pounds formed from anthracen by the addition of hydrogen and bromine behave like the corresponding derivatives of benzol and naphthalin. All the considerations, therefore, which go to prove that benzol consists of one closed circle and naphthalin of two, apply also to anthracen, and are opposed to the assumption of saturated or non-saturated lateral chains. Supported on the assumption founded in the foregoing as to the constitution of anthracen, we may take a further step, and consider the position of the atoms in the substitutionproducts of anthracen. Alizarin not only yields phthalic acid when oxidised, but can also be obtained from it synthetically. Hence it follows that the two atoms of oxygen must be combined with the two central atoms of carbon, and as alizarin is formed from phthalic acid and pyrocatechin, it further follows that the two hydroxyls are contained in the same benzol nucleus. If the two hydroxyls and the atoms of oxygen are replaced by hydrogen we obtain anthracen, and its formula may then be written simply as follows, the atoms of hydrogen belonging to one and the same circle being written together : H CH H We have then for alizarin the following formula:— H1 This formula is confirmed by the formation of alizarin from phthalic acid and pyrocatechin, and shows also simply how phthalic and oxalic acids are formed from alizarin. Further, we know that the two hydroxyls assume the position 1, 2, as in pyrocatechin. From this formula for alizarin it follows that also in tetrabromanthracen the four atoms of bromine take the same position as the atoms of oxygen and the hydroxyls; that therefore two atoms of bromine are connected with the two interior atoms of carbon, and two to the outer ring in the position 1, 2. That the two atoms of oxygen are linked to the interior atoms of carbon is no longer open to doubt, but it may perhaps be of interest to give the reasons which were formerly advanced in favour of this supposition. Benzol yields on direct oxidation no quinon; from naphthalin naphthaquinon may be with difficulty obtained in a direct manner; from anthracen the derivative in question may be obtained readily. This shows that a larger accumulation of carbon facilitates the formation of quinons. But in anthracen the two atoms of hydrogen belonging to the middle ring are in the greatest degree surrounded by carbon, and the ready formation of anthraquinon is therefore in favour of the above assumption, which is also supported by the great stability of anthraquinon. If the atoms of oxygen were contained in one of the outer nuclei, this would probably be more readily oxidised, and a naphthalin carbonic acid would be obtained. The formation of anthraquinon from bibromanthracen by means of oxidising agents, which do not convert bibrombenzol and bibromnaphthalin into the corresponding quinons, makes it probable that the atoms of bromine replace those atoms of hydrogen which take a peculiar position, and which in a certain manner are more aromatic than those of benzol and naphthalin. If these considerations are accepted, which may be done the more as there is much in their favour and nothing against them, the following formulæ may be assigned to the bromo and quinon derivatives of anthracen : Preparation and Properties of Anthracen. The synthetic methods for the preparation of anthracen have been indicated above. It is obtained, as was stated, on heating benzyl chloride with water to 180° in sealed tubes The action of water upon benzylchloride is perfectly analogous to that of zinc upon a mixture of benzylchloride and aromatic hydrocarbons. The water plays, in the first place, an introductory part; the reaction then begins between two molecules of benzylchloride, so that the benzol nucleus of the one molecule loses 1 H, and the other molecule 1 Cl, which both are thrown off as hydrochloric acid, whilst at the same time the chloride is formed. This latter enters again into reaction in various ways. It may next, losing HCl, become condensed, forming the hydrocarbon— CH-CH-CH-CH2 It may further react with one molecule of benzylchloride in the manner above described, forming a complicated chloride, CH–CH,-CH-CH,-CH-CH,CH, which, in turn, again undergoes the same reactions as the chloride of the second degree. There result, therefore, on the one hand, hydrocarbons of the composition CH, and on the other, chlorides of the composition C1,H3C1, C21H19C1, C28H25Cl, &c. 14 13 14 69 In the second place, the water, especially at elevated temperatures and on prolonged action, exerts another influence, it decomposes the chlorides formed, and converts a part of them into the corresponding alcohols and ethers. The ultimate product of the reaction is therefore a mixture of different hydrocarbons, chlorides, alcohols, and ethers, whose molecular weight increases with the duration of the reaction. Besides the chloride, C11H12Cl, and a hydrocarbon boiling at an elevated temperature, Zincke was unable to obtain any well-characterised compound; but all observations show that the reaction proceeds in the manner indicated above. Zincke became convinced, further, that in the crude product neither anthracen nor benzyltoluol are present, but that both these hydrocarbons are formed during distillation. The manner of their formation, indeed, is so peculiar and complicated, that it could not be discovered theoretically. Anthracen is formed on the distillation of its principal quantity, but decidedly only in the second place. It is no direct product of decomposition, but owes its formation to a previously formed and very complicated hydrocarbon, which is split up under the influence of heat into anthracen and other hydrocarbons. Benzyltoluol, on the other hand, is a direct product of decomposition, and is especially formed from the chloride, CH,,Cl, which is resolved into hydrochloric acid, benzylchloride, benzyltoluol, and hydrocarbons of high molecular weight; in small quantities it also seems 14 13 to be formed, along with anthracen, during the decomposition of the latter. The entire process of formation of anthracen and benzyltoluol from benzylchloride may be represented as follows:-When benzylchloride is heated with water there are formed the bodies mentioned above, among which the chloride, C1,H,,Cl, which boils in a vacuum at 204°-206° with partial decomposition, and hydrocarbons with a high boiling-point (from 280° to above 300° in a vacuum) predominate, whilst oxygenous bodies are only present to a small extent. On the distillation of these crude bodies, which in all cases should be freed from undecomposed benzyl chloride by means of a current of steam, there appear, if they are heated under ordinary atmospheric pressure, principally hydrochloric acid, water, benzylchloride, benzyltoluol, and resinous or tarry hydrocarbons, which on further heating yield anthracen and toluol, along with small quantities of other bodies. A small portion of the resinous or tarry hydrocarbons, which yield on decomposition anthracen, is possibly already present in the crude product. In any case, the latter approach the hydrocarbons formed directly from benzyl chloride. Like these, they correspond to the formula nC,H, and are only distinguished by their molecular weight. The decompositions above described may be traced very accurately in a vacuum, where the decomposition of the chloride C1HCl is less energetic, the formation of hydrochloric acid is less prominent; but water and benzylchloride with benzyltoluol, as well as the resinous hydrocarbons, are plentifully formed. The latter are generally obtained as solid masses, resembling colophonium, which melt below 100°, giving off a pleasant aromatic odour. In a vacuum they boil without decomposition, but if heated under atmospheric pressure they are split up into anthracen and toluol. The chloride C1,H,,Cl, as obtained by Zincke on distillation in vacuum, manifests similar phenomena of decomposition; if heated under atmospheric pressure, it yields hydro |