SEX-RATIO IN DROSOPHILA AMPELOPHILA. 361 



In the latter part of this experiment there was a diveation from 

 the usual methods. The element of time was sacrificed to be 

 able to make a larger number of matings in each generation. 

 A number of males and females were isolated from each mating, 

 being examined, and held to be mated after all of the counts for 

 the generation were finished. By this method it was necessary 

 to make matings only from the cultures which had given the 

 most desirable ratios but in so doing a considerable amount of 

 time was lost between each generation. 



This third attempt to modify, by selection, the sex-ratio did 

 not result in bringing about any very significant difference 

 between the two strains. (Tables IX. and X. give the results 

 of selection upon this stock.) The ratios computed from the 

 totals in each strain showed a slight difference in the desired 

 direction but this was so slight that it might be due to chance. 

 Therefore it seems that no definite conclusions can be drawn 

 from this experiment. In the low strain the totals were 9,673 

 males to 9,951 females, or a ratio of i: 1.0287 and in the high 

 strain 12,327 males to 12,898 females, or a ratio of 1 : 1.0463. 



CONCLUSION. 



Since three distinct and unrelated stocks of flies show no sig- 

 nificant effect of selection, it seems safe to conclude from these 

 data that Moenkhaus's conclusion concerning the amenability 

 of the sex-ratio in Drosophila ampelophila will not hold. 



DISCUSSION OF MOENKHAUS'S WORK. 



Moenkhaus concludes from his work on the effect of selection 

 on sex-ratios in Drosophila ampelophila that sex-ratios are 

 "strongly transmissible and amenable to the process of selection." 

 From his paper it is somewhat difficult to tell whether Moenkhaus 

 intended to say that he had developed by selection, strains char- 

 acterized by high and low female ratios or that by selection, he 

 simply maintained a high and low strain 'which he found in 

 nature. His results are given in the discussion which follows. 



By way of explanation it might be said that Moenkhaus called 

 the strain in which he selected for a relatively higher female 

 ratio the "female" strain, and the one in which he selected in 



