Sexual Conflict and Male - Female Coevolution in the Fruit Fly
Urban Friberg
2006
Department of Ecology and Environmental Science Umeå University
SE-901 87 Umeå Sweden
AKADEMISK AVHANDLING
som med vederbörligt tillstånd av rektorsämbetet vid Umeå Universitet för erhållande av filosofie doktorsexamen i ekologi kommer att offentligen försvaras torsdagen den
20:e april 2006, kl. 10.00 i Lilla hörsalen, KBC.
Examinator: Professor Lennart Persson, Umeå Universitet
Opponent: Professor Daniel E. L. Promislow, Department of Genetics, University of Georgia, USA
ISBN:91-7264-055-3
© Urban Friberg 2006
Printed by: Larserics Digital Print
ORGANISATION
Umeå University
Dept. of Ecology and Environmental Science SE-901 87 Umeå
DOCUMENT NAME
Doctoral Dissertation DATE OF ISSUE
April 2006 AUTHOR: Urban Friberg
TITLE:Sexual conflict and male - female coevolution in the fruit fly.
ABSTRACT
Harmony and cooperation was for long believed to dominate sexual interactions. This view slowly started to change 25 years ago and is today replaced with a view where males and females act based on what is best from a costs-benefits perspective. When sex specific costs and benefits differ, concerning reproductive decision influenced by both sexes, sexual conflict will occur. The basis for discordant reproductive interests between the sexes is that males produce many small gametes, while females’ produce few and large gametes. One result of this difference is that the optimal mating rate differs between the sexes. Males, with their many small sperm, maximize their reproductive output by mating with many females, while females often do best by not mating more frequently than to fertilize their eggs, since mating often entails a cost. Sexual conflict over mating is thus an important factor shaping the interactions between the sexes. In this thesis I study this and related conflicts between the sexes, using mathematical models, fruit flies and comparative methods.
Mathematical modelling was used to explore how males and females may coevolve under sexual conflict over mating. This model shows that sexual conflict over mating results in the evolution of costly female mate choice, in terms high resistance to matings, and costly exaggerated male sexual traits, aimed to manipulate females into mating. A key assumption in this model is that males which females find attractive also are more harmful to females. This assumption was tested by housing fruit fly females with either attractive or unattractive males.
Females kept with attractive males were courted and mated more, and suffered a 16 percent reduction in lifetime offspring production. In another study I measured genetic variation in two antagonistic male traits used to compete over females; offence - a male’s ability to acquire new mates and supplant stored sperm, and defence - a male’s ability to induce fidelity in his mates and prevent sperm displacement when remating occurs. Independent additive genetic variation and positive selection gradients were found for both these traits, indicating an ongoing arms race between these male antagonistic traits. This arms race also had a negative impact on females, since high values of offence compromised female fitness. Genetic variation in female ability to withstand male harm was also tested for and found, indicating that females evolve counter adaptations to reduce the effect of harmful male traits. Finally, the proposed link between sexual conflict and speciation was tested. Theory suggests that perpetual sexual arms races will cause allopatric populations to evolve along different evolutionary trajectories, resulting in speciation. This theory was tested using comparative methods by contrasting the number of extant species in taxa with high and low opportunity for sexual conflict. The study showed that taxa with high opportunity for sexual conflict, on average, has four times as many species as those with low opportunity, supporting that sexual conflict is a key process in speciation.
KEY WORDS: cryptic male mate choice, Drosophila melanogaster, female mate choice, multiple mating, sexual conflict, sexually antagonistic coevolution, sexual selection, speciation, sperm competition
LANGUAGE:English
SIGNATURE:
ISBN:91-7264-055-3 NUMBER OF PAGES:22 + 7 papers
DATE:20 March 2006
LIST OF PAPERS
This thesis is based on the following papers, which will be referred to in the text by their Roman numerals.
I Gavrilets, S., Arnqvist, G. and Friberg, U. 2001. The evolution of female mate choice by sexual conflict. Proc. R. Soc. Lond. B 268: 531-539.
II Friberg, U. and Arnqvist, G. 2003. Fitness effects of female mate choice:
preferred males are detrimental for Drosophila melanogaster females. J. Evol.
Biol. 16: 797-811.
III Friberg, U., Lew, T. A., Byrne, P. G. and Rice, W. R. 2005. Assessing the potential for an ongoing arms race within and between the sexes: selection and heritable variation. Evolution 59:1540-1551.
IV Friberg, U. 2006. Male perception of female mating status: its effect on copulation duration and male and female fitness in Drosophila melanogaster.
Anim. Behav., In press.
V Friberg, U. 2005. Genetic variation in male and female reproductive characters associated with sexual conflict in Drosophila melanogaster. Behav. Genet.
35:455-462.
VI Rice, W.R., Linder, J.E., Friberg, U., Lew, T.A., Morrow, E.H. and Stewart, A.D. 2005. Inter-locus antagonistic coevolution as an engine of speciation:
Assessment with hemiclonal analysis. Proc. Natl. Acad. Sci. USA 102: 6527- 6534.
VII Arnqvist, G., Edvardsson, M., Friberg, U. and Nilsson, T. 2000. Sexual conflict promotes speciation in insects. Proc. Natl. Acad. Sci. USA 97: 10460-10464.
Papers I-III and V-VII are published with the kind permission of the publisher
TABLE OF CONTENTS
1.INTRODUCTION……… 5
Sexual selection………. 5
Theories of sexual selection……….. 6
The three-way tug-of-war...………...……… 8
Sexual conflict and speciation………...……… 8
2.AIMS OF THE THESIS………...………… 9
3.STUDY SPECIES………...……... 9
4.RESULTS AND DISCUSSION………. 10
Evolution of female choice by sexual conflict (paper I)...….………... 10
Preferred males are more harmful to females (paper II)……...……….. 12
Antagonistic coevolution between male sexual traits (paper III)……...… 13
Cryptic male mate choice: its effect on male and female fitness (paper IV).. 14
Genetic variation in female resistance to male harm (paper V)…………... 14
Sexually antagonistic coevolution as an engine of speciation (paper VI)...… 15
Sexual conflict promotes speciation in insects (paper VII)……….. 17
5.CONCLUSIONS………. 17
6.ACKNOWLEDGEMENTS……… 19
7.REFERENCES………...………... 21
1.INTRODUCTION
Sexual selection
The fact that males and females (or male and female structures) are so different (sexually dimorphic) in many animal and plant species, commonly has its roots in a process known as sexual selection. Darwin (1859), how first acknowledged sexual selection wrote:
“This form of selection depends, not on a struggle for existence in relation to other organic beings or to external conditions, but on a struggle between the individuals of one sex, generally the males, for the possession of the other sex.
The result is not death to the unsuccessful competitor, but few or no offspring.”
Sexual selection has later been more stringently defined by, for example, Andersson (1994)
“Sexual selection of a trait can … be viewed as a shorthand phrase for differences in reproductive success, caused by competition over mates in relation to that trait.“
Sexual selection is normally a sex specific selective force, which shapes the phenotypes of males and females, in order to promote male and female success in the competition over mates. Selection in the wild can thus be viewed to operate at two different levels.
First there is natural selection, where the external physical and biological environment sets absolute and relative conditions for survival, and second there is sexual selection, where individuals compete to secure a mate (or mates). However, natural and sexual selection are often closely integrated, since many characters are subjected to both natural and sexual selection, which may diverge or coincide. Several authors also argue that natural and sexual selection should not be viewed as separate processes. Instead, sexual selection can be considered a component of natural selection (e.g. Endler 1986, Andersson 1994).
But why is it important to compete over mates and how does it relate to fitness? A general difference between the sexes, in polygamous species, is that female fitness is constrained by the number of offspring produced whereas male fitness is constrained by the number of sexual partners (Bateman 1948). This has its basis in that females commonly invest more per offspring than males do (eggs are large and sperm are small). Males consequently have the opportunity to produce many more offspring than any single female can, and females becomes a resource that limits male reproductive success (Trivers 1972).
Competition over mates can take two different forms. Either males compete among themselves and sort out which male(s) will mate with the available females (male-male competition), which leaves little opportunity for females to influence their partner(s).
This form of sexual selection has spurred the evolution of “weapons” (e.g. antlers, extended teeth) in males in many species. Alternatively, males present themselves to females and the outcome of such interactions determine if a male succeeds in securing a fertilization opportunity or not. This later form of sexual selection leaves much more opportunity for females to choose their partner(s). It is primarily this form of sexual selection that has resulted in the many extravagant male ornaments, seen especially in birds, and that has attracted most attention by evolutionary biologist.
Theories of sexual selection
Although the process of sexual selection has long been accepted by now, though controversial when first presented, the mechanisms by which sexual selection operates are still heavily debated. The controversy over this issue is not so much about which mechanisms male-male competition operates through, but rather what drives female mate choice.
Fisher (1930) was the first to suggest a mechanistic explanation for how female mate choice could result in exaggerated male ornaments. His idea was that if a genetic coupling exists between a female preference trait and an arbitrary male character, females mating with males that express the preferred trait would have an indirect fitness advantage, since their sons would express the preferred male trait and thus have higher mating success. The male trait would thus spread in the population, and since the female preference trait is genetically coupled to the male trait also the preference trait will raise in frequency. This creates a self-reinforcing system that continually will exaggerate the female preference and the male trait in the population. The exaggeration of the male trait, and hence the female preference, will eventually come to an halt when the size of the male trait compromises male survival to the extent that the extra mating advantage cannot account for the reduced survival (Lande 1981; Kirkpatrick 1982).
Although appealing in its elegance this explanation for the evolution of female mate choice and male ornaments suffers from the problem that no negative direct selection can act on the female preference trait. If female choice is costly the “runaway” process can never get started (e.g. Kirkpatrick 1985; Pomiankowski 1987a). Cost of choice is probably a reality for females in most species, since choice involves assessing, rejecting and sometimes also locating males. Fisher’s explanation for the evolution of female preference traits and male ornaments therefore seems less likely to be the general mechanism of sexual selection.
Another influential idea in this field is one that usually is referred to as the “good genes”, “indicator” or “handicap” model. Also this idea stems from Fisher’s early work (Fisher 1915), but it first became more generally known when discussed by Williams (1966) and others (Zahavi 1975, 1977; Hamilton and Zuk 1982). The basis of this idea is that if males vary in breeding value for fitness, females can gain an indirect genetic benefit by mating with genetically superior males, since their offspring will inherit some of their father’s qualities. This idea is intuitively very appealing, since sexual selection here reinforces natural selection by female choice. The outcome is however not entirely in line with what natural selection dictates. For females to be able to distinguish between males of differential genetic quality, one or several male phenotypic trait(s) must reflect their bearers’ genetic condition. The trait must therefore be costly for males to express, in order to limit its expression to males in good condition. If the trait would not be costly, cheaters would soon invade the system making the trait useless to females. Coevolution between a female preference trait and an initially rudimentary indicator trait would thus result in exaggerated male ornaments with sizable survival costs to its carrier. This model of sexual selection through female mate choice has now been shown to work on a theoretically level by several authors (e.g. Andersson 1982, 1986; Pomiankowski 1987b; Grafen 1990; Iwasa et al 1991).
An important assumption for this model to work is that fitness is heritable. The few measures of this parameter in the wild do however point to very low numbers, of only a few percent or less (Gustavsson, 1986; Merilä & Sheldon 2000; Kruuk et al 2000;
McCleery et al 2004). A related factor also complicating female choice for good genes is that the same alleles that produce superior male phenotypes might produce inferior female phenotypes (Chippendale et al 2001). Females preferentially mating with males with high phenotypic expression of “male traits” might therefore produce daughters of low quality.
Above two models of sexual selection both rely on indirect benefits (i.e. effects first expressed in the following generation) to females that exert mate choice. Female mate choice could instead have evolved in response to variation in direct effects on female fitness. Males may, for example, vary in the amount of parental care they provide, the quality of the territory they hold, or some other resources that directly influence female fitness. This could select for choosy females with (Price et al 1993) or without (Heywood 1989) selecting for any male trait, since variation in the benefits males provide can be entirely phenotypic. When females rely on a male phenotypic marker that is correlated with the direct benefits received, this model becomes very similar to the indicator model described above, and females are predicted to receive indirect as well as direct benefits from being choosy. However, this model cannot explain female mate choice when females receive nothing but sperm from their mates.
An explanation for the evolution of exaggerated male traits that have been suggested repeatedly is that females may have preferences for particular male traits or behaviours caused by mechanisms unrelated to sexual selection (West-Eberhard 1979, 1984; Ryan 1990; Endler & Basolo 1998). Females’ sensory systems may for instance have evolved in response to natural selection to detect food and predators, with the pleiotropic effect that they also react to these stimuli in other contexts. Males expressing phenotypes or behaviours that to some extent mimic the source to which the female sensory system has adapted may therefore become more interesting to females, giving such males a mating advantage, causing positive selection on this male trait.
This theory has been named sensory exploitation, since males evolve to exploit some pre-existing female sensory biases.
The theory of sexual selection, which the studies included in this thesis test, is called sexually antagonistic coevolution (SAC) (reviewed in Arnqvist & Rowe 2005). This theory is different from the previous ones in that it recognizes and focuses on that males and females often have conflicting reproductive interests. Conflict has for long been acknowledged to be the driving force behind direct competition among males, but has not until lately been thought of as a mechanism shaping male-female sexual interactions (Parker 1979). In an imaginary species, where mating is random and males and females live in lifelong monogamy, a male and his partner’s fitnesses are perfectly correlated and conflict absent (Rice 2000). However, as soon as any of these assumptions is violated, the fitnesses of the sexes will become partly decoupled, setting the stage for sexual conflict.
Polygamy, the prevalent mating system in nature, provides ample opportunity for sexual conflict, since individuals of both sexes commonly mate with multiple partners, which substantially decouples the fitnesses of interacting males and females. When
males and females have no long term interest in their partners reproductive output, selection will act to maximize the outcome of the opposite sex’s investments in the current activities. Since a male’s fitness commonly is dependent on the number of mates and the number of offspring produced by each mate, the theory of SAC predicts that males should evolve means to increase these, also when they compromise his partners’ lifetime fitness. If matng is associated with costs to females, such as a lost opportunity to forage, increased risk of predation, risk of receiving a sexually transmitted disease etc. (reviewed in Choe & Crespi 1997; Daly 1978), females are predicted to have an optimal mating rate (Arnqvist & Nilsson 2000). Males will thus often encounter females that would do best if they did not mate, and selection on males will hence act to find ways to persuade females to mate anyway. Females, on the other hand, will be selected to resist being persuaded into, for them, non-beneficial matings.
Parker’s (1979) first formulation of SAC for long received little attention but later authors’ (Rowe et al 1994 and especially Holland & Rice 1998) treatments of this subject have attracted a lot of attention.
The three-way tug-of-war
Sexual conflict is often portrayed to be driven by male-female disparate reproductive interest (see above). However, in species where males cannot monopolize females male-male competition is predominantly mediated through females. Female thus constitute the playing field where males try to outmanoeuvre one another. The means used by males to promote their success in this competition sometimes have negative consequences for females (Chapman et al 2003), and it is mainly this that motivates the evolution of female resistance with resulting SAC (see fig. 1).
Figure 1. When a male mates to a female in a polygamous species, there is substantial risk that the female will remate before all sperm provided by the male is used. The male should thus try to prevent the female from remalting. This can either be conducted through active mate grading, by transferring a mating plug or via seminal fluids that influence the female reproductive physiology to reduce her inclination to remate. If the female anyhow remates, the first males should try to influence the outcome to his benefit when competing with the other male’s sperm. This male is thus in a sperm defence position.
A male encountering an already mated female should first try to mate with her by overcoming her own, or induced reluctance to remate. If successful
Defence Offence
Reproductive tract &
physiology
Defence Offence
Reproductive tract &
physiology
this male should either physically remove already present sperm or by chemical means outcompet the sperm provided by the first male. This male is thus in an offence position. If any of these male activities have a negative effect on female fitness females should evolve means to neutralize or reduce their effect. (After Rice 1998).
Sexual conflict and speciation
Multiple factors will promote divergence between allopatric populations, including genetic drift, local adaptation to the physical and biotic environment and sexual selection (Coyne & Orr 2004). Sexual selection in general (Darwin 1871; Lande 1981;
West-Eberhard 1983) and sexual conflict in particular (Arnqvist 1998; Rice 1998;
Arnqvist et al 2000; Gavrilets 2000) has been suggested to be an especially important
“engine” of speciation. There are several reasons to believe this. Firstly, selection on sexual characters acts on a large proportion of a population every generation (West- Eberhard 1983). Secondly, sexual selection is more constant in direction because of its insensitivity to environmental fluctuations (West-Eberhard 1983). Thirdly, adaptation to the local physical environment will follow an asymptotic curve, while adaptations to the coevolving biotic environment (predators, preys, parasites etc.) will be continuously ongoing, creating more opportunity for diversification between populations over time (Van Valen 1973). In this respect sexual conflict is a particular potent force driving divergence, since perpetual arms races between the sexes is predicted to drive rapid evolutionary change in males and females (Rice 1998).
2.AIMS OF THE THESIS
The purpose of this thesis was to investigate the logic of sexual conflict and to test some of the main assumptions and predictions of this theory using mathematical modelling, the model organism Drosophila melanogaster and comparative methods.
The main questions addressed in this thesis are:
• Can sexual conflict drive the evolution of costly female mate choice? (Paper I)
• Are sexually preferred males more harmful to females? (Paper II, III)
• Is there genetic variation for male harm to females? (Paper III, V)
• Are male sexually antagonistic traits currently coevolving? (Paper III)
• Are males copulatory investments related to female mating status? (Paper IV)
• Is there genetic variation for female resistance? (Paper V, VI)
• Can sexual conflict drive speciation? (Paper VI, VII)
3.STUDY SPECIES
All empirical studies included in this thesis use the model organism Drosophila melanogaster (the fruit fly). This species has played a critical role in the development of genetics ever since Thomas Hunt Morgan started his work with this species in the early twentieth century. With time, D. melanogaster has found its way to almost all branches of biological science, and the field of sexual conflict is no exception.
D. melanogaster is a polygamous insect species. Males stay at feeding and oviposition sites and locate arriving females by their pheromonal profile (Ferveur 1997). Male courtship is complex but yet stereotypic in its performance, involving following the female, vibrating extended wings, taping the female with the fore tarsus and licking the female genitalia, before mounting is attempted (Hall 1994). Successful matings usually takes less than 20 minutes (Singh & Singh 2004).
Although surprisingly little is known about the life of D. melanogaster in the wild, it is safe to conclude that the laboratory environment differ from the natural in many
respects. Using the fruit fly, or any other laboratory animal or plant species, to study evolutionary biology, therefore has its limitations. The laboratory environment poses new selection pressures on wild caught animals, while it relaxes other. This has the effect that genetic variation that was neutral in wild may become selected upon in the laboratory and vice versa. New laboratory populations are therefore likely to undergo rapid evolutionary change during the first generations in the laboratory (Sgrò &
Partridge 2000). Such populations may therefore be far from their genetic equilibrium, and it may consequently be difficult to draw strong inferences from experiments with such populations. Populations that have a long history of adaptation to its laboratory environment may, on the other hand, differ substantially from the population it once was derived from. However, if general questions of evolutionary biology are studied, such populations should constitute reliable tools, since their new environment, in essence, has become their “natural” environment (Rice et al 2005). That said it is of course naïve to believe that such populations are perfectly adapted to the laboratory environment (Houle & Rowe 2003).
Fruit flies are normally cultured either on a two week discrete generation cycle or with overlapping generations. These culturing protocols both have their benefits and drawbacks. The use of overlapping generations better reflects the natural life history of fruit flies, but has the drawback that the culturing protocol is difficult to mimic during controlled experiments. Culturing flies on discrete generation cycles suffers from this artificial setting, but has the noticeable advantage that fitness is more easily defined and measured. Standing levels of genetic variation in fitness and life history traits is therefore relatively easy to measure on a relevant scale, as well as the fitness effects of manipulating life histories.
Three laboratory populations of fruit flies were used for the experiments in this thesis.
In paper II a population called Dahomey was used. This population is cultured with overlapping generations in fly cages and has an extensive history of adaptation to its laboratory environment. Before the onset of the experiments reported in this thesis it had been kept in the laboratory for at least 600 generations. The population used in paper V was also a population cultured with overlapping generations. Its history of adaptation to the laboratory was shorter, 40-60 generations. A population called LHM
was used for papers III, IV and VI. This population is reared on a discrete generation cycle and had adapted for at least 300 generations before the start of the experiments reported here.
4.RESULTS AND DISCUSSION
Evolution of female choice by sexual conflict (paper I)
In this paper we model the evolution of female mate choice under sexual conflict over mating rate, as verbally outlined by Holland & Rice (1998), in a simple quantitative genetic framework. In the “species” we model mating confers benefits as well as costs to females. The benefits can be thought of as getting sperm to fertilize the eggs, replenish sperm stores and avoiding fertilizing eggs with incompatible sperm, and costs as those described in §1. While the benefits of mating repeatedly are assumed to give diminishing returns, since a moderate mating rate will incorporate all the benefits, the
cost of mating is assumed to be constant per mating. These assumptions result in an optimal female mating rate. Mating is also assumed to convey costs and benefits to males. The marginal benefit is, however, assumed to be constant over matings, so that a male’s mating rate is linearly related to his fitness. There is thus sexual conflict over mating rate in this population; males benefit from mating as much as possible while females do best by mating at a moderate rate.
In the model males are characterised by a sexual trait that influence the probability of mating with females, where higher trait values confer higher mating success. Females are characterized by a preference functions that specifies the probability of mating to males of different trait values. The female preference (mating probability) function is assumed to have an S-shaped form, which approaches zero for males with very small trait values and one for males with very high trait values. The preference function is fixed in its shape but allowed to evolve by sliding along the x-axis to adjust female mating rate (see fig 2.). The female preference function is originally thought to have evolved in response to natural selection on females, to tune their sensory system to detect, for example, food or predators. This model is thus analogous to males exploiting a female sensory bias to increase their mating success, as described in §1.
Figure 2. Males are represented by the value of their sexual trait, y, and females by an S-shaped preference functions, Ψ(y-x), which gives the probability of mating between a female of trait value x and a male with trait value y. The figure depicts two populations with genetic variation in the male trait and the female preference function.
Given this limited set of assumptions, there will be sexual selection on the male trait to increase in value. This will confer a higher mating rate on a population level, which will knock females off their optimal mating rate and select for increased female resistance to mating, by shifting the preference function in fig. 2 to the right. If only sexual selection acts on the male and the female traits, constant escalation of these is expected. The outcome of this arms race depends on the relative strength of sexual selection and the relative amount of additive genetic variation for the male and female traits. If the male trait evolves faster than the female trait, the average mating rate in the population is ever increasing. If females can evolve faster than males the average mating rate in the population will be constant. The mating rate will, however, be higher
than the optimal female mating rate; to what extent is again dependent on the relative speed by which the male and female trait can evolve.
In a more realistic setting natural as well as sexual selection will act on the male and the female trait. If the male trait is an ornament, exaggerated trait values may reduce male survival, and if the value of the female trait influences females’ ability to detect food or predators, its evolution will also be constrained by natural selection. With these more realistic assumptions the dynamics of this model differ from above. Several outcomes are possible depending on the specific parameter values. The most realistic of these are perhaps when natural selection on the male trait is relatively stronger than on the female preference trait. With these assumptions the model has an equilibrium that is stable, or unstable (which will cause the system to cycle). On average the male trait will be exaggerated and female resistance costly. Whether the average female mating rate exceeds the female optimal mating rate, is again parameter dependent, but the displacement is in general small.
A later contribution, building on this model, in which also the slope of the female preference function was allowed to evolve, it was shown that antagonistic coevolution may be less common between a male sexual traits and a trait causing female sensory bias (Rowe et al 2005). Some evidence exists that the shape of the female preference function can evolve (e.g. Basolo 1996, 1998; Wagner 1998), but more research on the plasticity of female preference functions is needed before any general conclusions can be made regarding how frequent male-female arms races can be predicted to be.
Preferred males are more harmful to females (paper II)
This paper sets out to test one of the key predictions of sexual conflict; that male phenotypes females preferentially mate with inflict more harm to females than other male phenotypes. This prediction is in opposition to what is predicted by most other models of sexual selection. In this study size of D. melanogaster males was used as a proxy for male attractiveness, which several contributions have shown to correlate positively to male mating success in this species (e.g. Ewing 1961; Partridge &
Farquhar 1983; Partridge et al 1987; Pitnick 1991). Single females were randomly assigned to be housed for their entire life with either one or two males that either was small or large, while several female life history components were monitored and intersexual interactions quantified.
Females housed with large males suffered a substantial loss in offspring production (16%) compared to females that were housed with small males. This effect was primarily explained by differences in female lifespan between females housed with large and small males (see fig. 3). The effects of male size were to a large extent explained by the behavioural variables measured; courtship and mating rate. Large males courted females more frequently and induced a higher mating rate. Courtship rate was negatively correlated to female offspring production while offspring production peaked at an intermediate mating rate. These results were thus largely in concordance with what is predicted under SAC. The finding that large males are more harmful to female was also verified in another study with a similar experimental set up (Pitnick &
García- González 2002).
Figure 3. Female survival as a function of time in the four groups.
Antagonistic coevolution between male sexual traits (paper III)
In this study we tested for an ongoing arms race between male defence and offence (see fig 1. for explanation of these terms), in D. melanogaster, and measured its consequences to females. As described in §1, sexual conflict is predicted to, at least in part, be driven by male-male competition fought within females. To conclusively show that two traits currently are coevolving requires evidence of (i) additive genetic variation in both traits, (ii) that at least a part of this variation is independent between the traits, (iii) that selection acts on both traits and (iv) that the two traits interact and select upon one another.
To test for these criteria we clonally amplified 35 haploid genomes (hemiclones) and measured male defence and offence ability for these hemiclones. We found significant genetic variation in both these traits. Since nearly all genetic variation among hemiclones is additive we could conclude that male defence and offence fulfil criteria (i). To evaluate criteria (ii) we tested for a partial correlation between offence (defence) traits and another independently measured trait, while controlling for defence (offence).
Using this procedure we found independent additive variation in both offence and defence, fulfilling criteria (ii). Male fitness can be decomposed into three components egg-to-adult survival, male defence and male offence. Values of defence and offence were measured in offspring production and were thus directly related to fitness. Since no negative genetic correlations were found between these components we could conclude that positive selection acted on both defence and offence in this population, fulfilling criteria (iii). No explicit tests were conducted to evaluate criteria (iv).
However, the nature of offence and defence is such that an increase in one of them necessarily comes at the expense of the other, since males do not provide any nuptial gifts in this species (Chapman et al. 1994; Pitnick et al. 1997). The studied population thus fulfilled all these criteria, supporting that male offence and defence currently are antagonistically coevolving in this population.
To test if this male-male arms race potentially could drive an associated arms race between the sexes, we tested how the variation in offence and defence related to
female fitness. We found offence to influence female fitness negatively while the opposite pattern was found for defence. The positive effect by defence was partly explained by that high defence suppressed female remating, which was beneficial to females since remating proved costly. Defence also had a more direct positive effect on female fitness. Our evidence that high values of male offence are costly to females is indeed strong support for that the male-male arms race found could drive an associated arm race between the sexes.
Cryptic male mate choice: its effect on male and female fitness (paper IV)
When the risk of sperm competition varies males are predicted to adjust their reproductive investments accordingly (Bonduriansky 2001; Wedell et al 2002). Female mating status is a factor that covaries with this risk, and males are hence predicted to take this into account when mating with virgin and already mated females. In D.
melanogaster, females change their cuticular hydrocarbone (CH) profile after mating, and males adjust the intensity of their courtship thereafter, courting virgin females more passionately (Tompkins & Hall 1981). In this study I manipulated male perception of female mating status by coating virgin females with CHs from either virgin or already mated females. In response to this manipulation males mated significantly longer with virgin females that they perceived as mated, compared to females they perceived as virgin. Increased copulatory investment had a positive effect on male fitness. When focal males mated with virgin females they perceived as mated they were more successful in defending their sperm from being displaced, since these females remated to a lesser extent. This result also verified one of the assumptions of this study, that mating is costly to males, since males mated longer than normally to virgin females when fooled to believe these were mated, which improved their sperm defence.
Earlier contributions have shown that mating is costly to females in this species (e.g.
Fowler & Partridge 1989) and prolonged copulations could therefore be predicted to have a negative influence on female fitness. However, prolonged copulations were only about 10% longer, and a recent study have shown that remating once only reduces female fitness in this species with about 6% percent (unpublished data B. Kuijper, A.
D. Stewart & W. R. Rice). Female fitness was nevertheless affected in accordance with this prediction in one experiment. However, in a more controlled experiment the opposite pattern was found.
Genetic variation in female resistance to male harm (paper V)
Above empirical studies primarily focuses on the male part of SAC. However, for SAC to occur also genetic variation in female resistance to male harm must exist. In this study I investigated this part of SAC. To test for female resistance ten completely inbred lines of D. melanogaster was used in two experiments. In the first experiment, females from all ten lines were exposed to males from each of the ten lines for 25 hours (limited male exposure) and thereafter kept in individual vials and scored for the time of their death. The second experiment was identical to the first, with the one difference that the females were housed with one male each throughout their lives (lifetime exposure to males). Female resistance was here defined as the ability to withstand an increased exposure to males, which could be measured by the relative decline in lifespan between experiment one and two (see fig.4).
Figure 4. Mean female lifespan for the 10 female lines in the two experiments. Differences in reduction between lines in lifespan, over the two experiments, indicate that the female lines differed in their ability to withstand an increased level of exposure to males.
When combining experiment one and two, the interaction term “experiment x female line” was significant, indicating that there was genetic variation in female resistance to male harm. When each of the experiments were analysed separately, evidence for male harm was also found since the male-lines varied significantly in there effect on female lifespan in both experiments. This study tested for total genetic variation in female resistance, including additive, dominance and epistatic variation. For SAC to currently occur evidence of additive genetic variation is needed. Other studies have however verified that additive genetic variation for female resistance exists in this species (Linder & Rice 2003; Wigby & Chapman 2004; Lew et al 2006).
Sexually antagonistic coevolution as an engine of speciation (paper VI)
Sexually antagonistic coevolution is hypothesised to be an important engine of speciation. However, the evidence that SAC is currently operating is limited. In this paper we describe a method called “hemiclonal analysis”, with which current evolution can be estimated in laboratory populations. This method was applied when studying current evolution in males (paper III) and females (Linder & Rice 2005) from the same population, in two previous contributions. Results from these studies are here extracted in order to test for SAC.
Hemiclonal analysis is a substitute for measuring ongoing evolution in the wild, which with current methodology is a daunting task. Hemiclonal analysis consists of four steps.
1) Creating an “island population”. An “island population” is a large outbred laboratory population that has had plenty of time to adapt to its new (laboratory) environment, so that problems associated with inbred populations and populations that only recently was introduced to the laboratory environment are avoided (see §3). The population is further propagated in a way easily mimicked under experimental settings.
2) Clonally amplify a set of haploid genomes. With cytogenetic techniques nearly whole haploid genomes (hemiclones), including the X-chromosome and the two
major autosomes but excluding the small fourth chromosome, is cloned. These hemiclones constitute of different randomly drawn chromosomes copies from the base population and form a sub sample of the genetic material found in the base population.
3) Measuring genetic variation. Each hemiclone is expressed in many individuals, which has the other half of their genome randomly drawn from the base population. In this way each hemiclone is expressed in many different genetic backgrounds. When the effect of a hemiclone is averaged over many genetic backgrounds the hemiclone’s breeding value can be measured, and when a trait is measured in multiple hemiclones the additive genetic variation for that trait can be estimated.
4) Measuring net selection gradients. Since the population is cultured in a way that easily is mimicked under experimental settings, fitness of individual flies or genotypes can be measured. To estimate selection gradients acting on traits, fitness of each hemiclone is measured, which then is regressed on the variation of the focal trait.
Above method was used to measure additive genetic variation in mating rate in males (paper III) and females (Linder & Rice 2005), while the net selection gradient acting on this trait was measured separately for the sexes. Mating rate had an additive genetic component in both males and females. This variation was further uncorrelated between the sexes, suggesting that different genes code for these traits in males and females.
Mating rate was negatively selected in females, while positively selected in males (see fig. 5).
Figure 5. The distribution of average remating rate of 35 hemiclones when expressed in females (left) and males (right), and the selection gradients on this variation. The lightly stippled curves are normal distributions fit to the data, and they depict the phenotypic variation among hemiclones. The darker curves depict genetic variation: they are normal distributions centered at the sample means but with variance set equal to the estimated genetic variation among hemiclones for females and males, respectively.
The combined results of these two studies thus provide strong evidence for that there is sexual conflict over mating rate in this population and that this conflict drives a sexually antagonistic arms race. Mating rate is a complex trait and genes coding for this trait are likely to affect morphology, behaviour as well as physiology. Due to the complexity of the trait, and our evidence for that it is shaped by antagonistic coevolution, we think this finding provide strong support for the hypothesis that SAC can drive speciation.
Sexual conflict promotes speciation in insects (paper VII)
In this study we tested for an association between sexual conflict and the rate of speciation. To accomplish this we identified 25 phylogenetic contrasts. Each of these contrasts consisted of one clade of species where females predominately were classified as monandrous (mating only once) and one clade of species where females predominantly were classified as polyandrous (mating several times). As the potential for sexual conflict is higher when females mate frequently, compared to when females only mate once (see §1), we predicted that the rate of speciation would be higher among clades where females were polyandrous and that these clades thus would be more species rich.
The result from this test shows that polyandrous clades are significantly more species rich than monandrous clades. Comparing the averages between these two groups shows that polyandrous clades are about four times species rich as monandrous clades. We also tested if these results could have been explained by a number of other acknowledged factors affecting species richness, such as trophic ecology, range of geographic distribution and latitudinal distribution. Including these factors, in models testing for species richness, strengthened rather than weakened the positive association seen between the potential for sexual conflict and rate of speciation.
5.CONCLUSIONS
In this thesis I have studied sexual selection and speciation in the light of sexual conflict, using mathematical modelling, conducting empirical experiments with D.
melanogaster and using comparative methods. Overall I find strong evidence that conflicts within and between the sexes currently drive male-male and male-female antagonistic coevolution, and that these processes influence the rate of divergence between allopatric populations.
Indirect benefit models (the Fisher process and the good genes model) dominated research on sexual selection for several decades. These theories assume females to achieve genetic benefits through mating with certain males over others. Both these theories are however associated with difficulties. The good genes model is based on that females choose their partner after his genetic quality. But, as the heritability for fitness apparently is very low in nature, this seems like an unlikely process maintaining female choice. The problem with the Fisher process is yet more severe. For this model to work female mate choice cannot be costly, which seems like an unrealistic assumption. Models where females directly benefits from their mate choice therefore seem more plausible, especially if they have to compensate for costly mate choice. In systems where females receive nothing but sperm direct benefits are obviously absent.
However, direct effects of mate choice might still be present, since different mating strategies might be associated with costs of different magnitudes. From a female perspective sexual conflict is all about avoiding male induced costs.
In paper II in this thesis, females that were housed with males that provided high levels of courtship and induced higher mating rate suffered a lifetime offspring production reduction of 16 percent, compared to females that were housed with less “sexy” males.
In a study by Stewart et al (2005), an artificial female resistance allele was introduced into an experimental population. This artificial allele had the effect that males
“vanished” as soon as females had mated once. The effect of this new female resistance allele was tested by running this experimental population over several generations. On average females that expressed the allele had a 20 percent direct fitness advantage over females not expressing the allele, showing that direct cost by interacting with males are substantial. It thus seems like sexual conflict causes considerable harm to females and that females can avoid large direct costs by adopting a mate choice strategy that reduce male harm (i.e. resistance). As the evolution of female resistance is motivated by avoiding cost associated with mating, it by necessity also compensates for any costs associated with the process of female choice itself.
Sexual conflict seems like a particularly strong force driving male female coevolution, and good evidence for this is rapidly accumulating. However, evidence has yet only been sought for and found in a limited number of systems and studies on a wider set of taxa have to be conducted before we can conclude how general sexual conflict is in shaping male-female interactions, sexual dimorphism and divergence among populations.
6.ACKNOWLEDGEMENTS
It certainly helps to stand on the shoulders of giants, when to write a thesis. I have had the privilege to work with some of the most prominent researchers in my field of science during the course of my PhD-studies. Of these my supervisor, Göran Arnqvist, has played the most important role. Your expertise together with your ability and willingness to communicate your knowledge, have had a profound influence on this thesis and my personal development as a researcher. Thank you for all the help and support you have provided, especially during the first years. Sergey Gavrilets joined one of our projects at an early stage of my PhD training. Though this was only for a very short period of time, your skills and efficiency made me wish I was a theoretician.
Two years into my PhD-studies generous scholarships by the Sweden-American Foundation and STINT made it possible for me to visit Bill Rice’s lab for six months.
Thank you for inviting me Bill. I found those six months to be exceptionally stimulating and educational. I truly enjoyed working with you and look forward to do it again.
I also had to consort among dwarfs during the course of my PhD-studies. These, my other coauthors (Phil Byrne, Martin Edvardsson, Judy Linder, Tim Lew, Ted Morrow, Tina Nilsson and Andrew Stewart), I am sure, if not already, all will be considered giants themselves.
Coauthors and others have also influenced the outcome of this thesis in more indirect ways. Most valuable of these has Martin Edvardsson been. I found our many discussions on evolution, and everything else, during our undergraduates, travels and first graduate years most rewarding. Richard Dawkins, for your superb book “The selfish gene”, which introduced me to the truly exiting areas of evolutionary biology.
Locke Rowe, I wish you had chosen to have your sabbatical with us one or two years later, because then some of your genius might have rubbed off on me. At the time I was too scientifically immature to do anything but believe all your provocative statements, rather than seeing them as stimulating openings for scientific discussions – anyhow thanks for trying. Lars-Olov Rosenqvist, my gymnasium biology teacher. I cannot remember that you said a word about evolution but you made biology a highly pleasant subject. Thank you for your encouragement – it made a difference. Leve Linné! Ingrid Ahnesjö, Jens Andersson, José Andres, Johanna Arrendal, Sara Bergek, Anders Berglund, Johannes Bergsten, Mats Björklund – for providing space for me at the dept.
of Animal Ecology in Uppsala, Tomas Brodin, Phil Byrne – for much fun in the Rice lab and making me realize the thrill of surfing, Marnie Demandt, Damian Dowling – my fellow friend in the world of mitochondria and coauthor of the studies that did not fit into this thesis – thanks for putting up with me during a tough period of my life, for discussions, statistical advice and for sharing your good taste of music, Claudia Fricke - the kindest and most unselfish of us all – thanks for your friendship and all the help you have provided during these years– what would this thesis have been without you?
Ingela Ericsson, Åsa Eriksson, Barbara Giles – for your endless enthusiasm and support, Lars Gustafsson – for refreshing sarcasm, Åsa Hagner, Pelle Ingvarsson, Frank Johansson – for being such a cool guy and a role model where few are available, Mårten Hjernqvist, Björn Johansson, Mari Katvala, Niclas Kolm – for flattering words and being living evidence of that you can succeed also without a divine ancestry,
Natacha Kremer, Olivia Langhamer – for keeping track of all the breaks and cakes, or most of them at least…, Kjell Leonardsson – for interesting discussions and statistical advice, Tim Lew - for taking good care of a lost foreigner, Ylva Linghult, Jody Linder – for all the joy you spread, Maria Lundgren, Alexei Maklakov for friendship and endless scientific discussions, Sussi Mikaelsson – who nailed this thesis, the most helpful and efficient secretary who ever sat foot on this earth , Ted Morrow – friend of flies!, Anders Nilsson, Tina Nilsson - roommate of roommates, Lennart Persson, Anna Qvarnström, Åsa Rasmuson-Lestander - and the rest of the flygirls at the old genetics department in Umeå– thanks for giving me valuable support and help during my first fumbling tries with fruit flies, Emma Rova, Johanna Rönn my toughest critic (it was my birthday…), Sandra “mosquito food” South, Katherine Thuman, Jonas Victorsson, Chris Wiley – for always taking the discussion to a higher level, Jon Ågren – for inspiration, Nisse, Erik and the other boys in the basement – for all the practical help you have provided.
I would also like to take the opportunity to thank the funding bodies that have supported my PhD-work: Gustav och Hanna Winblads minnesfond för främjandet av zoologisk forskning, Helge Ax:son Johnsons stiftelse, J C Kempes Minnes Stipendiefond and Akademiska fond I, Magn. Bergwalls Stiftelse, Stiftelsen Lars Hiertas Minne, The Royal Swedish Academy of Sciences - P F Wahlbergs minnesfond and J A Ahlstrands testamentsfond, The Swedish Foundation for International Cooperation in Research and Higher Education, The Sweden-America Foundation and Wallenbergmedel -Umeå University.
There are also some people in the real world that I would like to thank. Anna and Ronny Falk, Ylva Friberg and Kristoffer Lindgren, Ingrid and Uno Friberg, Anna and Mattias Jonasson – for your good taste in food and wine and for often enough inviting us to take part of it, Jonas and Tove Kärvin, Stina and Staffan Lundstedt –for being the perfect family to share a holiday with – I am looking forward to the next, Kjell and Lisbeth Nyström – a dream for tired parents, Teresa Tiensuu and Patrik Ryden, Margaretha Willhelm and Håkan Hall, Aundrea Tavakkoly – for being the coolest housemate ever – keep that surfboard waxed – I’ll be back, Mats and Lisa Öström.
Last but not least I would like to thank my family. Josefin - not so much for your habit of always falling in deep dreamless sleep whenever I mention an especially important branch of science, but for all your love, care and support. What would life be without you? Alva and Nora - my very own clone, thanks for all the love and joy you spread, and for giving me daily evidence that the phenotype is so much more than the genotype.
7.REFERENCES
Andersson, M. 1982. Sexual selection, natural selection and quality advertisement. Biol. J. Linn. Soc. 17:
375-393.
_______ 1986. Evolution of condition-dependent sex ornaments and mating preferences: sexual selection based on viability differences. Evolution 40: 804-816.
_______ 1994. Sexual selection. Princeton, New Jersey: Princeton University Press.
Arnqvist, G. 1998. Comparative evidence for the divergence of genitalia by sexual selection. Nature 393:
784-786.
Arnqvist, G. & Nilsson, T. 2002. The evolution of polyandry: multiple mating and female fitness in insects. Anim. Behav. 60: 145-164.
Arnqvist. G., Edvardsson, M., Friberg, U. & Nilsson, T. 2000. Sexual conflict promotes speciation in insects. Proc. Nat. Acad. Sci. USA, 97: 10460-10464.
Basolo, A.L. 1996. The phylogenetic distribution of female preference. Sys. Biol. 45: 290–307.
_______ 1998. Evolutionary change in receiver bias: a comparison of female preference functions.
Proc. Roy. Soc. Lond. B 265:2223–2228.
Bateman, A.J. 1948. Intrasexual selection in Drosophila. Heredity 2: 349-368.
Bonduriansky, R. 2001. The evolution of male mate choice in insects: a synthesis of ideas and evidence. Biol. Rev. 76: 305-339.
Chapman, T., Trevitt, S. & Partridge, L. 1994. Remating and male derived nutrients in Drosophila melanogaster. J. Evol. Biol. 7: 51–69.
Chapman, T., Arnqvist, A., Bangham, J. & Rowe, L. 2003. Sexual conflict. Trends Ecol. Evol. 18: 41-47.
Chippendale, A.K., Gibson, J. & Rice. W.R. 2001. Negative genetic correlations for adult fitness between sexes reveals ontogenetic conflict in Drosophila. Proc. Nat. Acad. Sci. USA 98: 1671- 1675.
Choe, J.C. & Crespi, B.J. (eds) 1997. The evolution of mating systems in insects and arachnids.
Cambridge, Cambridge University Press.
Coyne, J.A. & Orr, H.A. 2004. Speciation. Sunderland, Sinauer Associates.
Daly, M. 1978. The cost of mating. Am. Nat. 112: 771-774.
Darwin, C. 1859. On the origin of species by means of natural selection. London, Murray.
_______ 1871. The descent of man, and selection in relation to sex. London, Murray.
Endler, J.A. 1986. Natural selection in the wild. Princeton, New Jersey: Princeton University Press.
Endler J.A. & Basolo, A.L. 1998. Sensory ecology, receiver biases and sexual selection. Trends Ecol.
Evol. 13: 415-420.
Ewing, A.W. 1961. Body size and courtship behaviour in Drosophila melanogaster. Anim. Behav. 9:
93–99.
Ferveur, J.F. 1997. The pheromonal role of cuticular hydrocarbons in Drosophila melanogaster.
BioEssays 19: 353-358.
Fowler, K. & Partridge, L. 1989. A cost of mating in female fruitflies. Nature 338: 760-761.
Gavrilets, S. 2000. Rapid evolution of reproductive barriers driven by sexual conflict. Nature 403: 886- 889.
Hall, J.C. 1994. The mating of a fly. Science 264: 1702-1714.
Grafen, A. 1990. Biological signals as handicaps. J. Theor. Biol. 144: 517-546).
Gustavsson, L. 1986. Lifetime reproductive success and heritabilities: empirical support for Fisher’s fundamental theorem. Am. Nat. 128: 761-764.
Heywood, J.S. 1989. Sexual selection by the handicap mechanism. Evolution 43: 1387-1397.
Holland, B. & Rice, W.R. 1998. Chase-away sexual selection: antagonistic seduction versus resistance.
Evolution 52:1-7.
Houle, D. & Rowe, L. 2003. Natural selection in a bottle. Am. Nat. 161: 50-67.
Iwasa, Y., Pomiankowski, A. & Nee, S. 1991. The evolution of costly mate preferences. II. The
“handicap” principle. Evolution 45: 1431-1442.
Kirkpatrick, M. 1982. Sexual selection and the evolution of female choice. Evolution 36: 1-12.
_______ 1985. Evolution of female choice and male parental investment in polygynous species: The demise of the “sexy son”. Am. Nat. 125: 788-810.
Kruuk, L.E.B., Clutton-Brock, T.H., Slate, J., Pemberton, J.M., Brotherstone, S. & Guinness, F.E. 2000.
Proc. Nat. Acad. Sci. USA 97: 698-703.
Lande, R. 1981. Models of speciation by sexual conflict on polygenic traits. Proc. Nat. Acad. Sci. USA 78: 3721-3725.
Lew, T.A., Morrow, E.H. & Rice, W.R. 2006. Standing genetic variance for female resistance to harm from males and its relationship to intralocus sexual conflict. Evolution 60: 97-105.
Linder, J.E. & Rice, W.R. 2005. Natural selection and genetic variation for female resistance to harm from males. J. Evol. Biol. 18: 568–575.
McClerry, R.H., Pettifor, R.A., Armbruster, P., Meyer, K., Scheldon, B.C. & Perrins, C.M. 2004.
Components of variance underlying fitness in natural populations of the great tit Parus major. Am.
Nat. 164: E62-E72.
Merilä, J. & Sheldon, B.C. 2000. Lifetime reproductive success and heritability in nature. Am. Nat. 155:
301-310.
Parker, G.A. 1979. Sexual selection and sexual conflict. In Sexual selection and sexual reproductive competition in insects. (eds. M. S. Blum & N. A. Blum), pp. 123-166. New York: Academic Press.
Partridge, L. & Farquhar, M. 1983. Lifetime mating success of male fruitflies (Drosophila melanogaster) is related to size. Anim. Behav. 31: 871–877.
Partridge, L., Ewing, A. & Chandler, A. 1987. Male size and mating success in Drosophila melanogaster: the roles of male and female behaviour. Anim. Behav. 35: 555–562.
Pitnick, S. 1991. Male size influences mate fecundity and remating interval in Drosophila melanogaster. Anim. Behav. 41: 735–745.
Pitnick, S. & García-González, F. 2002. Harm to females increases with male body size in Drosophila melanogaster. Proc. R. Soc. Lond. B 269: 1821–1828.
Pitnick, S., Spicer, G.S. & Markow, T.A. 1997. Phylogenetic examination of female incorporation of ejaculates in Drosophila. Evolution 51:833–845.
Pomiankowski. A. 1987a. Cost of choice in sexual selection. J. Theor. Biol. 128: 195-218.
_______ 1987b. The handicap principle does work – sometimes. Proc. Roy. Soc. Lond. B 231: 123-145.
Price, T., Schluter, D. & Heckman, N.E. 1993. Sexual selection when the female directly benefits. Biol. J.
Linn. Soc. 48: 187-211.
Rice, W.R. 1998. Intergenomic conflict, interlocus antagonistic coevolution, and the evolution of reproductive isolation. In Endless forms: species and speciation. (Eds. D. J. Howard & S. H.
Berlocher), pp. 261-270.
_______ 2000. Dangerous liaisons. Proc. Nat. Acad. Sci. USA 97: 12953-12955
Rice, W.R., Linder, J.E., Friberg, U., Lew, T.A., Morrow, E.H. and Stewart, A.D. 2005. Inter-locus antagonistic coevolution as an engine of speciation: Assessment with hemiclonal analysis. Proc.
Natl. Acad. Sci. USA 102: 6527-6534.
Rowe, L., Arnqvist, G., Sih, A. & Krupa, J.J. 1994. Sexual conflict and the evolutionary ecology of mating patterns: water striders as a model system. Trends Ecol. Evol. 9: 289-293.
Rowe, L., Cameron, E. & Day, T. 2005. Escalation, retreat, and female indifference as alternative outcomes of sexually antagonistic coevolution. Am. Nat. 165 (suppl.) S5-S18.
Ryan, M.J. 1990. Sexual selection, sensory systems, and sensory exploitation. Oxford surv. Evol. Biol. 7:
157-195.
Sgrò, C.M. & Partridge, L. 2000. Evolutionary response of the life history of wild-caught Drosophila melanogaster of two standard methods of laboratory culture. Am. Nat. 156: 341-353.
Singh, S.R. & Singh, B.N. 2004. Female remating in Drosophila: comparison of duration of copulation between first and second mating in six species. Current Science 86: 465-470.
Stewart, A.D., E.H. Morrow & W.R. Rice. 2005. Indirect benefits do not compensate direct costs of interlocus sexual conflict in a Drosophila model system. Proc. R. Soc. B 272: 2029-2035.
Tompkins, L. & Hall, J. C. 1981. The different effects on courtship of volatile compounds from mated and virgin Drosophila females. J. Insect Phys. 27: 17-21.
Trivers, L.R. 1972. Parental investment and sexual selection. In Sexual selection and the descent of man.
(ed. B. Campbell), pp. 136-179. Chicago, Aladine
Van Valen, L. 1973. A new evolutionary law. Evolutionary Theory 1: 1-30.
Wagner, W.E. 1998. Measuring female mating preferences. Anim. Behav. 55:1029–1042.
Wedell, N., Gage, M.J.D. & Parker, G.A. 2002. Sperm competition, male prudence and sperm-limited females. Trends Ecol. Evol. 17: 313-320.
West-Eberhard, M.J. 1979. Sexual selection, sexual competition, and evolution. Proc. Am. Philos. Soc.
123: 222-234.
_______ 1983. Sexual selection, social competition, and speciation. Quart. Rev. Biol. 58: 155-183.
_______ 1984. Sexual selection, competitive communication and species-specific signals in insects. In Insect communication. (ed. T. Lewis), pp. 283-342. New York: Acadmic Press.
Wigby, S. & Chapman, T. 2004. Female resistance to male harm evolves in response to manipulation of sexual conflict. Evolution 58:1028–1037.
Williams, G.C. 1966. Adaptation and natural selection: a critique of some current evolutionary thought.
Princeton, New Jersey: Princeton University Press.
Zahavi, A. 1975. Mate selection: selection for a handicap. J. Theor. Biol. 53: 205-214.
_______ 1977. Cost of honesty: (further remarks on the handicap principle). J. Theor. Biol. 67: 603-605
