| Summary: | The measurement and determinants of equine bodyweight were investigated with the aims of establishing the precision, accuracy and applicability of existing methods for estimating equine bodyweight and investigating some of the factors which influence bodyweight. An investigation of the accuracy of existing methods for the estimation of bodyweight found that all of the methods of estimating bodyweight gave estimates that were significantly different (P<O.OOl) from actual weights measured by the weighbridge and that the most accurate method for estimating bodyweight was the formula developed by Carroll and Huntington (1988) which had an error of -1.4 + 1.06 per cent. Weigh tape 1 and the visual estimate were the least accurate methods, with mean error values of 12.0 ± 0.93 per cent and -11.7 + 2.01 per cent respectively. When the population was divided into two height groups « 147 cm - the so called ponies - and > 147 cm - the horses) the formula was also still found to be the least erroneous method for estimating bodyweight of ponies with an improved error value of -p.4 + 0.52 per cent, whilst for horses weigh tape 1 with a mean error of +3.5 (0.91) per cent proved to be the best estimate of bodyweight. In order to provide an improved estimator of bodyweight the relationship between bodyweight and other body measurements was investigated. Girth measurements were found to be most closely correlated with bodyweight, followed by height and finally length (heart girth R=0.95, high heart girth R=0.94, umbilical girth R=0.85, formula length R=0.58,rug length R=0.62,height R=0.80). Exact R values were again affected by whether ponies or horses were being assessed. An improved method for measuring bodyweight was developed based on heart girth measurement and dividing the population into horses and ponies. The use of this method on a population of 2000 animals found that for both the pony and horse sub-groups, the newly developed weigh tapes provided estimates of bodyweight that were not significantly different (P>0.05) from the actual weights measured with the weighbridge. However, large individual variations in the error values were observed, particularly amongst the horse population. In order to further refine the recommendations for measuring bodyweight, factors likely to affect bodyweight and its measurement that were investigated were breed, time of day, exercise and diet. Breed was found cause significant (P<O.OOl) variations in body measurements and their' relationships with bodyweight. Arabs' bodyweights were best estimated using the formula BWT=67.S2+(1.80NG+3.10UG-2.82WH)R2=0.78,ThoroughbredsBWT=-1070.2+(3.43WH+ 3.37UG +17.2LMC) R2=0.84, Warmbloods BWT=-768.4+ (2.9SUG + 4.1FL) R2=D.96, and Welsh BWT=-764.2 +(5.19HG + 14.4 LMC) R2=0.96. Time of day was found to , significantly (P<O.OS) affect both bodyweight and heart girth measurements, however a constant relationship between heart girth and bodyweight was maintained irrespective of time of day (R=0.92). Exercise of moderate intensity and duration (4* three-day event phases A-D) was found to cause significant bodyweight losses (P<0.05), however once ag~in .a.constant relationship between bodyweight and heart girth was maintained (R70.78). with heart girth providing estimates of bodyweight losses post-exercise which ~:ere not significantly different (P>O.OS) to those losses measured by the weighbridge. A 4x4 Latin-square design feed trial was conducted to investigate the effect of diet on bodyweight and performance, in particular the effect of forage:concentrate ratios on ..11()?YlNeight and performance. Isocaloric rations ranging from 100% forage (100H) to 50% torage: 50% concentrate (SOH) were found to influence bodyweight with horses receiving ,100H ration being significantly (P<0.001) heavier than those receiving a SOH ration.
|
|---|
