Developing an energy evaluation protocol for horse feeds in South Africa.
The purpose of this study was to find the most accurate and reliable method available in South Africa to evaluate equine diets on an energy basis. Currently South African horse owners purchase food according to the crude protein content of the diet, not knowing the energy density of the diet, which they are feeding their animals. Energy is one of the most important measures of an animal feed, as the energy density determines how much of a diet needs to be fed to meet an animal's requirement. The level of feed intake determines the concentrations of all other nutrients in the diet, therefore one cannot formulate a diet correctly without knowledge of its energy content. Through domestication, there has been an increased demand for horses to perform under circumstances that require energy greater than that provided by its natural diet of grass alone. This has therefore lead to the inclusion of cereal grains and their by-products. These large grain meals can overwhelm the digestive capacity of the horse thus leading to various types ofdigestive disorders such as colic. Therefore by the development of an energy evaluation system, one could provide the horse with the correct amount of energy from the appropriate source without compromising its digestive system. Predicting digestibility of a diet is the basic step for energy evaluation of horse feeds. Currently horse diets in South Africa are formulated using ruminant total digestible nutrient (TDN) data. As large horses are difficult to work with in digestibility trials, a preliminary experiment was designed to see how accurate it would be to use miniature horses as predictors of digestible energy for large horses. By comparing the digestibility data with that of overseas predictive equations, where large horses were used, the results were found to be highly comparable. As it was established that miniature horses were a perfect pilot animal for digestibility studies on large horses, the next step was to determine the rate of passage in miniature horses so as to determine if the length of the collection period, in a digestibility trial, proposed by overseas researchers for large horses, was enough time to clear the digestive system in a miniature horse of the diet under investigation. An experiment was carried out using Celite® as an insoluble marker to determine rates of passage via the acid insoluble ash method. A mean retention time of 66.64 hours was obtained, therefore assuring that a collection period of 5 days, as recommended by overseas researchers, was sufficient time to clear the miniature horse's digestive system of the test diet. Following the preliminary trial, a digestibility experiment was designed to investigate the accuracy of using ruminant data to formulate equine diets as well as using rabbits as a possible pilot animal in horse digestibility trials. The trial involved five miniature horses, four male sheep and ten rabbits. Four commercial horse diets were investigated. From these results it was found that rabbits proved difficult too work with and did not favour the experimental conditions and therefore gave digestibility results very different to that of the equine. Ruminants proved accurate predictors of the digestibility of the fibre components for horses but not for the other digestible nutrients. Significant differences were found between the diets given only to the horses and diets that should have provided a higher digestible energy did not. This accentuates the need for the development of an energy evaluation protocol, so that equine diets can be formulated more precisely and thereby ensure that the energy requirements of the horse are met. In vivo digestibility results were compared to the same feeds incubated in vitro and significant differences (P<0.05) were found between the results obtained by the two methods. A possible reason for this could be the method employed for removing supernatant between the two stages of the Tilley and Terry (1963) method, leading to an overestimation of digestibility for feeds containing hay and incubated in vitro. No significant differences (P>0.05) were found between digestibilities, rates and maximum gas production between the sources of inoculum used. Significant differences (P<0.05) were found between digestibilities obtained by incubating concentrates alone or in an 80:20 ratio with hay. Further investigation is needed here as it was felt that the supernatant removal method contributed significantly to inaccurate in vitro results. This experimental work centres on discovering and developing the best method available to the South African feed industry for predicting digestible energy contents of horse feed, so as to improve defined performance within an equine discipline and reduce nutritionally-induced disorders.