Methods The results described in this paper were obtained with the'original Escherichia coli'National Collection of Type Cultures no. 86, and with Micrococcus lysodeikticus no. 2665. The basal medium used throughout was tryptic digest of casein with the addition of 0 5% NaCl and, in the case of M. lysodeicticus, 1% marmite. For the greater part of the work, growth took place in flasks incubated at 270 but M. lysodeikticus is difficult to grow in liquid-nedia and in order to get sufficient material this organism was grown in Roux bottles containing 70 ml. medium, lying on their sides and incubated at 37. E. coli will grow under these conditions at any pH between 4x5 and 9 although the yield of organism towards the limits of this range is small and it was sometimes impossible to harvest cultures grown at pH 4x5. The media were adjusted to the desired pH before growth by the use of a glass electrode and the pH was again checked when growth had finished. The metabolic activities of the organisms tend to alter the pH of the medium so that there is a drift towards neutrality from either side. This drift may amount to as much as 15 pH units at the end of the range and for the purpose of this work it was necessary to prevent the drift as far as possible. The best way to accomplish this was found to be theincorporation of buffer in the medium and the following final concen-trations of buffer were used: for mediaadjusted to pH 4-5 and 5, M/60 phthalate, for pH 6-8, M/45 phosphate, and for pH 8-5 and 9, M/60 borate. These additions do not interfere with the growth of the organisms and we have satisfied ourselves that the results to be presented are not due to, or affected by, the presence of the buffers. This point has been checked in all possible cases byrepetition ofthe estimations as far as possible with cultures grown in media adjusted without buffers and by checking the activity of the pH 7 culture with that of a culture grown atpH 7 in the presence ofJf/60 phthalate and borate; in no case has there been any significant effect due to the presence of buffers as such during growth. Even with the addition of these amounts of buffer to the media, there is still a small drift in pH during growth and the extent of this drift is shown for a typical example in Table 1. The organisms were harvested at the end of growth, washed