Wheat is often the largest single ingredient in piglet diets and therefore variation in quality may have a large effect on piglet performance. Currently there is no rapid method for the nutritive assessment of wheat. The livestock feed industry traditionally uses bushel weight as a predictor of nutritive value; however this assumption has recently been challenged by a literature review (Miller and Wilkinson, 1998) and by a weaner trial (Miller, Toplis, Arnold, Cook and Marshall, 2000). The aim of this experiment was to compare two more extreme bushel weights of Riband (64 v 78 kg/hl) than used in the previous experiment (Miller et al., 2000) when fed to weaned piglets with and without a xylanase enzyme. In order to amplify possible differences wheats were uncooked and included in the test diets at an atypically high level. We hypothesised that 78 kg/hl bushel weight would outperform 64kg/hl bushel weight (which is below the standard accepted by feed mills for feed wheat) and that enzyme supplementation would improve the performance of both wheats.