Abstract: The Eneolithic Botai culture of the Central Asian steppes provides the earliest archaeological evidence for horse husbandry, ~5,500 ya, but the exact nature of early horse domestication remains controversial. We generated 42 ancient horse genomes, including 20 from Botai. Compared to 46 published ancient and modern horse genomes, our data indicate that Przewalski's horses are the feral descendants of horses herded at Botai and not truly wild horses. All domestic horses dated from ~4,000 ya to present only show ~2.7% of Botai-related ancestry. This indicates that a massive genomic turnover underpins the expansion of the horse stock that gave rise to modern domesticates, which coincides with large-scale human population expansions during the Early Bronze Age.

Abstract: Horses involved in road traffic accidents (RTAs) are commonly presented to veterinarians with varying types of injuries. The aim
of this study was describe the pattern and severity of traffic accident-related injuries in horses in a single hospital population. Medical
records of horses either hit by a motorized vehicle or involved in RTAs whilst being transported from 1993 to 2015 were retrospectively
reviewed and the following data was extracted: Signalement, hospitalisation time, month in which the accident happened, cause of the
accident, place of the accident and type of vehicle hitting the horse. Further the different body sites injured (head, neck, breast, fore limb,
abdomen, back and spine, pelvis and ileosacral region, hind limb, tail and genital region), the type of injury (wounds, musculoskeletal
lesions and internal lesions) and the presence of neurological signs were retrieved from the medical records. 34 horses hit by motorized
vehicles and 13 horses involved in RTAs whilst being transported were included in the study. Most of the accidents where horses were hit
by motorized vehicles occurred during December (14.7%) and October (14.7%), horses were most commonly hit by cars (85.3%) and the
majority of accidents occurred on main roads (26.5%). In 29.4% of the cases, horses had escaped from their paddock and then collided
with a motorized vehicle. Most of the accidents with horses involved in RTAs whilst being transported occurred during April (30.8%) and
June (23.1%). In 76.9% of the cases the accident happened on a freeway. In the horses hit by motorized vehicles the proximal hind limbs
were the body site most commonly affected (44.1%), followed by the proximal front limbs (38.2%) and the head (32.4%). When horses
were involved in RTAs whilst being transported the proximal fore limbs (61.5%), the proximal hind limbs (53.8%) and the distal hind limbs,
back and head (38.5% each) were the most common injured body sites. Wounds were the most common type of injury in both groups
(85.3% hit by motorized vehicle, 76.9% transported ones). In horses hit by a motorized vehicle 35.3% suffered from fractures, in 20.6%
a synovial structure was involved and in 5.9% a tendon lesion was present. 14.7% suffered from internal lesions and 14.7% showed neurologic
symptoms (40% peripheral, 60% central neurologic deficits). On the other hand, in horses involved in a RTA whilst being transported
30.8% suffered from fractures. There were no synovial structures injured and no tendon injuries were present. Furthermore there were
no internal lesions present and only one horse involved in a RTA showed central neurologic symptoms. Injuries of horses being hit by a
motorized vehicle were more severe than when horses were protected by a trailer and involved in a RTA whilst being transported. The study
has been able to identify the different injury types of traffic accident-related injuries in horses. Awareness of the nature of these injuries is
important, to avoid underestimation of their severity.

Abstract: Large carnivores, such as gray wolves, Canis lupus, are difficult to protect in mixed-use landscapes because some people perceive them as dangerous and because they sometimes threaten human property and safety. Governments may respond by killing carnivores in an effort to prevent repeated conflicts or threats, although the functional effectiveness of lethal methods has long been questioned. We evaluated two methods of government intervention following independent events of verified wolf predation on domestic animals (depredation) in the Upper Peninsula of Michigan, USA between 1998-2014, at three spatial scales. We evaluated two intervention methods using log-rank tests and conditional Cox recurrent event, gap time models based on retrospective analyses of the following quasi-experimental treatments: (1) selective killing of wolves by trapping near sites of verified depredation, and (2) advice to owners and haphazard use of non-lethal methods without wolf-killing. The government did not randomly assign treatments and used a pseudo-control (no removal of wolves was not a true control), but the federal permission to intervene lethally was granted and rescinded independent of events on the ground. Hazard ratios suggest lethal intervention was associated with an insignificant 27% lower risk of recurrence of events at trapping sites, but offset by an insignificant 22% increase in risk of recurrence at sites up to 5.42 km distant in the same year, compared to the non-lethal treatment. Our results do not support the hypothesis that Michigan's use of lethal intervention after wolf depredations was effective for reducing the future risk of recurrence in the vicinities of trapping sites. Examining only the sites of intervention is incomplete because neighbors near trapping sites may suffer the recurrence of depredations. We propose two new hypotheses for perceived effectiveness of lethal methods: (a) killing predators may be perceived as effective because of the benefits to a small minority of farmers, and (b) if neighbors experience side-effects of lethal intervention such as displaced depredations, they may perceive the problem growing and then demand more lethal intervention rather than detecting problems spreading from the first trapping site. Ethical wildlife management guided by the “best scientific and commercial data available” would suggest suspending the standard method of trapping wolves in favor of non-lethal methods (livestock guarding dogs or fladry) that have been proven effective in preventing livestock losses in Michigan and elsewhere.

Abstract: Many studies have been carried out into both motor and sensory laterality of horses in agonistic and stressful situations. Here we examine sensory laterality in affiliative interactions within four groups of domestic horses and ponies (N = 31), living in stable social groups, housed at a single complex close to Vienna, Austria, and demonstrate for the first time a significant population preference for the left side in affiliative approaches and interactions. No effects were observed for gender, rank, sociability, phenotype, group, or age. Our results suggest that right hemisphere specialization in horses is not limited to the processing of stressful or agonistic situations, but rather appears to be the norm for processing in all social interactions, as has been demonstrated in other species including chicks and a range of vertebrates. In domestic horses, hemispheric specialization for sensory input appears not to be based on a designation of positive versus negative, but more on the perceived need to respond quickly and appropriately in any given situation.

Abstract: Predation on livestock is a source of human-wildlife conflicts and can undermine the conservation of large carnivores. To design effective mitigation strategies, it is important to understand the determinants of predation across livestock species, which often differ in husbandry practices, vulnerability to predators and economic value. Moreover, attention should be given to both predation occurrence and intensity, because these can have different spatial patterns and predictors. We used spatial risk modelling to quantify factors affecting wolf predation on five livestock species in Portugal. Within the 1619 parishes encompassing the entire wolf range in the country, the national wolf compensation scheme recorded 17,670 predation events in 2009-2015, each involving one or more livestock species: sheep (31.7%), cattle (27.7%), goats (26.8%), horses (14.8%) and donkeys (3.2%). Models built with 2009-2013 data and validated with 2014-2015 data, showed a shared general pattern of predation probability on each species increasing with its own density and proximity to wolf packs. For some species there were positive relations with the density of other livestock species, and with habitat variables such as altitude, and land cover by shrubland and natural pastures. There was also a general pattern for predation intensity on each species increasing with its own density, while proximity to wolf packs had no significant effects. Predation intensity on goats, cattle and horses increased with the use of communal versus private pastures. Our results suggest that although predation may occur wherever wolves coexist with livestock species, high predation intensity is mainly restricted to particular areas where husbandry practices increase the vulnerability of animals, and this is where mitigation efforts should concentrate.