This dissertation is the product of two behaviour studies and an in vitro fermentation trial. Both behaviour studies were conducted at the Ashburton Racehorse-Training Centre in Ashburton, near Pietermaritzburg in KwaZulu- Natal. The first behaviour study evaluated differences in behaviours obtained through feeding either twice or four times daily. This trial showed (P<O.OO1) that horses fed twice a day spend a greater proportion of their day in stereotypic or vice-like behaviours. Horses eating four times a day ate less hay (P<O.OO1) and more concentrate (P<O.05) than horses being fed twice a day. Horses in both yards ate more hay (P<O.05) when exercise intensity was increased. Defecation frequency was higher on days when exercise intensity was high (P<O.05) and in the yard where horses were fed four times per day (P<O.O1). Faecal weight was greater (P<O.05), horses lay down more frequently (P<O.05), spent more time eating concentrate (P<O.OO1) and less time eating hay (P<O.OO1) when horses were fed four times per day. Fillies spent more time (P<O.05) eating hay than geldings. The second behaviour study was conducted within one yard only and considered the effects of changes in management strategies on the incidence of stereotypic behaviour. The results did not indicate that changes in management related to exercise intensity would have an effect on behaviours exhibited. However this trial did demonstrate that a reduction in feed intake on days when exercise is reduced will reduce the incidence of stereotypic behaviours. Horses reduced the time spent eating hay when exercise was reduced except that when feeding frequency was reduced in conjunction with reduced exercise, more time was then spent eating hay (P<O.OO1). It was found that fillies spent more time licking surfaces (P<O.OO1) and weaving (P<O.O1) than geldings, which were more aggressive (P<O.OO1) and ate more bedding (P<O.05) than the fillies. All the horses were more alert (P<O.OO1) on days of moderate exercise except when feed was reduced in conjunction with reduced exercise, so that horses were less alert (P<O.OO1) and more time resting (P<O.OO1). The in vitro study was conducted at the department of Animal and Poultry Science at the University of KwaZulu-Natal, Pietermaritzburg. This was a dilution trial, using different ratios of maize and Eragrostis curvula. It was shown that as the proportion of maize in the ration was increased so the digestibility and the degradability of the ration increased (P<O.OO1). It was also shown that the adapted two stage digestion techniques described by Tilley & Terry (1963) had lower supernatant pH levels than the samples that underwent microbial digestion only. This was accounted for by a problem with the methodology. The trial had hoped to show a dramatic decrease in pH and increased rates of gas production when the maize portion of the sample was increased. From the results established during this trial it is clear that application exists in the adoption of this method in in vitro feed analysis in the horse industry.· The behaviour studies significantly linked the incidence of stereotypic behaviour to feeding and nutritional management in racehorses. Some explanations of the noted behaviours can be elucidated through the development of in vitro protocols, where hindgut pH, degradability and fermentation of different ration mixtures elicit responses in physical terms.

Thesis (M.Sc.)-University of KwaZulu-Natal, Pietermaritzburg, 2007.