In order to examine the previous assumption that the primary energy source for carp is not carbohydrate (glucose) but protein (amino acids), incorporation of radioactivity from glucose-U-14C or glutamate-U-14C into hepatopancreatic lipid and glycogen, blood glucose, and its excretion as 14CO2 in carp were measured. Carp used were bred for 45 days with two kinds of diet, HC (90% carbohydrate) and C (50% carbohydrate), which were composed of potato starch and casein. Excretions of 14CO2 from glutamate-U-14C of both diet groups were extraordinarily higher than those from glucose-U-14C. In both groups, glucose-U-14C was hardly incorporated into hepatopancreatic lipid but was incorporated into glycogen. In contrast, much more glutamate-U-14C was incorporated into lipid whereas little into glycogen. The blood glucose levels of both diet groups were unusually increased probably by injections of glutamate-U-14C with carrier glutamate, and the radioactivities in blood glucose 6 hours after the injections were 1/7 to 1/5 of the residual activities from glucose-U-14C. From these results it was suggested that oxidation of amino acids by carp is superior to glucose oxidation and that depot lipid is principally synthesized from amino acids while only a little glucose is converted to glycogen. Therefore, amino acids appear to be a more excellent energy source than glucose in carp and the energy utilization resembles that in diabetic mammals.