36 THE ROYAL SOCIETY OF CANADA 



tion, and the analogy with the murexide reaction is increased by the 

 fact that murexide gives a similar absorption band. 



The blue color given by hydrindantin and potassium or sodium 

 hydroxide differs in dilute solution very markedly from diketohydrin- 

 dylidene-diketohydrindamine ammonium salt. Thus 



(1) It gives no absorption band in the visible spectrum. 



(2) It remains blue when viewed by artificial light. 



(3) It is extremely readily oxidised to a colorless solution when 

 shaken with air. 



These tests enable one to distinguish between the two blue 

 colorations even in very dilute solution. 



THE REACTION BETWEEN TRIKETOHYDRINDENE HYDRATE AND 

 AMMONIUM SALTS. 



The results of Neuberg, who in contradiction to the earlier 

 results of Abderhalden and Schmidt, showed that many ammonium 

 salts gave a strong positive color reaction with triketohydrindene 

 hydrate, is of the highest importance as it destroys at once the specific 

 nature of the test as a means of recognising amino acids. Herzfeld 

 had also noticed that ammonium carbonate and ammonium oxalate 

 gave colorations with the triketone, and in the case of the latter salt, 

 during some of the very early work on this reaction in these laboratories 

 we ourselves had obtained contradictory results, By working quanti- 

 tatively, however, it has been found possible to explain these discrep- 

 ancies. We have examined the action of nearly twenty am-^ onium 

 salts of different types with uniform results, which can be expressed 

 as follows: — 



(1) In a concentration of one per cent, the ammonium salts of 

 weak acids react positively with triketohydrindene hydrate. 



(2) In very high concentration, the ammonium salts of strong 

 mineral acids give a positive reaction with "ninhydrin." 



(3) In a concentration of 1 c.c. = -05 mgs. nitrogen, no reaction 

 is obtained. 



(4) In presence of pyridine and in concentration of 1 c.c. = 

 •05 mgs. nitrogen, all ammonium salts react positively with triketo- 

 hydrindene hydrate. The amount of decomposition per c.c. is con- 

 stant at 0-019 mgs. nitrogen, and the blue coloration is due to the 

 ammonium salt of diketohydrindylidene-diketohydrindamine. 



It is at first sight inexplicable that a salt such as ammonium 

 chloride should give the same reaction as an amino acid like alanine, 

 for the mechanism demanded by Ruhemann's constitution of the blue 



