Initial velocity measurements of [’HIADP and [3H]ATP uptake in rat liver mitochondria were taken as a function of the mitochondrial energy state over a wide range of external nucleotide concentrations in an effort to understand kinetic anomalies ( ie . nonlinearity) often seen in published double reciprocal plots for adenine nucleotide transport. In addition, the mechanism of carrier transport was examined by varying the size of the exchangeable [ADP] plus [ATP] pool. Results obtained showed that plots of u-* versus [ADP]” or [ATP]” under both high energy and uncoupled conditions were nonlinear and biphasic, suggesting the presence of (a) two carriers with different kinetic properties; (6) negative cooperativity among adjacent carrier subunits or clusters of carriers in the membrane; or (c) endogenous, tight-binding competitive inhibitors (e.g. long chain acyl-CoA esters). Determinations of the apparent kinetic constants at higher substrate concentrations indicated that under energized conditions K , and V,,, values for ATP uptake are greater than K,,, and VmaX values for ADP uptake but that under de-energized conditions the opposite was observed. Changes in the K,,,/V,,,,, ratio, as a function of the mitochondrial energy state, were mainly due to the 10-20-fold increase and decrease in the K,,, for ADP and ATP uptake, respectively, upon a high energy to uncoupled state transition. Variation in the size of the intramitochondrial exchangeable pool was achieved by treating mitochondria with potassium arsentate which lowered the ATP/ ADP ratio as well as the total amounts of [ADP] plus [ATP] while increasing the level of [AMP]. The double reciprocal plots at various internal [ADP] plus [ATP] levels also showed biphasic patterns which extrapolated to yield two sets of intersecting lines on the abscissa. This pattern was observed for both ADP and ATP uptake and was independent of the mitochondrial energy state. These results indicate formation of a ternary complex as an integral part of the exchange process. Such an intermediate is inconsistent with the widely accepted ping-pong mechanism for nucleotide transport. A model for carrier transport which is consistent with the formation of a carrier-(nuc1eotide)a ternary complex and with previous results from inhibitor-binding studies is presented.