A procedure has been developed to assess whether polymerization of nucleotides by DNA polymerases is processive, that is, whether a succession of polymerization steps occurs without release of the enzyme from the template. The method involves measurement of the ratio of deoxyguanylate to deocycytidylate incorporated in the course of replicating a segment of the right hand cohesive end of phage lambaDNA with the sequence 5'-G-G-G-C-G-G-C-G-3'. In the case of the Escherichia coli DNA polymerase I, each enzyme molecule completes synthesis of the sequence before dissociation occurs. Furthermore, at both 6 and 22 degrees, the polymerase remains bound to the lambdaDNA template after synthesis has completed. Template challenge experiments, in which the polymerase is allowed to begin synthesis in the presence of a molar excess of lambdaDNA before addition of a very large excess of calf thymus DNA, show that under the conditions used, productive binding of polymerase to lambdaDNA is a slow process requiring 1 to 2 hours. After synthesis has been completed, polymerase remains bound to the lambdaDNA in spite of the availability of new primer termini. The association, polymerization, and dissociation rates measured in these experiments suggest that the polymerization reaction catalyzed by DNA polymerase I is processive, and that hundreds of nucleotides may be polymerized between each association and dissociation.