249 



The Effect of Artificial Selection on Bristle Number 

 IN Drosophila ampelophila. 



Fernandus Payne — Indiana University. 



The following brief abstract gives a summary of the results ob- 

 tained in an experiment designed to test the effect of artificial selection 

 on bristle number in Drosophila ampelophila, and to find out in what 

 way selection is active. 



The normal number of bristles on the scutellum is four. In a mass 

 culture which had been bred in the laboratory about three months, a 

 female was found with one extra bristle, or five in all. This female 

 was mated to a male from the same mass culture. Of the F ofi^spring, 

 two females had five bristles. These two females were mated to their 

 normal brothers, and gave in F', 935 normal flies, thirty-nine with 

 five bristles, and four with six bristles. The flies with extra bristles 

 were again mated and this method of selecting the high-grade par- 

 ent has been continued throughout the experiment. The per cent of 

 extra bristled flies and the mean bristle number have been gradually 

 increased until in the last generations of selection no normal flies were 

 found and the mean reached 9.089 in the twenty-eighth generation. 

 From the twenty-eighth to the thirty-eighth generations, the mean re- 

 mained practically the same. A back selection line started from the 

 eleventh generation was without effect. 



Selection then has produced decided results. The larger question 

 is, how have the results been produced? Have they been produced by 

 selecting somatic variations, by selecting the variations of the gene 

 which stands for bristle number, or have they been produced by piling 

 up or getting rid of modifying factors? The first possibility can be 

 dismissed without much consideration, as any character which is in- 

 herited must be germinal. Of the other two possibilities, my evidence 

 is in favor of the latter. It shows quite conclusively, I think, that 

 there is a factor in the X-chromosome and also one in the third chromo- 

 some which modifies bristle number. There may be more than two 

 such factors. One was no doubt present at the beginning of the experi- 

 ment. The others probably occurred as mutations during the course 

 of selection. 



