1913] Johnson: Pigment Formation in AmpMhian Larvae 65 



variations of colour when fed on cayenne pepper; but in all 

 cases the eifect is produced only if the pepper is given to the 

 young birds before their first moult. Moreover, he finds that a 

 moist atmosphere facilitates the change of colour, and that the 

 ruddy hue is discharged under the influence either of sunlight 

 or of cold. Lastly, he has observed that sundry other materials 

 .such as glycerine and aniline dyes, produce the same results ; so 

 there can be no doubt that organic compounds probably occur 

 in nature which are capable of directly affecting the colours of 

 plumage when eaten by birds." 



These facts indicate that significant variations in color are 

 produced by different kinds of food and the following experi- 

 ments with tadpoles bear out the suggestion. 



Both Bana and Eiila tadpoles were used for these experi- 

 ments. Different bunches of eggs were separated so that the 

 larvae used for one set of experiments were from a single egg 

 mass. As soon as the tadpoles hatched they were divided into 

 different groups and each group was fed on one kind of food. 

 The water in the dishes was changed every other day and the 

 tadpoles given fresh food. In the later experiments the table and 

 the sides of the dishes were covered with black paper. This 

 allowed the light to enter only from above and insured equal 

 illumination for all the dishes. The foods first used were beef 

 liver, egg yolk, egg albumen, beef suet, brown and white beans, 

 white bread, and fish. The foods were all cooked by boiling and 

 finely divided by being put through a sieve or rubbed between 

 the fingers. A few measurements given in the following tables 

 will show the size relations fairly well. The intensity of the 

 color is also noted (table 3). 



