Kostytscheff [1912, 1; 1913, 1, 2; 1914; Kostytscheff and Hübbenet, 1913; Kostytscheff and Scheloumoff, 1913; Kostytscheff and Brilliant, 1913] has examined the effect of the addition of zinc chloride, chosen with the idea that it might polymerise the aldehyde and thus remove it from the sphere of action. As pointed out by Neuberg and Kerb [1912, 1] this action is not very probable, and it was subsequently found [Kostytscheff and Scheloumoff, 1913] that the effect of added zinc salts was more probably specifically due to the zinc ion. Fermentation of sugar by dried yeast still proceeds when 0·6 gram of ZnCl{2} is added to 10 grams of the yeast and 50 c.c. of water, whereas it ceases in the presence of 1·2 gram of ZnCl{2}. Even the addition of 0·075 gram however greatly diminishes the rate of fermentation and the total amount of sugar decomposed. The most noteworthy effect is that the production of acetaldehyde is increased both in autofermentation and [p112] in sugar fermentation. The course of the reaction is further modified in the sense that the percentage of sugar used up which can be accounted for in the products decreases, in other words the "disappearing sugar" (p. 31) increases. In long continued fermentations moreover and particularly with high concentrations of zinc chloride less alcohol is produced than is equivalent to the carbon dioxide evolved. The interpretation of these results is difficult. Kostytscheff takes them to mean (1) that the zinc salt modifies one stage of the reaction so that a higher concentration of intermediate products is obtained, and (2) that the carbon dioxide and alcohol must be produced at different stages or their ratio, in the absence of secondary changes, would be unalterable. Alternative interpretations are, however, by no means excluded. Thus Neuberg and Kerb [1912, 1; 1913, 2] do not regard it as conclusively proved that the aldehyde really arises from the sugar since they have observed its production in maceration extract free from autofermentation. The method used by Kostytscheff for the separation of alcohol and aldehyde (treatment with bisulphite) has also proved unsatisfactory in their hands and the results obtained as to the reduction of acetaldehyde by yeast, etc., are not accepted. They also consider that in any case the small amounts produced (less than 0·2 per cent. of the sugar used) would not afford convincing evidence that the aldehyde is an intermediate product, although it must be admitted that no large accumulation of an intermediate product could be reasonably expected. It may also be pointed out that the increase in "disappearing sugar" may be simply due to the fact that in the controls the whole of the sugar was fermented, so that any polysaccharide formed at an earlier stage would have been hydrolysed and fermented, whereas in the presence of zinc chloride excess of sugar was present throughout the whole experiment.