280 THE AMERICAN NATURALIST [Vol. LII 



1903; Kellogg, 1906; Wright, Lee and Pearson, 1907). 

 One need only mention Kellogg's work on bees as a type. 

 Kellogg assumed that if amphimixis were the principal 

 cause of the continuous variations postulated by Darwin 

 and Weismann as the most important source of material 

 for the use of the natural selection, 2 parthenogenetically 

 produced individuals should be less variable than those 

 produced sexually. lAJ statistical investigation showed, 

 however, that the characters of drones probably are more 

 variable than those of worker bees of the same race. 

 Since Kellogg believes Darwin's judgment that "males 

 vary more than females" to have been disapproved, he 

 concludes that "amphimixis is not only not necessary in 

 order to insure Darwinian variation, but there is no evi- 

 dence (that I am aware of) to show that it increases 

 variation. ' ' 



It is hardly necessary to point out here the numerous 

 mathematical and biological pitfalls which should be con- 

 sidered before one could accept as valid the statistical 

 differences that appear to exist when coefficients of varia- 

 tion based on such data are examined. It should suffice to 

 note that the researches of Wright, Lee and Pearson 

 (1907) on wasps of the species Vespa vulgaris showed 

 just as great a difference in variability between workers 

 and drones in favor of the former. Apparently, the sta- 

 tistics in these two nearly related groups lead to opposite 

 conclusions ; in reality probably neither statistical differ- 

 ence is significant as far as the question we are discussing 

 is concerned. The only conclusion justified by such data 

 would seem to be that the coefficient of heredity is as high 

 in the production of asexual as it is in the production of 

 sexual forms. 



Moreover, one can not expect anything more definite 

 from this method of attack. Biologists may differ as to 



spores. Some preparation often occurs^hrouglfthe emission of one polar 

 body. This may be merely a kind of recapitulation, a vestigial process no 



