Evaluation of Western Australian canola meal for growing pigs
B. P. Mullan, J. R. Pluske, J. Allen and D. J. Harris
Australian Journal of Agricultural Research
51(5) 547 - 553
Twenty samples of canola seed collected throughout south-western Australia after the 1995 harvest were analysed for oil and total glucosinolate levels. Average oil and total glucosinolate levels were 378 g/kg and 14.0 mmol/kg (oil-free dry matter, DM), respectively. The subsequent single-press canola meal [containing an average of 96 g/kg oil and 10.5 mmol/kg total glucosinolates (oil-free DM)] extracted from 8 batches of the original canola seed was then used in 2 growth experiments. In Expt 1, canola meal was included in diets for growing–finishing pigs at levels of 0, 50, 100, 150, and 200 g/kg in the diet, and replaced Australian sweet lupins. In this study, feed was offered on an ad libitum basis between 23 and 110 kg liveweight (LW). In Expt 2, the same inclusion levels of canola meal were used; however, feed intake was standardised across treatments (approximately 3 times maintenance) to assess the impact of glucosinolate intake on performance and thyroid weight in pigs between 19 and 94 kg LW. In Expt 1, performance was similar (P > 0.05) in pigs offered up to 200 g/kg of canola meal. In Expt 2, average daily gain was similar between treatments (P > 0.05), although a linear decline in both daily gain (P = 0.244) and feed : gain (P < 0.05) with increasing canola meal levels was detected. Increasing amounts of canola meal in the diet were associated with heavier (P < 0.001) thyroid weights, suggesting a negative influence of the products of glucosinolate hydrolysis on thyroid function. Single-press canola meal is a suitable alternative to other protein sources; however, at levels >150 g/kg for growing–finishing pigs, higher intakes of glucosinolates were associated with a tendency for decreased pig performance and thyroid hypertrophy despite the canola varieties used being ‘low glucosinolate’. Keywords: thyroid, glucosinolates, production.
Full text doi:10.1071/AR99123
© CSIRO 2000