The expression "chemical structure," as commonly used by chemists, has, as is well known, nothing to do with the arrangement of atoms in space. The structural formula does not profess to represent spatial relations, but simply the connections which, after a careful study of the transformations and modes of formation of the compound represented, are believed to exist between the atoms. Nevertheless, although we do not commonly consider the question of space relations, it is clear that atoms must have some definite positions in space in the molecules, and the only reason why we do not represent these positions is because we know practically nothing about them. The most definite suggestion concerning space relations of atoms which has been made is that of Le Bel and Van't Hoff. The well known hypothesis of these authors was put forward to account for a certain kind of so-called physical isomerism which shows itself in the action of substances upon polarized light. Since this hypothesis was proposed, the number of cases of "abnormal isomerism," that is to say, of cases of isomerism which cannot be accounted for by the commonly accepted method of explaining structure, has increased to a considerable extent, and the necessity for some new hypothesis, or for some modification of the old ones, has come to be pretty generally recognized.

Among the cases of isomerism which it is at least difficult to explain by the aid of the prevailing views are those of maleic and fumaric acids; citraconic and mesaconic acids; certain halogen derivatives of crotonic acid and of cinnamic acid; and coumaric and coumarinic acids.

More than one hypothesis has been proposed to account for these cases of isomerism, but no one has shown itself to be entirely satisfactory. Quite recently Johannes Wislicenus, Professor of Chemistry in the University of Liepsic, has made what has the appearance of being an important contribution toward the solution of the problem referred to. The author shows that many of the facts known in regard to the relations between maleic and fumaric acids, and the other substances which furnish examples of "abnormal isomerism," may be explained by the aid of an extension of the Le Bel-Van't Hoff hypothesis. It is difficult without the aid of models to give a clear idea concerning the hypothesis of Wislicenus, but some idea of it may be gained from the following. If we suppose a carbon atom to exert its affinities in the directions of the solid angles of a tetrahedron, as is done in the Le Bel-Van't Hoff hypothesis, then, when two carbon atoms unite, as in ethane, the union will be between two solid angles of two tetrahedrons. If the two carbon atoms unite by the ethylene kind of union, the union will be along a line corresponding to one of the edges of each tetrahedron.

In the former case, in which single union exists, the two parts of the molecule represented by the two tetrahedrons can be supposed to be capable of revolving around an axis either in the same direction or in opposite directions. This axis corresponds to the straight line joining the two carbon atoms. In the case in which double union exists no such revolution is possible. Again, if, by addition to an unsaturated compound like ethylene, a saturated compound is formed, the kind of union between the carbon atoms is changed, and the possibility of revolution of the two parts of the compound is given. Whether such revolution take place or not will be determined largely by the structure of the compound. The tendency will be for those parts of the molecule which have the greatest specific affinity for one another to take those positions in which they are nearest to one another. Thus, suppose that chlorine is added to ethylene. By following the change on the model, it is seen that in the resulting figure the two chlorine atoms in ethylene chloride are situated at angles of the two tetrahedrons which are nearest each other. But chlorine has a stronger affinity for hydrogen than it has for chlorine, and therefore each chlorine atom would tend to get as near a hydrogen atom as possible.

This involves a partial revolution of the two tetrahedrons in opposite directions around their common axis. So also hydrogen would tend to take a position as near as possible to hydroxyl and to carboxyl, while hydroxyl would avoid hydroxyl, and carboxyl would avoid carboxyl. These views are suggested as a result of a careful application of the original Le Bel-Van't Hoff hypothesis, and are, of course, of little value unless they can be shown to be in accordance with the facts.

The chief merit of the work of Wislicenus consists in the fact that he has shown that a large number of phenomena which have been observed in the study of such cases of isomerism as were mentioned above find a ready explanation in terms of the new hypothesis, whereas for most of these phenomena no explanation whatever has thus far been presented. The most marked case presented is that of maleic and fumaric acids. One by one, the author discusses the transformations of these acids and their substitution products, and becomes to this conclusion: "There is not to my knowledge a single fact known in regard to the relations between fumaric and maleic acids which is not explained by the aid of the above geometrical considerations, not one which does not clearly support the new hypothesis." Among the facts which he discusses in the light of the hypothesis are these: The formation of fumaric and maleic acids from malic acid; the quantitative transformation of maleic into fumaric acid by contact with strong acids; the transformation of the ethereal salts of maleic acid into those of fumaric acid by the action of a minute quantity of free iodine; the formation of brommaleic acid and hydrobromic acid from the dibromsuccinic acid formed by the addition of two bromine atoms to fumaric acid; the formation of dibromsuccinic acid from brommaleic acid and of isodibromsuccinic acid from bromfumaric acid by the action of fuming hydrobromic acid; the conversion of brommaleic acid into fumaric and then into succinic acid by the action of sodium amalgam; the formation of one and the same tribromsuccinic acid by the action of bromine on brommaleic and on bromfumaric acid; and finally, the conversion of maleic into inactive tartaric acid, and of fumaric into racemic acid by potassium permanganate.

All these facts are shown to find a ready explanation by the aid of the new hypothesis. Further, it is shown that the decompositions of the salts of certain halogen derivatives of organic acids, which give up halogen salt and carbon dioxide, as well as the formation of lactones and of anhydrides of dibasic acids, are in perfect harmony with the hypothesis. But the only way to get a clear conception in regard to the mass of material which the author has brought together and which he has shown to support his hypothesis is by a careful study of the original paper, and the object of this notice is mainly to call the attention of American chemists to it.

As to the question what value to attach to the speculations which Wislicenus has brought to our notice, it is difficult to give any but a general answer. No one can well have a greater fear of mere speculation, which is indulged in independently of the facts, than the writer of this notice. Great harm has been done chemistry, and probably every other branch of knowledge, by unwarranted speculation, and every one who has looked into the matter knows how extremely difficult it is to emancipate one's self from the influence of a plausible hypothesis, even when it can be shown that it is not in accordance with the facts. It behooves every one, therefore, before accepting a new hypothesis, no matter how fascinating it may appear at first sight, to look carefully into the facts, and to endeavor to determine independently whether it is well founded or not. On the other hand, there is some danger to be apprehended from a tendency, sometimes observed, to denounce everything speculative, no matter how broad the basis of facts upon which it rests may be. Without legitimate speculation, it is clear that there could be no great progress in any subject.

As far as the hypothesis under consideration is concerned, the writer is firmly of the opinion that it is likely to prove of great value in dealing with a large number of chemical facts, and that, as it suggests many lines of research, it will undoubtedly in the course of a few years exert a profound influence on chemistry. Whether the evidence which will be accumulated will or will not confirm the view that the tetrahedron form is characteristic of the simplest molecules of carbon compounds is not the most important question to be asked under the circumstances. We should rather ask whether the testing of the hypothesis is or is not likely to bring us nearer to the truth. It is a proposition that admits of no denial that a hypothesis which can be tested by experiment, and which suggests lines of work and stimulates workers to follow them, is a gain to science, no matter what the ultimate fate of the hypothesis may be. - Amer. Chem. Jour.

[1]

Ueber die raumliche Anordnung der Atome in organischen Molekulen, and ihre Bestimmung in geometrisch-isomeren ungesattigten Verbindungen. Von Johannes Wislicenus. - Abhandlungen der mathemalisch-physischen Klasse der Konigl. Sachsischen Gesellschaft der Wissenechaften. Band XIV., No. 1.