CHEMISTRY five atoms of carbon—is combined with hydrogen cyanide, and the cyanide is hydrolysed, an acid isomeric with ordinary gluconic acid is obtained, and Kiliani had prepared from this acid the corresponding lactone. Kiliani’s lactone and that obtained from- mannonic acid were found by Fischer to be extraordinarily similar, but to differ in optical activity; by mixing the two lactones in equal proportions he obtained an optically - inactive product. The final step was taken at this stage. It was found that the lactones referred to could all be converted into corresponding sugars by combining them with hydrogen by means of sodium amalgam, the mannolactone giving ordinary mannose, the isomeric lactone an isomeric mannose of equal but opposite activity, and the inactive lactone an inactive mannose. Corresponding mannitols were obtained by subjecting these aldoses to further reduction, and it was found that the inactive mannitol was identical with a-acritol. Hence it followed that the a-acrose separated from acrosazone was an inactive mixture of ordinary fructose with an isomeride of equal but opposite rotatory power. A separation was effected by destroying the ordinary fructose by fermentation with beer yeast, and then adding phenylhydrazine. An osazone was thus obtained which was the stereoisomeride of glucosazone prepared from ordinary fructose. Although fructose had been obtained, it yet remained to prepare glucose. Bearing in mind the relation of mannose to glucose, it was to be expected that the one otcflucose wou^ be convertible into the other, but it could be foreseen that owing to their instability it would be difficult in practice to effect the conversion. Fischer, therefore, preferred to attempt ‘ to convert the corresponding monobasic acids into each other, as these were more stable, and Pasteur’s researches already provided a method, viz., that used in preparing lasvotartaric acid from ordinary dextrotartaric acid. He soon found that ordinary gluconic acid could be obtained from mannonic acid, and vice versa, by heating their solutions with quinoline under pressure at 140°. The final step in solving the problem was then easily taken by converting ordinary gluconic acid into its lactone and reducing this latter. Having thus succeeded in preparing substances identical with natural dextrose and laevulose entirely by artificial means, and starting from formic aldehyde—which can itself be produced from carbon dioxide, the material from which the plant derives its carbon—Fischer devoted his attention to the preparation of other stereoisomerides of glucose in order to obtain the data necessary to determine the exact configuration of the individual compounds. To understand his methods, it is necessary to bear in mind the manner in which the sixteen possible hexaldoses are structurally related. An asymmetric carbon system in which the four radicles attached to the carbon atom are arranged in a given order—say clockwise—may be termed a positive ( + ), and one in which hexaldoses. ^iey are arranged counter-clockwise a negative ( - ) system. When there are two such systems in a compound, four different arrangements are possible, provided the compound as a whole be asymmetric, viz.— But if, as in the case of tartaric acid, (C0.2H).CH(0H).CH(0H).(C02H), the compound is symmetric, being composed of two like systems, two of the four forms (those included in brackets) are identical, and the number of stereoisomerides is reduced to three ; moreover, in this latter case only two of the forms will manifest optical activity, the third being
723
inactive, as it consists of two equal systems of opposite sign. In the case of the hexaldoses, which are all represented by unsymmetrical formulae, symmetry is produced when the terminal groups become identical, as in the formation either of the corresponding hexhydric alcohol or of the corresponding tetrhydric di-acid :— Hexhydric alcohol . CH2(OH).[CH(OH)]4.CH2(OH) Hexaldose . . CH2(OH).[CH(OH)]4.COH Tetrhydric di-acid . CO(OH).[CH(OH)]4,CO(OH). Consequently, whereas there are sixteen possible hexaldoses, there are only ten corresponding hexhydric alcohols and ten di-acids. In the following table the possible configuration of the four asymmetric systems are arranged so as to indicate the pairs of forms (5 11, 6 12, etc.) which give rise to one and the same substance when the molecule becomes symmetrical.
Besides proving that the hexaldoses can be reproduced by reducing the lactones formed from the manobasic acids into which they are first converted by oxidation, Fischer has shown that it is possible to effect a similar reversal even in the case of the dibasic acids which are formed by continuing the oxidation. This discovery is of fundamental importance, as it gives a means of dealing, as it were, with both extremities of the molecule, and even of reversing the positions of the terminal groups in the hexaldoses. Taking the case of the saccharic acid, C02H.[CH(0H)]4.C02H, formed on oxidizing ordinary glucose, if the configuration be one of those represented under 1, 2, 3, or 4 in the above table, it must be a matter of indifference which of the two C02H groups is reduced to CH2(OH) and which to COH—the result will be the same; but not so in any other case. Experiment has shown that the hexaldose obtained from ordinary saccharic acid is not glucose, but an isomeride now known as gulose. The optical isomeride of this gulose is obtained by making use of the gluconic acid which is the optical isomeride of that prepared from ordinary glucose. The importance of such a result is obvious ; not only is the series extended, but it is placed beyond question that configurations 1-4 do not occur in either glucose or gulose, and in view of the simple relation which mannose bears to glucose, and seeing that the difference resides in the asymmetric system adjoining the COH group, it becomes possible to establish the configuration of glucose, gulose and mannose. Since d-glucose1 and d-gulose both yield d-saccharic 1 To distinguish the isomerides of opposite optical activity, it is usual to prefix the letters d- and 1-, but these are used only to indicate the genetic relationship, and not the character of the optical activity; ordinary fructose, for example, being represented as d-fructose—although it exercises a Isevorotatory power—because it is derived from d-glucose.
