Free roaming domestic cats (Felis silvestris catus) are widespread around the world. Whether they are owned or not, their presence and management raises strong opinions between supporters of “cat independence”, advocates of “cat safety” and defenders of wildlife, leading to suggestions that cats should have their freedom to roam outdoors restricted. The aim of this thesis was to get a precise understanding of the concepts of free roaming and containment in cats, their meaning for the relationship between cats and the society that they are living in, both from the cats’ ‘point of view’ and the perceptions of their owners, and the impact on cat welfare. Using a variety of approaches including a survey, GPS tracking, behavioural tests and an owner-based questionnaire, we showed that although many people’s perceptions about free roaming cats depended on their residential location and their ownership status, the risk of road traffic accidents was rated highly as a concern by people regardless of their residential location or cat ownership status. This perception was matched by the behaviour of free roaming cats themselves, who, regardless of their home location were observed to frequently engage in risky behaviour like frequent road crossing. This represents a high risk to their wellbeing and, combined with the other risks posed to cats that go outside, makes a clear case for the need to restrict the cat’s roaming behaviour, in order to protect its welfare. Given the problems of an indoor only lifestyle, which may perhaps be greater than is widely recognised, there is a case for examining the impact on cat welfare of an effective containment system, that restricts the cat to the boundaries of its owner’s property (both minimising risk to the cat and disruption to the community). Options for containment are limited, and little is currently known about their impact. Therefore I examined the impact of an electronic containment system that is widely thought to be effective but that also causes concern regarding its effect on cat welfare. This was the first research on such a system in cats, and I sought to establish if the welfare concerns were justified and outweighed by the potential benefits. First, I gathered initial information with a case series, and then I carried out a more comprehensive study with a larger sample of cats. The initial case series failed to establish clear evidence of a consistent negative impact on welfare. In the larger follow-on study, for the population studied, I found no evidence consistent with a long term negative impact of the electronic containment system on cat behaviour and welfare. Indeed, cats contained by the system interacted more with people, were more curious about novel objects and were no more sensitive to sudden noise than control cats, although whether these effects were due to the system or selection bias in the volunteers remains unknown. viii Although many research questions have been addressed, unanswered questions remain, and directions for future research have been identified and are discussed.

Arousal and distress are considered important factors when dogs show problematic behaviours and the crucial role of hormones and neurotransmitters involved in stress responses is widely recognized. Corticosteroids are important players in stress responses and, along with other hormones and neurotransmitters, contribute to the onset of both physiological and behavioural changes that can be either normal and adaptive or excessive and maladaptive depending on several factors. A literature review revealed that exogenous corticosteroids have been reported to change behaviour in human beings and laboratory animals but no data were available as far as similar e�ffects in dogs. The aim of this research was to identify possible behavioural changes in dogs treated with corticosteroids. In the fi�rst study, the perception of behavioural changes in dogs during corticosteroid therapy was investigated through semi-structured open interviews of the owners of 31 dogs of di�fferent breeds, genders, and ages. All dogs had received corticosteroid therapies in the past six months. Owners were asked to describe their dog's behaviours both on and o� corticosteroid therapy. Eleven owners reported behavioural changes in their dogs; nine dogs were reported to show more than one behavioural change. Six dogs reportedly showed nervousness and/or restlessness, three showed an increase in startle responses, three showed food guarding, two showed a decrease in their activity level, three showed an increase in avoidance responses, four showed irritable aggression, and two dogs increased barking. Semi-structured interviews can be useful preliminary tools for the identifi�cation of areas of future investigation, and the outcomes of the interviews were then used to investigate more rigorously the possible relationship between these signs and corticosteroid use in dogs. In the second study 99 dog owners were asked to complete a 12 item questionnaire, developed following the results of the previous survey. Owners were asked to evaluate their dogs' behaviour on and off� therapy, using a seven point scale. A sample of owners whose dogs were receiving treatment for dermatological, orthopaedic or other conditions completed the survey. The survey was completed by 44 dog owners with animals receiving treatment with a range of corticosteroid preparations (mainly prednisolone and methylprednisolone) and 54 dog owners with dogs receiving treatment with other drugs, mainly antibiotics and non-steroidal anti-infl�ammatory drugs. Dogs under corticosteroid treatment were reported to be signifi�cantly less playful, more nervous/restless, more fearful/less confi�dent, more aggressive in the presence of food, more prone to barking, more prone to startle, more prone to reacting aggressively when disturbed, and more prone to avoiding people or unusual situations. The last part of this study involved behavioural tests of dogs. Eleven �treatment� dogs were then tested twice: before and during corticosteroid treatment with either methyl-prednisolone or prednisolone to assess their sensitivity to a potentially aversive stimulus. Eleven control dogs, not receiving corticosteroid therapy, were also tested at the same time intervals in the same environment. Dogs were exposed to a brief dog growl while they explored some bowls containing food and their behaviour was video recorded. Treatment dogs investigated the area for signi�ficantly less time and ate signi�cantly less food in the second test trial when on corticosteroid compared to control dogs. In �final study, exploring relationships between corticosteroid therapy and dogs with behaviour problems, a review of the caseload of the author of 345 dogs reported for behaviour and management problems was analyzed. It was found that 16 % of them had a history of previous treatments with corticosteroids. Previous treatment with corticosteroid was found to be signi�ficantly associated with negative aff�ective states. These results support earlier preliminary �findings concerning possible adverse behavioural side e�ffects following the use of corticosteroids in dogs, and the possible need for concomitant behavioural advice when these drugs are used in general veterinary practice.

The study of cooperation has been crucial to research on the evolution of social living in human and animal societies. Grooming interactions have been used as model to investigate the exchange of services in animals. Using both established and novel methodologies, this thesis examines grooming interactions and cooperation in two populations of wild Barbary macaques living in the Middle Atlas Mountains of Morocco. It is important to have a comprehensive idea of the costs and benefits of grooming interactions, and of the effect of grooming interactions on the anxiety of the grooming partners. This thesis showed that, contrary to previous studies, anxiety increased after grooming interactions in both the donor and recipient. This highlights the need to further investigate the link between grooming and emotions. Individuals may also affect the grooming interactions of other group members. This thesis showed that individuals benefit from disrupting grooming interactions of group members by gaining grooming opportunities for themselves and by stopping the group members from grooming each other, although grooming disruptions may be risky. Monkeys may affect others’ grooming interactions to favour their own social and dominance positions. A key aspect of this thesis was also to assess whether grooming is reciprocated in the short-term and which type of reciprocity (i.e. direct, indirect and generalised) play a role in the exchanges of grooming. This study showed that direct but not indirect and generalised reciprocity play a role in the exchange of grooming. While there is a wide range of evidence that direct reciprocity plays a role in the exchange of services in animals, there is little evidence of indirect and generalised reciprocity. Additionally to exchanging grooming for grooming, animals also exchange grooming for other services such as tolerance around food resource and support during agonistic interactions. In this thesis, no evidence of short-term contingency between the exchange of grooming and food tolerance was found. The exchanges of services may be little affected by recent single events, and mechanisms involving an emotional mediation based on long-term social bonds between partners may play a more important role. The capacity to make effective choices among potential social partners is an important social skill, as choosing the best available partner improves the chances to establish successful cooperative interactions. This thesis highlighted, to some extents, the importance of factors such as tolerance and relationship quality between partners, in the performances of individuals and their choice of partners to solve a cooperative task. Tolerant relationships may have been a prerequisite for the evolution of cognitively complex cooperation. Testing a comprehensive framework of predictions, this thesis brings novel contributions to the understanding of grooming interactions and cooperation in wild Barbary macaques.

This thesis is based on the evidence of the use of range by laying hens from two separate studies that were carried out on commercial flocks. The first study involved a direct observation of 6 flocks of laying hens whereas the second study was carried out on a single flock of laying hens using a still digital camera. Mapping of the outdoor range into three zones, namely; apron (0-10metres), enriched (10-50metres) and range (beyond 50metres), was carried out in both studies determine whether it influenced the distribution, behaviour, nearest neighbour distance (NND) and feather condition of the hens. Based on the estimates of the total number of hens recorded outdoors, it was found that 14.5% of the hens used the range in the first study, with 6.0% of the hens seen in the apron, 4.8% in the enriched and the remaining 3.7% in the outer range zone. This ranging pattern was further confirmed by the quadrat head counts, where the hens used the apron most but showed a significant decrease in the use of other outdoor zones as distance increased from the shed (p < 0.001). The results of the two studies showed that range use peaked in the morning and decreased significantly in the afternoon (p < 0.001). Range use was found to increase significantly with age (p < 0.001) in the first study, although there was a significant decrease in the number of hens outside the shed as the flocks got larger (p < 0.001). The age effect reported in the six-flock study was complemented by the significant positive effect of the week of outdoor access on the ability of the hens to use the range (p < 0.001) in the single flock study. Range use increased as outdoor temperature rose in the six-flock study whereas increased outdoor temperature resulted in a decline in range use in the single flock study. NND of the hens was ii measured in the six-flock study and was found to decrease significantly towards the shed (p < 0.001) and as the hens aged (p < 0.001). There was evidence that the use of distant and less crowded outdoor locations by the hens in the first study was associated with improved feather conditions. The evidence presented in the two studies presented in this thesis showed that the hens were unevenly distributed in the range, with reduced hen density (hen/m2) towards the outer range and that an improvement in the use of distant parts of the range may have beneficial effects on the hens.