60 University of California Publications in Zoology [Vol. 11 



granules of the ehromatophores acted as reserve material in eases 

 of inanition. 



Experiments similar to these of Tornier were carried on by 

 me during two successive seasons with Hyla and Bana tadpoles. 

 The results do not confirm those of Tornier, since the tadpoles 

 receiving a small amount of food were no lighter than those 

 receiving an abundance. 



The tadpoles from a single egg mass were divided into three 

 lots and each lot given a different amount of food. As it is 

 impossible when feeding graduated amounts of food to be sure 

 that the tadpoles in a given dish share the food alike, instead of 

 placing different amounts of food in the different dishes, the food 

 was withheld from the di.shes for different length.s of time. On 

 days when the tadpoles received food they were given as much 

 as they could eat, and on other days all food was removed from 

 the dishes. One set of tadpoles was fed every day, the food of 

 the second set was withheld every third day and that of the third 

 set was withheld every second day. The three sets of tadpoles 

 thus received large, medium, and small amounts of food approxi- 

 mately in the ratio of 1 : % : I/2. 



The growth of the tadpoles was proportional to the amount 

 of food received and the amount of pigment seemed also to be 

 proportioned to size since the large, medium, and small tadpoles 

 on the same diet were all of the same color, those that were half 

 starved being no lighter than those that were large and ivell fed. 

 In order that this might not be a prejudiced judgment the dif- 

 ferent sets of tadpoles were repeatedly submitted to other 

 observers before explaining the object of the experiment. The 

 verdict always was that the tadpoles were all the same color or 

 that the smaller ones were a little darker. The experiments were 

 performed for two successive years with the same result (table 1) . 



