Three cooking treatments were applied to the longissimus muscle of beef and of pork and to deboned leg of lamb. Cooking treatments included: Intermittent energy application (3-min. cycle) with a microwave range operated at 220V and intermittent energy application (6-min. cycle) with a microwave range operated at 115V. Control roasts were cooked in a conventional gas oven (163+/-3 degrees C.). Cooking was adjusted so that roasts achieved an internal temperature of 70 degrees C. when cut for analyses. Cooking losses were significantly greater for microwave than for conventionally cooked beef. However, microwave cooking resulted in beef, pork, and lamb roasts with flavor of interior portions similar to those prepared conventionally. Flavor differences in samples from the edge of the slices of lamb and of pork and tenderness of lamb appeared to be related to cooking method. For these attributes, meat cooked conventionally was superior. In contrast, patterns in significant differences in tenderness and juiciness of beef and of pork were not consistent and were not related solely to method of cookery. Neither creatine nor creatinine was a good index of flavor of meat cooked by these methods. Aside from the time-saving aspect of microwave heating, there was no major advantage of one method of cooking over another. Thus, either high- or low- powered microwave equipment, operated at 2450 MHz, can be used satisfactorily for cooking tender cuts of beef, pork, and lamb.