We previously proposed that, during the infusion of either labeled leucine or its alpha-ketoacid, alpha-ketoisocaproate (KIC), the plasma specific activity (SA) of the transaminated product of the infused tracer ("reciprocal pool SA") may better reflect the intracellular leucine SA than the plasma SA of either infused tracer ("primary pool SA"). To test this hypothesis, 14 dogs were simultaneously infused intravenously with [3H]leucine and [14C]KIC, and blood and tissue compartments were sampled. The ratios of [3H]-leucine to [14C]leucine [( 3H]/[14C]leucine) in mixed tissue proteins and in the intracellular space of striated muscle were the same as the ratio of the isotope infusion rates and similar, although slightly lower (P less than 0.01), than [3H]KIC/[14C]leucine SA (ratio of reciprocal pool SA) in plasma. Plasma [3H]KIC/[14C]leucine SA were essentially identical to the [3H]/[14C] of leucine in 1) mixed liver proteins, 2) intrahepatic free leucine, and 3) fibrin. The [3H]/[14C]leucine in mixed renal proteins and in the intracellular space of kidney and erythrocytes were similar to those of the venous plasma [3H]/[14C]leucine SA. The plasma [3H]KIC and [14C]leucine SA (the reciprocal pool SA) were similar to the SA of [3H]- and [14C]leucine in the intracellular space of all organs investigated with the exception of kidney. Therefore, in postabsorptive dogs, the plasma SA of the transaminated product of the infused labeled KIC or leucine is an excellent predictor of the intracellular leucine SA in all tissues investigated with the exception of kidney.