THE AMERICAN NATURALIST [Vol. XLIII 



The results given in the following table are of feeding 

 tests made in order to see how much and how often a toad 

 will eat. The tests were made with sweepings and, as 

 grasshoppers formed the greater part, the experiment was 

 carried on largely with them. This, I think, does not af- 

 fect the experiment to any great extent, since toads show 

 very little preference, if any, for certain species of insects. 

 If toads are placed in cages with sweepings, various spe- 

 cies are taken at random, provided only that they be mov- 

 ing and of a size easily swallowed. 



Ten toads, half males and half females, were chosen 

 for each test. Each toad was confined separately in a 

 glass cage (4 X 5 X 7 inches) in which was about two 

 inches of moist earth. Throughout test no. I, insects 

 were kept constantly before the toads. The insects were 

 weighed and counted, then put into the cages with the 

 toads. In test no. II the toads were allowed to feed, one 

 at a time, in a wire cage containing insects. The cage 

 with the insects was weighed before the toad was put in 

 and then weighed again after the toad was removed. 

 Allowance had to be made for loss by evaporation in de- 

 termining the amount eaten. A more satisfactory cage 

 was constructed entirely of glass for test no. III. It 

 measured 8 X 10 X 10 inches and had on its bottom a half 

 inch of dry sand for the absorption of moisture. Evapo- 

 ration from this cage was less than a gram in ten hours. 

 Insects were placed in it and the process of weighing 

 followed as in test no. II. 



In nos. I and III the toads were not disturbed unless 

 they were out of the earth. But in no. II they were dug 

 up every day, if necessary, in order that they be offered 

 food at least once. 



The following are the results of the three feeding tests. 



The data show that for test no. II the toads feed oftener 

 than in the other two experiments, but that the amount 

 eaten, m proportion to the weight of the toads, is 3 and 

 2 per cent, less than in tests nos. I and III, respectivelv. 



The time represented in these experiments is equal to 

 that of one toad for 260 days, 96 of which were passed 



