RELATIVE PHOTOTROPISM. 355 



author's previous results it seemed to him that either this mixing 

 of the two responses, or perhaps some other unrecognized factor, 

 might have produced complications which would account for the 

 discrepancy in results. It was, therefore, determined to repeat 

 Cole's work, using so far as possible his own technique, but with 

 the addition of tests in horizontal tubes. The method employed 

 was as follows : 



A number of wild and vestigial flies were kept in bottles 

 containing a plentiful supply of food in which larvae were working, 

 until five days after hatching. On the eve of the sixth day from 

 ten to fifteen insects of each kind and of the same sex were 

 removed, etherized, and each animal placed in a separate vial 

 with a small amount of food. The next morning each insect 

 was given a preliminary test within its vial, to determine so 

 far as possible its general activity. This was done for the 

 vestigials by turning the glass one end up and then the other 

 four or five times. The five vestigial flies which responded most 

 readily by giving the negative reaction to gravity, and which 

 showed the greatest facility in crawling up the smooth glass of 

 the vials, were then selected for the critical tests. Five wild 

 flies were similarly chosen, except that their reaction to light 

 was used as a criterion of activity, and an effort was made to 

 pick those which reacted by crawling rather than by flight. In 

 this connection it might be thought that the use of light rather 

 than gravity as the stimulus for the winged flies would result in 

 the selection of particularly phototropic animals. However, 

 since virtually all wild type Drosophila are decidedly phototropic, 

 the fact that those which only crawled to the light were chosen, 

 would mean that if anything the less phototropic ones were 

 selected. This preliminary work was then immediately followed 

 by the critical tests of the ten flies in a manner to be described 

 below. A similar procedure was continued on successive days 

 until fifty wild males and fifty vestigial males had been tested; 

 the same was then done with fifty females of each kind, making 

 a total of 100 insects of each type. Each day's critical tests 

 involved the following apparatus and manipulation : 



Two tubes were used, one with no lining, and the other lined 

 with thin rice paper throughout the portion through which the 



