34^ Aaron Franklin Shull 



experiments of my own which, because they are too few to be con- 

 clusive and because other experiments along the same lines are 

 still in progress, are reserved for a future paper, seem to indicate 

 that temperature has some influence. But as it is probable that the 

 details of Maupas's conclusion can not stand, I have been led to 

 seek for an explanation of his results. Whitney, it will be remem- 

 bered, has already oflFered two possible explanations. First, he dis- 

 covered that at high temperatures a male-producer laid two to 

 four times as many eggs as did female-producers at the same 

 temperature. Maupas probably supposed that the out-put of 

 eggs was the same for each, hence his 97 per cent was in part 

 accounted for, Whitney believes, by larger families. This explana- 

 tion, however, would only account for an excess of males, whereas 

 Maupas obtained an excess of male-producers. It made no diff'er- 

 ence in Maupas's experiments whether a male-producer laid 

 15 eggs or 50, she counted only one toward the QJ per cent in the 

 result. To sustain Whitney's point it would be necessary to show 

 that a female whose ofi^spring are largely male-producers lays 

 more eggs at a high temperature than does a female whose off^- 

 spring are largely female-producers. Such evidence is not, I be- 

 lieve, forthcoming. 



The second explanation ofi^ered by Whitney to account for 

 Maupas's results was that at a high temperature shorter families 

 were produced than at a low temperature. He found from an 

 examination of 23 families that the male-producers occurred 

 chiefly in the first two-thirds of their respective families. If these 

 families were shortened by cutting off the last members, which 

 were nearly all female-producers, the percentage of male-producers 

 would be increased. It appears, however, from the examina- 

 tion of a very much larger number of families (Table XV) that 

 the male-producers are not accumulated at either end of the fami- 

 ly. It also appears that a short family is not a long family minus 

 its last portion. Families containing 46 to 50 members have their 

 maximum number of male producers among the 25th to 30th 

 members. If a family of 31 to 35 were the same as a family of 

 46 to 50 with its last 15 members omitted, the maximum number 



