74 PROCEEDINGS OF THE AMERICAN ACADEMY 



until the alcohol had evaporated, leaving the anthracene and picric 

 acid dissolved in the oily impurities. In point of fact, it was obtained 

 by imitating the above conditions ; that is, evaporating an alcoholic 

 solution of anthracene and picric acid, to which a few drops of toluol 

 had been added. The experiment would seem to show that phenan- 

 threne as such does not exist in the original product. 



To meet the objection, which might be urged, that the anthracene 

 obtained in the second experiment was not formed directly by the 

 action of sodium on the orthobrombenzylbromide, but during the dis- 

 tillation of the viscous mass, which resembled the substance obtained 

 by Zincke both from his reaction and the action of water at high tem- 

 peratures on benzylchloride, another portion was treated with ether 

 and hot alcohol, which removed a yellow oil and left anthracene iden 

 tified by crystallizing in white plates melting at 210°. The yellow oil 

 contained either phenanthrene or its dihydride,* as it gave phenan- 

 threnechinone by oxidation, but we were unable to determine which 

 of these substances was present on account of the small amount at 

 our disposal. We are therefore in doubt as to the way in which the 

 phenanthrene occurs in our product; for, on the one hand, we could 

 not obtain the picric acid compound either from the original product 

 or the yellow oil, while, on the other, a substance crystallizing in 

 plates melting between 170° and 190°, which has appeared more than 

 once in the course of this work as an intermediate product in the 

 purification of the anthracene, yields on oxidation anthrachinone and 

 phenantlirenechinone, and seems therefore to be a mixture of the two 

 free hydrocarbons. 



No anthracenedihydride could be foiind in any of the products, nor 

 have we succeeded in detecting benzyltoluol or ditolyl, altliough it is 

 highly probable that they were formed in addition to thedibenzyl; 

 but the acids resulting from the oxidation appeared in such small 

 quantities, and so contaminated with a brown impurity formed from 

 the viscous brominated substance, that it was impossible to obtain 

 any definite result from them. The viscous substances, too, containing 

 bromine, which were the principal products of the reaction, have re- 

 sisted thus far all our efforts to bring them into a form suitable for 

 analysis, but we hope to return to this part of the subject hereafter. 



* Plienantlirenediliytlride has never been prepared witli certainty, although 

 Barbier (Comptes Itendus, Ixxix. 121) thought it was formed by tlie reduction 

 of phenanthrene with sodium amalgam. 



