Abstract: Six pregnant mares were used to determine what level of water restriction causes physiological and/or behavioural changes indicative of stress. Nonlegume hay was fed ad libitum. During the first week of restriction, 5 l water/100 kg bwt was available, during the second week 4 l/100 kg bwt and, during the third week, 3 l/100 kg bwt. Ad libitum water intake was 6.9 l/100 kg bwt; at 3 l/100 kg bwt water intake was 42% of this. Daily hay intake fell significantly with increasing water restriction from 12.9 +/- 0.75 kg to 8.3 +/- 0.54 kg; bodyweight fell significantly for a total loss of 48.5 +/- 8.3 kg in 3 weeks. Daily blood samples were analysed; osmolality rose significantly with increasing water restriction from 282 +/- 0.7 mosmols/kg to 293.3 +/- 0.8 mosmols/kg bwt, but plasma protein and PCV did not change significantly. Cortisol concentrations fell from 8.1 ng/ml to 6.4 ng/ml over the 3 week period. Aldosterone fell from 211.3 +/- 74.2 pg/ml to 92.5 +/- 27.5 pg/ml at the end of the first week. The behaviour of 4 of the 6 mares was recorded 24 h/day for the duration of the study. The only significant difference was in time spent eating, which decreased with increasing water restriction from 46 +/- 3% to 30 +/- 3%. It is concluded that water restriction to 4 l/100 kg bwt dehydrates pregnant mares and may diminish their welfare, but is not life- or pregnancy-threatening.

Abstract: Lateralization of emotions has received great attention in the last decades, both in humans and animals, but little interest has been given to side bias in perceptual processing. Here, we investigated the influence of the emotional valence of stimuli on visual and olfactory explorations by horses, a large mammalian species with two large monocular visual fields and almost complete decussation of optic fibres. We confronted 38 Arab mares to three objects with either a positive, negative or neutral emotional valence (novel object). The results revealed a gradient of exploration of the 3 objects according to their emotional value and a clear asymmetry in visual exploration. When exploring the novel object, mares used preferentially their right eyes, while they showed a slight tendency to use their left eyes for the negative object. No asymmetry was evidenced for the object with the positive valence. A trend for an asymmetry in olfactory investigation was also observed. Our data confirm the role of the left hemisphere in assessing novelty in horses like in many vertebrate species and the possible role of the right hemisphere in processing negative emotional responses. Our findings also suggest the importance of both hemispheres in the processing positive emotions. This study is, to our knowledge, the first to demonstrate clearly that the emotional valence of a stimulus induces a specific visual lateralization pattern.

Abstract: Much evidence now suggests that the postnatal killing of young in primates and carnivores, and induced abortions in some rodents, are evolved traits exerting strong selective pressures on adult male and female behaviour. Among ungulates it is perplexing that either no species have developed convergent tactics or that these behaviours are not reported, especially as ungulates have social systems similar to those of members of the above groups. Only in captive horses (Equus caballus) has infant killing been reported. It has been estimated that 40,000 wild horses live in remote areas of the Great Basin Desert of North America (US Department of Interior (Bureau of Land Management), unpublished report), where they occur in harems (females and young) defended by males. Here I present evidence that, rather than killing infants directly, invading males induce abortions in females unprotected by their resident stallions and these females are then inseminated by the new males.

Abstract: Regulation of local overpopulations of free-roaming feral equids is in demand worldwide for ecological balance and habitat preservation. Contraceptive vaccines have proven effective in feral horses, which breed seasonally, but no data are available for equids such as the burro, which is reproductively active all year round. In the present study, 27 individually identified female feral burros (Equus asinus) roaming free in Virgin Islands National Park (St John, US Virgin Islands; Lesser Antilles) were remotely treated with pig zonae pellucidae (PZP) vaccine. Between January and May, 16 burros were darted with a 1 ml emulsion of PZP plus Freund's adjuvant. Ten to twelve months later each treated burro was given a single booster injection of PZP plus adjuvant to maintain contraception through a second year. Eleven adult untreated jennies served as controls. Beginning one year after initial vaccination, these burros were monitored for pregnancy and foal production. Collection of data to determine treatment effect was not begun until 12 months after initial treatment to ensure that pregnancies existing before vaccination were not included. Pregnancy was assessed using previously validated methods for steroid metabolite measurement in fresh faecal samples. None of the PZP-treated burros produced foals between 0 and 12 months after the last inoculation. One PZP-treated burro tested positive for pregnancy at 10 months after the final inoculation. During this same period, six of 11 untreated burros tested pregnancy-positive, and four were observed with foals. There was no difference in pregnancy rates among treated, control and randomly sampled jennies between 12 and 24 months after the last inoculation. The results demonstrate that, in free-roaming feral burros that are reproductively active all year round: (1) burros can be accessed for remotely delivered PZP vaccination; (2) PZP contraception is effective; (3) PZP contraception is reversible; and (4) pregnancy can be reliably detected by faecal steroid analysis.