The instability of penicillin in aqueous solution has been known since the time of its discovery by Fleming (1929). However, beyond the general recognition that excess acidity or alkalinity caused rapid inactivation, little work has been reported on the inactivation of penicillin over a wide pH range and at different temperatures. The maximum stability range of a barium salt of penicillin tested in aqueous solution by Abraham and Chain (1942) was between pH 5.5 to 7.5. Activity was retained at 2 C for several months, at 25 C for several weeks, and at 37 C for 24 hours, whereas most of the activity was lost in 30 minutes at 100 C. Rammelkamp and Helm (1943) studied the stability of penicillin during a 24-hour period at 5 C and 37 C in veal infusion broth at pH 2, 3, 4, 5, and 7.3. Although the original concentration of penicillin was low (0.625 Oxford units per milliliter), they were able to show rapid inactivation at pH 2 and 4 at both temperatures, partial inactivation at pH 5 and 37 C, and no inactivation at pH 4 and 5 at 5 C. Foster and Wilker (1943) conducted similar experiments in buffer solutions at pH 2.0, 2.6, 2.9, 4.8, 5.8, 6.8, 7.9, 9.4, and 10.3. The initial concentration of penicillin was only 0.168 Oxford units per milliliter. They concluded that penicillin is exceedingly labile in a medium below pH 4.8 or above 7.9, losing all activity in a matter of hours. The purity of the penicillin used by Rammelkamp and Helm and by Foster and Wilker was not given, but it is apparent from the low concentration (expressed as potency) that only partially purified preparations were used. Previous experiments conducted by the authors with crude penicillin had indicated that the optimum stability in aqueous solution was between pH 5.6 and 6.1 instead of at pH 7.0, which had been previously considered as the stability optimum. Furthermore, it was expected that crude penicillin would prove to be less stable than the crystalline material, and examination of the data presented here substantiates this supposition.