Assuming that his conclusion that all yeasts which ferment glucose also ferment fructose and mannose is correct, Armstrong has drawn attention to the fact that these three hexoses are also related by the possession of a common enolic form (p. 97) and has suggested that this enolic form is the substance actually fermented to carbon dioxide and alcohol [1904]. The idea that such an intermediate form is the direct subject of fermentation has much to recommend it. In the first place it is almost certain, as already pointed out, that the sugars in aqueous solution do exist, although to a very small extent, in this enolic form. The slow rate at which equilibrium is established in aqueous solution, however, must be taken as definite evidence that under these circumstances the enolic form is only produced very slowly [compare Lowry, 1903]. This has been used by Slator [1908, 1] as an argument against the probability of the preliminary conversion of the sugars into the enolic form before fermentation. It appears, however, quite possible that under the influence of the fermenting complex of the yeast-cell, or of special enzymes, this change might occur much more rapidly, [p134] and at different rates with the different sugars. This reaction might in fact control the observed rate of fermentation. This conception affords a simple explanation of the different rates of fermentation of mannose and glucose, and also of galactose, the enolic form of which is quite different, by yeast under different circumstances, but does not explain the uniformity of rate observed by Slator for glucose and fructose nor the results with mixtures of sugars. The direct fermentation of a common enolic form is also consistent with the fact that the same hexosephosphate is produced from all three hexoses. Slator himself prefers the view that the first stage of fermentation consists in the rapid combination of the sugar with the enzyme, producing a compound, which then breaks up at a rate which determines the observed rate of fermentation. This rate will of course vary with the nature of the compound, so that if two sugars form the same compound they will be fermented at the same rate; if they form different compounds, different rates may result. Slator supposes that glucose and fructose form the same compound with the enzyme. This, however, appears to involve an intramolecular change of the same order as the production of the enolic form, and moreover is not absolutely essential, as it is probably sufficient to suppose that the two compounds derived from glucose and fructose are very similar, although possibly not absolutely identical. Mannose and galactose, on the other hand, form stereoisomeric compounds, and the capacity of the fermenting complex to form these compounds may be affected by various agents to a different extent from its capacity for combining with glucose or fructose.