RELATIVE PHOTOTROPISM. 363 



further emphasis by one other observation. It was noted in 

 the horizontal tests of vestigial animals that a fly which had 

 reached the light end of the tube would often turn immediately 

 about and crawl quite as rapidly all the way to the dark end. 

 Such a performance, though it would appear very favorably on 

 the record, would obviously not denote any great degree of 

 phototropism. This occurred, however, in at least one of the 

 trials of each of 19 individuals. On the other hand, such action 

 was noted in only two trials of wild type animals. In the case 

 of vertical tests it never occurred with vestigials, but one of the 

 tw r o instances of wild flies took place in such a test. The reason 

 for this single instance is not clear, and in the present state of 

 our knowledge can only be ascribed to the erratic tendency 

 already noted in Drosophila. In a similar category, perhaps, 

 must be put the behavior of 9 vestigial insects which, though 

 recorded as active, made relatively poor records in the vertical 

 tests. 



How the results and conclusions indicated by Cole can be 

 reconciled with those herein presented it is rather difficult to see. 

 In any event it is perhaps worth while summing up certain 

 outstanding points which the two sets of experiments seem to 

 bring out. In the first place it does not appear that the mixing 

 of the light and gravity reactions, which was suggested as a 

 possible source of error, has any effect upon the main question 

 at issue. Secondly, though also apparently without bearing 

 upon this question, is the fact that Cole's wild type flies in 

 vertical tubes seem to have been much faster than those of the 

 writer. With 1,500 candle meters 50 per cent, of Cole's animals 

 averaged 6.17 seconds and with 750 candle meters 7.6 seconds. 

 Upon the other hand, the best average obtained by the writer 

 was 10.5 seconds, the remainder of them being from two to four 

 seconds slower, under a stimulus of 1,238 candle meters. Such 

 a discrepancy almost causes one to wonder whether there may 

 not have been some fundamental difference in the character of 

 the insects used or in the experimental conditions. Thus, for 

 example, the tests herein recorded were conducted at an average 

 temperature of 24.1 degrees centigrade, and it is known in a 

 general way that the activity of Drosophila varies more or less 



