Abstract: Females often aggregate at particular sites for feeding or shelter, thus giving adult males the opportunity to defend harems and increase male reproductive success. Rival males compete for control of harems via ritualized displays or direct combat using weaponry. Contests for harems or the resources required by females can be settled based on asymmetries in fighting ability or resource ownership. Males that accurately assess a rival's fighting ability prior to engaging in potentially costly combat should be favoured by selection. Game theory and optimality models provide three models to explain how individuals decide to persist in or flee from a fight. These models are the energetic war of attrition, the sequential assessment model and the cumulative assessment model. Using staged contests in the laboratory, I tested predictions of these models using the Wellington tree weta, Hemideina crassidens, a sexually dimorphic insect native to New Zealand. Male H. crassidens use their enlarged mandibles as weapons in fights for access to adult females that reside in cavities in trees. My results supported a prediction common to each assessment model: contest duration was negatively correlated with the asymmetry in opponent's weapon size. The sequential assessment model of contest settlement was partially supported but the strongest support was for the cumulative assessment model. Predictions of the latter model were supported because: (1) fights are apparently settled based on own-size assessment; (2) fights occur in a single phase and escalate; and (3) contests involve physical combat and injury. I suggest that, in nocturnal species, cumulative assessment will generally be most applicable.

Abstract: The social complexity hypothesis asserts that animals living in large social groups should display enhanced cognitive abilities along predictable dimensions. To test this concept, we compared highly social pinyon jays,Gymnorhinus cyanocephalus , with relatively nonsocial western scrub-jays, Aphelocoma californica, on two complex cognitive tasks relevant to the ability to track and assess social relationships. Pinyon jays learned to track multiple dyadic relationships more rapidly and more accurately than scrub-jays and appeared to display a more robust and accurate mechanism of transitive inference. These results provide a clear demonstration of the association between social complexity and cognition in animals. Copyright 2003 Published by Elsevier Science Ltd on behalf of The Association for the Study of Animal Behaviour.

Abstract: Methods of ranking individuals in a dominance hierarchy that use transitivity of relationships may obscure irregularities. Furthermore, these methods use only a small proportion of the information available from dominance encounters. This paper presents an intuitively appealing and easily implemented alternative to existing methods for ordering dominance data, developed from the work of Batchelder et al. (1992Journal of Mathematical Psychology36, 185-212). The procedure presented here is based on a mathematical model of paired comparisons and it involves only simple estimation procedures. We illustrate its use with data on dominance among red deerCervus elaphus, stags. The results indicate that dominance relationships are well characterized by the scale values that the model provides, and, because the method provides predictions for all pairings of animals, dominance predictions also exist for pairs of animals that have yet to be observed. Moreover, the dominance outcomes predicted by the model using the order scale are highly correlated with actual dominance observations at all levels. Overall, the procedure described provides a solution to the problem of identifying an appropriate order for a near-linear dominance hierarchy.

Abstract: A study of a herd of Camargue horsesEquus caballus, showed that while the majority of high-ranking stallions held single-male harems, some sons of low-ranking mares, being low ranking themselves, formed alliances that could last a lifetime. The two stallions were each other's closest associate and preferential grooming partner. Alliances were based on coalitions in which either both partners confronted an intruder synchronously or the dominant of the pair tended the female(s) while the subordinate simultaneously displayed towards the rival. Alliance partners were of similar age but were not more closely related to each other than to other stallions in the herd. Long-term paternity data revealed that subordinates sired close to a quarter of the foals born into the alliance group, and significantly more foals than low-ranking stallions in the herd adopting a `sneak'-mating strategy. The dominant appeared to benefit from the presence of his subordinate partner. Fights occurred all year round, and the subordinate stallion of each alliance pair fought outside competitors more than twice as often as the dominant. Forming short-term alliances before defending mares on their own may enhance long-term reproductive success for both partners. Other benefits to both partners include higher survivorship of their foals and increased access to proven reproductive mares. These results suggest that the relationship between alliance partners is based on mutualism, but several conditions for reciprocity seem to be fulfilled: the benefit to the dominant (assistance in fights), and the benefit to the subordinate (access to reproduction), are both costly to the other partner and delayed in time.

Abstract: We used comparative data to test functional hypotheses for 17 antipredator behaviour patterns in artiodactyls. We examined the literature for hypotheses about auditory and visual signals, defensive behaviour and group-related antipredator behaviour in this taxon and derived a series of predictions for each hypothesis. Next, we documented occurrences of these behaviour patterns and morphological, ecological and behavioural variables for 200 species and coded them in binary format. We then pitted presence of an antipredator behaviour against presence of an independent variable for cervids, bovids and all artiodactyls together using nonparametric tests. Finally, we reanalysed the data using Maddison's (1990, Evolution, 44, 539-557) concentrated-changes tests and a consensus molecular and taxonomic phylogeny. We found evidence that snorting is both a warning signal to conspecifics and a pursuit-deterrent signal, lack of evidence that whistling alerts conspecifics and indications that foot stamping is a visual signal to warn group members. Evidence suggested that tail flagging was a signal to both conspecifics and predators, that bounding, leaping and stotting were used both as a signal and to clear obstacles and that prancing functioned similarly to foot stamping. Analyses of tail flicking, zigzagging and tacking were equivocal. We confirmed that inspection occurs in large groups, freezing enhances crypticity, and species seeking refuge in cliffs tend to be small. Entering water and attacks on predators had few correlates. Finally, group living, a putative antipredator adaptation, was associated with large body size and species living in open habitats, confirming Jarman's (1974, Behaviour, 48, 215-267) classic hypothesis. Bunching and group attack apparently deter predators. Despite limitations, comparative and systematic analyses can bolster adaptive hypotheses and raise new functional explanations for antipredator behaviour patterns in general.