^4-0 Aaron Franklm Shiill 



tain necessary exceptions. The experiment was performed twice, 

 A and B, Table XIV. In A, it was necessary to use one eighth 

 member, and in B one twelfth and one eighteenth member instead 

 of a fifteenth. In B, one second-born was used instead of a first- 

 born. 



From these two experiments it appears that breeding from the 

 later parts of the family results in fewer male-producers unless 

 the percentage of male-producers yielded by the first-born is 

 very low. 



Influence of Size of Family on Percentage of Male-producers 



Whitney ('07, p. 13) endeavored to explain Maupas's high 

 percentage of male-producers in his temperature experiments as 

 due in part to the shortening of the families. He plotted the 

 position of the male-producers in 23 families, and found the great 

 majority of them to appear in the first two-thirds ot the family. 

 If the conditions of the experiment curtailed the family by simply 

 omitting the last third of each one, which consisted almost wholly 

 of female-producers, the proportion of male-producers would be 

 greatly increased. 



As Whitney's conclusion was based on a small number of fami- 

 lies, I have collected data from 349 families, varying in size from 

 II to 56, comprising about 12000 individuals, bred during the 

 summer and early autumn of 1909. Table XV groups this data 

 according to the size of the family. 



An examination of any one of these groups of famihes plainly 

 shows that there is no accumulation of male-producers near either 

 end of the family. The small numbers in the last four places in 

 each group are in part due to the fact that most of the families 

 did not reach the maximum length of their respective groups. 



It is conceivable that the difference in the proportion of male- 

 producers caused by starvation is referable to the shortened fami- 

 lies. In general, partial starvation was found to increase the pro- 

 portion of male-producers. It might be supposed that the num- 

 ber and position of the male-producers in the family was prede- 

 termined; if the middle third of the family were destined to be 



