Attempts were made to provide proof for the occurrence of cyclic 3', 5'-adenosine monophosphate in healthy and crown gall tissues of Vicia faba. Although our purified extracts gave positive readings in the Gilman binding assay for cyclic AMP, they were not digested by a specific cyclic 3', 5'-adenosine monophosphate phosphodiesterase from beef heart. The extracts were digested, however, by a partially purified cyclic nucleotide phosphodiesterase from carrot tissue, which attacks both cyclic 2',3'-and 3', 5'-nucleotides. The data indicate that the substances detected in the V. faba extracts are perhaps cyclic 2',3'-nucleotides, a possible RNA degradation product.Various laboratories have claimed they have detected cyclic 3', 5'-adenosine monophosphate in higher plant tissues (8, 1 1). Other reports have presented evidence for a change in cyclic AMP levels after treatment of plant tissues with gibberellic acid (7), auxin (9), or irradiation with far red (3).A cyclic nucleotide phosphodiesterase has been isolated and partially purified from a number of plant tissues (4,6,10 AMP recovery after purification. After homogenization for 1 min at high speed in a Waring Blendor, the extract was clarified by squeezing through four layers of cheesecloth. The filtrate was then boiled for 17 min, neutralized, and reboiled for 17 min. This step precipitated almost all the protein and polysaccharides. The pellet obtained by centrifugation (10,OOOg for 15 min) was saved for protein determination, while the supernatant extracts were acidified and purified by passage through 15 x 0.7 cm columns of Dowex AG-50-X8 and the cyclic AMP fraction eluted with 0.05 N HCI. This eluate was neutralized in 0.01 M tris-HCl, pH 7.4, and was applied to a second column (5 X 0.7 cm of Dowex-2 equilibrated with 0.01 M tris-HCI, pH 7.4), and the cyclic AMP fraction again eluted with 0.05 N HCI. Dowex-2 eluates were lyophilized; the residues were dissolved in 0.01 M tris-HCl, pH 7.4, and the pH was adjusted to 7.4 with 0.1 N NaOH. Recoveries were routinely 50 to 60%.Duplicate samples from each treatment and from column blanks were assayed for cyclic AMP by the method of Gilman (2). The experimental results are expressed as picomoles of cyclic AMP/mg protein + the standard error of the mean determined from three to seven separate experiments.