This experiment demonstrates that low frequencies are quite effective in passing the cuticle of insects. This fact should be borne in mind when one considers the function of insect ears at low frequencies. A large part of the attenuation in the 10-20 kHz range is apparently due

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION IN INSECTS

27 7

10 ;’if l+It (a)

(b)

(C)

(d

(el

(f 1

(9)

(h)

Fig. 1 7 . The different preparations used in various (rxperiments. (a-c) Shows the recording of nerve responses from the intact, the “opei-ated” and the “isolated” ear. (d-h) Shows the measurement of relative sound pressures by means of a 6 mm microphone operated into the body of locusts.

to a filtering effect of the internal tissues. This W,IS demonstrated in similar experiments where the tympanum and its surroundings had been removed (Fig. 17(e)). The rest of the body now surrounded a tube lined with the internal tissues of the animal and closed at one end by the microphone. The closed-tube resonances of this preparation could now be compared with those obtained when the tissues were coveied with a plastic tube (Fig. 17(f)). Here again there was a considerable sound absorption above 8 kHz. These crude experiments do not provide any exact values for the filtering, but they do demonstrate that very little 10-20 kHz sound can reach the rear of the tympanal membranes. The locust ear is capable of some frequency discrimination (see below). One of its groups of receptor cells covers the frequency range 10-20 kHz (see Fig. 24); and the experiments indicate that in this range the ear acts as a pressure receiver. b. Bush crickets. A curious example of sound conduction to the rear of the tympanum via a non-tympana1 route has been described by Lewis (1973). In some tettigoniids (Homorocoryphus) a part of the prothoracic spiracle gives rise to a trachea which ends at the tympanal region of the leg. The trachea has the form of an exponential horn with a calculated cut-off frequency of 3 kHz. These animals appear to be far more sensitive to sound directed at the open spiracle than to sound incident on the tympanal slits. An ideal horn of these dimensions should act as a transformer with a “gain” of about 3 4 d B , but nothing is known about the damping properties of tracheal tubes. Also, the biophysics of the tettigoniid ear-is not understood, so it is difficult t o say what one should expect in such a system. These problems deserve further study.

AXEL MICHELSEN AND HARALD NOCKE

278

5.3.3 Fat and soft

tissues

The acoustics of the locust ear is not so simple at frequencies below 8-10 kHz. The amount of fat between the ears has a remarkable influence upon the sensitivity of the ear (Fig. 18); at 3-4kHz the senstivity may vary about 30 dB. Damping of the ears in female locusts may also be caused by the presence of a part of the ovaries in the region of the ears. The physics of these damping effects is not understood. The force acting t o move the tympanum is unlikely to vary very much. Also, there is little, if any, variation in the best frequency in preparations with different amounts of fat. This seems t o rule out the additional mass as the cause of the damping. Thus, resistive damping seems to be the most likely cause. The physics of this phenomenon should be studied in more detail.

45

=E 35

,

t

4 7

l5 I0

I

i Frequency ( k H z 1

Fig. 18. The threshold (in dB relative to 2 x lo-’ N m-’) of an extremely meagre locust ( 0 ) and an extremely fat locust (a).

The tissues inside a locust are very soft, and apparently they are also easily moved by sound waves. The acoustically “soft” nature of the tissues surrounding the auditory air sacs can be demonstrated by calculation. The air sacs have a volume of about 0.1 cm3. If a membrane like the tympanum were backed by a rigid-walled box of this volume, then one should expect an increase of some kHz in its basic resonance frequency, because the “stiffness” of the air should add to the tension of the membrane. A minor change in frequency probably does occur, but it is unlikely t o exceed 500 Hz. A change of 500 Hz would correspond to about 0.4 cm3 “effective volume” of the air sacs. Because of the “soft” nature of the tissues, it is difficult to judge how much of the interior participates in the acoustics of the ears.

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION I N INSECTS

5.4

279

THE T W A N A L VIBRATIONS

In the previous sections we have been interested in the factors determining the force acting to move the tympanal membranes of insect ears. We have also seen that the surroundings may influence the tympanal vibrations in other ways: air spaces behind the membrane may contribute some stiffness, and additional mass and resistive damping may come from the radiation impedance (section 2.1) or from damping material (fat) close to the ear. It is therefore practical t o study the tympanal vibrations in an isolated ear preparation, where the tympanum and its cuticular rim have been cut out and mounted in free space. This kind of approach has been used in the locust (Michelsen, 1971b). In the cricket the'vibrations have been studied in the membrane in situ (Johnstone e t al., 1970), and in this case it is not possible to separate the influence of the surroundings from the properties of the tympanum itself. 5.4.1 Expected vibrations of membranes The expected modes of vibration for homogeneous membranes of various shapes can be calculated (see Morse, 1948). It is important to realize that the type of vibration depends upon the way in which energy is fed into the system. In the case of circular membranes, an evenly distributed harmonic force (sound) will cause circularly symmetrical modes of vibration (Fig. 19(a-c)) shows the first three of these). If the membrane is struck at a point, or if energy is fed in unsymmetrically, other modes of vibration become possible (Fig. 19(d-f) shows some examples). This should be borne

Fig. 19. The first three circularly symmetrical modes of vibration in a circular membrane (ac) and some other possible modes (d-f). The nodal lines are dotted. (From Morse, 1948.) (Copyright of and by permission of the McCraw-HillBook Company.) AIP-10

280

AXEL MICHELSEN AND HARALD NOCKE

in mind when studies on membrane vibrations are planned; the energy should, if possible, be supplied in the “natural” way. Circular membranes vibrate with their basic mode at low frequencies (Figs 20 and 19(a)). At resonance the amplitude is a maximum and the phase lag (between driving force and membrane displacement) is 90”. As the frequency is gradually increased, nodal lines will be formed at the border and move towards the centre. The distribution of the phase lags is now more complicated (Fig. 20), the central parts having the largest phase lag. This behaviour is true for an “ideal” membrane with mass and membrane tension (“springiness”) and very little damping. It should be noted that the ideal membranes of textbooks normally lack resistive damping. The introduction of this factor makes the behaviour much more complicated (see Michelsen, 1971b), but in practice the behaviour follows the same pattern as that described above.

( 1 9 00

I

45Oo=9O0

9 00

2

45Oo=9O0

2700

2 700 3

4

Fig. 20. The first four circularly symmetrical modes of vibration in a membrane. The approximate phase lags (between driving force and membrane displacement) are indicated at resonance. 1-3 are identical to a-c in Fig. 19.

The tympanal membranes of insects vary much in shape, and they are not always homogeneous. Thus, they cannot be expected to vibrate in the simple way shown in textbooks. One can, of course, use the determined mass and tension to calculate some expected vibrations, but their existence has t o be tested by experiments. 5.4.2 Observed vibrations of tympanal membranes

The actual modes of vibration of a tympanal membrane has been studied in the locust (Michelsen, 1971b). In this case the membrane is far from being homogeneous, and it has the shape of a bean (Fig. 21). Two-thirds of the total mass is concentrated in the “Miiller-organ” (the scolopophorous units and four cuticular bodies, to which the receptor cells attach). Furthermore, a minor part of the membrane is three times thicker than the rest. Anatomical considerations led to the hypothesis that at least two sets of

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION I N INSECTS

281

kNervec?n Styliform body ( b )

Imm

Folded body (c)

Thin membrane

c& Thin membrane

1

Thick membrane

Posterlor

Elevoted process (a)

Pyriform,vesicle ( d l

Fig. 21. The anatomy of the tympana1 organ. (I) Miiller-organ (right ear). The position of the receptor cells (ad) and the attachment parts of the membrane are shown. Arrows indicate the direction of the dendrites. (11) The left ear seen from outside the animal. The dark areas indicate the areas of attachment of the receptor cells. (111) The right ear seen from the inside. The arrow indicates the visual angle of I.

resonances (corresponding to the circularly symmetrical modes) might exist in this system. One set is due to the entire system, and the other is caused by the “thin membrane” (Fig. 21). The experiments confirmed this view, but in addition at least two other vibrations were observed. Their nature is unknown, but they may be similar to some of the vibrations shown in Fig. 19(d-f). The vibrations were observed by means of laser holography and capacitance electrodes. The former method provides photographic pictures of the spatial distribution of vibration amplitude (Fig. 22). The latter method gives information about the phase lag and relative amplitude at different frequencies for small areas of the tympana1 membrane. Both types of measurement were necessary in order to allow an identification of the vibrations. The observed resonance frequencies were in good agreement with those expected from theory. The most surprising feature of the vibrations is that most of them are extremely localized (see Fig. 22). Also, the spatial positions of the vibrations were far from those expected. They were not at the geometrical centre of the tympanum, and their positions differed for different modes of vibration. The centres of all the modes of vibration of the entire membrane are located in the thick-membrane end of the tympanum (Fig.21), but the different modes do not have the same position. The centres of the thin membrane system are located nearer to the geometrical centre of the tympanum (Fig. 22), but here again the position is not constant.

AXEL MICHELSEN AND HARALD NOCKE

282

Thin membrane Cuticular Insectrpin im

~:&w

1 ;

FTM

I-

Thlck membrane

SEM

Fig. 22. Holographic pictures of the vibrations in the frequency range 3-4 kHz. Above: the orientation of the isolated ear, and the approximate positions of the centres of the fundamental mode of the thin membrane (FTM) and of the second mode of the entire membrane (SEM). Below: the vibration patterns. The dark and light lines on the pictures are loci of equal amplitudes of vibration. Further explanation in the text.

The difference in the spatial position of the two sets of vibrations is also evident from the measurements with the capacitance electrode. At the thin membrane end of the tympanum, the phase lag and variation in amplitude are close to those expected for the thin-membrane system. In the area between the two centres of vibration, the phase lag and amplitude will be determined by both vibrations. The thin-membrane system will normally have the largest influence, because it is about three times less damped than the entire-membrane system. One of the groups of receptor cells (the c-cells) attach to the “folded body” (Fig. 21). This structure is about equally influenced by two vibrations. Figure 23 shows measurements of the amplitude and phase recorded with a capacitance electrode close to the folded body. Vibration I1 is the typical one while I and I11 are extreme types which almost correspond t o the fundamental modes of the entire- and thin-membrane systems respectively. The variation in phase of I1 is very close t o that expected from a system governed by the entire-membrane system at 1-2 kHz and by the thin-membrane system at 3-4 kHz. Between 2 and 3 kHz intermediate phase lags are observed. The variation in amplitude is close to that expected in the 1-3 kHz range, but

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION IN INSECTS

283

Frequency of sound (kHz1

Fig. 23. Variation in amplitude and phase lag measured close to the folded body (c-cells). I and 111 are extreme types which ;Jmost correspond to the fundamental modes of the entire and thin membrane, respectively. I1 is an intermediate type. Further explanation in the text.

the maximum at 3.5 kHz is lacking; the reason for this may be an interaction with the second mode of the entire-mode system (see Michelsen, 197 1b). It is evident that attachment areas situated at a few hundred pm from each other may experience quite different vibrations. Also, an attachment area may be close to one of the nodal circles at resonance and thus avoid being affected by the vibration. The ear is thlzrefore capable of frequency discrimination (Fig. 24). For at least two of the groups the excitation of the receptor cells is roughly proportional to the vibration of their attachment area. a. The cricket ear. The vibrations of the 1:wo tympana1 membranes of the cricket, Teleogryllus commodus, have been studied by means of the Mossbauer technique (Johnstone et al., 1970). In this method a Mossbauer source is fastened to the membrane, and this may influence the vibrations. The resonant characteristics of the two membranes are shown in Fig. 25. The vibration of the largest of the two tympana is a maximum at three frequencies (at 2, 5 and 12 kHz). It is interesting that Nocke (1972) observed some frequency selectivity in the ear of the cricket, Gryllus campestris, and that the preferred frequencies of three groups of receptor cells were close to those mentioned. However, th&e is no experimental evidence for a functional connection between the observed resonances and preferred frequencies. This problem should be studied further.

AXEL MICHELSEN AND HARALD NOCKE

284

-1

80 c 75 ~

0

5

-

7065-

0

2

60-

.-r 5 5 Z 50c II)

= 4540

-

-

Fig. 24. The threshold curves for the four groups of receptor cells in the isolated locust Broken lines indicate variations in threshold curves for different cells within each COUP. ear.

0.00011 I

1

2

I

1

1

1

5

1

1

1

1

I

10

20

Frequency (kHz1

Fig. 25. T h e vibration of the large (A) and small (B) tympana1 membrane of a cricket measured with the Mossbauer technique. (From Johnstone e t nl., 1970.)

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION IN INSECTS

285

The vibration of the small tympana1 membrane was observed in one case (Fig. 25). The amplitude of vibration is considerably less than that of the large membrane. This experiment should be repeated in order to establish the exact difference between the vibrations of the two tympana, since this problem is closely linked up with the directionality of the ear. These observations suggest that the cricket ear is almost a pressure receiver, but the effect of diffraction is very small in the individual leg of the cricket or tettigoniid. One should therefore not expect much directionality in the individual ear. Behavioural studies support this view: after destruction of one ear the animals make circus movements and cannot locate the sound source (Murphey and Zaretsky, 1971). The directionality of intact animals is probably due mainly to the diffraction effect of the entire body. This problem should also be studied in detail.

5.5

THE BEHAVIOUR OF THE RECEPTOR ORGAN

In this paper we do not intend to discuss rhe neural aspects of sound communication (for reviews see section 1). In the following we shall concentrate on the biophysical aspects of the receptor characteristics; we have not, however, tried to cover the cellular transduction processes. 5.5.1 Noctuid moths The characteristics of the most sensitive acoustic receptor cell in the ear of noctuid moths have been investigated by Adams (1970, 1971). In some of his experiments the intensity characteristics were studied by means of sound stimuli. The adaptation and response characteristics appeared to be highly dependent of the duration of the stimulus. The responses to stimuli of a few ms duration are determined by the time integral of stimulus power, and they often outlast the duration of the stimulus (the biophysics of this phenomenon is discussed below). No response “saturation” occurs with these short pulses. The stimulus-response curve measured with 50-ms sound pulses is sigmoid; the firing rate increases over a range of 40-50 dB before the response saturates. This dynamic range is larger than that found in most insect ears, and correspondingly the maximum firing rate (700-900 spikes per second) is about twice that found in the ear of the locust. The entire intensity-response curve does not fit exactly t o a logarithmic nor to a power function, although both functions fit reasonably well to limited parts of the curve. An almost perfect fit, however, may be obtained by means of an empirical equation proposed by Zwislocki et al. (1969). With long (10 s) stimuli there is an initial stage of exponential adaptation lasting for 2 s, during which the firing rate decreases by more than 75 per

286

AXEL MICHELSEN AND HARALD NOCKE

cent, followed by a period of slow adaptation. A monotonic intensityresponse function is found for a brief period after the onset of a long stimulus, but after a few seconds the function becomes non-monotonic. The reason for this behaviour is obscure. In simple RC-circuits the output in response t o a step input is M (1- exp ( - t / 7 ) ) and for the step back t o the baseline M exp ( - t / 7 ) , where M is the maximum response, t is the time and T the time constant. The response of the tympana1 organ was investigated by means of very brief sound stimuli with very short rise times (<0.015 ms) and low intensity (Adams, 1970). A rough estimate of the time constant, obtained from the occurrence of spikes at various times after the onset of low-intensity tone bursts, gives values around 1 ms. This gives us an idea about the ear’s ability for time resolution. The assumption that the ear membrane behaves as an RC-circuit is not far from being correct for the basic mode of vibration, but it is not correct for higher modes. The basic resonance frequency of the moth ear is not known. However, most of the time constant is probably biological, since the frequency response has a Q at 60 kHz of about 14, corresponding to a physical time constant of 0.1-0.2 ms (Adams, personal communication). The finding that the total time constant for the moth ear is about 1 ms explains why the response to very short stimuli is determined only by the time-integral of stimulus power. These experiments were performed with sound stimuli, and. the response was therefore determined by the properties of the tympanum, its surroundings, the receptor organ, and the receptor cells. In order to separate these factors, the tympanic membrane was driven directly with an electromechanical transducer. Figure 26 shows the amplitude threshold curve obtained. Three minima (at 2, 14, and 60 kHz) were found. The occurrence of these minima suggested that the filtering action of the

SIimulus frequency (kHz1

Fig. 26. The average threshold curve for the noctuid ear driven by electro-mechanical transducers. The ordinate gives the amplitude of vibration (-20 dB = 0.1 pm; -80 dE = 1 A). (From Adams, 1970.)

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION IN INSECTS

287

ear is not confined t o the tympanum and its surroundings, but that a transformation of the stimulus also occurs in the receptor organ. Anatomically, the receptor organ joins the 1.ympanum at a rather small angle, so the vibrations of the tympanum will also cause transverse vibrations of the beam-shaped receptor organ. Experiments with a model of the receptor organ suggest that the receptor organ may vibrate as a beam with transverse modes of the first, third, and fifth order at the three frequencies of maximum sensitivity in the experiment described above. The model proposed by Adams should be tested by experiments. Especially, the assumption that the filtering action of the tympanum and its surroundings had been excluded should be investigated. Similar experiments on the locust ear show that at [east in this preparation the assumption is not correct (see below). Therefore, it is possible that some of the minima observed by Adams are in fact caused by the tympanum. It is interesting that-apart from the three minima-the threshold curve follows a line of constant velocity (i.e. it drops 20 dB per decade when plotted in a frequency-displacement diagram, cf. equation 3). This behaviour is different from that of the locust ear (see below). The average threshold around 60 kHz corresponds t o a peak-to-peak vibration amplitude of 1 A . In one experiment the threshold was about 0.1 A . These values are somewhat smaller than those estimated for the locust ear (see below), but different criteria of threshold were used in these experiments.

5.5.2 Locusts The vibrations of the tympanal membrane in locusts have been described above (section 5.4). The excitation of the receptor cells is roughly proportional t o the vibration of their attachment area. However, the anatomy of the Miiller-organ (i.e. the receptor cells and the cuticular bodies to which they attach: see Fig. 21) is rather complicated. In fact, it seems to be much more complicated than necessary for the simple task of picking up the vibrations of the attachment areas of the receptor cells. Recent experiments suggest that the receptor cells are not only sensitive t o vibrations perpendicular to the membrane. In addition, they seem to pick up the vibrations in the plane of the tympanum (Michelsen, 1973b). In some experiments the ear was stimulated with vibrations from a mechanical stimulator. The threshold curve determined from the nervous response showed a number of the resonances of the tympanal membrane. This was also the case when the mechanical stimuli were delivered to the attachment areas of the receptor cells (cf. the noctuid moths). In order to eliminate the resonances, it was necessary to cut away most of the tympanum so that the preparation consisted of .the Miiller-organ surrounded by very little tympanal membrane; this preparation had a threshold curve which followed a line of constant acceleration down to a

288

AXEL MICHELSEN AND HARALD NOCKE

displacement amplitude of 5-10 A; the threshold curve then followed a line of constant displacement (see Michelsen, 1973b). The mechanisms responsible for this curious behaviour are not understood. However, the experiments indicate that the Miiller-organ is sensitive not only t o the vibrations perpendicular t o the membrane, but also to vibrations in the plane of the tympanum. This idea explains why the Miiller-organ, situated near the border of the membrane (see Fig. 21), can detect vibrations far away on the membrane. It also seems to explain most of the anatomical peculiarities of the Mulleragan. Thus the present working hypothesis on the function of the Miiller-organ is as follows: the organ responds t o the vibrations perpendicular t o the membrane, which occur in its neighbourhood. But in addition it also picks up the vibrations in different areas of the tympanum by responding to t h e variations in tensile strain in the tympanum. This behaviour is similar t o that performed by a spider sitting at the border of its web, and it may turn out t o be a sort of vectorial analysis. Furthermore, recent experiments with sound stimulation and recording of vibrations by means of the very sensitive laser-doppler-method show that the Miiller-organ is not merely a passive observer of the vibrations. It goes into vigorous vibrations oriented both perpendicular and parallel t o the plane of the membrane (see Michelsen, 1973b). A detailed study of these vibrations is now being made.

5.6 SOME ATYPICAL INSECT

EARS

5.6.1 The lacewing ear In the radial vein of the forewings of the green lacewing (Chrysopa, Neuroptera) a most remarkable tympanal organ is found. A swelling in the vein (Fig. 27) is bounded laterally and dorsally by thick cuticle and ventrally by thin membranous cuticle (Miller, 1970). About 25 scolopale units are attached in two groups t o the tympanum. In contrast to typical insect tympanal organs the trachea running through the swelling is unexpended, so the tympanal membrane lacks the tracheal component, and the swelling is largely filled with fluid. The tympanum is rippled. This organ responds t o the high-frequency cries of hunting bats. The ear lacks a secondary tympanic membrane and is probably a pressure receiver (Miller, 1971). It is not yet known how the fluid affects the vibration of the tympanum. If the tympanum alone vibrates, then the fluid is likely to cause considerable damping of the vibrations. An alternative possibility is that the mass of the membrane plus the fluid works against the elasticity of the air inside the tracheal tube. Calculations based upon this model give a resonance frequency of the correct order of magnitude (Miller and Michelsen, unpublished), but there are not enough physical data available to prove the hypothesis.

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION I N INSECTS

289

Unit In dirtol organ

Fig. 27. The lacewing ear. Ac, attachment cell. Den, dendrite. Sc, scolopale cells. SR, scolopale rod. TM-Rp and TM-Sm, rippled and smooth portion of the tympanic membrane. Tr, trachea. (From Miller, 1970.)

5.6.2 The pilifers of hawkmoths Some hawkmoths (Choerocampinae) carry enlxged labial palps, which are lightly poised and buoyant (specific gravity 0.18), since their first and second segments are almost entirely filled with air sacs (Roeder and Treat, 1970). The labral pilifers are nested in a depression of the palpal wall (Fig. 28). This arrangement, composed of appendages of two cranial segments, serves t o detect the cries of bats wiihin the range of 20-60 kHz. The receptors are situated on the pilifers, and. experiments show that the palps act as impedance matchers for the reception of ultrasound (Roeder et al., 1970): the medial surface of the second palpal segment seems to match the impedance of the air to that of the denser tissue fluids of the receptor mechanism in the manner of the vertebrate ear drum and middle ear. When the palp is abducted so as to break the contact with the distal lobe of the pilifer, the acoustic sensitivity drops about 40 dB. The acoustic sensitivity is restored when the system is reassembled, and it may also be partially restored by replacing the palpal contact with the distal lobe by the contact of a thin artificial membrane. The frequency range of the receptor is much broader when the distal lobe is vibrated mechanically than when the entire organ is excited acoustically (Roecler, 1972). Thus, apparently the physical characteristics of the palp limit the frequency response of the auditory organ. This is similar to the tympana1 membrane and receptor

290

AXEL MICHELSEN AND HARALD NOCKE PA

PI

PR

Fig. 28. Ear of a hawkmoth. Left labial palp (PA) deflected laterally to expose the pilifer (PI). The right palp transected. PR, proboscis. (From Roeder et al., 1970.)

organ of locusts (see above), but in hawkmoths the interface portion of the acoustic receptor can be reversibly disconnected from the transducer portion. In fact, the disconnection and reassembling of the acoustic receptor seems to be a natural movement caused by certain muscles (Roeder and Treat, 1970). This action must serve temporarily t o deafen the moth, but its adaptive significance is unknown. 5.6.3 Hairs and antennae In most insect ears the effective parameter of sound can only be estimated after detailed investigations. However, the antennae and hairs responding to sound are obviously movement receivers. The Johnstone organ of mosquitoes appears to be the most sensitive insect ear: the threshold of the summated response is around 0 dB in the male Anopheles (Tischner, 1953), but its physiology has been very little studied. It contains several thousand receptor cells which attach in two layers t o the second antennal segment and terminate at the base of the third. The rest of the antenna forms a flagellum covered with numerous hairs, which appear t o be essential for sound reception. The trains of nerve impulses are phase locked to the sine wave stimulus. The summated response thus contains the frequency of the stimulus, but it also contains harmonic components. The harmonic distortion is at a minimum when the direction of the sound wave is parallel t o the antennal flagellum. Vibrations in other directions cause harmonic distortions of up to 250 per cent in the nervous response. Thus, both the anatomical

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION IN INSECTS

291

arrangement of the receptor cells and the electrophysiological findings suggest that a complicated spatial vector analysis is performed in the organ; such an analysis may provide the animal with directional information (Tischner, 1953; Keppler, 1958). The entire. receptor organ is so complicated, however, that recordings of summai.ed responses are unlikely t o allow a complete analysis of its function. Variations in the harmonic distortion of a summated response only indicate that the synchronization of the nervous activity of the individual receptor cells has changed. Single cell studies should be performed and the biophysics should be investigated by modern methods. In cyclorrhaphous flies the third segment carries a feathery arista pointing sideways from an apical joint. The arista can be moved by sound waves or a flow of air. The perception of air laments during flight is very complicated (see Gewecke and Schlegel, 1970). In Drosophila the antennal arista can be set in resonant (?) vibration by sounds of 180-220 Hz (Bennet-Clark and Ewing, 1967). Sound waves travelling perpendicular to the plane of the arista and about 30" away from the sagittal axis cause a maximum amplitude of vibration (Bennet-Clark, 19 71). The antennal arista of Drosophila is used to record the air flow produced by wing movements of the male fly. During courtship he stands less than 5 mm away from the female, and in this extreme near field quite considerable particle velocities are produced by a wing upstroke: the particle velocity corresponds to that found in a far field of 75 dB (Bennet-Clark, 1971). The sound pressure level, however, is only about 39 dB. Obviously, the use of movement or pressure gradient receptors instead of pressure receptors is an advantage in near-field conditions. Most arthropods have numerous mechanoreceptive hairs on their surface. Many of these hairs respond t o intense sounds, but there is not much evidence for their use in sound communicaiion (see Michelsen, 1973a). Nevertheless, the possibility should always be borne in mind when responses t o sound are investigated in art!hropods. Almost all sound spectrograms of arthropod sounds have been made by means of pressure microphones. Local, low frequency air flows may not produce much audible sound at larger distances, but the magnitude of the local particle velocity may be fairly large. All close distance courtship songs should, therefore, be reinvestigated by means of pressure gradient (band) microphones. The mechanoreceptive hairs may tlhen, after all, turn out t o function as near-field sound receptors. Here again, the biophysics should be studied in detail.

6 Conclusions Throughout this paper we have tried to point out a number of problems which ought t o be studied in further detail. Some areas (for instance, the

292

AXEL MICHELSEN AND HARALD NOCKE

biophysics of hair receptors or the propagation of insect sounds) have not been studied at all, and most problems have only been investigated in a single species. Both the sound-producing mechanisms and the soundreceiving structures have evolved independently in a number of insects. We have already argued that the properties of a few systems are unlikely to be true for insects in general. The conclusions presented here are the synthesis of our limited knowledge about a few systems. Time will show how representative they are. The central question in the biophysics of sound communication is: How much freedom do the animals have in choosing the parameters of sound carrying the specific information? The information carried by sound may be coded as frequency, intensity, and time. The use of these parameters for encoding information is limited both by the neural machinery of the animal and by several physical factors. We shall mainly discuss the physical and acoustical limitations to the coding of specific information. The main problem of bioengineering in insect sound communication is that of matching the impedances of the sound transmitter, the medium, and the sound receptor. This problem is similar to the impedance matching of electrical equipment: for the transmission to be efficient, the output impedance of the sender should be low compared with the input impedance of the receiver. In sound emission the mechanical impedance of the transmitter (the vibrating parts of the insect) should be low compared with that of the air (the radiation impedance). In sound reception the impedance of the air should be low compared with that of the sound-receiving structure. The frequency range of insect sounds is limited mainly by the impedance problem because the radiation impedance depends very much upon the size of the object relative to the wavelength of sound. At low frequencies the sound radiation from small objects is inefficient because the radiation impedance of the sound radiator is small. The sound reception is also inefficient because the impedance of the sound receiver is large at frequencies much below the basic resonance; the amplitude of vibration will therefore be small. Thus, both sound emission and hearing are inefficient at low frequencies. At high frequencies, on the other hand, the sound receiver will be “damped” by its radiation impedance and the sound reception will be inefficient. The transmission properties of sound in air and the absorption by vegetation also tend t o limit the use of high frequencies, especially at large distances. Apparently, the useful band of sound frequency is very limited. Insects with high population densities might increase their frequency range by making their sound transmitters large and their ears small. However, in the former case the muscles must be able t o supply enough energy to set the large transmitter into vibration. In the latter case

BIOPHYSICAL ASPECTS OF SOUND COMMUNICATION IN INSECTS

293

the energy received (and the sensitivity) decreases when the area of the sound receiver becomes smaller. Thus, the frequency range used for communication is determined by simple physics. Most insects use a mechanism of frequency multiplication (e.g. stridulation) in order to communicate in the kHz range (section 3.2). Furthermore, many insects increase their acoustic output by using a resonant structure as sound radiator. However, the insects pay for this gain in acoustic output by giving up much of their control of the frequency and time parameters: lightly damped resonance systems are sharply tuned at certain frequencies (Fig. 7(b)). Therefore the amplitude of vibration-and thus the acoustic output-will decrease considerably if the singing insect does not maintain the resonance frequency. Additionally, lightly damped systems have long time constants (Fig. 7(c)), and the insect is therefore not able to produce quick changes in the sound intensity. In the sound receivers a low degree of damping will cause a high sensitivity and a good frequency selectivity. On the other hand, this results in a poor degree of time resolution. Here again, the insects have t o choose between the three parameters: intensity, frequency and time. Apparently, most insects select good time resolution of their hearing organs. Studies carried out on the biophysics and physiology of a few insect ears have demonstrated that the hearing organs of insects have very different properties. The forces acting on various ears may be caused by different parameters of sound waves: the pressure, pressure gradient, and movement of air particles. Also, the frequency responsr. of the receptor cells may be determined by the velocity, the acceleration, or displacement parameter of the movement. Furthermore, insect ears are influenced by their surroundings to varying extents. These preliminary observations all point t o the conclusion that “insect ears” differ as much am do their evolutionary origin. The common attempts t o generalize about the “properties of the insect ear” are therefore misleading. Probably, the hearing organs of insects vary more than, for example, the lateral organs o f fish differ from the ears of humans.

References Adams, W. B. (1970). Receptor characteristics in the tympanic organ of the noctuid moth. Special Report No. 8. Laboratory of Sensory Communication, Syracuse University. Adams, W. B. (1971). Intensity characteristics of the noctuid acoustic receptor. 1.gem Physiol. 5 8 , 562-579. Autrum, H. (1963). Anatomy and physiology of sound receptors in invertebrates. In “Acoustic Behaviour of Animals” (Ed. R. G . Busriel). Elscvier, London. Autrum, H. (1940). Das Richtungshoren von Locusta und Versuch einer Hortheorie fiir Tympanalorgane der Locustidentyp. 2. uergl. Physiol. 28, 326-352.

294

AXEL MICHELSEN AND HARALD NOCKE

Aylor, D. (1971). Noise reduction by vegetation and ground. J. acoust. SOC. A m . 51, 19 7-205. Bailey, W. J. (1970). The mechanics of stridulation in bush crickets. I. The tegminal generator. J. exp. Biol. 52, 495-505. Bailey, W. J. and Broughton, W. B. (1970). The mechanics of stridulation in bush crickets. 11. Conditions for resonance in the tegminal generator. J. exp. Biol. 52, 507-517. Barth, F. G. (1967). Ein einzelnes Spaltsinnesorgan auf dem Spinnentarsus: seine Erregung in Abhangigkeit von den Parametern des Luftschallreizes. Z . vergl. Physiol. 55,407-449. Bennet-Clark, H. C. (1970). The mechanism and efficiency of sound production in mole crickets. J. exp. Biol. 52, 619-652. Bennet-Clark, H. C. (1971). Acoustics of insect song. Nature, 234, 255-259. Bennet-Clark, ‘H. C. and Ewing, A. W. (1967). Stimuli provided by courtship of male Drosophila melanogaster. Nature, 215, 669-67 1. Beranek, L. L. (1954). “Acoustics”. McGraw-Hill, New York. Busnel, R.-G. (1963). “Acoustic Behaviour of Animals”. Elsevier, Amsterdam. Davis, W. J., Jr. (1968). Cricket wing movements during stridulation. Anim. Behav. 16, 72-73. Dubrovin, N. N. and Zhantiev, R. D. (1970). Acoustic signals of katydids (Orthoptera, Tettigoniidae). (In Russian.) Zool. J. 49, 1001-1014. Dumortier, B. (1963). Ethological and physiological study of sound emissions in Arthropoda. In “Acoustic Behaviour of Animals” (Ed. R.-G. Busnel), pp. 583-654. Elsevier, Amsterdam. Evans, E. J. and Bazley, E. N. (1956). The absorption of sound in air at audio frequencies. Acoustica, 6 , 238-245. Gewecke, M. and Schlegel, P. (1970). Die Schwingungen der Antenne und ihre Bedeutung fiir die Fhgsteuerung bei Calliphora erythrocephala Z . vergl. Physiol. 67, 325-362. Harris, C. M. (1967). Absorption of sound in air versus humidity and temperature. NASA, CR-647.Washington, DC. Huber, F. (1962). Central nervous control of sound production in crickets and some speculations o n its evolution. Evolution, 16, 429442. Huber, F. and Elmer, E. N. (1973). Neurale Grundlagen der akustischen Kommunikation bei den Orthopteren. Fortschr. Zool. (In press.) Johnstone, B. M., Saunders, J. C. and Johnstone, J. R. (1970). Tympanic membrane response in the cricket. Nature, 227, 625-626. Keppler, E. (1958). Uber d a s Richtungshoren von Stechmiicken. Z . Naturf. 13b, 280-284. Leston, D. and Pringle, J. W. S. (1963). Acoustical behaviour of Hemiptera. In “Acoustic Behaviour of Animals”. Elsevier, Amsterdam. Lewis, D. B. (1973). The prothoracic tracheal system and sound reception in Homorocoryphus dtidulus vicinus. (In press.) Lowry, W. P. (1970). “Weather and Life”. Academic Press, London and New York. Markl, H. (1965). Stridulation in leaf-cutting ants. Science, 149, 1392-1393. Markl, H. (1967). Die Verstandigung durch Stridulationssignale bei Blattschneiderameisen. I. Die biologische Bedeutung der Stridulation. Z . vergl. Physiol. 57, 299-330.

BIOPHYSICAL ASPECTS OF SOUND COMMUNlCAlrlON I N INSECTS

295

Markl, H. (1968). Die Verstandigung durch Stridulationssignale bei Blattschneiderameisen. 11. Erzeugung und Eigenschaften der Signale. Z. vergl. Physiol. 60, 103-150. Markl, H. (1971). Signale der Blattsneiderameisen. Bild der Wissenschaft, 984991. Markl, H. (1972). Neue Entwicklungen in der Bioakustik der wirbellosen Tiere. J . Om 113, 91-104. Michelsen, A. (1966a). Pitch discrimination in the locust ear: observations on single sense cells. J. Insect Physiol. 12, 1119-1131. Michelsen, A. (1966b). On the evolution of tactile stiniulatory actions in long-horned beetles (Cerambycidae, Coleoptera). Z . Tierpsychol. 23, 257-266. Michelsen, A. (1971a). The physiology of the locust ear. I. Frequency sensitivity of single cells in the isolated ear. Z . vergl. Physiol. 71, 49-62. Michelsen, A. (1971b). The physiology of the locust ear. 11. Frequency discrimination based upon resonances in the tympanum. Z . vergl. Physiol. 71, 63-101. Michelsen, A. ( 1 9 7 1 ~ ) The . physiology of the locust eiu. 111. Acoustical properties of the intact ear. Z. vergl. Physiol. 71, 102-128. Michelsen, A. (1973a). Hearing in invertebrates. In “Handbook of Sensory Physiology”, Vol. V, “Auditory Systems”. Springer-Verlag, Berlin. Michelsen, A. (1973b). An invertebrate frequency analyzer: the locust ear. In “Basic Mechanisms in Hearing” (Ed. A. Moller). Academic Press, New York and London. Miller, L. A. (1970). Structure of the green lacewing tympana1 organ (Chrysopa carnea, Neuroptera). J. Morphol. 131, 359-382. Miller, L. A. (1971). Physiological responses of green lacewings (Chrysopa, Neuroptera) to ultrasound. J. Insect Physiol. 17,491-506. Moore, T. E. and Sawyer, R. T. (1966). The mechanisri of cicada timbal action. A m . ZooL 6,509. Morse, P. M. (1948). “Vibration and Sound”. McGraw-Hill, New York. Murphey, R. K. and Zaretsky, M. D. (1972). Orientation t o calling song by female crickets. Scapsipedus marginatus (Gryllidae). J. exp. Biol. 56, 335-352. Nocke, H. (1971). Biophysik der Schallerzeugung durch die Vorderfliigel der Grillen. Z. vergl. Physiol. 74, 272-314. Nocke, H. (1972). Physiological aspects of sound communication in crickets (Gryllus campestris L.). J . comp. Physiol. 80, 141-162. Olson, H. F. (1957). Elements in acoustical engineering. Nostrand, New York. Pierce, G. W. (1948). The songs of insects. Harvard University Press, Cambridge, Mass. Pringle, J. W. S. (1954). A physiological analysis of ‘cicada song. J. exp. Biol. 31, 525-560. Pumphrey, R. J. (1940). Hearing in insects. Biol. Rev. 15, 107-132. Reichardt, W. (1968). Grundlagen der technischen Akustik. Akademische Verlagsgesellschaft Geest und Portig K.G., Leipzig. Reid, K. H. (1971). Periodical cicada: mechanism of sound production. Science, 172, 949-951. Roeder, K. D. (1972). Acoustic and mechanical sensitivity of the distal lobe of the pilifer in choerocampine hawkmoths. J. Insect Physiol. 18, 1249-1264. Roeder, K. D. and Treat, A. E. (1970). An acoustic sense in some hawkmoths (Choerocampinae). J. Insect Physiol. 16, 1069-1086. Roeder, K. D., Treat, A. E. and Berg, J. S. V. (1970). Distal lobe of the pilifer: an ultrasonic receptor in choerocampine hawkmoths. Science. 170, 1098-1099.

296

AXEL MICHELSEN AND HARALD NOCKE

Sivian, L. J. (1947). High frequency absorption in air and other gases. /. ucoust. SOC. A m . 19,914916. Shorey, H. H. (1962). Nature of the sound produced by Drosophilu melunoguster during courtship. Science, 137, 677-678. Skudrzyk, E. (1954). Die Grundlagen der Akustik. Springer-Verlag. Wien. Spieth, H. T. (1952). Mating behaviour within the genus Drosophila (Diptera). BulL Am. Mus. nut. Hist. 99, 401-474. Suga, N. (1966). Ultrasonic production and its reception in some neotropical tettigoniidae. /. Insect Physiol. 12, 1039-1050. Tischner, H. (1953). ijber den Gehorsinn von Stechmiicken. Acousticn, 3, 335-343. Young, D. (1972). Neuromuscular mechanism of sound production in Australian cicadas. J. comp. Physiol. 79, 343-362. Zwislocki, J. J., Adams, W. B., Barlow, R. B. and Kletsky, E. J. (1969). Intensity characteristics of sensory receptors. Abstr. 3rd Int. Biophys. Congr., p. 262, Cambridge, Mass.

Hormonal Mechanisms Underlying Insect Behaviour James W. Truman and Lynn M. Rialdiford Department of Zoology University of Washington Seattle, Washington. USA

1 Introduction 2 The insect endocrine system 3 The effects of hormones on behaviour . 3.1 Modifier effects 3.2 Releaser effects 4 The mode of action of hormones on behaviour . 5 Neurophysiological studies of hormone action 5.1 Hormonal effects on the central nervous system 5.2 Peripheral action of hormones: inhibition of firefly flashing 6 Role of hormones in the insect life history 6.1 Hormonal mechanisms underlying larval behaviour . 6.2 Activation of adult behaviour 6.3 Hormones and reproductive behaviour 6.4 Hormonal influences on migration and orientation . 6.5 Hormones and insect circadian rhythms 7 Conclusions . .

.

.

Acknowledgements References

.

. . . . . . . .

.

. . . . . . .

. .

297 298 300 301 303 303 305 305 309 311 311 314 316 333 337 340 342 342

1 Introduction The behavioural responses of an animal to its environment can be drastically modified by changes in its hormon,al milieu. Hormonal effects on behaviour have been long established in vertebrates and have been studied primarily with respect to the effects of hormones on sexual behaviour (Davidson and Levine, 1972). In insects the first demonstration of a hormonal influence on behaviour was in 1938 when Bowhiol reported that the removakof the corpora allata from the penultimate larval instar of Bombyx mori led to precocious pupation and, prior to this, the premature construction of a miniature 297

298

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

cocoon. But, during this time studies in insect endocrinology dealt primarily with effects on growth and metamorphosis. Therefore, investigations of the effects of hormones on behaviour were sparse until 1960,when studies in this area surged. Since that time interest has remained high, and today there is solid evidence for hormonal involvement in many diverse aspects of insect behaviour. As with vertebrates, the reproductive behaviour of many insects is hormonally controlled. Besides reproduction, endocrine influences have been extended to other types of insect behaviour ranging from simple photo- and geotaxis to the extremely complex behaviour involved in cocoon spinning. Indeed, insects use hormonal mechanisms in behaviour to a greater extent than do any other invertebrate group. Such a high degree of interaction between the endocrine and nervous systems most probably allows more information to be packaged in a nervous system of limited size. The extent of this relationship would therefore be an important factor in the development of the rich behavioural repertoires which are seen among insects. This paper is concerned with actions of the endocrine system on the nervous system which lead t o overt behavioural acts. In his review on hormones and insect behaviour, Highnam (1964) included hormonal effects on pigmentation because proper colouration is necessary for the effectiveness of cryptic behaviour. Since these effects do not represent hormone action on the nervous system, we have not included them in this review. Fuzeau-Braesch (1972) has recently reviewed the literature on colour changes in insects and includes a treatment of the hormonal control of these changes. We have also restricted ourselves to examples in which the behavioural changes cannot be attributed directly to a developmental modification of the nervous system by the addition of new neurons, etc. Therefore, the behavioural changes which eventually result from the ill not be hormonally induced metamorphosis' of the nervous system w examined. A detailed consideration of caste determination in the social insects and the associated behavioural changes is beyond the scope of this paper for the same reason. 2 The insect endocrine system The insect endocrine system is comprised of neuroendocrine and classically endocrine components (Wigglesworth, 1970). The insect brain serves to coordinate this system with sensory information from the environment. The brain contains two major paired clusters of neurosecretory cells-the median and the lateral neurosecretory cells. A small third group, the posterior neurosecretory cells, are sometimes also found in the protocerebrum (Herman and Gilbert, 1965). The major neurosecretory groups

HORMONES AND BEHAVIOUR

299

send axons that terminate in the paired corpora cardiaca which lie posterior to the brain. The corpora cardiaca (CC), besides functioning as neurohaemal organs for these cells, also contain a complement of intrinsic neurosecretory cells. Manifestly, the brain-corpora cardiaca complex of insects is roughly analogous to the hypothalamic-neurohypophysial complex of vertebrates (Scharrer and Scharrer, 1963). Through this system a number of neuroendocrine agents are released including those which regulate prothoracic gland activity, cuticular tanning, adult blood sugar levels, cardiac activity, water balance and various aspects of behaviour. Neurosecretory cells are also found in inost of the thoracic and abdominal ganglia. These cells release their products through unpaired, segmentally arranged, neurohaemal organs-the perivisceral organs (DupontRaabe, 1966; Brady and Maddrell, 1967). The prothoracic glands (ecdysial glands, ventral glands) lie in the anterior part of the thorax. In response t o a tropic hormone from the brain, these glands presumably secrete ecdysone which is responsible for the initiation of the moulting cycle in arthropods. The direction of the moult-whether it be to another larval stage, to the pupa, or to the adult-is determined by the titre of juvenile hormone (JH). This hormone is secreted by the corpora alkita (CA), the paired glands which are attached t o the corpora cardiaca. The juvenile hormone also functions in the adult stage of many insects to promote oogenesis (Engelmann, 1970). In many instances, hormones which serve diverse physiological or developmental functions have been captured for use by the nervous system. Consequently, a close coupling between physiology and behaviour is established. One example of this coupling is the action of ecdysone on the behaviour of Schistocerca gregurk larvae (Haskell and Moorhouse, 1963). In this locust ecdysone acts on the nervons system to reduce motor output. This apparently results in the relative quiescence observed during the vulnerable periods of ecdysone-induced apolysis and cuticle deposition. Another example involves the effects of the declining blood titre of JH in mature larval insects. The low JH concentraticm allows metamorphosis to occur and also promotes certain types of premetamorphic behaviour which then bring the larva into an environment suitable for pupation. Perhaps the best illustration of the capture of a hormone f i x a supporting behavioural role involves the effects of JH in the adult female. In many species this hormone directs not only ovarian development, but also the onset of sexual receptivity which will eventually result in the fertilization of the mature

eggs. In contrast to the cases illustrated above, certain neurosecretory hormones appear to have a solely behavioural function. Significant quantities of the eclosion hormone can be extracted from Manduca sexta

300

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

brains and corpora cardiaca prior to adult eclosion and not at any other time in the life history of the insect (Truman, 1973). Similarly, there are hormones such as those which elicit the “calling” posture (Riddiford and Williams, 1971) or stimulate oviposition in silkmoths (Mokia, 1941; Truman and Riddiford, 1971; Riddiford and Ashenhurst, 1973) which can be associated with no known physiological changes. These hormones have apparently been evolved for the sole purpose of long-distance communication within the nervous system. The claim that there are hormones which serve only behavioural functions must be made with caution because of the scant information on the chemistry of insect neurohormones. It is possible that some behavioural hormones may serve two or more unrelated functions at different times in the insect life history. For example, Gersch and Richter (1963) and Unger (1965) have shown that the cockroach phallic nerve-stimulating hormone (Milburn et al., 1960) is identical with Neurohormone D , , a prothoracicotropic hormone (Gersch, 1962). Besides the neuroendocrine system and the classical endocrine glands, other organs show secondary endocrine functions which have behavioural implications. In response to insemination, the oviposition behaviour of the female is modified due to humoral substances released from the spermatheca in Rhodnius prolixus (Davey, 1965) or from the bursa copulatrix in Hyalophora cecropia (Riddiford and Ashenhurst, 1973). The testis of the firefly beetle, Luciola lusitanica, has been implicated in releasing noradrenaline which inhibits the flashing of the light organ (Brunelli et al., 1968a, 1968b, 1970; Bagnoli et al., 1970, 1972). In mosquitoes the male accessory glands produce matrone (Craig, 1967), a substance which produces unreceptive behaviour in mated females. Sensu strictu, this last agent should be categorized as a pheromone; but since it functions even when injected into the haemocoel of females (Craig, 1967), it is included here. Leopold et al. (1971a) have recently demonstrated that in Musca a similar agent moves from the reproductive tract of mated females into the blood and gradually accumulates in the head (see below). Thus, these agents act as hormones.

3 The effects of hormones on behaviour The effects of hormones on insect behaviour are somewhat analogous to the effects of pheromones on behaviour. The latter have been classified by Wilson and Bossert (1963) into two types: releasers and primers. In a hormonal context, a releaser effect denotes a relatively rapid behavioural response which is directly triggered by the hormone (Fig. 1). Additional sensory stimuli are not necessary for the behaviour to occur. This action is similar to that of a releaser pheromone.

HORMONES AND BEHAVIOUR

Releaser hormone

301

-

IMotor response

Stimulus

f

Motor

response I

t

Motor

response 2

Fig. 1 . Schematic representation of the releaser and the modifier effects of hormones on the insect nervous system.

A primer effect of a hormone is slower to appear and, by some unknown mechanism, results in a change in responsiveness of the nervous system. Consequently, before exposure of the animal to the hormone, a given stimulus provokes one behavioural response, whereas after exposure the same stimulus triggers a different response (Fig. 1). This action of hormones differs from that of a primer pheromone in that the pheromone usually works through a neuroendocrine or endocrine relay (Wilson and Bossert, 1963). By contrast, the primer effect 3f a hormone is possibly a direct action on the nervous system; and often such a hormone provides the link between the pheromone and the behaviour. Therefore, in order to avoid confusion as to possible modes of action, we will refer to this type of hormone effect as a modifier effect. As we will describe below, some hormones have modifier effects, others show releaser effects, and still others have both actions. 3.1

MODIFIER EFFECTS

In most insects the individual is capable of a variety of behavioural options when faced with a given stimulus. It can attack, flee, attempt to copulate, ignore the stimulus, etc. Consequently, one respcinse must be promoted and the others repressed, depending upon the “mood” of the insect. Hormones are often involved in the determination of “mood” and, indeed, this is the most common behavioural effect of hormones. As humoral agents which have access to all cells, they are well suited to effect t o k c alterations in the responsiveness of the nervous system. A classic example of a hormoneinduced alteration in behaviour is illustrated by the effect of JH on the

302

JAMES W. TRUMAN AND LYNN M. RIDDIFORD

response of a female Gomphocerus rufus to a courting male (Loher and Huber, 19 66). After the imaginal ecdysis, the female grasshopper displays primary defence-the approaches of males are met with kicking movements of the hind legs and escape reactions. After exposure to JH, primary defence disappears. Now the female faced with the same stimulus of a courting male stridulates in answer to the male’s call and allows him to mount and copulate (Loher and Huber, 1964, 1966; Loher, 1966). The eclosion hormone has a similar effect on the mating response of the silkmoth, Antheraea pernyi. In these moths the male is attracted to the female by a volatile sex pheromone that is perceived by sensory cells in the sensilla trichodea which cover the antennae of the male moth. These receptors are finely tuned t o respond to the emanations of conspecific females. The irrelevant odours of most heterospecific females are, therefore, effectively filtered out at the receptor level (Priesner, 1968; Schneider, 1969). In the case of the Pernyi moth, males invariably mate with virgin females within 15 minutes of being placed with them in a cage under dim illumination (Riddiford, 1970). But this response to the sex pheromone does not occur if the male has not been exposed to the eclosion hormone. When the pupal cuticle was removed from a pharate moth which had finished adult development but had not yet emerged, the animal showed no behavioural response t o the sex pheromone, even under optimal conditions. This behavioural block was not in the ability to detect pheromone, since the antennae from pharate male moths gave normal electroantennogram responses to the female pheromone (Riddiford, unpublished). Instead, this block must reside centrally. Whatever the nature of the block, it is rapidly removed when adult behaviour is “turned on” by the eclosion hormone (section 6.2). Thus, in the absence of the proper hormonal modification, the CNS of the pharate moth does not respond to the sensory input caused by the conspecific pheromone. As a consequence, motor patterns which are irrelevant to the physiological state of the insect are not activated. In vertebrates hormone-induced changes in behaviour are usually transient and depend upon the contemporaneous presence of the hormone. For example, estrous behaviour in many female mammals declines after circulating estrogen has dropped below a critical level (Lisk, 1967; Komisaruk, 1971). In insects the need for the contemporaneous presence of hormone for the maintenance of a particular behavioural state has also been demonstrated. In Gomphocerus removal of the corpora allata from mature receptive females leads to the disappearance of sexual receptivity and the reappearance of defensive behaviour (Loher, 1962, 1966). In other cases, the hormonally-induced changes appear to be permanent and persist long after the hormone has presumably vanished from the insect. Matrone acts on mated females of Aedes aegypti to render them

HORMONES AND BEHAVIOUR

303

refractory to mating for the remainder of their lives which may last for weeks or months (Craig, 1967). Similarly, the eclosion hormone “turns on” adult behaviour at the time of eclosion (‘Truman, 1971a). It then presumably disappears from the blood but adult behaviour persists. 3.2

RELEASER EFFECTS

Some hormones serve to trigger directly a specific piece of behaviour. This action can be seen in the capacity of the phallic nerve stimulating hormone to release copulatory movements in cockroaches (Milburn et al., 1960). Similarly, the hormonal control of “calling” behaviour in silkmoths (Riddiford and Williams, 197 1 ) is presumably mediated through a releaser effect. The eclosion hormone also acts as a releaser because it directly triggers the performance of the pre-eclosion behaviour (Truman, 1971a; Truman and Sokolove, 1972). 4 The mode of action of hormones on behaviour Thirty-five years ago, Lashley (1938) proposed four hypotheses to account for the effects that hormones had on sexual behaviour in rats. 1. Hormones might stimulate the growth or formation of new nervous connections. 2. The hormone could act by merely increasing the general excitability of the organism. 3. The hormone might induce specific changes in various peripheral organs. This altered condition would then initiate sensory impulses which would facilitate the performance of the behaviour. 4. The hormone could act upon the central nervous system to increase the excitability of the sensori-motor mechmisms specifically involved in the activity. He then concluded that the fourth alternative best accounted for the sexual behaviour of the rat. Nature has not clung to only one type of hormonal action, however. There are now examples which fall into each (of the four categories (see Hinde, 1969). Among insects, the action of ecdysone and JH in promoting metamorphosis is an extreme example of the first alternative. A more subtle example involves the hormonally controlled changes in the nervous system and in behaviour which are observed during the process of caste formation in termites (Liischer, 1962). The second type of action, that of inducing a specific piece of behaviour through a general increase in excitability, is technically difficult to demonstrate. One example which may fall into this category is the report

304

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

by Ozbas and Hodgson (1958) that stereotyped walking movements were induced in cockroaches after injections of corpora cardiaca homogenates. These extracts also served t o alter the spontaneous rate of interneuron discharge in the nerve cord. Most cases of hormone-mediated behaviour cannot be attributed to the third mechanism. For example, in Gomphocerus rufus the JH-induced onset of receptive behaviour does not depend upon stimuli from the developing ovaries since gonadectomy will not interfere with the normal appearance of this behaviour (Loher and Huber, 1964). Similarly, ovariectomized females of the moth, Portheria dispar, will nevertheless perform normal oviposition behaviour (Klatt, 1920). The most extreme example is seen after application of JH mimics to last instar larvae of the bug, Pyrrhocoris apterus. Zdarek and Slima (1968) reported that 50 per cent of the resulting supernumerary larvae showed sexual behaviour despite the fact that the accessory glands and external genitalia were completely undeveloped. A case in which hormonally induced peripheral changes do influence behaviour is seen in the premetamorphic behaviour of Hyalophora cecropia. In this species the onset of metamorphosis is signalled when the caterpillar ceases to feed, wanders from the food plant, and begins to spin a cocoon. These events are potentiated by a low JH level and perhaps triggered by a neurosecretion from the brain (section 6.1). The construction of the cocoon is accomplished by simple figure-eight movements of the head and by more complex movements of the body (Van der Kloot and Williams, 1953a). Caterpillars whose spinnerets were blocked, nevertheless, performed the spinning movements (Van der Kloot and Williams, 1953b). Thus, the actual spinning of the silk thread is not a prerequisite for the behaviour to occur. By contrast, when the silk glands were removed from early last instar larvae, the caterpillars fed and behaved normally until fully grown, but then did not show wandering behaviour. Spinning behaviour was also omitted (Van der Kloot and Williams, 1953b). From this evidence it appears that low JH titres influence spinning behaviour at least partiolly through the JH-controlled growth of the silk glands. Most probably the hormone titres also have a direct influence on the CNS of the larva. Most modifier effects in insects probably arise through the facilitation of specific sensori-motor pathways. In the case of the mosquito, Aedes aegypti, the pathways involved in female receptivity are confined to the terminal abdominal ganglion (Gwadz, 1972). Newly emerged A . aegypti females are refractory t o the approach of males (Gwadz and Craig, 1968; Lea, 1968), but within two to three days JH promotes the onset of receptivity in the mature virgin (Lea, 1968; Gwadz et a f . , 1971b). After insemination the matrone received from the male permanently switches off receptivity (Craig, 1967). The differences in behaviour between receptive

HORMONES A N D BEHAVIOUR

305

and refractory females are, in large part, confined to subtle changes in the movements of the genitalia of the female (Gwadz et al., 1971a). In response to genital contact with the male, a receptive female extends her cerci, thereby ensuring the male a firm grasp and copulation ensues. A refractory female retracts her cerci; consequently, she denies the male a good grasp and avoids copulation. These important responses to the contact of the male are mediated within the abdominal ganglia. Severance of the ventral nerve cord anterior to the second abdominal ganglion influenced neither the refractory behaviour of young virgins, the JH-induced receptivity of mature virgins, nor the matrone-induced refractoriness of mated females (Gwadz, 19 72). Similarly, sectioning of the nerve cord anterior to the terminal ganglion had no effect on receptivity. The importance of the terminal ganglion was underscored by the effects of removing it and, thus, denervating the genitalia. This operation abolished the differences between receptive and refractory females (Gwadz, 1972). When treated in this manner, approximately one-third of each group of young virgins, mature virgins, and matrone-injected females were inseminated after exposure to males. These experiments indicate that in mosquitoes the sensori-motor pathways which respond to JH and to matrone and which mediate the reaction of the female to contact by the male are located in the terminal abdominal ganglion. The releaser effects of hormones, on the other hand, do not appear to be mediated through sensori-motor mechanisms. Sensory iqput is unnecessary and the presence of the hormone leads directly to motor output as in the following two examples. The phallic nerve-stimulating hormone of cockroaches removes the descending inhibition of the centres which generate the copulatory movements of the phallomeres (Milburn et al., 1960). Similarly, the eclosion hormone acts on the abdominal nerve cord of silkmoths to release the pre-eclosion behaviour (Truman and Sokolove, 1972). 5 Neurophysiological studies of hormone action When one considers the profound and dramatic effects which hormones have on insect behaviour, it is surprising that only a few attempts have been directed towards an electrophysiological investigation of their action. 5.1

HORMONAL EFFECTS ON THE CENTRAL NERVOUS SYSTEM

5.1.1 Changes in levels of spontaneous activity One approach to the study of the effects of hormones on the nervous system has involved the observation of changes in spontaneous activity

306

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

after the addition of various hormonal materials to different ganglionic preparations (Strejckova et al., 1965). Studies of this sort have shown that hormones can alter the level of spontaneous activity in the insect nervous system, but they have been of doubtful value since the experimental results cannot be related t o the behaviour or to the normal physiology of the animal. The earlier work of Ozbas and Hodgson (1958) presented a more useful approach. These authors showed that extracts of cockroach corpora cardiaca markedly inhibited spontaneous activity when applied to the isolated nerve cord. When injected into intact cockroaches, these same agents caused a decrease in coordination but an increase in locomotor activity. High doses of extract promoted complete quiescence. This work of Ozbas and Hodgson is significant because it demonstrated for the first timk the parallel effects of neuroendocrine secretions on both the neurophysiological and the behavioural levels. Also, along the same line in the case of the desert locust, Schistocerca gregaria, Haskell and Moorhouse (1963) demonstrated that the blood from moulting nymphs caused an increase in interneuronal activity but a decrease in spontaneous motor output. This example is considered in detail in section 6.1. 5.1.2 Release of patterned behaviour a. Copulatory movements in male cockroaches. In the praying mantis locomotion and copulatory movements of the male arise after he is decapitated by the female. The onset of this behaviour is due to the removal of inhibitory centres in the suboesophageal ganglion (Roeder, 1935). Similar behaviour was also observed in the cockroach after removal of the head (Roeder et al., 1960). A few minutes after decapitation, the male cockroaches showed an increased leg tonus, a lowering of the abdomen, slight peristaltic movements of the terminal abdominal segments, bending of the abdominal tip, twitching of the cerci, and coordinated movements of the phallomeres. In the nervous system, decapitation provoked bursts of action potentials in the phallic nerves and increased efferent activity from the metathoracic ganglion (Roeder et al., 1960). The neural responses which are observed after decapitation could also be induced by application of corpora cardiaca extracts t o the intact insect (Milburn et al., 1960; Milburn and Roeder, 1962). When these extracts were placed on the nerve cord or injected into the head capsule of semi-dissected cockroaches, these preparations showed the initiation of rhythmic bursting in the phallic nerve. With respect to the magnitude and the time of onset of the response, the effects of the extract were very similar t o that of decapitation. But, unlike decapitation, the response to the hormone was transient and subsided after about 50 min. With additional applications of corpora cardiaca extract, the rhythmic bursting could be

HORMONES AND BEHAVIOUR

307

maintained for longer periods. Thus, the phallic nerve stimulating hormone apparently acts on certain inhibitory centres in the cockroach nervous system which normally repress the motor programs that are responsible for copulation. b. The pre-eclosion behaviour of silkmoths. Another example of a hormonal triggering of a specific piece of behaviour is the action of the eclosion hormone in releasing the pre-eclosion behaviour-a stereotyped series of abdominal movements which immediately precede adult ecdysis. In H. cecropia this behaviour has three distinct phases which differ as to the relative frequency and type of movements involved: (1) a hyperactive period which lasts about 0.5 h and which involves primarily abdominal rotations; (2) a quiescent period of approximately 0.5-h duration and during which the frequency of rotations is greatly reduced; (3) a period of hyperactivity which consists of a series of strong peristaltic waves that move anteriorly along the abdomen. This complex piece of behaviour occurred 10 t o 30 min after a pharate Cecropia moth was injected with extracts containing the eclosion hormone (Truman, 197 la). Moreover, abdomens isolated from pharate moths also showed the full pre-eclosion behaviour after hormone injection (Truman, 1971a). Thus, the information necessary for the performance of the pre-eclc'sion behaviour must be encoded in the abdominal ganglia of the pharate moth. The timing and patterning of movements that are seen during the pre-eclosion behaviour are due to a neural program which is built into the circuitry of the abdominal ganglia (Truman and Sokolove, 1972). This was shown by recording the motor output from the deafferented abdominal nerve cord of Cecropia. Similar results have since been obtained using the isolated chain of abdominal ganglia (Truman, unpublished). In the absence of the eclosion hormone the deafferented nerve cord showed a low level of firing with a few (c. one per hour) spontaneous bursts. By contrast, addition of the eclosion hormone to similar preparations evoked a complex pattern of motor outpul (Fig. 2) from the dorsal nerves (Truman and Sokolove, 1972). (These motor roots supply the longitudinal bands of intersegmental muscle (Libby, 1961), which produce most of the movement seen during the pre-eclosion behaviour.) The onset of bursting began 20 to 40 min after the addition of the hormone. The first set of bursts showed a patterning which would have generated rotational movements of the abdomen if the nerves had been attached t o their corresponding muscles. Each burst was characterized by a strong right-left alternation of volleys, and by identical motor output from the ipsilateral roots of sequential ganglia (Fig. 2). The rotational bursts occurred at a relatively high frequency for the first 0.5 h. During thennext 0.5-h period, the frequency of rotational bursts dramatically declined. Indeed, in a few preparations all spontaneous motor activity ceased. After the quiet period

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

308

frequent bursting resumed, but these new bursts showed a “peristaltic” patterning. Each burst consisted of one major volley which occurred simultaneously in the right and left roots of a given ganglion. These bilateral volleys first occurred in the most posterior ganglion and moved anteriorly in a sequential fashion (Fig. 2). Manifestly, this pattern of efferent discharge would generate the anteriorly directed peristaltic waves observed during the last portion of the pre-eclosion behaviour.

0.5h

5s

1st phase

3 r d phase

Fig. 2. The effect of the eclosion hormone on the motor activity of the deafferented abdominal nerve cord of pharate Cecropia moths. (Top) Integrated efferent activity from the right dorsal nerve of ganglion Az. Hormone was added to the preparation approximately 40 min before the onset of the first burst. (1) The fist hyperactive period; (2) the quiet period; (3) the second hyperactive period. (Bottom) The “fine-structure’’ of the bursts which were typically recorded during the first and the last phases of the response t o the eclosion hormone. Records are representations of the integrated neural activity which was recorded from the deafferented abdominal nerve cord. The letters refer to the electrode placements shown on the diagram to the right.

In short, the peristaltic movements and the rotational movements are generated by central motor programs (as originally described for locust flight by Wilson (1961)). The temporal arrangement of these programs as well as the switch from one motor program to the next are also centrally coded. In response t o the eclosion hormone the entire pre-eclosion program is then activated and “read off”. It is not yet known if the hormone is needed only to trigger the program or whether its presence is required during the entire display.

309

HORMONES AND BEHAVIOUR

5.2

PERIPHERAL ACTION OF HORMONES: INHIBRION OF FIREFLY FLASHING

In contrast to the central action of the insect hormones just described, the flashing of the firefly beetle Luciofa is modulated by a hormonal mechanism which acts peripherally. In fireflies an exposure to light will inhibit the flashing of the beetle. Magni (1967) showed that in Luciofa italicn this photically induced flash inhibition was mediated by two mechanisms: a central one which acted at the level of the neural pacemaker and a peripheral mechanism which acted on the lantern. The involvement of a hormone in this peripheral inhibition was shown by connecting the haemocoels of two beetles (Luciola lusitanica) and then illuminating the eyes of one (Brunelli et al., 1968a, 1968b). This treatment markedly depressed the intensity of spontaneous flashing in the unilluminated partner. When the testes were removed from the stimulated insect, the inhibition seen in the attached partner was abolished (Fig. 3).

I

A

2

0 I

1 I lllllll

1 1

IIIIIIIIIIII IllIllllIII

1

I Ill I

II

IIIIIII ll IIII

IIIll I I I I I I I I IIIIII IIIIIII I

Fig. 3. The effect of illumination of the eyes on the sporltaneous flashing of parabiosed pairs of fireflies. (Top) Set-up used to record the flashing of (1) the sighted firefly and (2) its partner whose eyes were covered with an opaque paste. CRO, oscilloscope; L, light guide; P, photomultiplier; S , saline bridge. (Bottom) Representation of oscilloscope records of spontaneous flashing of beetles 1 and 2 before, during: and after illumination of beetle 1. Illuminated firefly either (A) unoperated or (B) castrated. (Modified from Brunelli et al., 1968a.)

310

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

The role of the testes was then examined using the technique of direct electrical stimulation of the lantern (Bagnoli et al., 1970). When the light organ was stimulated directly, the intensity of the elicited flash was unaffected by either decapitation or denervation of the lantern. But if the eyes of the denervated firefly were illuminated, the intensity of the electrically induced flash was markedly reduced. Inhibition was again abolished by removal of the testes. Innervation of the testis is essential in this inhibition. Transection.of the nerve cord anterior to the ganglion which innervates the testis destroyed the inhibitory response to illumination of the eyes. Consistent with this finding was the result that electrical stimulation of the nerve leading to the testis inhibited the flash elicited by electrical stimulation of the lantern (Bagnoli et al., 1970). The material responsible for flash inhibition is apparently stored in granules found in the cortex of the testis (Brunelli et al., 1970). Histological studies showed a marked depletion of these cortical granules after the insect was exposed to light. Moreover, these granules, when isolated from testes homogenates by centrifugation in a sucrose density gradient, inhibited the induced flashing of the lantern when injected into males. The flash-inhibiting substance was identified as noradrenalin (Bagnoli et al., 1972). This identification was based on the low doses ( 6 x g) of noradrenalin needed to inhibit flashing, the fluorimetric measurements of high levels (75 x 10-I2g per male) of noradrenalin and of the absence of adrenalin in the gonads, and autoradiographic evidence of the uptake of 3H-dopamine into the cortical granules. Although the evidence for the involvement of noradrenalin is compelling, one would also like to know if the peripheral inhibition of flashing could be blocked by pharmacological agents which specifically block the synthesis or action of noradrenalin. The action of noradrenalin must be peripheral since its application inhibits flashing but does not affect the size or frequency of efferent volleys coming to the lantern (Bagnoli et al., 1972). Since the neurotransmitter released from the neuroeffector organ within the lantern also appears to be adrenergic in nature (Carlson, 1969), there are several possible sites where noradrenalin could exert its inhibitory effect. Bagnoli et al. (1972) present two alternative hypotheses to explain its inhibitory action: ( 1 ) noradrenalin may act to block transmitter release at the neuroeffector junction; or (2) it may act somewhere between transmitter release and light production. The authors prefer the former alternative because noradrenalin has been shown to block the giant fibre response to input from the cercal afferents in the cockroach (Hodgson and Wright, 1963). But the firefly inhibition has a different specificity than does the cockroach inhibition; both are sensitive t o noradrenalin but the latter is

31 1

HORMONES AND BEHAVIOUR

also sensitive to adrenalin, whereas the former is not (Bagnoli et al., 1972). Also, noradrenalin interferes with the response of the lantern to direct electrical stimulation (Bagnoli et al., 1972). This effect would not be seen if noradrenalin were acting only presynaptically.

6 Role of hormones in the insect life history Hormonal involvement in behaviour has been reported for many aspects of insect behaviour. These claims are usually based on one or more of the following criteria: 1. The correlation of the onset of hormont: secretion with the appearance of a specific behaviour. 2. The precocious induction of a behaviour by hormone application or by implantation of active glands. 3. The disappearance of a particular behaviciur after the removal of the endocrine source. 4. The manipulation of the behaviour by removal and replacement therapy, i.e. the demonstration that removal of an endocrine centre abolishes a specific behaviour which can then be restored through hormone application or gland implantation. Of these four lines of evidence the first is the weakest and, in the absence of additional data, cannot be taken to show any causal relationships. The first approach has the additional disadvantage that the claim for hormone release is often based on histological change,; in the secretory cell. As indicated by Highnam (1965), the static picture presented by histology must be interpreted with caution if one is trying t o represent the true dynamic state within the cell. The third method also presents some pitfalls when applied to neurosecretory structures because ablation of the source of neurosecretion may also result in the removal or damage of nonneurosecretory neurons. The observed effects may then be due to the destruction of neural centres which are crucial to the performance of the behaviour rather than to the removal of an important endocrine component. The last method guards against this pitfall and provides the strongest evidence for hormonal involvement. The relative merits of the four approaches should be kept in mind while reading the remainder of this section. 6.1

HORMONAL MECHANISMS UNDERLYING LARVA:L BEHAVIOUR

The influence of hormones on larval behaviour has received only occasional interest. In comparison with the adult stage, larval insects show a much reduced behavioural repertoire. That of holometabolous larvae, in '

AIP-11

312

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

particular, is often extremely modified for a single purpose-the ingestion of food. 6.1.1 Larval activity The larval stages of insects are punctuated with periods during which the larva temporarily ceases feeding. This condition often occurs during larval diapause or during the moult. As a rule, moulting larvae display decreased locomotor activity or, as in the case of lepidopteran larvae, complete quiescence. This periodic cessation of activity was investigated in the desert locust, Schistocerca gregaria, by Haskell and Moorhouse (1963). These authors studied the response of a standard nerve cord preparation obtained from male locusts to applications of haemolymph taken from nymphs during the moult or intermoult periods. Changes in levels of interneuron activity were determined by recording from the commissures between the first and second thoracic ganglia; motor neuron activity was monitored in the nerve which innervates the extensor tibialis muscle of the metathoracic leg. When haemolymph from intermoult nymphs was used to bathe the nerve cord, little change was noted in levels of either interneuron or motorneuron activity. By contrast, haemolymph obtained from larvae 12 h prior to ecdysis caused a marked increase in interneuron firing, but substantially decreased motor output (Fig. 4). This effect of blood from moulting larvae could be mimicked by adding to the intermoult blood 'Bombyx mori extracts which had high concentrations of a-and 0-ecdysones. Evidently, the hormonal agent which acts on the epidermis to provoke the moult also acts on the CNS to suppress locomotor activity prior to ecdysis. 6.1.2 Larval behaviour at the outset of metamorphosis

In holometabolous larvae the endocrine activity of the corpora allata begins to decline midway through the last instar and reaches a low level by the time of pupation (Williams, 1961). Accordingly, towards the end of the last instar, the blood titre of JH decreases. During this period of endocrine changes, there are also major behavioural alterations. In Lepidoptera, the mature last stage larva typically ceases feeding and leaves the food plant either to burrow into the ground or to begin cocoon construction. In the linden moth, Mimas tiliae, Piepho et al. (1960) showed that the mature larva becomes positively geotactic and descends from its tree t o the ground. This positive geotaxis could be reversed by the implantation of active corpora allata. Therefore, the sign of larval geotaxis is influenced by the JH titre in the blood (also see section 6.4). The declining titre of JH apparently facilitates the performance of the premetamorphic behaviour but does not trigger its occurrence. Thus, in another sphinx moth, Manduca sexta, the cessation of feeding is followed

HORMONES AND BEHAVIOUR

31 3

2

Fig. 4. The effect of blood from moulting larvae on spontaneous neural activity in the locust nervous system. (A) Simulated records from the connective between the pro- and mesothoracic ganglion; (B) simulated records from nerve which innervates the extensor tibialis. 1, before addition of blood; 2, a f t a adding blood. Drawings of records based on data given in Haskell and Moorhouse (1963).

by increased locomotor activity, voiding of the gut, appearance of a pink pigment in the dorsal epidermis, and exposure of the heart. All of these events occur at approximately the same time. Experiments involving neck ligation of mature Manduca larvae showed that at least the last two events occur in response to a factor released from the head (Truman et al.. unpublished). Presumably, the hormonal ccmtrol of red pigmentation in Manduca is similar to the control reported for Cerura vinula (Buchmann, 1959; Hintze, 1968) in which the pigmentation change is mediated through the brain-prothoracic gland system. The behivioural events associated with the appearance of pink pigment in Mamluca (i.e. gut emptying and increased locomot ion) are probably also triggered by the same endocrine signal. One might suspect that some of the premefamorphic behaviour of larvae is potentiated by the low JH titre and then released in response to the prothoracicotropic hormone or to ecdysone.

314

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

6.1.3 Cocoon construction The role of J H in cocoon construction was first indicated by the classic experiments of Bounhiol (1938) using Bombyx mori. Allatectomy early in the penultimate larval instar caused precocious pupation and, as a prelude to this, the precocious spinning of a miniature cocoon. By contrast, implantation of active corpora allata into last stage larvae caused a supernumerary larval moult and the delay of cocoon construction. In the greater wax moth, Galleria melonella, a silken structure is spun at the end of each instar but the shape changes depending upon the instar. The penultimate (4th) stage larva spins a thin-walled moulting tube. The mature larva constructs a tough-walled cocoon. When active corpora allata from fourth instar larvae were implanted into a last stage larva, the latter spun a structure which was intermediate between a cocoon and a moulting tube (Piepho, 1950, 1960; Wiedbrauck, 1955). In some other lepidopterans the untimely appearance of J H prevents spinning. The application of J H mimics to mature Cecropia larvae delayed the onset of spinning for 1 to 3 days but did not alter the type of cocoon which was eventually spun (Riddiford, 1972). Application of JH mimics to last instar Cerura larvae also disrupted spinning behaviour but the data indicate that the disruption may simply be due to the abnormal development of the silkgland (Hintze-Podufal and Fricke, 19 7 1). Neurosecretion may also play a role in the type of cocoon constructed. In the silkworm, Samia Cynthia, a hormone from the brain reportedly directs the construction of either a summer or a winter cocoon (Pammer, 1966; Nopp-Pammer and Nopp, 1968). The type of cocoon spun is correlated with whether or not the insect will diapause. N ADULT BEHAVIOUR 6.2 A ~ I V A T I O OF

In the wild silkmoths, a hormone also regulates the behavioural event which signals the end of metamorphosis-the eclosion of the adult. After the completion of adult development, the pharate moth waits until a speciesspecific time of day in order t o emerge (Truman and Riddiford, 1970). When the pupal cuticle was removed from pharate moths prior to the normal time of emergence, the peeled moths did not show proper adult behaviour (Bastock and Blest, 1958; Blest, 1960; Truman, 1971a). In the case of Antheraea pernyi, peeled pharate moths exhibited little, if any, locomotor activity. Also, other typical post-emergence behaviours, such as wing-spreading and flight, were not observed. These peeled moths characteristically showed only the rotations of the abdomen that were typical of pupae (Truman, 1971a). At the normal time of eclosion, peeled Pernyi moths performed the entire emergence sequence (pre-eclosion movements, eclosion, post-ecdysis activity, and wing-spreading), even though the pupal cuticle was removed

HORMONES AND BEHAVIOUR

31 5

hours earlier. By the time that this sequence was complete, the moths showed the full adult behaviour and the pupal behaviour was lost (Truman, 1971a). It was as if the “adult nervous system” had been turned on and the “pupal nervous system” shut off. This activation of adult behaviour could be provoked at any time of day by the injection of the eclosion hormone into a pharate moth. Since brainless silkmoths usually emerge, the eclosion hormone is not an absolute requirement for eclosion t o occur. Apparently, in the absence of the eclosion hormone, some of the neural centres which are responsible for the various bits of behaviour can eventually “turn on”. Under these conditions the smooth coordination between the behavioural parts is lost. In the Cecropia moth, for example, most debrained moths escaped from the pupal skin but only 12 per cent displayed the pre-eclosion behaviour (Truman, 1971a). In our experience, no brainless moth has ever spread its wings. Yet, a debrained moth which has a brain implant in its abdomen (and therefore is supplied with the eclosion hormone) goes through the entire emergence sequence including the spreading of the wings. These data suggest that a lack of proper emergence behaviour in brainless moths is due, not to a disruption of the nervous connections with the brain, but to the absence of the eclosion hormone. On the cellular level, the action of the eclosion hormone in “turning-on” adult behaviour and “turning-off’’ p u p d behaviour can be seen in the degeneration of the intersegmental muscles. A.s noted above, these muscles line the third through sixth abdominal segments and are primarily responsible for the movements seen during eclosion. After emergence they begin to degenerate and disappear within the following three days (Lockshin and Williams, 1965a; Finlayson, 1956). The signal for these muscles to undergo histolysis is apparently given by the motor neurons which supply them. When the appropriate nerves were chronically stimulated either electrically or pharmacologically (e.g. with physostigmine), the muscles were retained (Lockshin and Williams, 1965b, 1 9 6 5 ~ )Thus, . the fate of the intersegmental inuscles appears to reflect the “activity” of the neurons which supply them. Lockshin (1969) later found that when the abdomens of A. polyphemus were isolated prior to emergence, the intersegmental muscles were not only retained, but remained functional for at least three days after the normal time of eclosion. He concluded that a signal from the anterior portion of the moth was required for the “shut-down” of the motor neurons. Injection of the eclosion hormone into isolated Polyphemus abdomens subsequently demonstrated that this signal was supplied by the eclosion hormone (Truman, 1970). As part of its action of “turning-off” pupal behaviour, the eclosion hormone apparently shuts down the motor neurons which supply the intersegmental muscles. Subsequent studies (Taylor and Truman, in ]~qm.ration) have shown that

316

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

eclosion is followed by the selective degeneration of specific motor neurons in the abdomen of the adult moth. In addition certain groups of small neurons (presumably interneurons) are also rapidly lost. The eclosion hormone, therefore, appears to trigger the destruction of the neural circuitry which coded for pupal behaviour in the pharate moth.

6.3 HORMONES

AND REPRODUCTIVE BEHAVIOUR

The influence of hormones on insect behaviour has been most thoroughly studied in connection with sexual maturation and reproduction. During its life the adult insect often goes through several transitions from one behavioural state to another-i.e. a non-receptive virgin becomes receptive. Since all aspects of reproductive behaviour from mate attraction to the deposition of fertilized eggs must be smoothly coordinated with both the internal physiology and the external environment, it is not surprising that hormones play a large role in this coordination. Hormones typically serve to link behaviour with internal happenings, whereas pheromones serve as major external cues for behaviour. The latter are known from most orders of insects (Jacobson, 1972) and with the advent of microanalytical methods, their isolation and identification has burgeoned during the past few years (see Law and Regnier, 1971, for recent review). The spectrum of action of pheromones on behaviour is outside the scope of this paper except in so far as they may mediate the release of hormones, or as hormones may mediate the behaviour involved in their release. Our present concern is the hormonal control of sexual receptivity, mating, and oviposition.

I

6.3.1 Male sexual behaviour Compared t o the female, the male of most insect species shows rather simplified behaviour. Mate-finding, copulation and, perhaps, feeding are his main pursuits. In accordance with this reduced behavioural repertoire, one also finds a paucity of cases of hormonal involvement in either the “maturation” of male behaviour or its actual expression. In most instances male behaviour matures in concert with the metamorphosis of the nervous system. Thus, at adult ecdysis or shortly thereafter, suitable stimuli will evoke the full copulatory behaviour. This behavioural state then persists until death. Many members of the Orthoptera are in contradistinction to this general picture. In some species the CA are essential for the appearance of male - - - of CA t o sexual behaviour. But in other related species, the importance behaviour decreases or becomes nil. This rather confusing picture of endocrine control can be somewhat clarified by considering the role of the sexual behaviour in the biology of the insect (see Barth and Lester. 1973).

HORMONES AND BEHAVIOUR

317

For example, one might anticipate that lon!g-lived insects which have a restricted breeding season might be expected t o be able to turn reproductive behaviour “on and off”, whereas short-lived or continuously breeding species would probably not be able to do so. A second factor is undoubtedly the wide range of criteria which different investigators use to appraise the behavioural status of their insects. Often assessments are based simply upon whether or not an introduced fernale becomes inseminated or upon chance or random observations of the behaviour of the male. The recent development of quantitative measures of male behaviour (e.g. Wajc and Pener, 1969) promises to defhe better thc: roles of endocrine agents in the regulation of behaviour. a. Grasshoppers and locusts. The endocrine control of male behaviour has been most widely studied in the gasshoppers and locusts. In the desert locust, Schistocerca gfegaria, and the red locust, Nomadacris septetnfasciata, the corpora allata are required for the maturation of sexual behaviour in males. In both Schistocerca and Nomaducris allatectomy of the male soon after ecdysis completely prevented the onset of sexual behaviour which normally occurred .about 10 days post-ecdysis (Loher, 1961; Pener, 1965, 1967, 1968; Odhiambo, 1066a; Pener and Wajc, 1971). The implantation of several pairs of active corpora allata into an allatectomized male promoted the reappearance of this behaviour (Loher, 1961; Pener, 1965, 1968). The regulation of J H secretion in Schistoccwa is apparently controlled by the brain via the nervus corporis allati (Pener, 1965, 1967; Odhiambo, 1966b). Sectioning of the nerve in freshly emerged males severely inhibited the appearance of male behaviour. But, eventually about 10 per cent of the insects displayed some sexual activity (Pener, 1967). This low level of sexual behaviour was presumably due t o a small amount of JH which is secreted from the denervated gland. Brain control of the activation of the corpus allatum was also suggested by the obwrvation that flight activity accelerated sexual maturation in these males (Norris, 1962). Highnam and Haskell (1964) have shown that similar activit.y in Schistocerca or crowded Locusta females caused accelerated oocyte growth through the increased release of the corpora allata-activating neurosecretion. One would then suspect that the activity-induced maturation of male locusts reported by Norris (1962) was also due to the release of the same allatatropic neuro secret ion . Allatectomy of Schistocerca males has diverse effects including an inhibition of production of the pheromone which accelerates sexual maturation in other adult male locusts (Loher, 1961) and the prevention of accessory gland development (Cantacuzine, 1967, 1968). Accordingly, the action of J H on malq behaviour could be an indirect result of the maturation of some other system such as the accessory glands. However,

318

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

Loher (1961) showed that sexual behaviour occurs even after removal of the accessory glands and the gonads. The same has been demonstrated for the grasshopper, Chorthippus parallelus (Haskell, 1960). In strains of Locusta which undergo reproductive diapause, the male accessory glands remain well developed but sexual behaviour is not displayed (Cantacuzine, 1968). Consequently, the effects of JH on male behaviour cannot be explained solely by influences from the maturing accessory glands on the insect’s behaviour. As suggested by Odhiambo (1966a), it is very likelythat JH acts directly on the central nervous system of Schistocerca t o control the development or unmasking of the neural circuitry involved in courtship and mating. This unmasking of the proper circuits appears to require the continuous presence of JH since allatectomy of a mature male will lead to the disappearance of sexual behaviour (Loher, 1961; Pener, 1967). In the male migratory locust, Locusta miFatoria migratorioides, JH has a similar role but its control is not as complete as that seen in Schistocerca (Girardie, 1966; Girardie and Vogel, 1966; Wajc and Pener, 1969; Pener and Wajc, 1971). The CA appears to regulate the intensity of sexual behaviour in Locusta males. The necessity of the corpora allata has been challenged by Strong (1968a, 1968b) who claims that his allatectomized Locusta showed no diminution of sexual behaviour. But Strong (1968a) only observed the mounting behaviour of allatectomized males with other similarly treated males. Later Wajc and Pener (1969) used more rigorous quantitative criteria for sexual behaviour and observed the interaction of allatectomized males not only with each other, but also with normal males and with females. They found that the intensity of sexual behaviour towards females or towards normal males was greatly diminished by allatectomy (Fig. 5 ) . Among groups of operated males the frequency of mounting each other was greater than with females or with normal males. This last behaviour was apparently due t o the sluggishness of such males and the fact that they reacted less violently to this type of approach (Pener and Wajc, 1971). In Locusta the C-type neurosecretory cells of the pars intercerebralis appear to be responsible for the activation of the corpora allata which in turn leads to the onset of sexual behaviour (Girardie, 1966, 1970; Girardie and Vogel, 1966; Broza and Pener, 1972). Cauterization of the pars intercerebralis completely prevented the appearance of sexual behaviour. Importantly, when 6 partes intercerebrales were implanted into cauterized males, both sexual behaviour and the mature yellow pigmentation were restored (Girardie and Vogel, 1966). Although the action of the C-cells of the brain is partially through their stimulation of JH secretion, they also have direct behavioural effects. This additional role was indicated by the fact that allatectomy only partially reduced male sexual behaviour, whereas cauterization of the pars inter-

HORMONES AND BEHAVIOUR

319

Weeks after treatment

Fig. 5. The effects of allatectomy on the sexual activity of L. m. migratorioides males. The determination of time spent in sexual behavioux for each group was based on a series of 2-h observation periods during which females were introduced. Attacking or mounting a female, copulation, and mounting or jumping on to a male were considered as male sexual behaviour. A, sham-operated males; B , unoperated males; C, allatectomized males. (Redrawn from Pener and Wajc, 1971.)

cerebralis completely abolished it (Pener et al., 1972). Moreover, the behavioural effects of cauterization could bc partially reversed by pars intercerebralis implants but not by CA implants. Accordingly, Pener et al. (1972) concluded that the C-type neurosecretory cells are required for the performance of sexual behaviour and that the 1CA then regulate the level of this behaviour. The interaction between the corpora allata and the neurosecretory system of the brain provides a control over the onset of sexual behaviour which is responsive to environmental conditions. In the grasshopper, Oedipoda miniata, as well as in some subspecies of L. mipatoria (the Kazalinsk and the Panzano strains of L. m. cin.erascens), the males show a photoperiodically controlled diapause (Cantacuzkne, 1967; Darjo, 1969; Perez et al., 1971; Broza and Pener, 1972). In these insects the sexual behaviour of the male was severely inhibited or eliminated under long-day conditions, but it was restored by the implantation of active corpora allata (Dajo, 1969; Broza and Pener, 1972). Presumably, as has been shown in Leptinotarsa decemlineata (de Wilde and de Boer, 1969), the photoperiod controls the release of neurosecretion which then activates the corpora allata. Along the same lines, in the Egyptian grasshopper, Anacridium aegyptium, Geldiay (1967, 1970) has shown that reproductive diapause is correlated with the cessation of neurosecretory activity in the brain.

320

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

In contrast t o the above species, the grasshoppers Gomphocerus mfus (Loher and Huber, 1966; Hartmann, 1971) and Syrbula fuscovittata (Loher, unpublished, as cited in Hartmann et al., 1972) have no CA control over the maturation of male behaviour. These latter species have neither a facultative nor an obligatory reproductive diapause, as do the species which exhibit an endocrine control over sexual activity. Pener (1970) has suggested that the tightness of control which the CA exercises over the maturation and maintenance of male sexual behaviour is an adaptation to allow for adult diapause. This endocrine regulation then permits the behaviour of the male to shift with changing environmental conditions. It will be interesting to see if this hypothesis holds for other groups of insects which have an adult reproductive diapause. b. Cockroaches. Studies of the role of the endocrine system on the behaviour of male cockroaches are relatively few. The corpora allata are apparently not needed for the onset of male sexual behaviour, but this conclusion is based primarily upon studies of only two species-Byrsotria furnigata (Barth, 1962, 1968) and Gromphadorhina brunneri (Ziegler, 1972a). Although the onset of male behaviour may not be controlled by JH, the neurosecretory hormone discovered by Milburn et al. (1960) almost certainly has a role during courtship and mating. As described above, the phallic nerve stimulating hormone elicits abdominal movements which are similar to those observed in males during normal copulation. Unfortunately, there are no data concerning the effects of removal of the corpora cardiaca or of cauterization of the pars intercerebralis on the mating ability of male cockroaches. It also is unknown whether the hormone is released from the corpora cardiaca and is present in the blood during mating. Although this hormone is most probably involved in cockroach mating behaviour, its relative importance to this behaviour remains to be explained. c. Diptera. In Diptera the only evidence for hormonal involvement in male behaviour is in the yellow dung fly, Scatophaga stercoraria (Foster, 1967). When the fly was deprived of food after emergence, both sexual behaviour and the accessory glands remained undeveloped.. This same syndrome was also seen in fed flies which were allatectomized immediately upon emergence. The effects of prey deprivation or of allatectomy could be reversed simply by implanting active corpora allata. Thus, in males of this fly, food intake apparently serves to activate the corpora allata which, in turn, promote the onset of sexual behaviour. 6.3.2 The development of receptivity in virgin females a. Involvement of the corpora allata. Whereas the corpora allata seem not to be necessary for male sexual activity in most insects, the reverse is true in the case of the female. But one must carefully distinguish between

HORMONES AND BEHAVIOUR

321

direct effects of J H o n the behaviour of the female toward the courting male and indirect effects of JH in mediating sex pheromone production and thereby attraction of the male. The distinction is especially important because in many studies the decision that the female has acquired receptive behaviour is not based upon a careful observation of her behaviour during courtship, but rather upon whether she has been inseminated or not. The classic example of the distinction between JH effects upon behaviour and upon pheromwe production is seen in the cockroach, Byrsotria fum*ata (Barth, 1961, 1962). When allatectomized females were placed with males, no mating occurred. This effect was not due to lack of receptivity on the part of the female, but rather to her failure to produce the volatile sex pheromone. When such females were coated with extracts containing pheromone, they were vigorously cclurted and copulated readily (Barth, 1961, 1962; Roth and Barth, 1964). Thus, the CA of Byrsotria are not involved in the onset of sexual receptivity and, indeed, the JH-mediated onset of pheromone production occurs two days prior to the normal appearance of sexual behaviour (Barth and Bell, 1970). In addition, the corpora allata do not appear to be involved in sexual maturation in the oviparous cockroach, Diploptera punctata (Engelmann, 1960b), the cricket, Gryllus bimaculatus (Rtmssel, 1967), or the lepidopterans, Galleria mellonella (Roller et 01.. 1963), Bombyx mori (Bounhiol, 1938; Fukuda, 1944), Antheraea p e m y i (Barth, 1965; Riddiford and Williams, 1971), A. polyphemus, and Hyalophora cecropia (Riddiford and Williams, 1971). Some of these latter species, however, show another type of endocrine control over their receptive behaviour (see part c). At the other end of the spectrum one finds insects in which the CA completely control the onset of female receptivity. The best known example is that of Gomphocerus (Loher, 1962, 1966; Loher and Huber, 1964, 1966; Hartmann et al., 1972). Loher and his coworkers have shown that allatectomy of last instar or newly emerged females completely prevented the onset of sexual receptivity. Al1att:ctomized females remained in a permanent state of “primary defence” arid responded to a courting male by kicking and escape. The implantation of active CA into such females then led to a state of “copulatory readiness”-courting males were allowed t o mount and copulate. The application of JH mimics such as farnesyl methyl ether also served t o promote receptivity in allatectomized females (Loher, 1966). Copulatory readiness was not only initiated but also maintained through the action of the CA. Allalectomy of sexually mature females led to the loss of receptivity and the reappearance of primary defence behaviour within six days. As in the case of some male acridids, the neurosecretory cells of the pars intercerebralis of the female Gomphocerus have a role in turning o n the CA

322

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

(Loher, 1966). They thereby indirectly regulate both oocyte maturation and the appearance of sexual receptivity. It is of interest that when active CA were implanted into females from which both these neurosecretory cells and the CA had been removed, yolk deposition and accessory gland development were stimulated but sexual behaviour was not restored (Hartmann et al., 1972). This experiment reconfirms the fact that the onset of receptivity is not simply the outcome of the activation of the gonads since the latter can proceed without inducing sexual behaviour. Furthermore, the data raise the possibility that a brain neurosecretion has a direct role in the development or expression of receptivity. However, in the absence of reimplantation experiments, one cannot exclude the possibility that cauterization of the pars intercerebralis simply resulted in destruction of neural centres which were crucial to the performance of receptive behaviour. In another grasshopper, Euthystira brachyptera, the CA do not appear to have as strong a control. Miiller (1965) reported that allatectomy of this species 2 to 24 h after adult ecdysis did not prevent the onset of receptive behaviour. But some J H secretion had apparently already occurred by the time of the operations because the operated females also showed traces of egg maturation. This small amount of secreted JH may have been sufficient to trigger the later appearance of receptivity. This interpretation gains additional support from the fact that implants of active corpora allata or application of farnesol will induce precocious readiness in young females (Miiller, 1965). Cockroaches provided the first evidence that hormones were involved in the onset of sexual behaviour in insects. Engelmann (1960a) demonstrated that allatectomy of newly emerged Leucophaea maderae females led to a marked reduction in the number of females which became receptive and mated (30 per cent). When CA were implanted into such females, normal sexual receptivity ensued. These results were later challenged by Roth and Barth (1964) who reported that their allatectomized females became receptive and mated normally. These conflicting results were later resolved by Engelmann and Barth (1968). Using Engelmann’s cockroaches, allatectomy again led to a low percentage of mated females. Moreover, the operated females which mated did so only after a long bout of courtship by the males. These authors concluded that the conflicting results obtained by the two laboratories arose from strain differences within Leucophaea. This confusing picture of the role of the CA is also evident when other species of cockroaches are examined. In Nauphoeta cinerea the onset of receptivity was correlated with the JH titre as estimated by the size of the gowing oocytes (Roth, 1962). Later Roth and Barth (1964) showed that allatectomy delayed the onset of receptivity but did not completely prevent it. In the species Bhttellagermanica, B. vaga (Roth and Stay, 1962)

HORMONES AND BEHAVIOUR

323

and Byrsotria f u m k a t a (Barth, 1961, 1962), sexual receptivity arose normally, even in the absence of the CA. By contrast with the above species, allatectomy of last instar larvae or freshly ecdysed females of Gromphadorhina brunneri completely prevented the onset of receptivity (Ziegler, 1972b). Although this operation may also have interfered with the production of a non-volatile pheromone, allatectomized females actively warded off the approaches of courting males (Ziegler, 1972b). The differences seen among the cockroach species may in part be due simply to differences in the time that the CA become active. Indeed, as indicated above for the grasshopper, Euthysti#m(Miiller, 1965),JH released at the time of ecdysis (when operations are iinpossible) or even before this event may be responsible for the turning-on of sexual receptivity. Therefore, species in which the CA do not appear to be necessary may simply be those in which adequate JH is rel..ased prior to or during ecdysis. Besides the role of the CA, there is evidence that the neurosecretory cells of the brain may also be needed for the onset of receptivity in cockroaches. In Nauphoeta cinerea, Roth and Barth (1964) hypothesized that the delayed onset of receptivity which was observed in allatectomized females may be due to a neurosecretion from the brain. In Byrsotria a histological examination of the brain revealed a change in newosecretory activity at the time of parturition (a time when females again become receptive) (Barth, 1968). However, in this species one finds no comparable neurosecretory changes during the initial maturation of sexual receptivity. Barth (1968) suggested that the prolonged courtship necessary t o induce receptivity in allatectomized Leucophaea as well as the delayed onset of receptivity in allatectomized Nauphoeta indicated that neural centres controlling this behaviour may be turned on by environmental influences, presumably working through the neuroendocrine system. Similarly, he suggested that the retardation of sexual receptivity in Blattella germanica, B. uaga (Roth and Stay, 1962), and Leucoplzaea (Roth and Barth, 1964) which can be induced by starvation may be due t o the prevention of release of the proper neuroendocrine agent. The only direct evidence for the involvement of a brain neurosecretion has been reported for Leucophaea (Engelmann and Barth, 1968). Cauterization of the pars intercerebralis of the female completely abolished mating behaviour. The proper behaviour could not be restored by implantation of active CA. Therefore, the absence of receptive behaviour was not due simply to the lack of a tropic factor which is necessary for JH release. Rather, a neuroendocrine agent was probably directly involved in the appearance of the behaviour. But, as discussed above for female Gomphocerus, without additional implantation data this evidence cannot be taken as unambiguous support for a neuroendoaine involvement. In Diptera the role of the CA in promoting female receptivity was first

324

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

demonstrated in Drosophila melanogaster (Manning, 1966, 1967). The female of this insect becomes receptive to courting males during the second day of adult life. When active corpora allata-corpora cardiaca complexes were implanted into pharate females 1 7 t o 19 h before eclosion, the females became precociously receptive by 24 d after eclosion (Manning, 1966). Sexual receptivity in Musca domestica is likewise influenced by the CA (Adams and Hintz, 1969). Allatectomy of freshly emerged females did not prevent sex pheromone production and, consequently, attraction of the male occurred normally. But operated females did not usually respond to the mounting male by protrusion of the ovipositor. Thus, successful copulation was prevented. JH application to allatectomized females caused the onset of receptive behaviour. In Diptera the most complete studies of the role of the CA in promoting female sexual behaviour have been carried out on mosquitoes, especially Aedes aegypti. The newly emerged female is sexually unreceptive and, although males may attempt to copulate with her, successful insemination does not occur (Lea, 1968;Gwadz and Craig, 1968;Spielman et al., 1969). By one to two days after emergence, depending upon the particular strain, females accept males and are inseminated (Lea, 1968; Gwadz and Craig, 1968).The onset of receptivity could be substantially delayed by removal of the CA from newly emerged females (Lea, 1968).But this condition was not permanent, and such females showed a considerable percentage of inseminations by 12 days after eclosion. The eventual appearance of receptivity might again be due to the secretion of a small amount of JH prior to CA removal. This possibility is likely since Lea (1968)has shown that the CA of female mosquitoes are active within 5 min of emergence. As with other insects, implantation of CA completely restored receptivity t o allatectomized females (Lea, 1968). Similarly, application of a JH mimic to female A . aegypti within 30 min after emergence caused the precocious onset of sexual receptivity (Gwadz et al., 1971b; see also Spielman et al.. 1969). b. Influence of the ovaries. Although the non-involvement of the gonads in the appearance of sexual behaviour appears to be a general rule (Regen, 1909; Husain and Baweja, 1936; Loher and Huber, 1966), there are exceptions. In the grasshopper, Chorthippus parallelus, the removal of the ovaries typically resulted in the disappearance of the “responsive state” within 24 to 48 h (Haskell, 1960).When blood from receptive females was injected into castrated unreceptive females, a transient return of receptivity was observed. The latter experiment should be considered as preliminary since only three individuals were involved and all died within a few days of treatment. But taken in conjunction with the other data, they indicate an involvement of the ovaries in the maintenance of receptivity and suggest the possibility of hormonal mediation.

HORMONES AND BEHAVIOUR

325

c. Corpora cardiaca control. Female wild silkmoths emerge with ovaries full of mature eggs and in a sexually receptive condition. I n response to specific environmental cues (Riddiford and Williams, 1967, 1971; Riddiford, 1968), the female assumes the “calling” posture. This behaviour involves the protrusion of the last two abdominal segments which exposes the pheromone glands and allows the release of the sex attractant. Extirpation of the CA had no effect on the calling behaviour of virgin females of A. polyphemus or H. cecropia. But when the corpora cardiaca were also removed, calling was essentially abolished (Riddiford and Williams, 1971). This behavioural deficiency is most probably due to the lack of the intrinsic cells of the CC because under the conditions of the above experiments the neurohaemal portion of the CC routinely regenerates (Stumm-Zollinger, 1957). Transection of the small nerves between the brain and the CC was as effective as corpora cardiacectomy in preventing calling behaviour (Riddiford and Williams, 1971). Apparentlv, the release of the “calling hormone” is controlled neurally by the brain. The calling hormone has been shown to be present in the blood of “calling” Polyphemus females. When injected into virgin females in the absence of appropriate stimuli, blood from calling moths routinely provoked calling in 1% to 2 h. The blood of non-calling virgin females did not have this effect (Riddiford, 1973).

6.3.3 The onset of refiactoriness in mated females In most female insects a successful mating is followed by the termination of receptivity t o the courtship of subsequent males. In some species this refractoriness is only temporary, whereas in others it lasts for life. The stimuli for the switch to refractory behaviour are various, ranging from the mechanical stimulus of the mounting male (Van den Assem, 1970) or the spermatophore (Roth, 1962) t o presumably chemical cues provided by the male accessory gland material (Craig, 1967)or by the sperm (Davey, 1965). Most cockroaches and some acridids use a purely neural mechanism to enforce non-receptivity. In Nauphoeta cineraea, the physical presence of the spermatophore in the bursa copulatrix turns-off receptivity (Roth, 1962, 1964). Transection of the nerve cord anterior to the terminal abdominal ganglion led t o the reappearance of receptivity in mated females (Roth, 1962). Manifestly, in this instance the information concerning the presence of the spermatophore must be relayed through neural channels to the brain. A similar mechanism which induces “secondary defence” has been reported in Gomphocerus (Loher, 1966; Loher and Huber, 1964, 1966). In other acridids the ovaries may play a role in the onset of refractory behaviour. In Euthystira bachyptera secondary defence begins exactly at

326

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

the time of ovulation when the spermatophore is dissolved (Renner, 1952). This switch in behaviour could be prevented by ovariectomy (Renner, 1952), but was not affected by transection of the ventral nerve cord (Miiller, 1965). Thus, Muller suggested that unreceptive behaviour in Euthystira arose from the action of a humoral substance from maturing eggs. In the closely related Chorthippus parallelus, Haskell (1960) found that the ovaries had the opposite effect. As described above, gonadectomy of receptive virgin females rapidly led to the disappearance of receptivity. Whether these conflicting results arise from species differences is uncertain at this time. Clearly, further studies are needed to confirm the role of the ovaries in receptive behaviour and to clarify the mode of interaction of the ovaries with the nervous system. The effects of the ovaries are possibly mediated by the corpora allata. This was indicated by the experiments by Miiller (1965) which showed that allatectomy of Euthystira females prevented the onset of secondary defence. Similarly, in Chorthippus curtipennis, Hartmann and Loher (in preparation, cited in Hartmann et al., 1972) never observed secondary defence in allatectomized females. But since allatectomized Euthystira cease egg maturation (Miiller, 1965), it is also possible that the CA influence is indirect and that allatectomized females remain receptive because they have no mature eggs. In some Diptera the onset of refractoriness is a two-step process; mating often results in a rapid, neurally mediated refractory condition followed by a slower, more persistent refractory state which is often hormonally induced. In mosquitoes, the immediate switch-off of receptivity is due to the filling of the bursa copulatrix with seminal fluid (Gwadz and Craig, 1970; Gwadz et al., 1971a) and is most likely neural. But, once inseminated, A. aegypti females remain behaviourally unreceptive t o all future matings. This long-lasting refractoriness is caused by matrone from the male accessory glands (Craig, 1967; Fuchs e t al., 1969; Fuchs and Hiss, 1970). Presumably, matrone passes through the walls of the bursa copulatrix into the haemolymph and then acts on the nervous system. Indeed, injections of matrone into the haemocoel were just as effective as that received in the normal manner. In Drosophila there are also multiple mechanisms involved in the control of receptivity. In the female receptivity is immediately turned-off by the mechanical stimuli associated with copulation, but this effect lasts only 2 h (Manning, 1967). A long-term change is presumably caused by the sperm because mated Drosophila females again become receptive after the sperm are depleted (Manning, 1962, 1967). The reports that females which were mated t o castrated males did not regain receptivity until 24 h later (Smith, 1956; Manning, 1962) indicated that a third mechanism was also involved. Merle (1968) then demonstrated that implants of the male paragonial

HORMONES AND BEHAVIOUR

327

glands into virgin females inhibited sexual receptivity for at least 5 days. Thus, in Drosophila there exist three stimuli which cause the switching-off of receptivity: the neural stimulus caused by mating, the paragonial gland secretion, and the sperm. In Musca a secretion from the male ejaculatory duct is deposited in the vaginal pouches of the female during mating (Riemann et al., 1967; Adams and Nelson, 1968; Riemann and Thorson, 1969). This protein then rapidly enters the haemolymph, the female becomes unreceptive, and mating terminates (Leopold et al., 1971a). Using male accessory secretion labelled with 3H-arginine or 3H-lysine, Leopold et al. ( 1971a, 1971b) and Terranova et al. (1972) have followed the movement of the secretions into the haemolymph and have shown that some of the material can be subsequently found in the head-the proposed site which controls female receptivity. These hormone-like accessory gland substances exert their effects on various portions of the central nervous system In A. aegypti, both matrone and JH presumably act on centres in the female terminal abdominal ganglion (Gwadz, 1972). In Drosophila a deca.pitated female is unreceptive and keeps her vaginal plates closed (Spieth, 1966). If a male breaches these defences, however, the female then shows normal behaviour and begins ovipositing. Thus, some control over the behaviour must be relegated to the fused thoracic-abdominal ganglia. By contrasi:, decapitated Musca females do not lose receptivity after mating. Thus, the brain is apparently most important for controlling receptivity in Musca (Leopold et al., 1971b). 6.3.4 Owposition behuviour In some insects, the effect of mating on receptivity has not been studied, but its effect o n oviposition is well documented. The stimulation of oviposition and the cessation of receptivity are closely related behaviours which most probably represent two manifestat ions of the same process, i.e. a switchover from the “virgin mode” to the “mated mode”. Thus, changes in oviposition rate can provide a quantitative measure of the extent of this behavioural switch. This type of quantitative assessment of the effects of mating on female behaviour is complicated by the fact that mating often also stimulates oogenesis (Engelmann, 1970). Consequently, ;in enhanced oviposition rate might represent not a behavioural change but merely the increased rate of egg maturation. Only a few workers (e.g. Eienz, 1969) have made the distinction between a change in oogenesis rate and a change in oviposition rate. In many insects there seems to be a release of neurosecretory material at the time of oviposition. On the basis of histological studies, this relationship has been claimed for Schistocerca gregaria (Highnam, 1962a,

328

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

1962b; Delphin, 1965), the cockroach, Trichoblatta sericia (Adiyodi and Nayar, 1965), the pyrrhocorid bug, Iphita limbata (Nayar, 1958), the beetles, Leptinotarsa decemlineata (Schooneveld, 19 70) and Galeruca tanaceti (Siew, 1965), and the silkmoth, Bombyx mori (Arvy et al., 1953). But only in Schistocerca and Iphita has there been any direct demonstration of oviposition-stimulating activity by implantation of neurosecretory cells or injection of their extracts. In Iphita pairs of bugs copulate for 3 weeks during which time the eggs of the female mature. When blood was taken from females during the period around the time of oviposition and injected into partially gravid mating females, mating was terminated and oviposition induced (Nayar, 1958). These injections also served to deplete the neurosecretory cells of the brain. Since a similar depletion was noted after injection of extracts of mature ovaries, it was concluded that the factor in the blood was derived from the ovary. This material then acted on the neurosecretory cells of the brain to cause the release of an oviposition-stimulating hormone. Further evidence for this hypothesis was obtained by the implantation of two sets of neurosecretory cells from a mating female into a nearly gravid female. This treatment rapidly stimulated oviposition. When neurosecretory cells were implanted into young females which lacked eggs, quivering movements of the genital plates (a characteristic behaviour which accompanies oviposition) were triggered. In Schistocerca the injections of blood or corpora cardiaca homogenates from ovipositing females as well as the implantation of the corpora cardiaca caused oviposition movements in virgin females (Highnam, 1962a, 1962b). Electrical stimulation of the central nervous system or enforced hyperactivity (both of which cause release of neurosecretory material (Delphin, 1965; Highnam and Haskell, 1964)) also induced oviposition (Highnam, 1962a). Besides the neurosecretory cells of the pars intercerebralis (Highnam, 1962a, 1962b), those in the glandular lobe of the corpora cardiaca (Highnam, 1962b) and in the third thoracic ganglion (Delphin, 1965) also seem to be involved. At this time the respective roles of the various neurosecretions in Schistocerca is unclear. The median neurosecretory cells of the mated female of Rhodnius prolixus similarly release a substance which influences oviposition (Davey, 1967). In the walking sticks, Carausius morosus and Clitumnus extradentatus, the control over oviposition is more complex. The neurosecretory cells of the pars intercerebralis have a role (Dupont-Raabe, 1952), but so also do those in the suboesophageal, thoracic, and first four abdominal ganglia (Thomas and Mesnier, 1973). In the above insects, as well as in Schistocerca (Highnam, 1962a), extracts of the active component of the nervous system have a direct myotropic effect on the ovarioles. Though this

I

329

HORMONES AND BEHAVIOUR

action can be mimicked by 5-hydroxy-tryptamine (Highnam, 1962a), the actual active agent is unknown. The switch in oviposition rate observed in mated H. cecropia females is mediated by a hormone which is apparently released from the intrinsic cells of the corpora cardiaca (Truman and Riddiford, 1971). An increased oviposition rate after mating was seen in both intact and allatectomized Cecropia females. But removal of both the CA and the CC abolished the response to mating (Fig. 6). As with the calling response of Cecropia (section 6.2), implantation o f CC-CA complexes into a cardiacectomized female did not restore this oviposition response. Thus it appears that the intrinsic cells of the CC also mediate the change in oviposition behaviour.

Days

Fig. 6. The effects of various experimental manipulations on the oviposition rates of H. cecropia females. A, mated females; B, mated allatectomized females; C, virgin females which received implants of a copulatory bursa from a mated female; D, mated allatectomized-cardiacectomized females; E, virgin females; F, virgin females which received implants of a copulatory bursa from a virgin female. The lin s are drawn through the averages of the respective groups. (Data redrawn from Tru an and Riddiford, 1971; Riddiford and Ashenhurst, 1973.)

L

The stimulus which provokes the switch to the mated oviposition rate is the presence of sperm. This was shown by the fact that matings t o castrated males, which resulted in the deposition of a spermatophore devoid of sperm, were completely ineffective in eliciting the mated response (Truman and Riddiford, 1971). The manner by which the sperm exert their effect was then examined by implanting various portions of the reproductive tract from mated females into virgin females (Riddiford and Ashenhurst, 1973). Implantation of the spermatheca did not alter the. virgin behaviour. But when the bursa copulatrix was removed from mated females, emptied of its

330

JAMES W. TRUMAN AND LYNN M. RIDDIFORD

contents, and implanted, the recipients showed an oviposition pattern which was essentially identical to mated females. By contrast, control experiments which involved the implantation of the bursa copulatrix from virgin females were without effect (Fig. 6). Thus, an interaction of sperm with the bursa appears to cause the release of a humoral factor which eventually leads to the release of the oviposition-stimulating hormone from the CC. In Bombyx mori injections of blood from mated females caused increased oviposition (Mokia, 1941). Similarly, the blood from mated Cecropia females stimulated egg-laying when injected into virgin females (Riddiford and Ashenhurst, 1973). The blood injections did not elicit an immediate oviposition response, but rather caused a marked increase in the number of eggs deposited per night for the next one to two days. The effects of the blood were similar to those of the bursa implants except that they were transient. Most probably, the active agent in the blood was the factor which is secreted by the bursa of a mated female. This factor then presumably acts through the neuroendocrine system of the female to cause the release of the oviposition-stimulating hormone from the corpora cardiaca. The humoural involvement of the copulatory bursa in the mated response of Hyalophora cecropia is just one example of the mechanisms which are utilized by insects to bring about an increased oviposition respose. The crucial stimuli provided by the male and the ihternal relays in the female are many and varied. Some of these have been considered in the preceding section with regard to the onset of female refractoriness. The mechanisms which are involved in the loss of receptivity apparently also mediate the changes in oviposition behaviour. In mosquitoes an accessory gland substance from the male is responsible for an increased oviposition rate (Leahy and Craig, 1965) and also for the onset of refractoriness. In Drosophila a paragonial gland acidic peptide triggers a greatly increased oviposition rate when injected into virgin females (Garcia-Bellido, 1964; Merle, 1968; Chen and Biihler, 1970). Secretions from these same glands also function to turn-off female receptivity (section 6.3.3). A humoral relay mechanism which has not yet been considered is illustrated by Rhodnius (Davey, 1965, 1967). In this bug mating provokes increased oviposition. Implantation of seminal vesicles or accessory glands from unmated males had no effect on the egg-laying behaviour of virgin females. Though spermathecae from virgin females were also ineffective, those from mated females provoked the full oviposition response when implanted into virgin females. All evidence indicates that the sperm from the male acts on the spermatheca to cause the release of a humoral substance which “tells” the female that she has been mated. This agent then acts on the median neurosecretory cells of the brain to cause the

HORMONES AND BEHAVIOUR

331

synthesis and release of a hormone which has myotropic action on the ovaries (Davey, 1967). This neurosecretory hormone apparently also has direct behavioural effects.

6.3.5 Summary The studies considered in the first four parts of this section show that the effects of hormones on reproductive behaviaur are many and varied. In many species only fragmentary information is available. Moreover, some of the larger insect orders such as the Homoptera, Coleoptera, and Hymenoptera have received, at best, scant attention with regard to endocrine effects on reproductive behaviour. Among the examples given above, there appears to be some conservatism of certain hormonal mechanisms. But, in some cases there are already exceptions noted and in others exceptions will undoubtedly arise as more groups are studied. This section attempts t o focus on the few generalizations that can be made concerning the role of hormones in the sexual behaviour of insects. The story of hormonal involvement in male hehaviour is relatively simple (Fig. 7). In most insects the corpora allata are not involved in the

Fig. 7. Schematic representation of the hormonal influences which m a y be involved in the regulation of reproductive behaviour in a “generalized” insect. (Top) Male behaviour. (Bottom) Female behaviour. Modifier effects occur along the horizontal axis; releaser effects along the vertical axis.

332

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

maturation of adult behaviour. In these instances the onset of sexual behaviour is developmentally programmed into the nervous system so that it matures coincidentally with adult development. Other insects rely on the corpora allata t o cause the appearance of the proper behaviour. The effect of JH is undoubtedly a modifier effect-the hormone lowers the behavioural threshold of the male with respect to his response to the female. Besides JH, a neurosecretory hormone from the brain also appears t o be important. The exact behavioural role of this hormone is unclear, but its action may be analogous to the apparent function of the phallic nerve-stimulating hormone of cockroaches, i.e. it may be secreted during the time of courtship to trigger (or to facilitate) the performance of copulatory behaviour. Such a neurosecretory link in the chain of events which lead to copulation demands that a certain amount of time has t o be spent in courtship behaviour before titres are high enough t o permit copulation. This would then ensure that sufficient time was expended for correct mate recognition. Alternatively, the neurosecretory hormone might function to establish the high level of excitability which the male must maintain during prolonged courtship. The role of these neurosecretory hormones in the male remains a major question to be answered. The behaviour of the female is more complex than that of the male and typically is comprised of three distinct behavioural states: (1) the young virgin which is unreceptive to courtship; (2) the mature virgin which is responsive to the advances of the male; ( 3 ) the mated female which again displays refractory behaviour and which also shows intense oviposition behaviour (Fig. 7). In virgins the switch from non-receptivity to receptivity appears to be an abrupt switch (e.g. in Drosophilu (Manning, 1967)). Typically, receptivity is due to the action of the corpora allata and depends upon the continued presence of JH. As with the male, the action of JH in the female is undoubtedly a modifier action. A neurosecretory hormone is also apparently required for the expression of receptivity, but the nature of the effects of this hormone is unclear. One possibility is that it serves to trigger part of the copulatory behaviour. Thus, in a manner similar t o that proposed for the male, it would thereby ensure that sufficient time had been spent in courtship before mating could then occur. Mechanisms which are involved in the switch from the virgin to the mated modes are extremely varied and defy generalization. In many insects refractoriness arises from the withdrawal of juvenile hormone; in others a factor from the maturing ovaries appears to be involved. The signal that mating has occurred is most commonly provided by the secondary reproductive structures-the accessory glands of the male or the spermatheca and bursa of the female. These structures typically release humoral factors which probably exert their primary effects through the neuro-

HORMONES AND BEHAVIOUR

333

endocrine system of the female. Some of these factors presumably also have direct behavioural effects. The one aspect of the behaviour of mated females which appears to be relatively consistent is the association of a neurosecretion from the brain with the onset of oviposition. This hormone has direct myotropic action on the musculature of the ovarioles, but also has behavioural effects which presumably arise from its action on the nervous system. The same agent controls both the movement of eggs out of the ovariole and the behaviour involved in depositing these eggs in a suitable spat. Therefore, in some male and most female insects, hormones are intimately associated with adult behaviour. Juvenile hormone typically acts in a tonic fashion to set the behavioural mode which is appropriate to the reproductive condition of the adult. Certain neurosecretory hormones then appear to be associated with the performance of specific behavioural acts. They may have a phasic action or, in response to suitable environmental cues, they may exert a tonic change in behaviour which may last for a number of hours. 6.4 HORMONAL INFLUENCES ON MIGRATION AND ORIENTATION

Certain environmental conditions provoke migration of some insects from their old habitat to a new one. This move may be of single individuals over small distances, such as the movement of aestivating insects from their host plant into the soil. Or it may consist of the mass movements of insects over hundreds of miles as is seen for locusts. Irrespective of the extent of the migration, the endocrine state of the insecl appears to be of great importance. 6.4.1 Gregarious and migratory behaviours in 1oi:usts Most studies of hormonal influences o n migratory behaviour have been focused on the migratory locusts, Schistocerca gregaria and Locusta migratoria migratorioides. When reared in isolation, the hoppers of these species show a moderate level of activity; the adults tend to be solitary and usually do not perform long sustained flights. Crowding of young locusts results in highly active, gregarious hoppers which display marching behaviour and which show rapid oriented locomotion in response to certain stimuli. The adults are gregarious and undertake long sustained flight uohnson, 1969). An indication of an endocrine basis for the differences in locomotor activity observed in the two phases was provided by the observation that solitary hoppers have larger prothoracic glands than do gregarious hoppers (Carlisle and Ellis, 1959). Moreover, in the adult the prothoracic glands persist in the solitary form but rapidly disappear in gregarious locusts.

334

JAMES W. TRUMAN AND LYNN M. RIDDIFORD

When prothoracic gland homogenates were injected into gregarious hoppers, marching was decreased (Carlisle and Ellis, 1963), thus supporting the hypothesis of a possible role for these glands in the control of activity. Similarly, implantation of prothoracic glands from solitary adults into gregarious adults also decreased the duration of sustained flight of the recipient (Michel, 1972). But in neither study was the decrease striking (50 per cent at most) nor was it comparable to the marked difference in locomotor activity exhibited by the two phases. Since activity is a labile behaviour which can be influenced by many factors, including injury due to injection or implantation of foreign tissues, one cannot be confident that the above effects were specifically due to ecdysone. Another line of evidence has been supplied by the experiment of Haskell and Moorhouse (1963). As described above (section 6.1.1), they demonstrated that ecdysone reversibly reduced the spontaneous motor output from the metathoracic ganglion of Schistocerca adults. Although this study is more convincing than the previous ones, it still does nqt clarify the relative importance of ecdysone in the regulation of the activity of locusts. What is needed is a demonstration thai gregarious behaviour can be produced in solitary individuals by partial or complete extirpation of the prothoracic glands (an operation which is feasible in locusts). The corpora allata have also been implicated in the control of locust activity, but again the data are not complete. Cassier (1963: 1964a, 1964b) first showed that CA implants decreased the latency of the phototropic response shown by gregarious Locusta, and led to an increase in the speed of movement. In Schistocerca males, Odhiambo (1965, 1966c) then demonstrated that allatectomy severely decreased spontaneous locomotor activity. In addition, these allatectomized insects showed abnormally large deposits of glycogen and lipids in the fat body. Odhiambo (1965) hypothesized that JH enhanced locomotor activity, as well as sexual activity, by a direct action on the nervous system and that the increased fat and glycogen reserves resulted from the persistent inactivity of the operated males. This hypothesis was partially confirmed by the demonstration that allatectomized Locusta which were forced t o exercise 2 h each day deposited less fat than did unexercised, allatectomized controls (Strong, 1968a). But Strong (1968a, 1968b) also reported that, unlike Schistocerca, Locusta did not show any obvious reduction in locomotor activity after removal of the CA. Twenty-four-hour measurements of activity failed to indicate a difference between operated and control groups. Wajc and Pener (1971) subsequently tested the performance of allatectomized Locusta males on a flight roundabout. The allatectomized locusts consistently flew 66 less intensely” than did sham-operated controls. The difference in activity levels was significant but not very pronounced. Therefore, in Locusta

HORMONES AND BEHAVIOUR

335

allatectomy decreases locomotor activity, but the reduction is relatively small as compared to Schistocerca. Strong (1968a) suggested that the reported effects of allatectomy on locomotor activity were simply the result OF the reduction in sexual behaviour. Since caged locusts spent most of their time in sexual activity, he concluded that operations which inhibited this behaviour would also appear to reduce locomotor behaviour. This hypothesis was disproved by Wajc and Pener (1971) who showed that flight, a behaviour not related to sex, was also diminished by allatectomy. However, the converse of Strong’s hypothesis was not considered. As described above, Odhiambo (1966a) concluded that JH specifically unmasked neural circuitry which was involved in courtship and mating. JH also acts to maintain a high level of general activity (Odhiambo, 1966c; Wajc and Pener, 1971). But it is possible that the CA have no direct role in the development of sexual behaviour. Indeed, in the allatectomized locusts sexual behaviour may be fully developed but siinply not expressed because of the sluggishness and inactivity caused by allatectomy. The significance of the control of normal locomotor activity by juvenile hormone is somewhat unclear. In larval locusts the green pigmentation of the integument is induced by JH (Staal, 1961). Yet the green solitary hoppers, which have a higher JH titre, nevertheless have a lower level of activity than do the black gregarious forms. In the adults the marked differences between the effects of allatectomy i n Schistocerca and those of the same operation in Locusta raise a question as to the relative role of the CA in the normal physiology of locomotion. Most probably, both JH and ecdysone do play some role in the modulation of the level of activity. But an attempt t o explain the complex changes which occur during phase determination simply in terms of concentration changes of these two hormones is still premature and, indeed, may be too simplistic an approach.

6.4.2 Migration and behaviour In his work on insect migration, Johnson (1969) put forth a model in which adult migration was triggered when ecdysone was absent and the JH titre was low. The rise in JH titre then caused both the cessation of migratory behaviour and the onset of oogenesis. This model was recently tested in the milkweed bug, Oncopeltus fasciatu.~.In this bug the migratory behaviour of the adult is determined primarily by the photoperiod to which it was exposed during larval life (Dingle, 1968). Long-day reared Oncopeltus show little or no tendency to make sustained flights, and oogenesis begins in the female relatively soon after adult ecdysis. By contrast, short-day bugs typically undertake long sustained flights, and the onset of reproduction is delayed.

336

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

Rankin (1973) examined the role of the CA in controlling migratory behaviour in Oncopeltus. The migratory status of the bugs was determined by the percentage of animals which performed sustained, tethered flight (Dingle, 1965). When a JH mimic was applied to either long-day or short-day males, and the insects tested three days later, a greater proportion showed sustained flight than did controls of comparable age. In a parallel experiment, the implantation of three CA significantly increased the number of males which made sustained flights. JH initially appeared to have different effects in the female (Rankin, 1973). When applied to 5-day-old females, the JH mimic did not stimulate flight behaviour. When it was administered t o very young (2-day-old) females, a significant but transient increase in flight activity was observed. Since the applied JH caused the rapid onset of oogenesis in the day 3 females, these results were not surprising because migratory behaviour in Oncopeltus is terminated with the onset of oogenesis. This explanation was then tested using ovariectomized females. The operated females responded to application of JH mimics and to CA implants in a manner which was essentially identical to that seen in males. Thus, in Oncopeltus the onset of migratory behaviour is apparently caused by high titres of JH. In the females migratory behaviour is later shut off in response to the JH-induced maturation of the ovaries. The nature of this effect of the ovaries is unknown. A slightly different control is observed in the Colorado beetle, Leptinotarsa decernlineata. Short-day photoperiods promote reproductive diapause and migration of Leptinotarsa adults from the host plant into the soil. This “soil-positive” behaviour characteristic of diapause could also be induced in long-day beetles by allatectomy (de Wilde and de Boer, 1961). Active CA implants restored normal “soil-negative” behaviour to these allatectomized long-day beetles. But CA implants (up t o 5 to 6 active glands) into short-day beetles did not counteract the diapause syndrome and soil-positive behaviour persisted (de Wilde and de Boer, 1961). Only massive implants of 11 to 12 active CA served to reverse the effects of short days, but only temporarily (de Wilde and de Boer, 1969). The difficulty in restoring “soil-negative” behaviour in short-day beetles was explained by the fact that the activity of the CA is maintained by a neurosecretion from the brain which is only released under long-day conditions (de Wilde and de Boer, 1969). After implantation into a short-day host, the active glands quickly become inactivated because of the absence of this neurosecretion; therefore, the diapause condition is maintained. Unlike Oncopeltus, the ovaries do not influence the behaviour of Leptinotarsa (de Wilde and de Boer, 1969). Long-day females which were ovariectomized nevertheless continued to show “soil-negative behaviour”. Also, allatectomized and ovariectomized females switched from “soil-

HORMONES

A N D BEHAVIOUR

337

positive” to “soil-negative” behaviour after implantation of active CA, suggesting that the action of JH is not mediated through its effects on the ovaries. Effects of JH o n orientation have also been reported in other insects. As described above, positive geotaxis is induced in larvae of Mimas.tiliae by low JH titre (Piepho et al., 1960). Also, in the scarab beetle, Mefolontha mefolontha, implantation of active female CA intl3 migratory males reverses the compass direction of flight (.Stengel and Schubert, 1972). Both JH and ecdysone influence the sign of phototaxis in thr: caterpillar of Smerinthus ocellata (Beetsma et al., 1962).

6.5

HORMONES AND INSECT CIRCADIAN RHYTHMS

As in most other organisms, insects possess internal clocks which can be synchronized and entrained by daily environmental periodicities (Biinning, 1967). In the absence of external cues these clocks continue to run, but with a periodicity which is slightly different from 24 h. Thus, they are “circadian” clocks. These clocks have been widely implicated in the control of many diverse types of behaviour. Generalized behaviour such as locomotor activity is influenced by a complex array of factors. In this instance the behaviour has numerous potential functions and therefore is used in various environmental contexts (e.g. escape, food-gathering, mate-finding, etc.). Consequently, the daily stimulus provided by the clock must interact with a network of diverse external and internal stimuli w.hich also influence locomotion. A simpler type of control is often seen with certain specialized behaviours. These are stereotyped behaviour patterns which are displayed under restricted environmental or developmental contexts. The performance of the behaviour is often directly coupled to the output of the clock with little or no influence from other stimuli. In many instances the coupling between the driving circadian clock and the behaviour appears to be hormonal. These will be only briefly considered here because there are recent reviews on the subject (Brady, 1969,1973; Truman, 1971b). 6.5.1 Circadian rhythms of locomotion The most extensive studies of the hormal involvement in insect activity rhythms have utilized the cockroach as an experimental subject. The earliest work was that of Harker (1956, 1960) who examined the role of the suboesophageal ganglion in the control of this rhythm. According to Harker, decapitation rendered cockroaches arrhythmic, but the implantation of a suboesophageal ganglion into the headless insect restored

338

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

rhythmicity. Importantly, the phase of the new rhythm was similar to the prior rhythm of the donor of the ganglion, but not to the rhythm that had previously characterized the host. Thus, the ganglion was responsible for the intrinsic timing of activity and did so through a humoral pathway. Subsequent workers were unable to confirm the central importance of the suboesophageal ganglion but, instead, directed attention to the brain as the site of the driving clock (see Brady, 1969, for review). Again, in the case of the brain, the experiments of Nishiitsutsuji-Uwo and Pittendrigh (1968) indicated a hormonal mediation of the rhythm. They reported that the transection of the circumoesophageal connectives (thereby severing the nervous connections between brain and thorax) did not influence the cockroach activity rhythm. This experiment was subsequently challenged by Roberts et al. (1971) and Brady (1969) on the basis that the surgical procedures which were described led t o the cutting of the labial nerves instead of the circumoesophageal connectives. Severance of the latter led to hyperactivity with no indication of rhythmicity (Roberts et al., 1971). This latter result led Brady (1969) to conclude that hormones were not involved in this response and that the rhythm was driven throughstrictly neural pathways. The data presented for the cockroach do not support a hormonal driving of the rhythm, but also do not exclude it. Cymborowski and Brady (1972) have recently shown that when a legless cockroach was parabiosed to the back of a decapitated animal which had its legs, the headless cockroach showed a distinct locomotor rhythm. The authors are cautious in their interpretation, however, because a small number of the control pr,eparations which did not have a common blood supply nevertheless showed the establishment of a locomotor rhythm in the decapitated partner. The involvement of neurosecretory hormones in driving the cockroach activity rhythm is still unresolved and remains an active question. Klug (1958) described daily changes in nuclear size of cells of the CA in a carabid beetle, and Biinning and Joerrens (1962) reported daily changes in gland volume in larval Pieris brassica. However, there is no direct evidence that the CA are involved in the mediation of locomotor rhythms. In cockroaches they can be removed without affecting the rhythm (Roberts, 1966). Even in locusts, a group in which the CA has been shown to influence the level of activity, allatectomy does not disturb the daily distribution of the residual locomotion (Odhiambo, 1 9 6 6 ~Strong, ; 1968b). In the noctuid moth, Noctua pronula, microcautery of the median neurosecretory cells abolished the nocturnal circadian component of the adult flight rhythm (Hinks, 1967). Injections of serotonin extended the duration of flight activity in normal individuals. Furthermore, high amounts of tryptophan-positive material were found to be present in the A-cells of the median neurosecretory cluster. Hinks (1967) therefore

HORMONES AND BEHAVIOUR

339

concluded that a portion of the flight rhythm was mediated or enhanced by secretion of serotonin from the A-cells. Clearly. further work is necessary to establish hormonal control of the moth flight rhythm. An indication of hormonal involvement in an arachnid activity rhythm was provided by work on the locomotor rhythm of the scorpion, Heterometrus fulvipes (Rao and Gropalakrishnareddy, 1967). Using a standardized nerve cord preparation, they studied the effects of blood and ganglion extracts on the level of electrical activi1.y. As seen in Fig. 8, blood or extracts of the cephalothoracic nerve mass which were obtained from animals from 5 or 8 p.m. (times when the scorpions were normally active) caused a marked increase in the rate of spontaneous electrical activity. By contrast, blovd or extracts sampled during the rest phase' (-2 a.m.) had inhibitory effects. O n this evidence, the authors suggest that activity in the scorpion is controlled by two factors, one which enhances locomotor activity and one which inhibits it. Both factors are then released at different times of day and thereby regulate thc animal's daily pattern of activity.

'

Fig. 8. Effect of blood ( 0 ) or cephalothoracic nerve mass extracts (A) on the electrical activity of the isolated ventral nerve cord of the scorpion. Tissues were collected at the times specified on the abscissa. The control values ( 0 ) give the activity of the preparation in Ringers after being tested with the respective extracts. (Data from Rao and Gropalakrishnareddy, 1967.)

6.5.2 Circadian control of specialized behaviours As mentioned above, specialized behaviours are performed under restricted environmental or developmental conditions. Some of these specialized behaviours such as pheromone release (Sower et al., 1970; Truman et al., unpublished, cited in Riddiford, 1973) may occur repeatedly at a specific

340

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

time on successive days. Other behaviours such as eclosion (Pittendrigh, 1966) or cocoon spinning (Lounibos, unpublished) occur but once in the lifetime of the indiyidual, but always during ayestricted period of the day. Although it may seem contradictory to speak of a circadian rhythm with respect to “one-shot” behaviour, an analysis of this behaviour in the context of a population clearly shows the underlying circadian control in the individual (see Pittendrigh, 1966). Some specialized behaviours such as “calling”, oviposition, and eclosion are apparently triggered by neurosecretory hormones and have been considered in detail above. The circadian organization of these behaviours presumably arises from the circadian release of the corresponding hormones. This has been shown in the case of silkmoth eclosion. As described above, silkmoths emerge during a species-specific time of day (Truman and Riddiford, 1970). Removal of the brain from silkmoth pupae yielded moths which showed abnormal eclosions that were randomly distributed throughout the day and night. When a brain was implanted into the abdomen of each debrained moth, the proper timing of the behaviour was restored. Moreover, when brains were interchanged between pupae of Hyalophora cecropia and Antheraea pernyi (two species which had very different eclosion times), the resulting moths emerged at the time of day which was characteristic of the species which donated the brain (Truman and Riddiford, 1970). Thus, the brain controls the time of release of the eclosion hormone which in turn activates the proper behaviour.

7 Conclusions A review on hormonal influences in behaviour is, by its very nature, biased towards the importance of hormones and away from purely neural mechanisms. Thus, one may be left with the impression that the nervous system functions only to feed sensory information into the endocrine system and to respond to the subsequent blood-borne commands. Obviously, behaviour cannot be explained solely in terms of circulating chemi6ry. But neither can one ignore the profound effects which hormones have on the functioning of the nervous system. Hormones are well suited for modulating the general responsiveness of the nervous system because they have access to all of the neurons. For a widespread change to be mediated through purely neural means requires an extensive series of excitatory and inhibitory networks. The extra neural circuitry will provide rapid and “fine-tuned” modulation of response thresholds, but it will also increase the complexity and size of the CNS. Hormonal modulation, b y contrast, apparently either facilitates or suppresses neural pathways according to the response characteristics of the

HORMONES AND BEHAVIOUR

341

particular neurons involved. The response levels change more slowly and probably cannot be as finely adjusted as with neural control. One then has a trade-off between speed of response and the complexity of the neural components. The insects with their size limitations apparently cannot afford the luxury of extensive neural modulation and have, instead, relied heavily on hormonal means. Behaviours involved in alarm and escape reactions require split-second responses that only nervous mechanisms can provide. But other behaviours such as mating and oviposition are typically trig,gered only after sufficient environmental information has been gathered tc assure a proper mate or a suitable oviposition site. The "decision" to rcspond often requires the summation of information over a time period which is measured in minutes. One possible mechanism for accomplishing this long-term summation could involve the release of hormones in response to the proper environmental stimulus. Since hormones typically have a longer half-life than neurotransmitters, successive releases could readily summate and trigger the appropriate behaviour. The manner by which hormones exert their effects on the nervous system is unknown. They may alter the permeability characteristics of the neurons, affect the level of transmitter syni.hesis, facilitate or block synaptic transmission, etc. Invertebrates, with their simplified nervous systems and their large, identifiable neurons, have proved to be excellent animals in which to answer many questions concerning the functioning of the nervous system. In regard to hormone-nerve interactions, insects seem especially attractive because of their rich and complex behaviour and because of the extensive involvement of hormones in regulating this behaviour. Insects have the additional benefit that among the invertebrates the chemistry of their hormones is by far the best known. Hormones such as ecdysone and juvenile hormone are available in pure form for experimental use. Therefore, from the standpoint of behaviour, the simplicity of the nervous system, and the extent of endocrinological knowledge, insects appear to be the most proniising animals in which to determine the mode of action of hormones on the nervous system. In order to provide systems for neurophysiological study, the behavioural effects of the hormone must be well defined. Thus far, many studies have dealt only with the end result (e.g. failure of the female to mate) without a detailed consideration of the behavioural alterations involved. This lack is especially evident in cases i n which two hormones are implicated in the occurrence of a particulu behavioural response. Typically, in these cases no attempt is made to (define the respective roles of the two hormones. Yet a clear knowledge of both the. behaviour involved as well as the endocrine basis is a prerequisite for a fruitful study of the neural effects of the hormone.

342

JAMES W. TRUMAN AND LYNN

M. RlDDlFORD

ACKNOWLEDGEMENTS

We are indebted t o A. M. Ajami, L. P. Lounibos, and H. F. Nijhout for a critical reading of the manuscript. We are also grateful to Mrs M. J. Randell for excellent secretarial assistance. The original work reported here was supported by the Harvard Society of Fellows and grants from the Rockefeller Foundation and NSF and was carried out at Harvard University. References Adams, T. S. and Hintz, A. M. (1969). Relationship of age, ovarian development, and the corpus allatum to mating in the house-fly, Musca domesticus. J. Insect Physiol. 15, 201-215. Adams, T. S. and Nelson, D. R. (1968). Bioassay of crude extracts for the factor that prevents second mating in female Musca domestica. Ann. ent. SOC.A m . 61, 112-116. Adiyodi, K. G. and Nayar, K. K. (1965). Some neuroendocrine aspects of reproduction in the viviparous cockroach Trichoblatta sericea (Saussure). Zool. Jb. (Physiol.), 72, 453-4 62. Amy, I.. Bounhiol, J. J. and Gabe, M. (1953). Dtrochement de la neuroskattion protocktbrale chez Bombyx mori L. au cows du dkeloppement post-embryonnaire. C.T.Acad. S C ~Paris, , 236, 627-629. Bagnoli, P., Brunelli, M., D’Ajello, V. and Magni, F. (1970). Further evidence for peripheral inhibition of flashing and for role of the male gonads in Luciola lusitanica (Charp). Arch. Ital. Biol. 108, 180-206. Bagnoli, P., Brunelli, M., Magni, F. and Viola, M. (1972). The identification of a flash-inhibiting substance from the male gonads of Luciola lusitanica (Charp). Arch. Ital. Biol. 110, 16-35. Barth, R. H., Jr. (1961). Hormonal control of sex attractant production in the Cuban cockroach. Science, 133. 1598-1599. Barth, R. H., Jr. (1962). The endocrine control of mating behavior in the cockroach Byrsotria fumigata (Gukin). Gen. comp. Endocr. 2, 53-69. Barth, R. H., Jr. (1965). Insect mating behavior: Endocrine control of a chemical communication system. Science, 149. 882-883. Barth, R. H., Jr. (1968). The comparative physiology of reproductive processes in cockroaches. Part I. Mating behavior and its endocrine control. Adu. Reprod. Physiol. 3, 167-201. Barth, R. H., Jr. and Bell, W. J. (1970). Physiology of the reproductive cycle in the cockroach Byrsotria fumigata (Gutrin). Biol. Bull. mar. biol. Lab., Woods Hole, 139, 447460. Barth, R. H. and Lester, L. J. (1973). Neurohormonal control of sexual behavior in insects. A . Rev. Ent. 18, 445-472. Bastock, M. and Blest, A. D. (1958). An analysis of behavior sequences in Automeris aurantiaca Weym (Lepidoptera). Behauiour. 12, 243-284. Benz, G. (1969). Influence of mating, insemination, and other factors on oogenesis and oviposition in the moth Zeimphera diniana. J. Insect Physiol. 15, 55-71. Beetsma, J., de Ruiter, L. and de Wilde, J. (1962). Possible influence of neotenine and ecdysone on the sign of phototaxis in the eyed hawk caterpillar (Smerinthus ocellata L.). J. Insect Physiol. 8, 251-257.

HORMONES AND BEHAVIOUR

343

Blest, A. D. (1960). The evolution, ontogeny, and quantitative control of settling movements of some new world saturniid moths, with some comments on distance communication by honey bees. Behaviour, 16, 188-253. Bounhiol, J. J. (1938). Recherches experimentales sur le determinisme de la metamorphose chez les Lepidopteres. Bull. biol. Fr. Befg. (Suppl.), 24, 1-199. Brady, J. (1969).How are insect circadian rhythms con:rolled? Nature 223, 781-784. Brady, J. (1973).The physiology of insect circadian rhythms. Adv. Insect Physiol. 10, 1-1 15. Brady, J. and Maddrell, S. H. P. (1967).Neurohaemll organs in the medial nervous system of insects. Z. Zeflforsch. 76. 389-404. Broza, M. and Peper, M. P. (1972).The effect of the corpora allata on mating behaviour and reproductive diapause in adult males of the grasshopper, Oedipoda miniata. Acrida, 1, 79-96. Brunelli, M., Buonamici, M. and Magni, F. (1968a). Mechanisms for pho\ic inhibition of flashing in fireflies. Arch. Ital. Biol. 106, 85-99. Brunelli, M., Buonamici, M. and Magni, F. (1968b).Effects of castration on the inhibition of flashing in fireflies. Arch. f t a f .Biol. 106, 100-112. Brunelli, M., Buonamici, M., Magni, F. and Viola, M . (1970).The role of the male gonads in the peripheral inhibition of flashing in Lirciola lusitanicu Arch. Itnl. Biol.

108, 1-20. Buchmann, D. (1959).Die Auslosung der Umfarbung durch das Hautungshormon bei Cerura uinula L. (Lepidoptera, Notodontidae).]. fns
Cantacuzhe, A.-M. (1968).Recherches morphologiques et physiologiques sur les glands annexes m&s des Orthopthes. 11. Les glandes annexes miles des Acridiens apris I’ablation des corps allates et au cows de la diapause imaginale. Bull. SOC. 2001. Fr.

93,545-557. Carlisle, D. B. and Ellis, P. E. (1959).La persistance des glands ventrales ckphaliques chez les Criquets solitaires. C.r. Acud. S c i Paris, 249, 1059-1060. Carlisle, D. B. and Ellis, P. E. (1963).Prothoracic glmd and gregarious behaviour in locusts. Nature, 200, 603-604. Carlson, A. D. (1969). Neural control of firefly luminescence. I n “Advances in Insect Physiology” (Eds J. W. L. Beament, J. E. Treherne m d V. B. Wigglesworth), Vol. 6, pp. 51-96.Academic Press, London and New York. Cassier, P. (1963). Action des implantations de corps allates s u r la rkactiviti phototropique de Locusta migatoria migratorwides (R. & F.), phase gregmia (Insecte Orthoptkroidc). C.r. Acad. Sci Paris, 257,4048-4049. Cassier, P. (1964a).La riactiviti phototropique de Locusta migratoriu m@ztorioides (R. et F.), phase grigaire, aprks implantation abdominale de corps allates. Insectes SOC.

11, 131-140.

AIP-I2

344

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

Cassier, P. (1964b). fitude et interpretation des effets i long terme, des implantations abdominales de corps allates sur la rkactiviti phototropique de Locustu migrntoriu migutorioides (R. et F.), phase grtgaire (Insecte Orthopteroide). C.r. Acud. Sci Paris, 258, 723-725. Chen, P. S. and Biihler, R. (1970). Paragonial substance (sex peptide) and other free ninhydrin-positive components in male and female adults of Drosophilu melunoguster. J. Insect Physwl. 16, 615-627. Craig, G. B., Jr. (1967). Mosquitoes: Female monogamy induced by male accessory gland substance. Science, 156, 1499-1501. Cymborowski, B. and Brady, J. (1972). Insect circadian rhythms transmitted by parabiosis-a re-examination. Nature, New Biology. 236. 221-222. Darjo, A. (1969). RBle des corpora allata dans le conditionnement photopiriodique de la maturation chez Locustu migratoriu L. souche “Kazalinsk.” C.r. hebd. Sknnc. Acad. S C 268D, ~ 337-340. Davey, K. G. (1965). Copulation and egg production in Rhodnius prolixus: the role of the spermathecae. J. exp. Biol. 42, 373-378. Davey, K. G. (1967). Some consequences of copulation in Rhodnius prolixus. J. Insect Physiol. 13, 1629-1636. Davidson, J. M. and Levine, S. (1972). Endocrine regulation of behavior. Ann. Rev. Physiol. 34, 375-408. Delphin, F. (1965). The histology and possible functions of neurosecretory cells in the ventral ganglia of Schistocercu greguriu Forskal (Orthoptera: Acrididae). Trans. R . ent. SOC. Lond. 117, 167-214. Dingle, H. (1965). The relation between age and flight activity in the milkweed bug, Oncopeltus. J. exp. Biol. 42, 269-283. Dingle, H. (1 968). Life history and population consequences of density, photoperiod, and temperature in a migrant insect, the milkweed bug Oncopeltus. A m . Nut. 102, 149-163. Dupont-Raabe, M. (1952). Contribution i l’itude du role endocrine du cerveau e t notamment de la pars intercerebralis chez les phasmides. Archs 2001. exp. gin. 89, 128-138. Dupont-Raabe, M. (1966). i t u d e des phenomenes de neurosecretion au niveau de la chaine nerveuse ventrale des phasmids. Bull. SOC. rool. Fr. 90, 631-654. Engelmann, F. (1960a). Hormonal control of mating behaviour in a n insect. Experientk, 16, 69-70. Engelmann, F. (1960b). Mechanisms controlling reproduction in two viviparous cockroaches (Blattaria). A n n N.Y. Acud. Sci 89, 516-536. Engelmann, F. (1970). “The Physiology of Insect Reproduction”. Pergamon Press, New York. Engelmann, F. and Barth, R. H., Jr. (1968). Endocrine control of female receptivity in Leucophueu maderue (Blattaria). A n n ent. SOC.A m . 61, 503-505. Finlayson, L. H. (1956). Normal and induced degeneration of abdominal musculature during metamorphosis in the Lepidoptera. Q. JL microsc. S c i 97, 215-234. Foster, W. (1967). Hormone-mediated nutritional control of sexual behavior in male dung flies. Science, 158, 1596-1597. Fuchs, M. S., Craig, G. B., Jr. and Despommier, D. D. (1969). The protein nature of the substance inducing female monogamy in Aedes uegypti J. Insect Physiol. 15. 701-709.

HORMONES AND BEHAVIOUR

345

Fuchs, M. S. and Hiss, E. A. (1970).Partial purification and separation of the protein components of matrone from Aedes aegypti J . Insect Physiol. 16, 931-940. Fukuda, S. (1944).The hormonal mechanism of larval molting and metamorphosis in the silkworm. J . Fac. Sci T o k y o Uniu. sec. IV, 6,477-532. Fuzeau-Braesch, S. (1972).Pigments and color change. A . Rev. Ent. 17, 403-424. Garcia-Bellido, H. von (1964).Das Sekret der Paragonien als Stimulus der Fekunditat bei Weibchen von Drosophila melanogaster. Z. NatzrJ 19, 491-495. Geldiay, S. (1967).Hormonal control of adult reproductive diapause in the Egyptian grasshopper, Anacridium aegyptium L. J. Endocr. 37, 63-71. Geldiay, S. (1970).Photoperiodic control of neuroiecretory cells in the brain of the Egyptian grasshopper, Anacridium aegyptium L.Cen. comp. Endocr. 14,35-42. Gersch, M. (1962). The activation hormone of the mepmorphosis of insects. Cen. comp. Endocr. (Suppl.), 1, 327-329. Gersch, M. and Richter, K. (1963). Auslosung von Nervenimpulsen durch ein Neurohormon bei Periplaneta americana. Zool. Jb. (Physiol.), 70,301-308. Girardie, A. (1966).Controle de 1’activitC gknitale chez Locusta m$ratoria. Mise en hidence d’un facteur gonadotrope et d’un facteur allatrope dans la pars intercerebralis. Bull. SOC. Zool. Fr. 91,423-439. Girardie, A. (1970).NeurosCcrCtions ckibrales chez les Acridiens. Bull. SOC. zool. Fr.

95, 783-802. Girardie, A. and Vogel, A. (1966). fitude du contrble neurohumoral de I’activitC sexuelle mile de Locusta migratoria (L.). C.r. herid Skanc. Acad. Sci.. Paris, 263D,

543-546. Gwadz, R. W. (1972).Neuro-hormonal regulation of sexual receptivity in female Aedes aegypti. J . Insect Physiol. 18, 259-266. Gwadz, R. W. and Craig, G. B. (1968). Sexual receptivity in female Aedes aegypti Mosquito News, 28, 586-593. Gwadz, R. W. and Craig, G. B. (1970). Female polygamy due to inadequate semen transfer in Aedes aegypti Mosquito News, 30, 355-360. Gwadz, R. W., Craig, G. B., Jr. and Hickey, W. A. (1971a). Female sexual behavior as the mechanism rendering Aedes aegypti refractory to insemination. Biol. Bull. mm. biol. Lab., Woods Hole, 140, 201-214. Gwadz, R. W. Lounibos, L. P. and Craig, G. B., Jr. (1971b). Precocious sexual receptivity induced by a juvenile hormone analog in females of the yellow fever mosquito, Aedes aegypti Cen. comp. Endocr. 16,47-51. Harker, J. E. (1956). Factors controlling the diurnal rhythm of activity in Periplaneta americana L. J. exp. Bwl. 33, 224234. Harker, J. E. (1960).Endocrine and nervous factors in insect circadian rhythms. Cold Spring Hmb. Symp. quant. Biol. 25, 279-287. Hartmann, R. (1971). Der Einfluss endokrinen Falrtoren auf die mannlichen Akzessorischen Driisen und die Ovarien bei der Keulenheuschrecke Comphocerus rufus L. (Orthoptera, Acrididae). Z. uergl. Physbl. 74, 190-216. Hartmann, R.,Wolf, W. and Loher, W. (1972).The influence of the endocrine system o n reproductive behavior and development in LTasshoRpers. C e n comp. Endocr. Suppl. 3, 518-528. Haskell, P. T. (1960).Stridulation and associated behaviour in certain Orthoptera: the influence of the gonads. Anim. Behau. 8, 76-81.

346

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

Haskell, P. T. and Moorhouse, J. E. (1963). A blood-borne factor influencing the activity of the central nervous system of the desert locust. Nature, 197, 56-58. Herman, W. S. and Gilbert, L. I. (1965). Multiplicity of neurosecretory cell types and groups in the brain of the saturniid moth Hyalophora cecropia (L.). Nature, 205, 926-927. Highnam, K. C. (1962a). Variations in neurosecretory activity during oocyte development. J . Endocr. 24, IV-V. Highnam, K. C. (1962b). Neurosecretory control of ovarian development in Schistocerca gregarin. Q. Jl. microsc. Sci. 103, 57-72. Highnam, K. C. (1964). Hormones and behaviour in insects. Viewpoints in Biology, 3, 219-260. Highnam, K. C. (1965). Some aspects of neurosecretion in arthropods. Zool. J b . (Physiol.). 71. 558-582. Highnam, K. C. and Haskell, P. T. (1964). The endocrine systems of isolated and crowded Locusta and Schistocerca in relation to oocyte growth, and the effects of flying upon maturation. J . Insect Physiol. 10, 849-864. Hinde. R. A. (1969). “Animal Behaviour”, 2nd ed. McGraw-Hill, New York. Hinks, C. F. (1967). Relationship between serotonin and the circadian rhythm in some nocturnal moths. Nature, Lond. 214, 386-387. Hintze, C. (1968). Histologische Untersuchungen iiber die aktivitat der inkretorischen Organe von Cerura vinulu L. (Lepidoptera) wahrend der Verpuppung. Wilhelm Roux Arch. EntwMech. Org. 160, 313-343. Hintze-Podufal, C. and Fricke, F. (1971). The effect of farnesol derivatives o n the mature larva of Cerura uinula L. (Lepidoptera). J. Insect Physiol. 17, 1925-1932. Hodgson, E. S. and Wright, A. M. (1963). Action of epinephrine and related compounds upon the insect nervous system. G e n comp. Endocr. 3, 519-525. Husain, M. A. and Baweja, K. D. (1936). Studies o n Schistocerca gregariu Forsk. IV. Colour changes and sexual behaviour in desexualized Schistocerca greguria adults. Ind. J . agric. Sci 6 , 586-590. Jacobson, M. (1972). “Insect Sex Attractants”. John Wiley, New York. Johnson, C. G. (1969). “Migration and Dispersal of Insects by Flight”. Methuen, London. Klatt, B. (1920). Beitrage zur Sexualphysiologie des Schwammspinners. Biol. Z . 40, 539-558. Klug, H. (1958). Neurosekretion und Aktivitatsperiodik bei Carabiden. Natunuissenschaften, 45, 141-142. Komisaruk, B. R. (1971). Strategies in neuroendocrine neurophysiology. A m . ZooL 11, 741-754. Lashley, K. S. (1938). Experimental analysis of instinctive behavior. Psychol. Rev. 45, 445-471. Law, J. H. and Regnier, F. E. (1971). Pheromones. A . Rev. Biochem. 40,533-548. Lea, A. 0. (1968). Mating without insemination in virgin Aedes aegypti. J. Insect Physiol. 14, 305-308. Leahy, M. G. and Craig, G. B., Jr. (1965). Accessory gland substance as a stimulant for oviposition in A e d e s aegypti and A. albopictus. Mosquito News, 25, 448-45 2. Leopold, R. A., Terranova, A. C., Thorson, B. J. and Degrugillier, M. E. (1971a). The biosynthesis of the male housefly accessory secretion and its fate in the mated female. J. Insect Physiol. 17, 987-1003.

HORMONES AND BEHAVIOUR

347

Leopold, R. A., Terranova, A. C. and Swilley, E. M. (1971b). Mating refusal in Musca domesticn: Effects of repeated mating and decerebration upon frequency and duration of copulation. J. exp. Zool. 176, 353-360. Libby, J. L. (1961). The nervous system of certain abdominal segments and the innervation of the male reproductive system and genitalia of Hyalophora cecropia (Lepidoptera: Saturniidae). Ann. ent. SOC. Am. 54, 887-896. Lisk, R. D. (1967). Sexual behavior: hormonal control. In “Neuroendocrinology” (Eds L. Martini and W. F. Ganong), Vol. 2, pp. 163-196. Academic Press, New York and London. Lockshin, R. A. (1969). Programmed cell death. Activation of lysis by a mechanism involving the synthesis of protein. J. Insect Physiol. 15, 1505-1516. Lockshin, R. A. and Williams, C. M. (1965a). Programmed cell death. I. Cytology of degeneration in the intersegmental muscles of the Prmyi silkmoth. J. Insect Physiol. 11, 123-133. Lockshin, R. A. and Williams, C. M. (1965b). Programmed cell death. 111. Neural control of the breakdown of the intersegmental muscles of the silkmoths. J. Insect PhysioL 11, 601-610. Lockshin, R. A. and Williams, C. M. ( 1 9 6 5 ~ )Programnied . cell death. IV. The influence of drugs on the breakdown of the intergegmental muscles of silkmoths. J. Insect Physiol. 11, 803-809. Loher, W. (1961). The chemical acceleration of the maturation process and its hormonal control in the male of the desert locust. Proc. R. SOC. B 153, 380-397. Loher, W. (1962). Die Kontrolle des Weibchengesanges von Comphocerus rufus L. (Acridinae) durch die Corpora allata. Natunuissenschctften, 49, 406. Loher, W. (1966). Die Steuerung sexueller Verhaltensweisen und der Oocytenentwicklung bei Comphocerus rufus L. Z . vergl. Physiol. 53, 277-316. Loher, W. and Huber, F. (1964). Experimentelle Untersuchungen am Sexualverhalten des Weibchens der Heuschrecke Comphocerus rufus L. (Acridinae). J . Insect Physiol. 10, 13-36. Loher, W. and Huber, F. (1966). Nervous and endocrine control of sexual behavior in a grasshopper (Comphocerus rufus L., Acridinae). SOC. Exp. B i d . Sympos. 20, 381-400. Liischer, M. (1962). Social control of polymorphism in termites. In “Insect Polymorphism” (Ed. J. S. Kennedy), Symp. No. 1, R . Enl. SOC. London, 57-67. Magni, F. (1967). Central and peripheral mechanisms in the modulation of flashing in the firefly Luciola itulica L. Arch. Ital. Biol. 105, 339-360. Manning, A. (1962). A sperm factor affecting the receptivity of Drosophila melanogaster females. Nature, 194. 252-253. Manning, A. (1966). Corpus allatum and sexual receptivity in female Drosophila melunogaster. Nature, 211, 1321-1322. Manning, A. (1967). The control of sexual receptivity in female Drosophilu Anim. Behav. 15, 239-250. Merle, J. (1968). Fonctionnement ovarien et riceptivitt sexuelle de Drosophila melunogaster a p r b implantation de fragments de I’appareil gtnital mile. J. Insect Physiol. 14, 1159-1168. Michel, R. (1972). 6tude expirimentale de I’influence des glandes prothoraciques sur l’activiti de vol du Criquet Ptlerin Schistocerca grvgariu C e n comp. Endocr. 19, 96-101.

348

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

Milburn, N. S. and Roeder, K. D. (1962). Control of efferent activity o n the cockroach terminal ganglion by extracts of the corpora cardiaca. Gen. comp. Endocr. 2, 70-76. Milburn, N., Weiant, E. A. and Roeder, K. D. (1960). The release of efferent nerve activity in the roach, Periplaneta americana. by extracts of the corpus cardiacum. Biol. Bull. mar. biol. Lab., Woods Hole, 118, 111-119. Mokia, G. G. (1941). Contribution to the study of hormones in insects. C.r. (Doklady) Acad. Sci. URSS, 30, 371-373. Miiller, H. P. (1965). Zur Frage der Steuerung der Paarungsverhaltens und der Eireifung bei der Feldheuschrecke, Euthystira brachyptera Osck., unter besonderer Beriicksichtipng der Rolle der corpora allata. 2. vergl. Physiol. 50, 447-497. Nayar, K. K. (1958). Studies o n the neurosecretory system of lphita limbata Stahl. V. Probable endocrine basis of oviposition in the female insect. Proc. Indian Acad. S c i B 47, 233-251. Nishiitsutsuji-Uwo, J. and Pittendrigh, C. S. (1968). Central nervous system control of circadian rhytbmicity in the cockroach. 111. The optic lobes, locus of the driving oscillation? Z. vergl. Physiol. 58, 14-46. Nopp-Pammer, E. and Nopp, H. (1968). Gehirnhormon und Spinnverhalten bei Philosamia Cynthia Dru. Verh. d t . zool. ges. 33, 508-519. Norris, M. J . (1962). Group effects on the activity and behaviour of adult males of the desert locust (Schistocerca gregaria Forsk.) in relation to sexual maturation. Anim. Rehav. 10,275-291. Odhiambo, T. R. (1965). Metabolic effects of the corpus allatum hormone in the desert locust, Schistocerca grrgaria. Nature, 207, 13 14- 13 15. Odhiambo, T. R. (1966a). Growth and the hormonal control of sexual maturation in the male desert locust, Schistocerca ~qegariu(Forsk.). Trans. R . ent. SOC.Lond. 118, 393-412. Odhiambo, T. R. (1966b). Ultrastructure of development of the corpus allatum of the adult male of the desert locust.]. Insect Physiol. 12, 995-1002. Odhiambo, T. R. ( 1 9 6 6 ~ ) The . metabolic effects of the corpus allatum hormone in the male desert locust. 11. Spontaneous locomotor activity. J. exp. Biol. 45, 51-63. Ozbas, S. and Hodgson. E. S. (1958). Action of insect neurosecretion upon central nervous system in vitro and upon behavior. Proc. natn. Acad. Sci. USA, 44, 825-830. Pammer, E. (1966). Auslosung und Steuerung des Spinnerhaltens und der Diapause bei Philosurniu Cynthia Dru. (Saturniidae, Lep.). Z. v r y l . Physiol. 53, 99-1 13. Pener, M. P. (1965). On the influence of corpora allata on maturation and sexual behaviour of Schistocerca .greguria. J. Zool. Lond. 147, 119-136. Pener, M. P. (1967). Effects of allatectomy and sectioning of the nerves of the corpora allata on oocyte growth, male sexual behavior, and colour change in adults of Schistocerca RTegaria J. Insect Physiol. 13, 665-684. Pener, M. P. (1968). The effect of corpora allata on sexual behaviour and “adult diapause” in males of the red locust. Ent. exp. A p p l . 11, 94-100. Pener, M. P. (1970). The corpus allatum in adult acridids: the interrelation of its functions and possible correlations with the life cycle. Proc. f n t . Study Conf. Current and Future Problems of Acridology, London, 135-147. Pener, M. P. and Wajc, E. (1971). The effect of allatectomy on male sexual behaviour in adult Schistocerca gregaria and Locusta migratoria migratorioides. In “Insect Endocrines” (Eds V. J. A. Novak and K. Slima). Suppl. A c t a ent. Bohemo. 37-43. Academia, Praha, Czechoslovakia.

HORMONES AND BEHAVIOUR

349

Pener, M. P., Girardie, A. and Joly, P. (1972). Neurosecretory and corpus allatum controlled effects o n mating behavior and color change in adult Locusta migratoria migratorioides males. G e n comp. Endocr. 19, 494508. Perez, Y., Verdier, M. and Pener, M. P. (1971). The effect of photoperiod o n male sexual behavior in a North Adriatic strain of the migratory locust. Ent. exp. AppL 14, 245-250. Piepho, H. (1950). Hormonale Grundlagen der Spinnta tigkeit bei Schmetterlingsraupen. Z. Tierpsychol. 7, 424434. Piepho, H. (1960). Hormonal control of moulting bchavior and scale development in insects. Ann. N. Y.Acad. Sci 89, 564-571. Piepho, H., Boden, E. and Holz, I. (1960). iiber die Hormonabhangigkeit des Verhaltens von Schwirmeraupen vor den Hiiutungen. Z. Tierpsychol. 17, 261-269. Pittendrigh, C. S. (1966). The circadian oscillation in Drosophila pseudoobscura pupae: a model for the photoperiodic clock. Z. Pflanzenphysiol. 54, 275-307. Priesner, E. (1968). Die interspezifischen Wirkungen der Sexuallockstoffe der Saturniidae (Lepidoptera). Z. vergl. Physiol. 61, 263-297. Rankin, M. A. (1973). The hormonal control of flight in the milkweed bug, Oncopeltus fasciatus. In “Experimental Analysis of Insect Behaviour” (Ed. L. Barton Browne). (Inpress.) Springer-Verlag, Berlin. Rao, K. P. and Gropalakrishnareddy, T. (1967). Blood borne factors in circadian rhythms of activity. Nature, 213, 1047-1048. Regen, J. (1909). Kastration und ihre Folgeerscheinunigen bei Gryllus campestris L. d. I. Mitteilung. ZooL Anz. 34, 477-478. Renner, M. (1952). Analyse der Kopulationsbereitcchaft der Weibchens der Feldheuschrecke Euthystira brachyptera Ocsk in ihrer Abhangigkeit von Zustand des Geschlechtsapparat. Z. Tierpsychol. 9, 122-154. Riddiford, L. M. (1968). Trans-2-hexenal: mating stimulant for Polyphemus moths. Science, 158, 139-141. Riddiford, L. M. (1970). Antenna1 proteins of saturniid moths: their possible role in olfaction. J. Insect Physiol. 16, 653-660. Riddiford, L. M. (1972). Juvenile hormone in relation to the larval-pupal transformation of the Cecropia silkworm. Biol. Bull. mar. biol. Lab., Woods Hole, 142. 310-325. Riddiford, L. M. (1973). The role of hormones in the reproductive behavior of female wild silkmoths. In “Experimental Analysis of Insect Behaviour” (Ed. L. Barton Browne). (In press.) Springer-Verlag, Berlin. Riddiford, L. M. and Ashenhurst, J. (1973). The witchover from virgin to mated behavior in female Cecropia moths; the role of the bursa copulatrix. Bwl. BulL mar. biol. Lab., Woods Hole, 144, 162-171. Riddiford, L. M. and Williams, C. M. (1967). Volatile principle from oak leaves: role in sex life of the Polyphemus moth. Science, 155, 589-590. Riddiford, L. M. and Williams, C. M. (1971). Role of the corpora cardiaca in the behavior of saturniid moths. I. Release of sex pheromone. Biol. Bull. mar. bwL Lab., Woods Hole, 140, 1-7. Riemann, J. G., Moen, D. J. and Thorson, B. J. (1967). Female monogamy and its control in houseflies. J. Insect Physiol. 13, 407-418. Riemann, J. G. and Thorson, B. J. (1969). Effect of male accessory material o n oviposition and mating by female houseflies. A n n eiat. SOC.A m . 62, 828-834.

350

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

Roberts, S. K. de F. (1966). Circadian activity rhythms in cockroaches. 111. The role of endocrine and neural factors. J. cell. come. Physiol. 67, 473-486. Roberts, S. K. de F., Skopik, S. D. and Driskill, R. J. (1971). Circadian rhythms in cockroaches: does brain hormone mediate the locomotor cycle? In “Biochronometry” (Ed. M. Menaker), pp. 505-514. National Academy of Sciences Press, Washington. Roeder, K. D. (1935). An experimental analysis of the sexual behavior of the praying mantis. Biol. Bull. mar. biol. Lab., Woods Hole, 69, 203-220. Roeder, K. D., Tozian, L. and Weiant, E. A. (1960). Endogenous nerve activity and behaviour in the mantis and cockroach. J. Insect Physiol. 4, 45-62. Roller, H., Piepho, H. and Holz, I. (1963). Zum Problem der Hormonabhangigkeit des Paarungsverhaltens bei Inseckten. Untersuchungen an Galleria mellonella (L.). J. Insect Physiol. 9, 187-194. Roth, L. M. (1962). Hypersexual activity induced in females of the cockroach Nauphoeta cinerea. Science, 138, 1267-1269. Roth, L. M. (1964). Control of reproduction in female cockroaches with special reference to Nauphoeta cinerea. I. First preoviposition period. J. Insect Physiol. 10, 915-945. Roth, L. M. and Barth, R. H., Jr. (1964). The control of sexual receptivity in female cockroaches. J. Insect Physiol. 10, 965-975. Roth, L. M. and Stay, B. (1962). Oocyte development in Blattella germanica (Linn.) and Blattella uaga Hebard (Blattaria). Ann. ent. SOC. A m . 55, 633-642. Roussel, J. P. (1967). Fonctions des corpora allata et controle de la pigmentation chez Gryllus bimaculatus de Geer. J. Insect Physiol. 13, 113-130. Scharrer, E. and Scharrer, B. (1963). “Neuroendocrinology”. Columbia University Press, New York. Schneider, D. (1969). Insect olfaction: deciphering system for chemical messages. Science, 163, 1031-1037. Schooneveld, H. (1970). Structural aspects of neurosecretory and corpus allatum activity in the adult Colorado beetle, Leptinotarsa decemlineata Say. as a function of day length. Arch. Neerl. Zool. 20, 151-237. Siew, Y. C. (1965). The endocrine control of adult reproductive diapause in the chrysomelid beetle, Galeruca tanaceti (L.). 111. J. Insect Physiol. 11, 973-981. Smith, J. H. (1956). Fertility, mating behaviour, and sexual selection in Drosophila subobscura. J. Genet. 54, 261-279. Sower, L. L., Shorey, H. H. and Gaston, L. K. (1970). Sex pheromones of noctuid moths. XXI. Light: dark cycle regulation and light inhibition of sex pheromone release by females of Trichoplusiu ni. Ann. ent. SUC.A m . 63, 1090-1092. Spielman, A., Leahy, M. G. and Skaff, V. (1969). Failure of effective insemination of young female Aedes aegypti mosquitoes. J. Insect Physiol. 15, 1471-1479. Spieth, H. T. (1966). Drosophilid mating behaviour: the behaviour of decapitated females. Anim. Behau. 14, 226-235. Staal, G. B. (1961). “Studies o n the Physiology of Phase Induction in Locusta migratoriu migratorioides R. and F”. H. Veenman and N. V. Zonen. Wageningen. Stengel, M. and Schubert, G. (1972). Influence des corpora allata de la femelle pondeuse de Melolontha melolontha L. (Coleoptera Scarabidae) sur le comportement migrateur du mile. C.r. Acad. S c i Paris, 274, 568-570.

HORMONES AND BEHAVIOUR

351

Strejckova, A., Swevit, 2. and Novak, V. J. A. (1965). Lffect of neurohormones Cl and D I on spontaneous electrical activity of the central nervous system of the cockroach. J. Insect Physiol. 11, 889-896. Strong, L. (1968a). The effect of enforced locomotor activity on lipid content in allatectomized males of Locusta migratoria m*atorzoides. /. exp. Biol. 48, 625-630. Strong, L. (1968b). Locomotor activity, sexual behavior, and the corpus allatum hormone in males of Locusta. J. Insect Physiol. 14, 1685-1692. Stumm-Zollinger, E. ( 195 7). Histological study of regenerative processes after transection of the nervi corporis cardiaci in transplanted brains of the Cecropia silkworm (Platysamia cecropia L.). J. exp. Zool. 134, 315-326. Terranova, A. C., Leopold, R. A., Degrugillier, M. I.. and Johnson, J. R. (1972). Electrophoresis of the male accessory secretion and its fate in the mated female. J. Insect Physiol. 18, 1573-1591. Thomas, A. and Mesnier, M. (1973). Le rBle du systime nerveux central sur lee mechanismes de l’oviposition chez Carausius morosus et Clitumnus extradentatus. /. Insect Physiol. 19, 383-396. Truman, J. W. (1970). The eclosion hormone: its release by the brain and its action on the central nervous system of silkmoths. Am. 2001. 10, 51 1-512. Truman, J. W. (1971a). Physiology of insect ecdysis I. The eclosion behaviour of saturniid moths and its hormonal release. /. cxp. Biol. 54, 805-814. Truman, J. W. (1971b). Circadian rhythms and physiology with special reference to neuroendocrine processes in insects. In “Proceedings of the International Symposium o n Circadian Rhythmicity”, pp. 111-135. Pudoc Press, Wageningen, Netherlands. Truman, J. W. (1973). Physiology of insect ecdysis. 11. ‘The assay and occurrence of the eclosion hormone in the Chinese oak silkmoth, An!heraea pernyi. BioL BULL mar. biol. Lab., Woods Hole, 144, 200-21 1. Truman, J. W. and Riddiford, L. M. (1970). Neuroendocrine control of ecdysis in silkmoths. Science. 167, 1624-1626. Truman, J. W. and Riddiford, L. M. (1971). Role of the corpora cardiaca in the behavior of saturniid moths. 11. Oviposition. Biol. BulL mar. biol. Lab., Woods Hole, 140. 8-14. Truman, J. W. and Sokolove, P. G . (1972). Silkmoth eclosion: hormonal triggering of a centrally programmed pattern of behavior. Science, 1:’5, 1491-1493. Unger, H. (1965). Der Einfluss korpereigener Wirkstoffe auf die Nerventatigkeit von Periplaneta americana. L. ZooL Jb. (Physiol.), 71, 727-740. Van den Assem, J. (1970). Courtship and mating in Lmiophagus distinguendus (Forst.) Kurdy (Hymenoptera, Pteromalidae). Arch. Neerl. Zool. 20, 329-352. Van der Kloot, W. G. and Williams, C. M. (1953a). Cocoon construction by the Cecropia silkworm. I. T h e role of the external environment. Belaauiour, 5, 141-156. Van der Kloot, W. G. and Williams, C. M. (1953b). Cocoon construction by the Cecropia silkworm. 11. The role af the internal environment. Behaviour, 5, 157174. Wajc, E. and Pener, M. P. (1969). The effect of the corpora allata o n the mating behavior of the male migratory locust, Locusta m e a t o r i a m&atorioides (R. and F.). Israel J. Zool. 18, 179-192. Wajc, E. and Pener, M. P. (1971). The effect of the corpora allata on the flight activity of the male African migratory locust, Locusta m w a t o r i a migrntorioides (R. and F.). G e n comp. Endocr. 17, 327-333.

352

JAMES W. TRUMAN AND LYNN M. RlDDlFORD

Wiedbrauck, S. (1955). Vom Spinnin der Schmetterlingsraupen und seiner Abhangigbeit von Metamorphosehormonen. Z . Tierpsychol. 12, 176-202. Wigglesworth, V. B. (1970). “Insect Hormones”. W. H. Freeman and Co., San Francisco. Wilde, J. de, and Boer, J. A. de (1961). Physiology of diapause in the adult Colorado beetle. 11. Diapause as a case of pseudoallatectomy. J. Insect Physiol. 6 , 152-161. Wilde, J. de and Boer, J. A. de (1969). Humoral and nervous pathways in photoperiodic induction of diapause in Leptinotursa decemlineata. J. Insect PhysioL 15, 661-675. Williams, C. M. (1961). The juvenile hormone. 11. Its role in the endocrine control of molting, pupation, and adult development in the Cecropia silkworm. Biol. Bull. mar. biol. Lab., Woods Hole, 121, 572-585. Wilson, D. M. (1961). The central nervous control of flight in a locust. J. exp. BioL 38. 47 1-490. Wilson, E. 0. and Bossert, W. H. (1963). Chemical communication among animals. Rec. R o g . Horm. Res. 19, 673-716. Zdkek, J. and Slima, K. (1968). Mating activity in adultoids or supernumerary larvae induced by agents with high juvenile hormone activity. /. Insect Physiol. 14, 563-567. Ziegler, R. (1972a). Sexual- und territorialverhalten der Schabe Gromphudorhina brunner Butler. Z. Tierpsychol. 31, 531-541. Ziegler, R. (1972b). Kontrolle des Fortpflanzungszyklus und des Sexualverhaltens bei der ovoviviparen Schabe Cromphadorhinn brunneri Butler. Zool. Jb. Physiol. 76, 417-433.

Author Index Numbers with an asterisk refer to .Paaes - on which references are fisted at the end of the paper

A

B

Acree, F., Jr., 31, 97*, 114* Adams, T.S., 324,327,342* Adams, W. B., 285, 286, 293*,

Baglioni, C., 180, 183, 184, 186,

225* Bagnoli, F., 300, 310, 311,342* Bailey, W.J., 263, 294* Baillie, D. L., 180, 181, 183, 185,

296* Adiyodi, K, G.,328, 342* Adkisson, P. L., 26, 32, 45, 101*,

186, 187 225*

Bakker, K..,86,98* Ball, H.J., 44,46,58,98* Barlow, R. B., 285,296* Barnes, F. J., 216,225* Barritt, L.C.,87,98* Bartell, R.J., 11, 15,98* Barth, F. G.,274, 294* Barth, R. H.,316,342* Barth, R. H. Jr., 320, 321, 322,

115*

Aguirre, L., 122,242*,243* Ajami, A. M., 140, 141, 159, 160,

165,169,225*

Albrecht, W.,124,227* Allegri, G.,194,230* Almeida, F. F., de, 147, 159, 196,

208,225*,236*

Amouriq, L.,42,97* Anders, G.,126, 178, 244* Anderson, R. L., 131, 226* Arai, Y.,176,240* Arcaya, G.,222,225* Arkchiga, H.,66,68,91, 97* Amy, I., 328,342* Aschoff, J., 4,47. 48, 51, 92, 97* Ashenhurst, J., 300, 329, 330.

349* Atkinson, D. E., 218, 225* Atkinson, R.J. A., 67, 73, 108* Autrum, H.,162, 168, 225*,235*,

271,272,273,293* Aylor, D., 269,294* Azaryan. A. G.,60,98*

323,342*,344*, 350*

Barton Browne, L.,34,98* Bastock, Id., 314,342* Bates, M.,10,98* Bauer, A. C.,216, 225* Baumann, F., 197,240* Baumann, U., 135, 137, 138, 140,

148,159,161,225*,229*

Baumberger, J. P., 165,234* Baweja, K.D.,324, 346* Bazley, E. N.,268, 270,294* Beadle, G W., 119, 126, 128, 131,

186,225*,2%*, 231*, 243*

Beament,J. W. L., 74,98* Beattie, T. M.,62,66,98* 353

354

Becher, D., 130, 227* Beck, J., 168, 176, 227* Beck, S. D., 1, 2, 23, 24, 26, 47, 98* Becker, E., 119, 125, 134, 135, 138, 159, 162, 170, 226*, 228* Beckmann, R., 127, 138, 226*, 229 * Beckwith, J. R., 187, 232* Beier, W., 9, 98* Belcher, K. S., 23, 98* Bell, W. J., 321, 342* Bellamy, D., 164, 165, 226* Benassi, C. A., 121, 226* Bennett, M. F., 60, 98' Bennett-Clark, H. C., 248, 265, 266,267,268,291,294* Beetsma, J., 337, 342* Benz, G., 327,342* Benzer, S., 75, 76, 78, 80, 106* Beranek, L. L., 248, 249, 251, 253, 268,275,294* Berg, J. S. V., 289, 290, 295* Berthold, G., 174, 178, 226* Biedermann, W., 178, 226* Biekert, E., 120, 131, 135, 137, 138, 139, 142, 144, 145, 146, 148, 149, 150, 159, 160, 161, 176, 177: 180, 193, 194, 228*, 229* Binkley, S., 89, 98* Birringer, H., 131, 242* Birt, L. M., 87, 98* Birukow, G., 7, 13, 79,95,99* Blest, A. D., 314, 342*, 343* Bliss, D. E., 69, 99* Block, R. J., 118, 226* Boden, E., 312, 337, 349* Boer, J. A., de, 319, 336, 352* Boggust, W. A., 120, 231* Bonse, A., 124, 125, 126, 128, 159, 160, 178, 208, 209, 212, 226*, 245 *

AUTHOR INDEX

Boistel, J., 34, 109* Bossert, W. H., 300, 301, 352* Bounhoil, J. J., 297, 314, 321, 328, 342*, 343* Bouthier, A., 137, 138, 139, 144, 159, 177,226* Bowman, J. T., 194, 240* Bowness, J. M., 165, 226* Brady, J., 2, 3, 6, 7, 8, 13, 14, 15, 25, 30, 34, 35, 37, 55, 56, 57, 58, 59, 60, 61, 62, 64, 65, 66, 74, 77, 79, 87, 89, 95, 99*, 101*, 299, 337, 338, 343*, 344* Bratton, A. C., 121, 180, 226* Bregazzi, P. K., 69, 99* Brett, W. J., 17, 23, 76, 98*, 99* Britten, R. J., 217, 227* Broughton, W. B., 263, 294* Browse-Gaury, P., 44, 99* Brown, F. A., 91, 100* Brown, K. S., 130, 227*, 243* Brown, R. H. J., 6, 100* Brown, R. R., 121, 194, 227*, 235* Broza, M., 318,319,343* Bruce, V., 35, 111* Brun, F., 223, 233* Brunelli, M., 300, 309, 310, 311, 342*, 343* Brunet, P. C. J., 133, 224, 227* Brunken, W., 75, 76, 111* Brunnelli, M., 300, 310, 342* Bryant, T. R., 88, 100* Biickmann, D., 120, 124, 159, 160, 166, 171, 173, 175, 176, 177, 178, 196, 200, 204, 205, 211, 217, 227*, 235*, 237*, 313, 343* Buffington, J. D., 23, 100* Biihler, R., 330, 344* Bull,D. L., 27, 31,32,95, 100* Bullock, T. H., 62, 100*

AUTHOR INDEX

355

Biinning, E., 74, 88, loo*, 337,

338,343*

Buonamici, M., 300, 309, 310,

343* Burden, G.S.,31,97* Burkhardt, D., 168,169,244* Bumet, B., 168.221,227* Busch, E., 13,99* Busnel, R-G.,248,294* Butenandt, A., 119, 120, 124, 125,

Chinzei.,Y., 202, 204,230* Chovnick, A., 180, 181, 183, 185,

186, 187,225*

Chua, K. E., 29, 110* Clayton, D. L., 75, 101* Cloudsli-y-Thompson, J-L.,

44,

101* Cole, C-L., 26, 32, 101* Colombo, G.,128, 131, 132, 159,

199,230*

127, 131, 135, 137, 138, 139, 140, 142, 144. 145, 146, 148, 149, 150, 159, 160, 161, 176, 177: 180, 193, 194,227*,228*, 229* Butz, A., 27, 30, 71, 95, 103*, 105*

Coluzzi, M., 74, 101* Condoulis, W. V., 21, 107* Connolly, K.,168,227* Conroy, R. T.W. L., 92, 101* Cook, B. J., 55, 70, 74, 101* Cook, D. J., 44,101* Corbet, P. S., 1, 7, 8, 12, 16, 95,

C

Costa, C:., 194,230* Cott, H.B., 171, 230* Craig, G . B., 304, 324, 326, 345* Craig, G.B., Jr., 300, 303, 304, 305,

Calam, D. H., 132,229* Caldwell, R. L., 7, 8, 12, 20, 21,

72, loo*, 102*, lll* Campbell, B. O.,24, 100* Cantacuzene, A-M., 317, 318, 319,

343* Carlisle, D. B., 333, 334, 343* Carlson, A-D., 310,343* Carter, C. E., 12, 107* Casini, E., 127, 239* Caspari, E., 119, 164, 181, 185,

229*, 231* Cassady, W. E., 192,229* Cassier, P., 334,343*, 344* Cawley, B., 27, 113* Chambers, D. C., 121, 242* Chambers, D. L., 6, 100* Chandrashekaran, M. K., 12, 17,

18,45,79, loo*, 107* Chaudhury, M. F. B., 58,98* Chauvin, R., 134, 177, 230* Chen, P. S., 330, 344* Chiba, Y.,7: 10, 78, 79, 100* Chino, H.,165,233*

101*,103*, 104*

324, 325, 326, 330,344*,345*, 346* Crescitelli, F., 91, 105* Cromartie, R. J. T., 120, 142, 144, 228*,230* Cubbin, C. M., 7, 48, 80, 81, 95, 105*,106* Cuesta, M.,de la, 55, 70, 74, 101* Cymborowski, B., 3, 15, 32, 37, 38, 39, 4.1, 46, 55, 58, 66, 67, 85, 95,101*,102*,338,344* D Dadd, P. H., 133, 230* D’Ajello, V.,300, 310, 342* Dales, R.. P., 150, 162,230* Dalgliesh, C. E., 122, 132, 233* Danilevsky, AS., 1, 2, 102* Danneel, R., 128, 129, 131, 192,

208,230*,245*

356

D a j o , A., 319,344* Davey, K. G.,300, 325, 328, 330,

331,344* Davidson, E. H., 217,227* Davidson, J. M.,297,344* Davis, G. R. F., 221, 230* Davis, W.J., Jr., 255, 294* Dawkins, P. D., 212, 238* Dean, G.J. W., 14,19,102* De Antoni, A., 194, 230* De Besse, N.,37, 102* Degrugillier, M. E., 300, 346*, 351* Deguchi, N., 162, 245* Delphin, F., 36, 102*,328, 344* Dennhofer, U.,159,230* Derjugin, W.,von, 125, 228* Despommier, D. D., 326,327, 344* Dimicoli, J-L., 223, 230*, 233* Dingle, H.,7, 8, 12, 15, 20, 21, 72, loo*, 102*, 111*, 335, 336, 344* D’Orazio, G., 120, 121, 125, 128, 130,200,240* Dreisig, H.,12,51, 72, 102* Driskill, R. J., 59, 60, 112*, 338, 350* Dubrovin, N. N., 269,294* Dumortier, B., 12, 45, 80, 102*, 254,294* Dupont-Raabe, M., 299, 328, 344* Diirrwachter, G.,13, 102* Dustmann, J. H., 129, 137, 144, 158, 159, 171, 173, 194, 196, 227*, 230* Dutkowski, A., 32, 38, 39, 66, 85, 95,101*,102* Dutkowski, A. B., 15, 37, 38, 41, 46,102* Dutky, S. R., 6, 60,93,112*, 113* E Edney, E. B., 7, 102*

AUTHOR INDEX

Egelhaaf, A., 120, 124, 125, 126,

130, 132, 135, 140, 159, 160, 172, 180, 181, 183, 184, 185, 191, 197, 206, 211, 214, 230*, 231*,236* Ehret, C. F., 87, 102* Eichelberg, D., 120, 126,128,131, 132,208,231*,245* Eidmann, H.,77, 102* Ellis, J., 220,221, 233* Ellis, P. E., 8, 34, 103*, 333, 334, 343* Elsner, E. N., 248, 294* Engelmann, F., 299,321,322,323, 327,344* Engelmann, W., 47,1 0 P Enright, J. T.,88, 103* Ephrussi, B., 119,225*,231* Eskin, A., 90, 103* Esther, H.,27, 1 1 1 * Evans, D. G., 20,108* Evans, D.R.,34,98* Evans, E. J., 268, 270, 294* Ewing, A. W.,291,294* F

Fabro, S., 191,232* Fearon, W.R., 120, 231* Feigelson, P., 181,231* Fernandez, A. T.,27, 103* Feron, M.,159,246* Fingerman, M., 23, 56, 103*, 168,

231* Finlayson, L. H., 56, 73, 113*,

315,344* Finley, J. W., 122,231* Fisher, R. W.,26, 103* Fletcher, B. S., 10, 56, 79, 95,

103*, 114* Flisinska-Bojanowska, A., 37, 39,

101*

Fondacaro, J. D., 27, 71, 95, 103*

357

AUTHOR INDEX

Forrest, H. S., 172, 173, 181, 184, 191, 192, 195, 213, 232*, 240* Foster, W., 320, 344* Fowler, D. J., 33, 39, 40, 41, 66, 95, 103* Franco, P., 159, 230* French, W. L., 214, 241* Fricke, F., 314, 346* Fried, R., 120, 242* Friedman, M., 122, 231* Frudden, L., 27, 103* Fuchs, M. S., 326,344*, 345* Fuge, H., 162, 163, 172, 231* Fujii, S., 221, 235* Fujimori, E., 222, 231* Fukuda, S., 321,345* Fukuda, T., 202,231* Futch, D. G., 185, 242* Fuzeau-Brabsch, S., 120, 159, 164, 173,174, 231*, 232*, 298,345*

G Gabe, M., 328, 342* Galarza, A. M., 120, 159, 232*, 243 * Galun, R., 8, 103* Garcia-Bellido, H., von, 330, 345* Gardner, E. J., 221, 242* Gartenmann, G., 168, 244* Gaston, L. K., 10, 11, 13, 109*, 113*, 339,350* Gaston, S., 89,95, 103* Gaude, H., 57, 114* Geldiay, S., 319, 345* George, J. E., 9, 103* Geronimo, J., 90, 105* Gersch, M., 300, 345* Gewecke, M., 291,294* Geyer, I., 141, 142,241* Ghelelovich, S., 221, 232*

Gholson, R. K., 122, 132, 133, 224,233*, 236* Ghosh, D., 172, 173, 181, 184, 191,192,213,232* Giersberg, H., 178, 232* Gilbert, L. I., 298, 346* Gillett, J. D., 7, 8, 12, 16, 95, 103*, 104* Ginoulhiac, E., 191, 232* Girardie, A., 318, 319, 345*, 349* Glass, B., 221, 241* Glassman, E., 179, 189, 190, 213, 232* Goda, Y.,129, 198, 235* Goldsmith, T. H., 46,47,49, 115*, 162,232* Good!ellow, R. D., 216, 225* Goodinan, L. J., 15, 104* Goodnight, C. J., 33, 39, 40, 41, 66,95, 103* Goodwin, T. W., 134, 152, 175, 232* Gorini, L., 187, 232* Goryshin, N. I., 1, 2, 102* Gotz, K. G., 168, 232*, 233* Grace, T. D. C., 39, 104* Graf, G. E., 126, 178, 232*, 244* Grassmader, H. H., 122, 124, 208, 211,232*, 236* Graszynski, K., 192, 232* Gratz. N. G., 21, 112* Green, G. W., 7, 8, 79, 104* Green, M. M., 122, 124, 125, 186, 208,220,232*, 240* Grell, E. H., 186, 187, 237*, 243* Grigolo, A., 131, 132,237* Gripenberg, J., 144, 161, 232*, 233* Gropalakrishnareddy , T., 40, 41, 69,95, 111*, 339,349* Gutmann, H. R., 194,233* Gwadz, R. W., 304, 305, 324,326, 32?, 345*

358

AUTHOR INDEX

H Haagen-Smit, A. J., 125, 243* Haddow, A. J., 7, 8, 12, 15, 95, 103*, 104*

Hadorn, E., 120, 172, 233* Halberg, F., 24, 27, 71, 105*, 112*, 113*

Hallmann, G., 127,227*, 229* Hamilton, W. J., 1,92, 107* Hanser, G . , 131, 163, 166, 228*, 233*

Hara, A., 194, 202, 234* Harano, T., 165,233* Hardeland, R., 10, 12, 23, 40, 75, 76,80,86,95,104*,

111*

Harker, J. E., 1, 2, 3, 6, 18, 25, 29, 34, 35, 39, 42, 43, 44, 55, 56, 57, 58, 74, 92, 93, 104*, 337, 345* Htirle, E., 148, 229* Harmsen, R., 179, 233* Harmsen, Ri, 162, 172,246* Harris, C. M., 268, 294* Harris, J. E., 13, 104* Harshbarger, J. C., 42, 104* Hartmann, R., 320, 321, 322, 326, 345* Hasegawa, K., 129, 197, 245* Haskell, J. B., 20, 21, 102* Haskell, P. T., 299, 306, 312, 313, 317, 318, 324, 326, 328, 334, 345*, 346* Hastings, J. W., 72, 88, 92, 105* Haverty, M. I., 28, 29, 105* Hayaishi, O., 191, 192, 239* Hayes, D. K., 24, 27, 71, 105*, 113* Htltne, Cl., 223, 230*, 233*, 238* Henderson, L. M., 122, 132, 233* Hendrichs-Hertel, U., 128, 191, 192,213,214,233*, 237* Hengstenberg, R., 72, 105*, 168, 233*

Henke, K., 220,236* Henning, H. D., 128, 233* Henze, M., 174, 226* Herman, W. S., 298,346* Hertel, U., 128, 203, 233*, 238* Hertweck, H., 178, 233* Hickey, W. A., 305,345* Hiehnam, K. C., 298, 311, 317, 327,328,329,346*

Hill, M., 7, 8, 15, 79, 106* Hilliard, S. D., 30, 95, 105* Hinde, R. A., 303, 346* Hinks, C. F., 33,42, 67, 105*, 338, 346*

Hinton, T., 220, 221, 233* Hintz, A. M., 324, 342* Hintze, C., 313, 346* Hintze-Podufal, C., 314, 346* Hiraga, S., 121, 159, 180, 189, 190, 233*

Hirata, Y.,127, 233*, 234* Hiss, E. A., 326, 345*. Hodeson, E. S., 304, 306, 310, 346*

Hoffmann, K., 7, 51, 88, 105* Hoffmann, Ch., 168,236* Hoglund, G., 165, 166, 168, 234* Holz, I., 312, 321, 337, 349*, 350* Honkanen, E., 144, 161, 233* Honegger, H. W., 47, 103* Hope, A.M., 7, 8, 15, 79,106* Hopsu-Haw, V. K., 189,235* Horikawa, M., 129,192,234* Horn, U., 191,234* Horowitz, N. H., 165, 234* Horridge, G. A., 62, 100* Horstmann, G., 163,234* Horton, J., 24, 71, 105* Hoyle, G., 8,34, 103*, 105* Huber, F., 60, 64, 65, 105*, 248, 265, 294*, 302, 304, 320, 321, 324, 325, 347* Hurshman, L. F., 21, 107* Husain, M. A., 324,346*

AUTHOR INDEX

359

I Inagami, K., 121, 123, 127, 128, 130, 131, 132, 154, 159, 179, 194, 198, 199, 201, 202, 234*, 238* Isaak, D., 29, 110* Isenberg, I., 222, 234* Ishiguro, O., 120, 121, 127, 130, 132, 159, 161, 165, 179, 194, 202, 203, 204, 211, 219, 234*, 237*, 242* Iwata, S., 127, 139, 235*, 239* Ives, D., 17, 47, 115*

J Jacklet, J. W., 81, 90, 105* Jacobson, M., 316, 346* Jackson, P. J., 12, 109* Jacobson, K. B., 187, 188, 235*, 243* Jaenicke, L., 164, 246* Jahn, T. L., 91, 105* Janda, V., 23, 105* Jansen, C. T., 189, 235* Jegannathan, N. S., 194, 243* Jezewska, M., 122, 235* Joerrens, G., 338, 343* Johnson, C. G., 7, 15, 105*, 333, 335, 346* Johnson, J. R., 327, 351* Johnstone, B. M., 279, 283, 284, 294* Johnstone, J. R., 279, 283, 284, 294* Joly, P., 319, 349* Jones, M. D. R., 6, 7, 8, 15,48, 79, 80,81,95, 105*, 106*, 113* Joshi, S., 194, 235* Junei, Th., 168, 244* K Kalmus, H., 67, 106*, 168,235* Kambysellis, M. P., 86, 106* Kashin, P., 8, 106*

Kanchisa, T., 221, 235* Kaplan, W. D., 185,242* Kaplanis,J. N., 86, 106* Kappus, K. D., 10, 115* Karlson,P., 86, 87, 112*, 131, 175, 21 7,228*, 235* Katayama, M., 130, 211, 243* Katz, E., 193, 235*, 244* Kaufman, S., 183, 184, 185, 186, 212,235* Kawase, S., 135,154,159,235* Keck, J., 193, 228* Keeley, L. L., 57, 112* Keller, J. C., 60, 93, 113* Kellner, O., 202, 235* Keppler, E., 291, 294* Kikkawa, H., 125, 127, 129, 132, 133, 197, 198, 199, 202, 233*, 234*, 235* Kirimura, J., 202, 231 * Klatt, B., 304, 346* Kletsky, E J., 285, 296* Klose, W., 159, 170, 176, 241* Klotter, K., 4, 98* Klug, H., 34,106*, 338,346* Kluth, E., 139, 98* Knox, W. E;., 189,238* Koga, N., 1.21, 126, 129, 137, 139, 144, 159, 198, 229*, 235*, 239*, 24O* Kolb, G., 162, 236* Koltermanri, R., 9, 106* Komisaruk. B. R., 302, 346* Konopka, R. J., 75, 76, 78, 80, 106* Kontopp, Ei., 40, 106* Kommann, P., 120, 135, 137, 138, 139,176,228* Kotake, Y . , 130, 236* Krauss, D., 131, 2$2* Krieger, F., 170, 174, 236* Kiibler, H., 137, 138, 139, 144, 159, 160, 162, 177, 195, 229*, 236*

360

AUTHOR INDEX

Kiihn, A., 119, 120, 135, 140, 159, 160, 169, 171, 172, 181, 195, 196, 197,220,233*, 236* Kuwana, H., 133, 235* I,

Labrie, M. M., 33, 40, 41, 66, 95, 103* Lago, A. D., 23, 56, 103* Lan, S. J., 133, 224, 236* Langer, H., 124, 1E2, 163, 164, 165, 166, 167, 168, 208, 211, 234*, 236*, 237* Larimer, J. L., 46, 69, 109* Lashley, K. S., 303, 346* Laudani, U., 131, 132, 237* Law, J. H., 316,346* Lea, A. O., 304,324,346* Leahy, M. G., 324, 330, 346*, 350* Leclerc, M. M., 24, 81, 106* Lees, A. D., 2, 22,44,93, 106* Leibenguth, F., 131, 159, 183, 185, 193,209,211,214,237* Leklem, J. E., 218,237* Leopold, R. A., 300, 327, 346*, 347*, 351* Leppla, N. C., 27, 28, 106* Lester, L. J., 316, 342* Leston, D., 254, 294* Leuthold, R., 56, 60, 106* Levenbook, L., 216, 225* Levine, S., 297, 344* Lewis, D. B., 277,294* Lewis, T., 8, 106* Libby, J. L., 307,347* Lienhard, J. P., 24, 81, 106* Lindauer, M., 9, 15, 25, 44, 98*, 106*, 107* Lindquist, D. A., 27, 31, 32, 95,

loo*

Lindsley, D. L., 186, 237*

Linzen, B., 120, 121, 122, 124, 127, 128, 130, 132, 133, 134, 135, 137, 138, 139, 140, 144, 145, 146, 147, 149, 150, 159, 160, 161, 162, 163, 164, 171, 173, 175, 176, 177, 180, 183, 185, 191, 192, 194, 195, 196, 200, 202, 204, 205, 206, 211, 213, 214, 217, 218, 219, 227*, 228*, 229*, 233*, 234*, 237*, 238*, 241*, 242* Lipton, G. R., 8, 107* Lisk, R. D., 302,347* Locke, M., 21, 107* Lockshin, R. A., 315, 347* Loher, W., 12, 17, 18, 45, 61, 79, loo*, 107*, 302, 304,317,318, 320, 321, 322, 324, 325, 326,. 345*, 347* Lohmann, M., 7, 39,107* Louloudes, S. J., 86, 106* Lounibos, L. P., 304, 324, 326, 345* Lowe, M. E., 23,56, 103* Lowry, W. P., 271,294* Lucas, F., 202,238* Lum, P. T. M., 16, 17, 20, 110* Liischer, M., 303, 347* M

McArthur, J. N., 212, 238* Macaulay, E. D. M., 6, 107* McCluskey, E. S., 12, 107* McComb, M., 27,28,29,114* McFarlane, J. E., 12, 107* McMenamy, R. H., 212,238* Maddrell, S. H. P., 299, 343* Maeno, H., 181, 231* Magni, F., 300, 309, 310, 311, 342*, 343*, 347* Magnus, D., 170,238* Maier, W.,163, 238*

AUTHOR INDEX

Makino, K., 127, 238* Mandel, M., 222, 241* Manning, A., 324, 326, 332, 347* Marco, R., 222, 242* Margalit, J., 8, 103* Markl, H., 248, 254,255,256,257. 262,294*, 295* Marler, P., 1, 92, 107* Marsh, D., 7, 11, 48, 80, 81, 95, 105*, 106*, 107* Marshall, E. K., 121, 180,226* Martin, H., 15, 106* Marzluf, G. A., 180, 183, 184, 185, 187,238* Matuda, M., 202,231* Maxwell, J., 131, 226* Mayer, G., 189, 191,238* Medugorac, I., 25,44, 107* Mehler, A. H., 189, 238* Mellanby, K., 55, 107* Menaker, M., 89,98*, 103* Merle,J., 326, 330, 347* Merlini, L., 159, 238* Mesnier, M., 328, 351* Michel, R., 57,66, 107*, 334,347* Michelsen, A., 248, 250, 253, 265, 271, 272, 274, 275, 276, 279, 280,283,287,288,295* Milburn, N. S., 300,303,305,306, 320,348* Miliani, M., 192,193,240* Miller, L. A., 274, 288, 289, 295* Milligan, J. V., 44, 107* Mills, J. N., 92, 101* Minis, D. H., 5, 12, 15, 16, 17, 22, 2 8 , 2 9 , 4 a , 75,77,78, 95, io7*, 110* Mittler, S., 221, 238* Moccellin, E., 159, 230* Moen, D. J., 327,349* Mohlmann, E., 130, 159, 208, 238* Mokia, G. G., 300,330,348* Montenay-Garestier, Th., 223, 238*

361

Moore, T. E., 257, 295* Moorhouse, J. E., 299, 306, 312, 313,334,346* Moran, V. C., 8, 74, 79, 80, 115* Morean, L. R., 194,238* Morse, P. hl., 248, 260, 275, 279, 295* Mothes, G., 41, 71, 107* Mrciak, M., 23, 105* Muckenthaler, F. A., 126, 129, 238* Miiller, H. F'., 322, 323, 326, 347* Muralikrishna Dass, P., 32, 114* hlurphey, R. K., 285,295* Musajo, L., 127, 239* Muth, F. W.,164, 195,239* N

Naegele, J. A., 12, 24, 25, 26, 109*, 110* Nagarnura, Y.,159, 161, 165, 194, 202, 203, 204, 211, 219, 234* Nagasawa, H. T., 194, 233* Nair, P. M., 193, 239* Nakagaki, M., 139,239* Nakamura, A., 126, 243* Nakanishi, K., 127,233*, 234* Nasini, G., 159, 238* Nawa, S., 180,239* Nayar, J. K., 3, 7, 16, 1 7 , 20, 31, 54, 77, 78, 79, 81, 95, 107*, 108* Nayar, K. K., :328,342*, 348* Naylor, E., 62, 67, 68, 69, 73, 108*, 115* Needham, A. E., 135, 239* Neese, V., 159,168,169,239* Neubert, G., 135, 137, 138, 139, 140, 144, 1145, 146, 159, 161, 193,228*, :?29*, 239* Nelissen, F. X., 86, 98* Nelson, D. R., 327,342*

362

AUTHOR INDEX

Neumann, D.,19, 20, 74,95,108*,

109*

Neville, A. C., 20, 21, 46, 71, 72,

95,108*,115* Nickerson, B., 171, 239* Nielsen, E.T., 10,20, 51, 72, 102*, 108* Neilsen, H.T.,10,108* Nishiitsutsuji-Uwo, J., 42, 44, 45, 56, 58, 59, 61, 62, 66, log*, 109*,338,348* Nocke, H., 258, 259, 260, 263, 265,283,295* Nolte, D. J., 162, 163, 170, 220, 239* Nopp, H.,314,348* Nopp-Pammer, E., 314,348* Norris, M.J., 317, 348* Novak, V.J. A., 69, 70,113*,306, 351* Nowosielski, J. W., 6 , 12, 24, 25, 26, 30, 45, 46, 79, 109*, 110* Noyes, D.T.,220, 221, 233* Nozaki, M., 192,239* 0

O’ConnelI, T.B., 6, 100* Odhiambo, T. R., 7, 109*, 317,

318,334,335,338,348*

Ogata, K.,127, 235* Ohnishi, E., 129, 197, 242* Ohtsuki, H.,150, 165,245* Okada, M.,121,244* Okamoto, H.,191,192,239* Olson, H. F., 248,295* Oncley, J. L.,212,238* Orci, L., 197,240*

M., 121, 126, 129, 137, 139, 159, 167, 176, 194, 198, 235*, 239*, 240* Osborn, J., 72, 109* Ottaviani, F., 191, 192, 240* Ozbas, S., 304, 306, 348* Osanai,

P Page, T. L., 46,69, 109* Paget, J., 14,102* Paietta, J. V.,75, 101* Pammer, E., 314, 348* Pantoja, M.E., 222,225* Parker, A. H.,7, 27, 79, 109* Parsons, P. A., 220, 240* Patoharju, O.,144,161,233* Patton, R. L., 6, 24, 30,45,46,79,

109* Pavlidis, T., 72, 73, 88, 90, 109*,

115*

Payne, T.L., 11, 109* Pener, M. P., 317, 318, 319, 320,

334, 335, 343*, 348*, 349*, 351* Perez, Y., 319, 349* Perrelet, A., 197,240* Petris, A., 121,180,183,184,191, 192, 193,240* Petropulos, S. F., 5 8 , 59, 66, 109* Pfliiger, W.,19, 109* Phelps, R.J., 12, 109* Phillips, J. P., 194, 195, 240* Philpott, D.E., 162,232* Pichon, Y.,34, 109* Pieber, M.,222,225*, 240* Piepho, H., 312, 314, 321, 337, 349*, 350* Pierce, G.W., 255, 295* Pilson, B. M.,55, 109* Pilson, R. D.,55, 109* Pinamonti, S., 120, 121, 125, 128, 130, 131, 132, 180, 183, 184, 191, 192, 193, 199, 200, 230*, 240* Pittendrigh, C. S., 5, 15, 16, 17, 18, 22, 25, 42, 44, 45, 48, 49, 56, 58, 59, 61, 62, 66, 72, 73, 75, 77, 78, 79, 80, 83, 95, 107*, 108*, 109*, 110*, 113*, 115*, 338,340,348*,349*

363

AUTHOR INDEX

Pizzarello, D. J., 29, 110* Plagge, E., 119, 197, 236*, 240* Plaine, H. L., 221, 241* Polcik, B., 12, 25, 26, 110* Pomonis, J. G., 55, 70, 74, 101* Powell, B. L., 68, 69, 110* Prasad, C., 214, 241* Price, J. M., 121, 227* Priesner, E., 302, 349* Pringle, J. W. S., 254, 257, 260, 262,294*, 295* Provost,M. W.,16, 17,20, 110* Przelecka, A., 37, 102* Pumphrey, R. J., 248, 271,295*

R Raabe, M., 37, 71, 72, 110* Ralph, C. L., 25,40, 110* Ramsay, J. A., 128, 241* Randolph, N. M., 27, 103* Rankin, M. A., 12, 72, 111*, 336, 349* Rao, K. P., 40, 41, 69, 70, 95, 111*,339,349* Raszka, M., 222,241* Rechsteiner, M. C., 216, 241* Regen, J., 324, 349* Regnier, F. E., 316, 346* Reichardt, W.,265, 295* Reid, K. H., 257, 295* Reinhardt, R., 27, 111* Reisener-Glasewald, E., 172, 241* Remmert, H., 19, 111* Renner, M., 9,95, 111*,326, 349* Rensing, L., 12, 13, 14, 15, 16, 19, 23, 24, 29, 35, 36, 37, 39, 40, 41, 53, 54, 66, 71, 75, 76, 85, 86,87,95,111* Rhyne, A. L., 29, 110* Richards,A.G.,23, 24, 111*, 112* Richards, J., 164,229* Richter, K., 300, 345*

Riddiford, ,L. M., 11, 12, 44, 45, 47, 52, 53, 112*, 114*, 140, 141, 159, 160, 165, 169, 225*, 300, 302, 303, 314, 321, 325, 329, 330, 339, 340, 349*, 351* Riemann, J. G., 327,349* Ringelberg,J., 13, 112* Rizki, M. 1'. M., 125, 126, 221, 241* Rizki, R. M., 126, 181, 183, 185, 241* Rizki, T. M., 126, 181, 183, 184, 185,186,241* Robbins, W. E., 86, 106* Roberts, S. K de F., 3, 6, 7, 23, 27, 44, 56, 57, 59, 60, 61, 72, 73, 79,95, 112*, 338,350* Roeder, K. I)., 60, 112*, 289, 290, 295*, 300, 303, 305, 306, 320, 348-350* Rohner, M. C., Senr, 194, 241 * Roller, H., 321, 350* Romero, C., 222, 225*, 240* Rosenthal, J . , 27, 113* Roth, H., 206, 241* Roth, L. M., 321, 322, 323, 325, 350' Rothert, H., 27, 112* Rounds, H. I)., 31, 112* Roussel, J. P., 321, 350* Rowell, C. H. F., 173, 174, 241* Riidiger, W., 159, 170, 176, 241* Rudkin, G. 'I. 133, , 242* Ruiter, L., de, 171, 241*, 337, 342* S

Sakan, T., 130, 243* Sako, T., 2051, 235* Sang, J. H., 221,227* Santti, R. S., 189, 235* Satoh, K., 127, 238* Sauerman, D. M., Jr., 3, 7, 17, 78, 79,108*

364

Saunders, D. S., 22, 112* Saunders, J. C., 279, 283, 284, 294*

Sawano, J., 202, 235* Sawyer, R. T., 257, 295* Schafer, W., 120, 137, 139, 141, 142, 144, 146, 160, 161, 177, 180,228*, 229*, 241* Schallek, W., 62, 67, 68, 69, 112* Scharrer, B., 299,350* Scharrer, E., 299, 350* Schartau, W., 121, 122, 124, 128, 132, 133, 159, 185, 206, 213, 219,241* Schechter, M. S., 6, 24, 60, 71, 93, 104*, 112*, 113* Scheibe, G., 139, 245* Schieber, J. P., 24, 81, 106* Schiedt, U., 120, 131, 135, 137, 138, 139, 142, 145, 176, 193, 228* Schildknecht, H., 131, 171, 241*, 242* Schlegel, P., 291, 294* Schlein, Y., 21,22,52, 112* Schlossberger, H., 119, 127, 228* Schlossberger, H. G., 127, 242* Schmid, H., 130, 244* Schmidt, H., 171,241* Schneider, D., 302, 350* Scholtan, W., 139, 242* Schooneveld, H., 328,350* Schubert, G., 337,350* Schultz, J., 133, 242* Schulz, G., 131, 228* Schwabl, G., 162, 163, 164, 242* Schweiger, H. G., 88, 112* Schwinck, I., 150, 242* Scoffone, E., 121, 226* Seecof, R. L., 185, 242* Semm, K., 120, 242* Senoh, S., 130,243* Servaas, H., 13, 112*

AUTHOR INDEX

Servit, Z., 69, 70, 113* Seshadri, T. R., 135, 159, 162, 242*

Shaaya, E., 86, 87, 112* Shapard, P. B., 124, 242* Shaw, J. T. B., 202,238* Shepard, M., 57, 112* Shichri, G., 130, 236* Shinohara, R., 189, 242* Shorey, H. H., 10, 11, 13, 15, 95, 98*, 109*, 113*, 264,296*

Shorey, M. M., 339, 350* Shoup, J. R., 162, 163, 172, 242* Siew, Y. C., 328, 350* Simmons, J. R., 194, 221, 240*, 242*

Singh, H., 135, 159, 162, 242* Sivian, L. J., 269, 270,296* Skaff, V., 324, 350* Skangiel-Kramska, J., 32, 102* Skopik, S. D., 18, 59, 60, 110*, 112*, 113*, 338,350*

Skudrzyk, E., 257, 296* Slima, K., 304, 352* Smith, G., 68,108* Smith, J. H., 326, 350* Smith, S. G., 202,238* Smith, U., 71,113* Smittle, B. J., 31, 98* Sokolove, P. G., 52, 114*, 303, 305, 307,351* Sols, A., 222, 242* Sonobe, H., 129,197,242* Sower, L. L., 10, 11, 113*, 339, 350* Spada, A., 127,239* Spangler, H. G., 9, 27, 28, 106*, 113* Spielman, A., 324,350* Spies, J. R., 121, 242* Spieth, H. T., 264, 296*, 327,350* Srisukh, S., 134, 152, 232* Ssenkubuge, Y., 8, 12, 15. 104*

365

AUTHOR INDEX

Staal, G. B., 335,350* Stamm, M. D., 120, 122, 159,

232*,242*, 243* S t a g e , G., 10,23, 80,95, 104* Staudinger, H., 189, 191, 234*,

238* Stay, B., 322, 323, 350* Stengel, M.,337, 350* Stierlin, E., 172, 244* Streck, P., 168, 169,243*,244* Strejzkovi, A., 69, 70, 113*, 306, 351* Strong, L., 318, 334, 335, 338, 351 * Strother, G. K., 166,243* Strumwasser, F., 81, 90, 113* Struwe, G., 162, 164, 165, 166, 167,168,234*,236* Stumm-Zollinger, E., 325,351* Subba Rao, P. V., 194,243* Subramanian, G. B. V., 135, 159, 162,242* Suga, N., 263, 264,296* Suguri, S., 150, 165, 245* Sullivan, W. N., 6,27,60,93, 112*, 113* Sukc-Michieli, S., 175,243* Sutherland, D.J., 8, 107* Suzuki, T., 202, 231* Sweeney, B. M., 39, 88, 92, 113* Swevit, Z., 306,351* Swilley, E. M., 327,347* Szent-Gybrgyi, A., 222,234* Szymanski, J. S.,6, 113* T Takahashi, H., 127,238* Tamura, Z.,121, 244* Tartof, K. D., 181, 183, 186, 187,

189,243* Tatum, E. L., 125, 186, 243* Taylor, B., 81, 113*

Taylor, L. R., 8, 106* Taylor, R. L.,42, 104* Tenconi, L. T., 191, 232* Terranova, A. C.,300, 327, 346*,

347*,350* Thach, B., 35, 111* Thomas, A, 328, 351* Thomas, IZ., 56, 73, 113* Thompson, M. J., 86, 106* Thorell, B., 165, 166, 168, 234* Thorson, B. J., 300, 327, 346*,

349* Tiedt, S., 125, 128, 183, 185, 190,

200,201,243* Tirapegui, C., 222, 240* Tischner, :H., 290, 291, 296* Tohi, J., :!22, 240* Tohi, J. C.,222, 225* Tojo, S., 202, 204, 230* Tokuyama., T., 130,243* Tozian, L., 306,350* Traub, P., 137, 138, 141, 144,

229*, 243* Traynier, It. M. M., 10,113* Treat, A. E;., 289, 290, 295* Trucco, E.,87, 102* Truman, J-W., 2, 12, 19,44,45,47,

51, 52, 53, 54, 76, 77, 82, 83, 84, 85, 86, 95, 113*, 114*,300, 303, 305, 307, 314, 315, 329, 337,340,351* Tsuchitani, K., 131,244* Turner, R. B., 31, 97*, 114* Twardzik, 11. R., 187,243* Tychsen, P.H.,10,79,95,114* Tyshchenko, V. P., 1, 2, 60, 98*, 102* U Uchida, T., 139, 159, 160,244* Udenfriend, S., 126, 243* Ullrich, V., 189, 191, 234*, 238*

366

AUTHOR INDEX

Umebachi, Y . , 126, 130, 131, 139, 159, 160, 21 1, 243*, 244* Unger, H., 300, 351 * Unwin, D. M., 6, 100* Ursprung, H., 126, 178, 244* V Vaidyanatnan, C. S., 193, 194, 239*, 243* Vanable, J. W., Jr., 33, 114* Van Cassel, M.,, 44, 114* Van den Assem, J., 325, 351* Van der Kloot, W. G., 304, 351* Venard, C. E., 10, 115* Venkatachari, S. A. T., 32, 33, 40, 70, 71, 114* Verdier, M., 319, 349* Veronese, F. M., 120, 121, 125, 128, 130,200,226*, 240* Vining, L. C., 193, 239* Viola, M., 300, 310, 310, 342*, 343* Viscontini, M., 130, 172, 244* Vogel, A., 318, 345* Vuilleaume, M., 159, 174, 244*

W Wachmann, E., 197, 244* Wagner, R. P., 192, 229* Wajc, E., 317, 318, 319, 334, 335, 348*, 351 * Ware, G. W., 27, 28, 29, 105*, 114* Washizu, Y.,168, 169, 244* Watanabe, M., 121, 244* Watanabe, Y . , 121, 244* Weber, F., 131, 244* Weber, W., 57, 114* Wehner, R., 168,244* Weiant, E. A., 300, 303, 305, 306, 320, 348-350*

Weichert, R., 125, 228* Weidel, W., 119, 125, 127, 228* Weimorts, D. M., 194,238* Weiss, K. W., 118, 226* Weissbach, H., 193, 235*, 244* Wellso, S. G., 27, 103* Wenil, M., 137, 139, 244* Wessing, A., 120, 124, 126, 128, 131, 132, 159, 160, 178, 192, 208,209,212,231*, 244*, 245* Wever, R., 4, 98* Wiedbrauck, S., 314,352* Wiersma, C. A. G., 91,97* Wigglesworth, V. B., 1, 114*, 298, 352* Wilde, J., de, 319, 336, 337, 342*, 352* Wilkins, M. B., 24, 73, 74, 114* Willey, R. B., 59, 114* Williams, B. G., 62, 67, 68, 69, 73, 108*, 115* Williams, C. M., 11, 45, 86, 106*, 112*, 115*, 300, 303,304,312, 315, 321, 325, 347-349*, 351*, 352* W X g, A., 120, 124, 173, 174, 175, 176, 204, 205, 211, 227*, 245* Wilson, D. M., 308, 352* Wilson, E. O., 9, 115*, 300, 301, 352* Wilson, F., 14,19, 102* Wilson, L. P., 220, 245* Winfree, A. T., 18, 49, 50, 51, 73, 77,87,88,115* Wiss, O., 131, 244* Witkop, B., 130, 243* Wobus, U., 49,80, 115* Wolf, W., 320, 321, 322, 326, 345* Wolfram, R., 200, 245* Wolken, J. J., 165, 226* Wolsky, A., 194, 245* Wortham, S., 19, 102* Worz, O., 139,245*

367

AUTHOR INDEX

Wright, A. M.,310,346* Wright, J. E., 10, 115* Wulff, V.J., 91, 105* Wurtman, R.J., 4, 115* Wyatt, G.R.,219, 245*

Y Yamafuji, K., 202, 245* Yamamoto, R. T.,86, 106* Yamamoto, S., 192,239* Yamamoto, T.,202, 203,204,211,

219,234* Yamashita, O.,129, 197, 245* Yasuzumi, G.,162,245* Yoshida, K.,126, 244* Yoshida, M.,150, 165, 245*

Young, D., 2!i7,296* Youthed, G. J., 8, 74, 79, 80, 115* Z Zaretsky, M. I]., 285, 295* Zdirek, J., 304,352* Zeigler, R., 320,323, 352* Zelazny, B., 21, 115* Zeutzschel, B., 162, 163, 245* Zhantiev, R. I)., 269,294* Ziegler,I., 119, 159, 162, 164, 172,

245*,246*

Zillig, W., 131,228* Zimmerman, W. F., 17, 46, 47,49, 115* 72, 73, 109:*, Zimmermann, B., 131, 208, 230* Zwislocki, J. J., 285, 296*

This Page Intentionally Left Blank

Subject Index A

Abricta curvicosta, tymbal, 25 7 Acanthacris ruficornis fulva, ommochromes, 152 Accessory glands and female refractoriness, 327 andmale behaviour, 31 7-318 and oviposition behaviour, 330, 331,332 Acetylcholinesterase activity, rhythms, 31-33,91,92,95 Acheta domesticus, circadian rhythms acetylcholinesterase, 32 endocrine cells, 3 7 , 3 8 haemolymph metabolites, 30 locomotor activity, 3 , 6 control, 55,66-67 narcotic sensitivity, 24 pharmacological rhythms, 42 role of brain hormone, 60 Acrididae, female reproductive behaviour, 321,325 Acridiommatins as waste products, 177 chromatography, 140 distribution, 136, 138, 151, 152, 161-162 special data, 143 Actinomycin D, effect on circadian rhythms, 40,88 Adaptive colouration, ommochromes, 1 7 1

Adrenalin, and giant fibre response, 311 Adsorption of ommochromes, 138-139 Aedes, circadian rhythms, 3,16 Aedes’aegypti circadian rhythms inconstant light, 81 feeding, 8 oviposition, 1 2 hormonal control, female behaviour receptivity, 304-305,324 refractoriness, 326,327 Aedes taeniorhynchus, circadian rhythms eclosion, 78 haemolymph metabolites, 31 hatching, 16 in constant light, 79 pupation, 16-17,20,54 types of clock., 77,78,81 .4edes triseriatus, swarming rhythm, 10 Aeschna, ommochromes A . cyanea, 1 5 1 A.juncea, 151 A . mixta, 151 Aggregation of ommochromes, 138-139 Aglais (Vanessa) urticae, ommochromes, 1585 Agrotis comes, cmmochromes, 157

369

370

Amphetamine, and locomotor rhythms, 42 Anacridium aegyptium female sexualbehaviour, 319 ommochromes, 152 Anaesthesia, rhythmic response to, 24-25 Anagasta kiihniella, circadian response topheromones, 10 Anax imperator, ommochromes, 151 Animals other than insects amphipod, locomotor rhythm, 69 Anthomedusa, ommochromes, 150,165 Aplysia, circadian rhythms, 90-91,97 Aplysaa californica, driving oscillators, 81 Arachnid, locomotor rhythm, 339 Arthropoda, hormones and rhythm control, 66-71 bat, hawkmoth detection, 289-290 bird, driving oscillators, 89 Cambarus, locomotor rhythm,68 Carcinas maenas. locomotor rhythm, 68-69,73 cephalopoda, ommochromes, 150 crab, locomotor rhythm, 68, 73 crayfish locomotor rhythm, 68 mitochondrial membrane, 192 optic lobe rhythm, 91 Crustacea driving oscillator, 89 extra-retinal en trainmen t, 46 Crustacea, decapod, locomotor rhythm, 62 Daphnia magna, phototaxis rhythm, 1 3 earthworm, circadian rhythms, 60

SUBJECT INDEX

Animals other than insects-cont. Echiurida, ommochromes, 150 Ceocareinus lateralis, locomotor rhythm, 69 Heterometris fulvipes, locomotor rhythm, 40-41, 339 Limulus, ommidin, 161 mammals glutarate pathway, 132 kynurenine hydroxylase, 192 oestrous behaviour, 302 Neurospora, kynureriine hydroxylase, 192 Polychaeta, ommochromes, 150, 162 Protista, circadian rhythms, 39, 51,92 Protozoa, circadian clocks, 77 rat kynurenine formamidase, 189 sexual behaviour, 303 scorpion, circadian rhythms, 32, 40-41, 69-71, 339 sea hare, driving oscillators, 81 Sepia, ommins, 161 spider, slit sense organ, 274 Talitrus saltator, locomotor rhythm, 69 vertebrates glutarate pathway, 133 kynureninase, 193 tryptophan metabolism, 218 tryptophan oxygenase, 184, 189 Anopheles, Johnstone organ, 290-29 1 Anopheles gambiae, circadian rhythms behavioural phase-response curve, 80,81 bimodal locomotor rhythms, 7 feeding rhythms, 8 ffight phase-response curve, 48 oviposition, 12, 1 5

SUBJECT INDEX

Anopheles stephensi, genetics of clock, 74, 75 Anopheles superpictus, swarming rhythms, 10 Ant circadian rhythms, 9, 12 sound production, 262, 273 leaf-cutting, ultrasonic frequencies in sound, 257 stridulating, sound production radiation, 262-263 through earth, 254, 263 Ant-lion, circadian rhythms, 8, 79 Antenna, as sound receiver, 271, 290-29 1 Antheraea pernyi circadian rhythms eclosion, clock-gating, 52-53 eclosion, hormonal control, 340 protocerebral clock, 82-84 hormonal control adult behaviour, activation, 314-315 eclosion hormone, 302 female receptivity, 321 ommochromes, 154 Antheraea polyphemus, hormones eclosion hormone, 3 1 5 female receptivity, 321, 325 Anthonornus grandis, circadian rhythms, 26, 31-32 Anthranilic acids during metamorphosis, 202, 206 properties, 1 31-132 tumour induction, 221 Anticon'xa, phototaxis rhythm, 13 Aphid, circadian rhythms morph determination, 22 photoperiod measurement, 93 photoperiodic induction, 44, 47 response to pheromones, 11 A p k . ommochromes, 144, 161, 168

371

Apis mellifera absence of glutarate pathway, 133 feeding rhythms, 9 Apis mellifica kynurenine.3-hydroxylase, 19 1 ommochrornes, 158 Aporia cratuegi, ommochromes, 156, 177 Aruschnia laeimzu, ommochromes, 155 Argynnis papliia, ommochromes biosynthesis, 195 in meconia, 177 localization, 155 as pattern pigments, 170 Arista, sound reception by, 291 Attu, sound radiation of, 262-263 through earth, 254, 263 A t t a sexdens, ommochrome distribution, 158 Autonomous rhythmicity, individual cells, 92, 9 5 Azide, sodium, circadian response to, 26 Azinphosmethyl, circadian response to, 27, 28 B Baffle, use in sound emission, 264-267 Bee circadian rhythms, 15, 25 kynurenine-3-hydroxylase, 191, 193 Bee, honey feeding rhythm, 9 ommochromes deposition, 162 3-hydroxy kynurenine, 129, 130 asscreening pigments, 168-169

SUBJECT INDEX

372

Beetle circadian rhythms daily growth layers, 21 endocrine cells, 34 in ERG, 9 1 insecticide susceptibility, 27 temperature effects, 73 oviposition behaviour, 328 sound emission, 265 Beetle, carabid, locomotor rhythms, 338 Beetle, Colorado, migratory behaviour, 336 Beetle, dung, orientation rhythm, 13 Beetle, scarab, migratory behaviour, 33 7 Beetle, water, 8-hydroxy-quinaldic acid, 131 Behaviour, see Hormonal control Behavioural circadian rhythms changes in responsiveness, 12-15 control by oscillator, 54-7 1 brain hormones, 57-60 cardiaca-allata complex, 57 optic lobes, 61-63 other arthropods, 66-71 suboesophageal ganglion, 55-57 ventral nerve cord, 60-61 feeding rhythms, 8-9 locomotor activity, 6-8 sexual rhythms, 9-12 Biliverdin, in colour changes, 174-175, 176 Bimodal entrained rhythms, 6-7 Biochemical circadian rhythms, 29-34 Biston cetularia, ommochrome distribution, 156 Blaberus craniifer, circadian rhythms, 58, 80 Blaps, suboesophageal ganglion, 56

Blaps mucronata, locomotor rhythm, 73 Blatta orientalis, ommochrome distribution, 153 Blattella germanica female receptivity, 322, 323 insecticide susceptibility rhythm, 26 oxygen consumption rhythm, 23 Blattella vaga, female receptivity, 322,323 Blood ions, rhythms, 95 neuropharmacological rhythms, 41 sugar level, 30, 31, 92, 95, 299 Blood-sucking Diptera, feeding rhythms, 7 Blowfly daily growth layers, rhythms, 22 kynureninase, 132 landing response, rhythm, 15 metamorphosis, tryptophan during, 206 Bollworm, pink, circadian rhythms hatching, 1 5 insecticide susceptibility, 27, 28 oviposition, 1 2 X-ray sensitivity, 29 Bombyx mori hormonal control cocoon construction, 314 ecdysone, larval activity, 312 female receptivity, 321 oviposition, 328, 330 tryptophan-, ommochrome pathway absence of glutarate pathway, 133 anthranilic acids, 132 cinnabarinic acid. 136 enzyme ontogeny, 214, 217,

218

373

SUBJECT INDEX

Bombyx mori-cont. tryptophan + ommochrome pathway-cont. 3-hydroxy kynurenine, 127130 kynureninase, 193 kynurenine, 125, 127 kynurenine-3-hydroxylase, 191,192 kynurenine formamidase, 190 ommochrome localization, 154 protein tryptophan, 122 quinoline derivatives, 130 tryptophan balance, 219 tryptophan during metabolism, 202-204, 210 Brain and circadian rhythms cells, 35, 37, 38, 85 cells, photosensitivity, 44, 46 eclosion rhythm, 340 extract, 41 hormones, 53, 54, 57-60, 63-67, 85 locomotor activity, 63-65, 70, 338 corpora allata system, and ommochrome synthesis, 174 Bug, milkweed daily growth layers, 20 locomotor rhythms, 7 migratory behaviour, 335 Bug, pyrrhocorid, oviposition behaviour, 328 Bupalus piniarius, ommochromes, 156 Bursa copulatrix, in oviposition behaviour, 329, 330, 332 Bursicon, gated release of, 52 Butterfly, ommochromes daily growth layers, 21 deposition, 162

Butterfly, onimochromes-cont. extraction, 135 as pattern pigments, 170 Byrsotria fumigata, sexual behaviour. 320, 321, 323 C Callandra p z n i a (Sitophilus granarius), locomotor rhythms, 7 “Calling” posture, hormonal control, 300, 303, 325, 340 Calliphora ery t hrocep hala ecdysone, pre-pupal, 86 tryptophan -+ ommochrome pathway enzyme ontogeny, 2 13 free tryptophan, 124 3-hydroxy-kynurenine, 127, 128 kynurenine-3-hydroxylase, 191.,192 during nietamorphosis, 208, 211 ommochrome biosynthesis, 194-196 ommochrome deposition, 162164 ommochrome localization, 158,160 screening pigments, 166-168 xanthommatin, 135 Calpodes ethlius, daily growth layers, 21 Cancer, and rhythmic hormone secretion, 42 Carabus nemoralis, endocrine cell rhythm, 34 Carausius morosus circadian rhythms colour change, 71-72 locomotor activity, 77 pharmacological, 41

SUBJECT INDEX

374

Gzrausius rnorosus-cant. ommochromes as pattern pigments, 171, 172 as waste products, 178 association with melanin, 17 1, 172 distribution, 153 3-hydroxy kynurenine, 12; in morphological colour change, 173 turnover, 196 oviposition behaviour, 328 Cardiac activity, regulation, 299 fl-Carotene, role in entrainment, 47, 49 Carotenoids, in colour changes, 174 Caterpillar colour change, 175 phototactic rhythm, 1 3 Celerio euphorbiae, ommochromes, 167-168 Cell multiplication, and tryptophan metabolism, 21 7 Cellular circadian rhythms clocks, 93-97 endocrine cells, 34-39 tissue culture, 39-40 Central nervous system, hormonal effects on, 305-309 copulatory movements, 306-307 level of spontaneous activity, 305-306 pre-eclosion behaviour, 307-308 Ceratitis capitata, ommochrome distribution, 157 Cerura vinula hormonal control of cocoon construction, 314 pigmentation, 313 ommochromes as pattern pigments, 170-172 as waste products, 177, 178 egg, tryptophan metabolites, 199

Cerura vinula-cont. ommochromes-cont. enzyme ontogeny, 217 free tryptophan, 124 3-hydroxy-kynurenine, 127 in morphological colour change, 173, 175 larva, tryptophan metabolites, 200 localization, 157, 160 metamorphosis, tryptophan metabolites, 204-206,

211 oxidation-reduction, 165-166 turnover, 196 Choerocampinae, pilifers, 289-290 Cholinesterase inhibitors, control of susceptibility, 71 Chorthippus curtipennis circadian rhythms, 12, 4 5 female refractoriness, 326 Chorthippus paraltelus, sexual behaviour, 318, 324 Chromatography ommochromes, 135-137, 139140 tryptophan metabolites, 120-121 Chrysopa ear, 288-289 ommochromes, 170, 176 C. carnea, ommochromes, 153 C. vulgaris, ommochromes, 153 Cicada, sound emission baffled sound radiator, 20-25 click mechanism, 257 radiating tymbal, Q value, 260 Cicindela sp., ommochrome distribution, 159 Cinnabarinic acid distribution, 136, 161 spectral data, 144 Circadian rhythms hormonal control, 337-340 principles, 2-6 I

SUBJECT INDEX

Circadian rhythms-cont. timing processes, 43-9 1 controI of behavioural rhythms, 54-71 control of gated events, 52-54 control of other rhythms, 71-72 genetics, 74-76 mechanisms of driving oscillators, 81-91 phase adjustment, 47-51 photoreception, 44-47 temperature effects, 72-74 two clock types, 76-81 types, 6-43 adult eclosion, 17-20 biochemical, 29-34 cellular, 34-40 changes in responsiveness, 12-15 daily growth layers, 20-22 feeding, 8-9 hatching, 15-16 insecticide susceptibility, 25-29 locomotor activity, 6-8, 337-339 metabolic, 23-24 narcotic sensitivity, 24-25 pharmacological, 40-42 photoperiodism, 22 pupation, 16-17 sexual, 9-12 tumours, 4 2 4 3 X-ray sensitivity, 29 Cliturnnus extradeatatus, circadian rhythms, 37, 328 Clock types, 93-97 Clunio marinus, circadian rhythms eclosion, 19-20 genetics of clock, 74, 75 Cockroach behaviour, hormonal control circadian locomotor rhythms, 337-338 AIP-13

375

Cockroach-cont. behaviour, hormonal control-cont. female sexual behaviou, 321-323, 325 male sexual behaviour, 306-307, 320 oviposition, 328 phallic nerve-stimulating hormone, 300,332 circadian rhythms brain hormone, 57-60 constant light, 80 control system, 63-66 driving oscillator, 89 endocrine cells, 34-37 entrainment, 44-45, 49, 51 insecticide susceptibility, 26,27 locomotor activity, 3, 6-7, 93 narcotic sensitivity, 25 optic lobes, role, 61-63 oxygen consumption, 23, 24 sodium and potassium, 34 optic lobe clclck, 81-82 pharmacologically active substances, 40 photosensitiw ty, 44-45 suboesophageal ganglion, role, 55-57 temperature effects, 72, 74 tumour induction, 42-43 ventral nerve cord, role, 60-61 glutarate pathway, absence of,133 Cocoon spinning circadian rhythm of, 340 hormonal contral, 304, 314 tryptophan metabolism, 202-204 Coleoptera daily growth layers, 21 ommochrome distribuJion, 159 Colias edusa, ommochromes, 156 Colour change ommochromes, 171, 173-176, 204 rhythms, 71-72

376

Copulatory movements, hormonal control, 306-307 Corixn, phototaxis rhythm, 13 Corn borer, European, oxygen consumption rhythm, 24 Corpora allata and behaviour cocoon spinning, 3 14 female behaviour, 320-324, 326 juvenile hormone, 297, 299 larva, 312 male behaviour, 316-320 migratory behaviour, 334-337 oviposition, 329 reproductive behaviour, 331, 332 and circadian rhythms cellular rhythm, 34, 35 locomotor activity, 338 silkmoth calling, 11 and ommochrome synthesis, 174 Corpora cardiaca and behaviour, 299 CNS spontaneous activity, 306 female receptivity, 325 male behaviour, 306-307, 320 oviposition, 3 28-330 walking movements, 304 and circadian rhythms cardiaca-allata complex, 57, 63 cellular, 35 silkmoth calling, 12 COSSUS C O S S U S , ommochrome distribution, 153 Courtship rhythm, 8 0 sound emission, 264-265 Cricket circadian rhythms, 6 acetylcholinesterase rhythms, 32

SUBJECT INDEX

Cricket-cont. circadian rhythms-cont. brain hormone, 60, 63 brain cell changes, 85 control of locomotor rhythm, 38, 55, 66-67, 85 effect of drugs, 42 endocrine cell ultrastructure, 37 entrainment phase adjustment, 51 haemolymph sugar, 30 in constant light, 79 narcotic sensitivity, 24 optic lobes, role, 61 RNA synthesis, 15, 38 sexual rhythms, 12 stridulation rhythms, 45, 61 female receptivity, hormonal control, 321 sound communication conductiofi by non-tympanal route, 277 harp, properties resonant sound radiators, 263 stridulation mechanism, 254 tympana1 vibrations, 283-285 tryptophan' ommochrome pathway absence of glutarate pathway, 133 in larva, 200 kynurenine, 125 turnover of ommochromes, 197 Cryp tommidin distribution, 136, 137, 152, 161 spectral data, 144 C-type cells, pars intercerebralis, 319 Culex pipiens circadian rhythms, 23, 78 ommochromes, 157

377

SUBJECT INDEX

Cultured tissues, autonomous rhythmicity, 92, 95 Cuticle deposition, rhythm of, 71, 89; 299 daily growth layers, 20-22 temperature effects, 72 Cyanide, circadian response to, 26 Cyclorrhaphous flies, sound reception, 291 Cytochrome system, ommochrome synthesis, 194 Cytoplasm, circadian changes, 35-37

D

Dacus tryoni, sexual rhythms, 10, 79 Dasyneura, indole acetic acid, 132 Defence, by 8-hydroxy-quinaldic acid, 131 Degradation reactions, ommochromes, 145-150 Deilephila (Pergesa) elpenor, ommochromes, 156 Deposition of ommochromes, 162164 Dermaptera, daily growth layers, 20 Detoxication mechanisms, 3-hydroxy kynurenine, 130 Development circadian rhythms, 15-22 adult eclosion, 17-20 daily growth layers, 20-21 hatching rhythms, 15-16 photoperiodism, 22 pupation, 16-17 colour changes during, 173-176 tryptophan metabolism during, 197-2 20 egg and embryo, 197-199 larva, 199-201

Development-con t. tryptophan metabolism duringcont. metamorphosis, 201-212 ontogeny of enzymes, 212-218 tryptophan balance, 218-220 DDT, circadian response to, 26, 27 DDVP, circadian response to, 26 Diapause hormone, and 3-hydroxykynurenine, 129 Dichlorvos, circadian response to, 26 Dicofol, circadian response to, 26 Dictyoptera daily growth layers, 20 ommochrome distribution, 153 Die1 rhythms, 4, 6-7 Dieldrin, circadian response to, 27 Diffraction, sound reception, 275-276 Dihydro-xanthommatin in colour change, 175-176 distribution, 157, 158 Redox properties, 140, 142 spectral data, 143 Dimetilan, circadian response to, 26 Dioctria atricapilla, ommochromes, 157 Diploptera punctata, female receptivity, 321 Diptera circadian rhythms in constant light, 79 daily growth layers, 2 1 ecdysis, 52 feeding, 8 locomotor activity, 7 sexual, 9 type I clocks, 79-81 hormonal control ,behaviour female, 323-324, 326 male, 320

378

Diptera-cont. ommochromes deposition, 162 distribution, 157-158, 161 3-hydroxy kynurenine, 128 xanthommatin, 136, 160 Dissosteira longipennis, absence of glutarate pathway, 133 Disulfoton (Di-syston), circadian response to, 27-28 DNA association with tryptophan, 222-223 during metamorphosis, 202, 204 DNP, circadian response to, 26 Dociostaurus maroccanus, ommochromes, 152 DOPA, in ommochrome synthesis, 194 Dormancy, metabolic level rhythm, 71 Driving oscillators, mechanisms, 5! 81-91 cockroach optic lobe clock, 81-82 Drosophila clock, 85-87 silkworm protocerebral clock, 82-85 Drosophila circadian rhythms clock, Rensing’s model, 85-87 clock, two types, 77, 78, 80 courtship, 10 eclosion, 17-18, 96 endocrine cells, 35, 36, 37 entrainment, 85 genetics of clock, 75-76 5-HT, 40 insecticide susceptibility, 2 7 locomotor activity, 27 narcotic sensitivity, 25 oviposition, 12 oxygen consumption, 27, 96

SUBJECT INDEX

Drosophila-cont. circadian rhythms-cont. phototaxis, 13 pupation, 53-54 synchrony of different tissues, 89 temperature effects, 72 hormonal control female behaviour, 326, 332 oviposition behaviour, 330 ommochromes absence of glutarate pathway, 133 biosynthesis, 195 deposition, 162-164 formyl kynurenine, 125 3-hydroxy kynurenine, 128, 129 kvnurenine, 125-126, 199 localization, 160 metamorphosis, tryptophan metabolites, 208, 209 as pattern pigments, 172 as screening pigments, 168 as waste products, 1 7 7 sound communication emission, mechanism, 264-265 reception by arista, 291 Drosophila melanogaster circadian rhythms eclosion, 1 7 , 4 6 , 4 7 5-HT, 33-34 oxygen consumption, 23, 24, 29,81 salivary gland cells, 39 X-ray sensitivity, 29 female receptivity, control, 324 tryptophan -+ommochromepathway detrimental effects of ommochromes, 220-221 enzyme ontogeny, 212, 214,

217

SUBJECT INDEX

379

Drosophila melanogaster-cont. tryptophan + ommochrome pathway-cont. 3-hydroxy kynurenine, 128, 129 kynurenine formamidase., 189, 190 kynurenine-3-hydroxylase, 191,192 ommochrome distribution, 157 quinoline derivatives, 131 tryptophan content, 122, 124-125 Drosophila pseudo0 bscura, circadian rhythms eclosion, 17, 19, 48-51 emergence, 73 phase-response curve, 48-5 1 temperature effects, 73 Drosophila victoria, eclosion rhythm, 19 Drosophila uirilis, tryptophan metabolism, 122, 190 Dung fly, yellow, female behaviour, 320 Dytiscus fasciventris, circadian rhythms, 91

E Ear as sound receiver, 271-291 atypical ears, 288-291 forces acting on ears, 274-275 influence from surroundings, 275-279 parameters of sound, 271-274 receptor organ, 285-288 tympanal vibrations, 279-285 Ecdysis, rhythm of, 52, 54 Ecdysone and apolysis, 299

Ecdysone-cont. and nuclear rhythm, cultured glands, 40 and premetamorphic behaviour, 313 and promotion of metamorphosis, 303 gating of, 53 in circadian clock, 85-87 in migratory behaviour, 334, 335,337 Eclosion circadian rhythm of, 17-20, 91, 95,96,340 entrainment, 46 phase-response CUNe, 48 protocerebral clock. 82-85 hormone, 299 mcidifyer effects, 302, 303 neurophysiological studies, 3 0 7-308 regulation, 3 14-316 releaser effect, 305 Egg ommins, 161 tryptophan metabolism, 197-199 Embryo, tryptophan metabolism, 195-199 Endocrine system, 298-300 and tumour induction, 42-43 cells, circadian rhythms, 34-39 Endoplasmic reticulum, circadian rhythm of, 37 Endrin, circadian response to, 27 Entrainment of rhythms, 4-5,43-52 Enzymes. kynurenine pathway, 179-193 kynureninase, 193 kynureninase” transaminase, 193 kynurenine formamidase, 189 kynurenine-3-hydroxylase,189193 tryptophan oxygenase, 180-189

380

Ephedrine, and locomotor rhythm, 42 Ephestia kuhniella, tryptophan + ommochrome pathway anthranilic acids, 132 detrimental effects, 220 egg, tryptophan metabolism, 197 enzyme ontogeny, 214 kynurenine content, 125: 126 kynurenine-3-hydroxylase, 19 1, 192 ommins, 140, 144 ommochromes as pattern pigments, 172 early experiments, 119 biosynthesis, 195 deposition, 163-164 in testis sheath, 169 localization, 154, 160 quinoline derivatives, 130, 131 “skotommin”, 135 tryptophan content, 122, 124 tryptophan oxygenase, 180-185 Ep h ippiger, stridulation rhythms , 45 Epiphyas postvittana, circadian response t o pheromones, 11 Epidermis, tryptophan oxygenase, 184 Euthystira bmchyptera, female sexual behaviour, 322, 323, 325-326 Excreta, ommochromes in, 161 Excretory pigments, rhodommatin, 136 Excretory rhythm, 92 Eye and circadian rhythms, 69, 90-91 as photoreceptor for entrainment, 44-47 rhythm of action potential, 90-9 1 ommochromes as pattern pigments, 170

SUBJECT INDEX

Eye-cont. ommochromes-cont. as screening pigments, 166 distribution, 151-158, 160161 kynurenine, 192 ommidins and ommins, 136 Eyestalk, and crustacean locomotion rhythm, 62, 67-69 F Famesyl methyl ether, and female receptivity, 321, 322 Fat body cells, circadian rhythm, 35 tryptophan+ ommochrome pathway kynurenine, 126, 192 kynurenine transaminase, 193 ommochromes, 161, 175 tryptophan oxygeriase, 184, 185 Fat tissues, effect on sound conduction, 278 Feeding, circadian rhythms of, 8-9, 95 Female sexual behaviour, hormonal control receptivity, 3 20-32 5 refractoriness, 325-32 7 File, cricket stridulation, 255-257 Filters, acoustic, body parts as, 276-278 Firefly flashing, hormonal control, 309-311 Flea, water, phototaxis rhythms, 1 3 Flight central motor programs, 308 phase-response curve, 48 Fluorescent tryptophan metabolites, 120-132 anthranilic acids, 131-132

SUBJECT INDEX

Fluorescent tryptophan metab olites-cont. formyl kynurenine, 125 3-hydroxy kynurenine, 127-130 kynurenine, 125-126 methodology, 120-122 quinoline derivatives, 130-131 Fluoride, circadian response to, 26 Fly anthranilic acids, 132 ecdysone titres, 87 3-hydroxy kynurenine, 127 Formyl kynurenine, 125 Frequency multiplication mechanisms click mechanism, cicada, 257 stridulation, 254-257 Fructose diphosphate aldolase, 216 Fruit fly, sexual circadian rhythms, 10,79

Galeruca tanaceti, oviposition behaviour, 328 Gall induction, by IAA, 132 Galleria, nuclear volume rhythm, 40 Galleria mellonella hormones and behaviour, 314, 321 ommochromes, 154 Ganglia ommins, 160, 161 ommochromes, pigmentation by, 169 Ganglion abdominal and oviposition behaviour, 328 neurosecretory cells, 299 photosensitivity, 44

381

Ganglion-cont. suboesop hageal arid 3-hydroxy kynurenine, 129 arid tumour induction, 42-43 circadian rhythm of cells, 35, 37,38 role in behavioural rhythms, 55-57, 63, 65, 67, 70 role in locomotor rhythms, 337-338 role in oviposition, 328 thoracic neurosecretory cells, 299 role in oviposition. 328 Gated circadian rhythms, 5 clocks controlling, 52-54 developmental, 9 1-97 Gene dosage, and tryptophan oxygeriase activity, 185-189 Genetic!; of clocks, 74-76, 87 Geotrupes sylvaticus. orientation rhythm, 13 Gerris fucustris, ommochromes, 153 Glossina morsitans, circadian rhythms locomotor activity, 3, 7 mating, 79 visual response, 13, 14, 1 5 Glowing, endogenous timing of, 12 Glutaratt: pathway, absence of, 132-134 Golgi region, circadian rhythms, 37 Gornphocerus rufus, hormonal control female sexual behaviour, 321-322, 325 juvenile hormone, 302.304 male st:xual behaviour, 320 Gonads, tryptophan oxygenase, 184 Granules, pigment, ommochrome deposition, 162-164

382

Grasshopper circadian rhythms oviposition, 12, 45, 79 oxygen consumption, 23 sexual, 12 glutarate pathway, absence, 133 hormones and female receptivity, 324 and male sexual behaviour, 3 17-320 juvenile hormone, 302 Gromphadorhina brunneri, sexual behaviour, 320, 323 Growth layers, daily, rhythms of, 20-22 Gryllidae, stridulation mechanism, 254 Gryllotalpa vinae, sound production, 265-268 Gryllotalpa vulgaris, ommochromes, 15 1 Gryllus assimilis, absence of glutarate pathway, 133 Gryllus bimaculatus female receptivity, 321 tryptophan + ommochrome pathway 3-hydroxy kynurenine, 128 kynurenine, 125 kynurenine formamidase, 190 larva, 200, 201 ommins, 144 ommochromes, 151, 161 pattern pigments, 1 7 2 Cry llus campes tris sound communication baffled sound radiator, 265 ear, frequency selectivity, 283 harp, properties, 258-262 resonant sound radiation, 263 singing efficiency, 268 Gryllus domesticus, ommochromes, 151

SUBJECT INDEX

Guide, sound, use in sound emission, 264-267 Gut extract, rhythmic sensitivity to, 31 ommochromes, 161, 175, 176-177 tryptophan oxygenase, 184 H

Habrobracon juglandis, tryptophan + ommochrome pathway enzyme, ontogeny, 214 in metamorphosis, 209, 21 1 kynurenine transaminase, 193 ommochrome localization, 158 tryptophan oxygenase, 182-183, 185 quinoline derivatives, 131 Haemol ymph circadian variation ' i n metabolites, 30-3 l 3-hydroxy kynurenine, 127 ommochromes, 161, 176 Haemophysalis leporispalustris, feeding rhythms, 9 Hair receptors, sound, 2 7 1, 290-29 1 Harp, cricket baffled sound radiator, 265 mechanism, 255 sound radiating properties, 260-262 vibrational properties, 258-260 Hatching, circadian rhythm of, 15-16,91, 9 5 Hawkmoth, sound, 271, 289-290 Heliconiinae, wing pigmentation, 129-130 Heliconius, ommochromes, 156, 168,170 H. errato, 162

SUBJECT INDEX

Heliothis zea, insecticide susceptibility rhythm, 28 Hemimetabola, larva, tryptophan, 199-201 Hemiptera daily growth layers, 20 ommochrome distribution, 153 Heptacosadiene, circadian variations in, 31 Hestina japonica, ommochromes, 155, 167, 176 Holometabola absence of glutarate pathway, 133 larva, behaviour, 3 11 metamorphosis, tryptophan, 201212,219 Homorocoryphus, sound conduction, 277 Hormonal control of behaviour, 297-352 during life history; 31 1-340 activation of adult behaviour, 3 14-315 circadian rhythms, 337-340 larva, 311-314 migration and orientation, 333-337 reproductive behaviour, 3 16333 effect on behaviour, 300-303 modifier effects, 301-303 releaser effects, 303 endocrine system, 298-300 mode of action, 303 neurophysiological studies, 305311 CNS, 305-309 peripheral action, 309-3 11 Hormones and circadian rhythms cancer induction, 42 synthesis, cycles, 91, 92, 95 Horn, mole cricket, sound emission, 265-267

383

Housefly insecticide susceptibility rhythms, 27 omxriochromes, 165 Humidity, sound attenuation, 26'9-271 Hyalophora cecropia circaldian rhythms, 11-12, 52-53. 340 hormones and behaviour adult, 315 cocoon construction, 314 eclosion, 307-308 female receptivity, 321, 325 juvenile hormone, 304 oviposition, 300, 329, 330 ommochromes ganglia pigmentation, 169 localization, 154, 160 ommins, 141 Hydrocarbons, circadian variations in, 3 1 Hydrocyrius, daily growth layers, 21 3-Hydroxy anthranilic acid, 131-132 and absence of glutarate pathway, 133-134 determination, 12 1 in ornmochrome biosynthesis, 136, 194-195 3-Hydroxy kynurenine, in tryptophan + ommochrome pathway, 127-130 and ommochrome deposition, 164 as ommochrome precursor, 136 during colour change, 176 determination, 121 detrimental effects, 220 early rcxogniiion, 119 egg, 197-199 3 - h y d r ~ y kynureninase, 189193

384

SUBJECT INDEX

3-Hydroxy kynurenine-cont. in larva, 200 in metamorphosis, 202-205, 207-212 in ommochrome biosynthesis, 195, 196 tryptophan balance, 219 5-Hydroxy tryptamine and locomotor rhythm, 42 and oviposition behaviour, 329 circadian rhythms of, 33 die1 rhythm, 40 Hymenoptera circadian rhythms, 9, 21 ommochrome distribution, 158 Hypodermis ommochromes, 160, 161 photosensitivity, 44

I Ilybius fenestratus, 8-hydroxyquinaldic acid, 131 Znachis (Vanessa) io, ommochrome distribution, 155 Insecticide susceptibility rhythms, 25-29. 9 1 Integument 3-hydroxy kynurenine, 127 ommochromes, 161, 175 Zphita limbata. oviposition behaviour, 328 Ithomiinae, wing pigmentation, 129-130

J Johns tone organ, behaviour , 290-291 Juvenile hormone, 299 and cocoon construction, 314 and female behaviour, 320 324, 327

Juvenile hormone-cont. and male behaviour, 3 17-320 and migratory behaviour, 334-337 and reproductive behaviour, 331-333 in larva, 312-313 mode of action, 303-305 modifier effect, 302 K Kynurenic acid, 130-131 as waste product, 178 in colour changes, 174 in egg, 199 in larva, 200 in metamorphosis, 206 Kynureninase, 131, 179, 193 Kynurenine, 125-126 early recognition, 119 in egg, 197, 198 in larva, 200 in metamorphosis, 202-209 pathway, enzymes, 179-193 kynureninase and kynurenine transaminase, 193 kynurenine formamidase, 189 kynurenine-3-hydroxylase, 189-193 tryptophan oxygenase, 180189 tryptophan balance, 219 Kynurenine formamidase (aryl formylamine amidohydrolase), 179, 180, 189,213 Kynurenine-3-hydroxylase, 189193 in ovary, 197 ontogeny, 213-215,217 tryptophan balance, 219 Kynurenine transaminase, 179, 193

385

SUBJECT INDEX

L Labial palps, as sound detectors, 289-290 Lac larva, ommochromes, 162 Laccifer lacca, ommochromes, 153 Lacewing, ear, 274, 288-289 Larnpyns noctiluca, glowing, endogenous timing of, 12 Laphnu gibbosa, ommochromes, 157 Larva behaviour, hormonal control, 311-314 activity, 312 cocoon construction, 314 onset of metamorphosis, 312-313 ecdysis, clock gating, 54 tryptophan metabolism, 199-201 Larviposition, circadian rhythms of, 12 Laspcyrcsia pornonella, oxygen consumption rhythms, 24 Leiobunurn longipes, 5-HT circadian rhythm, 33 Lepidoptera circadian rhythms oxygen consumption, 24 sexual, 9 hormonal control of behaviour cocoon construction, 3 14 female receptivity, 321 larva, 312 moulting, 3 12 ommochromes as pattern pigments, 170 as waste products, 176-177 distribution, 153-157, 160161 in colour changes, 175, 176 rhodommatin, 136 xanthommatin, 138

Leptiiiotarsa decernlineata, hormones female behaviour, 3 19 migratory behaviour, 336 oviposition behaviour, 328 Leucophaea rnaderae circadian rhythms brain hormone, 59 endocrine cells, 3 7 optic lobes, 61 photoreception, 44 vzntral nerve cord, 60 female receptivity, 322, 323 Light and dark rhythms, 3-5 Locomotor activity rhythms, 6-8, 31., 95, 337-339 Locust circxdian rhythms chitin lamellogenesis, 77 daily growth layers, 20, 21 locomotor activity, 7 marching, 34 hormonal control of behaviour central motor programs, 308 female sexual behaviour, 31 7-320 grt-garious and migratory behaviour, 333-335 laival activity. 312, 313 sound communication acoustic filters, body parts as, 276-277 forces acting on ears, 274-275 fat and soft tissues, effects of, 278 parameters of sound, 271-273 receptor organ, behaviour, 285, 287-288, 290 tympana1 vibrations, 279-284 tryptophan metabolism acridiommatins, 140 egg,, 199 eye pigments, 134

386

SUBJECT INDEX

Locust-cont. tryptophan metabolism-cont. 3-hydroxy anthranilic acid, 132 kynurenine transaminase, 193 ommatins, 135 ommochromes in colour change, 174-175 ommochrome deposition, 162 ommochrome localization, 161 waste products, 177 Locusta accessory gland, role, 3 18 allotropic neurosecretion, 31 7 Locusta pegaria, colour changes, 175 Locusta migratoria L. migratoria cinerascens. male sexual behaviour, 319 L. mipatoria mipatorioides male sexual behaviour, 3 18, 319 migratory behaviour, 333-335 locomotor activity rhythms, 7 ommochromes, 152 Locusta solitaria, colour change, 175 Lucilia caesar, ommochrome binding, 165 Lucilia sericata, landing response rhythm, 15 Luciola, hormonal control of flashing, 300, 309-311 L. italica, 309 L. lustianica, 309 Lunar periodicity, ant-lion, 8 Lysergic acid, and locomotor rhythms, 42 M Mapcicada, 257

tymbal, mechanism,

Malathion, circadian response to, 27 Male sexual behaviour, hormonal control, 316-320 cockroach, 320 Diptera, 320 grasshopper and locust, 317-320 Malpighian tubules self-sustaining oscillation, 92 tryptophan + ommochrome pathway accumulation of kynurenine, 126 during metamorphosis, 203, 208,209,212 3-hydroxy kynurenine, 127, 128,192 in larva, 200 ommochromes, 159, 161, 175, 177, 179 tryptophan oxygenase, 184 Manduca sexta eclosion hormone, 299 gating of pupation, 53 larval behaviour, hormones, 312-313 Mantis religiosa, ommochromes in colour changes, 171, 174-175 localization, 153 Matrone, and female refractoriness, 300, 302, 304-305, 326, 327 Mating rhythms, 12, 9 5 Mealworm, oxygen consumption rhythm, 24 Meconium, ommochromes, 154157,160, 176 Megoura viciae circadian response t o pheromones, 11 morph determination, 22 photoperiod measurement, 93 Melanin, relationship with ommochromes, 171-172, 174, 175, 179, 194

387

SUBJECT INDEX

Melolontha melolontha, migratory behaviour, 33 7 Melolontha vulgaris, ommochromes, 159 Metabolic circadian rhythms, 23-24 Metamorphosis, tryptophan metabolism, 201-212 Midge, circadian rhythms eclosion, 19-20 swarming, 1 0 Migration and orientation, hormonal influence, 333-337 Mimas tiliae, hormones larval behaviour, 312 migratory behaviour, 337 Mirror, cricket, vibrational properties, 258 Mite, spider, circadian rhythms insecticide susceptibility, 26 narcotic sensitivity, 24 oviposition, 12 Modifier effects, hormones, 301-303 Mole cricket, horn, 265-268 Morph determination, aphid, 22 Mosquito circadian rhythms clock types, 78 eclosion, 20, 78 feeding, 8 haemolymph metabolites, 31 hatching, 16 in constant light, I9 locomotor activity, 79 oviposition, 12 oxygen consumption, 23 pupation, 17 swarming, 10 hormonal control of behaviour corpora allata, 324 female receptivity, 304-305, 324 female refractoriness, 326

Mosquito-cont. hormonal control of behaviourcont. matrone, 300, 302, 304-305, 326, 327 oviposition, 330 Johnstone organ, 290-291 sound reception, 271 Moth, circadian rhythms, 19, 42

Moth apple, Circadian response to pheromcnes, 11 cabbage looper, circadian response to pheromones, 10 codling, oxygen consumption rhythm, 24 flour, circadian response to pheromones, 10 linden. larval behaviour, 312 noctuid circadian r'zsponse to pheromones, 1 0 ear, 285-28'7 flight rhythin, 338 tryptophan in neurosecretory d s , 33 sphinx, larval hehaviour, 312-313 Moulting larva, decreased motor activity, 312 tryptophan metabolism, 200 Miiller-organ, locust ear, 28 1, 287-288 Murelron obscurus, feeding rhythm, 8 Musca dornestica circadian rhythms insecticide susceptibility, 27 mating, 79 female sexual behabiour, 324, 327 ommochromes as screening pigments, 166

388

Musca dom estica-con t . ommochromes-cont. localization, 157, 160 quinoline derivatives, 131 Muscles, hormonal control of histolysis, 3 15 Mushroom bodies, role in circadian rhythms, 65

SUBJECT INDEX

Odonata daily growth layers, 20 ommochromes acridiommatin, 136 as pattern pigments, 170 distribution, 151, 162 in colour change, 174 Oedipoda coerulescens, ommochromes, 152 Oedipoda miniata, female sexual N behaviour, 319 Ommatin D Narcotic sensitivity rhythms, 24-25, biosynthesis, 195 91,92 deposition, 162 Nasonia vitripennis, diapause inducdistribution, 155-157, 160-161 tion, 22 in colour changes, 175-176 Nauphoeta cinerea, female sexual in egg, 199 behaviour, 322,323,325 in meconia, 176-177 Neurohormone D 1 , 300 redox properties, 140 Neuroptera spectral data, 143, 147 ear, 288-289 Ommatins, 134-135 ommochromes, 153, 170, 176 distribution, 136, 138 Nicotinic acid, need for, 133 in meconia, 176-177 Noctua pronula, flight rhythm, 338 splitting, 138 Nomadacris septemfasciata Ommidins female sexual behaviour, 3 17 biosynthesis, 195, 196 ommochrome distribution, 152 chromatography, 140 Noradrenalin, and firefly flashing, deposition, 162 300,310 distribution, 136, 137, 151,152, Nucleic acids, association with 161 tryptophan, 222-223 solubility, 138 Nuclear circadian changes spectral data, 144,149 membrane, 37 Ommins volume, 34-36,71, 85, 86, 89, aggregation and adsorption, 139 91,95 as screening pigments, 167 Nymphalid butterfly biosynthesis, 195, 196 ommatin excretion, 134 chromatography, 140 ommochrome distribution, 160 deposition, 162, 163 degradation reactions, 145-147 distribution, 136, 137, 151-161 0 in colour change, 173, 174 Ocellus, as photoreceptor for enin egg, 198 trainment, 44-46 redox potential, 141

389

SUBJECT INDEX

Ommins-cont. spectral data, 144, 149 Ommochromes, see Tryptophan ommochrome pathway Oncopeltus fasciatus circadian rhythms daily growth layers, 20, 21 feeding, 8 locomotor activity, 7 oviposition, 8, 12, 72 sexual, 12 temperature effects, 72 Oogenesis, and juvenile hormone, 299 Optic lobes clock, mechanism, 81-82 rhythm, crayfish, 91 role in circadian rhythms, 61-65, 85 Organochlorides, circadian response to, 26, 27 Organophosphates, circadian response to, 26, 27, 29 Orientation and migration, hormonal influences, 333-337 Ornithacris cyanea. ommochrome distribution, 152 Orthoptera daily growth layers, 20 male sexual behaviour, 316 ommochromes as pattern pigments, 170, 171 acridiommatin, 136 distribution, 151, 152, 161 ommidins, 136 stridulatory mechanisms, 265 Oryzaephilus surinamensis, effects of tryptophan, 221 Oscillators, circadian and photoreceptor, separation, 77 control of behavioural rhythms, 54-71 -+

Oscillators, circadian-cont. control of gated events, 52-54 control of other rhythms, 71-72 driving, mechanisms, 5, 81-91 DrosophilA clock, 85-87 optic lobe clock, 81-82 protocerebral clock, 82-85 Ostrinia nubilar'is, oxygen consumption rhythm, 24 Ovaries kynurenine content, 192 role in female behaviour, 324, 326 Oviposition behaviour, hormonal control, 300, 327-331, 340 circadian rhythms of, 8, 12, 15, 29,91, 340 and response to insecticides, 26 entrainment, 45 phase-response curve, 48 temperature effects, 72 Oxidative metabolism, role in circadian oscillators, 88, 97 Oxygen consumption, circadian rhythmicity of, 23, 34, 91, 92. 96,97 P Palps, sound reception by, 27 1 Papilio, tryptophan ommochrome pathway, 169 P. machaon, 1.56 P. xuthus, 156, 160, 210 Papilionid butterflies, wing pigmentation, 126, 130 Paragonial gland and female refractoriness, 326-327 and oviposition behaviour, 330 Parathion, circadian response to, 26,27 +

390

Parnassius apollo, ommochromes, 156 Pars intercerebralis and female receptivity, 321-322, 323 and male reproductive behaviour, 318-319, 320 and oviposition behaviour, 328 Pattern pigments, ommochromes as, 169-173 Pectinophora gossypiella, circadian rhythms clock types, 77, 78 eclosion, 78 emergence, 77 hatching, 15, 78 insecticide susceptibility, 27, 28 oviposition, 1 2 , 4 8 , 78 phase-respose curve, 48 photoperiodism, 22 X-ray sensitivity, 29 Pentacosane, circadian variations in, 31 Perikaryon, endocrine cells, rhythm, 37 Peripheral action of hormones, 309-311 Penplaneta americana absence of glutarate pathway, 133 circadian rhythms brain hormone, role, 58-59 daily growth layers, 21 endocrine cells, 37 entrainment, photoreception, 44 feeding, 8 haemolymph metabolites, 30, 31 locomotor activity, 3, 6, 55-57 narcotic sensitivity, 25 optic lobes, role, 61-62 oxygen consumption, 23

SUBJECT INDEX

Periplaneta americana-cont. circadian rhythms-cont. sodium and potassium, 34 nerve cord, role, 60 X-ray sensitivity, 29 Perivisceral organs, neurosecretion, 299 Phallic nerve stimulating hormone, role in male sexual behaviour, 303,305,307,320,332 Pharmacological circadian rhythms, 40-42 Phase angle, circadian rhythms, 4-5 Phase adjustment during entrainment, 47-51 Phasmida daily growth layers, 20 ommochrome distribution, 153 Phenylalanine during colour change, 176 Pheromones and female receptivity, 323 and reproductive behaviour, 316 and sexual circadian rhythms, 10-11 effects on behaviour, 300-301 in allatectomized locusts, 317 release, rhythmicity of, 339 responsiveness to, rhythmicity of, 15 role of juvenile hormone, 321 Phlugis, sound emission, 263, 264 Phobocampe unicincta, ommochromes, 158, 162 Phormia regina, mating rhythm, 79 Phormia terraenovae, tryptophan + ommochrome pathway absence of glutarate pathway, 133 during metamorphosis, 206,207,

211 enzyme ontogeny, 2 1 3 3-hydroxy kynurenine, 128

SUBJECT INDEX

Phormia terraenovae-cont. kynureninase, 132 kynurenine, 125 kynurenine formamidase, 190 ommochrome localization, 158 tryptophan content, 122, 124 tryptophan balance, 219 tryptophan oxygenase, 126, 181-183, 185 Photochemical hourglass, 93, 96 Photoperiodism circadian rhythms of, 22 measurement, aphid, 93 Photoreception, role in circadian rhythms, 44-47, 52, 67, 78-80 Phototaxis, circadian rhythm of, 1 3 Phryne fcnestralis, ommochromes, 157 Physiological circadian rhythms, 22-43 biochemical, 29-34 cellular, 34-40 insecticide susceptibility, 25-29 metabolic, 23-24 narcotic sensitivity, 24-25 pharmacological, 40-42 tumours, 42-43 X-ray sensitivity, 29 Pieris brassicae locomotor activity rhythm, 338 ommochromes, 156 phototactic rhythm, 13 "Pigment IV" degradation reactions, 146-148 in ommochrome biosynthesis, 195 spectral data, 144, 148 Pigmentation hormonal control, 298, 313 3-hydroxy kynurenine, 129-130 influence of juvenile hormone, 335 kynurenine, 126

391

Pigmmtation-con t. onimochromes as pattern pigments, 169-173 orrimochromes as screening pigments, 166-169 rhjrthmicity, control, 71-72 Pilifer, hawkmoth, 289-290 Platyrieura capitata, sound radiating rymbal, 260 Plodic interpunctella ommochromes, 154 tryptophan metabolism in metamorphosis, 208 xanthurenic acid, 130 Pogonom yrm ex californicus, sexual circadian rhythm, 12 Pond r.kater, phototactic rhythms, 13 Porcellio scaber, drugs and loconiotor rhythm, 42 Portheria dispa, oviposition behaviour, 304 Potassium and sodium, diel changes in, 34 Praying mantis copulatory movements, control, 306 ommochromes in colour change, 174 Pressure receivers, ears as, 274-275, 276 Proline, during colour change, 176 Protein!; binding of ommochromes to, 164-165 synthesis, diel rhythm, 38 Prothoracic gland and premetamorphic behaviour, 313 cells, rhythmicity, 35 regulation, 299 role in migratory behaviour, 3 33-334 timing by brain hormone, 5 3

392

SUBJECT INDEX

Protocerebrum clock, mechanism, 82-85 neurosecretory cells, 298 Pteridines as cofactor, kynurenine-3hydroxylase, 191 as pattern pigments, 172-173 as screening pigments, 166, 167 association with tryptophan, 222 in colour change, 174 excretory role, 179 inhibition of tryptophan oxygenase, 181 Pteronemobius heydeni, ommochromc turnover, 197 Ptychopoda seriata, ommochromes as waste products, 17 7 biosynthesis, 195, 196 distribution, 154 ganglia pigmentation, 169 larva, tryptophan metabolism, 200 Pupation, circadian rhythmicity of, 16-17, 95 clock gating of, 53-54 Drosophila clock, 85-87 Pyrumeis, ommochromes P. atalanta, 155, 160, 177 P. cardui, 155, 177 Pyrethrum, circadian response to, 27 Pyrrhocoris apterus, juvenile hormone, 304

Q Quinoline derivatives, tryptophan + ommochrome pathway, 130-131 R Receptivity of female, 304-305 role of corpora allata, 320-324

Receptivity of female-cont. role of corpora cardiaca, 325 role of ovaries, 324 Redox properties, ommochromes, 140-145 Refractoriness of female, hormonal control, 325-327, 332 Releaser effects, hormones, 303 Reproductive behaviour, hormonal control, 316-333 female receptivity, 320-325 female refractoriness, 325-327 male behaviour, 316-320 oviposition, 327-331 Reserpine, and locomotor rhythms, 42 Respiration, function of ommochromes, 165-166 Response threshold, central modulation of, 91, 95 Responsiveness, rhythmicity of, 12-15 Rhodnius prolixus 3-hydroxy kynurenine, 130 ommochromes, 153 oviposition behaviour, 328, 330 Rhodommatin biosynthesis, 195 degradation reactions, 150, 155, 156, 157 deposition, 162 distribution, 136, 137, 160-161 in meconia, 176-177 in morphological colour change, 175-176 redox properties, 140-141 spectral data, 143, 146 Rhopaloceran butterflies, ommochromes, 161, 170 RNA die1 rhythm of, 15, 38 protein synthesis cycles, 91, 95 role in cellular oscillators, 88

SUBJECT INDEX

393

RNA-cont. role in Drosophilo clock, 85 and tryptophan oxygenase activity, 187-189 Romalea microptera circadian rhythms oxygen consumption, 23 role of suboesophageal ganglion, 56 ommochromes, 152, 161

S Salivary gland cells, rhythmicity of, 35, 3940, 86-87 Samia Cynthia, cocoon construction, 314 Sarcophaga bullata, ommochromes, 162 Sarcophaga falculata, daily growth layers, 22 Sasakia charonda, ommochromes, 155, 176 Suturnia p y n , protein tryptophan, 122 Scatophaga stercorariu, male sexual behaviour, 320 behaviour, hormonal control CNS spontaneous activity, 306 larval activity, 312 male sexual behaviour, 317-318 migratory behaviour, 333-335 oviposition, 327, 328 locomotor activity, 7 tryptophan+ ommochrome pathway 3-hydroxy kynurenine, 128 in egg, 199 in larva, 200 kynurenine, 125

Scatophaga stercoruriu-cont. tryptophan -+ ommochrome pathway-cont. kynurenine-3-hydroxylase, 191, 192 kynurenine transaminase, 193 ommochromes, 152, 170 quinoline derivatives, 130 tryptophan oxygenase, 182183, 184 Scoliopterix libatrix, ommochromes, 15 7 Scraper, role in stridulation, 2 55-25 7 Screening pigments, ommochromes IS, 166-169 Sexual circadian rhythms, 9-12 Silkmoth circadian rhythms brain as gating clock, 96 driving oscillator, mechanism, 89 eclosion, clock gating, 52-53 eclosion, entrainment, 45, 47 eclosion, hormonal control, 340 eclosion, type I1 clock, 77 larval ecdysis, 54 photochemical hourglass, 93 pupation, 53 rhythmic pheromone release, 11-12 role of protocerebrum, 51 hormonal control of behaviour adult behaviour, 314-315 “calling” behaviour, 300,303, 325,340 corpora cardiaca, role, 325 oviposition, 300, 328 pre-ecloiion, 307-308 Silkworm protocerebral clock, mechanism, 82-85

394

Silkworm-cont. tryptophan + ommochrome pathway anthranilic acids, 132 cinnabarinic acid, 16 1 egg, 121, 197-199 enzyme ontogeny, 214 larva, 201 metamorphosis, 202-204 ommochrome biosynthesis, 194 quinoline derivatives, 131 Singing efficiency , 2 6 7 -26 8 endogenous timing, 12 Sitophilus granan'us (Callandra p a n i a ) , locomotor activity rhythms, 7 Skotommin, distribution, 135, 136 Smerinthus ocellata, migratory behaviour, 337 Sodium and potassium, die1 changes, 34 Solubility, ommochromes, 138-139 Sound communication, 247-296 ears as receivers, 271-291 atypical ears, 288-291 forces acting on ears, 274-275 influence from surroundings, 275-279 parameters of sound, 271-274 receptor organ, behaviour, 285-288 tympana1 vibrations, 279-285 production of sound, 253-268 driving vibration and radiated sound, 262-264 efficiency of singing, 267-268 frequency multiplication mechanisms, 254-257 sound guide, 264-267 sound radiator, 258-262 propagation of sounds, 268-271

SUBJECT INDEX

Sound communication-cont. properties of sound, 248-253 sound fields, 251-253 vibrations, impedances and radiation, 248-251 Spectral data, ommochromes, 143-145 Sperm role in female refractoriness. 326-327 role in oviposition behaviour, 3 29-330 Spermatheca, role in reproductive behaviour, 332 Spermatophore production, endogenous timing of, 12 Sphinx, ommochromes S. ligustn', 156, 170 S. pinastn', 156 Spinning gland, tryptophan oxygenase, 184 Spiracle, prothoracic, role in sound conduction, 277 Spontaneous activity, CNS, hormonal changes, 305-306 Stick insect circadian rhythms colour change, 71-72 endocrine cells, 37 ommochromes as pattern pigments, 171 as waste products, 178 3-hydroxy kynurenine, 128 in morphological colour change, 173 oviposition behaviour, 328 Stress factor, neuro-active, 55 Stridulation biophysical aspects, 254-257 circadian rhythm of, 45, 61 Suboesophageal ganglion and 3-hydroxy kynurenine, 129 and tumour induction, 42-43

SUBJECT INDEX

Suboesophageal ganglion-cont. circadian rhythm of cells, 35, 37, 38 role in behavioural rhythms, 55-57, 63, 65, 67, 70 role in locomotor rhythms, 337-338 role in oviposition, 328 Sugar levels, blood circadian variations, 30, 31, 92, 95 regulation, 299 Sympetrum, ommochrome distribution, 151 S. flaueolum, 151 S. sanguineurn, 151 S. uulgatum, 151 Syrbula fuscouittata, male sexual behaviour, 320 Syrphus pyrastri, ommochromes, 157, 160 T Tabanus sp., ommochromes, 157 Taeniorhynchus (Mansonia) fuscopennata, feeding rhythm, 8 Tanning of cuticle, regulation, 299 Tegmen, distribution of sound level over, 260, 262 Tegminal resonators, vibrational properties, 258 Teleogryllus commodus stridulation rhythm, 45 tympanal vibrations, 283 Temperature and sound attenuation, 270-271 effect on circadian rhythms, 5, 72-74 Tenebrio molitor absence of ghtarate pathway, 133 insecticide susceptibility rhythm, 27

395

Tenebrio molitor-cont. ommochromes, 159 oxygen consumption rhythm, 24 Termite, caste [ormation, 303 Testis, ommochromes, 160, 161, 169 Tettigoniidae, sound communication conduction, non-tympanal route, 277 emission, biophysics of, 263-264 stridulation, imechanism, 254 Tetranychus urticae, circadian rhythms insecticide susceptibility, 26 narcotic sensitivity, 24-25 oviposition, 12 Tetrodotoxin, effect on locomotor rhythm, 81-82 Thio-organo-phosphate, circadian response to, 28 Tick, feeding rhythm, 9 Tipula oleraceti, ommochromes, 157-158 Tissue culture, circadian rhythms in, 39-40 Trehalose, circaclian variation in, 3 1 Tryptophan -+ oinmochrome pathway, 117-246 absence of +tarate pathway, 132-134 detrimental efkcts, 220-223 during development, 197-220 egg and emhryo, 197-199 larva: hemimetabola, 199-201 metamorphosis: holometabola, 201-21.2 ontogeny of enzyme activities,

2 12-218 tryptophan balance, 218-220 kynurenine pathway, enzymes, 179-193 kynureninase, 193

396

Tryptophan + ommochrome pathway-con t. kynurenine pathway-cont. kynurenine formamidase, 189 kynurenine-3-hydroxylase, 189-193 kynurenine transaminase, 193 tryptophan oxygenase, 180-189 ommochromes as pattern pigments, 169-173 as screening pigments, 166169 as waste products, 176-179 binding to proteins, 164-165 bios yn thesis, 193-19 7 deposition, 162-164 distribution and localization, 150-162 in morphological colour change, 173-176 isolation, 135-138 nomenclature, 134-135 properties, 138-150 tryptophan metabolites, fluorescent, 120-132 anthranilic acids, 131-132 formyl kynurenine, 125 3-hydroxy kynurenine, 126130 kynurenine, 125-126 methodology, 120-122 quinoline derivatives, 130-131 tryptophan, 33, 122-125 Trichoblatta sericia, oviposition behaviour, 328 Trichoplusia ni, circadian response to pheromones, 10-11 Tsetse fly, circadian rhythms activity rhythm, 7, 55, 93 eclosion, 19 larviposition, 12 visual response, 13, 14, 15

SUBJECT INDEX

Tubifera pendula, ommochromes, 157 Tumour induction rhythmic hormone secretion, 4243 tryptophan, 221 Tylotropidus speciosus, ommochromes, 152 Tymbal, cicada, mechanism;257 Tympana1 organ, sound-receiving properties forces acting on ears, 274-275 influence from surroundings, 275-279 parameters of sound, 271-273 tympal vibrations, 279-285 Tyrosine, during colour change, 176

U Ultrasonic frequencies detection, hawkmoth, 289-290 in insect sounds, 257, 263, 270 Ultrastructure, circadian rhythmicity in, 37 Urechis caupo, ommochromes, 165 Uric acid, circadian variations in, 30

V Vanessa sp., ommochromes, 160, 170 V. (Pyrameis) atalanta, 155 V. (Pyrameis) cardui, 155 V. urticae, 160, 176 Vegetation, sound attenuation by, 269-270 Velia currens, phototaxis rhythm, 13

397

SUBJECT INDEX

Ventral ncrvc cord, role in circadian rhythms, 60-63 Vibrations, in sound communication, 248-251 isolated locust ear, 279-285

W Wasp, parasitic diapause induction, -22 ommochromcs, 162 quinolinc derivatives, 131 Waste products, ommochromes as, 176-179 Water balance, regulation, 299 Water boatman, phototaxis rhythm, 13, 15 Weevil, circadian rhythms boll activity of AChE, 31-32 response to parathion, 26 grain in constant light, 79 locomotor activity, 7 Wing distribution of sound level over, 260,262 ommochromes as pattern pigments, 170 deposition, 162 distribution, 154-156, 160-161 rhodommatin, 136 sound production by, Drosophila, 264-265 Woodlouse, drugs and locomotor rhythm, 42

X Xanth om matin as patten pigment, 170-171 as screening pigment, 166-168 as waste product, 178 binding to proteins, 165 biosynthcsis, 194-197 chromatography, 140 degradation reactions, 150 deposition, 162, 164 distribution., 136, 137, 150-158, 160-162 early elucidation, 120 in egg, 198 in larva, 200 in metamorphosis, 203, 204, 206, 288 in morphological colour change, 173-176 in tryptophan balance, 219 redox properties, 142 reduction, 165-166 solubility and aggregation, 1 38- 139 spectral data, 143, 145 Xanthurenic acids, 130-131 X-ray sensitivity, rhythmicity of, 29, 71, 9 5 Y Yeast cells, Ei weakly oscillators. 89 Z Zeitgeber, definition, 4

coupled

This Page Intentionally Left Blank

Cumulative List of Authors Numbcrs in bold face indicate the volume number of the seryes

Aidley, D. J., 4, 1 Andersen, Sven Olav, 2, 1 Asahini, E., 6, 1 Ashburner, Michael, 7, 1 Baccetti, Baccio, 9, 315 Bcamcnt, J. W. I.., 2, 67 Berridge, Michael J., 9 , 1 Boistel, J., 5, 1 Brady, John, 10, 1 Bridges, R. G., 9, 51 Burkhardt, Dietrich, 2, 131 Bursell, E., 4, 33 Burtt, E. T., 3, 1 Carlson, A. D., 6, 51 Catton, W. T., 3, 1 Chen, P. S., 3, 5 3 Colhoun, E. €I., 1, 1 Cottrell, C. B., 2, 175 Dadd, R. H., 1 , 4 7 Dagan, D., 8, 96 Davey, K. G., 2, 219 Edwards, John S., 6 , 9 7 Eisenstein, E. M., 9, 11 1 Fraser Rowell, C. H., 8, 146 Gilbert, Lawrence I., 4, 69 Goodman, Lesley, 7, 97 Harmsen, Rudolf, 6, 139 Harvey, W. R., 3, 133 Haskell, J. A., 3, 133 Hinton, H. E., 5, 65 Hoyle, Graham, 7 , 3 4 9

Kilby, B. A., 1, 111 Lawrcncc, Peter A., 7, 197 I,ces, A. I)., 3, '207 Linzen, Bernt, 10, 117 Maddrell, S. H. P., 8, 200 Michelsen, Axel, 10, 247 Miles, P. W., 9, 183 Miller, P. L., 3, 279 Narahashi, Toshio, 1, 175; 8, 1 Neville, A. C., 4., 213 Nocke, Harold, 10, 247 Parnas, l., 8, 96 Pichon, Y . , 9, 257 Prince, William T., 9 , 1 Pringle, J. W. S., 5, 163 Riddiford, Lynn M., 10, 297 Rudall, K. M., I , 257 Sacktor, Bertram, 7, 268 Shaw, J., 1 , 3 1 5 Smith, D. S., I , 401 Stobbart, R. H.,, 1, 315 Treherne, J. E., 1,401; 9, 257 Truman, James W., 10, 297 Usherwood, P. N. R., 6, 205 Waldbauer, G . P., 5 , 229 Weis-Fogh, Torkel, 2, 1 Wigglesworth, V. B., 2, 247 Wilson, Donald M., 5, 289 Wyatt, G. R., 4, 287 Ziegler, Irmgard, 6, 139

399

This Page Intentionally Left Blank

Cumulative List of Chapter Titles Numbers in bold face indicate the uolume number of the serivs

Active Transport and Passive Movement of Water in I n s x t s , 2 , 67 Amino Acid and Protein Metabolism in Insect Development, 3, 5 3 Biochemistry of Sugars and Polysaccharides in Insects, 4, 287 Biochemistry of the Insect Fat Body, 1 , 1 11 Biology of Ptcridines in Insects, 6 , 139 Biophysical Aspects of Sound Communication in Insects, 1 0 , 247 Cellular Mechanisms Underlying Behaviour-Neuroetho logy, 7, 349 Chitin Orientation in Cuticle and its Control, 4, 213 Chitin/Protcin Complexes of Insect Cuticles, 1 257 Choline Metabolism in Insects, 9, 51 Colour Discrimination in Insects, 2, 131 Comparative Physiology of the Flight Motor, 5, 163 Consumption and Utilization of Food by Insects, 5, 229 Control of Polymorphism in Aphids, 3, 207 Control of Visceral Muscles in Insects, 2, 219 Effects o f Insecticides in Excitable Tissues, 8, 1 Electrochemistry of Insect Muscle, 6, 205 Excitation of Insect Skeletal Muscles, 4, 1 Excretion of Nitrogen in Insects, 4, 3 3 Feeding Behaviour and Nutrition in Grasshoppers and Locusts, 1, 47 Frost Resistance in Insects, 6, 1 Function and Structure o f Polytene Chromosomes Dui-ing Insect Development, 7, 1 Functional Aspects o f the Organization of the Insect Nervous System, 1 , 40 1 Functional Organizations of Giant Axons in the Cential Nervous Systems of Insects: New Aspects, 8 , 9 6 Hormonal Mechanisms Underlying Insect Behaviour, 1 0 , 2 9 7 Hormonal R e p l a t i o n of Growth and Reproduction in Insects,2, 247 Image Formation and Sensory Transmission in the Compound Eyc, 3, 1 Insect Blood-Brain Barrier, 9 , 257 Insect Ecdysis with Particular Emphasis on Cuticular Hardening and Darkening, 2, 175 40 1

402

CUMULATIVE LIST OF CHAPTER TITLES

Insect Sperm Cells, 9, 315 Learning and Memory in Isolated Insect Ganglia, 9, 111 Lipid Metabolism and Function in Insects, 4, 69 Mechanisms of Insect Excretory Systems, 8, 200

Metabolic Control Mechanisms in Insects, 3, 133 Nervous Control of Insect Flight and Related Behaviour, 5 , 289 Neural Control of Firefly Luminescence, 6, 5 1 Osmotic and Ionic Regulation in Insects, 1, 315 Physiology of Insect Circadian Rhythms, 10, 1 Physiological Significance of Acetylcholine in Insects and Observations upon other Pharmacologically Active Substances, 1, 1 Polarity and Patterns in the Postembryonic Development of Insects, 7, 197 Postembryonic Development and Regeneration of the Insect Nervous System, 6, 9 7 Properties of Insect Axons, 1, 175 Regulation of Breathing in Insects, 3, 279 Regulation of Intermediary Metabolism, with Special Refrrunce to the Control Mechanisms in Insect Muscle, 7, 268 Resilin. A Rubberlike Protein in Arthropod Cuticle, 2, 1 Role of Cyclic AMP and Calcium in Hormone Action, 9. 1 Saliva of Hemiptera, 9, 183 Spiracular Gills, 5 , 65 Structure and Function of the Insect Dorsal Ocellus, 7, 97 Synaptic Transmission and Related Phenomena in Insects, 5 , 1 Tryptophan + Ommochrome PatGway in Insects, 10, 117 Variable Coloration of the Acridoid Grasshoppers, 8, 146