This paper presents the first analysis of dental microwear textures of carnivorans. Carnassial microwear is examined for three large carnivorans, the cheetah Acinonyx jubatus, African lion Panthera leo and spotted hyaena Crocuta crocuta using dental microwear texture analysis, which combines confocal microscopy with the study of scale-sensitive fractal geometry. Results indicate significant differences in the microscopic wear textures of these carnivores consistent with dissimilarities in their reported feeding behaviours. Acinonyx jubatus carnassial shearing facets are characterized by low surface texture complexity and high anisotropy, while P. leo and C. crocuta evince less wear texture anisotropy and more complexity. Panthera leo and C. crocuta have more heavily pitted surfaces, a wider size range of wear features and scratches that vary in their orientations relative to the long axis of the carnassial blade. Further, C. crocuta is most variable in overall surface complexity and also has the highest average complexity values. These results are consistent with differences in bone consumption rates among the three species, wherein cheetahs typically avoid bone, lions triturate it on occasion and spotted hyaenas comminute it more often. Incidences of bone consumption in carnivores reflect degree and/or type of carcass utilization and can be used as a general guide for niche partitioning. Thus, the application of microwear analyses to carnivores can be used to interpret competition and niche position within a guild of fossil carnivores across space and through time.

Two types of mathematical models incorporating several types of competition between two species of African carnivores were constructed and analysed. The heuristic model focuses on population dynamics of both carnivores and their shared prey. The individual-based model focuses on the role of habitat heterogeneity and unique features of ecology of one of the predators, using the African wild dogs and hyaenas as a case study.

The study of lateralisation has taken several forms ranging from investigating morphological asymmetries to research on lateralised motor and perceptual functions with many studies successfully evidencing lateralisation in a variety of species. This study, featuring three species (olive baboons, rhesus macaques, and spotted hyaenas) investigated visual field biases with the aim of determining whether emotional valence underpins these biases whilst also considering the influence of a number of other factors such as emotional intensity, age, sex, rank, and, for the first time, oestrus cycles (olive baboons only). This study aimed to establish whether Campbell’s (1982) Right Hemisphere Hypothesis or Silberman & Weingartner’s (1986) Valence Hypothesis offered the more valid theory for the lateralisation of emotion by considering interactions across the full spectrum of emotion – a question the almost exclusive investigation of negatively affective scenarios by previous studies has been unable to answer. Furthermore, this study provided a new methodology for investigating behavioural lateralisation by suggesting that separating the visual spectrum into five fields (extreme left, mid left, centre, mid right and extreme right) allows a more accurate insight into the lateralisation of visual perception than the traditional hemifield model. Finally, a more conservative method is proposed for analysing behavioural data in future studies from this field and suggests that these methods provide a more accurate representation of the lateralisation of emotion than those previously employed. A population-level left side bias was found for the spotted hyaenas, thus providing the first evidence of significantly lateralised behaviour in a large carnivore and, for this species at least, lending some support to Campbell’s (1982) Right Hemisphere hypothesis but as population-level biases were not found for either of the other species it may be premature to suggest this support is unequivocal. Significant age effects were found in two species as adult olive baboons and spotted hyaenas were both found to express significant left side biases. Spotted hyaenas were also found to express significant left side biases for females, dominant individuals, high intensity interactions, and sexual valence interactions whilst olive baboons expressed a significant left side bias during negative valence behaviours but no significant lateral biases were found in any context for rhesus macaques. In olive baboons behaviours performed by males and those of a low intensity were found to occur more frequently in the mid and central visual fields and neutral valence behaviours were less occurrent in the extreme visual fields whilst in spotted hyaenas sexual, positive and negative valence behaviours were significantly less centralised than neutral valence behaviours. Non-oestrus adult female olive baboons were significantly more strongly lateralised than in-oestrus females, thus suggesting an influence of sex hormones upon lateralisation that may also have been apparent from the hyaena data, particularly regarding the significant lateral biases observed for females and dominant individuals. Finally, this thesis discusses a number of methodological issues that were encountered during this study and provides recommendations for future research in this field. Namely, this thesis provides an updated method for calculating laterality bias that is much more suitable for species with binocular vision and details a novel method of assessing visual field preferences by considering central and peripheral visual fields as separate entities. Furthermore, this thesis suggests that the weighted method designed and implemented for this study provides a much more accurate methodological foundation for analyses which avoids the caveats that may have affected previous research and thus provides a considerably more robust template that should be encouraged for any similar subsequent studies.

Intraguild competition and predation have been recognised as important ecological factors influencing the population dynamics of carnivores. The effects of these interactions are often asymmetrical due to a size-related dominancy hierarchy. However, it has been suggested that competitively subordinate carnivores can minimise the costs of predation and competition through spatial and temporal avoidance. Here I investigate the ecological and behavioural mechanisms by which cheetahs (Acinonyx jubatus) coexist with competitively stronger lions (Panthera leo) and spotted hyaenas (Crocuta crocuta). Fieldwork was carried out in the Okavango Delta, northern Botswana, between October 2008 and August 2011. A total of 20 Global Positioning System (GPS) radio-collars were fitted on all known cheetahs (n=6), lion prides (n=5) and spotted hyaena clans (n=6) in the study area (approx. 3 000 km²). Pre-programmed radio-collars recorded locations and activity continuously for each individual and these data were complemented with direct behavioural observations. Cheetah data were analysed with respect to the temporal and spatial likelihood of encountering lions and spotted hyaenas. Results suggest that the response to the risks posed by other predators is species-specific, habitat-specific and dependent on the immediacy of the risk. Resource partitioning was not the main mechanism for coexistence as cheetahs overlapped extensively with lions and spotted hyaenas in time, space and habitat use. Instead, cheetahs adjusted their spatial distribution in response to immediate risks or adapted their habitat use depending on their vulnerability (e.g. behaviours such as feeding or with differing levels of moonlight at night). In general, cheetah temporal and spatial distribution is a hierarchal process, firstly driven by resource acquisition and thereafter fine-tuned by predator avoidance. In addition, habitat heterogeneity seemed to be key in facilitating coexistence. Understanding the behavioural mechanisms that interacting apex predators adopt to regulate these negative interactions could be crucial to carnivore conservation, especially as human-related habitat loss is forcing species into ever smaller areas.