Abstract: The question “Why does an animal behave as it does?” can be answered in terms of ontogeny, function, phylogeny and causation. The achievements of applied ethology relative to those four approaches are reviewed, gaps in our knowledge are identified and predictions for fruitful avenues of future research are made. Ontogenic studies have been useful in the past and it is suggested that studies of the effects of early experience on the sexual behaviour of animals used in artificial breeding schemes might pay dividends. It is proposed that functional studies should be approached cautiously. More information is required on the process of domestication in order to increase the chances of success in the trend to farm exotic species. Studies on causation are likely to continue to be the mainstay of applied ethological research. It is suggested that within this category, studies on states of suffering, motivation and cognition are urgently required to answer the most pressing questions on animal welfare.

Abstract: Abstract
In response to a request published in Equus, a magazine for those interested in horses, 85 owners of older geldings exhibiting sexual behavior completed history forms. The mean age of geldings was 16 f 5 years. Only 39 of the owners had had the gelding for at least a year before the behavior was noted. These cases could be used to determine the true age of onset of the problem. When log survivorship was used to determine whether there were one or two different populations, a break or change in the slope at age 16 indicated that there are two populations. One population shows the behavior from the time of castration and the other first exhibits the behavior in old age, possibly in response to an ACTH secreting pituitary adenoma. A total of 40% of the horses were Quarterhorses, the most numerous breed in the US; 78% of the horses were purebreds. Fewer than half the owners knew the age at which their horse had been castrated because they did not own the horse at the time.
The mean age at castration, when known, was 3.3 f 2.5 years. The reason for contacting us was sexual behavior (70%), aggression (24%). or some other problem ( 1 o/o). Whether or not aggression was the presenting problem, most of the horses showed aggression (95%), particularly towards other geldings (88%)) but also towards people (3 1%). Copulatory behavior (mounting) was shown by 69% of the geldings and half of those were able to intromit. These findings indicate that the sexual behavior of geldings is a problem for owners and that aggression usually accompanies sexual behavior.
The owners were encouraged to send serum samples taken before and after human chorionic gonadotropin (HCG) administration for testosterone and estrone sulfate analysis to determine whether residual testicular tissue was responsible for the horse’s behavior. Of the 14 horses tested, only one had elevated levels of testosterone indicating that there was residual testicular tissue. A total of six of the owners agreed to treat their horses with cyproheptadine at a dose of 8 mg day- ’ gradually increased to 88 mg day- ’ per horse. A total of three of the horses showed a decline in sexual and aggressive behavior, one got worse and two had side effects and treatment was withdrawn.

Abstract: Stress in humans has been reported to be associated with a decrease in the salivary immunoglobulin A (s-IgA) levels enabling the possible use of s-IgA to assess stress. Prolonged stress, if reliably assessed in a non-invasive manner, may be used to assess animal welfare. This study analysed groups of dogs undergoing physical and temperamental training and s-IgA levels were measured by rocket immunoelectrophoresis in prospective samples. Behavioural assessment was carried out and cortisol levels in saliva were measured by ELISA. A significant negative correlation (P < 0.007) between the logarithmic cortisol concentrations and s-IgA levels in saliva was recorded. The behavioural assessment of the dogs agreed well with the biochemical markers. It is concluded that IgA levels in saliva may be a useful marker of dog well-being and that stress results in decreased s-IgA levels.

Abstract: Non-invasive faecal oestrogen and progesterone metabolite evaluations are well established approaches for monitoring reproductive function in a variety of mammalian species. The route of excretion of steroid hormone metabolites varies considerably among species, and also between steroids within the same species. Steroid concentrations in faeces exhibit a similar pattern to those in plasma, but have a lag time, which depending upon the species, can be from 12 h to more than 2 days. Faecal steroid metabolites in mammals are mainly unconjugated compounds. Faecal oestrogens consist predominantly of oestrone and/or oestradiol-17α or -17β. Therefore, specific oestrogen antibodies or antibodies against total oestrogens can be used for their determination. Progesterone is metabolised to several 5α- or 5β-reduced pregnanediones and hydroxylated pregnanes prior to its faecal excretion. Therefore, relevant antibodies for their determination show considerable cross-reactivities with several pregnane metabolites, whereas specific progesterone antibodies are less suitable. Faecal oestrogen evaluations have been used as reliable indicators of pregnancy in several ungulate and some primate species. They have also been used to determine the preovulatory period in carnivores, corpus luteum activity in New World primates, and to diagnose cryptorchidism in horses. Faecal progesterone metabolite analysis has been successfully used for monitoring corpus luteum function and pregnancy, abortion, seasonality and treatment therapies in an ever expanding list of species.