Abstract The NADP-dependent malate dehydrogenase (decarboxylating) ( l -malate: NADP oxidoreductase (decarboxylating) EC 1.1.1.40), combined hexose monophosphate shunt dehydrogenases∗ and isocitrate dehydrogenase ( l S-isocitrat: NADP oxido-reductase (decarboxylating), EC 1.1.1.42) activities were assayed in mitochondrion-free fractions of liver obtained from normal fed, starving, starving but X-irradiated, and refed rats. Starvation for 48 h decreased the first two enzyme activities to about half that of the normal liver, whereas refeeding for 48 h with carbohydrate rich diets after starvation increased these enzyme activities over five times the normal liver value. In starving but X-irradiated (2400 R) rats the combined hexose monophosphate shunt dehydrogenase activity was similar to, and malate dehydrogenase (decarboxylating) activity was higher than, those of normal fed liver. There was no such marked effect of these experimental conditions on the isocitrate dehydrogenase activity which remained relatively unaltered. Compared with liver, in epididymal fat pads of normal rat the combined hexose monophosphate shund dehydrogenases, malate dehydrogenase (decarboxylating) and malate dehydrogenase ( l -malate: NAD-oxidoreductase, EC 1.1.1.37) were considerably more active, whereas lactate dehyrogenase ( l -lactate: NAD oxidoreductase, EC 1.1.1.27) and isocitrate dehydrogenase were less active. Thus in addition to the dehydrogenases of hexose monophosphate shunt, malate dehydrogenase (decarboxylating), but not isocitrate dehydrogenase, showed a close association with lipogenesis. Malate dehydrogenase (decarboxylating) is conceivably involved as a source of NADPH2 formation. Experimental evidence is presented for the conversion of NADH2 to NADPH2 in preparations from liver and epididymal fat pads which involve malate dehydrogenase (decarboxylating) and malate dehydrogenase.