Summary1. Differences in individual behaviour affect social interactions and contribute to the spatial structuring of animal populations. However, disturbance should also affect spatial networks by altering habitat heterogeneity and resource availability. Variation in resource availability should perturb the frequency and nature of social and ecological interactions within a population by affecting the spatial distribution of individuals. 2. In disturbed habitats where resources are limiting, spatial relationships should reflect behavioural differences among individuals, with higher-quality resources controlled by dominant individuals. In contrast, all individuals may exploit preferred resources in resource-rich habitats. Environmental variation and population reorganization may also result in variation in morphological, behavioural and ecological traits, which ultimately affect fitness. 3. We addressed these considerations for male tree lizards (Urosaurus ornatus) at three sites that differ in levels of disturbance. The habitats at these localities differed in the availability of live trees, the preferred microhabitat of U. ornatus. In addition, male U. ornatus exhibits a polymorphism in dewlap colour linked with differences in aggression, which should influence their position in a network and access to resources. We applied a network framework to characterize the spatial organization of male morphs at each site and quantified male aggressive behaviour in the laboratory. We also compared body size, body condition, number of bite marks, parasite load and the microhabitat use and diet, of males among the sites. 4. We detected no significant differences in spatial network structure between unburned and infrequently burned sites. However, at a frequently burned site, the network shifted towards geographically closer, heteromorphic male neighbour associations. Males at this site were also larger, more aggressive and had more bite marks but fewer parasites than males at the other sites. Moreover, we detected divergence in microhabitat use and diet among the morphs at the frequently burned site that reflected the shift in spatial network structure and differences in morph behaviour. That is, only more aggressive morphs usurped trees and consumed prey from higher trophic levels. 5. We conclude that environmental variation may influence animal spatial network structure. Jointly, behavioural and environmental variation may promote despotic social dynamics and ecological divergence in resource-limited habitats.

Discrete colour morphs associated with alternative mating tactics are assumed to be ecologically equivalent. Yet suites of behaviours linked with reproduction can also favour habitat segregation and exploitation of different prey among morphs. By contrast, trophic polymorphisms are usually attributed to morphs exhibiting habitat or prey selectivity. An alternative hypothesis is that habitat variation generates a trophic polymorphism driven by differences in morph reproductive behaviour, the spatial dispersion of morphs in a landscape and their exposure to different prey types. In this scenario, morphs are allowed to vary in habitat or diet selectivity (e.g. specialist or generalist) as they do in behaviour, rather than being assumed to exhibit equivalent levels of ecological specialization. We test this hypothesis using male Urosaurus ornatus lizards that exhibit a discrete dewlap colour polymorphism that reflects alternative mating tactics. We found blue morphs specialize on prey at higher trophic levels, yellow males display plasticity in trophic and morphological attributes and orange males are trophic generalists. Our results also demonstrate that morph diet differences are enhanced in resource-limited habitats. We conclude that discrete behavioural morphs may also diverge in morphology and trophic niche. Jointly, these processes may enhance speciation rates in colour polymorphic taxa.

Sexual size dimorphism (SSD) is often assumed to reflect the phenotypic consequences of differential selection operating on each sex. Species that exhibit SSD may also show intersexual differences in other traits, including field-active body temperatures, preferred temperatures, and locomotor performance. For these traits, differences may be correlated with differences in body size or reflect sex-specific trait optima. Male and female Yarrow's spiny lizards, Sceloporus jarrovii, in a population in southeastern Arizona exhibit a difference in body temperature that is unrelated to variation in body size. The observed sexual variation in body temperature may reflect divergence in thermal physiology between the sexes. To test this hypothesis, we measured the preferred body temperatures of male and female lizards when recently fed and fasted. We also estimated the thermal sensitivity of stamina at seven body temperatures. Variation in these traits provided an opportunity to determine whether body size or sex-specific variation unrelated to size shaped their thermal physiology. Female lizards, but not males, preferred a lower body temperature when fasted, and this pattern was unrelated to body size. Larger individuals exhibited greater stamina, but we detected no significant effect of sex on the shape or height of the thermal performance curves. The thermal preference of males and females in a thermal gradient exceeded the optimal temperature for performance in both sexes. Our findings suggest that differences in thermal physiology are both sex- and size-based and that peak performance at low body temperatures may be adaptive given the reproductive cycles of this viviparous species. We consider the implications of our findings for the persistence of S. jarrovii and other montane ectotherms in the face of climate warming.

Non-random female mating preferences may contribute to the maintenance of phenotypic variation in color polymorphic species. However, the effect of female preference depends on the types of male traits used as signals by receptive females. If preference signals derive from discrete male traits (i.e., morph-specific), female preferences may rapidly fix to a morph. However, female preference signals may also include condition-dependent male traits. In this scenario, female preference may differ depending on the social context (i.e., male morph availability). Male tree lizards (Urosaurus ornatus) exhibit a dewlap color polymorphism that covaries with mating behavior. Blue morph males are aggressive and defend territories, yellow males are less aggressive and defend smaller territories, and orange males are typically nomadic. Female U. ornatus are also polymorphic in dewlap color, but the covariation between dewlap color and female behavior is unknown. We performed an experiment to determine how female mate choice depends on the visual and chemical signals produced by males. We also tested whether female morphs differ in their preferences for these signals. Female preferences involved both male dewlap color and size of the ventral color patch. However, the female morphs responded to these signals differently and depended on the choice between the types of male morphs. Our experiment revealed that females may be capable of distinguishing among the male morphs using chemical signals alone. Yellow females exhibit preferences based on both chemical and visual signals, which may be a strategy to avoid ultra-dominant males. In contrast, orange females may prefer dominant males. We conclude that female U. ornatus morphs differ in mating behavior. Our findings also provide evidence for a chemical polymorphism among male lizards in femoral pore secretions.

Social dynamics in territorial species often reflect underlying variation in aggression and other aspects of social dominance among individuals. In ornate tree lizards (Urosaurus ornatus), males differing in dewlap color differ in social dominance: while blue males are the dominant, aggressive morph and always territorial, yellow males tend to exhibit a less‐aggressive satellite behavioral tactic. However, in habitats with fewer available territorial resources, yellow males defend territories and increase in relative abundance. These observations suggest that consideration of social dominance alone may be insufficient to explain U. ornatus' territorial dynamics in the wild. Here, we tested how both dominance and another important behavioral trait, boldness, contribute to the outcome of territorial disputes in tree lizards. We recorded the territorial behavior of blue and yellow male tree lizards (entered in pairs) in an experimental arena. At the end of each trial, we then approached each male and recorded whether it fled (shy) or not (bold) in response to our approach. As expected, dominant blue males exploited the higher quality perch more often than yellow males. However, when approached by a simulated predator, blue males were more likely to flee than yellow males. Thus, while blue males are more dominant, yellow is likely the bolder morph. As a result, this morph may be better equipped to defend territories in riskier environments. We conclude that although dominance asymmetries may predictably drive initial territorial interactions among competing males, variation in other behaviors (like boldness) may perturb the long‐term outcome of these interactions across variable environments.