Although the concentration of carbon dioxide or of oxygen in the storage atmosphere have a considerable effect upon the sprouting of potatoes, it has been shown that changes in the carbon dioxide and oxygen tension in the internal atmosphere of the tuber, prior to the natural break of dormancy, are negligible (Burton, I950, I95i). Further, where sprout inhibition has resulted from storage in high concentrations of carbon dioxide these high concentrations have usually been achieved by means which allowed also the accumulation in the storage atmosphere of volatile respiratory products other than carbon dioxide. Thus Kidd (I9I9) inhibited the sprouting of non-dormant potatoes by storing them in three artificial gas mixtures containing zo % of carbon dioxide, and 5, io and 20 % of oxygen respectively. He did not pass these mixtures continuously through the desiccators in which the tubers were contained, but adjusted their composition every two days (see Kidd, I914). The accumulation, to some extent, of volatiles other than carbon dioxide was thus possible. In experiments in which a concentration of io% carbon dioxide was used, the reduction of sprouting which normally resulted was due more to a reduction in the number than in the size of sprouts, and there were instances in which the sprouting in I0 % carbon dioxide and in the absence of carbon dioxide did not markedly differ. Braun (I93I) used three methods of adjusting the composition of the atmospheres in which he stored potato tubers. The first, which he dismissed as unsatisfactory, was to allow the respiratory carbon dioxide to accumulate for from i to 3 days, the highest concentration reached being zo2z 0. The atmosphere was then replaced by air-it is in this complete replacement that the method differs from that of Kidd-and the process repeated. In the second method, which was that adopted by Braun for most of his work, carbon dioxide was allowed to accumulate as above, and it was then attempted to keep it at the desired level by continuous ventilation. The technique was not altogether satisfactory. Thus a concentration of carbon dioxide described as 9g I3 % was in reality the mean, weighted for time, of concentrations ranging from 3.5 to I2z6 %. The third, and most satisfactory, method was continuously to ventilate, with artificial gas mixtures, the desiccators in which the tubers were contained. It will be seen that, in the first of the methods described above, volatiles other than carbon dioxide would accumulate for a few days and would then be removed; in the