On the Proteid Reaction of Adamkiewicz, <&c. 



27 



was slowly distilled nearly to dryness, and the distillate returned and 

 again distilled. The second or third distillate usually showed freedom 

 from peroxide when tested with chromic acid • if not, distillation was 

 repeated. 



One-tenth of the final distillate was set aside, and the remainder 

 neutralised with potash. The still acid portion being then mixed with 

 the rest, the whole was distilled as low as possible, avoiding, however, 

 any separation of potassium acetate in the retort. The distillate 

 always gave an abundant proteid reaction, and if any trace of free 

 peroxide had been left at the previous stage, it always disappeared 

 during the distillation of the partially neutralised mixture as just 

 described. A small trace of free peroxide will interfere with the 

 proteid reaction. On adding phenyl hydrazine hydrochloride (without 

 acetate) to the distillate thus obtained, a light yellow precipitate 

 begins to separate almost at once, and after standing it becomes con- 

 siderable in amount, and is crystalline. But although, as we were able 

 to show, the hydrazone of glyoxylic acid is present in this precipitate, 

 it is mixed with a considerable proportion of a compound much less 

 soluble in acetic ether and in hot water. If the original precipitate be 

 recrystallised from a minimal quantity of acetic ether, the substance 

 which separates first consists of perfectly colourless glistening plates, 

 which after recrystallising from acetic ether may assume the form of 

 rosetted prismatic needles. These melt sharply at 184°. 



The nature of this substance became clear after the publication of 

 certain recent observations. Gerhard Ollendorff has shown that 

 formic aldehyde is formed when glycollic acid is oxidised with per- 

 oxide of hydrogen, and Fenton* calls attention to the fact that gly- 

 oxylic acid must in this case be the intermediate product. The 

 product we obtained from acetic acid was undoubtedly the compound 

 of formaldehyde described by Wellington and Tollens.f 



A portion repeatedly recrystallised from acetic ether and showing a 

 constant melting point (184°) was analysed. 



0-147 gramme gave 27'4 c.c. moist N, at 12°, and 758 mm. N = 

 22-07 per cent. 



Another preparation, recrystallised from a mixture of alcohol and 

 oluol, melted at 182—183°; of this 

 0*21 1 gramme gave 39"3 c.c. moist N, at 14°, and 758 mm. N = 

 21-87 per cent. 



Calculated for 

 i. n. C 15 H 16 N 4 . 



22-07 21-87 22-22 



This hydrazone can be obtained pure in the above manner with 



* Fenton, 1 Journ. Chem. Soe.,' 1900, vol. 77, p. 1296. 

 f 'Deutsch. Chem. Gres. Bericlite,' 1835, vol. 18, p. 3330. 



