easily oxidizable substance, and that the reduction of carbon dioxide is effected, wholly or in part, by the quinol of chlorophyll—one at least of these changes taking place under the influence of light. Assuming that formaldehyde, ch2o, is the immediate product, it is highly probable that this is at once converted into a carbohydrate—in many cases into starch. To appreciate the importance of this change, it is necessary to consider the phenomena of fermentation; it is then obvious that the energy Fermen- stored up in the carbohydrate system plays an tation. all-important part. Fermentation is certainly at bottom a process of hydrocatalysis, and it can scarcely be doubted that the function of the enzyme is to introduce water into the circuit of change—to establish a circuit in which hydrolytic changes can occur, although not of the ordinary kind, but reductive on the one hand and oxidative on the other. Baeyer, in 1870, was the first to point out that the resolution of glucose during fermentation into alcohol and carbon dioxide, and, indeed, the phenomena of fermentation generally, may be accounted for by the assumption that translocation of hydrogen and oxygen atoms is induced through the alternate separation of the elements of water and tlyfir addition in a different order— changes which would involve the accumulation of oxygen atoms at certain parts of the molecule, thus constituting points of weakness where disruption of the molecule would inevitably occur. This is shown in the following formulae, in the second of which the separating elements are indicated by dots :— ch3 CH2...OH CH.,OH CH.OH C.OH...H CH.OH CH.OH
C(OH)2 I OHg III.
It will be seen that in the product of rehydration (III.), the terminal groups are fully reduced, whilst there has been an accumulation of OH groups in other parts of the system. In a molecule so constituted, changes might set in similar to those involved in the resolution of oxalic acid into carbon dioxide and formic acid. The conception is one of remarkable beauty, and undoubtedly a sufficient explanation, as changes precisely similar to those attending fermentation may be effected by simple hydrolysis. Thus glucose may be converted into lactic acid by heating it with alkali, and on boiling fructose with muriatic acid an acid is formed, i.e., kevulinic acid, CH3.C0.CH2.CH2.C02H, the production of which clearly involves a reduction of some of the CH2.OH and CH.OH groups, and oxidation of others, in the manner pictured by Baeyer. The changes attending fermentation are presumably functions of the protoplasm—in other words, they probably occur within very complex molecular systems of extreme instability, perhaps under the influence of, in contact with, the very same hydrolyst (enzyme) which, when separated from the cell, is so active in promoting the hydrolysis of cane sugar, or, if not, of substances of a similar nature. On inspecting Baeyer’s formulae it will be noticed that during rehydration the elements of the water molecule become added on to noncontiguous carbon - atoms; this appears to be a special feature of such changes. Supposing that the protoplasmic hydrolyst were to condition the formation of a conducting circuit in which any two of the cai'bon systems (CH2.OH, CH.OH or COH) of the glucose molecule and water mole-
cules were included, if the total hydrolytic change which could take place in such a circuit were exothermic, even if the change affecting the one group involved an expenditure of energy, water would be “ electrolysed,” and its hydrogen would effect the withdrawal of OH from the one group and its displacement by hydrogen, while oxygen would be added to the other group, it may be directly, or in consequence of the displacement of H by OH. The different effects produced by different organisms, on this hypothesis, would be the consequence of the hydrolysis affecting different systems. As most, if not all, fermentable compounds are asymmetric, and the enzymes are undoubtedly also asymmetric bodies, the direction of attack would depend on the character of the asymmetry of both hydrolyte and hydrolyst, and only compatible hydrolysis, i.e., those compatible with the hydrolyte, would condition the hydrolysis, i.e. fermentation. If only the conversion of glucose into alcohol and carbon dioxide be considered, there is a danger of regarding fermentation as a purely destructive process, but it must not be forgotten that in some cases products are obtained which afford the clearest evidence of the occurrence of constructive changes. It is, however, probable that some of these are the outcome of interactions occurring, as it were, naturally and independently between products of enzymic change, and that the direct function of constructhe enzyme in fermentation may after all be only tive aspect to induce simplification, not complication. Thus, of termento explain the production of isobutyl-carbinol ta ,on’ (the inactive amyl alcohol of fusel oil) from glucose, it may be supposed that in the first instance the glucose is resolved by enzymic action into a mixture of glyceraldose, CH0(OH).CH(OH).COH, with the isomeric glyceroketose, CH2(OH).CO.CH2(OH), which interact spontaneously, a new hexose, (CH2.OH)2.C(OH).CH(OH), CH(OH).COH, being formed, and that this in turn becomes transformed by enzymic influences into amyl alcohol. Again, to account for the production of butyric and caproic acids from lactic acid, it may be supposed that the latter is partly reduced by “fermentation” to the aldehyde CH3.CH(OH).COH, and that an interaction then occurs spontaneously between this aldehyde and unchanged acid, giving rise to the complex acid CH3.[CH(OH)]4.COOH, which on further fermentation yields acetic, butyric, and caproic acids. The formation of fat from carbohydrates is doubtless a process of the same kind, in which at an early stage at least three molecules of glucose, or some substance closely related thereto, become condensed into one, much as three molecules of acetone C3H60, yield a single molecule of phorone, C;)HuO; the reduction of the complex so produced is then effected by enzymic influence. Recent observations, however, prove beyond question that the ordinary hydrolytic changes effected by enzymes are reversible, and therefore that the enzymes must be credited with synthetic powers. Croft Hill -was the first to observe that glucose is partially converted into a biose by diastase, and he supposed that maltose was formed; later observations seem to show that it yields isomaltose : it may be that both are formed. Emil Fischer and E. Frankland Armstrong have since found that a mixture of glucose and lactose in contact with the kephir enzyme yields a biose closely related to if not identical with milk sugar. Ordinary alcoholic fermentation is a nicely-balanced action in which the products of the presumed “ electrolysis ” of water are fully utilized, so that neither oxygen nor hydrogen is liberated. In butyric and many other fermentations, however, hydrogen is freely evolved, and in these cases somewhat more highly oxidized products (acids instead of alcohols) are formed; in fact, the evolution of hydrogen involves the occurrence of an amount of oxida-