1913] Johnson: Pigment Formalioii in Amphihinii Larvae 67 



pigment formation by a decrease in the amount or power of the 

 oxidase or by a decrea.se in the amount of the chromogen or by 

 the presence of something which inhibits the action of the oxidase. 



The oxidase in these experiments was certainly not admin- 

 istered to the animal directly, since all of the foods were boiled 

 before being given, and oxidases are destroyed at this tempera- 

 ture. Therefore the oxidase was produced by the organism. If 

 the difference in pigmentation was due to a greater amount of 

 oxidase, the later could have been furnished only indirectly by 

 the foods. 



In considering the amount of chromogen contributed by the 

 foods, one must consider their proteid content, since tyrosm, a 

 common chromogen, is a decomposition product of the digested 

 proteids. A study of the analyses of the various foods shows that 

 the pigmentation is not correlated with the amount of proteid, 

 since egg yolk which produced little pigment contains a large 

 percentage of proteid, larger than beans, which produced more 

 pigment. From the table it will be seen that while pigmentation 

 may depend upon the kind of proteid fed, it cannot depend 

 upon the gross amount of proteid (table 4). A few tabulations 

 of the cleavage products of these foods are available (see table 

 10). It is notable that in these ov-albumen shows a larger per- 

 centage (2.4%) of ty rosin than does vitellin (1.6%). The 

 differences are very slight, but it will be remembered that the 

 differences in pigmentation are also small, and that tyrosm is so 

 slightly soluble in water that artificial melanins are formed with 

 very dilute solutions of the chromogen. 



Table 4 

 Analyses of the foods used in experiments, from Hammarsten, ' ' A Text 

 Book of Physiological Chemistry," trans, by Mandel (1911, p.882-3). 

 1000 parts contain Relation of 1 :2 :3 



