120 



J. PAUL VISSCHER. 



3. EXPERIMENTS ON SELECTION OF FOOD. 

 A. Inanimate Substances. 



In experiments on inanimate substances it was desired to use 

 only insoluble and non-toxic materials. The following substances 

 were tested : carmine, chalk, sand, powdered glass, and india ink. 

 Three of these, namely, carmine, glass, and india ink, appeared to 

 be more favorable than the others, because of the fact that vacuoles 

 filled with these substances were readily distinguishable. The re- 

 sults obtained in ten experiments with each of these three sub- 

 stances are given in Tables I., II., and III. 



Table I. contains the results obtained with carmine. This table 



TABLE I. 



EXPERIMENTS ON CARMINE. 



Table showing results obtained in feeding Dileptus on carmine. In the five 

 columns under the headings 0-4, are indicated the number of individuals which 

 formed o, 1,2, 3, 4, or more vacuoles containing carmine, respectively, during 

 the twenty minutes of the experiment. Carmine was ingested by 97.5 per cent, 

 of the individuals but in only 17 per cent, was there more than one vacuole 

 containing carmine formed. 



shows that only three of the one hundred and forty-nine indi- 

 viduals tested did not ingest carmine, that more than eighty per 

 cent, of the total number tested formed one vacuole, while fifteen 

 per cent, formed two vacuoles each. Thus it is evident that car- 

 mine is eaten, but only in small quantities, for in only seventeen 



