OF ARTS AND SCIENCES. 109 



The distillate, as above with steam, is at first perfectly colorless, 

 but becomes light blue in a few minutes, and in an hour is changed 

 to a deep dark blue liquid, which gets darker and darker; in the 

 course of a few weeks it becomes viscous, and at the same time 

 purple red needles begin to appear. After three months standing, 

 the whole mass is changed to a solid resin, which, brought on clay- 

 plates and pulverized, goes over into a brown powder. This is 

 probably a polymeric product, and from the above it is evident 

 that phenylisocyanide is an extremely unstable substance. For all 

 the reactions mentioned in this paper freshly distilled phenylisocy- 

 anide must therefore be used. At higher temperatures, already 

 at 100°, polymerization takes place very quickly. At ordinary 

 pressure phenylisocyanide boils at 165°-166°, and although at first 

 a small portion goes over colorless, much polymerization takes place. 



Hofmann's product evidently contained some aniline, since he 

 says * that on distillation it is converted into an odorless substance 

 crystallizing in needles and partly volatilizing at 230°. The latter 

 substance was in all probability diphenylformamidine formed by 

 the action of the aniline present on phenylisocyanide (see below). 

 Weith f obtained the same substance on heating phenylisocyanide 

 in sealed tubes at 220°. 



The isonitriles $ all burn with very great difficulty, § because of 

 the formation of polymeric products, and it was possible to obtain 

 good analytical results only by mixing with lead chromate. The 

 nitrogen determinations also come low, unless the substance is 

 mixed with copper oxide, and heated long and very hard (see 

 analyses I. and II.). 



0.1145 gram substance gave 0.3411 gram C0 2 and 0.0554 gram 



H 2 0. 

 0.1452 gram substance gave 16.2 c.c. moist nitrogen at 18° and 



757 mm. 

 0.1803 gram substance gave 22 c.c. moist nitrogen at 20° and 



754 mm. 



* Ann. Chem. (Liebig), CXLIV. 118. 



t Ber. d. chem. Ges., VI. 213. 



f Hofmann, Ann. Chem. (Liebig), CXLIV. 117, has analyzed phenylisocy- 

 anide, but he gives no analytical data. 



§ It is therefore possible that azulmic acid, which is formed by the polym- 

 erization of prussic acid, is free from oxygen, especially since no concordant 

 results have been obtained in its analysis by different chemists. Cf. Gautier, 

 loc. at., XVII. 158 and 161. 



