1916] Beeves: Inheritance of Extra Bristles in Drosophila 513 



also have some influence on the factor or factors involved; or it may 

 simply be in agreement with Stark's (1915) work showing the in- 

 creased occurrence of lethal factors in inbred stock. In that case there 

 should be the same results in normals, as they were also inbred. 

 Hyde's (1914) conclusions do not seem to have any bearing on the 

 results found. The sterile females in his experiments contained eggs, 

 but these were not laid, apparently owing to an obstruction in the 

 oviduct. In all cases examined in my cultures, however, no eggs were 

 found in sterile females. 



The results of selection only in part agree with the work of Zeleny 

 and Mattoon (1915). They disagree in that selection does not increase 

 the mean number, but agree in that the individuals in any genera- 

 tion differ as regards germinal constitution. Also the conclusion is 

 the same, namely, that probably other factors were concerned, or the 

 variability is due to the presence of original differences in factorial 

 composition, rather than to the addition of new ones or the loss of 

 any factor. 



Selection of extra bristles did not result in the establishment of a 

 pure line as by Lutz (1911) in his experiments in selection of abnormal 

 venation of the wings of Drosophila. Nor do our results in selection 

 compare with Dexter 's (1914) results in selection on beaded stock. 

 He established a line which bred practically pure. However, if un- 

 selected it returned to the normal condition. Results similar to those 

 of Lutz were obtained as regards position of inheritance, namely, that 

 there was no relation of offspring to parent in regard to the position 

 of inheritance. As in the case of Tower's (1906) experiments on 

 Leptinotarsa, if .selection was stopped the flies returned to the normal 

 mean. 



Lethal factors, while probably present, yet seemed to have no 

 influence on the sex ratios, but the fact that the extras were less viable 

 might support the theory of Liff (1915) in his paper on a peculiar 

 Mendelian ratio. He was working with a pink-eyed mutant and the 

 peculiar ratios were found to be due to lethal factors. Morgan has 

 done extensive work with lethal factors appearing in Drosophila. Per- 

 haps in this case the lethal factor was introduced by both extra- 

 determining chromosomes, and that is why in the hatching we have 

 such small numbers. 



It is not possible from the results to decide finally either in favor 

 of a Mendelian interpretation or of a fluctuating variation, as neither 

 explanation is completely satisfactory. 



