AbstractCooking losses and eating quality were determined for semitendinosus (ST) and longissimus dorsi (LD) roasts obtained from bulls and steers fed three levels of dietary roughage (20, 50 and 80% alfalfa–brome hay) and from animals slaughtered at approximately 450 or 580 kg liveweight.Cooking losses of either type of roast did not appear to be affected by level of dietary roughage, but were affected by castration. For either weight group, trained panelists generally determined no significant effects of roughage level on eating quality of either type of roast. However, consumer panelists noted significant differences in the flavor, tenderness, and overall acceptability but not juiciness in ST roasts with those from the animals fed 20% roughage superior to those from the 80% group. Objective measurements of juiciness (water holding capacity (WHC)) and tenderness generally supported the data of the trained panel. Only one objective test for each type of roast, rated roasts from animals fed 80% roughage to be significantly less tender than roasts from animals fed lower roughage levels. All other objective tests detected no differences in tenderness attributable to roughage level. Cooking losses were generally greater for roasts from bulls than from steers. Fat percentage was lower and moisture higher for roasts from bulls than those from steers. A detectable difference in pH, with bull samples higher than steer samples, was found for ST roasts from the heavy weight group, and for LD roasts from both slaughter weight groups. Trained panelists detected few differences in palatability characteristics between ST roasts from bulls vs steers; more differences were detected between bull and steer LD roasts, and the differences were greater for the heavy slaughter weight group. Consumer panelists rated steer ST roasts superior to those from bulls for flavor, tenderness and overall acceptability, but the rating of roasts from bulls was acceptable. Water holding capacity was similar for bull and steer roasts in this study. Objective tests for tenderness did not detect differences between bull and steer ST roasts but did show highly significant differences between bulls and steers for LD roasts. Objective color difference between sexes was found only for ST roasts from the heavy slaughter group. Thus these studies indicate that beef from animals fed roughage levels from 20 to 80% of the diet was similar and acceptable. Differences in eating quality of beef from bulls and steers were detected but generally these differences were small and beef from bulls was judged acceptable.