Monthly Archives: July 2014

Emerging issues in the evolution of animal nuptial gifts

This week Biology Letters published a short opinion piece by Sara Lewis and colleagues (including me) on the state of play for research on nuptial gifts in animals. Substances transferred from one mating partner to another are not limited to gametes, and this piece tries to clarify the conceptual significance of this behaviour and stimulate productive future research by carefully defining what a nuptial gift is and clarifying  different categories of gifts on the basis of their source and transmission mode. The piece is described for a more popular audience here.

Screen Shot 2014-07-19 at 10.35.14

Graduation 2014

One of the best and worst parts of supervision is the mixed feeling of fledging a crop of graduates. This is definitely a happy occasion, and one worth marking. It represents the culmination of a lot of hard and industrious work. And while the collaborations will continue (we need to publish most of this work!), it’s still bittersweet to have so many great folks leave the lab. Best of luck to this year’s graduates! Please stay in touch regularly….

The 2014 graduating crew, in fancy dress, from left to right: Claudia Santori, Gregor Hogg, Toby Hector, Luc Bussière, Tom Houslay, & Sam Paterson. Photo by Miles Houslay.

The 2014 graduating crew, in fancy dress, from left to right: Claudia Santori, Gregor Hogg, Toby Hector, Luc Bussière, Tom Houslay, & Sam Paterson. Photo by Miles Houslay.

Fritzsche & Arnqvist 2013: Homage to Bateman: sex roles predict sex differences in sexual selection

(NB: In spite of the authorship flag above, the section below is Claudia’s work, not mine! I am posting it on her behalf because she’s off somewhere sunny scuba diving, I think…. Luc)

I am synthesizing one last paper to conclude our “Journal Pub” session – I apologise it took so long! I read Karoline Fritzsche and Göran Arnqvist‘s paper “Homage to Bateman: sex roles predict sex differences in sexual selection”, published in Evolution in 2013.

In this paper the authors review the classic sex-role theory, which assumes sexual selection to be stronger in males in taxa with conventional sex roles (systems where the females are the choosy sex and contests for mates are mainly between males) and in females in systems with reversed sex roles. Little empirical work has been done to assess the relative strength of sexual selection on males and females, and no previous study has been able to provide measures in order to compare the strength of this process between sexes and between different taxa. For example, research is often limited to comparing sexual dimorphic traits between the two sexes of a species to evaluate the asymmetry of selection between males and females. The challenges of this type of study are reinforced by the debate regarding what type of measure should be used to quantify the strength of sexual selection (Fitze and Le Galliard 2011). To address these knowledge and evidence gaps, Fritzsche and Arnqvist aim to (i) clarify whether there is a single measure that is best for quantifying sexual selection, (ii) determine if it is best to base measures of sexual selection on phenotypic traits or variance across individuals, and finally (iii) assess how important is it to quantify the strength of sexual selection in both males and females.

The authors discuss these issues by presenting an experiment in which they observed the mating behaviour and reproductive success of both sexes of four seed beetle species. They used four related species of beetles: two of the Callosobruchus genus, females of which are the choosy sex, and two of the Megabruchidius genus, where females often compete against each other for mates.  Fritzsche and Arnqvist then attempted to compare the relative strength of sexual selection between the two sexes, and test the validity of various measures such as variance-based measures – measures unrelated to differences between morphological traits, like Bateman gradients (which they also refer to as sexual selection gradients: these are estimates of the slope of a regression line of reproductive success on mating success, and therefore describe how much fitness each individual gains per successful mating) and the opportunity for sexual selection (variance in reproductive success). Other measures analysed by the authors were trait-based measures like the selection differential (covariance between a standardized trait and fitness), the mating differential (covariance between a standardized trait and mating success), and the residual selection differential (covariance between a standardized trait and the residual reproductive success calculated from the Bateman gradient).

The authors used body size as the trait against which to compute selection. They made this choice because body size tends to covary positively with mating success, and also because it is a comprehensive measure that reflects both phenotypic and genetic variation. Furthermore, the conspicuous sexual dimorphism in the seed beetles strongly suggests a history of sexual selection on this trait. In their mating trials, they presented five females to five males, four of which were sterile. Therefore, each mating assay yielded data on mating and reproductive success for all the females and only one male. The authors quantified mating success by observing how many times each individual was successful at mating, female reproductive success by counting how many offspring each of them produced, and male reproductive success by summing all the offspring produced by the five females in his assay group (which are necessarily the offspring of the lone fertile male).

The Bateman gradients were identified as the best measure to quantify the strength of sexual selection for comparisons between sexes and/or across species. This study’s results support the idea that variance-based measures of sexual selection are very accurate at representing the sexual dimorphism (behavioural and morphological) of these species of beetles. Indeed, Bateman gradients show both male and female mating adaptations, since changes in adaptations are depicted by changes in gradients, representing the properties of the specific mating system. Bateman gradients were identified as the most relevant because they were very good predictors of sexual dimorphism of both secondary sexual traits and mating behaviour across the different species. Bateman gradients were generally steeper in males than in females, and steeper in species with species where males are the choosy sex (Figure 1A). In Callosobruchus species sexual dimorphism was more pronounced, as were the male and female Bateman gradients. In contrast, the opportunity for sexual selection was larger in males of Callosobruchus species, in which females are choosy, but similar or stronger in females of species where the males are choosy (Figure 1B).




Figure 1. The different Bateman values and opportunity for sexual selection for the different species of seed beetles. CM = C. maculatus; CC = C. chinesis; MD = M. dorsalis; MT = M. tonkineus; white bars represent males, black bars represent females. SRR+ indicates sex-role-reversed species and SRR− denotes species with conventional sex roles. Error bars = SE (Figure by Fritsche and Arnqvist 2013).

The authors’ results are consistent with the classic sex role theory, as the calculated Bateman gradients support that the strength of sexual selection is greater in males than in females. The strength of sexual selection varied with mating system. Sex role reversal is often associated with male provision of nutritious gifts to females, increasing female direct benefits with mating success and thereby increasing the female Bateman gradient in such systems. Indeed, male Megabruchidius provide nutritious ejaculates to females, which directly increase the number of eggs a female lays (Takakura 1999). In contrast, there is no conspicuous benefit to additional matings beyond the first one among female Callosobruchus (Arnqvist et al. 2005).

The authors argue that, in the absence of a trade-off between male nuptial gift provisioning and mating success, increased male nuptial gift-giving should increase the Bateman gradients not only of females but also of males. As expected, in this study the Bateman gradients of gift-giving males (Megabruchidius spp.) were the highest. Moreover, even though one might predict that in the Megabruchidius species, female Bateman gradients would be steeper than those of males (because of the sex-role-reversal), this does not necessarily need to be the case. Indeed, stronger sexual selection on females requires a male trade-off between mating success and pre-copulatory fertilization success, which is likely to occur under low resource levels due to the costliness of mating and of providing a nuptial gift. However, this trade-off could have been affected by the experimental design, because the provision of ad libitum food may have provided the male beetles with greater energy levels that they would have in nature. Also, the study males varied in resource acquisition, a fact that could have further weakened the potential trade-off, or perhaps made it harder to detect experimentally. Despite these issues, the higher sexual selection opportunity in female Megabruchidius (Figure 1B) conforms to the sex-role-reversed system.

The authors discuss the importance of post-copulatory sexual selection, which is mediated by female cryptic choice and/or sperm competition. This type of selection is an important component of the sexual selection acting on seed beetles. Residual selection is a measure of the “strength of selection on a trait due to factors other than mating success”, and is calculated as the covariance between a standardized trait and the residual reproductive success obtained from the Bateman gradient. Residual selection has been regarded as a measure of post-copulatory sexual selection and/or fecundity selection based on a trait (z)under sexual selection (in the case of this study, body size), including components from both natural (e.g. fecundity) and sexual selection (e.g. sperm competition). In males, this measure represents the covariance of z with both fecundity and success in sperm competition. Instead, female residual selection represents only the covariance of z and female fecundity.

Fritzsche and Arnqvist conclude their paper by highlighting how the data they gathered, and the measures of strength of sexual selection they calculated, were consistent with the sexual dimorphisms in behaviour and morphology in the studied species of seed beetles. Variance-based measures appeared to be more accurate at representing the sexual dimorphisms observed across the different beetle species. Bateman gradients in particular were the most informative measure of the strength of sexual selection, allowing comparisons between sexes and across species. However, the authors underline the importance of gathering data from both pre- and post-copulatory reproductive competition to provide deeper knowledge on how variation in the strength of sexual selection within and across species affects mating systems.



Reviewed paper: Fritzsche, K., Arnqvist, G., 2013. Homage to Bateman: sex roles predict sex differences in sexual selection. Evolution67(7): 1926–1936.

Arnqvist, G., Nilsson, T., Katvala, M., 2005. Mating rate and fitness in female bean weevils. Behavioural Ecology16: 123–127.

Fitze, P. S., Le Galliard, J. F., 2011. Inconsistency between different measures of sexual selection. American Naturalist178: 256–268.

Takakura, K., 1999. Active female courtship behavior and male nutritional contribution to female fecundity in Bruchidius dorsalis (Fahraeus) (Coleoptera : Bruchidae). Researches on Population Ecology41: 269–273.