Partial purification of ribonucleoside diphosphate reductase from rabbit bone marrow was achieved by size exclusion HPLC of the crude homogenate. This step, requiring < 15 min, led to 9- to 13-fold purification of the reductase and removal of 64% of the contaminating kinase/phosphatase activities, which in the crude extract degrade > 95% of substrate CDP when reductase is assayed. A systematic study was conducted to evaluate the influence of contaminating kinase/phosphatase activities on CDP concentration during the reductase-catalyzed reaction with either ATP or its kinase-inhibiting analog, 5'-adenylylimidodiphosphate (AMP-PNP), as the allosteric effector. Our studies demonstrated that in the presence of ATP, CDP levels fell instantly to < 24% but thereafter remained fairly constant due to recycling via CTP. In contrast, in the presence of AMP-PNP, CDP levels decreased continuously. The Km values of the reductase for CDP determined in the presence of ATP were significantly higher than those in the presence of AMP-PNP. Furthermore, we also found that the concentration of the ultimate electron donor dithiothreitol (DTT) required for optimum activity of the reductase varies significantly with the level of purity of the reductase preparation. Interestingly, DTT is an inhibitor of the reductase above the optimum concentration. This purification method and the optimized assay together with the understanding of the fate of CDP in partially purified preparations should find application in studies with reductases from other eukaryotic sources.