82 ANNALS OF THE MISSOURI BOTANICAL GARDEN 



[Vol 



red or black coloration with tyrosin solutions, which may be 

 precipitated in the same fraction with the deaminase. 



Although toluol was used to cover the surface of the solu- 

 tions, in one of the flasks which stood for several days be- 

 tween the beginning of the experiment and the analysis, the 

 toluol evaporated and a slight black pellicle was formed on 

 the surface. As the solutions had been plugged with cot- 

 ton, suspicion was aroused and nutrient agar plates were 

 poured from each flask, although toluol still covered the sur- 

 face of the remaining flasks. The plates were virtually pure 

 cultures of Bacillus sp. which were shown to produce ammonia 

 rapidly in Dunham's peptone solution. Hence I think the am- 

 monia found in all of the experiments was the result of bacte- 

 rial action, since it never formed a constant portion of the 

 amino-nitrogen present nor of the nitrogen which did not 

 enter into the reaction with nitrous acid under the conditions 

 of a Van Slyke determination of amino-nitrogen. These bac- 

 teria have thus far withstood all attempts to kill them without 

 injuring the enzyme, resisting thymol-chloroform, alcohol, 

 heat, and poisons. 



The percentages of nitrogen loss and gain shown in table 

 ii were computed on the basis of tyrosin nitrogen only, in 

 order to find a common measure to study the question of 

 deamination. This is not wholly desirable, as the solutions 

 contained some amino- and ammonia-nitrogen extracted from 

 the fungous flour, but as the possible bacterial action was un- 

 known, it is felt that the figures furnish as reliable data as 

 could be obtained under the circumstances. 



The enzyme preparations were also tried with glycin, 

 leucin, alanin, and phenylalanin. Here again the amount of 

 ammonia found bears no relation to the amount of amino- 

 nitrogen lost after the proper corrections have been made, 

 and no ammonia was found in the solutions kept at ± 5° C. 

 for two weeks, where bacterial action would be extremely 

 slow, although a few colonies were isolated from these flasks. 

 Here, however, the enzyme action would also be slower if it 

 followed the Van't Hoff rule of temperature coefficients, but 

 a characteristic tyrosinase coloration was obtained in the 



