While the hydrolytic action of enzymes is thus clear the synthesis in the cell is still a riddle. An interesting suggestion was made by van’t Hoff, who in 1898 expressed the idea that the hydrolytic enzymes should also act in the opposite direction, namely synthetically. Thus it should not only be possible to digest proteins with pepsin but also to synthetize them from the products of digestion with the aid of the same enzyme. This expectation was based on the idea that the enzyme did not alter the equilibrium between the hydrolyzed and non-hydrolyzed part of the substrate but only accelerated the rate with which the equilibrium was reached. Van’t Hoff’s idea omitted, however, the possibility that in the transitory combination between enzyme molecule and substrate a change in the molecular configuration of the substrate or in the distribution of intramolecular strain may take place. The first apparently complete confirmation of van’t Hoff’s suggestion appeared in the form of the synthesis of maltose from grape sugar by the enzyme maltase, which decomposes maltose into grape sugar. By adding the enzyme maltase from yeast to a forty per cent. solution of glucose Croft Hill[17] obtained a good yield of maltose. It turned out, however, that what he took for maltose was not this compound but an isomer, namely isomaltose, which has a different molecular configuration and cannot be hydrolyzed by the enzyme maltase. [17] Hill, C., _Jour. Chem. Soc._, 1898, lxxiii., 634. Lactose is hydrolyzed from kephyr by an enzyme lactase into galactose and glucose; by adding this enzyme to galactose and glucose a synthesis was obtained not of lactose but of isolactose; the latter, however, is not decomposed by the enzyme lactase. E. F. Armstrong has worked out a theory which tries to account for this striking phenomenon by assuming “that the enzyme has a specific influence in promoting the formation of the biose which it cannot hydrolyze.”[18] The theory is very ingenious and seems supported by fact. This then would lead to the result that certain hydrolytic enzymes may have a synthetic action but not in the manner suggested by van’t Hoff. [18] Armstrong, E. F., _Proc. Royal Soc._, 1905, B. lxxvi., 592. The principle enunciated by Armstrong, that in the synthetic action of hydrolytic enzymes not the original compound but an isomer is formed which can not be hydrolyzed by the enzyme, may possibly be of great importance in the understanding of life phenomena. It shows us how the cell can grow in the presence of hydrolytic enzymes and why in hunger the disintegration of the cell material is so slow. It was at first thought that the formation of isomers contradicted the idea of the reversible action of enzymes, but this is not the case; on the contrary, it supports it but makes an addition which may solve the riddle of what Claude Bernard called the creative action of living matter. We shall come back to this problem in the last chapter.