Causes and consequences of sexual dimorphism in mandrill (Mandrillus sphinx) feeding and spatial behavior

Abstract

Sexual dimorphism is a widespread phenomenon, the origins of which have long intrigued evolutionary ecologists. Classic evolutionary theory attributes the evolution of sexual dimorphism to sexual and fecundity selection, that arise due to sex differences in optimal reproductive strategy. However, cross-species comparative analyses have found mixed support for sexual and fecundity selection as drivers of dimorphism. Other hypotheses may therefore be required to explain the full diversity of sexual dimorphisms found in nature. One alternative is the evolution of sexual dimorphism via ecological causation. This thesis explores the possibility of ecological causation of sexual dimorphism, with a particular focus on mandrills (Mandrillus sphinx). Mandrills are the most sexually dimorphic primate, with males over three times larger than females, possessing 45 mm canines and displaying extravagant red, blue, and violet ornamentation on their faces and rumps. Males also abandon social groups outside of the breeding season, foraging and travelling separately to females, which could suggest an ecological component to mandrill dimorphism. One of the key predictions of the ecological causation hypothesis is that greater sexual dimorphism should be associated with greater ecological divergence between males and females. In Chapter 2, I conducted a meta-analysis of 163 species to examine associations between sexual dimorphism and ecological sex differences, measured using stable isotopes. Across species, greater size dimorphism was associated with greater sex differences in trophic level. Chapter 2 thus supports a key prediction of the ecological causation hypothesis and suggests that the strength of relationships between sexual dimorphism and ecology may be underappreciated. Mandrills are challenging to study in the wild, due to the dense forests in which they are found. I therefore applied research tools that avoided the need to directly observe individuals, to study the socioecology of males and females and investigate the possibility of ecological causation of mandrill dimorphism. In Chapter 3, I analyze the contents of 4024 mandrill fecal samples, collected over an eight-year period, in conjunction with eight years of fruiting phenology data. I find that mandrills are primarily frugivorous, with consumption of fruit tracking phenological patterns of fruit production. The focal group also demonstrated high dietary flexibility, consuming animal prey, leaves, crushed seeds, and other plant fibers when fruit availability was low. However, because male mandrills spend much of year solitary, the feeding niche information gained from Chapter 3 largely relates to females. To investigate sex differences in feeding niche, I therefore applied stable isotope analysis to mandrill hairs. In Chapter 4, I detail the results of a diet-switch experiment, that quantified mandrill hair growth rates, isotopic tissue turnover times and isotopic tissue-diet discrimination factors. This experiment showed that stable isotope analysis of mandrill hairs can reveal dietary information on wild mandrills at a temporal resolution of 5.5 days and allowed the stable isotope ratios of wild mandrills to be correctly adjusted for accurate dietary inference. In Chapter 5 I applied stable isotope analysis to the hairs of wild male and female mandrills. I found that the diets of individual male mandrills were more consistent than individual females, who instead showed more dietary variation, at the individual level. This result suggests that individual male mandrills had more consistent access to preferred resources than individual females, and therefore that males experience differing levels, or outcomes, of resource competition. Males may therefore seasonally leave mandrill social groups to avoid resource competition. In Chapter 6, I used GPS collars to examine sex differences in mandrill spatial behavior. I found that, during the breeding season, male habitat selection, home-range size and Brownian motion variance was similar to females, but distinct from females during the non-breeding season. These results indicate that collared males were not always within the social group during the non-breeding season. Furthermore, during the non-breeding season, males travelled shorter distances, at slower speeds, compared to the breeding season. Males therefore appeared to expend less energy on travel during the non-breeding season, suggesting that male mandrills may also leave groups to reduce energetic expenditure. The results of Chapters 5 and 6 suggest that ecological factors compel male mandrills to abandon social groups outside of the breeding season. However, the intensity of sexual selection on male traits may be increased as a result of this behavior, as males must establish dominance and attract unfamiliar females when they re-join social groups to breed. An interaction between sexual and ecologically mediated selection may therefore offer the most complete explanation of why mandrill dimorphism is so extreme.