UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Phylogenetic analysis of behavioural evolution : a case study using gasterosteid fishes McLennan, Deborah Ann 1989

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1989_A6_7 M04.pdf [ 9.66MB ]
Metadata
JSON: 831-1.0302341.json
JSON-LD: 831-1.0302341-ld.json
RDF/XML (Pretty): 831-1.0302341-rdf.xml
RDF/JSON: 831-1.0302341-rdf.json
Turtle: 831-1.0302341-turtle.txt
N-Triples: 831-1.0302341-rdf-ntriples.txt
Original Record: 831-1.0302341-source.json
Full Text
831-1.0302341-fulltext.txt
Citation
831-1.0302341.ris

Full Text

PHYLOGENETIC ANALYSIS OF BEHAVIOURAL EVOLUTION: A CASE STUDY USING GASTEROSTEID FISHES By Deborah Ann McLennan B.Sc, Simon Fraser University, 1983 A thesis submitted in partial fulfillment of \ the requirements for the degree of MASTER OF SCIENCE in T H E FACULTY OF GRADUATE STUDIES (DEPARTMENT OF ZOOLOGY) We accept this thesis as conforming to the required standard T H E UNIVERSITY OF BRITISH COLUMBIA September, 1989 ® Deborah Ann McLennan, 1989 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Z QOL.0G^) The University of British Columbia Vancouver, Canada Date ^ a ° T S 3 1^ 89 DE-6 (2/88) ABSTRACT ii The hypothesis that intersexual selection has been the most important force shaping the male nuptial signal in Gasterosteus aculeatus is based upon experiments in which females were offered a choice between "red" and "nonred" males. However, only territorial males assume nuptial colouration; therefore, results from these experiments provide support for mate recognition rather than differentiation among potential mates. In this thesis I examined the intersexual selection hypothesis further by first performing a phylogenetic systematic analysis of the Gasterosteidae using only behavioural characters (chapter 2). The resultant cladogram was congruent with trees utilizing morphological and electrophoretic data. Based upon the macroevolutionary relationships between breeding colours and breeding behaviours in the gasterosteids, three predictions relevant to a discussion of nuptial colouration in G. aculeatus were proposed: nuptial colouration is (1) weakly associated with male/male interactions (intra-sexual selection), (2) moderately associated with parental care (natural selection) and (3) strongly associated with male/female interactions (intersexual selection). These predictions were investigated by scoring seven colour variables (intensity, hue and distribution of red body colour; intensity and hue of blue eye colour; intensity and distribution of black body colour) for 19 males across an entire breeding cycle (chapter 3). The variables interacted to produce four distinct male colour mosaic signals corresponding with the stage a male had reached in the breeding cycle. No single variable was sufficient for the delineation of all breeding stages; however, the distribution of red body colour reliably distinguished the courting male from nesting and parental males. The phylogenetic predictions were confirmed for the component of the nuptial mosaic showing the greatest degree of intermale variability: the intensity of red body colour peaked maximally during courtship, and secondarily during fry guarding. The relationship between male colour and male behaviour was examined by recording a solitary male's reactions to a captive, receptive female over a 5 minute test period (chapter 4). The intensities of red body and blue eye colour in dull (and possibly bright) males reliably signalled their behavioural vigour. Medium intensity males signalled that they were more vigorous than their dull conspecifics; however, there was no association between colour and behavioural intensity within this group. Medium males may be more strongly influenced by stochastic factors such as previous experiences or colour relative to neighbours. The increased importance of personal history introduces a source of disorganization into the mating system that may oppose the directional force of truth in advertising and thereby increase the ambiguity of the male colour signal. The relationship between male colour and female behaviour was examined in a sequence of experiments culminating in a female's choice between two competing territorial males (chapter 5). Cladistic analysis of 16 behavioural characters recorded during the choice tests revealed three groups of males in this population: inactive losers (losing male did not participate in choice test), active losers (losing male active during choice test) and fighting losers (subgroup of active losers defined by female behaviour during choice test). The "inactive losers" group represents a "no choice" situation. In the remaining two groups, females responded preferentially to the most intensely coloured member of the competing male pair. This preferential response was strongest during the pre-choice, captive presentation in which the majority of females oriented head-up to, and tracked, the brighter red male. Once released, the female's initial response to the brighter courtship signal was overridden by the behavioural actions of the duller intensity male in 2 out of 12 trials: 10 of 12 females spawned with the brighter red male. The three structural components of red body colour thus provide information at all levels of mate choice: hue identifies sex, species and territorial status (mate recognition), distribution across the entire ventral/lateral surface signals a courting male (breeding status recognition) and intensity provides the basis for female discrimination among courting males (differentiation among potential mates). This series of micro-evolutionary studies supports the macroevolutionary prediction of a strong association between colour and courtship in this species and confirms the original intuition of previous researchers: the intensity of red is an important component of female mate choice in Gasterosteus aculeatus. IV T A B L E OF CONTENTS Abstract 11 - iii List of Tables vii - viii List of Figures ; i x - x Acknowledgements xi CHAPTER ONE General Introduction 1-4 General Methods The collecting site 4 - 7 Stock maintenance 7-9 Experimental design 9-10 Scoring colour 10 - 12 C H A P T E R TLUO: THE R E L A T I O N S H I P B E T W E E N B R E E D I N G C O L O U R S A N D B R E E D I N G B E H A U I O U R S THROUGH M f l C R O E U O L U T I O N f l R V T I M E Introduction '• 13-19 Materials and Methods The method: phylogenetic systematics 19-21 The study group 21-23 Character description 23 - 24 Rationale for character states 24-25 Results 25 - 29 Discussion Reconstructing phylogenetic relationships with behavioural characters 30 - 31 The phylogenetic approach to studying behavioural evolution 31-36 Summary —— 36 C H A P T E R T H R E E : T E M P O R A L C H A N G E S IN THE S T R U C T U R E OF THE M A L E M O S A I C S I G N A L Introduction — 37 Signal Production (i) what do sticklebacks eat? — ~~~ — 37-38 (ii) carotenoids in the food — — 38 - 39 (iii) carotenoids in sticklebacks 40 - 42 Materials and Methods 43 - 44 Results General observations 44 - 45 Data analysis Temporal fluctuations in colour: red body (i) total colour score (intensity) 45 - 48 (ii) colour distribution 48-51 (iii) hue 51-52 Temporal fluctuations in colour: blue eye (i) intensity 52 - 53 (ii) hue : 54 The interaction of red body and blue eye colour 54 - 55 Temporal fluctuations in colour: black body 55 - 56 Discussion 57-61 Summary — 61 C H A P T E R FOUR : THE R E L A T I O N S H I P B E T W E E N M A L E C O L O U R A N D M A L E BEHRUIOUR Introduction 62 - 63 Materials and methods 63 - 65 Results Red body colour 65 - 66 Blue eye colour 66 - 67 Associations between red body and blue eye colour 67 - 68 Discussion 68 - 70 Summary 70 C H A P T E R F I U E : THE R E L A T I O N S H I P BETUJEEN M A L E C O L O U R A N D F E M A L E BEHHUIOUR Introduction 71 - 72 Materials and Methods General experimental design — 72 - 73 Scoring behaviour 74 - 76 Character description 77 Results Cladistic analysis of choice tests 78 - 83 Behavioural differences between winners and losers (i) Initial presentation: solitary male/captive female 84 (ii) Second presentation: two interacting males/captive vi female 84 - 85 Colour differences between winners and losers (i) comparisons between winning and losing males within the same experimental stage 86 - 88 (ii) comparisons within groups of winning and losing males between experimental stages 88 - 89 (iii) captive female orientation based on male colour 89 - 90 Discussion Male/female interactions 90 - 93 The reinforcement of female choice 93 - 94 Male colour and male behaviour: what is so special about winners? ; 94 - 95 Summary 95 C H A P T E R S I K : F I N A L E The colour hierarchy in the male's nuptial signal 96 - 100 Mechanisms of colour elaboration: Fisherian run-away sexual selection or truth in advertising? 100 - 102 Concluding remarks 102 - 103 Literature Cited 104- 119 Appendices (A) Data matrix for phylogenetic analysis (chapter 2) 120 (B) Raw colour scores for individuals across the breeding cycle(chapter 3): (a) red body 121 (b) blue eyes 122 (c) black body 123 (C) Measurements of colour and behaviour during solitary male/captive female presentation (chapter 4) 124 - 125 (D) Measurements of colour and behaviour during sequence of experiments culminating in female choice (chapter 5): (a) intensity of red 126 (b) behavioural interactions focal male/female 127 (c) orientation times 128 (d) interacting males 129 (E) Data matrix for cladistic clustering of behavioural characters scored during the choice trials (chapter 5) 130 vii LIST OF TRBLES Table 1.1: Number and breeding condition of anadromous G. aculeatus individuals collected at 12 trap sites in Cow Pasture Creek, Salmon River in 1987. Table 3.1: % composition of major food types in the diet of various G. aculeatus populations. i Table 3.2: Carotenoid distribution in representative prey items of Gasterosteus aculeatus. Table 3.3: Percent distribution of carotenoids in three populations of Gasterosteus aculeatus. Table 3.4: Spearman rank correlations between red body intensity scores at different stages of the breeding cycle. Table 3.5: Results of Wilcoxon analyses of the differences in colour scores (red) between nest-building and the remaining stages of the breeding cycle for the six areas measured on each male G. aculeatus. Table 3.6: Spearman rank correlations between eye colour intensity at different stages of the breeding cycle. Table 3.7: Spearman rank correlation values for red body and blue eye colour scores at the same stage of the breeding cycle. Table 3.8: Relationship between six structural components of nuptial colouration across the breeding cycle. Table 3.9: Relationship between territory acquisition and development of nuptial colouration. Table 4.1: Spearman rank correlation coefficients between intensity of red body colour and aggressive/sexual behaviours of male three-spined stickle-backs. Table 4.2: Spearman rank correlation coefficients between intensity of blue eye colour and aggressive/sexual behaviours of male three-spined stickle-backs. Table 4.3: Spearman rank correlation coefficients within the same test stage between intensity of red body and blue eye colour of male three-spined sticklebacks. Table 5.1: Temporal sequence of experimental tests. Table 5.2: Wilcoxon analyses of differences between future winning and losing members of competing male pairs based upon behavioural interactions with a captive female prior to barrier removal. vii i Table 5.3: Wilcoxon analyses of the time a captive female spent oriented towards the future winning male of a competing male pair. Table 5.4: Wilcoxon analyses of differences in intensity of red between winners and losers at various stages in the experimental sequence. Table 5.5: Wilcoxon analyses of the time a captive female spent oriented towards the brighter male of a competing male pair. Table 6.1: Factors potentially involved in the pathway from pigments in prey items to nuptial colouration in male Gasterosteus aculeatus. ix LIST OF FIGURES Figure 1.1: Area map showing location of the Salmon River and Cow Pasture Creek; (insert) location of trap sites in Cow Pasture Creek. Figure 1.2: Six anatomical landmarks scored for the intensity of "red" nuptial colouration on all male Gasterosteus aculeatus test individuals. Figure 2.1: Hypothetical example outlining the applicability of phylogenetic systematics to questions concerning the evolution of behaviour. Figure 2.2: Phylogenetic tree for the Gasterosteidae based upon behavioural characters. Figure 2.3: Phylogenetic trees for the Gasterosteidae based upon non-behavioural characters. Figure 2.4: Relationship between diversification of colour and courtship in the gasterosteids. Figure 2.5: Relationship between diversification of colour and aggressive behaviours in the gasterosteids. Figure 2.6: Relationship between diversification of colour and parental behaviours in the gasterosteids. Figure 3.1: Change in total colour score (intensity of red) of Gasterosteus aculeatus males (n=19) across a complete breeding cycle. Figure 3.2: Histogram of the intensity scores (red) of G. aculeatus males at different stages in the breeding cycle. Figure 3.3: Changes in colour intensity (red) at each of the six areas on G. aculeatus males scored across a complete breeding cycle. Figure 3.4: Relationship between colour intensity and perceived hue. Figure 3.5: Change in eye colour intensity of Gasterosteus aculeatus males (n=19) across a complete breeding cycle. Figure 3.6: Histogram depicting the distribution of colour score intensities for the blue eye (striped bars) and red body (solid bars) components of the nuptial signal in courting male G. aculeatus. Figure 3.7: Change in the intensity of dorsal/lateral melanism in G. aculeatus males across the breeding cycle. Figure 4.1: The relationships between a message and its signals. Figure 5.1: Tree produced by cladistic analysis of behavioural characters recorded during female choice trials. X Figure 5.2: Distribution of characters associated with the activity of the losing male. Figure 5.3: Distribution of characters associated with female activity. Figure 5.4: Hierarchical structuring of groups of males within a test population of G. aculeatus. Figure 5.5: Differences in red body colour scores between winners and losers (winners - losers) plotted for each pair of males. Figure 5.6: Changes in male colour through time in response to different behavioural interactions. xi The end. product of every creative enterprise represents the work* ideas and support of more than just one person. This thesis is no exception. I would like, to than^the following people, for helping me in more tuays than it is possible, to recount: Sinn and (Rudy (Kpenig, (Donald and Josephine (McLennan, Delia, Donald, Janice, (David and Ian (McLennan, Vera (tfamlyn, Donald and 'Ethel (Keller, Bernice Good, %gbert and (Mae (Hamlyn, (Penni Grey-Mien, Wayne Qoodey, 'Bob Gregory (the bold), <Peter Watts (the capricious), INjancy (Butler (the righteous), Linda Qlennie (the venturous), Ziphiid (the befuddled), Gary (Birch, Chris foote, Gayle (Brovm, ddc^Taylor, Laura (Hooker, (PatrickLavin, Jim Caldwell, (Pete Cahoon, Lee Qass, John Borden, J.9{3d> Smith and9{.%. Liley. I am eternally grateful to my supervisor J. (D. (Mc(Phailfor teaching me that every biological question encompasses its own realm of excitement and wonder, to Daniel Brooks for opening up whole new areas of exploration in evolutionary theory and to Gordon (Haas for helping me stay (relatively) sane throughout this entire process. This thesis is dedicated to the memory of my grandfather, Donald (McLennan. CHRPTER ONE G E N E R A L INTRODUCTION 1 The three-spined stickleback is a small, seemingly unimportant bony-plated teleost. It has played no direct role in the development of North American culture or society, is of little or no commercial importance and has definitely not contributed to the sport fishing or tourist industry in British Columbia. However, the ubiquitous three-spined has been leading a double life in the New World for its ancestors enjoyed prominent roles in cultures older than our own. The historical precedent for the commercial importance of Gasterosteus aculeatus was set by the English peasantry who used to "scoop them [sticklebacks] out by the millions and apply them as manure to the land" (at the lofty price of one halfpenny per bushel: Maxwell, 1904). Sticklebacks were prized for their oil in France and Sweden (Maxwell, 1904), as fish meal and duck food in Holland (Baggerman,1958), and "amazing shoals" of migrating individuals formed the basis of a thriving fishing industry in Japan (at yet another lofty price of 10 sen per 100-200 fish: Ikeda, 1933). These little fish also provided sustenance to people during the final days of the second world war and contributed to the development of the German language. Children referred to their younger, smaller counterparts as "Stichlinge" (R. Koenig.pers. comm.). But perhaps the threespine's most glorious role comes from Hebrew mythology, where this unassuming fish acquires heroic proportions: On the fifth day of creation, God took fire and water, and out of these two elements He made the fishes of the sea . . . The ruler over the sea animals is Leviathan. With all the fishes he was made on the fifth day. Originally he was created male and female like all the other animals. But when it appeared that a pair of these monsters might annihilate the whole earth with their united strength, God killed the female. So enormous is Leviathan that to quench his thirst he needs all the waters that flow from the Jordan to the sea. His food consists of all the fish which go between his jaws of their own accord. When he is hungry, a hot breath blows from his nostrils, and it makes the waters of the great sea seething hot. Formidable though Behemoth, the other monster, is he feels insecure until he is certain that Leviathin has satisfied his thirst. The only thing that can keep him in check is the stickleback, a little fish which was created for this purpose, and of which he stands in great awe. - The Haggadah So the stickleback, despite its diminutive size, is in fact a culturally, economically and theologically important species deserving of the attention bestowed upon it by generations of researchers. This scientific attention has occurred, not because of the threespine's glorious past, but because of an aspect of its unpretentious present: it has an alter ego. During spring, stickleback males undergo a transformation from inconspicuous, silver -green fishes into flamboyant mosaics of intense flame scarlet and flashing aquamarine blue. Homo sapiens is a very visually oriented species. This affects both the kinds of questions we ask and the way we search for answers. Because of our perceptual bias, we have tended to investigate other organisms through the filter of our own sensory system. Fortunately current wisdom has changed this approach. Nevertheless, while Homo sapiens as scientist attempts to minimize these "observer" effects, the scientist as Homo sapiens remains strongly drawn to, what is for him, the most obvious signal in the stickleback system: colour. This fascination culminated in a series of elegant experiments by Pelkwijk and Tinbergen (1937) designed to discover the "function" of sexually dimorphic breeding colouration in Gasterosteus aculeatus. Subsequent researchers have added details of the three-spined sticklebacks' complex mating system to these initial studies of the mating signal. In brief, the male three-spined stickleback participates in resource defense polygamy (Emlen and Oring, 1977): males defend territories around a nest site and both sexes may spawn with more than one partner during the breeding season. The extent of polygamy in this species has not been rigorously documented. Although females can produce a clutch every 3 to 4 days under laboratory conditions (Wootton, 1974), interspawning intervals were substantially longer in the field : i. e., 19 days for the only population measured to date (Reiffers,1984). This difference between physiological capability and field performance is mirrored in the breeding behaviour of males. Unlike individuals in the lab, who obtain 1 to 5 clutches in one cycle and renest repeatedly within the breeding season (Baggerman, 1958; van Iersel, 1953), males in the field do not appear to renest as readily (Kynard.1978; Whoriskey and Reebs.1986; my unpublished data, 1986). Like many teleost species, male three-spined sticklebacks provide all the parental care. The female simply deposits her eggs in the the nest then flees. There may be a physiological basis for this female "spawn and dash" strategy. High caloric substances are transported from somatic tissue into the ovaries as the eggs are developing and, once complete, one clutch can occupy up to 30% of the female's total body mass (Wootton,1974). Females therefore expend energy on predominantly physiological tasks, with some investment in the mate seeking and courtship process. Male expenditure, on the other hand, consists of the adoption of nuptial colouration and the energetic stresses of territory maintenance, nest building, courtship and parental care; it is primarily behavioural. Since parental investment in this species is not sex biased, sexual selection should operate to some degree upon both sexes (Darwin, 1871; Fisher, 1930;Trivers, 1972). Current research has begun to provide corroborating evidence for the role of sexual selection in shaping both male and female breeding characteristics (reviewed by Li and Owings,1978a;Rowland, 1982b). Given the complexities of this breeding system, it is not surprising that several different, but not neccessarily mutually exclusive, hypotheses have been proposed to explain the function of male nuptial colouration in this species. Researchers have concluded that "red", as a signal, is involved to various degrees in (1) territory acquisition and maintenance (intra-sexual selection), (2) mate acquisition (intersexual selection) and (3) parental care (natural selection) (Pelkwijk and Tinbergen, 1937;Tinbergen, 1948;Semler, 1971; Rowland, 1982a, 1984; Bakker and Sevenster, 1983). Support for the intersexual selection hypothesis is based upon experiments in which receptive females have been offered a choice of "red" and "nonred" males (Pelkwijk and Tinbergen, 1937; McPhail, 1969; Semler, 1971). Since all territorial G. aculeatus {sensu stricto) males develop some degree of red nuptial colouration, offering females a choice of males differing in the intensity of red provides a stronger test of the hypothesis. Additionally, the nuptial signal is a mosaic of at least two colours, red and blue, and is thus more complex than a simple contrast of "red" versus "nonred" would indicate. Gasterosteus aculeatus is one of the best studied species in ethology and thus one of the few organisms where evolutionary scenarios are based upon the fruitful interaction of experimental data and theoretical models. Nevertheless, there are still gaps in our knowledge, one of which centers . around the perception and treatment of "red" as an all or nothing signal. My study begins with an investigation of colour at the macroevolutionary (among 4 species) level. Attention will then be focussed upon colour changes and their associations at the microevolutionary (within species) level in the three-spined stickleback. The thesis will be presented in a series of four chapters, each dealing with one specific question: (1) What is the relationship between breeding colour and breeding behaviours through macroevolutionary time? This problem is addressed by performing a phylogenetic analysis on the five genera within the family Gasterosteidae. (2) Are these macroevolutionary relationships mirrored in colour change across the G. aculeatus breeding cycle? Colour at four stages of the breeding cycle - nest building, courtship, egg guarding and fry guarding - are scored to elucidate the temporal and morphological structure of the male colour signal. (3) What is the relationship between male colour and male behaviour in this population? The assumption that bright males are more vigorous behaviourally than their dull counterparts (Rowland, 1984) is investigated by recording the intensities of both male colour and behavioural response to a captive, gravid female. (4) What is the relationship between male colour and female behaviour in this population? In this final experimental chapter the receiver's responses to variation in the signal are investigated by giving females a choice between males differing only in the intensity of the red signal. GENERAL METHODS The Collecting Site Anadromous three-spined sticklebacks [Gasterosteus aculeatus, forma trachurus : Cuvier cited in Maxwell, 1904) were collected in unbaited minnow traps during May, 1986 and March to June, 1987 from the Cow Pasture Creek branch of the Salmon River, Langley, British Columbia (Fig. 1.1). Cow Pasture Creek is a shallow and sluggish body of water surrounded by pastures on the north western shore closest to the river and a tangle of alders, blackberry thickets and various other deciduous shrubs and saplings along the remaining banks. The terrestrial vegetation is sufficiently thick, particularly during summer months, to interfere with the transmission of light into the creek. This initial decrease in the intensity of light reaching the surface of the water is accentuated by the fact that the creek is extremely turbid. Unlike the clear or tea coloured streams elsewhere in the valley, this water is the light, muddy beige (Munsell colour 27: drab) characteristic of a high concentration of suspended soil particles. Secchi disk readings taken in the river in 1988 ranged from 50 to 75 centimeters. Salmon River stickleback populations have been studied for nearly 10 years (McPhail.pers. comm.). Sampling over these years has demonstrated that anadromous individuals rarely disperse more than 100 meters up the creek, therefore 10 trapping sites were established below this maximum range; 3 along the western shore of the river immediately north of the creek, and 7 in the creek itself (Fig. 1.1 insert). The majority of the experimental data were collected during the 1987 breeding season. Unfortunately this coincided with a "bad" year for returning sticklebacks so two additional sites (#11 and 12) were established further upstream on April 28th. Water temperature, trap site and the number, reproductive condition and type (anadromous versus freshwater) of trapped individuals were recorded. Few individuals were ever collected in the river itself and sites within the creek varied consistently in the numbers of fish trapped (Table 1.1). Gravid females appear long before coloured males in this population. Male sticklebacks nest in the soft mud amongst patches of Elodea and emergent grasses near the edges of the stream. The turbidity, combined with the luxuriant growth of aquatic vegetation in Cow Pasture Creek, made field observation of breeding behaviour impossible. Equally problematical is the fact that the red pigmentation occurs along the lateral and ventral surfaces of the fish, rendering it impossible to accurately score either the extent or intensity of the colour signal in the field. All the experimental data were therefore collected in the laboratory. Fig. 1.1: A r e a map showing location of the Sa lmon River a n d Cow Pasture Creek; (insert) location of trap sites in Cow Pasture Creek. 7 SITE DATE TEMP. 1 2 (°C.) 3 4 5 6 7 8 9 10 11 12 TOTAL GRAVID FEMALES COLOUREI MALES Mar 5 9.5 - - 0 20 9.0 - - 0 - -26 7.5 - - - 1 - - - - - NA NA 1 NO NO Apr 2 10.5 - - NA NA 0 - -9 9.5 - - 1 NA NA 1 NO NO 16 10.0 - - 3 - 1 NA NA 4 NO NO 23 11.0 1 2 6 1 - - 4 3 5 NA NA 22 NO NO 28 12.0 - - 4 • 23 8 18 4 20 1 73 147 YES NO® May 7 14.0 - - - - 2 1 1 - 2 1 35 42 YES NO® 14 14.5 - - - 2 6 - 5 - 2 - 17 32 YES NO® 21 16.0 - - 5 9 14 YES NO® 27 17.0 - - - 10 13 5 10 - 4 1 6 49 YES YES Jun2 15.5 - - - 3 5 - 8 - - - - 16 YES YES 23 17.0 - - 1 - 2 - 1 - - - - 4 NO NO T a b l e 1.1: N u m b e r a n d b r e e d i n g c o n d i t i o n of a n a d r o m o u s G. aculeatus i n d i v i d u a l s c o l l e c t e d at 12 t r a p s s i t e s i n C o w P a s t u r e Creek, S a l m o n R i v e r i n 1987. * = m i s s i n g d a t a f o r t h e s e sites; @ = s o m e m a l e s w i t h b l u e eyes, r e d o n t h e i n s i d e f l o o r of t h e i r m o u t h a n d a f a i n t h i n t o f r e d i n t h e lo w e r j a w are a . N A = n o t a p p l i c a b l e ; t r a p s were n o t set at t h e s e s i t e s d u r i n g t h i s p e r i o d . Stock Maintenance Fish were housed in plexiglass aquaria measuring 62 x 58 x 26 cm. Samples collected at different times were combined into five separate groups: Mar 5 -28, May 7, May 14, May 23 and June 2-23. Densities were restricted to a maximum of 50 individuals per holding tank; however, there was no attempt to control actual density or sex ratio below this upper limit. The bottom of each tank was covered with coarse gravel, with refuge and environmental spatial structuring provided by clay flower pots, large rocks, and floating water sprite. Dechlorinated water flowed continuously through each tank, temperature fluctuated between 10 - 12 ° C . and the stock room was illuminated by fluorescent lighting on a 10 hour light-14 hour dark photoperiod. These conditions do not suppress either the completion or maintenance of gonadal maturation but do favour a decrease in the rate of maturation (Baggerman,1958). The appearance of breeding behaviour is discouraged by the lack of suitable nesting substrate/nesting materials and the unnaturally high densities in each tank (Baggerman, 1958; Reisman, 1968b). During the course of the experiments none of the males in the stock tanks developed red nuptial colouration or displayed any tendency to build nests or court females. Once introduced into the test aquaria however, only 5/83 (6%) males failed to complete nest building within 48 hours. Since only recently ovulated females respond to male courtship (Lam et al, 1978) a constant supply of gravid individuals was required throughout the summer. Female stock tanks were established in an environmental chamber where conditions were more condusive to a rapid maturation rate. Water temperature in the aquaria fluctuated between 19.0 - 19.5 °C. and the photoperiod was automatically set to the stimulating 16 hour light/8 hour dark regime (Baggerman, 1958). Ovulated females were identified by the presence of a jelly-like ovarian exudate extruding from the gonopore (Lam et ai.,1978), the relaxed and expanded nature of the gonopore and the ability to actually see the ovulated eggs during the application of very gentle pressure to the sides of the fish. Although studies of nuptial colouration have traditionally centered around the male, females in this population also undergo a colour change during breeding; passing from silvery green, through gold dorsally/silver ventrally to an intense metallic gold over the entire body of the courting female. The existence of gold nuptial colouration has been reported for one other gasterosteid: Gasterosteus wheatlandi males. The degree of this colour development in three-spined stickleback females appears to be associated with behavioural sexual receptivity; however, analysis of the relationship between female colour and behaviour awaits further study. The "red" nuptial colouration characteristic of G. aculeatus males is based upon the deposition of carotenoids in specialized integument cells. Since animals cannot synthesize carotenoids de novo, they must sequester these pigments from their food and will lose colour if deprived of an appropriate pigment source (Fox, 1976). Circumstantial evidence from preliminary tests in 1986 indicated that three-spined stickleback males may lose pigment fast enough to confound studies of the relationship between colour and behaviour. Males collected on May 4, 1986 were allowed to complete one breeding cycle in two sequential groups; the first on May 19-June 12, the second on June 16-July 10. Peak breeding colour in group 1 ranged from a total score of 5 to 55 (details of colour scoring procedure will follow) with a mean value of 27.6. Colour scores decreased in the second group to a mean of 20.8 and a range of 13 to 30. This presented the problem of preventing colour loss without artificially inflating colour scores. Two steps were taken in an attempt to solve this puzzle: (1) fish were collected on a weekly basis, segregated according to collection date and used as soon as possible and (2) the regular, nutritionally adequate but carotenoid-poor diet of Tubifex and chopped earthworms was supplemented once a week with a carotenoid-rich food source: live adult brine shrimp, Artemia salina (see chapter 3). Experimental Design Experiments were conducted in nine 200 liter aquaria measuring 122 x 51 x 33 cm. Five of these tanks were housed within a separate room in the laboratory. A behavioural chamber was created around the remaining four tanks by the installation of floor to ceiling curtains. "Daylite" fluorescent tubes suspended 30 cm. above the water surface provided the only light source during filming sequences and colour scoring. The wavelength composition of this light approximated natural daylight; all wavelengths were represented, with a skew towards the blue end of the spectrum. Baggerman (1958) investigated the rate of gonadal maturation under the light regimes of several artificial and natural sources and found no differences based upon spectral quality. Aquaria were covered on three sides with neutral grey construction paper and lined with a 3 cm. base of coarse, grey gravel. Light intensities, measured through the light meter of a Canon A l camera at nine positions in each tank, were virtually identical in each test chamber. Aquaria were divided in half by a 0.6 cm. thick, moveable, black plexiglass sheet. This divider was removed during the female choice tests. Although each male was spatially and visually isolated from his opponent during the initial set of experiments in the choice sequence (chapters 4 and 5), it is possible that chemical and/or auditory communication was occurring. Studies by several authors (Bruning, 1906; Westerfield,1922; Segaar,1961; Freihofer,1963; Bannister, 1965) have suggested that olfaction and sound do not play major roles in the breeding system of this species. Conditions for the final experiment, the examination of endogenous colour changes over a complete breeding cycle (chapter 3), required more stringent inter-male isolation so the tanks were drained and the divider siliconed into glass slots. A 20 cm. green plastic plant saucer, filled with soft mud, fine gravel and filamentous algae was placed against the far side of each test compartment. Since G. aculeatus males use visual cues to locate their nests and define territorial boundaries (van Iersel,1958; van den Assem, 1967; Tschanz and Sharf,1971), the filter and a series of small rocks were placed along the divider in order to encourage the males to learn specific borders. Water temperatures, maintained by laboratory heat, fluctuated between 18.0 and 19.5 °C. The photoperiod was automatically set on a 16 hour day/8 hour night regime (lights on at 5:00 and off at 21:00 hours). Curtains were suspended along the front of the tanks and small 5 x 5 cm. flaps were cut to align with the center of each test compartment. Colour was scored by looking through these flaps. Filming required a larger viewing area so, after scoring both of the members of one male pair, the curtains were drawn back. Males were fed ad libitum on Tubifex at 6:00 every morning. Colour scoring and the behavioural tests detailed in chapters 4 and 5 were started at 8:00 hours; colour scoring for the complete breeding cycle (chapter 3) was started at 13:00 hours. Behavioural interactions were filmed with a Hatachi video camera. The colour resolution on the resultant videotapes was too poor to record colour accurately. Scoring Colour The colour signal in G. aculeatus exists in the temporal and three spatial (structural) dimensions: hue, intensity and distribution. Colour data were recorded for three components of the nuptial signal: ventral/lateral red, dorsal/lateral black and blue eyes. Seven colour measurements were recorded daily for each individual male: (1) red: hue, intensity and distribution; (2) blue: hue and intensity (distribution is not a relevant parameter of eye colour); (3) black : hue and distribution (hue = intensity for achromatic "colours"). Previous studies have either recorded the "presence/absence" of red (Semler,1971) or have ranked the intensity of red on a scale of 1 (low) to 5 (high) (Rowland, 1984;Ward and Fitzgerald, 1987). I refined this measure-ment to an intensity score of 0 - 10 for each of six anatomical landmarks on the fish (Fig. 1.2). Summing these values produced an overall individual colour score ranging from 0 (no colour) to 60, and provided information concerning both the overall intensity and distribution of red during the breeding cycle. I found it more difficult to differentiate eye colour intensities and scored these on a scale of 0-5. Hue was determined by comparing males to colour swatches in the Naturalist's Color Guide (Smithe, 1975). Preliminary studies in 1986 indicated that the colour range of red of this population was very narrow, therefore all but four of the red/orange swatches presented in the Guide (colours #14 - 17) were covered with neutral grey paper. The guide sheet was then placed outside the aquarium by the central divider. This procedure was duplicated for measurements of eye colour (blue) and dorsal/lateral melanism (black). Seven shades (intensities) of black were scored: pale (colour 86), light (colour 85), medium (colour 84) and dark (colour 83) neutral gray and the intermediate values of 85.5, 84.5 and 83.5. In the case of variegated individuals, only the dominant colour was scored. This analysis is thus coarse and is used here only to establish trends in melanism fluctuations. To counteract the perceptual biases of the observer, recording sequences were randomized daily, none of the data were examined until after the termination of the experiments, a concerted effort was made not to form any recognition images of test individuals, and one randomly chosen individual was rescored at the end of every test period. Concordance between the initial and repeated scores was tested using a Spearman rank correlation analysis corrected for ties (Siegel,1956). Initial and repeated measurements of both red body and blue eye intensity scores were significantly correlated (red: n= 23; p = 0.99; p < 0.001; blue: n= 23; p = 0.89; p < 0.001); therefore, this scoring procedure is precise within a given breeding season. 12 VPS Fig. 1.2: Six anatomical landmarks scored for the intensity of red nuptial colouration on all male Gasterosteus aculeatus test individuals. Capital letters refer to the following areas on the fish: LJ = lower jaw; OP = operculum; VPS = ventral area between the lower jaw and the pelvic spines; PP = pectoral plate; PSC = ventral area from the pelvic spines to the caudal fin; LP = lateral plates. Each area was assigned ah intensity score from the range of 0 - 10 and the six scores added to produce an overall intensity score for each male. 13 CHAPTER TUJO THE R E L A T I O N S H I P B E T W E E N B R E E D I N G COLOURS AND B R E E D I N G B E H A U I O U R S THROUGH M A C R O E U O L U T I O N H R V T I M E INTRODUCTION Ethology, as a science, was founded upon a tradition of investigating behaviour within an explicitly phylogenetic framework. Darwin (1872: 346) laid the foundations for this tradition when he wrote " . . . the characters which naturalists consider as showing true affinity between any two or more species, are those which have been inherited from a common parent, all true classification being genealogical." This idea was expanded in an entire chapter dedicated to instinct, in which Darwin compared, among other things, the behaviour of two species within the genus Formica in an attempt to trace the evolution of slave-making in ants. Having presented a variety of behavioural examples, he concluded that his theory was strengthened by examples of "closely allied, but distinct, species, when inhabiting distant parts of the world and living under considerably different conditions of life, yet often retaining nearly the same instincts." (1872: 219). Following this example, the "founding fathers" of ethology, Oskar Heinroth and Charles O. Whitman, proposed that there were discrete behavioural patterns which, like morphological features, could be used as indicators of common ancestry. Whitman's (1899) views mirrored Darwin's, "Instincts and organs are to be studied from the common viewpoint of phyletic descent." This perspective served as the focal point for a plethora of studies in the early 20th century. Behavioural data were examined with an eye to their phylogenetic significance for the Anatidae (Heinroth,191 l;Herrick,191 la,b,c), the social Vespidae (Ducke, 1913), weaver birds (Chapin, 1917), social insects in general (Wheeler, 1919), spiders (Petrunkevitch, 1926), cowbirds (Friedmann, 1929), bumblebees (Plath, 1934), birds of paradise (Stonor, 1936), termites (Emerson, 1938) and caddisfly larvae (Milne and Milne, 1939). Wheeler (1928: 20) reiterated Darwin's and Whitman's perspective and reaffirmed the basis of ethological studies at the time: Of late there has been considerable discussion... as to the precise relation of biology to history...and what most of us older investigators have long known seems now to be acceded, namely that biology in the 14 broad sense and including anthropology and psychology is peculiar in being both a natural science and a department of history (phylogeny). Comparative behavioural studies flourished under the direction of Konrad Lorenz and Niko Tinbergen during the 1940's and 1950's. Both of these ethologists repeatedly emphasized two distinct but related points: (1) behavioural patterns are as useful as morphology in assessing phylogenetic relationships and (2) behaviour does not evolve independently of phylogeny. Lorenz (1941) stated ". . . all forms of life are, in a way, phylogenetic attainments whose special objects would have to remain completely obscure without the knowledge of their phylogenetic development." and (1958), "Every time a biologist seeks to know why an organism looks and acts as it does, he must resort to the comparative method." Tinbergen (1964) outlined the comparative method thusly: The naturalist . . . must resort to other methods. His main source of inspiration is comparison. Through comparison he notices both similarities between species and differences between them. Either of these can be due to one of two sources. Similarity can be due to affinity, to common descent; or it can be due to convergent evolution. It is the convergences which call his attention to functional problems. . . . The differences between species can be due to lack of affinity, or they can be found in closely related species. The student of survival value concentrates on the latter differences, because they must be due to recent adaptive radiation. In other words, phylogeny provides the pattern from which explanations of processes responsible for behavioural evolution must be derived. Although the comparative approach to studying behavioural evolution flourished during the late 1950's and 1960's, skepticism mounted about Lorenz's assertation that species-specific behavioural characters were valuable systematic characters. By the centenary of the publication of Darwin's book, two widely divergent viewpoints were held: To assume evolutionary relationships on the basis of behavior patterns is not justifiable when such findings clearly contradict morphological considerations. The methods of morphology will therefore remain the basis for the natural system [of classification] (Starck, 1959 cited in Eibl-Eibesfeldt, 1975, p. 223). If there is a conflict between the evidence provided by morphological characters and that of behavior, the taxonomist is increasingly inclined to give greater weight to the ethological evidence (Mayr, 1958, p. 345). This difference in opinion was founded, in part, upon continuing unresolved debates among ethologists. Two questions re-occurred; first, how well can sequences of ancestral and derived traits be determined for attributes that left no fossil record and, second, how well can similarities due to common ancestry (homology) be distinguished from similarities due to convergent evolution (analogy) (Boyden,1947;Lorenz,1950;Tinbergen, 1951 ;Schneirla, 1952;Michener, 1953)? The question of homology was problematical because homologous characters were defined by their common origin, and at the same time, were used to reconstruct phylogenetic relationships. The risk of circularity was high. Remane (1956) proposed a set of criteria for testing hypotheses of common origin (homology) without a priori reference to phylogeny. These were (1) similarity of position in an organ system, (2) special quality (e. g., commonalities in fine structure or development), and (3) continuity through intermediate forms. Although authors did not agree about the universal applicability of Remane's criteria to behaviour, the majority accepted that the criterion of special quality, studied at the level of muscle contractions (fixed action patterns), was the fundamental tool for establishing behavioural homologies (Baerends, 1958; Remane, 1961; Wickler, 1961; Albrecht, 1966). Initial attempts to homologize behaviour in this way were admittedly vague and simplistic when compared to the more quantitative methodology of comparative morphology, but this reflected more the youth of the discipline than a fundamental flaw in the behavioural traits themselves. Time and again, phylogenies reconstructed using behavioural characters mirrored those based solely on morphology. However, in a scathing review of the ethologists' research program, Atz (1970) made only a cursory reference to these successes when he concluded: The number of instances in which behavior has provided valuable clues to systematic relationships has continued to grow but it should be made clear that the establishment of detailed homologies was seldom, if ever, necessary to accomplish this ... Functional, and especially behavioral, characters usually do not involve demonstrable homologies, but depend instead on resemblances that may be detailed and specific but nevertheless cannot be traced, except in a general way, to a common ancestor. ... Until the time that behavior, like more and more physiological functions, can be critically associated with structure, the application of the idea of homology to behavior is operationally unsound and fraught with danger, since the history of the study of animal behavior shows that to think of behavior as structure has led to the most pernicious kind of oversimplification. This review marked the end of attempts to homologize behaviour, and the beginning of the "eclipse of history" in ethology. Organisms are the result of a dynamic interaction amongst genetical, developmental, physiological, morphological and behavioural systems through time. There is no a priori reason to believe that one system contains more information about evolutionary relationships than any other. Since information contained within a system (history) constrains the way that system is able to respond to current selection pressures (Brooks and Wiley, 1988), the elimination of history from evolutionary explanations produces an incomplete picture of underlying processes. Nevertheless, the past 15 years have witnessed a decline in the use of behavioural characters in phylogenetic analyses. Even more disconcerting is the shift away from asking evolutionary questions within a phylogenetic framework. Lorenz (1941) cautioned, "The similarity of a series of forms even if the series structure arises ever so clearly from a separation according to characters, must not be considered as establishing a series of developmental stages." In his opinion, without reference to phylogenetic relationships, the criterion of similarity was, of itself, a dangerously misleading evolutionary marker. Unfortunately, the Gordian knot of behavioural homology drove ethologists towards a new methodology based, in direct contrast to Lorenz' warning, upon arranging behavioural characters as a "plausible series of adaptational changes that could easily follow one after the other" (Alcock, 1984:432). Although intuitively pleasing, this method relies heavily on subjective, a priori assumptions concerning the temporal sequence of ethological modifications and dissociates character evolution from underlying phylogenetic relationships. This dissociation of history from behavioural evolution (see Ricklefs, 1987 for a recent discussion of the "eclipse of history" in ecology in general) has had an important impact on both the nature and direction of ethological research. Ethological questions are frequently examined via sophisticated cost-benefit analyses which often do not differentiate between mechanisms affecting the evolutionary origin of a character from those involved in its maintenance, once established, in the population. Consider the hypothetical situation presented in Fig. 2.1. Question: Why does species "C" show maternal care? The answer is straightforward using phylogenetic systematic reasoning. Traits which "look the same" are evolutionary homologues as long as they covary with the phylogenetic relationships of the taxa derived from other characters. In Fig. 2.1, the character "female parental care" occurs in taxa A, B and C. These taxa, in turn, form a group consisting of all the descendants of a common ancestor (a clade). In evolutionary terms, the existence of maternal care in species A, B and C is the result of the character arising in the common ancestor of the clade and persisting in all the descendants. The answer to the question "why does C show maternal care?" is thus simply "because its ancestor did" or, in the terminology of phylogenetic systematics, "the presence of this behaviour is a manifestation of phylogenetic constraints within the group". In this particular example, any additional explanations, such as cost-benefit analyses, involve character maintenance, not character evolution, for this trait. Cost-benefit analyses which endeavour to explain the original success of the trait, as opposed to the current success, must be based on assessments of the ancestor's biology in its environment. CHARACTER O U T G R O U P A B c (a) PARENTAL CARE BIPARENTAL FEMALE ONLY FEMALE ONLY FEMALE ONLY 18 OUTGROUP A CF") B ("F") C ("F") (b) OUTGROUP A B C (C) Fig. 2.1: Hypothetical example outlining the applicability of phylogenetic systematics to questions concerning the evolution of behaviour: (a) type of parental care present in the outgroup and in each member of the study group; (b) phylogenetic tree constructed from characters other than the parental care trait (i.e.: morphological data). Species accompanied by an "F" are those which show female parental care; (c) the same phylogenetic tree as in (b) with notation (*) indicating that distribution of female only parental care is correlated with the phylogenetic relationships of species A, B and C. The evolutionary interpretation is that the trait originated in the common ancestor of A + B + C (marked with an *), and has persisted in all the descendant species. This thesis is concerned with studying the evolutionary significance of male nuptial colouration in Gasterosteus aculeatus. The first step in such a study is the delineation of the origin and diversification of the character under investigation. The remainder of this chapter therefore presents a phylogenetic systematic analysis of the Gasterosteidae based solely upon behavioural characters. From this analysis two questions are examined: (1) Do behavioural characters contain phylogenetic information, i. e., are the patterns of behavioural evolution congruent with the patterns of morphological and genetical evolution in this family? (2) What are the macroevolutionary relationships between breeding colours and breeding behaviours in the Gasterosteidae? MATERIALS AND METHODS The Method: Phylogenetic Systematics Changes in systematic biology in the late 1950's and early 1960's were triggered as a reaction against a perceived lack of repeatable methodology and quantitative rigor in the discipline. Two major problems were addressed during this period: (1) homology and (2) levels of generalities in similarities. As previously discussed, Remane proposed a set of criteria for testing hypotheses of common origin (homology) without a priori reference to phylogeny. These criteria work well for establishing that some traits which appear to be "the same" are, or are not, "the same". However, certain traits that are homologous under Remane's criteria could conceivably be non-homologous evolutionarily. This would occur, for example, if two species showing the same ancestral polymorphism experienced similar selection pressures leading to fixation of the same trait. Because the fixed trait arose more than once evolutionarily its various instances among species are not evolutionary homologues. To implement an "evolutionary homology criterion" (Wiley, 1981), systematists needed a method for reconstructing phylogeny independent of assumptions of phylogenetic history. Taxa could not be grouped according to overall similarity, because similarity embodies three different phenomena. First, there is similarity in general homologous traits (e.g. humans, gorillas and elephants all have vertebrae, and vertebrae appear to have evolved only once, but the presence of vertebrae does not help determine that humans and gorillas are more closely related to each other than either is to the elephant). Second, there is similarity due to convergent and parallel evolution (jointly termed homoplasv). which conflicts with phylogenetic relationships. And third, there is similarity due to special homologous traits (e.g. birds and crocodilians have sub-mandibular fenestrae, a trait found in no other vertebrates), which is evidence of phylogenetic relationships. Two problems must therefore be solved: (1) how to distinguish general from special traits and (2) how to distinguish homology from homoplasy. A solution to these problems was provided by the German entomologist, Willi Hennig (1950,1966). Noting that there are different kinds of similarities and relationships, Hennig emphasized that the primary goal of systematics should be the delineation of a special type of similarity (homology) which, when used to reconstruct relationships, would provide a general reference system for comparative biology. He reasoned that this system should be based on reconstructing phylogenetic relationships from shared homologous traits because all homologies co-vary with each other and with phylogeny. Other types of similarity (homoplasy), although evolutionarily interesting, are not phylogenetically informative because homoplasies need not co-vary with anything. Hennig recognized that there are three types of homologous characters: (1) shared general characters, which identify a collection of taxa as a group; (2) shared special characters, which indicate relationships among taxa within the group; or (3) unique characters, which identify particular taxa within the group. In addition to these three categories of homology there is a separate category, homoplasy or "false homology", which tells us nothing about relationships among taxa. Since only shared special homologies denote particular phylogenetic relationships within a study group, characters must be assigned to one of the three homology categories before their usefulness in a phylogenetic study can be determined. Hennig suggested that this determination be made by comparing the state of each character in the study group to the state of the same characters in one or more species outside the study group (outgroups). In this way, each character is independently assigned a particular homology status (general, special, or unique) depending upon properties of species for which the phylogenetic relationships are not being assessed (the outgroups). Outgroup comparison allows researchers to distinguish traits that are shared between the outgroups and at least some members of the study group (shared general or plesiomorphic traits), traits that are restricted to some members of the study group (shared special or synapomorphic traits), and traits unique to single members of the study group (unique or autapomorphic traits). The members of the study group are then clustered according to their special shared traits. If there are conflicting groupings, it means that some traits assumed to be evolutionary homologies on the basis of non-phylogenetic criteria are actually homoplasies. Because all evolutionary homologies co-vary, and homoplasies do not co-vary, the pattern of relationships supported by the largest subset of special similarities is adopted as the working hypothesis of phylogenetic relationships. As more and more traits are sampled, there will be progressively more support for a single phylogenetic pattern. Traits that are inconsistent with this pattern are interpreted, post hoc, as homoplasies. Thus, the phylogenetic systematic method works in the following way: (1) assume homology, a priori, whenever possible; (2) use outgroup comparisons to distinguish general from special homologous traits; (3) group according to shared special homologous traits; (4) in the event of conflicting evidence, choose the phylogenetic relationships supported by the largest number of traits; (5) interpret inconsistent results, post hoc, as homoplasies. So, homologies, which indicate phylogenetic relationships, are determined without reference to a phylogeny while homoplasies, which are inconsistent with phylogeny, are determined as such by reference to the phylogeny. The Study Group The Gasterosteidae are restricted to temperate and sub-polar habitats in the northern hemisphere. Of the five genera, two, Pungitius and Gasterosteus, are both geographically widespread and morphologically variable. Two species of Pungitius are recognized, P. pungitius (the nine-spined stickleback) and P. platygaster (the Ukranian stickleback). The P. pungitius complex includes several subspecies, the status of which remains undetermined (Wootton, 1976). Within this complex the widely distributed P. pungitius pungitius has been the subject of intense behavioural examination; P. platygaster and the other P. pungitius subspecies are geographically restricted and largely undescribed in the ethological literature. This pattern is repeated in Gasterosteus where at least two distinct species, G. aculeatus and G. wheatlandi, have been described, and the existence of many more debated (Penczak, 1966;Hagen, 1967;McPhail, 1969, 1984;Miller and Hubbs, 1969;Hagen and McPhail,1970,Wootton,1976). In fact, the current taxonomic status of the G. aculeatus species complex is still best summarized by Hubbs (1929): "the systematic interpretation of the sticklebacks has long been - and still remains - one of the most perplexing problems of ichthyology." Although morphological differences between populations are often quite striking, behavioural differences, when mentioned, are usually very subtle; with the exception of male nuptial colouration. Red is the originally described and most widespread colour state but recently populations ranging from no colour to white to black have been reported (McPhail, 1969,pers.comm.;Semler, 1971;Moodie,1972;Hagen and Moodie,1979; Hagen et al.,1980). Whatever the resolution of the species relationships within this complex, red will be the condition for the species called Gasterosteus aculeatus [sensu stricto), by virtue of nomenclatorial priority. Behavioural data were collected through a literature search and supplemented with observations on anadromous populations of G. aculeatus. The following references were consulted: Spinachia spinachia (fifteen-spined or sea stickleback: Prince, 1885;Sevenster, 1951); Apeltes quadracus (four-spined stickleback:Reisman, 1963;Rowland, 1974a,b); Culaea inconstans (five-spined or brook stickleback: Winn, 1960;Reisman, 1961 ;Thomas, 1962; Reisman and Cade, 1967;McKenzie,1969a,b,1974;Moodie, 1986); Pungitius pungitius (nine-spined stickleback: Leiner, 193 l;Barraud,1955;Morris, 1952, 1958;McKenzie and Keenleyside,1970;Wilz,1971); Gasterosteus aculeatus (three-spined stickleback: Pelkwijk and Tinbergen, 1937; van Iersel, 1953, 1958;Sevenster, 1961; van den Assem, 1967;Wilz, 1970a,b, 1971,1972,1973; Wootton, 1976); Gasterosteus wheatlandi (black-spotted stickleback: Reisman, 1968a;McInerney,1969;Coad and Power, 1973). Initial determinations of plesiomorphic and apomorphic traits used the Aulorhynchidae (tube-snouts) as the outgroup. Nelson (1984) placed the Gasterosteidae, the Aulorhynchidae and, provisionally, the monotypic Hypoptychidae (sand eels: for which no behavioural data could be found) in the Gasterosteiformes. Wootton (1976) considered the Aulorhynchidae to be the sister-group of the Gasterosteidae. Although this family is composed of two monotypic genera, Aulorhynchus and Aulichthys, sufficient behavioural data are available only for Aulorhynchus jlavidus (the tube-snout: Limbaugh, 1962;Marliave, 1976;Gotshall, 1981). 23 CHARACTER DESCRIPTION 1. Nest building; 0=glue network only; l=glue network then collect/glue; 2=collect/glue only. 2. Nest placement; O=vegetation; l=substrate 3. Superficial glueing; 0=absent; l=present 4. Insertion glueing; 0=absent; l=present 5. Tunnel entrance; 0=absent; l=present 6. Tunnel exit; 0=absent; l=created by male 7. Male response to female; 0=lunge/bite; l=lunge/pummel; 2=zigzag 8. Female response to male; 0=turn towards; l=heads-up 9. Dorsal pricking during courtship; 0=absent; l=present 10. Head down threat; 0=absent; l=present 11. Circle fighting; 0=absent; l=present 12. Dorsal roll during nest showing; 0=absent; l=present 13. Nest showing; 0=hover; l=snout into nest; 2=snout above nest 14. Fan during nest showing; 0=absent; l=present 15. Male response to female in nest; 0=bite caudal peduncle area; 1= quiver flank 16. Extent of male nuptial colouration; 0=pelvic fins; l=whole body; 9=no colouration 17. Male nuptial colouration; 0=red; l=black; 9=no colouration 18. Male nuptial colouration; 0=red; l=gold; 9=no colouration 19. Number of mates per male; l=more than one 20. Male territorial; l=yes 21. Nursery construction; 0=absent; l=present 22. Nest ventilation; 0=absent; l=fanning; 2=sucking 23. Orientation during ventilation; O=horizontal; l=head down 24. Parental care; l=male only 25. Courtship dance; 0=absent; l=zigzag component; 2=tail flagging 26. Dorsal role submission; 0=absent; l=present 27. Fry retrieval; 0=absent; l=present Rationale For Character States Aulorhynchids and gasterosteids share three behavioural traits also found in other groups of fishes that are pertinent to a discussion of the evolution of mating behaviours: polygyny (character #19), male territoriality (character #20) and male only parental care (character #24). They are listed on the cladogram shown in Fig. 2.2 below the outgroup and the ingroup to indicate their origins prior to the evolution of the ancestor of the aulorhynchids and the gasterosteids. In addition, only aulorhynchids and gasterosteids are known to produce a substance in their kidneys that is used by males to glue plant material into a nest (character #3). This is a synapomorphy for aulorhynchids plus gasterosteids, and is listed as such in Fig. 2.2. Sixteen of the remaining twenty-three characters (characters # 2, 4, 5, 6, 8, 9, 10, 11, 12, 14, 15, 16, 21, 23, 26 and 27) are binary characters for which one state is found in the outgroup and in at least one member of the ingroup (the plesiomorphic state), and the other state is found among the rest of the ingroup members (the apomorphic state). For these characters "0" represents the plesiomorphic condition and "1" represents the apomorphic condition. Male nuptial colourations documented for gasterosteids include red, gold and black. Spinachia spinachia lacks nuptial colouration, whereas Aulorhynchus Jlavidus and Apeltes quadracus have red pelvic fins. Consequently red is considered the plesiomorphic colour. Members of Pungitius pungitius and Culaea inconstans have black nuptial colouration, members of Gasterosteus wheatlandi are gold and members of G. aculeatus (s. s.) are red. Functional outgroup comparisons (Watrous and Wheeler, 1981;Maddison et aL,1984) suggest that black and gold are independently derived from red. They have therefore been treated as two separate characters (characters 17 and 18). S. spinachia is coded as "9" for nuptial colouration characters 16, 17 and 18. This code is commonly used to indicate either missing data or the absence of a character, and is meant to indicate the ambiguous nature of synapomorphies based on the absence of a trait. Coding for characters 1, 7, 13, 22 and 25 is more complicated because there are three states involved. For characters 7 and 25, the plesiomorphic condition ("0") can be determined readily because it is found in A. Jlavidus and S. spinachia. For the other three characters the state found in the outgroup does not occur in the ingroup, so the plesiomorphic condition is unclear. These five characters were considered to be unordered, and the polarities found in the character description are based upon the degree of fit to the other characters and functional outgroup comparisons (see Wiley, 1981). The data matrix (Appendix A) constructed from the above characters was analyzed quantitatively using the PHYSYS computer program for phylogenetic systematic analysis developed by James S. Farris (State University of New York) and Mary F. Mickevich. RESULTS The phylogenetic tree based on the greatest degree of support by special similarities (Fig. 2.2) has a consistency index (Kluge and Farris, 1969) of 90.3% and an F-ratio (Farris, 1970) of 7.3. These are goodness-of-fit measures designed to indicate the degree of support for a particular tree by a particular data set. The consistency index ranges from 0 to 100%, and indicates the percent of all characters that support the given tree. A consistency index of 100% indicates that all the characters agree with the tree (i. e., no homoplasy). Fig. 2.2: Phylogenetic tree for the Gasterosteidae based upon behavioural characters. Numbers accompanying slash marks on the tree refer to characters and are coded in the following manner: the number preceding the parenthesis refers to a particular character; the number enclosed within parentheses refers to the state of that character. "9" indicates missing data. For convenience, homoplasious traits are marked with an asterisk. See data matrix (Appendix A) and methods section for a description of character states. The F-ratio is an open-ended measure for which a value of 0 indicates no discrepancy between the raw data and their representation on the phylogenetic tree. The high consistency index and low F-ratio indicate a high degree of historical constraint and a low degree of plasticity in the traits used to construct the tree. Four characters (# 3, 19, 20 and 24) are present in only one state in the gasterosteids. They do not contribute to the goodness-of-fit statistics, but they are relevant to the evolution of behavioural repertoires within the group. Three postulated cases of homoplasy occur, representing reversals to the ancestral state in Culaea inconstans: (a) nest construction via first laying down a network of glue then collecting plant material (character #1), (b) lack of a male-produced exit from the nest (character #6) and (c) pummel response by territorial males to females (character #7). Several phylogenetic hypotheses for the Gasterosteidae have been proposed utilizing morphological (Gill, 1885; Bertin, 1925; Leiner, 1934; Nelson, 1971) and electrophoretic (Chen and Reisman, 1970; Hudon and Guderley, 1984) characters. In addition, relationships based on a combination of morphological, biochemical, ecological, behavioural and paleontological data have been discussed (Reisman and Cade, 1967;Mural, 1973;Wootton, 1976). Wootton (1976) commented that " . . . a series of studies making use of modern techniques for studying phylogenetic relationships and which embrace both the Gasterosteidae and the Aulorhynchidae are required to clarify the situation." Two phylogenetic systematic studies have been published subsequently, the first by Paepke (1983, Fig. 2.3b) based upon predominantly morphological traits and the second by Hudon and Guderley (1984, Fig. 2.4a) utilizing chromosomal data. When the phylogenetic tree derived from ethological data (Fig. 2.2) is compared with these trees, no conflicts are found. Resolution of phylogenetic relationships is either improved (cf. Hudon and Guderley) or retained (cf. Paepke). In addition, when chromosome number (Chen and Reisman, 1970) is mapped onto the tree, Culaea inconstans demonstrates a reversal to the ancestral condition similar to the behavioural reversals noted for characters 1, 6 and 7. APELTES PUNGITIUS G. WHEATLANDI G. ACULEATUS Fig. 2.3: Phylogenetic trees for the Gasterosteidae based upon non-behavioural characters, (a) electrophoretic data (Hudon and Guderley, 1984; after their figure la) ; (b) primarily morphological data (Paepke, 1983; after his figure 9). 30 D I S C U S S I O N RECONSTRUCTING PHYLOGENETIC RELATIONSHIPS WITH BEHAVIOURAL CHARACTERS Given the tractability of ethological data, it is curious that behavioural characters are rarely used by systematists. In fact, the current state of behavioural systematics is still best summarized in a paper presented 25 years ago by R. D. Alexander (1962) during a symposium on the usefulness of nonmorphological data in systematic studies: . . . anyone with more than a passing curiosity about the study of animal behavior soon acquires the feeling that it has been neglected too frequently in many aspects of zoology, but especially among the systematists, who have almost a priority on the comparative attitude. . . Behavioral attributes are . . . too often at the core of diverse problems in animal evolution to allow us to get by with the vague feeling that structure and physiology can be compared but behavior cannot - that a structural description is important information but that a behavioral description is a useless anecdote. Systematist's apprehensions stem, in part, from the perception that behavioural characters are actions and therefore ephemeral. Many authors have voiced concern over the structural and temporal plasticity of behaviour, arguing that it is far more environmentally sensitive than morphology and thus difficult to characterize (Parsons, 1972;McClearn and DeFries, 1973; Dunford and Davis, 1975). On the surface, this concern seems valid. Verbs are intuitively more labile than nouns. These perceptions, however, are based on an assumption that variability within a species automatically disqualifies a character from examination of relationships among species. Consider the following example: male three-spined sticklebacks court females with an elegant, horizontal zigzag dance. The form, sequence and duration of the dance components are highly variable both within and among individuals. However, no male Gasterosteus aculeatus ever executes the staccato, vertical zigzag courtship dance typical of the related species Pungitius pungitius, and no other members of the Gasterosteidae dance like G. aculeatus. The character is diagnostic for the species and is therefore a useful systematic tool at that level. The congruence among cladograms derived from biochemical, morphological and ethological data for the Gasterosteidae, clearly demonstrates that behavioural characters are useful in assessing phylogenetic relationships. In light of the success of the early comparative ethologists, this result should not seem surprising. Behaviour is an aspect of an individual's phenotype. Closely intertwined with ontogeny, physiology and morphology and thus subject to the same constraints, it should change in ways that reflect underlying phylogenetic relationships. Behaviour is also one of the cohesive forces holding individuals of a species together; though in many cases plastic, it is never chaotic. By analyzing and comparing structural (biochemical/ morphological) and functional (ecological/behavioural) aspects of the phenotype, future phylogenetic studies will develop a more comprehensive picture of the patterns of biological evolution, for evolution, like behaviour, is a dynamic process. THE PHYLOGENETIC APPROACH TO STUDYING BEHAVIOURAL EVOLUTION A phylogenetic tree is an hypothesis of historical relationships based on sequences of character origin and modification. Phylogenetic systematic analysis therefore offers ethologists the opportunity to trace both the evolutionary fate of individual characters (hypotheses concerning origin and modification) and the potential interactions among those characters (hypotheses concerning co-adapted trait complexes). Tracing macro-evolutionary relationships among characters can be a useful way to disentangle the potential contributions of various selection pressures in the evolution of sexually dimorphic nuptial colouration in gasterosteids. For example, the initial elaboration of male nuptial colouration (character #16) is preceded by an increase in the complexity of courtship (character #25) on the phylogenetic tree. This result agrees with intersexual selection theory. An increase in the intricacy of signal exchange (a) focuses the receiver's attention on the sender and (b) increases the amount of information available in the interaction. This creates the potential for the evolution of differential female response (female choice) to variability in a male character. Apparently this potential was realized in the ancestor of the Pungitius + Culaea + Gasterosteus clade, where nuptial colouration underwent an exaggeration (from pelvic spines to the entire body) consistent with Fisher's (1930) runaway sexual selection scenario. The phylogenetic tree supports an interpretation that initially colour was not necessary in courtship {Spinachia spinachia lacks colouration) but that it later became intimately involved with male/female interactions. Certainly 32 the elaboration of colour and courtship, once associated, are tightly coupled on the tree (Fig. 2.4). Fig. 2.4: Relationship between diversification of colour and courtship in the gasterosteids. Capital letters refer to taxa: X = outgroup; GW = Gasterosteus wheatlandi ; GA = Gasterosteus aculeatus. Boxes represent the following nuptial colors: white= red pelvic spines only; striped= no color; black= black body; lightly stippled= gold body; boldly stippled= red body. The orientation of the zigzag courtship dance is mapped beneath each appropriate species. * indicates that the courtship dance has been replaced by a tail flagging display in Culaea inconstans. Overall, there are 5 colour states associated with 5 courtship states. Colour may also play a role in male/male interactions (intra-sexual selection) as part of a threat display, allowing an individual to assess the social status, experience and motivational state of an opponent. Depending on the system employed, the information exchanged during male/male interactions may contain elements of both truth and bluff. However, once an encounter has escalated into fighting, it is difficult to envision a role for colour; the emphasis should shift to factors directly involved with fighting performance (stamina, skill, strength). Because colour elaboration is initially associated with the appearance of fighting behaviour (character #11) and not with threat behaviour (character #10) on the phylogenetic tree (Fig. 2.5), it appears that any subsequent function of colour in male/male interactions is secondary to the more direct coupling of colour and male/female interactions. This proposition is strengthened by the observation that, unlike courtship, once threat/fight behaviours arose they remained very conservative, whereas colour continued to be elaborated (compare Fig. 2.4 and 2.5). So, macroevolutionary analysis suggests that, although colour may have some function in intra-sexual selection, the evolution of nuptial colouration in gasterosteids is better correlated with intersexual selection. Fig. 2.5: Relationship between diversification of colour and aggressive behaviours in the gasterosteids. Capital letters refer to taxa: X = outgroup: GW = Gasterosteus wheatLandi ; GA = Gasterosteus aculeatus. Boxes represent the following nuptial colors: white= red pelvic spines only; striped= no color; black= black body; lightly stippled= gold body; boldly stippled= red body. Behaviours involved in male/male interactions have been mapped onto the tree. Overall, 5 color states are associated with 3 changes in aggressive behaviours. The relationship between colour and parental care falls between the intra- and intersexual selection patterns. Like courtship, the initial elaboration of colour is associated with an increase in parental care (prolonged fry retrieval; character #27). This supports the hypothesis that natural selection played a role in the elaboration of the colour signal in this ancestor. However, past this point in phylogeny, changes in colour and parental care are not as closely associated as the macroevolutionary relationship between colour and courtship. Parental care has remained 34 relatively conservative among sticklebacks, whereas colour has diverged extensively (Fig. 2.6). • • • • • X Spinachia Apeltes Pungitius Culaea GW G A Fig. 2.6: Relationship between diversification of colour and parental behaviours i n the gasterosteids. Capital letters refer to taxa: X = outgroup; GW = Gasterosteus wheatlandi ; G A = Gasterosteus aculeatus. Boxes represent the following nuptial colors: white= red pelvic spines only; striped= no color; black= black body; lightly stippled= gold body; boldly stippled= red body. Parental care behaviours have been mapped onto the tree. Overall, 5 color states are associated with 3 changes i n parental care behaviours. The preceding discussion emphasizes a major asset of the use of phylogenetic trees in comparative ethology. There are a number of microevolutionary studies that suggest colour has been influenced by intrasexual selection, natural selection and intersexual selection. At the macroevolutionary level, as indicated by the phylogenetic tree, the diversification of colour is associated weakly with the diversification of agonistic behaviours (intrasexual selection), moderately with parental care behaviours (natural selection) and strongly with courtship behaviours (intersexual selection). Apparently, the evolutionary diversification of colour has been tied to the evolution of specific mate recognition systems, although colour has other positive functions within species as well. Thus, by using the phylogenetic tree, we need not opt for explanations using only one mechanism; we may ask the degree to which each of a series of processes may have played a role in evolution. The importance of the phylogenetic approach is further emphasized by examination of the following statement in light of the gasterosteid tree: "One of the early important insights about mating system evolution was the recognition that the form of mating systems is more closely correlated with environmental contexts than it is with phylogenetic heritage." (Vehrencamp and Bradbury, 1984). Within the Gasterosteidae, Pungitius pungitius and Gasterosteus aculeatus are habitat generalists. G. aculeatus, for example, exists as marine, estuarine, anadromous and freshwater populations. Among the freshwater populations, habitats range from ephemeral, weed-choked ditches to large, oligotrophic lakes. At the other end of the spectrum, the fifteen and five-spined sticklebacks are habitat specialists: Spinachia spinachia is restricted to marine habitats and Culaea inconstans to freshwater ones. Apeltes quadracus falls somewhere between these extremes, preferring marine/brackish habitats but venturing into freshwater areas on occasion. If type of mating system is closely correlated with environmental factors, then a variety of mating systems should have evolved under the diverse selection pressures encountered by members of this family. However, all gasterosteids exhibit a male territorial (character #20), polygynous (character #19), paternal care (character #24) mating system. Apparently these behavioural patterns are a reflection of phylogeny and not a response to specific local environments. In conclusion, then, the results of this study have demonstrated that behavioural traits can be employed to reconstruct phylogenetic relationships. As with morphology, some traits are more amenable to analysis than others. But the technical problems and method of tree construction seem no different whether one is using morphological or behavioural characters. When ecological, behavioural, morphological and geographical data are combined in a phylogenetic framework, a robust characterization of evolutionary history is produced. The resultant phylogentic tree, in turn, provides a framework of predictions for further ethological studies. Ethology, as a science traces its origins to Darwin. This chapter was opened with a quotation from The Origin of Species. In closing, the quotation is completed: . . . community of descent is the hidden bond which naturalists have been unconsciously seeking, and not some unknown plan of creation, or the enunciation of general propositions and the mere putting together and separating of objects more or less alike. (1872:346) SUMMARY The cladogram produced by phylogenetic systematic analysis of the Gasterosteidae based solely upon behavioural characters is congruent with trees based upon morphological and electrophoretic data. Behaviour, like other aspects of an organism's phenotype, contains information of phylogentic relationships. Current behavioural patterns are therefore an interaction of past (historical constraints) and present (environmental selection). Studies of the evolutionary significance of a character should take both factors into account to avoid confusing character maintenance (stasis) with character origin and diversification (evolution). The macroevolutionary examination of patterns and relationships between characters provides a framework of predictions for micro-evolutionary (within species) experimental examination. Based upon the macroevolutionary relationships between breeding colours and breeding behaviours in the gasterosteids, three predictions, relevant to a discussion of nuptial colouration in G. aculeatus, are proposed: nuptial colouration is (1) weakly associated with male/male interactions (intra-sexual selection), (2) moderately associated with parental care (natural selection) and strongly associated with male/ female interactions (intersexual selection). 37 CHRPTER THREE T E M P O R A L C H A N G E S IN THE STRUCTURE OF M A L E THE N U P T I A L S I G N A L INTRODUCTION Nuptial signals of fishes are often composed of a mosaic of colours varying across both structural and temporal dimensions (Kortmulder, 1972; Baldaccini, 1973;Lanzing and Bower, 1974;Neil, 1984). Gasterosteus aculeatus males in breeding livery are a patchwork of red, blue and black. Several authors have noted (1) an intensification of ventral/lateral red and concomitant decrease in dorsal/lateral melanism during courtship and (2) a decrease in red and intensification of melanism during parental care stages (Titschack,1922; Wunder,1930; Craig-Bennett, 193l;Ikeda, 1933; van Iersel, 1953;Sevenster, 1961). In this chapter I examine the endogenous rhythm of changes in three colours (red body, blue eyes, black body) across a complete breeding cycle to characterize more fully the temporal and morphological structure of the male nuptial colouration signal in G. aculeatus. S I G N A L P R O D U C T I O N The bright colours displayed by male G. aculeatus during the breeding season result from the deposition of carotenoids in special integument cells. The biosynthetic origin of carotenoids lies in plants, bacteria and fungi. Animals cannot synthesize carotenoids de novo and thus ultimately depend upon dietary sources of these pigments. Sexual selection has been postulated to involve several genetic mechanisms (Fisher, 1930; Zahavi, 1975; Kodric-Brown and Brown, 1984). To investigate the potential importance of these mechanisms in shaping the three-spined sticklebacks' nuptial colouration signal, it is necessary to understand the basis of signal production in this species. (i) W H A T DO S T I C K L E B A C K S E A T ? Gasterosteus aculeatus exists as marine, estuarine, anadromous and freshwater populations. Amongst the freshwater populations, habitats range from ephemeral, weed-choked ditches to large, oligotrophic lakes. One of 38 the factors contributing to the success of this species is its opportunistic feeding behaviour; the diet includes algae, diatoms, molluscs, oligochaetes, insect larvae and pupae, crustaceans and fish eggs (Hartley, 1940,1948; Hynes,1950;Walkey,1967; Manzer,1976; Adalsteinsson,1979; Snyder, 1984). Of these prey items, crustaceans - primarily cladocerans and copepods - and chironomid larvae dominate the diet (Table 3.1): SITE FOOD TYPE A B C D E F G H 2 4 1 3 2 4 1 4 4 4 4 % CRUSTACEA 307 64.1 57.7 36.7 32.2 13.2 81.0 60.0 38.5 90.0 24.0 % CHIRONOMID 16.6 22.8 14.6 56.2 37.8 57.8 10.0 6.1 60.7 7.2 51.0 TOTAL % CHIRONOMID + CRUSTACEA 47.3 86.972.3 92.9 70.0 71.0 91.0 66.1 99.2 97.2 75.0 Table 3.1: % composition of major food types i n the diet of various G. aculeatus populations. Numbers refer to seasons: 1= autumn; 2= winter; 3= spring; 4= summer. Capital letters refer to the following collecting sites: A= Birket stream, England (Hynes,1950); B= Monkton Pond, England fWalkey.1967); C= San Pablo Creek, California (Snyder, 1984); D= Windemere Lake, England (Hynes,1950); E= Easdale Quarry, England (Hynes,1950); F= Lake Myvatn, Iceland (Adalsteinsson.1979); G= Great Central Lake, B . C . (Manzer.1976); H= River Cam, England (Hartley, 1940; 1948). (ii) CAROTENOIDS IN THE FOOD a-carotene, lutein, p*-carotene and its derivatives, canthaxanthin and astaxanthin, are the most widely distributed carotenoids in the major prey items of G. aculeatus (Table 3.2). Within a species, astaxanthin exists in the highest concentrations relative to other pigments in the majority of the crustaceans. Canthaxanthin, is important in the anostracans (i. e., Artemia) and isopods. 39 SPECIES SPECIFIC PIGMENTS 1 2 3 4 5 6 7 8 OTHER REFERENCES PIGMENTS Branchiopoda Artemia salina Daphnia magna D. longispina Holopedium gibberum x x x x x x x x X 10.12,13 a-f 14,15 14,15 g-j Herring(1968);i Coneooda Cyclops s. strenuus x x x Diaptomus bacillifer x x D. castor x x Hemidiaptomus amblydon x x Eudiaptomus amblydon x x x Ostracoda Heterocypris incongruens x x Cyclocypris laevis x x x x x x x x x x x I m m m Czeczuga and Krywuta,1981 Green, 1959 Czeczuga, 1976 Amph ipoda Gammarus lacustris G. pulex x x x x x 10,11,14 Czeczuga. 1980b 16.17,18 x x Beatty,1949;k Insecta Chironomus annularis 11 Czeczuga, 1970 Stickleback eaqs 15,16 Czeczuga, 1980a Table 3.2: Carotenoid distribution in representative prey items of Gasterosteus aculeatus. Numbers refer to the following pigments: 1= B-carotene; 2= Isozeaxanthin; 3= Astaxanthin; 4= Echininone; 5= Canthaxanthin; 6= Violaxanthin; 7= a-carotene; 8= Lutein; 9= y-carotene; 10= Isocryptoxanthin; 11= B - cryptoxanthin; 12= Phoeniconone; 13= Crustaxanthin; 14= Phoenicoxanthin; 15= Neothxanthin; 16= oc-Doradexanthin; 17= Tunaxanthin; 18= Mutatochrome. Small case letters refer to references : a= Gilchrist and Green (1960); b= Krinsky (1965); c= Davies et al(1970); d= Czeczuga (1971); e= Hsu et al (1970); f= Gilchrist (1968); g= Green (1957); h= Gross and Budowski (1966); i= Paanakker and Hallegraeff (1978); j= Partali et al (1985); k= Barrett and Butterworth (1968); 1= Czeczuga and Czerpak (1968); m= Czeczuga and Czerpak(1966). 40 (ii i) C A R O T E N O I D S IN S T I C K L E B A C K S Several researchers have attempted to identify the pigments in G. aculeatus. Investigations of carotenoid distributions both within and among populations have yielded several interesting results (Table 3.3): FRESHWATER POLAND] BALTICl FRESHWATER2 SEA JAPAN PIGMENTS JUVENILE MALE MATURE FEMALE MATURE MALE MATURE MALE MATURE MALE p-Carotene 5.4 7.3 0.0 0.0 0.0 TOTAL 54 7.3 0.0 0.0 0.0 cryptoxanthin 16.6 14.1 1.0 0.0 0.0 zeaxanthin 35.9 0.0 0.0 15.1 12.3 diatoxanthin 0.0 0.0 0.0 0.0 3.3 cynthiaxanthin 0.0 0.0 0.0 0.0 7.2 TOTAL # 525 141 1.0 15.1 22.8 epoxide 0.0 9.6 2.6 41.1 0.0 mutatochrome 0.0 8.3 0.5 8.6 0.0 TOTAL ** 0.0 17.9 3.1 49.7 0.0 canthaxanthin 0.0 0.0 12.8 0.0 0.0 astaxanthin 24.1 8.6 60.7 22.8 16.0 TOTAL *** 24.1 8.6 735 22.8 16.0 a-cryptoxanthin 0.0 0.0 0.0 0.0 1.5 lutein 15.0 33.0 8.4 4.9 20.4 TOTAL * 15.0 33.0 84 49 21.9 doradexanthin 3.0 1.0 13.2 7.5 0.0 TOTAL *** 3.0 1.0 13.2 75 0.0 e-neothxanthin 0.0 11.2 0.0 0.0 0.0 tunaxanthin 0.0 0.0 0.0 0.0 39.3 TOTAL * 0.0 11.2 0.0 0.0 39.3 Table 3.3: Percent distribution of carotenoids i n three populations of Gasterosteus aculeatus. Subscripts refer to the following studies: 1= Czeczuga, 1980a; 2 = Matsuno and Katsuyama,1976. Aster isks refer to chemical classes of pigments: * = hydroxy-compounds; ** = epoxy-compounds; *** = ketones. (a) intersexual differences: within a population, mature males and females store virtually the same pigments but in different relative amounts: females retain hydroxy/ epoxy compounds (65% of total content) and males retain ketones (87% of total content). (b) intra-sexual differences: developmental changes in the male's total pigment profile entails both a minor alteration in the types of carotenoids present and a major shift in the relative proportions of the remaining pigments. In breeding males, the concentration of the ketones increases while the concentrations of all other pigments decrease. Brush and Reisman's preliminary analysis (1965) of California freshwater populations also suggests that nuptial colouration results primarily from quantitative rather than qualitative changes in carotenoids. (c) nuptial colouration: intense red colouration has been linked to the presence of two p-ketones, astaxanthin and canthaxanthin, in a variety of teleost species (Kanemitsu and Aoe, 1958; Crozier, 1970; Czeczuga, 1975, 1979a,b). These pigments predominate in the mandibular region of the breeding male (81% of total content) and are the major carotenoids stored in the skin, muscles and liver (70% of total content: Czeczuga, 1980a). Breeding males thus carry an internal pool of carotenoids which may be available for nuptial colouration development independent of daily changes in foraging success. (d) interpopulation differences: Astaxanthin (the "red" pigment) has been detected in all G. aculeatus populations analyzed to date, establishing a common basal breeding colour for the species, with interpopulation variation arising from either the addition of new pigments or changes in the relative concentrations of old ones (Crozier, 1967). For example, the two Polish populations share 67% of their pigments, and these pigments, in turn, comprise approximately 85% of the total carotenoid content. The fact that these groups inhabit ecologically diverse freshwater and marine environments further highlights the striking similarities between them. The Japanese population is radically different both in the type and proportion of stored pigments. This group shares only 2 of 12 carotenoids with the Polish freshwater population; approximately 36% of the total content. (e) fate of dietary carotenoids: Of the thirteen pigments identified in this species, 11 are found in their principal prey items (Table 3.2). The remaining two carotenoids, diatoxanthin and cynthiaxanthin, have not been identified in any of the primary prey items; however, they are common pigments in flagellated, autotrophic protistans and in diatoms (Rabourn et al., 1954; Jeffrey, 1961). Since these organisms have been reported in the diet of some populations (Saunders, 1914; Hynes,1950), their presence does not unequivocally demonstrate the existence of carotenoid metabolic pathways in three-spined sticklebacks. This observation is important to studies concerned with the genetic basis of nuptial colouration. For example, the progeny of Salmon River anadromous populations were raised to maturity in the laboratory. During this time individuals were fed ad libitum on the usual laboratory diet of Tubifex and Artemia. Tubificids contain no carotenoids; Artemia naupliae contain only echinone and its metabolite canthaxanthin (Beddard,1892;Krinsky, 1965). Males raised on this diet spawned but did not develop the nuptial colouration characteristic of their natal population. The extent of breeding colour development varied slightly among individuals but was never more than a hint of red in the mandibular region. Canthaxanthin plays a minor role in the development of nuptial colouration in at least one population (Table 3.3) therefore it is reasonable to assume that the colour noted in the laboratory individuals was due to the presence of that pigment. The lack of intense colour given the constant supply of canthaxanthin in Artemia, indicates that this population of Gasterosteus aculeatus is not capable of converting canthaxanthin to astaxanthin. Interestingly, progeny of a freshwater population collected from Enos Lake, Vancouver Island developed normal breeding colouration on the same diet. The relationship between absolute and relative amounts of pigments is a subtle one with interesting implications for studies of sexual selection in G. aculeatus. If differences amongst individuals in a population are largely a result of shifts in relative proportions of the same pigments, then males may differ in both intensity (variation in absolute amount of pigment) and hue (variation in relative proportions of pigments). Further studies are necessary to resolve this matter for an understanding of the mechanism of signal production is a necessary component in explanations about the evolution of the signal's function. MATERIALS AND METHODS 43 The breeding cycle of Gasterosteus aculeatus can be divided into four behaviourally distinct stages (van Iersel,1953;Sevenster,1961;Kynard,1978): (i) Territory acquisition and maintenance: high levels of aggression directed towards all conspecifics; the appearance of nest-related activities (sand digging, collecting, glueing, boring, pushing, some fanning). (ii) Courtship: appearance of courtship behaviours (zigzag dance, dorsal pricking, circling, leading); decrease in aggression directed towards receptive females with concomitant increase in aggression directed towards male conspecifics; sporadic nest-building and fanning bouts. (iii) Egg guarding: decrease in propensity to court receptive females; decrease in aggression towards conspecifics in general; major increase in duration and number of fanning bouts. (iv) Fry guarding: increase in aggression directed towards all conspecifics; disappearance of courtship behaviours; decrease in fanning. Males from the stock group collected on May 14, 1987 were placed in the test aquaria on June 8th and allowed 5 days to establish territory boundaries and build a nest possessing both a tunnel entrance and exit (a complete nest). Males which had not started nest construction within 48 hours were removed and replaced with individuals from the stock tank. An extra three days were included in this stage to examine whether the presence of a completed nest would have any effect on colour. Two ovulated females were released into each male's territory on the morning of day 6. Three days of courtship (days 6-8) were allowed to determine the constancy of the signal during this stage. "Spawned" females were replaced with gravid individuals at 8:00 hours every day. Gravid females respond to male courtship but do not spawn (Lam et al, 1978). The parental phase of the breeding cycle is divided into two stages: egg and fry guarding. The length of the egg guarding stage is a function of temperature and the time between successive clutches. Under the temperature regime maintained throughout this experiment, embryo development to hatching occurred in approximately 8 days (van 4 4 Iersel, 1953). Colour during the fry guarding stage was measured over a 10 day period, beginning with the appearance of fry outside the nest. Although the experiment was terminated at this point, all males were still guarding the fry. The phylogenetic analysis presented in chapter 2 suggested that the evolutionary elaboration of colour is primarily associated with changes in courtship behaviour and secondarily with changes in fry guarding behaviour. If macroevolutionary changes are mirrored in microevolutionary events, the same relationships between colour and breeding behaviours should occur in a within species (microevolutionary) study: colour will reach a maximum peak during courtship and a second peak during fry guarding. RESULTS (R) G E N E R R L OBSERURTIONS Two of 26 males did not begin nest building within the allotted 48 hours and were removed from further analysis, and 5 more died before completion of the breeding cycle. The remaining 19 males completed their nests by the afternoon of day 2, with the majority of completions occurring within the first day. Every male spawned within 5 hours of the ovulated females' introduction into his territory and courtship was frequently observed in the remaining 2 days of the courtship stage. The remaining gravid females had not ovulated therefore 3 days of courtship colour data were obtained without the confounding problem of intermale variability in number and timing of spawning bouts. Since a male was guarding only one clutch under constant temperature conditions, hatching was relatively synchronous both within and between clutches. This served to standardize the length of the egg guarding stage (5 days) and to sharply define the transition from egg to fry guarding. The appearance of fry coincided with an increase in male aggressive behaviour. Lunges and bites were directed towards the observer and, occasionally, bubbles of air rising from the filter. The intensity of this behaviour was not measured; as a general observation, all males showed the initial increase but to differing degrees and for different lengths of time. All these results agree with data collected for other populations (Wunder,1930;Jepps,1938; van Iersel,1953; Baggerman, 1958; Sevenster,1961; Segaar,1961; Peeke et al.,1969; Wootton, 1971; Huntingford, 1976b). (B) DATA ANALYSIS (raw intensity scores are presented in Appendix B). Temporal Fluctuations in Colour Red Body (i) Total colour score (intensity) Fig. 3.1 depicts the changes in total colour score over one complete breeding cycle. During the nest building/maintenance stage there is a low level fluctuation of male nuptial colouration (days 1-5); males do not respond to the presence of a completed nest by increasing their colour. Colour intensity peaks and remains constant during courtship (days 6-8), declines during egg guarding (days 9-13) and peaks again briefly at a time that coincides with the appearance of fry and an increase in aggression. This second peak is followed by a gradual decline in colour intensity; reaching a plateau from approximately day 17 to the end of the test period. This experiment was designed to investigate whether a male's colour score at any given stage x of the breeding cycle is different from his score at stage y. Since information about both the magnitude and direction of change in colour scores is available for these related samples, a Wilcoxon matched-pairs signed-ranks test (one-tail, corrected for ties) is a powerful test of the significance of these changes (Siegel,1956; Sokal and Rohlf,1969: 399). Five comparisons from this experiment produce unequivocal results: total colour score was (a) higher for all individuals during courtship, egg and fry guarding when compared with their scores during nest building and (b) lower for all individuals during egg and fry guarding in comparison to courtship. The increase in colour intensity associated with the appearance of fry (colour on day 13 and 14: * on Fig. 3.1) is significant (N=17; T=23; p < 0.01). Of 19 males, 13 increased in intensity (2-30 point increase), 4 decreased (1-4 point decrease) and 2 remained constant. Differences between the courtship and fry guarding peaks are more difficult to assess because the duration of the former was controlled during the course of this 46 experiment while the latter was not; the "fry guarding peak" was only defined as a result of the initial data analysis. Since there is no standard method for such an a posteriori comparison, the mean values for equivalent time periods - 3 days of courtship (days 6-8) and the first 3 days of fry guarding (days 14-16) - were chosen as the most objective way to test this problem. Wilcoxon analysis demonstrates a significant difference between the courtship and fry guarding peaks (** on Fig. 3.1): individuals are more intensely coloured during courtship (N=18; T= 6.5; p < 0.001). A more rigorous test was performed by comparing courtship with the actual peak in fry guarding (days 14 and 15). Once again colour intensity during courtship is significantly greater than colour intensity during the initial stages of fry guarding (N= 19; T=20: p < 0.005). 25 -1 ** 20" 15 " 10 -5" 0 1 T — i — i — r 2 3 4 5 - nest —) T — r 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 (court) (egg guard) ( fry guard ) DAY Fig. 3.1: Change i n total colour score (intensity of red) of Gasterosteus aculeatus males (n=19) across a complete breeding cycle. The breeding cycle is broken into four distinct stages: days 1-5: nest-building/maintenance; days 6-8: courtship; days 9-13: egg guarding; days 14-23: fry guarding. A l l nests were completed by the end of day 2. Asterisks refer to the results of two Wilcoxon analyses: * = significant difference (p < 0.01) between colour scores of individuals on day 13 (last day of egg guarding) and day 14 (first day of fry guarding); ** = significant difference (p < 0.001) between courtship peak (days 6-8) and fry guarding peak (days 14-16). 47 The reliability of the information transmitted by the colour signal depends, in part, upon the continuity of the signal between different stages in the breeding cycle. Spearman rank correlation analyses reveal a significant positive association between an individual's colour score at all stages of the breeding cycle (Table 3.4). This association is strongest between courtship and all subsequent stages. STAGE COURT EGG FRY FRY PEAK NEST 0.75 0.83 0.76 0.68 COURT - 0.93 0.95 0.92 EGG - - 0.87 0.84 Table 3.4: Spearman rank correlations between red body intensity scores at different stages of the breeding cycle. All scores are significant at p < 0.001. Intermale variability in total colour score (intensity) is superimposed upon the pattern of endogenous colour change across the breeding cycle (Fig.3.2). The range in colour scores across the population of nesting males is narrow and the distribution of scores clusters around the low end of the intensity scale (skewed to the right). During courtship, both the range and distribution of scores expand to maximum levels. Following courtship there is a trend towards a decrease in intermale variability; however, given the small sample size, it is difficult to assess the degree of this decrease. An important prediction of intersexual selection theory is thus corroborated by the distribution of colour scores in this population; intermale variability in one component of the male nuptial signal (intensity of red) reaches its maximum level during male/female interactions. 48 16 - • NEST | C O U R T H E G G G U A R D • FRY G U A R D m 1 2 3 4 5 6 COLOUR INTENSITY CATEGORIES (RED BODY) Fig. 3.2: Histogram of the intensity scores (red body) of G. aculeatus males at different stages in the breeding cycle. Mean values of colour scores for each male within a particular stage of the cycle are plotted. Nest = days 1-5; court = days 6-8; egg guard = days 9-13; fry guard = days 14-23. Intensity categories represent the following colour scores: 1= 0-9.5; 2=9.6-19.5; 3=19.6-29.5; 4=29.6-39.5; 5=39.6-49.5; 6=49.6-60.0. (ii) Colour distribution Changes through time in the colour intensity of the six anatomical landmarks mirror the total score pattern depicted in Fig. 3.1. Colour increases during courtship, falls during egg guarding, peaks briefly when the fry appear, and then decreases gradually in each area measured (Fig. 3.3). Wilcoxon matched-pairs signed-ranks tests of the differences in colour score between the base-line nest building level and the three remaining breeding stages were performed for each of the six areas scored. Fry guarding was divided into two stages: (1) the initial peak following the appearance of fry (days 14-16) and (2) the following plateau (days 17-23). The colour scores of the lower jaw and opercular region were higher for all individuals during courtship, egg and fry guarding than during nest building. To minimize errors introduced by performing a large number of independent tests for the remaining comparisons, the critical T values for each N were divided by the total number of tests; 16. 49 ca. O U CO a. (a) § d 6-5" 4-3" 2 -1 • • • a • • • • • D • • a -|—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 DAY • • (b) a. O QZ ZD 9 o o 2H H • + + • • • " • + + • SB • • • • • • • # * • • • • ~~l—I T—I—I 1 1 1—I 1 1 1 1 1—I 1 1 1—I 1 1—I 1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 DAY Fig. 3.3: Changes in colour intensity (red) in each of the six areas scored on G. aculeatus males across a complete breeding cycle, (a) mandibular area: solid squares = lower jaw and operculum; open squares = ventral area from the operculum to the pelvic spines; (b) posterior two-thirds offish: open squares = pectoral plate; diamonds = ventral area from pelvic spines to caudal fin; crosses = lateral plates (see chapter 1, Fig. 1.2 for a diagram of these areas). A consistent pattern for changes in colour distribution emerges from this analysis (Table 3.5). During courtship, intensity increases in all areas. Following courtship there is a gradual reduction of colour towards base-line distribution levels. This reduction follows an orderly sequence beginning with the posterior third of the fish during egg guarding (lateral plates and ventral area from pelvic spines to caudal fin), the pectoral plate region during the initial stages of fry guarding and finally the ventral area under the operculum to the pelvic spines during the latter stages of fry guarding. Throughout this period colour is retained in the lower jaw/opercular region. Time lapse photography would create a dynamic series of colour images beginning with a burst of "red" over the entire ventral and lateral surface of the fish, followed by colour loss sweeping forward from the caudal fin and stopping at the mandibular area. AREA COURT EGG FRY PEAK FRY 2 u * « » »•* * « • ftftft OP tt ft * ttttft *ft* • ftft VPS ft ft ft * NS pp * ft *ft NS NS PSC NS NS NS LP NS NS NS Table 3.5: Results of Wilcoxon analyses of the differences in colour scores (red) between nest-building and the remaining stages of the breeding cycle for the six areas measured on each male G. aculeatus. LJ= lower jaw; OP= operculum; VPS= ventral area from beneath the operculum to the pelvic spines; PP= pectoral plate; PSC= ventral area from the pelvic spines to the caudal fin; LP= lateral plates. Asterisks refer to the following probability values: * p<0.005; ** p<0.001; *** all males increase; no statistical test necessary. NS =no significant difference. If colour distribution is decreasing across the breeding cycle, what accounts for the second peak in overall intensity during the initial stages of fry guarding (Fig. 3.1)? At any given time colour is most intense in the anterior third of the male (lower jaw / operculum = mandibular region); however, this intensity is not static through time (Fig. 3.3a). The decrease in intensity in the mandibular region during egg guarding is significant (N=18; T=ll; p < 0.001), as is the increase from the last day of egg guarding (day 13) to the first day of fry guarding (day 14) (N=17; T= 24.5; p < 0.01). There is no significant difference between the courtship and fry peaks. The second peak in overall colour score can thus be attributed to a surge of colour to courtship levels in the mandibular region, combined with a loss of colour in the posterior third of the fish and the pectoral plate. Colour in this second peak is therefore less intense and differently distributed than it is during courtship. (iii) Hue Two colours, chrome orange and flame scarlet could be identified reliably. These colours lie close to one another on the hue scale for orange/red: chrome orange falls near the end of the orange scale (colour 16; Munsell notation = 2.5 YR), whereas flame scarlet is placed at the beginning of the red (colour 15; Munsell notation = 10.0 R). Many observations fell between these two colours, therefore the score 15.5 was added to the colour repertoire. These subjective scores allow a preliminary analysis of the changes in hue during the breeding cycle in this population. The relationship between colour hue and colour intensity is a complex one, therefore only a coarse analysis is presented here (Fig. 3.4). Hue and intensity ranks collected for the 19 individuals over the 23 day breeding cycle were examined (total n=437; scores of 0 are not included). Ninety percent of the chrome orange scores are associated with low colour intensities (1-10), whereas 85% of the flame scarlet scores are associated with high colour intensities (31-60). The intermediate hue is coupled with intermediate intensities (87%: 11-30), which overlap the scores associated with both chrome orange and flame scarlet. The association between hue and intensity is reflected in intramale variability in colour across the breeding cycle. Male hue lies at the orange end of the spectrum during the initial stages of nest building and, as intensity gradually increases, moves towards the red. This relationship is also reflected in intermale variability. Spearman rank correlation of an individual's mean colour score and hue over the entire breeding cycle demonstrates a strong positive association between the two (p = 0.94). Bright males appear red, dull males appear orange. 10.0% 1.00?? 12.00% 15.00% 85.00% 90.0% 69.00% CHROME ORANGE (16) INTERMEDIATE (15.5) FLAME SCARLET (15) colour score H i-io • 11-20 • 21-30 ^ 31-60 Fig. 3.4: Relationship between colour intensity and perceived hue; data from all males pooled. Each graph represents a particular hue with its associated distribution of colour intensity scores. For example, 90% of all males recorded as chrome orange (16) had a total intensity of 10 or less. Temporal Fluctuations In Colour: Blue Eye (i) Intensity Eye colour changes across the breeding cycle are shown in Fig. 3.5. Wilcoxon matched-pairs signed-ranks analyses corrected for ties were performed to test the significance of these changes (once again, appropriate T values were divided by the number of tests: 8). Mean eye colour intensities during courtship, egg and fry guarding are all significantly higher than scores during nest building (nest/courtship: N=19, T=0, p < 0.005; nest/egg guarding: N=17, T=1.5, p < 0.005; nest/fry guarding: N=18; T=1.5, p < 0.005). Differences between courtship/egg guarding, courtship/fry guarding, courtship peak/fry peak (days 14-16), egg/ fry guarding and days 13/14 are not significant. In contrast to the cyclical behaviour of red body intensity, eye colour intensity plateaus during courtship then fluctuates slightly around that plateau for the remainder of the breeding cycle. 1 H—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 ( — nest — ) (court) (egg guard) ( fry guard ) DAY Figure 3.5: Change in eye colour intensity of Gasterosteus aculeatus males (N=19) across a complete breeding cycle. The breeding cycle is broken into four distinct stages: days 1-5: nest building/maintenance; days 6-8: courtship; days 9-13: egg guarding; days 14-23: fry guarding. Spearman rank correlation analysis reveals that, while the consistency of a male's rank in terms of red extends across the entire breeding cycle (Table 3.4), intensity of eye colour is only a reliable signal of relative colour rank on a brief time scale (Table 3.6). There is a significant, positive association between nest and courtship rank, between courtship and egg guarding, and between egg/fry guarding, other males over an extended time period. STAGE COURT E G G FRY NEST 0.46* 0.33 0.15 COURT - 0.62** 0.45 E G G - - 0.77*** Table 3.6: Spearman rank correlations between eye colour intensity at different stages of the breeding cycle. Probability values are indicated with asterisks: * p < 0.05; ** p < 0.01; ** p < 0.001. Unmarked scores are not significant. (ii) Hue Eye colour hue does not change either within an individual male across the breeding cycle or between males at the same stage. Colour for all individuals was scored as cerulean blue (colour 67: Munsell notation 8.0 B) regardless of corresponding intensity. The Interaction of Red Body and Blue Eye Colour Variability in signal intensity between courting males is greater for red than blue (Fig. 3.6): 14 l 1 2 3 4 5 6 RANGE IN MEAN COLOUR INTENSITY SCORES Fig. 3.6: Histogram depicting the distribution of colour score intensifies for the blue eye (striped bars) and red body (solid bars) components of the nuptial signal i n courting male G. aculeatus. Intensity categories represent the following scores: "red": 1 = 0-9.5; 2 = 9.6-19.5; 3 = 19.6-29.5; 4 = 29.6-39.5; 5 = 39.6-49.5; 6 = 49.6-60.0: "blue": 1 = 0-1; 2 = 1.1-2; 3 = 2.1-3; 4 = 3.1-4; 5 = 4.1-5. Spearman rank correlation analyses of the relationship between red body and blue eye colour scores during the four stages of the breeding cycle indicate that the strongest association between the two variables occurs 55 during courtship (Table 3.7). During this stage a pulse of red sweeps across the entire fish accompanied by a surge in eye colour intensity. This association, although significant, is only moderate, indicating some degree of decoupling between the two colours. The decoupling of red and blue is exaggerated during the parental stages as red follows a strongly cyclical pathway and blue plateaus at courtship levels. Thus, red and blue function as an integrated "colour unit" only in the early stages of breeding. STAGE C O M P A R E D R H O P-VALUE NEST/NEST 0.51 0.05 COURT /COURT 0.54 0.02 E G G / E G G 0.23 NS FRY/FRY 0.35 NS Table 3.7 : Spearman rank correlation values for red body and blue eye colour scores at the same stage of the breeding cycle. Temporal Fluctuations in Colour: Black Body (Melanin) Changes in the intensity (shade) of melanism through time are presented in Fig. 3.7. During the courtship period there is a marked reduction in dorsal melanism. Although not depicted in Fig. 3.7, this trend is exaggerated during the actual courtship sequence. Immediately upon completion of a creeping through or glueing bout, the entire male flushes snowy white. As the male pales, he performs a very intense, lateral zigzag approximately one-quarter to one-third of the way above his nest towards the female. Prior to this striking colour change, glueing and creeping through are usually followed by a zigzag dance all the way up to the female, a combined dance/lunge or simply a lunge. Both the white flush and abrupt zigzag are extremely distinctive. If the female begins to drop, males in this state turn, swim rapidly to the nest and perform (if given a chance by their partners) a dorsal roll, nest showing display. Under normal courtship conditions, the snowy flush is a brief, powerful predictor that the male is now "ready to spawn". Although red nuptial colouration is still visible during this courtship flush, the overall effect, to my eye at least, is one of covering the red with a film of fine, white gauze. Once the female is in the nest and the male has begun quivering, a wave of grey sweeps forward from his caudal fin. This colour change intensifies during and immediately following fertilization. After the courtship period, melanism begins to increase until, during fry guarding, the entire area surrounding the remaining red pigmentation reaches an intense medium to medium dark gray. 0 H—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 (—nest—) (court) (- egg guard -) ( fry guard ) Fig. 3.7: Change in the intensity of dorsal/lateral melanism in G. aculeatus males across the breeding cycle. The breeding cycle is broken into four distinct stages: days 1-5: nest building / maintenance; days 6-8: courtship; days 9-13: egg guarding; days 14-23: fry guarding. Numbers along the y-axis refer to the following hues: 0 = pale grey (Munsell # 86); 1= light grey (# 85); 2= medium grey (# 84); 3= dark grey (# 83). 57 D I S C U S S I O N The three-spined stickleback's nuptial signal is a mosaic of six components (summarized in Table 3.8). Although each of the six components follows a separate pattern of change, when these are superimposed upon the breeding cycle, a temporally fluid but distinct series of nuptial colouration mosaics emerges. BLUE EYE RED BODY BLACK BODY STAGE IN CYCLE INTENSITY HUE INTENSITY DISTRIBUTION INTENSITY DISTRIBUTION NONTERRITORIAL + sky blue 0 0 VARIABLE 1 TERR./NEST BUILD + + 4 ++ 4 COURTING +++ +++ 2 + 3 SPAWN +++ +++ 2 0 1 (white) EGG GUARD +++ ++ 5 ++ 5 FRY GUARD +++ +++ 4 +++ 4 Table 3.8: Relationship between six structural components of nuptial colouration across the breeding cycle. Intensities are represented by crosses; 0= no colour; += low; ++ = medium; +++ = high. Note that intensity and hue are the same for black. Distribution of colour is represented by numbers: 1 = the entire fish; 2 = all ventral and lateral surfaces; 3 = dorsal surface; 4 = red mandibular area, remaining areas black; 5 = red mandibular and mid ventral/lateral areas, remaining areas black. Nesting males in this population of Gasterosteus aculeatus are characterized by (1) a low intensity fluctuation of chrome orange colour predominantly in the mandibular region, (2) gradually increasing, low intensity cerulean blue eye colour, and (3) a light to medium neutral gray dorsal surface .This low level of nest-associated colour in isolated individuals was also observed in a series of preliminary experiments in which 40 males competed for 20 nest sites. Over the course of two trials, the appearance of red breeding colouration lagged behind territory establishment/defense and the completion of nest building in all males (Table 3.9). Red is therefore not a necessary prerequisite for territory acquisition in this population. This result has been reported for other three-spined populations (Wunder,1930; Craig-Bennett, 1931; van Iersel, 1953; McPhail, 1969). The evidence accumulating from a variety of microevolutionary studies thus supports the phylogenetic prediction based upon the macroevolutionary relationships of breeding colour and behaviours: intra-sexual selection does not appear to have played a major role in the evolutionary elaboration of the nuptial signal in the gasterosteids. TERRITORIAL MALES TRIAL #1 TRIAL #2 DAY 1 3 6 8 10 11 12 13 18 19 20 1 2 5 9 10 11 15 20 Z 1 - - - - - - - - - - - - - - - - - - - -2 - - - - - - - - - - - @ - @ @ - - - - -3 4 - @ @ - - - - @ - - - - - - - - @ - - -5 6 - - - - - - - - - - - - - - - @ . . @ . 7 @_---@--_-. - - - - - - - - -8 - - - - - - - - - - - - - - - - - @ - @ 9 @ - - - - - - . 10 - - - - @ - - - @ - - - - - * - - - - -H - - - - - - @ - - @ - * - - * * - - - -1 2 - * - ® - - - - - - ® * - - * * - - - -2^ * * * * _ _ _ * * _ _ * * _ * * Table 3.9: Relationship between territory acquisition and development of nuptial colouration. "@" marks the day when a male was first observed on his territory; asterisks (*) represent the appearance of "red" in a given male. Courting males are characterized by a peak in intensity of both red body and blue eye colours and a decrease in the intensity of the dorsal surface from medium/light to light/pale gray. Within this mosaic, changes in red are by far the most complex aspect of the colour signal. To state that males achieve their maximum intensity during courtship is ambiguous, because colour score could peak in manifold ways. For example, an individual with a total score of 30 could possess a high intensity (10) in three areas or a moderate intensity (5) in six. Data from these experiments support the second alternative: colour surges across all ventral and lateral surfaces of the fish. In addition to these changes in red alone, the correlation between the intensity of red body and blue eye colour is higher at this stage than at any other time in the breeding cycle. The male thus presents his most intense, widely dispersed and cohesive mosaic signal during the brief courtship phase. The strong association between the evolutionary elaboration of colour and courtship in the gasterosteids in general, is thus mirrored in G. aculeatus in particular. Change in colour intensity is also associated with change in hue. As intensity increases, hue shifts from chrome orange towards flame scarlet. All males pass through this series of colour changes across the breeding cycle; in this sense, the association between intensity and hue may be thought of as the "ontogeny" of colour. Changes in both the absolute amount of pigment present and the relative concentrations of different pigments may be responsible for these observed hue shifts. Czeczuga (1980a) outlined the pigment profile of a population of freshwater three-spined sticklebacks in Poland. Within this population, developmental changes in the males' integumentary pigments entailed both a minor alteration in the types of carotenoids present and a major shift in the relative proportions of the remaining pigments. In breeding males, the concentration of ketones (astaxanthin, canthaxanthin) increased, whereas the concentrations of all other pigments decreased. Brush and Reisman's preliminary analysis (1965) of California freshwater populations supports the observation that nuptial colouration results primarily from quantitative rather than qualitative changes in carotenoids. Unfortunately no data are available concerning intermale variability in the types or concentrations of pigments, therefore it is impossible at this stage to determine the cause of the colour changes observed in breeding Salmon River fish. For example, a "scarlet red" male may have both more pigment absolutely and a higher concentration of pigment "a" relative to pigment "b" than his orange counterpart. Crozier (1967) demonstrated that increasing proportions of the carotenoid astaxanthin were associated with increasing the intensity of red colouration in Sebastes species. Astaxanthin is the major nuptial pigment in all populations of G. aculeatus examined to date (Brush and Reisman, 1965; Matsuno and Katsuyama, 1976;Czeczuga, 1980a), so the suggestion that intermale variability in relative pigment concentrations may be reflected in intermale variability in nuptial colouration is not an unreasonable one. Since 60 the effects of selection on the elaboration of the signal will be constrained by the mechanisms underlying signal production, knowledge of these mechanisms is of paramount importance. Detailed biochemical and physiological analyses of pigment structure, sequestration and deposition are required to add depth to the two dimensional portrait of colour presented so far for three-spined sticklebacks. The parental care stages are characterized by an increase in melanism to medium/dark gray levels accompanied by an overall decrease in both the intensity and distribution of red body colour. These changes may occur as a result of the close association between dermal melanophores and the carotenoid containing cells. Although the body of the melanophore lies within the dermal layer, extensions project up and over the epidermal chromatophores (Bagnara and Hadley, 1973). Since expansion of melanin in these extensions covers the chromatophores, changes in colour associated with parental care may be due to masking of red rather than the metabolism of pigment or the destruction of pigment cells. This permits a rapid return to courtship colouration following completion of or interference with one spawning cycle. Van Iersel (1953) and Baggerman (1958) noted this masking of erythrophores by melanocytes in populations of anadromous sticklebacks in Holland. The damping of red, combined with the increase in dorsal/lateral melanism, renders the male more cryptic overall (van Iersel, 1953) and presumably decreases detection by piscivores (i. e., trout, char: Frost, 1954; Hagen and Gilbertson,1972; Moodie,1972; Moodie etal, 1973); while replacement of the white dorsal surface with more sombre shades parries the effectiveness of avian predation (i.e.. herons,diving ducks, gulls : Owen, 1960;Bengston, 1971 ;Whoriskey and Fitzgerald, 1985). As melanism increases from the caudal fin anteriorly, fluctuations occur in the intensity of red in the mandibular regions. A small decrease in colour during egg guarding is followed by a surge in intensity to courtship levels coinciding with the appearance of fry and an increase in aggression. The similarity between the cycling of colour and of aggression over this period is striking, and perhaps indicative of a common physiological pathway (reviewed in Bakker, 1986). The second peak in overall colour is lower than the level achieved during courtship as colour increases in the anterior third of the fish are countered by losses in the posterior and middle regions. Nevertheless, this peak suggests some role for colour during the 61 initial stages of fry guarding and supports the phylogenetic hypothesis that the two have been moderately associated over the course of evolutionary time. Four very vivid and distinct male images emerge from the interplay between red, blue and black in this species. First, the nesting male tends to be a low intensity (dull), flame orange, medium/light gray individual. He is, in fact, barely distinguishable from a nonterritorial male. Second, the courting male tends to be a flamboyant mixture of bright cerulean blue eyes and intense flame scarlet distributed across the entire ventral and lateral surfaces, displayed against a background of light/pale grey. Third, as courtship proceeds, a snowy white flush washes over the male, perhaps signalling to his partner that the transition from predominantly aggressive to predominantly sexual motivation is complete and he is now "ready to spawn". And fourth, the parental male presents a more sombre image as a wave of medium/dark gray sweeps forward from the caudal fin, masking all colours except the bright blue of the eyes and intense flame scarlet in the throat region. This variation in the colour mosaic reliably advertises each male's position along the pathway leading from nest to offspring. S U M M A R Y The three-spined stickleback male's nuptial signal is a complex mosaic of at least three colours: blue eyes, black dorsal/lateral body and red ventral/lateral body. Seven variables - the intensity, hue and distribution of red body colour, the intensity and hue of blue eye colour and the intensity and distribution of black body colour - interact to produce four distinct male colour mosaic signals corresponding with the stage a male has reached in the breeding cycle. Variability in signal intensity between courting males is greatest for the red component of the nuptial signal and both intra- and intermale variability in the overall intensity of red is greatest during courtship. The intensity of courtship red is positively associated with the intensity of red during the parental care stages. The results of this experiment confirm the phylogenetic predictions outlined in chapter 2: the intensity of red nuptial colouration peaks maximally during courtship and secondarily during fry guarding. 62 CHAPTER FOUR THE RELATIONSHIP BETWEEN MALE COLOUR RND MALE BEHHUIOUA INTRODUCTION Communication, in its broadest sense, is "any discriminatory response of an organism" (Diebold, 1968). All biological interactions, from molecular to organismic recognition of "likeness" thus involve some form of communication. Within this very general framework, ethologists have traditionally concentrated on interactions involving the exchange of signals (Tinbergen, 1959,1962). All signals share one important feature: they are all an external manifestation of an encoded message (Smith, 1968,1977; Sebeok,1972; Hailman,1977). Since we introduce our perceptual biases into any study, certain signal types lend themselves more easily to analysis of message/signal relationships than others. For example, levels of aggression (message) are frequently measured via frequency and duration of biting (first level signal). Because the relationship between message and signal exists within a phylogenetic framework, species specific signals (e. g. colour) are difficult to decipher for an observer outside that framework. These signals tend to be interpreted in terms of first level signals: i. e., researchers investigate the association between biting and aggression (Fig. 4.1a) and the association between colour and biting (Fig. 4.1b), and infer from this the relationship between colour and aggression (Fig. 4.1c). I A M A G G R E S S I V E (message) / > \ (a) / (c) \ \ (d) / (b) X * B I T E ^ — • C O L O U R (first level signal) (second level signal) Fig. 4.1: The relationships between a message and its signals are represented in this heuristic diagram. Factors involved in the transformation of the message into different signals are represented by (a) and (d); correlation between signals by (b). The relationship between the second level signal and the message (d) is inferred (c) from measurements of (a) and (b). The macroevolutionary elaboration of male nuptial colouration within the Gasterosteidae is strongly associated with the diversification of courtship behaviours (chapter 3). This corroborates the hypothesis, based upon intra-specific (microevolutionary) studies, that intersexual selection has played a major role in shaping the nuptial signal in Gasterosteus aculeatus (Pelkwijk and Tinbergen, 1937;McPhail,1969;Semler, 1971). Since the focus of such selection is on the male signal, it is important to document the messages potentially being conveyed in nuptial colouration. However, since "message" is an internal property of the sender and thus difficult to measure, this aspect of the communicative interchange is more easily studied at the level of associations between signals. The three-spined stickleback male's nuptial signal is a complex mosaic of at least six components: intensity and distribution of red body colour, intensity and hue of blue eye colour and intensity and distribution of black body colour. Males do not differ significantly in eye colour (hue), the distribution of red or the intensity and distribution of black (chapter 3). Rowland (1984) measured the response of territorial males differing in intensities of red body colour to an intruder and reported a positive association between male colour and male behaviour. This chapter presents the results of a similar investigation using the intensities of blue eye and of red body colour for a population of G. aculeatus geographically and ecologically distinct from that studied by Rowland (1984). MATERIALS AND METHODS Test individuals were chosen from the stock groups collected on April 23/28th (trial #1), May 7th (trial #2) and May 14th (trial #3). Aquaria were drained and cleaned, nesting materials were replaced and light meter readings repeated between trials. Males were placed in test aquaria (details in chapter 1) and allowed five days to establish territory boundaries and build a complete nest. To facilitate the development of peak breeding colouration, gravid females, enclosed within a 250 ml. beaker, were presented for 5 minutes to all males on day 3 of the nesting period. Test presentations of gravid females began at 8:00 hours of day 6. All females were chosen to be approximately 10% larger than the males. The female, enclosed within a 250 ml. beaker, was placed 30 cm. from the male's nest and the male's responses recorded on a Hitachi video camera for five min. starting from his first reaction. Any male that did not respond within 20 minutes was tested again at the end of the experimental period. The intensities of red body and blue eye colour were recorded before and after female presentation. Videotapes were analyzed and the following behaviours were scored: (i) Frequency of bites: the number of bites per presentation is widely used to quantify aggressive behaviour in three-spined sticklebacks (Sevenster, 1961 ;Sevenster-Bol, 1962 ;Peeke et al, 1969;Wootton, 1970,1971; Huntingford, 1976a,b;Rowland, 1982a, 1983,1984;Gaudreault and Fitzgerald, 1985). (ii) Duration of biting: Because of the method of female presentation, intense bites involving one approach and one contact may last for several seconds as the male circles the jar, mouth open and in contact with the glass; attempting to bite. These intense bouts can, by definition, only be scored as one bite. To add depth to the measure of aggression, the total time spent in aggressive interactions was recorded for each 5 minute test period. (iii) Frequency of zigzags: the number of zigzags per presentation is the accepted measure of male sexual motivation in G. aculeatus (van Iersel, 1953; Sevenster, 1961 ;Sevenster-Bol, 1962;Wootton, 1971 ;Huntingford, 1976b 1982; Rowland, 1982b, 1984). (iv) Frequency of bites + zigzags: to the extent that aggression and courtship share similar hormonal (Baggerman, 1968) and neural (Tinbergen, 1951;Sevenster, 1961) pathways, they should be positively associated on a long time scale; a more aggressive male should court more intensely (Rowland, 1982b, 1984). However, on the very short time scale of a 5 minute observation period, number of bites and zigzags can be negatively correlated, implying that some of the physiological controls underlying the two behavioural systems are separate and mutually inhibitory (Sevenster, 1961, 1968; Symons,1965;Wilz,1972;Rowland,1984).This negative correlation will confound studies of the relationships between colour and behaviour over a short time period therefore the frequencies of bites and zigzags were combined to produce one overall measure of male activity level (Rowland, 1984). (v) the 5 second criterion: the duration of a male's orientation towards the observer (eye, body and movement) was recorded. Individuals spending more than 5 seconds of the test period oriented in this fashion were removed from analyses (n=3). Raw colour and behavioural scores are tabulated in Appendix C. RESULTS Red Body Colour Spearman rank correlation analyses demonstrate a significant, positive association between colour score and the four behavioural variables (Table 4.1). Rowland's (1984) data are included in this table for comparison. The similarities in correlation coefficients are striking given the differences in behavioural response rates between the New York and Salmon River populations (aggression: New York: 18.4 bites/ (male- min.) versus Salmon River: 7.7 bites/(male • min.) and courtship: New York: 2.0 zigzags/(male-min.) versus Salmon River: 3.7 zigzags/(male • min.). The Salmon River males were divided into three colour intensity categories determined before the experiment (dull: intensity score of 0-19; medium: 20-39; bright 40-60), and the preceding analyses were repeated. The "bright" (high intensity) category was not large enough for statistical analysis (n=2). Within the two remaining groups, colour score is significantly associated with behavioural intensity in dull males, but shows no association with behaviour in medium coloured males (Table 4.1). Both dull males (Wilcoxon matched-pairs signed-ranks test corrected for ties: N=22; T=0; p< 0.001) and medium males (N=16; T=6.5; p< 0.001) increase their intensity of red following the presentation of a receptive female. However, while a male's colour rank before female presentation (initial colour) is significantly correlated with his rank after female 66 presentation (final colour) in the dull intensity group, the association is not significant in medium males. GROUP COLOUR # DURATION # # BITES + FINAL STAGE BITES BITING ZIGZAGS # ZIGZAGS COLOUR INITIAL 0.42** 0.49** 0.33* 0.49** 0.86** ALL MALES (n=43) FINAL 0.52** 0.58** 0.47** 0.60** ROWLAND 0.50** - 0.33* 0.48** (n=29) DULL MALES (n=25) INITIAL 0.24 0.38* 0.27 0.34* 0.54** FINAL 0.44* 0.54** 0.60** 0.57** MEDIUM MALES (n=16) INITIAL -0.15 -0.07 0.11 0.01 0.20 FINAL 0.07 0.13 0.30 0.14 Table 4.1: Spearman rank correlation coefficients between intensity of red body colour and aggressive/sexual behaviours of male three-spined sticklebacks. Colour was scored before (initial) and after (final) the five minute female presentation period. Rowland's data are included for comparison. Asterisks refer to the following probability values: * p< 0.05; ** p< O.Ol. B l u e E y e C o l o u r Relationships between the intensity of blue eye colour and behaviour mirror those discussed for red body colour. There is a significant, positive association between colour score and the four behavioural variables when the associations between colour and behaviour are examined at the population level. Within the intensity groups originally distinguished by red body colour, blue eye colour score is significantly associated with behavioural intensity in dull males, but shows no association with behaviour in medium intensity males (Table 4.2). 6 7 Both dull males (Wilcoxon test: N=17; T=0; p< 0.001) and medium males (N=12; T=4.5; p< 0.01) increase their eye colour signal in response to the presence of a receptive female. However, unlike the results for red, a male's colour rank before female presentation (initial colour) is significantly correlated with his rank after female presentation (final colour) in both the dull and medium intensity groups. GROUP COLOUR STAGE # BITES DURATION BITING # # BITES + ZIGZAGS # ZIGZAGS FINAL COLOUR ALL MALES (n=43) INITIAL FINAL 0.35* 0.51** 0.41** 0.57** 0.42** 0.50** 0.47** 0.64** 0.76** DULL MALE (n=25) INITIAL FINAL 0.08 0.38* 0.25 0.49** 0.68** 0.77** 0.44* 0.62** 0.69** MEDIUM MALES (n=16) INITIAL FINAL 0.24 0.38 0.16 0.20 -0.35 -0.14 -0.07 0.17 0.43* Table 4.2: Spearman rank correlation coefficients between intensity of blue eye colour and aggressive/sexual behaviours of male three-spined sticklebacks. Colour was scored before (initial) and after (final) the five minute female presentation period. Asterisks refer to probability values: * p< 0.05; ** p< O.Ol. Associations Between Red Body and Blue Eye Colour Analysis at the population level confirms the observation that red and blue are acting as a cohesive "colour unit" during courtship (see chapter 3). There is a significant association between the two colours both prior to and following female presentation (Table 4.3). When this association is analyzed from the perspective of low and medium intensity males, a familiar pattern emerges. Dull males present a moderately cohesive colour "mosaic" signal before and after testing and the association between the two colours is strongest following interactions with the female. In contrast, there is no correlation between body and eye colour intensity in medium males. If 68 anything, the trend is towards a decoupling of the two signals in the presence of a female. GROUP RED BODY BLUE EYE ALL MALES INITIAL 0.72** (n=43) FINAL 0.76** DULL MALES INITIAL 0.34* (n=25) FINAL 0.53** MEDIUM MALES INITIAL 0.40 (n=16) FINAL 0.26 Table 4.3: Spearman rank correlation coefficients within the same test stage between intensity of red body and blue eye colour of male three-spined sticklebacks Colour was scored before (initial) and after (final) female presentation. Asterisks refer to probability values: * p< 0.05; ** p< 0.005. D I S C U S S I O N Rowland (1984) measured the response of territorial males differing in intensities of red to a constant stimulus intruder and reported a positive association between male colour and behaviour. Strikingly similar results were obtained in this study using a Gasterosteus aculeatus population differing in geographical location, ecological/life history parameters (anadromous vs. brackish/marsh) and overall behavioural intensities. Since breeding colour and behaviours are closely associated through macroevolutionary (phylogenetic) time, the similarity between these two populations is not unexpected. It reflects the conservation of an ancestral character association at the population level even though the two populations are ecologically distinct. Rowland (1984) interpreted the significant association between colour and behaviour as evidence for a truth in advertising mechanism (Kodric-Brown and Brown, 1984) operating on colour in G. aculeatus. Analysis of Salmon River sticklebacks suggests that this interpretation may require further research. In this population, females will only receive reliable information about male "vigour" for dull, and possibly brightly coloured males. Red and blue are positively associated in dull males; therefore, a reliable and redundant message that "I am not a behaviourally vigorous male" is projected both before and during an interaction. Redundancy counteracts environmental noise which might otherwise interfere with signal transmission and increases the effectiveness of information transfer (Shannon and Weaver, 1949; Hailman,1977). Since a female encountering a dull male receives information about his future behavioural intensity from at least two colour signals prior to a direct encounter, she may avoid wasting energy and time interacting with a potentially unresponsive male. Additionally, the low levels of aggression demonstrated by dull males may greatly reduce their chances of acquiring and maintaining a territory (van den Assem,1967;Black, 1971; Sargent and Gebler, 1980) or attracting females (Ward and Fitzgerald, 1987). As predicted by a truth in advertising scenario, all these factors would combine to select against low behavioural/ colour intensity males. Similar reasoning could be applied to bright males. Evidence from Rowland's (1984) experiments suggests that bright males are also reliably advertising their behavioural vigour. If this is true, then female three-spined sticklebacks are potentially presented with enough information in the nuptial signal to discriminate between males at opposite ends of the "behavioural vigour" spectrum without the need for a prolonged encounter with each male. Medium males are more problematical. The intensities of red body and blue eye colour are not significantly correlated with any of the measured behavioural variables or with each other. Of the two colours, red, which is both more labile and variable among males than blue (chapter 3), shows the weakest associations. This pattern of predictability at the high and low ends of the colour spectrum and unpredictability in the middle is mirrored in the relationship between territory size and reproductive success. Van den Assem (1967) reported that males with large territories were extremely successful, whereas males with small territories were relatively unsuccessful (but still better than nonterritorial males). Within the largest subset of the breeding population, males holding intermediate sized territories, territory-size and reproductive success were not associated. More research is required to document whether this is a common pattern and, if so, the causal factors involved. Although medium males transmit ambiguous information in their colour signal about both future colour and behavioural intensity relative to other medium males, they are truthfully advertising their behavioural responsiveness relative to dull males in the population. Medium males are both more intensely coloured and are more behaviourally vigorous (bites + zigzags: T = -2.29; p < 0.03) than their dull counterparts. For a medium male, then, a truth in advertising mechanism is a weak force relative to other medium males (within group comparison) but a potentially strong force relative to dull males (between group comparison). If the outcome of interactions is predictable for males at the top and bottom of the colour hierarchy in G. aculeatus, the disorganizing effects of an individual's history (experience) may not be very strong. Medium males, on the other hand, have passed some theoretical threshold for the appearance of moderate colour and behavioural intensities but have not reached the level of the most vigorous males. Outcomes of interactions may be determined, in part, by chance (e. g., which males they encounter during territory establishment or who their closest neighbours are). This, in turn, introduces an element of stochasticity or unpredictability into the system which counters the strength of a truth in advertising mechanism promoting the evolution of a reliable colour signal in medium males. S U M M A R Y The intensities of red body and blue eye colour in dull (and possibly bright) males reliably signals their behavioural vigour. Medium intensity males signal that they are more vigorous than their dull conspecifics; however, there is no association between colour and behavioural intensity within this group. A truth in advertising mechanism is thus potentially very powerful at either end of the colour intensity spectrum but is less effective on the central section of the population. Medium males may be more strongly influenced by stochastic factors such as previous experiences or colour relative to neighbours. The increased importance of personal history introduces a source of disorganization into the mating system that may oppose the directional force of truth in advertising and thereby increase the ambiguity of the male colour signal. 71 CHAPTER FIUE THE R E L A T I O N S H I P B E T W E E N M A L E COLOUR RND F E M R L E BEHRUIOUR INTRODUCTION Researchers investigating the effects of intersexual selection on the elaboration of male nuptial colouration in Gasterosteus aculeatus have traditionally offered females a choice of "red" and "nonred" males (Pelkwijk and Tinbergen, 1937;McPhail,1969;Semler, 1971). Pelkwijk and Tinbergen (1937) presented females with red and nonred models, and demonstrated an absolute preference, measured in terms of the "follow-male" response, for red. McPhail (1969) extended these findings in a study utilizing monomorphic red and black populations of freshwater sticklebacks. When given a choice, females from allopatric red and black populations demonstrated a significant preference, defined by the "head-up" response, for red, while black females from areas of sympatry between the two forms showed no preference. McPhail interpreted this result as the decay of preference for red in response to selection against interpopulation hybridization. Semler (1971) documented the existence of several nuptial morphs ranging from red to black in three-spined sticklebacks collected from Wapato Lake, Washington. When females from this region were allowed to spawn with their choice of red or nonred males the response was unequivocally red. Based upon Semler's observation that breeding red and nonred males were spatially segregated, further studies are required to determine whether this variability in colour represents one polymorphic population or several sympatric, monomorphic populations. The above studies have been cited as evidence for the effect of intersexual selection on the evolution of red male nuptial colouration in three-spined sticklebacks (O'Donald, 1980, 1983; Kodric-Brown, 1983; Kirkpatrick, 1987). However, the investigations are confounded by the confused taxonomic status of the G. aculeatus species complex which encompasses more than one species (Penczak,1966;Hagen, 1967;McPhail, 1969,1984; Miller and Hubbs, 1969; Hagen and McPhail, 1970; Wootton, 1976). If different coloured populations represent different species of Gasterosteus, then experiments which offer females a choice between males from such populations are investigating the first level of mate choice in any species: mate recognition. Within red populations, all sexually mature G. aculeatus males have the capacity to develop nuptial colouration; however, only the breeding subset of the population - the territorial males - do so. A female will only spawn with a territorial male, all of whom are red, so a response to "red" versus "nonred" enables her to recognize an appropriately coloured individual as a territorial male of her own species. It does not allow her to discriminate among all such potential mates. Experiments to date thus provide support for the existence of a mate  recognition, but not a mate discrimination, response in the threespine. If females are differentiating among potential mates on the basis of variability in the intensity of red, this aspect of male colour and female response will have played a vital role in shaping the structure of the nuptial signal during the evolutionary history of the species. In this chapter I will investigate two questions: (1) when given a choice of two competing males, do female three-spined sticklebacks prefer the brighter red male and (2) based on the outcome of the choice trials, are there any behavioural or colour differences between winning and losing males which can be detected in pre-choice tests? MATERIALS AND METHODS Males, identified by the presence of serrated dorsal spines, were placed in test aquaria and allowed five days to establish territory boundaries and build a complete nest. Individuals on either side of a divider were matched within 1 mm. for standard length. To facilitate the development of peak breeding colouration, gravid females, enclosed within a 250 ml. beaker, were presented for 5 minutes to all males on day 3 of the nesting period. Ovulated test females were presented in a water-filled 250 ml. glass jar capped with an inverted glass petri dish. A thread attached to the top of the petri dish via a drop of clear silicon enabled the experimenter to remove the lid while filming the interaction. The following series of three experiments, culminating in the female's choice of a spawning partner, were started at 8:00 hours of day 6 (Fig. 5.1): 73 5 MIN. 30 MIN. 3 HOURS 6 MIN. < 30 MIN. SOLITARY MALE SOLITARY MALE TERRITORY BORDERS 2 TERRITORIAL FREE "REST PERIOD" ESTABLISHED MALES MALES CAPTIVE FEMALE NO FEMALE NO FEMALE CAPTIVE FEMALE FREE FEMALE (a) (b) (c) (d) (e) Table 5.1: Temporal sequence of experimental tests. (1) Solitary male/captive female (Table 5.1a) : All females were chosen to be approximately 10% larger than the test male pairs. The enclosed female was placed 30 cm. from the male's nest and the male's responses recorded on a Hitachi video camera for five minutes starting from his first reaction. Colour was recorded before (initial) and after (final) female presentation. Males were allowed a 30 minute rest period then the divider was removed (Table 5.1b). Although territory boundaries were usually established within a two hour period following divider removal, fights between presumptive owners were intense and often prolonged, leaving both combatants exhausted and unresponsive (Semler, 1971 ;pers. obs.). The territorial phase was therefore extended by one hour to provide recovery time (Table 5.1c). (2) Two interacting males/captive female (Table 5.Id): Colour was scored at the end of the three hour territorial period. Following scoring, a novel, captive, ovulated female was placed in the middle of the aquarium. This corresponded with either the territorial boundary (two competing males) or the middle of the territory (only one territorial male). Interactions were videotaped for a 6 minute period: 3 minutes per male. (3) Two interacting males/free female (female "choice") (Table 5.1e): Colour was rescored at the end of the 6 minute pre-choice period. This procedure, by design, startled the males slightly, driving them back to their nest area. The female was released when both males appeared over their nests. "Choice" was defined as the completion of spawning and interactions were filmed to that point. Any triads which had not spawned within 30 minutes were removed from later analysis. Scoring Behaviour Videotapes from the three experimental stages were analyzed and the following behaviours scored (raw colour and behaviour scores are tabulated in Appendix D): (1) Solitary male/captive female: the number of bites, duration of biting and the number of zigzags. Individuals spending more than 5 seconds of the test period oriented towards the observer were removed from analysis (n=3). (2) Two interacting males and captive female (a) Interactions between focal male and captive female : the number of bites, number of zigzags and the frequencies and durations of the nest oriented behaviours dig, fan, glue and creeping through (see van Iersel,1953; Sevenster, 1961 for behavioural descriptions). Lengthy aggressive bouts directed towards the female were never observed in this experiment, therefore duration of biting was not recorded. fb) Interactions between males (i) number of bites directed towards a rival male (ii) number of chases initiated by each male (iii) number of rebounds: rebounds generally occur along or close to the territorial boundary, when the focal male rapidly approaches his rival then, by reversing the direction of pectoral fin movement, stops sharply and backs up before contact is made. (iv) number of head down threats: a threatening male assumes a head down position at approximately a 60 - 90 degree angle with the substrate, broadside to his opponent. The pelvic spine closest to the rival is fully extended; however, there is considerable variability in the degree of dorsal spine extension (see Symons, 1965). (v) number of circle fights: (van Iersel,1953;Wootton, 1976). Males, aligned head to tail, circle rapidly together, attempting to bite each other's caudal area. The pelvic spine nearest the opponent - on the inside of the circle - is fully extended, whereas the degree of dorsal spine erection is variable (see Pelkwijk and Tinbergen, 1937; van Iersel, 1953). (vi) number of intrusions: all entries by one male into his rival's territory. (c) Female's response to the focal male : Every female assumed a head-up posture in response to the presence of the territorial males, however, the orientation of this posture varied throughout the 6 minute period. To assess whether a female was displaying a preferential response to one male, the durations spent oriented to each male were recorded. (3) Two interacting males and free female : female choice It is often difficult to distinguish male/male from male/female interactions during female choice experiments (Halliday,1983). One way to control male behaviour is to present captive males to a female and measure her response in the absence of male-male interactions. This approach is useful in fish species where visual stimulation by the male is sufficient to induce spawning behaviour in the female. However, courtship in Gasterosteus aculeatus is composed of a spatially and temporally complex interchange of signals. From the female's perspective, static visual (zig-zagging, circling), active visual (leading) and contact (quivering) stimulation are generally required to induce spawning. Presentation of confined males will therefore yield, at best, data concerning the amount of time spent within a specified distance of each male; a measurement of association. Without a detailed study of the relationship between "association" and mating preference, it is misleading to extrapolate from the strength of the former to the selective effect of the latter. Since "choice" can only be examined within the G. aculeatus system by allowing the female access to freely interacting males, how can the effects of male-male interactions be removed from the analysis? I approached this problem by attempting to discern a priori clusters of males based on a suite of behavioural traits they exhibited during the choice tests. I then asked if there were differences in colour and behaviour between winning and losing males within and among the groups. Male pairs were clustered using cladistic analysis. This method of analysis, developed originally for use in reconstructing phylogenetic relationships, can also be seen as the clustering method that both provides the most parsimonious representation of the data set (Farris,1979) and explicitly depicts the distribution of characters (variables) among the clusters. Lambshead and Paterson (1986) examined the feasibility of using cladistic analysis to cluster ecological data and reported that the results compared favourably with standard ecological techniques. They argued that such a clustering method can be used to uncover patterns in population-based data because it does not incorporate any assumptions about the processes responsible for producing those patterns. Similar arguments have been applied to the use of cladistics as a clustering method in studies of biogeography (Wiley, 1988a,b;Mayden, 1988) and coevolution (Brooks, 1988). Phylogenetic analysis has already uncovered a strong association between breeding colour and courtship in the gasterosteids. It is important to re-emphasize, however, that the use of cladistic methods to cluster data from individuals is not a test of the associations between those data, only a description of the patterns of the associations. Once these patterns have been delineated, a variety of data can then be examined using standard statistical analyses. Videotapes of each choice test were analyzed and test pairs clustered based upon 16 behavioural responses (characters). These characters are generally not used in a behavioural analysis because, since they are not measurements in the sense that "numbers of bites" is a measurement, they are difficult to analyze statistically. Yet these behaviours are an important component of the entire suite of interactions leading to spawning. These characters thus provide a useful foundation for the clustering of male pairs which can be used, in turn, to examine more traditional data such as the number of bites or zigzags. In this way, arguments of circularity are also avoided because the data we are interested in examining statistically are not incorporated into the clustering analysis. Evolutionary models of female "choice" generally begin with a randomly mating population. Changes in female preference/male character are then hypothesized to sweep through the population due to the fitness advantage conferred upon chosen males and, possibly, choosey females (reviewed by Kirkpatrick,1987). The "no choice" situation was therefore chosen as the "plesiomorphic" (general) state for characters in this analysis. Lambshead and Paterson (1986), applying analogous biogeographic arguments to their data, concluded that their outgroup consisted of a theoretical station containing no species (i. e., absence of a character was plesiomorphic). 77 CHARACTER DESCRIPTION 1. BOTH MALES VISIBLE WHEN FEMALE EMERGES: 0 = NO; 1 = YES. 2. BOTH MALES WITHIN 30 CM. OF FEMALE WHEN SHE EMERGES: 0 = NO; 1 = YES. 3. BOTH MALES MOVE TOWARDS FEMALE PRIOR TO "CHOICE": 0 = NO; 1 = YES. 4. BEHAVIOUR OF "CHOSEN" MALE IMMEDIATELY PRIOR TO "CHOICE": 0 = PASSIVE; 1 = ACTIVE TOWARDS FEMALE; 2 = ACTIVE TOWARDS MALE. 5. BEHAVIOUR OF RIVAL MALE IMMEDIATELY PRIOR TO "CHOICE": 0 = PASSIVE; 1 =ACTIVE TOWARDS FEMALE; 2 = ACTIVE TOWARDS MALE. 6. FEMALE'S INITIAL RESPONSE UPON EMERGENCE FROM THE FRM: 0 = TURNS IMMEDIATELY TO ONE MALE; 1 = HOVER. 7. FEMALE ULTIMATELY SPAWNS WITH FIRST MALE TO APPROACH HER: 0 = YES; 1 = NO. 8. FEMALE ULTIMATELY SPAWNS WITH INITIAL "CHOICE": 0 = YES; 1 = NO. 9. NUMBER OF TIMES FEMALE LEAVES HER PARTNER'S TERRITORY BUT DOES NOT SWITCH PARTNERS : 0 = NONE; 1 = ONE; 2 = MORE THAN ONE. 10. NUMBER OF TIMES FEMALE SWITCHES PARTNERS: 0 = NONE; 1 = ONE; 2 = MORE THAN ONE. 11. NUMBER OF TIMES RIVAL MALE INTRUDES ON TERRITORY BUT DOES NOT APPROACH FEMALE: 0 = NONE; 1 = ONE; 2 = MORE THAN ONE. 12. NUMBER OF TIMES RIVAL INTRUDES ON TERRITORY AND APPROACHES FEMALE DIRECTLY: 0 = NONE; 1 = ONE; 2 = MORE THAN ONE. 13. NUMBER OF TIMES FEMALE TURNS TOWARDS RIVAL: 0 = NONE; 1 = ONE; 2 = MORE THAN ONE. 14. NUMBER OF TIMES FEMALE REJECTS ULTIMATE MATE AT NEST: 0 = NONE; 1 = ONE; 2 = MORE THAN ONE. 15. NUMBER OF TIMES FEMALE REJECTS ULTIMATE RIVAL AT NEST: 0 = NONE; 1 = ONE; 2 = MORE THAN ONE. 16. FEMALE SPAWNS WITH THE BRIGHTEST MALE: 0 = YES; 1 = NO. 78 RESULTS Cladistic Analysis of Choice Tests One cladogram (Fig. 5.1), with a consistency index (Kluge and Farris, 1969) of 49%, is produced by Camin - Sokal (1965) parsimony analysis of the data presented in the data matrix (Appendix E). The low consistency index obtained in this analysis indicates a high degree of behavioural variability in the data; however, as noted, only one tree was produced. This occurs because of the configuration of the data set and the constraints placed on the analysis by the use of the Camin-Sokal algorithm. This algorithm allows parallelisms but not reversals in homoplasious characters. This is important because allowing reversals to the plesiomorphic condition would involve clustering groups based on behaviours which did not happen. This algorithm thus incorporates the greatest amount of actual behavioural data onto the cladogram. The use of spawning as the criterion for female choice does not adequately address the influence of male/male interactions on female responses. A stringent definition of what actually constitutes "choice" in this experiment is required. Cladistic clustering of the behavioural data provides one method for separating "no choice" from "potential choice" trials by highlighting various traits on the tree and examining the relationships amongst them. This, in turn, allows us to frame our definition of "female choice" within the context of the behavioural system of the interacting organisms. (i) Examination of male behavioural characters: From the female's perspective, the opportunity for choice occurs when both males are visible to her (character #1). Using "both males visible" as the minimum criterion for choice eliminates pairs #16 and 18 from the analysis. In fact, these trials almost represent the "no choice" situation. All males in the remaining pairs were at least visible when the female emerged. Characters #11 and 12 provide a measure of the persistence of a rival male in terms of territorial intrusions and attempts to attract the female's attention. Highlighting these characters on the tree reveals two groups of males, based upon the intensity of the losing male's behaviour (Fig. 5.2): (1) inactive losers - only one territorial male remained; displaced male either was not visible or hovered Fig. 5.1: Tree produced by cladistic analysis of behavioural characters recorded during female choice trials. Numbers at the end of branches refer to particular pairs of test males. "X" represents the "plesiomorphic" (no choice) trial. Numbers accompanying slash marks on the tree refer to characters and are coded in the following manner: the number preceding the parenthesis refers to a particular character; the number enclosed within parentheses refers to the state of that character. See data matrix and methods section for a description of character states. Character states represent a series of changes in a character beginning with the plesiomorphic ("no choice") state. The first occurrence of a character condition different from "no choice" is indicated by the number "1" in parentheses. For example, for character #11 (the number of times a rival male intrudes on a courting male's territory): 11(0) = the plesiomorphic state. "0" states, as a rule are not depicted on the tree. 11(1) = rival male intrudes once during the choice test. Character state 11(1) is marked with an @ on the tree. All male pairs prior to this point (#2,4,8,10,11,12,16,17 and 18) display the condition 11(0): the rival male does not intrude during the course of the choice test. All male pairs past this point on the tree display at least one rival male intrusion during courtship (#14,1,13,9,15 and 3). The remaining six pairs (# 6,7,19,20,21 and 5) display the second state of this character, 11(2): the rival male intruded more than once on the courting male's territory during the choice period. Asterisks above pairs # 6 and 5 draw attention to the only trials in which the female spawned with the duller male: character state 16(1). 80 81 on the ground during the entire spawning sequence and (2) active losers -both males held approximately equal sized territories; losing male attempted to attract the female from the rival. This group offered females a choice based on two competing males and thus represents the broadest acceptable criterion for female "choice". Fig. 5.2: Distribution of characters associated with the activity of the losing male. Character 11 = Number of times rival intrudes on winner's territory but does not approach female; character 12 = Number of times rival intrudes on winner's territory and approaches female directly. Numbers i n parentheses refer to the coding of each character state: (1) = one time; (2) = more than one time. Pair numbers for the inactive male group are not included on this tree. (ii) Examination of female behavioural characters: Females did not turn immediately to one male upon emergence in the majority of trials within the active losers group (character #6: see Fig. 5.1); they tended to swim up out of the jar, then pause briefly. Given this, perhaps "female choice" is nothing more than "female moves towards the first male who approaches her". One of the characters used in this analysis was "first male to approach female is eventually chosen" (#7). When the distribution of this character is examined on the tree a third group of males - the fighting losers - appears within the active losers, delineated by the fact that the female did not spawn with the first male to approach her (Fig. 5.3). Fig. 5.3: Distribution of characters associated with female activity. Character 7 (1) = female does not spawn with first male to approach her. The remaining two characters are involved with female movement: character 9 = number of times a female leaves her partner's territory but does not switch partners; character 10 = number of times a female switches partners. Numbers in parentheses after traits #9 and #10 refer to the coding of each character state: (1) = one time; (2) = more than one time. Although pair #14 shares this character with the rest of the fighting losers, it is not clustered closely with them because the winning male did not approach or interact with the female until after his rival had begun courting. This, in fact, is the strongest choice test present. The female responded initially to the first male to approach her (male #11). After a brief follow, she paused and remained in the head-up position close to the territory boundary. Male #12 left his nest, approached the female and was driven off by #11. The female continued holding head-up, dropped down to #11 once more when he approached then gradually began drifting towards the territory boundary. Both males approached at the same time; just before contact by one or the other was imminent, she turned, dropped to male #12 and remained with him for the duration of the spawning bout despite attempts by # 11 to attract her back to his territory. Because of the existence of both female movement and loser persistence, trial #14 is included within the group of "fighting losers". Active losers are defined by the high degree of male activity during the choice interactions. Highlighting female activity (characters # 9 and 10) on the tree indicates that female movement is strongly associated with the "fighting loser" pairs (Fig. 5.3). Females left the territory of courting males 18 times within the active group; of these 18 movements, 15 occur within pairs of fighting losers. Increasing the stringency of the choice criterion reveals a hierarchical arrangement of trials within this study (Fig. 5.4). Level one is outcome biased; all trials are included based solely upon the criterion of "female spawns", ignoring behavioural interactions between female release and spawning. Level two requires the female's exposure to two competing males (inactive versus active group). Level three, representing the most stringent level of choice, requires the active participation of both competing males in the choice process and a demonstration that the female is not passively moving to the first male to approach her (fighting group). CHOICE FILTER ALL INDIVIDUALS 1 (n = 21 PAIRS) i 1 1 INACTIVE LOSERS ACTIVE LOSERS 2 (n = 9 PAIRS) (n = 12 PAIRS) 4 - 1 1 ?? LOSERS FIGHTING LOSERS 3 (n = A PAIRS) (n = 8 PAIRS) Fig. 5.4: Hierarchical structuring of groups of males within a test population of G. aculeatus. Groups determined by phylogenetic clustering of test male pairs based upon 16 behaviours associated with female choice trials. Each choice filter represents an increasingly stringent definition of what actually constitutes a choice trial in this experiment. Behavioural Differences Between Winners and Losers 84 (1) Initial presentation: solitary male/captive female There is no significant difference between future winning and losing members of a pair in the number of bites, total duration of aggression, number of zigzags and total number of bites + zigzags directed towards the captive, gravid female (Table 5.2). When behavioural differences are examined within the three groups of males delineated by cladistic analysis, the population level result is mirrored within the active and fighting groups: there is no difference between the behaviour of winners and losers towards a captive female. Inactive losers, on the other hand, are significantly less aggressive (# of bites, duration of aggression) than their winning partners, but demonstrate no difference in the frequency of zigzagging. ALL INACTIVE ACTIVE FIGHTING BEHAVIOUR PAIRS LOSERS LOSERS LOSERS (n=19)* (n=9) (n=10)" (n=8) # BITES NS 0.03 NS NS (0.21) (0.12) (0.20) DURATION NS 0.04 NS NS BITING (0.28) (0.10) (0.22) # ZIGZAGS NS NS NS NS (0.46) (0.30) (0.32) (0.39) # BITES + NS 0.005 NS NS # ZIGZAGS (0.25) (0.07) (0.15) Table 5.2: Wilcoxon analyses of differences between future winning and losing member of competing male pairs based upon behavioural interactions with a captive female prior to barrier removal. NS= not significant (significance values are included in parentheses). * = missing data from pairs # 1 and 7. (2) Second presentation: interacting males/captive female (a) interactions between the focal male and the female: There are no significant differences between winners and losers in either female-oriented (number of bites, zigzags and bites + zigzags) or nest - oriented behaviours (total number of dig + fan+ glue + creeping through bouts) within the active and fighting groups. 85 (b) interactions between males: The number of bites directed towards a rival, elastic rebounds, head down threats and intrusions onto a rival's territory are not significantly different between winners and losers in either the active or fighting group. If any dominance differences exist between territory holders, they are too subtle to detect at this level of investigation. The frequencies of chase and circle fight are too low for analysis. (c) captive female's response to competing males : There was no significant difference in female orientation towards winning/losing males in the fighting losers group over any time interval. Within the active losers group, females spent significantly more time oriented towards the future winning male in the second three minutes, but not the first three minutes, of presentation (Table 5.3). Females holding in a head-up position initially turned towards the closest behaviourally active male. Since male approaches to the female were generally staggered (male "a" interacted with the female while "b" was engaged in nest - maintenance activities and vice versa), a female was provided with the opportunity to view both males. As time progressed the female began to demonstrate a preferential response to her future spawning partner. Instead of turning towards the closest male, the female maintained her orientation while the "preferred" male was at his nest, ignoring all but the most aggressive approaches by the rival. TIME ACTIVE FIGHTING LOSERS LOSERS First 3 minutes NS NS (0.06) (0.10) S e c o n d 3 minutes 0.05 NS (0.06) Total per iod 0.05 NS (0.07) Table 5.3: Wilcoxon analyses of the time a captive female spent oriented towards the future winning male of a competing male pair. Differences between males are significant at the given probability values. Colour Differences Between Winners and Losers 86 (1) Comparisons between winning and losing males within the same experimental stage: The results of Wilcoxon matched-pairs signed-ranks analyses of differences in the intensity of red nuptial colouration of winners and losers are unequivocal: at all levels of investigation, from the population to the most stringent criterion of choice, the fighting group, winners are more intensely coloured than losers at the time of female release (Table 5.4). In the tests preceding choice, this difference is maintained only within the inactive losers group (mirroring the behavioural results presented in Table 5.2). STAGE IN EXPERIMENTAL SEQUENCE ALL PAIRS (n=21) INACTIVE LOSERS (n=9) ACTIVE LOSERS (n=12) FIGHTING LOSERS (n=8) Solitary male/capt ive female: initial colour NS (0.08) 0.03 NS (0.50) NS (0.34) Solitary male/capt ive female: final colour NS (0.13) 0.03 NS (0.38) NS (0.36) Interacting males/no female (3 hours after divider removed) 0.005 0.005 NS (0.22) NS (0.22) Interacting males/just prior to female release (COLOUR AT CHOICE) 0.0002 0.005 0.01 0.02 Table 5.4: Wilcoxon analyses of differences in intensity of nuptial red between winners and losers at various stages in the experimental sequence. A "not significantly different" result from a Wilcoxon test allows two interpretations: the variable under investigation is (i) identical (or nearly so) between samples (winners and losers are the same colour) or (ii) unpredictable between samples (some winners are brighter, some the same and/or some duller than losers). In the latter case, opposing results on a pair by pair basis negates any overall trend towards an association between colour and future spawning success. To investigate this, the losing male's colour score was subtracted from the score of his winning opponent and the 87 resultant differences plotted for each pair over the three experimental stages (Fig. 5.5): -40 "H—I—i I I—i—i—i—i—i i i i—i—i—i—I—i—i—i—r 1 2 3 4 5 6 7 8 9 IC1112131415IC1718192C21 TRIALS -20 "h—i—i i i—i—i—i—i i i i—i—i—i—i—i—i—i—i—r 1 2 3 4 5 6 7 8 9 IC 1112131415161718192C21 TRIALS -20 -h—i—i i i—i—i—i—i—i i i—i—i—i—i—i—i—i—i—r 1 2 3 4 5 6 7 8 9 1C11121314151617 18 192C 21 TRIALS Fig. 5.5: Differences in red body colour scores between winners and losers (winners - losers) plotted for each pair of males. Positive values indicate that the winning male was brighter than the losing male ; negative values indicate that the loser was brighter than the winner. Three experimental stages are plotted: (a) = solitary male/captive female: final colour; (b) = no female/ interacting males 3 hours after divider removal; (c) = captive female/interacting males: colour immediately prior to female release (colour at choice). Arrows highlight trial #9; all trials up to and including #9 represent the inactive losers group. Trials #10-21 are the active losers and, within this, trials #14-21 represent the fighting loser group. Males from the inactive group are included for comparison. Within the inactive losers, winning males are brighter than losers at all stages, with one exception. The colour mismatch (represented by the height of the bar) is greatest between these males. Within the active and fighting groups, there is initially a large degree of unpredictability in the differences between winners and losers (Fig. 5.5a). By the end of the male/male interaction period following divider removal, differences have shifted in favour of winners (Fig. 5.5b). At choice, all but two of the winning males are brighter than their counterparts (Fig. 5.5c). (ii) Comparisons within groups of winning and losing males between experimental stages: Isolated males, whether "future" winners or losers, increased their colour in response to the presence of a captive, gravid female (initial colour versus final colour: Fig. 5.6). Of 21 winners, 19 increased their colour and 2 remained constant; within the losers, 20 increased and 1 remained constant. There was no significant change in the colour response to the presence of another territorial male (final colour versus divider colour) within the winners and losers of either the active or fighting groups; however, within the inactive group colour scores of both losers and winners decreased following three hours of interaction. The "boost your colour" response documented in the first female presentation was repeated in the second presentation even though the males were freely interacting with one another (divider colour versus choice colour). This colour increase is significant for all groups of winning and losing individuals (at p<0.05). When a male's colour score following the first (solitary) and second (competing) captive female presentation were compared (final colour versus choice colour), losing males in the active and fighting groups demonstrated no significant change while their winning counterparts showed a significant increase in the intensity of their colour signal. In the inactive group, neither losers nor winners increased their colour signal past the level achieved during solitary female presentation (Fig. 5.6). 89 40 n (a) (b) Q U J CtL \^ 1X1 Ca o or 9 o o Q L U CL ca O O co ca O INACTIVE ACTIVE FIGHTING NS *** illl.iil.ril I F D C . I F D C . I F D C STAGE IN EXPERIMENTAL SEQUENCE (WINNERS) 25 • 20 ' 15 ' 10 • 5' 0 INACTIVE ACTIVE FIGHTING NS NS i F NS i C T NS T T T NS I D C . I F D C . I F D C STAGE IN EXPERIMENTAL SEQUENCE (LOSERS) Fig. 5.6: Changes In male colour through time in response to different behavioural interactions within the three groups. I = colour before first captive female/solitary male interaction; F = colour after interaction; D = colour in response to male/male interactions 3 hour after divider removal; C = colour after 6 minutes of female presentation, immediately prior to female release and choice. Wilcoxon analyses of differences in colour score for both (a) winners and (b) losers were performed within a group between adjacent stages. The results of two comparisons are represented here: (i) colour score after three hours of male interactions following divider removal (final colour versus divider colour) and (ii) colour score after first female presentation versus colour after second female presentation (final colour versus choice colour). Asterisks indicate the following probability values: * p< 0.05; ** p< 0.01; *** p < 0.005. (iii) Captive female orientation based upon male colour: The demonstration that free females are preferentially responding to the brighter male of a competing pair is strengthened by re-examination of the captive female's orientation during the pre-choice period. Initial analysis was based upon the division of males into winners and losers (Table 5.3). When orientation time is re-analyzed on the basis of male colour, females spent more time in a head-up posture oriented towards the brighter of the two males in both the active and fighting groups (Table 5.5). TIME ACTIVE FIGHTING LOSERS LOSERS First 3 minutes 0.02 0.02 S e c ond 3 minutes 0.002 0.005 Total per iod 0.002 0.01 Table 5.5: Wilcoxon analyses of the time a captive female spent oriented towards the brighter male of a competing male pair. Differences between males are significant at the given probability values. D I S C U S S I O N The female's ability to discriminate between potential mates and non-mates (mate recognition) has been documented for a variety of Gasterosteus aculeatus populations (Pelkwijk and Tinbergen, 1937;McPhail,1969;Semler, 1971). The results of this study provide evidence for another level of female mate choice in this population: discrimination among potential mates based on differences in the intensity of red nuptial colouration. This preference was strongest when the female was prevented from interacting with the males. However, since courtship in this species contains both agonistic and sexual components (Wilz, 1972), the captive female's head-up behaviour may be a fear-based submissive reaction to a stronger threat stimulus projected by the brighter male rather than a sexual response to a potential mate. The head-up display can be construed to convey two different messages depending upon the associated movement vector. Unreceptive females do not move with (track) the male, they simply hold in the head-up posture. Receptive females track the courting male; therefore, no movement (static) 91 = appeasement; movement (tracking) = appeasement + receptivity. Since all test females displayed tracking behaviour it appears that they are responding to the brighter male in a manner more strongly indicative of a sexual rather than a fear based motivation. Once the female was released, the response to the brighter male was weaker than that documented during the pre-choice period in 3 of the 21 trials. It is possible that some aspect of a duller red male's behaviour may have attracted the female during the preliminary stages of the male-female interaction in these cases. As courtship proceeded, however, two of these initially successful males lost the female to their more intensely coloured rival; either through an inability to hold the female's attention (pair #14), or through an inability to prevent the other male from interfering during courtship (pair #20). Only one female was observed to switch from a brighter to a duller coloured male (pair #6) during the latter stages of courtship. Persistent intrusions by the "unchosen" male and interference with his rival's courtship attempts may have been responsible for this partner switch. As in the initial stages of attraction, once courtship has started it is possible, albeit unlikely, for a duller red male to acquire a mate based upon some aspect of his behaviour which attracts a female away from a courting rival. Overall, then, although females respond strongly to more intensely coloured males in this G. aculeatus population, male/male and male/female interactions play a role in modulating the cumulative strength of this response at the population level. Given the brief breeding season and the advantage enjoyed by the most brightly coloured males in mate attraction, a high level of territorial intrusion is one way for an individual to attract a mate if he is surrounded by more intensely coloured neighbours. Behavioural interactions during courtship enhance the potential for female choice based on the intensity of the male's nuptial red signal. Direct interactions between male and female are primarily composed of four male behaviours - "lunge", "nudge/bite", "zigzag" and "circle" - and two female responses - "head-up/stationary" and "head-up/move with (track) male". Lunge and nudge/bite involve a direct approach towards the female which, because of the stickleback's fusiform shape, highlights the mandibular region of the male. The aggressive component of courtship, therefore, exposes the female to only a portion of the signal. In contrast to this, the female is briefly exposed to varying head-on, dorsal and lateral views of her partner throughout the zigzag dance. During circling, she is presented with a prolonged display of the male's lateral surface as a result of male orientation and the receptive female's propensity to track the male through the circle. The sexual component of courtship, therefore, introduces the potential for female exposure to the entire signal. The appearance of male sexual behaviour is associated with a change in the structure of the nuptial signal. A courting male three-spined stickleback is characterized by the rapid appearance and intensification of red pigmentation across the entire ventral and lateral surfaces of the fish. Overall then, amplification of the physical characteristics of the signal (intensity and distribution) is associated with the appearance of behaviours which promote signal display (see Endler, 1983 for a discussion of similar relationships between colour and courtship in guppies). The amplification of the male signal coincides with changes in the female's ability to detect the signal. Gasterosteus aculeatus views the world through a mixed rod/cone retina (Wunder, 1926; Ali et al, 1968;Wagner, 1972) containing one pigment matched to the background ("green") and one offset from the background ("red")(Cronly-Dillon and Sharma, 1968). The relative sensitivities of the visual pigments in gravid three-spined females shifts strongly towards red during the spring/summer breeding season. During this time, females generally travel close to the substrate (Li and Owings, 1978a;pers. obs.), a world dominated by horizontally transmitted green spacelight (see e. g., Levine et al, 1980;Lythgoe, 1980). Since the searching female's peak spectral sensitivity is in the orange-red area, an increase in the number of orange-red photons reflected from the male will increase the probability of signal detection. If the background absorption does not rapidly swamp the signal, increasing the number of photons reflected may also increase the distance at which the signal is first detected. The wavelength of the photons reflected by the male's nuptial signal also overlaps the peak spectral sensitivity range of the female's offset pigment. This emphasizes the contrast between the male (close, bright object) and the background (green spacelight). The importance of nuptial colouration in emphasizing the conspicuousness (contrast) of the courting individual has been documented for several teleost species (Baylis, 1974; Barlow, 1976; Endler, 1980,1983; Neil, 1984;Kodric-Brown, 1985). 93 THE REINFORCEMENT OF FEMALE CHOICE A male's courtship colour is a powerful predictor of his parental colour relative to other males in the population (McLennan and McPhail, 1989a). If red plays a positive role in parental care then the effects of natural selection will reinforce the effects of sexual selection. During the egg and fry guarding periods the distribution of red is maintained in the lower jaw/ opercular areas where it is maximally displayed during aggressive interactions (lunge/bite) and broadside head down threat displays. Since threat signals decrease aggressive interactions between males by stimulating avoidance responses in intruders (Pelkwijk and Tinbergen, 1937;Rowland, 1982), it is conceivable that red may function as a deterrent to potential territorial intruders. Such deterrence an important component of the stickleback breeding system. Nest raiding by males and females, individually or in roving packs, has been documented through direct observation of the phenomenon and implied through discovery of conspecific eggs in stickleback stomach analyses (Hynes, 1950; Black and Wootton, 1970; Semler, 1971). The intensity of such raids varies from unsuccessful, through minor egg stealing to total destruction of both eggs and nest, followed by eviction of the territorial male (Guiton,1960; Black and Wootton, 1970; Black, 1971; Li and Owings, 1978a,b; unpubl. data). Additionally, removal of decaying eggs and maintenance of a current of water over the eggs are essential components in embryo development, and development is favoured by fewer, longer (uninterrupted) fanning bouts (Van Iersel, 1953; van den Assem, 1967). A more subtle effect of territorial intrusion, then, is manifested through a reduction in the time available for nest and egg care (Sargent and Gebler,1980). Following egg hatching, parental males move into a frenzy of fry retrieval and patrol which coincides with an increase in both levels of aggression and levels of colour in the mandibular region (Segaar, 1961; Black, 1971; Huntingford, 1976). At this stage, the young are particularly susceptible to predation by other fishes (McPhail, 1969; Moodie.1972) and invertebrates (Moodie,1972). So, shifting the balance of energy and time expenditure from territory defense to brood care contributes to an increase in the percentage of zygotes which successfully reach the juvenile stages of development. Given that (i) nuptial red as a deterrent favours the shift towards increased parental care time and (ii) signal intensity during courtship is a reliable predictor of the signal intensity during parental care, 94 a female who spawns with a brighter male may be maximizing her clutch's chance of survival to at least the early juvenile stage. MALE COLOUR AND MALE BEHAVIOUR: WHAT IS SO SPECIAL ABOUT WINNERS? Cladistic analysis revealed three levels of differences between winning and losing males in this test population of G. aculeatus. At the first level, typified by the inactive group, winners are more behaviourally vigorous and brightly coloured than their losing counterparts. Such exaggerated asymmetries occur during the early stages of territory acquisition and maintenance, filtering dull intensity males out of the breeding subset of the population (van den Assem, 1967; Black, 1971; Sargent and Gebler, 1980; Ward and Fitzgerald, 1987). This initial level, then, reflects differences between territorial and non-territorial males. The second level is based upon differences among the territorial males (active and fighting groups). At this level, relative intensities of the red signal are critical: winners are brighter than losers. There are no significant differences in the behavioural vigour of these opponents at any stage; nor are there differences in colour except at the time of choice. It is therefore impossible to predict the outcome of a female choice test based upon any characteristic other than the male's colour intensity in the presence of both a competing male and a receptive female. This decoupling of male colour and behaviour mirrors the results obtained when the relationship between the two was examined in isolated medium intensity males (Rowland, 1984; McLennan and McPhail, 1989b). The third level represents a subtle difference between winners and losers. Social facilitation of colour development (Reisman,1968) and colour maintenance (McLennan and McPhail, 1989a) have been documented for this species. This study demonstrated that the magnitude of a male's colour response to a female is dependent upon the context of female presentation in winning, territorial males. These individuals attain two colour peaks: one in isolation, in response to a captive female; and the second, and higher peak during male/male competition for a free, courting female (Fig. 5.6a). Losing rivals reach the first, but not the second peak in colour intensity (Fig. 5.6b). This result suggests that the presence of other territorial males enhances the colour signal for winners, but not losers, over the brief time interval of this experiment. Studies are required to elucidate the mechanism(s) underlying such a subtle difference between males in this population. S U M M A R Y A series of female choice trials revealed three groups of males in a population of anadromous Gasterosteus aculeatus : inactive losers (losing male did not participate in choice test), active losers (losing male active during choice test) and fighting losers (subgroup of active losers defined by female behaviour during choice test). The "inactive losers" group represents a "no choice" situation. In the remaining two groups, females responded preferentially to the most intensely coloured member of the competing male pair. This preferential response was strongest during the pre-choice, captive presentation where the majority of females oriented head-up to, and tracked, the brighter red male. Once released, the female's initial response to the brighter courtship signal was overridden by the behavioural actions of the duller intensity male in 4 out of 12 trials. Although two of these males eventually lost the female to their rival, the remaining two individuals succeeded in spawning. Overall 10 of 12 females spawned with the brighter red male. Given the brief breeding season and the advantage enjoyed by the most brightly coloured males in mate attraction, a high level of territorial intrusion is one way for an individual to attract a mate if he is surrounded by more intensely coloured neighbours. 96 CHAPTER SIR FINRLE THE COLOUR HIERARCHY IN THE MALE'S NUPTIAL SIGNAL The mosaic nuptial colouration signal potentially provides information for three levels of mate choice in Gasterosteus aculeatus. These levels reflect, in part, an interaction between nuptial signal characteristics which are unique to G. aculeatus (red body and blue eyes) and characters which are shared with other members of the gasterosteidae (red pelvic spine membranes). The unique status of red and blue is reflected in their importance as transmitters of information at each of the three levels of mate choice postulated for this species. Females initially discriminate "potential mates" from "non-mates" (mate recognition). At this point, the combination of at least two of the structural components of sexually dimorphic body colour, the absolute hues red and blue, transmit the same message for all males: "I am a sexually mature, territorial male three-spined stickleback." The presence of the plesiomorphic trait red pelvic spines further reinforces the "male" component of the signal. A fair amount of redundancy therefore exists at this level, and this redundancy is hierarchical in nature: while all signal components (spine colour, blue and red body colour) identify sex, both blue eye and red body identify species and sexual maturity and red identifies territorial status. Since signal redundancy increases the probability that the encoded information (message) will be received (Shannon and Weaver, 1949; Hailman,1977), all components play a major role in the early stages of mate recognition in G. aculeatus. Given that even nonterritorial males develop blue eyes and red pelvic spines, it is unlikely that these characters will be important on any other discriminatory level. Note, however, that since blue eye is a species specific signal (autapomorphy), it conveys relatively more information than spine colour and thus should be involved to a greater degree in the next level of mate choice. Red, on the other hand, is a territory specific signal, therefore if females are discriminating between males on the basis of nuptial colouration, intermale variability in the intensity and distribution of red should increase in importance over the remaining two levels of choice. Once a subset of potential mates has been recognized females may-differentiate amongst them based upon criteria pertaining to courtship status (breeding status recognition). For example, males with incomplete nests, a large number of clutches, or fry will attack rather than court an intruding female (Segaar, 1961; Wootton, 1971, 1972, 1976; Huntingford, 1976, 1977). If a female responds indiscriminantly with the receptive head-up signal to every territorial male she encounters, she is risking injury from a male attack and is expending both time and energy in searching, responding, and fleeing from unresponsive males. Information about male breeding status is potentially available in the relationship amongst the structural components of the signal; i. e., the association of these structural dimensions provides temporal information. Once again, just as red and blue are absolute markers of species and sex, the relationship between red, blue and black in this context is an absolute and reliable signal of the stage a particular male has reached in his spawning cycle. The relationships on this second level of choice retain information available from the first level of mate recognition; however, the plesiomorphic character, pelvic spine colour, adds no information at this second level of mate choice. Blue eye on its own contributes little information past identification of species and sex. It is impossible to determine the breeding status of an individual based upon eye colour (hue) or intensity alone. Although the remaining two colours, red and black, are more variable across the cycle in terms of both intensity and distribution, no single component is sufficient for the delineation of all breeding stages. For example, distribution of red across the entire ventral and lateral surfaces of an individual reliably signals a courting male; however, without intensity data, pigmentation in only the mandibular region could indicate either a fry guarding or nest building male. Since males in these two stages will attack all intruders, from the female's perspective they are equally unresponsive to her courtship overtures. At this level, then, the distribution of red across the entire ventral and lateral surfaces of an individual reliably distinguishes a courting male from parental and nest building males. Although blue eye does not contribute any additional information, it does reinforce the message encoded within the distribution of red body colour. Female discrimination among the subset of organisms perceived as "potential courting mates" represents the third level of mate choice in this population (differentiation among mates). Questions of the relationship between male colour and female behaviour have traditionally been investigated by offering females a choice between red and nonred males. As discussed previously, this distinction falls at the first level of mate choice, mate recognition, where the "hue" components of the nuptial signal play an important role in species and sex identification. For researchers interested in exploring the third level of mate choice, it is critical to test the female's ability to differentiate between members of a more select subset of the population: courting males. How much intermale variability is expressed in each of the six components of the nuptial mosaic signal? All males converge upon four of the six components: eye colour (hue), distribution of red, degree and distribution of black. This leaves two potential sources of intermale variability; the intensity of blue eye and red body colour. Since the intensity of eye colour is a less powerful predictor of male breeding status than the intensity of red body colour at the second level of mate choice (two different intensity states for eye colour versus four states for red body colour), it is unlikely that intermale differences in the intensity of blue will be important on this third level of choice. Not surprisingly, variability in signal intensity between courting males is greater for red than blue and both intra- and intermale variability in the overall intensity of red is greatest during courtship. So the original intuition of previous researchers is supported by the results of this study: the intensity of red is potentially an important signal at the third level of mate choice in G. aculeatus. The three structural components of red provide information at all levels of choice: hue identifies species, distribution signals a courting male, and intensity provides the basis for differentiation among courting males. This conclusion is supported by information available in the phylogenetic tree. The presence of "entire body" breeding coloration represents the persistence of an ancestral trait in three-spined sticklebacks. The actual event associated with the initial elaboration of the distribution of colour occurred in the ancestor of the Pungitius + Culaea + Gasterosteus clade. If, as this rapid elaboration within one species suggests, this event was driven by run- away sexual selection, then it should be possible to document the effects of female choice based upon differences in intensity of body colour in G. aculeatus. A series of female choice experiments confirmed this phylogenetic prediction. In contrast to the evolutionary patterns delineated for the diversification of colour distribution, the elaboration of the second structural component of colour, hue, remains unresolved by phylogenetic analysis. There are three equally parsimonious transformation series for colour: (1) black gives rise to red and gold, (2) gold gives rise to red and black or (3) red gives rise to black and gold. McLennan et al (1988) assumed that transformation (3) was the more likely scenario based upon the observation that "red" is the plesiomorphic colour of the ventral fins (spines) in this family. The transformation series for pelvic fin colour is "red plesiomorphic, with secondary loss of colour in two genera: Spinachia spinachia and the ancestor of the Pungitius + Culaea clade". Based on the premise that the nuptial signal in G. aculeatus was simply "red", pelvic spine colour and body colour were treated as equivalent components of one breeding signal and the evolutionary elaboration of colour postulated to have been "red spines to red body". However, the experimental studies reported herein demonstrate that the nuptial signal is a complex structural and temporal mosaic. The complexity of the signal casts doubt upon the original postulate that the red colour in the spines and body are homologous. The hypothesis that "red pelvic spines" and "red body" are distinct characters is supported by attempts to rear offspring of bright and dull males in the laboratory. Although the diet of Artemia nauplii did not provide the pigment types necessary for the development of body colour, all individuals developed red pelvic spines. Evidently spine and body colour, although scored as "red", are not based upon the same pigment. Hence, the experimental studies have led to a modification of the phylogenetic assessment for colouration. While distribution of the colour signal represents the persistence of an ancestral trait, signal hue is associated with the more recent G. aculeatus speciation event; both "red body" and "blue eyes" are autapomorphic (unique) traits for G. aculeatus [s. s.). This suggests that some of the structural components of the nuptial signal have been dissociated in evolutionary time. It appears, then, that the original analysis of the macroevolutionary elaboration of colour hue was too simplistic. However, removal of these characters (#17 and 18) will not affect the topology of the tree. Nor will it affect the macroevolutionary predictions since they are based upon the associations between breeding behaviours and (1) the elaboration of colour distribution and (2) the existence of differences in colour between genera. 100 In summary then, there is a hierarchy of female mate choice decisions within G. aculeatus which, in turn, is associated with a hierarchy based upon the importance of four colours - red pelvic spine membranes, blue eye, black body and red body - in these decisions. All colours provide important and redundant information at the first level of choice, mate (sex and species) recognition. Pelvic spine colour adds no information at the second level (breeding status), and blue eye colour adds no information at the third. Superimposed upon this colour hierarchy lies the overall importance of red to female mate choice decisions in this population. The three structural components of red provide information at all levels of choice: hue identifies species, distribution signals a courting male, and intensity provides the basis for differentiation among courting males. MECHANISMS OF COLOUR ELABORATION: FISHERIAN RUN-AWAY SEXUAL SELECTION OR TRUTH IN ADVERTISING? The pigments underlying the sticklebacks' flamboyant nuptial signal are all extracted, in one form or another, from their food. Intuitively, this establishes a positive link between pigment display and success at capturing carotenoid-containing prey items. Given such a direct relationship between colour and feeding, several authors have argued that the intensity of an individual's carotenoid-based signal is a truthful advertisement of his foraging abilities (Endler, 1980,1983;Kodric-Brown and Brown, 1984,1985; Rowland, 1984). Recent investigations of the mechanisms involved in encoding carotenoids in the nuptial signal have demonstrated that the system is more complex than a simplistic "foraging efficiency = colour intensity" relationship in G. aculeatus. Sticklebacks are opportunistic feeders (Hartley, 1948;Hynes, 1950). Since they consume a wide variety of prey items and appear to be capable of rapid switches in prey preferences, food availability may not be a consistent, limiting factor in three-spined stickleback populations (Hynes,1950;Worgan and Fitzgerald, 1981 a, b; Ward and Fitzgerald, 1983; Walsh and Fitzgerald, 1984). More importantly though, the majority of the three-spined sticklebacks' prey items are a potential source of carotenoids, so pigment availability may not be a limiting factor in colour development either. In addition, there have been no investigations of the relationship between carotenoid content and either the energetic or 101 nutrient value of prey items. It is therefore impossible to determine exactly what an increase in the intensity of red is signalling in terms of foraging abilities: i. e., it could mean "I have been able to capture (1) a greater number of prey absolutely, (2) a greater number of carotenoid rich prey, (3) prey items of high calorific value". Ultimately, we may discover that these messages are all variations on the underlying pronouncement that "I am a good forager". However, in the absence of data concerning the coding rules and relationships between several messages and one signal, the assumption that colour intensity = foraging ability is premature. Once prey have been captured, the pathway from prey pigments to subsequent predator colour in Gasterosteus aculeatus involves a series of complex and poorly understood steps (Table 6.1): 1. PREY CAPTURE may be a reflection of foraging ability, which may in turn reflect overall "genetic quality" of the male (Kodric-Brown and Brown,1984); no research in this area for Gasterosteus aculeatus 2. PIGMENT EXTRACTION 3. PIGMENT TRANSFORMATION differences at the level of physiological control: 4. PIGMENT STORAGE enzymatic, hormonal, neurological mechanisms 5. PIGMENT MOBILIZATION FROM STORAGE AREAS (currently no research in these areas) 6. PIGMENT DEPOSITION IN CHROMATOPHORES differences in absolute number and distribution of chromatophores: hormonal, neurological mechanisms (Titschak, 1922;Craig-Bennett, 1931 ; Ikeda 1933;Reisman,1968). 7. PHYSICAL CONDITION interaction between genotype.physiology and behaviour to produce phenotype;degree of nuptial colouration a direct reflection of ability of an individual to meet environmental challenges;some evidence that parasites may draw carotenoids from G. aculeatus host (Czeczuga, 197lb); no other evidence for a link between health and colour. Table 6.1: Potential factors Involved in the pathway from pigments i n prey items to nuptial colouration i n male Gasterosteus aculeatus . The pathway between prey carotenoids and male nuptial colouration is a result of the dynamic interactions between genetical, physiological and 102 behavioural influences. Given the complexity of the rules underlying the encoding of the message in the nuptial signal, it is unlikely that only one intersexual selection mechanism has been responsible for shaping the signal. It seems more reasonable to postulate that the evolutionary elaboration of the signal which has resulted in the production of intense red G. aculeatus males, represents the outcome of an interplay between both runaway (acting at steps 2-6: Fig. 6.1) and truth in advertising (acting at steps 1 and 7: Fig. 6.1) mechanisms. CONCLUDING REMARKS The tradition of studying behavioural evolution by combining phylogenetic and experimental information was developed by Tinbergen and Lorenz many years ago. However, as discussed in chapter 2, this tradition foundered as both ethologists and systematists became deeply emeshed within their own theoretical revolutions. This thesis follows in the steps of other researchers attempting to re-establish a dialogue between systematics and ethology, to the mutual benefit of both. It is a tribute to the original ethologists' insights, because it refines and re-emphasizes the power of a phylogenetic/experimental integration, and because it reinforces the fundamental views of stickleback behaviour unveiled by Pelkwijk and Tinbergen's elegant experiments in 1937. Gasterosteus aculeatus is a fascinating, flamboyant creature. Although there is an intangible perception that everything important has been already said about colour in this system, phylogenetic analysis combined with experimental investigation has revealed a wealth of questions awaiting further research: What is the relationship between colour and territory "quality"? Is the potential for female choice realized in the field? What was responsible for the surge of evolutionary changes which appeared in the ancestor of the Pungitius + Culaea + Gasterosteus clade? How can we resolve the order of the multistate character red-black-gold? This thesis was opened with a discussion of the valuable role played by G. aculeatus in the development of Western and Eastern cultures. It is therefore appropriate to 103 close with some eloquent words befitting one of evolution's most glorious creations: the three-spined stickleback. The c o u r t s h i p o f f ishes . . . is usua l l y a c o l d - b l o o d e d a f fa i r . N o t so t h e s t i c k l e b a c k ' s . Ear ly in t h e y e a r h e p r e p a r e s a s u i t a b l e p l a c e fo r his e s t a b l i s h m e n t b y c l e a r i n g t h e n e i g h b o u r h o o d o f a l l p o s s i b l e r ivals. The w e a k e r g o t o t h e wa l l ; a t l eas t , t h e y a r e c o m p e l l e d t o d e c a m p , a n d p a s s e n g e r s h a v e t o p r e p a r e fo r b a t t l e a t short n o t i c e . The c h a l l e n g e is d e v o i d o f f o rma l i t y ; t h e l o rd in po s s e s s i on d a s h e s u p o n t h e s t r a n g e r w i t h o u t a n y w a r n i n g a n d a t t a c k s h im t o o t h a n d na i l - t h a t is, w i t h n i m b l e m o u t h a n d t r e n c h a n t sp ine . P e r a d v e n t u r e t h e first o w n e r o f t h e c o v e t e d n o o k is n o t a b l e t o h o l d his o w n , for a c h a n c e thrust o f t h e e r r an t knight 's s p e a r m a y f i nd a jo int in t h e ha rness e v e n o f a l a r ge r a n d m o r e p o w e r f u l f i sh. W h i c h e v e r w i n s , t h e m e e d o f v i c t o r y is s o o n a p p a r e n t . T h e v a n q u i s h e d c o m b a t a n t f a d e s in to a s h e n g r e y , a n d slinks a w a y , p e r h a p s t o d i e o f his w o u n d s ; b u t t h e v i c to r , b e g i n n i n g t o g l o w w i t h b r i gh te r hue s a n d t o sh i ne w i t h c l e a r e r lustre, s eek s t h e l e g i t i m a t e r e w a r d o f v a l o u r in t h e f avou r s o f t h e fair sex. M a x w e l l , 1904 104 LITERATURE CITED ADALSTEINSSON, H. 1979. Size and food of arctic char Salvelinus alpinus and stickleback Gasterosteus aculeatus in Lake Myvatn. Oikos 32: 228-231. ALBRECHT, H. 1966. Zur Stammesgeschichte einiger Bewegungsweisen bei Fischen, untersucht am Verhalten von Haplochromis (Pisces, Cichlidae). Z. Tierpsychol. 23: 270-302. ALCOCK, J . 1984. Animal behavior. 3rd. ed. Sinauer Associates, Sunderland, MA. ALEXANDER, R. D. 1962. The role of behavioral study in cricket classification. Syst. Zool. 11: 53-72. ALI, M.A., ANCTIL, M. and MOHIDEEN, H. M. 1968. Structure retinienne et la vascularisation intraoculaire chez quelques poissons marins de la region de Gaspe. Can. J . Zool. 46: 729-745. ASSEM, J . VAN DEN. 1967. Territory in the three-spined stickleback, Gasterosteus aculeatus L. An experimental study in intra-specific competition. Behaviour Suppl. 16: 1-164. ATZ, J . W. 1970. The application of the idea of homology to behavior. In Development and Evolution of Behavior. Edited by L. R. Aronson, E. Tobach, D. S. Lehrman, and J . S. Rosenblatt. W. H. Freeman & Co.,San Francisco, pp. 53-74 BAERENDS, G. P. 1958. Comparative methods and the concept of homology in the study of behaviour. Arch. Neerl. Zool. 13 (Suppl. 1): 401-417. BAGGERMAN, B. 1958. An experimental study on the timing of breeding and migration in the three-spined stickleback {Gasterosteus aculeatus L.). Arch. Neer. Zool. 12: 105-317. 1968. Hormonal control of reproductive and parental behaviour in fishes. In Perspectives in endocrinology: hormones in the lives of lower vertebrates. Edited by E. J . W. Barrington and C. Barker Jorgensen. Academic Press, London, pp. 351-404. BAGNARA, J . T. and HADLEY, M. E. 1973. Chromatophores and color change: the comparative physiology of animal pigmentation. Prentice-Hall, Inc., New Jersey. BAKKER, Th. C. M. 1986. Aggressiveness in sticklebacks (Gasterosteus aculeatus L.): a behaviour-genetic study. Behaviour 98: 1-143. and SEVENSTER, P. 1983. Determinants of dominance in male sticklebacks (Gasterosteus aculeatus L.) Behaviour 86: 55-71. BALDACCINI, N. E. 1973. An ethological study of reproductive behaviour, including the colour patterns of the cichlid fish, Tilapia mariae (Boulenger). Monit. Zool. Italy NS 7: 247-290. 105 BANNISTER, L. H. 1965. The fine structure of the olfactory surface of teleostean fishes. Quart. J . Microscop. Sci. 106: 332-342. BARLOW, G. W. 1976. The midas cichlid in Nicaraguan Lakes. In Investigations of the ichthyofauna of Nicaraguan Lakes. Edited by T. B. Thorson. University of Nebraska Press, Lincoln, pp. 332-358. BARRAUD, E. M. 1955. Notes on the territorial behaviour of captive ten-spined sticklebacks (Pygosteus pungitius). Brit. J . Anim. Behav. 3:134-136. BARRETT, J . and BUTTERWORTH, P. E. 1968. The carotenoids of Polymorphus minutus (Acanthocephala) and its intermediate host, Gammarus pulex. Comp. Biochem. Physiol. 27: 575-581. BAYLIS, J . R. 1974. The behavior and ecology of Heterotilapia multispinosa (Teleostei: Cichlidae). Z. Tierpsychol. 34: 115-146. BEATTY, R. A. 1949. The pigmentation of cavernicolous animals III. The carotenoid pigments of some amphipod Crustacea. J . Exp. Biol. 26: 125-130. BEDDARD, F. E. 1892. Animal coloration: an account of the facts and theories relating to the colours and markings of animals. Swan Sonnerschein and Co., London. BENGSTON, S. 1971. Food and feeding of diving ducks breeding in Lake Myvatn, Iceland. Ornis. Fenn. 48: 77-92. BERTIN, L. 1925. Recherches bionomiques, biometriques et systematiques sur les Epinoches (Gasterosteides). Ann. Inst. Oceanogr., Monaco 2: 1-204. BLACK, R. 1971. Hatching success in the three-spined stickleback (Gasterosteus aculeatus) in relation to changes in behaviour during the parental phase. Anim. Behav. 19: 532-541. and WOOTTON, R. J . 1970. Dispersion in a natural population of three-spined sticklebacks. Can. J . Zool. 48:1133-1135. BOYDEN, A. 1947. Homology and analogy: a critical review of the meanings and implication of these concepts in biology. Amer. Midi. Nat. 37: 648-660. BROOKS, D. R. 1988. Behavioural comparisons of host and parasite phylogenies. Ann. Rev. Ecol. Syst. 19: 235-259. and WILEY, E. O. 1988. Evolution as entropy: toward a unified theory of biology. 2nd. Ed. The University of Chicago Press, Chicago. BRUNING, C. 1906. Versuche uberdas Horen der Fische. Natur u. Haus 14: 312-313. BRUSH, A. H. and REISMAN, H. M. 1965. The carotenoid pigments in the three-spined stickleback, Gasterosteus aculeatus. Comp. Biochem. Physiol. 14:121-125. CAMIN, J . H. and SOKAL, R. R. 1965. A method for deducing branching sequences in phylogeny. Evolution 19: 311-326. 106 CHAPIN, J . P. 1917. The classification of the weaver-birds. Bull. Amer. Mus. Nat. Hist. 37: 243-280. CHEN, T. R. and REISMAN, H. M. 1970. A comparative chromosome study of the North American species of sticklebacks (Teleostei: Gasterosteidae). Cytogenet. 9: 321-332. COAD, B. W. and POWER, G. 1973. Observations on the ecology and phenotypic variation of the threespine stickleback, Gasterosteus aculeatus L , 1758, and the blackspotted stickleback, G. wheatlandi Putnam, 1867, (Osteichthyes: Gasterosteidae) in Amory Cove, Quebec. Can Field Nat. 87: 113-122. CRAIG-BENNETT, A. 1931. The reproductive cycle of the three-spined stickleback, Gasterosteus aculeatus, Linn. Phil. Trans. Roy. Soc. London, B, 219:197-279. CRONLY - DILLON, J . and SHARMA, S. C. 1968. Effect of season and sex on the photopic spectral sensitivity of the three-spined stickleback. J . Exp. Biol. 49: 679-687. CROZIER, G. F. 1967. Carotenoids of seven species of Sebastodes. Comp. Biochem. Physiol. 23: 179-184. 1970. Tissue carotenoids in prespawning and spawning sockeye salmon (Oncorhynchus nerka). J . Fish. Res. Bd. Canada, 27: 973-975. CZECZUGA, B. 1970. Some carotenoids in Chironomus annularis Meig. larvae (Diptera: Chironomidae). Hydrobiol. 36: 353-360. 1971. Studies on the carotenoids in Artemia salina L. eggs. Comp. Biochem. Physiol. B., 40: 47-52. 1975. Carotenoids in fish. IV. Salmonidae and Thymallidae from polish waters. Hydrobiol. 46: 223-249. 1976. Carotenoid content in some Crustaceans from the Baltic Sea. Bull. Acad. Pol. Sci. 24: 349-353. 1979a. Carotenoids in fish. XIX. Carotenoids in the eggs of Oncorhynchus keta (Walbaum). Hydrobiol. 63: 45-47. 1979b. Carotenoids in fish. XXI. Percidae from polish waters. Acta. Hydrobiol. 21: 1-7. 1980a. Carotenoids in fish. XXVI. Pungitius pungitius (L.) and Gasterosteus aculeatus L. (Gasterosteidae). Hydrobiol. 74: 7-10. 1980b. Changes occurring during the annual cycle in the carotenoid content of Gammarus lacustris G. O. Sars (Crustacea: Amphipoda) specimens from the River Narew. Comp. Biochem. Physiol. 66: 569-572. . and Czerpak, R. 1966. Carotenoids in certain Diaptomidae (Crustacea). Comp. Biochem. Physiol. 17: 523-534. 107 and Czerpak, R. 1968. Carotenoids in Cyclops strenuus strenuus Fischer (Crustacea: Copepoda). Comp. Biochem. Physiol. 26: 101-110. and KRYWUTA, S. 1981. Blue astaxanthin-proteins of Eudiaptomus amblydon. Biochem. Syst. Ecol. 9: 339-340. DARWIN, C. 1871. The descent of man, and selection in relation to sex. John Murray, London. 1872. The origin of species. 6th. ed. J . Murray, London. DAVIES, B. H., HSU, W-J. and CHICHESTER, C. O. 1970. The mechanism of the conversion of (3-carotene into canthaxanthin by the brine shrimp Artemia salina L. (Crustacea: Branchiopoda). Comp. Biochem. Physiol. 33: 601-615. DIEBOLD, A. R., JR. 1968. Anthropological perspectives. In Animal communication. Edited by T.A. Sebeok. Indiana University Press, Bloomington, III. pp. 525-571. DUCKE, A. 1913. Uber Phylogenie und Klassification der sozialen Vespiden. Zool. Jahrb. Abt. Syst. Oekol. Geogr. Tiere 36: 303-330. DUNFORD, C. and DAVIS, R. 1975. Cliff chipmunk vocalizations and their relevance to the taxonomy of coastal sonoran chipmunks. J . Mamm. 56: 207-212. EIBL-EIBESFELDT, I. 1975. Ethology: The Biology of Behavior. New York: Holt, Rhinehart & Winston. 2nd ed. EMERSON, A. E. 1938. Termite nests- a study of the phylogeny of behavior. Ecol. Mono. 8: 247-284. EMLEN, S. T. and ORING, L. W. 1977. Ecology, sexual selection, and the evolution of mating systems. Science (Washington, D. C.) 197: 215-223. ENDLER, J . A. 1980. Natural selection on color patterns in Poecilia reticulata. Evolution 34: 76-91. 1983. Natural and sexual selection on color patterns in poeciliid fishes. Env. Biol. Fish. 9: 173-190. FARRIS, J . S. 1970. Methods for computing Wagner trees. Syst. Zool. 19: 83-92. 1979. The information content of the phylogenetic system. Syst. Zool. 28: 483-519. FISHER, R. A. 1930. The genetical theory of natural selection. The Clarendon Press, Oxford. FOX, D. L. 1976. Animal biochromes and structural colours. 2nd ed. Univ. California Press, Berkeley. 108 FREIHOFER, W. C. 1963. Patterns of the ramus lateralis accessorius and their systematic significance in teleostean fishes. Stanford Ichthyol. Bull. 8: 1-189. FRIEDMANN, H. 1929. The Cowbirds. Springfield. Illinois: Charles C. Thomas. FROST, W. E. 1954. The food of pike Esox lucius L. in Windemere. J . Anim. Ecol. 23: 339-360. GAUDREAULT, A. and FITZGERALD, G. F. 1985. Field observations of intraspecific and interspecific aggression among sticklebacks (Gasterosteidae). Behaviour 94: 203-211. GILCHRIST, B. M. 1968. Distribution and relative abundance of carotenoid pigments in anostraca (Crustacea: Branchiopoda). Comp. Biochem. Physiol. 24: 123-147. and GREEN, J . 1960. The pigments of Artemia. Proc. Roy. Soc. B. 152: 118-136. GILL, T. N. 1885. On the mutual relationships of the hemibranchiate fishes. Proc. Acad. Nat. Sci. Phila. 1884: 154-166. GOTSHALL, D. W. 1981. Pacific coast inshore fishes. Sea Challengers and Western Marine Enterprises. Ventura, CA. GREEN, J . 1957. Carotenoids in Daphnia. Proc. Roy. Soc. B. 147: 392-401. 1959. Pigmentation of an ostracod, Heterocypris incongruens. J . Exp. Biol. 36: 575-582. GROSS, J . and BUDOWSKI, P. 1966. Conversion of carotenoids into vitamins A i and A2 in two species of freshwater fish. Biochem. J . 101: 747-754. GUITON, P. 1960. On the control of behaviour during the reproductive cycle of Gasterosteus aculeatus . Behaviour 15: 163-184. HAGEN, D. W. 1967. Isolating mechanisms in threespine sticklebacks {Gasterosteus ). J . Fish. Res. Bd. Canada 24:1637-1692. and McPHAIL, J . D. 1970. The species problem within Gasterosteus aculeatus on the Pacific coast of North America. J . Fish. Res. Bd. Canada 27:147-155. and GILBERTSON, L. G. 1972. Geographic variation and environmental selection in Gasterosteus aculeatus L. in the Pacific northwest, America. Evolution 26: 32-51. and MOODIE, G. E. E. 1979. Polymorphism for breeding colors in Gasterosteus aculeatus. I. Their genetics and geographic distribution. Evolution 33: 641-648. , MOODIE, G. E. E. and MOODIE, P. F. 1980. Polymorphism for breeding colors in Gasterosteus aculeatus II. Reproductive success as a result of convergence for threat display. Evolution 34: 1050-1059. 109 HAILMAN, J . P. 1977. Optical signals: animal communication and light. Indiana University Press, Bloomington, III. HALLIDAY, T. R. 1983. The study of mate choice. In Mate choice. Edited by P. Bateson. Cambridge University Press, Cambridge, pp 3-32. HARTLEY, P. H. T. 1940. The food of coarse fish: being the interim report on the coarse fish investigation. Freshw. Biol. Assoc. Sci. Publ. No. 3:1-33. 1948. Food and feeding relationships in a community of fresh-water fishes. J . Anim. Ecol. 17:1-14. HEINROTH, O. 1911. Beitrage zur Biologie, namentlich Ethologie und Psychologie der Anatiden. Verh. Ver. Int. Ornithol. Kongr. (Berlin) 1910: 589-702. HENNIG, W. 1950. Grundzuge einer Theorie der phylogenetischen Systematik. Deutscher Zentralverlag, Berlin. 1966. Phylogenetic systematics. University of Illinois Press, Urbana. HERRICK, F. H. 1911a. Nest and nest-building in birds. Part I. J . Anim. Behav. 1:159-192. 1911 b.Nest and nest-building in birds. Part II. J . Anim. Behav. 1: 244-277. 191 Ic.Nest and nest-building in birds. Part III. J . Anim. Behav. 1: 336-373. HERRING, P. J . 1968. The carotenoid pigments of Daphnia magna Strauss - 1. The pigments of animals fed Chlorella pyrenoidosa and pure carotenoids. Comp. Biochem. Physiol. 24: 187-203. HSU, W-J., CHICHESTER, C. O. and DAVIES, B. H. 1970. The metabolism of (3-carotene and other carotenoids in the brine shrimp, Artemia salina L. (Crustacea: Copepoda). Comp. Biochem. Physiol. 32: 69-79. HUBBS, C. H. 1929. The Atlantic American species of the fish genus Gasterosteus. Occas. Papers, Museum Zool. Univ. Mich. 200:1-9. HUDON, J . and GUDERLEY, H. 1984. An electrophoretic study of the phylogenetic relationships among four species of sticklebacks (Pisces: Gasterosteidae). Can. J . Zool. 62: 2313-2316. HUNTINGFORD, F. A. 1976a. The relationship between anti-predator behaviour and aggression among conspecifics in the three-spined stickleback, Gasterosteus aculeatus. Anim. Behav. 24: 245-260. 1976b. An investigation of the territorial behaviour of the three-spined stickleback {Gasterosteus aculeatus) using principal components analysis. Anim. Behav. 24: 822-834. 1977. Inter- and intraspecific aggression in male sticklebacks. Copeia 1977: 158-159. 110 1982. Do inter- and intra-specific aggression vary in /elation to predation pressure in sticklebacks? Anim. Behav. 30: 909-916. HYNES, H. B. N. 1950. The food of fresh-water sticklebacks (Gasterosteus aculeatus and Pygosteus pungitius), with a review of methods used in studies of the food of fishes. J . Anim. Ecol. 19: 36-58. IERSEL, J . J . A. VAN. 1953. An analysis of the parental behaviour of the male three-spined stickleback (Gasterosteus aculeatus L) . Beh. Suppl. 3:1-159. 1958. Some aspects of the territorial behaviour of the male three-spined stickleback. Arch. Neer. Zool., Suppl. 13: 383-400. IKEDA, K. 1933. Effect of castration on the secondary sexual characters of anadromous three-spined stickleback, Gasterosteus aculeatus. Jap. J . Zool. 5: 135-157. J E F F R E Y , S. W. 1961. Paper-chromatographic separation of chlorophylls and carotenoids from marine algae. Biochem. J . 80: 336-342. JEPPS, M. W. 1938. Notes on the breeding of sticklebacks. Proc. Zool. Soc. London, A 108: 253-255. KANEMITSU, T. and AOE, H. 1958. Studies on the carotenoids of the salmon. II. Determination of muscle pigments. Bull. Jap. Soc. Sci. Fish. 24: 555-558. KIRKPATRICK, M. 1987. Sexual selection by female choice in polygynous animals. Ann. Rev. Ecol. Syst. 18: 43-70. KLUGE, A. G. and FARRIS, J . S. 1969. Quantitative phyletics and the evolution of anurans. Syst. Zool. 18:1-32. KODRIC - BROWN, A. 1983. Determinants of male reproductive success in pupfish (Cyprinodon pecosensis). Anim. Behav. 31: 128-137. 1985. Female preference and sexual selection for male coloration in the guppy (Poecilia reticulata). Behav.. Ecol. Sociobiol. 17: 199-205. and BROWN, J . H. 1984. Truth in advertising: the kinds of traits favored by sexual selection. Am Nat. 124: 309-323. 1985. Why the fittest are prettiest. The Sciences (Sept/Oct.): 26-33. KORTMULDER, K. 1972. A comparative study in colour patterns and behaviour in seven asiatic Barbus species (Cyprinidae, Ostariophysi, Osteichthyes): A progress report. Behaviour Suppl. 19: pp. 1-331. KRINSKY, N. I. 1965. The carotenoids of the brine shrimp, Artemia salina. Comp. Biochem. Physiol. 16: 181-187. KYNARD, B. E. 1978. Breeding behavior of a lacustrine population of threespine stickleback (Gasterosteus aculeatus L.). Behaviour 67:178-207. 111 LAM, T. J . , NAGAHAMA, Y., CHAN, K. and HOAR, W. S. 1978. Overripe eggs and postovulatory corpora lutea in the threespine stickleback, Gasterosteus aculeatus L , form trachurus. Can. J . Zool. 56: 2029-2036. LAMBSHEAD, P. J . D. and PATERSON, G. L. J . 1986. Ecological cladistics - an investigation of numerical cladistics as a method for analysing ecological data. J . Nat. Hist. 20: 895-909. LANZING, W. R. J . and BOWER, C. C. 1974. Development of colour patterns in relation to behaviour in Tilapia mossambica (Peters). J . Fish. Biol. 6: 29-41. LEINER, M. 1931. Der Laich und Brutflegeinstinkt des Zwergstichlings Gasterosteus (Pygosteus) pungitius L. Z. Morph. Okol. Tiere 21: 765-788. 1934. Die drei Europaischen Stichlinge (Gasterosteus aculeatus L., Gasterosteus pungitius L. und Gasterosteus spinachia L. ) und ihre Kreuzungsprodukte. Z. Morphol. Okol. Tiere .28:107-154. LEVINE, J . S., LOBEL, P. S. and MacNICHOL, E. F. JR. 1980. Visual communication in fishes. In Environmental physiology of fishes. Edited by M. A. Ali. Plenum Press, New York. pp. 447-475. Ll, S. K. and OWINGS, D. H. 1978a. Sexual selection in the three-spined stickleback: I. Normative observations. Z. Tierpsychol. 46: 359-371. 1978b. Sexual selection in the three-spined stickleback: II. Nest raiding during the courtship phase. Behaviour 64: 298-304. LIMBAUGH, C. 1962. Life history and ecological notes of the tubenose, Aulorhynchus flavidus, a hemibranch fish of western North America. Copeia 1962: 549-555. LORENZ, K. 1941. Vergleichende Bewegungstudien an Anatinen. J . Ornith. 89: 194-294. 1950. The comparative method in studying innate behaviour patterns. Symp. Soc. Exp. Biol. 4: 221-268. 1958. The evolution of behaviour. Sci. Amer. 199: 67-78. LYTHGOE, J . N. 1980. Vision in fishes: ecological adaptations. In Environmental physiology of fishes. Edited by M. A. Ali. Plenum Press, New York. pp. 431-445. McCLEARN, G. E. and DEFRIES, J . C. 1973. Introduction to behavioural genetics. W. H. Freeman and Co., San Francisco. MclNERNEY, J . E. 1969. Reproductive behaviour of the blackspotted stickleback, Gasterosteus wheatlandi. J . Fish. Res. Bd. Canada 26: 2061-2075. McKENZIE, J . A. 1969a. A descriptive analysis of the aggressive behavior of the male brook stickleback, Culaea inconstans . Can. J . Zool. 47:1275-1279. 112 1969b. The courtship behavior of the male brook stickleback, Culaea inconstans (Kirtland). Can. J . Zool. 47:1281-1286. 1974. The parental behavior of the male brook stickleback, Culaea inconstans (Kirtland). Can. J . Zool. 52: 649-652. and KEENLEYSIDE, M. H. A. 1970. Reproductive behavior of the ninespine sticklebacks (Pungitius pungitius (L.)) in South Bay, Manitoulin Island, Ontario. Can. J . Zool. 48:55-61. McLENNAN, D. A., BROOKS, D. R. and McPHAIL, J . D. 1988. The benefits of communication between comparative ethology and phylogenetic systematics: a case study using gasterosteid fishes. Can. J . Zool. 66: 2177-2190. McLENNAN, D. A. and J . D. McPhail. 1989a. Experimental investigations of the evolutionary significance of sexually dimorphic nuptial colouration in Gasterosteus aculeatus (L.): Temporal changes in the structure of the male mosaic signal. Can. J . Zool. 67:1767-1777. 1989b. Experimental investigations of the evolutionary significance of sexually dimorphic nuptial colouration in Gasterosteus aculeatus (L.): The relationships between male colour and male behaviour. Can. J . Zool. 67: 1778-1782. McPHAIL, J . D. 1969. Predation and the evolution of a stickleback (Gasterosteus). J . Fish. Res. Bd. Can. 26: 3183-3208. 1984. Ecology and evolution of sympatric sticklebacks (Gasterosteus): morphological and genetic evidence for a species pair in Enos Lake, British Columbia. Can. J . Zool. 62: 1402-1408. MADDISON, W. P., DONOGHUE, M. J . and MADDISON, D. R. 1984. Outgroup analysis and parsimony. Syst. Zool. 33: 83-103. MANZER, J . I. 1976. Distribution, food, and feeding of the threespine stickleback, Gasterosteus aculeatus, in Great Central Lake, Vancouver Island, with comments on competition for food with juvenile sockeye salmon, Oncorhynchus nerka. Fish. Bull. 74: 647-688. MARLIAVE, J . B. 1976. A theory of storm-induced drift dispersal of the gasterosteid fish Aulorhynchus flavidus . Copeia 1976: 794-796. MATSUNO, T. and KATSUYAMA, M. 1976. Comparative biochemical studies of carotenoids in fishes - XI. Carotenoids of two species of flying fish, mackerel pike, killifish, three-spined stickleback and Chinese eight-spined stickleback. Bull. Jap. Sco. Sci. Fish. 42: 761-763. MAXWELL, H. 1904. British fresh-water fishes. Hutchinson and Co., London, pp. 79-86. MAYDEN, R. L. 1988. Biogeography, parsimony, and evolution in North American freshwater fishes. Syst. Zool. 37: 329-355. 113 MAYR, E. 1958. Behavior and systematics. In Behavior and Evolution. Edited by A. Roe and G. Simpson. Yale University Press, New Haven, CT. pp. 341-366. MICHENER, C. D. 1953. Life-history studies in insect systematics. Syst. Zool. 2: 112-118. MILLER, R. R. and HUBBS, C. L. 1969. Systematics of Gasterosteus aculeatus, with particular reference to intergradation and introgression along the Pacific coast of North America: a commentary on a recent contribution. Copeia 1969: 52-69. MILNE, M. J . and MILNE, L. J . 1939. Evolutionary trends in caddis-worm case construction. Ann. Entomol. Soc. Amer. 32: 533-542. MOODIE, G. E. E. 1972. Predation, natural selection and adaptation in an unusual threespine stickleback. Heredity 28: 155-167. 1986. The population biology of Culaea inconstans, the brook stickleback, in a small prairie lake. Can. J . Zool. 64:1709-1717. , McPHAIL, J . D. and HAGEN, D. W. 1973. Experimental demonstration of selective predation on Gasterosteus aculeatus. Behaviour 47: 95-105. MORRIS, D. 1952. Homosexuality in the ten-spined stickleback {Pygosteus pungitius L) . Behaviour 4: 233-261. 1958. The reproductive behaviour of the ten-spined stickleback (Pygosteus pungitius L.). Behaviour Suppl. 6:1-154. MURAL, R. J . 1973. The pliocene sticklebacks of Nevada with a partial osteology of the Gasterosteidae. Copeia 1973: 721-735. NEIL, S. 1984. Color pattern variability and behavioral correlates in the firemouth cichlid, Cichlosoma meeki. Copeia 1984: 534-538. NELSON, J . S. 1971. Comparison of the pectoral and pelvic skeletons and of some other bones and their phylogenetic implications in the Aulorhynchidae and Gasterosteidae (Pisces). J . Fish. Res. Bd. Canada 28: 427-442. 1984. Fishes of the world. 2nd ed. John Wiley and Sons, New York. O'DONALD, P. 1980. Genetic models of sexual selection. Cambridge University Press, Cambridge. 1983. Sexual selection by female choice. In Mate choice. Edited by P. Bateson. Cambridge University Press, Cambridge, pp. 53-66. OWEN, D. F. 1960. The nesting success of the heron Ardea cinerea in relation to the availability of food. Proc. Zool. Soc. Lond. 133: 597-617. PAANAKKER, J . E. and HALLEGRAEFF, G. M. 1978. A comparative study of the carotenoid pigmentation of the zooplankton of Lake Maarsseveen 114 (Netherlands) and of Lac Pavin (Auvergne, France) - I. Chromatographic characterization of carotenoid pigments. Comp. Biochem. Physiol. 60: 51-58. PAEPKE, H. J . 1983. Die Stichlinge. A. Ziemsen Verlag, Wittenberg, Lutherstadt. PARSONS, P. A. 1972. Genetic determination of behaviour (mice and men). In Genetics, environment and behaviour. Edited by L. Ehrman, S. Omenn and E. Caspari. Academic Press, New York. pp. 75-103. PARTALI, V., OLSEN, V., FOSS, P. and LIGAEN-JENSEN, S. 1985. Carotenoids in food chain studies - I. Zooplankton (Daphnia magna) response to a unialgal {Scenedesmus acutus) carotenoid diet, to spinach, and to yeast diets supplemented with individual carotenoids. Comp. Biochem. Physiol. 82: 767-772. PEEKE, H. V. S., WYERS, E. J . and HERZ, M. J . 1969. Waning of the aggressive response to male models in the three-spined stickleback {Gasterosteus aculeatus L) . Anim. Behav. 17: 224-228. PELKWIJK, J . J . TER and TINBERGEN, N. 1937. Eine reizbiologische Analyse einiger Verhaltensweisen von Gasterosteus aculeatus L. Z. Tierpsychol. 1: 193-200. PENCZAK, T. 1966. Comments on the taxonomy of the three-spined stickleback, Gasterosteus aculeatus Linnaeus. Ohio J . Sci. 66: 81-87. PETRUNKEVITCH, A. 1926. The value of instinct as a taxonomic character in spiders. Biol. Bull. 50: 427-432. PLATH, O. E. 1934. Bumblebees and their ways. New York: MacMillan Publ. Co. PRINCE, E. E. 1885. On the nest and development of Gasterosteus spinachia at the St. Andrews Marine Laboratory. Ann. Mag. Nat. Hist. 16: 487-496. RABOURN, W. J . , QUACKENBUSH, F. W. and PORTER, J . W. 1954. Isolation and properties of phytoene. Arch. Biochem. Biophys. 48: 267-274. REIFFERS, C. 1984. Reproduction et comportement des femelles de trois especes d'epinoches (Gasterosteidae) sympatrique en milieu naturel. M.Sc. thesis, Laval University, Quebec, Canada. REISMAN, H. M. 1961. The reproductive behavior of the five-spined stickleback, Eucalia inconstans (Kirtland). Am. Zool. 1: 468. 1963. Reproductive behaviour of Apeltes quadracus, including some comparisons with other gasterosteid fishes. Copeia 1963: 191-192. 1968a. Reproductive isolating mechanisms of the blackspotted stickleback, Gasterosteus wheatlandi. J . Fish. Res. Bd. Canada 25: 2703-2706. . 1968b. Effects of social stimuli on the secondary sex characters of male three-spined sticklebacks, Gasterosteus aculeatus. Copeia 1968(4): 816-826. and CADE, T. J . 1967. Physiological and behavioral aspects of reproduction in the brook stickleback, Culaea inconstans . Am. Midi. Nat. 77: 257-295. REMANE, A. 1956. Die Grundlagen des naturlichen Systems der vergleichenden Anatomie und Phylogenetik. 2. Geest und Portig. K. G., Leipzig. 1961. Gedanken zum Problem: Homologie und Analogie, Preadaptation und Parallelitat. Zool. Anz. 166: 447-470. RICKLEFS, R. E. 1987. Community diversity: relative roles of local and regional processes. Science (Washington, D. C. ) 235: 167-171. ROWLAND, W. J . 1974a. Reproductive behavior of the fourspine stickleback, Apeltes quadracus. Copeia 1974: 183-194. 1974b. Ground nest construction in the fourspine stickleback, Apeltes quadracus. Copeia 1974: 788-789. 1982a. The effects of male nuptial coloration on stickleback aggression: a reexamination. Behaviour 80: 118-126. 1982b. Mate choice by male sticklebacks, Gasterosteus aculeatus. Anim. Beh. 30: 1093-1098. 1983. Interspecific aggression and dominance in Gasterosteus. Env. Biol. Fish. 8: 269-277. 1984. The relationships among nuptial coloration, aggression, and courtship of male three-spined sticklebacks, Gasterosteus aculeatus. Can. J . Zool. 62: 999-1004. SARGENT, R. C. and GEBLER, J . B. 1980. Effects of nest site concealment on hatching success, reproductive success, and paternal behavior of the threespine stickleback, Gasterosteus aculeatus. Behav. Ecol. Sociobiol. 7: 137-142. SAUNDERS, J . T. 1914. A note on the food of freshwater fish. Proc. Camb. Phil. Soc. 17: 236-239. SCHNEIRLA, T. C. 1952. A consideration of some conceptual trends in comparative psychology. Psychol. Bull. 49: 559-597. SEBEOK, T. A. 1972. Perspectives in zoosemiotics. Mouton and Co., The Hague. SEGAAR, J . 1961. Telencephalon and behaviour in Gasterosteus aculeatus males. Behaviour 18: 256-287. SEMLER, D. E. 1971. Some aspects of adaptation in a polymorphism for breeding colours in the threespine stickleback (Gasterosteus aculeatus). J . Zool. Lond. 165: 291-302. SEVENSTER, P. 1951. The mating of the sea stickleback. Discovery 12: 52-56. 116 1961. A causal analysis of a displacement activity (fanning in Gasterosteus aculeatus L) . Beh. Suppl. 9:1-170. 1968. Motivation and learning in sticklebacks. In The central nervous system and fish behaviour. Edited by D. Ingle. University of Chicago Press, London, pp. 233-245. SEVENSTER-BOL, A. C. A. 1962. On the causation of drive reduction after a consummatory act (in Gasterosteus aculeatus L.). Archs neerl. Zool. 15: 175-236. SHANNON, C. E. and WEAVER, W. 1949. The mathematical theory of communication. Univ. Illinois Press, Urbana. SIEGEL, S. 1956. Nonparametric statistics for the behavioral sciences. McGraw-Hill, New York. SMITH, W. J . 1968. Message-meaning analyses. In Animal communication. Edited by T.A. Sebeok. Indiana University Press, Bloomington, III. pp. 44-60. 1977. The behavior of communicating. Harvard University Press, Cambridge, Ma. SMITHE, F. B. 1975. Naturalist's Color Guide. The American Museum of Natural History, New York SNYDER, R. J . 1984. Seasonal variation in the diet of the threespine stickleback, Gasterosteus aculeatus, in Contra Costa County, California. Calif. Fish and Game 70: 167-172. SOKAL, R. R. and ROHLF, J . 1969. Biometry: The principles and practice of statistics in biological research. W. H. Freeman and Company, San Francisco. STONOR, C. R. 1936. The evolution and mutual relationships of some members of the Paradiseidae. Proc. Zool. Soc. London 1936: 1177-1185. SYMONS, P. E. K. 1965. Analysis of spine raising in the male three-spined stickleback. Behaviour 26: 1-74. THOMAS, B. O. 1962. Behavioral studies of the brook stickleback, Eucalia inconstans (Kirtland). Am Zool. 2: 452. TINBERGEN, N. 1948. Social releasers and the experimental method required for their study. Wilson Bull. 60: 6-52. 1951. The study of instinct. Clarendon Press, Oxford. 1959. Comparative studies of the behavior of gulls (Laridae): a progress report. Behaviour 15: 1-70. 117 1962. The evolution of animal communication - a critical examination of methods. Symp. Zool. Soc. London 8:1-6. 1964. On aims and methods of ethology. Z. Tierpsychol. 20: 410-433. TITSCHACK, E. 1922. Die sekundaren Geschlechtsmerkmale von Gasterosteus aculeatus L. Zool. Jahrb. Physiol. 39: 83-148. TRIVERS, R. L. 1972. Parental investment and sexual selection. In Sexual selection and the descent of man: 1871-1971. Edited by B. Campbell. Aldine Press, Chicago, pp. 136-179. TSCHANZ, B. and SCHARF, M. 1971. Nestortwahl und Orienteirung zum Nestort beim Dreistachligen Stichling. Rev. Suisse Zool. 78: 717-721. VEHRENCAMP, S. L and BRADBURY, J . W. 1984. Mating systems and ecology. In Behavioural ecology: an evolutionary approach. 2nd ed. Edited by J . R. Krebs and N. B. Davies. Sinauer Associates, Sunderland, MA. pp. 251-278. WAGNER, H. J . 1972. Vergleichende Untersuchungen uber das Muster der Sehzellen und Horizontalen in der Teleostier Retina (Pisces). Z. Morph. Tiere 72:77-130. WALKEY, M. 1967. The ecology of Neoechinorhynchus rutili (Muller). J . Parasit. 53: 795-804. WALSH, G. and FITZGERALD, G. J . 1984. Resource utilization and coexistence of three species of sticklebacks (Gasterosteidae) in tidal salt-marsh pools. J . Fish. Biol. 25: 405-420. WARD, G. and FITZGERALD, G. J . 1983. Fish predation on the macrobenthos of tidal salt-marsh pools. Can. J . Zool. 61: 1358-1361. 1987. Male aggression and female mate choice in the threespine stickleback, Gasterosteus aculeatus L. J . Fish. Biol. 30:679-690. WATROUS, L E. and WHEELER, Q. D. 1981. The out-group comparison method of character analysis. Syst. Zool. 30:1-11. WESTERFIELD, F. 1922. The ability of mudminnows to form associations with sounds. Comp. Psychol. 2: 187-190. WHEELER, W. M. 1919. The parasitic Aculeata, a study in evolution. Proc. Amer. Philos. Soc. 58: 1-40. 1928. The Social Insects: Their Origin and Evolution. New York: Harcourt, Brace & Co. WHITMAN, C. O. 1899. Animal Behavior. In Biological Lectures, Wood's Hole. ed. C. O. Whitman, pp. 285-338. Boston: Ginn and Co. 118 WHORISKEY, F. and FITZGERALD, G. J . 1985. The effects of bird predation upon an estuarine stickleback (Pisces: Gasterosteidae) community. Can. J . Zool. 63: 301-307. and REEBS, S. G. 1986. The breeding-season population structure of three sympatric, territorial sticklebacks (Pisces: Gasterosteidae). J . Fish. Biol. 29: 635-648. WICKLER, W. 1961. Okologie und Stammesgeshichte von Verhaltensweisen. Fortschr. Zool. 13: 303-365 WILEY, E. O. 1981. Phylogenetics: the theory and practice of phylogenetic systematics. John Wiley and Sons, New York. 1988a. Vicariance biogeography. Ann. Rev. Ecol. Syst. 19: 513-542. 1988b. Parsimony analysis and vicariance biogeography. Syst. Zool. 37: 271-290. WILZ, K. J . 1970a. Causal and functional analysis of dorsal pricking and nest activity in the courtship of the three-spined stickleback, Gasterosteus aculeatus. Anim. Behav. 18: 115-124. 1970b. The disinhibition interpretation of the 'displacement' activities during courtship in the three-spined stickleback, Gasterosteus aculeatus. Anim. Behav. 18: 682-687. 1971. Comparative aspects of courtship behavior in the ten-spined stickleback, Pygosteus pungitius (L.). Z. Tierpsychol. 29: 1-10. 1972. Causal relationships between aggression and the sexual and nest behaviours in the three-spined stickleback (Gasterosteus aculeatus). Anim. Behav. 20: 335-340. 1973. Quantitative differences in the courtship of two populations of three-spined sticklebacks, Gasterosteus aculeatus. Z. Tierpsychol. 33:141-146. WINN, H. E. 1960. Biology of the brook stickleback, Eucalia inconstans (Kirtland). Am. Midi. Nat. 63: 424-438. WOOTTON, R. J . 1970. Aggression in the early phases of the reproductive cycle of the male three-spined stickleback (Gasterosteus aculeatus). Anim. Behav. 18: 740-746. 1971. Measures of the aggression of parental male three-spined sticklebacks. Behaviour 40: 228-262. 1972. Changes in the aggression of the male three-spined stickleback after fertilization of eggs. Can. J . Zool. 50: 537-541. 1974. The inter-spawning interval of the female three-spined stickleback, Gasterosteus aculeatus L. J . Zool. London 172: 331-342. 1976. The biology of the sticklebacks. Academic Press, London. 119 WORGAN, J . P. and FITZGERALD, G. J . 1981a. Habitat segregation in a salt marsh among adult sticklebacks (Gasterosteidae). Env. Biol. Fish. 6: 105-109. 1981b. Diel activity and diet of three sympatric sticklebacks in tidal salt marsh pools. Can. J . Zool. 59: 2375-2379. WUNDER, W. 1926. Uber den Bau der Netzhaut von SuBwasserfischen, die in groBer Tiefe leben. Z. Vergl. Physiol. 4: 22-36. 1930. Exper imented Untersuchungen an dreistachligen Stichling {Gasterosteus aculeatus L.) Wahrend der Laich zeit. Z. Morph. Okol. Tiere 16: 453-498. ZAHAVI, A. 1975. Mate selection - a selection for handicap. J . Theor. Biol. 53: 205-214. APPENDIX A 120 Data matrix of behavioural data collected for the Gasterosteidae (chapter 2). See chapter two for explanation of character transformation series. Taxa are represented by capital letters: A = Spinachia; B = Apeltes ; C = Pungitius ; D = Culaea ; E = Gasterosteus aculeatus ; F = Gasterosteus wheatlandi ; X = the outgroup, Aulorhynchus Jlavidus. TAXON CHARACTER A B c D E F X 1 1 2 2 1 2 2 0 2 0 0 0 0 1 1 0 3 1 1 1 1 1 1 1 4 0 0 1 1 0 5 0 0 1 1 1 1 0 6 0 0 1 1 1 0 7 0 1 1 0 8 0 1 1 1 1 1 0 9 0 0 1 0 10 0 1 1 1 1 1 0 11 0 0 1 1 1 1 0 12 0 0 0 0 1 0 13 1 1 1 1 0 14 0 0 1 1 0 0 15 0 0 1 1 1 1 0 16 9 0 1 1 1 1 0 17 9 0 1 1 0 18 9 0 0 0 1 0 19 1 1 1 1 1 1 1 20 1 1 1 1 1 1 1 21 0 0 1 1 0 0 22 1 2 1 1 1 1 0 23 0 0 0 0 1 1 0 24 1 1 1 1 1 1 1 25 0 1 1 1 1 0 2f> 0 0 1 1 1 1 0 27 0 0 1 1 1 1 0 APPENDIX B 121 Raw colour scores for individual male G. aculeatus across a complete breeding cycle (data from chapter 3). The breeding cycle is divided into four stages: days 1-5 = nest building and maintenance; days 6-8 = courtship; days 9-13 = Egg guarding; days 14-23 = fry guarding. Three colours were scored: (a) red body; (b) blue eyes; (c) black body. (a) TOTAL INTENSITY SCORES (RANGE 0-60) FOR RED BODY MALE NUMBER »AY 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 1 0 16 2 0 2 8 18 6 17 6 0 0 18 7 0 7 0 0 11 2 2 17 2 0 8 10 16 15 13 3 0 0 12 10 2 7 0 4 10 3 1 22 2 1 12 6 16 4 5 0 2 1 15 11 3 0 0 1 3 4 4 33 6 5 16 8 21 13 6 2 0 1 18 8 0 0 0 1 9 5 2 29 4 2 8 10 18 9 2 0 4 0 11 8 5 0 0 2 6 6 10 44 14 20 22 20 40 16 16 31 12 16 51 43 15 12 5 6 25 7 13 44 16 15 26 23 34 16 21 31 12 16 50 49 18 10 4 8 25 8 13 40 34 16 21 28 34 12 20 28 8 15 48 48 16 10 4 8 23 9 12 46 28 18 22 21 34 12 16 15 8 11 36 34 16 8 4 6 20 10 13 38 14 16 21 22 34 12 12 15 8 9 36 34 16 10 4 4 19 11 6 34 14 15 23 14 30 12 12 18 6 3 26 38 12 8 0 4 20 12 14 46 10 11 17 18 28 10 12 16 4 3 18 21 12 4 0 4 17 13 10 43 21 11 17 14 26 10 11 15 6 0 17 15 16 8 0 4 17 14 12 46 20 11 13 18 23 15 14 23 8 12 43 45 16 10 4 6 21 15 13 46 26 11 13 15 24 14 15 22 8 15 43 43 14 10 4 6 20 16 11 31 24 14 13 16 16 11 16 22 8 14 43 44 12 8 2 6 21 17 8 34 17 13 16 16 15 10 17 22 6 14 33 31 10 8 3 7 15 18 8 36 13 7 11 15 15 10 16 22 6 12 33 30 10 8 4 6 15 19 8 39 16 9 11 11 19 15 14 21 5 11 33 30 10 8 1 2 17 20 8 38 17 9 10 15 18 12 10 21 5 10 33 29 11 8 1 2 15 21 8 38 17 10 11 16 16 10 10 20 6 10 34 29 14 10 0 1 14 22 8 33 18 9 11 18 24 10 10 20 5 11 30 30 9 5 1 3 10 23 4 32 20 12 11 20 16 10 8 14 5 8 31 31 8 5 0 1 10 (b) TOTAL INTENSITY SCORE (RANGE 0-5) FOR BLUE EYES MALE NUMBER »AY 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 1 2 3 0 0 3 2 4 4 3 0 3 3 2 1 0 0 2 0 4 2 2 4 0 1 4 2 4 3 4 1 4 3 2 1 1 0 2 0 3 3 3 J 4 1 1 3 4 4 2 3 0 3 2 3 3 3 1 2 1 3 4 4 4 1 2 4 2 4 3 3 1 3 2 4 2 1 1 2 3 4 5 2 3 1 1 3 3 4 3 3 1 3 4 4 3 1 0 3 3 2 6 4 5 3 2 4 4 5 3 4 4 4 5 4 4 3 3 5 3 4 7 4 4 4 3 4 5 5 4 4 5 4 5 5 5 4 3 4 4 4 8 5 5 4 2 4 5 5 3 3 5 5 4 5 4 4 2 4 4 5 9 4 3 4 4 3 3 4 3 2 4 5 5 4 4 3 3 5 4 4 10 4 4 3 3 4 4 4 3 3 4 4 3 5 4 2 3 4 3 4 11 2 4 2 3 4 4 4 3 3 4 4 4 5 4 3 4 4 3 5 12 4 4 3 4 4 5 4 3 4 4 5 4 4 4 2 3 4 3 4 13 4 4 4 4 4 4 3 3 4 4 5 2 3 4 2 3 5 3 5 14 4 5 3 4 5 5 5 3 5 4 3 3 5 5 3 4 5 4 5 15 4 4 4 4 4 5 3 4 4 5 4 4 5 5 4 4 5 5 5 16 4 4 4 4 5 5 3 3 5 5 4 4 5 5 3 4 5 4 5 17 3 4 4 5 4 3 5 3 4 5 5 3 5 5 3 3 5 4 4 18 4 3 4 3 5 5 5 3 5 5 4 4 5 5 3 2 5 3 5 19 4 4 4 4 4 3 5 3 3 4 5 4 5 4 4 4 5 3 5 20 4 3 3 4 5 5 4 3 3 4 4 4 4 5 4 4 5 4 4 21 2 3 4 3 3 3 5 2 3 4 4 3 5 4 3 4 4 3 5 22 2 2 4 3 3 3 4 3 3 3 4 3 5 5 3 4 4 3 4 23 2 2 3 2 3 4 3 3 3 3 3 2 4 4 2 3 3 4 4 123 (c) TOTAL INTENSITY SCORE (RANGE 83-86) FOR BLACK BODY MALE NUMBER DAY 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 1 86 85 85.5 85 85 84 85 84.5 85 85.5 84.5 84.5 84.5 84 84.5 85 85 84 84 2 86 85 85.5 85 85 84 85 84.5 85 85.5 84.5 84.5 84.5 84 84.5 85 85 84 84 3 86 85 85.5 85 85 84 85 84.5 85 85.5 84.5 84.5 84.5 84 84.5 85 85 84 84 4 86 85 85.5 85 85 84 85 84.5 85 85.5 84.5 84.5 84.5 84 84.5 85 85 84 84 5 86 85 85.5 85 85 84 85 84.5 85 85.5 84.5 84.5 84.5 84 84.5 85 85 84 84 6 86 86 85 86 86 85 85 85 86 85 84.5 86 85 85 85 86 86 85 85 7 86 86 85 86 85 86 86 85 86 86 84.5 86 85 85 85 86 86 85 84.5 8 85 85 85 86 86 86 86 86 86 85 85 86 85 85 85 85 85 85 84.5 9 84 84 84.5 84 85 85 85 85 86 85 85 84.5 84 84 85 84 85 84 84 10 84 84 84.5 84 85 85 85 84 84 85 84.5 85 84 84 84 84 85 84 84 11 84 84 84.5 84 85 85 85 84 84 85 84.5 85 85 84 84 84 84 84 84 12 84 84 84.5 84 85 84 85 84.5 85 84.5 84.5 85 84 84 84 84 84 84 84 13 84 84 84.5 84 84 84 85 84.5 85 84.5 84 84 84 84 84 84 84 84 84 14 84 84 84.5 84 84.5 84 84 84.5 84.5 84.5 84.5 84 84 84 84 84.5 84 84 84 15 83 84 84.5 84 84 84 84 84 84.5 84.5 84.5 83.5 84 84 84.5 83 84 84 84 16 83 84 84.5 84 84 84 83.5 84 84.5 84.5 84 83.5 84 83 84 83 84 84 84 17 83 84 84.5 84 84 84 83.5 84.5 84.5 84.5 84 83.5 84 83 84 83 84 84 84 18 83 84 84.5 84 84 84 83.5 84.5 84.5 84.5 84 84 84 83 84 83 84 84 84 19 83 84 84.5 84 84 84 83.5 84.5 84.5 84.5 84 84 84 83 84 83 84 84 84 20 83 84 84.5 84 84 84 83.5 84.5 84.5 84.5 84 84 84 83 84 83 84 84 84 21 83 84 84.5 84 84 84 83.5 84.5 85 84.5 84 84 84 83 84 83 84 84 84 22 83 84 84.5 84 84 84 83.5 84.5 85 85 84 84 84 83 84 83 84 84 84 23 83 84 84.5 84 84 84 83.5 84.5 85 85 84.5 84 84 83 84 83 84 84 84 124 APPENDIX C Measurements of behavioural intensity and colour scores before (initial) and after (final) presentation of a captive, gravid female to solitary males for a 5 minute test period (chapter 4 and experimental stage 5.1a in chapter 5). Pair number represents male pairs in female choice tests (chapter 5). @ = missing data (data also missing for pair #7); DNS = did not spawn; * = winning member in the choice test (female spawned with this male). MALE INITIAL FINAL # DURATION # PAIR # COLOUR COLOUR BITES BITING ZIGZAGS # RED BODY (BLUE EYE) (seconds) 1 * 9 (4) 15 (5) 26 15.77 110 2 @ @ @ 1 3 15 (4) 25 (5) 89 81.61 25 4 * 15 (4) 28 (5) 104 62.92 13 2 5* 19 (4) 26 (5) 24 7.57 23 6 1 (2) 8 (3) 37 15.13 6 3 7 * 16 (2) 36 (4) 39 53.14 19 8 0 (2) 7 (4) 22 7.49 5 4 9 # 0 (4) 6 (4) 13 2.10 28 10 22 (4) 38 (4) 45 19.29 26 5 11 1 (2) 9 (4) 58 28.94 5 12* 1 (2) 11 (3) 0 0.0 0 6 15 28 (4) 30 (5) 52 24.12 1 16 6 (2) 12 (4) 37 17.45 28 DNS 17 # 19 (2) 32 (4) 18 11.42 85 18 0 (3) 0 (4) 44 15.60 39 8 l b 2 (2) 4 (3) 1 0.39 4 2b * 17 (3) 20 (4) 64 27.93 18 9 3b 0 (2) 14 (2) 24 8.32 32 4b * 2 (2) 14 (4) 102 78.39 9 10 5b 0 (0) 6 (2) 13 6.29 2 6b * 0 (2) 8 (4) 17 6.94 14 11 APPENDIX C CONTINUED 125 MALE INITIAL FINAL # DURATION # PAIR # COLOUR COLOUR BITES BITING ZIGZAGS # RED BODY (BLUE EYE) (seconds) 7b 0 (2) 7 (3) 54 48.21 4 8b * 20 (4) 33 (5) 135 68.35 18 12 9b 0 (2) 2 (2) 10 3.66 0 10b* 0 (2) 0 (2) 6 2.07 0 13 l i b * 2 (2) 2 (2) 0 0.0 0 12b 0 (2) 5 (3) 12 3.64 2 14 13b 20 (4) 12 (3) 0 0.0 0 14b 12 (2) 38 (5) 47 13.69 1 DNS 15b 26 (3) 34 (5) 22 10.93 52 16b* 19 (3) 24 (3) 34 15.58 15 15 17b* 4 (2) 11 (2) 29 9.81 2 18b 14 (4) 24 (5) 31 10.65 0 16 3c 6 (4) 19 (5) 25 19.67 71 4c * 23 (5) 30 (5) 98 55.29 3 17 5c * 31 (4) 40 (5) 62 83.59 24 6c 0 (1) 1 (1) 0 0.0 0 18 7c 2 (2) 18 (4) 48 14.64 7 8c * 0 (0) 5 (2) 56 15.06 0 19 11c 32 (5) 43 (5) 60 23.95 80 12c * 22 (5) 36 (5) 26 14.91 8 20 17c * 0 (2) 12 (3) 25 9.51 9 18c 4 (2) 7 (2) 40 16.33 4 21 126 A P P E N D I X D Measurements of male colour and behaviours during the captive, gravid female presentation to two competing territorial males (raw data from chapter 5). (a) The intensity of the male's red colour signal: COLOUR AFTER COLOUR AT CHOICE DIVIDER (stage 5. Id) (stage 5. le) MALE PAIR # LOSER WINNER LOSER WINNER 1 6 16 8 26 2 4 19 6 32 3 6 30 20 45 4 0 34 0 38 5 40 31 44 42 6 1 1 20 9 7 4 6 15 27 8 0 12 0 26 9 8 21 12 27 10 0 17 0 32 11 0 2 0 3 12 8 24 20 36 13 14 4 16 22 14 4 11 8 16 15 28 25 30 38 16 0 7 0 22 17 11 16 16 30 18 15 38 20 50 19 11 1 15 17 20 28 35 37 60 21 4 9 5 12 127 (b) Behavioural interactions by focal male with captive, gravid female (3 minutes of behavioural data recorded per male) # NEST-RELATED # BITES # ZIGZAGS (#dig/fan+#glue+ #creep through) MALE PAIR LOSER WINNER LOSER WINNER LOSER WINNER ACTIVE LOSERS 1 0 16 2 2 0 0 3 10 10 14 17 2 0 5 20 6 2 2 0 0 6 0 4 31 14 5 0 7 0 20 6 9 0 0 9 35 18 8 16 0 2 13 0 2 1 1 0 1 14 0 3 5 7 0 0 15 0 0 42 14 7 0 19 5 17 8 0 3 1 20 0 19 7 0 0 0 21 ' 1 1 0 3 0 0 INACTIVE LOSERS 2 1 9 0 15 1 0 4 * * * tt * 8 * » # tt • * 10 • * » tt • • 11 * • » * tt * 12 0 29 0 33 0 0 16 # * tt * • • 17 0 16 13 51 0 11 18 * • • * * * indicates that the comparison is not applicable for this male pair (losing male either remained hidden for duration of test, or was visible but did not participate). (c) Behavioural interactions between males. 128 MALE # BITES # ELASTIC # HD # CIRCLE # INTRU- # CHASES PAIR # REBOUNDS THREAT FIGHTS® SIONS W L W L W L W L W L W L ACTIVE GROUP 1 0, 0 4 8 7 3 0 0 4 0 0 0 3 5 11 8 6 5 17 0 8 10 1 1 2 5 1 2 23 11 9 8 1 0 2 3 0 1 6 0 1 2 0 5 0 0 0 0 1 0 0 7 0 0 6 9 3 6 0 0 5 1 0 0 9 2 7 4 16 12 14 0 0 2 5 1 1 13 0 0 3 2 3 0 0 0 0 0 0 0 14 1 0 0 3 1 0 0 0 1 0 0 0 15 2 1 0 0 0 3 0 0 2 0 0 0 19 0 0 1 1 5 4 0 0 0 0 0 0 20 0 0 1 9 0 12 0 0 1 0 0 0 21 6 1 0 2 0 2 0 0 11 0 2 4 INACTIVE GROUP 2 10 0 6 8 6 4 0 2 8 7 1 6 0 4 * * * » • * • . • • » . g « * * « « « « * IQ » » * • * * * • • * * > 11 * » * * * * * • * * * * 12 0 0 1 5 2 4 0 0 5 0 0 0 1Q » * * * * * * * ft* * * 17 1 0 3 4 3 3 0 0 3 0 0 0 10 » » * f t < • * • * * • « * indicates that the comparison is not applicable for this male pair (losing male either remained hidden for duration of test, or was visible but did not participate). @: fights on the territorial boundaries not included. W = winning male; L = losing male 129 (d) Time captive female spends oriented to each male during the six minute pre- choice presentation (data for inactive group pairs #2,4, 8,10,11,12,16, 17 and 18 not applicable): FIRST THREE MINUTES SECOND THREE MINUTES MALE PAIR # WINNER LOSER WINNER LOSER ACTIVE GROUP 1 180.0 0.0 175.69 4.31 3 45.59 64.87 60.97 60.60 5 36.59 90.39 51.74 110.00 6 # * 40.49 115.32 7 37.98 77.06 114.28 65.72 9 120.92 59.08 120.04 59.96 13 167.54 12.46 130.47 25.23 14 150.24 0.0 139.38 37.29 15 106.55 73.45 94.35 85.65 19 94.64 71.32 112.18 60.33 20 23.13 0.0 29.87 0.0 21 29.80 22.18 33.12 0.0 * indicates missing data APPENDIX E 130 Data matrix of behavioural characters recorded during female choice tests (chapter 5). See methods section of chapter 5 for a description of the characters. CHARACTERS MALE PAIR 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 ! 1 0 1 0 1 0 0 0 0 1 1 0 0 0 0 2 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 3 1 1 1 2 2 1 1 0 2 0 1 2 1 0 0 0 4 1 0 0 1 0 1 0 0 2 0 0 0 0 2 0 0 5 1 1 1 2 2 1 1 0 0 2 2 2 1 0 2 1 6 1 0 0 1 0 0 0 1 0 1 2 1 1 0 0 1 7 1 1 1 1 1 1 0 0 2 0 2 2 0 1 0 0 8 1 0 0 1 0 0 0 0 2 0 0 0 0 1 0 0 9 1 1 1 2 2 1 1 0 0 0 1 0 0 0 0 0 10 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 11 1 0 0 1 0 1 0 0 0 0 0 0 0 0 0 0 12 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 13 1 1 0 1 0 1 0 0 0 0 1 0 0 1 0 0 14 1 0 0 1 0 0 1 1 1 1 1 2 1 1 0 0 15 1 1 1 2 2 1 1 0 0 0 1 1 0 0 0 0 16 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 17 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 18 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 19 1 1 1 1 1 1 1 0 0 0 2 0 0 0 0 0 20 1 1 1 1 1 1 1 1 1 1 2 1 1 0 0 0 21 1 1 1 1 1 1 1 0 2 0 2 1 0 0 0 0 (A) PUBLICATIONS: (i) Published or in press Lindgren, B. S., Borden, J. H., Gray, D. R., Lee, P. C , *Palmer, D. A. and Chang, L. 1982. Evaluation of two trap log techniques for ambrosia beetles in timber processing areas. J . Econ. Entomol. 75: 577-586. McLennan, D. A., Brooks, D. R. and McPhail, J. D. 1988. The benefits of communication between comparative ethology and phylogenetic systematics: A case study using gasterosteid fishes. Can. J. Zool 66: 2177-2190. McLennan, D. A. and McPhail, J. D. 1989a. Experimental investigations of the evolutionary significance of sexually dimorphic nuptial colouration in Gasterosteus aculeatus (L.): Temporal changes in the structure of the male mosaic signal. Can. J. Zool 67: 1767-1777. McLennan, D. A. and McPhail, J. D. 1989b. Experimental investigations of the evolutionary significance of sexually dimorphic nuptial colouration in Gasterosteus aculeatus (L.): The relationship between male colour and male behaviour. Can. J. Zool 67: 1778-1782. McLennan, D. A. and McPhail, J . D. in press. Experimemtal investigations of the evolutionary significance of sexually dimorphic nuptial colouration in Gasterosteus aculeatus (L.): The relationship between male colour and female behaviour. Can. J. Zool Brooks, D. R. and McLennan, D. A. in press. Historical ecology as a research program in macroecology. In Systematics, Historical Ecology and North American Freshwater Fishes. Ed. by R. L. Mayden. Stanford University Press, pp. 000-000. Brooks, D. R. and McLennan, D. A. in press. Historical Ecology: mastering the possibilities. University of Chicago Press. (iii) Published Abstacts McLennan, D. A. Phylogenetic analysis of behavioural evolution: A case study using gasterosteid fishes. American Zoologist. * Please note: D. A. Palmer = D. A. McLennan 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0302341/manifest

Comment

Related Items