Nos. 618-619] THE BOLE OF REPRODUCTION 



281 



the definition of fluctuation, mutation, etc., but they are 

 generally agreed that germinal variations, be they great 

 or small, are inmost species so rare they can not be gauged 

 by the use of ordinary statistical methods. For this rea- 

 son, a comparison between the variability of the drones 

 and of the workers of a pure race of bees is not likely to 

 show any difference between these two modes of repro- 

 duction in the matter of the frequency or the type of the 

 germinal variation produced, and can not answer the ques- 

 tion as to whether sexual reproduction contributes more 

 material for the use of natural selection than asexual re- 

 production. A study of variability in crossed races, 

 where the effect of Mendelian recombination can be con- 

 sidered, would be a more logical attack upon the second 

 problem, but is hardly necessary in view of the other 

 evidence available. 



One is then justified in claiming there is no experimental 

 evidence to show that sexual reproduction in itself is not 

 an exact equivalent of asexual reproduction in the matter 

 of a heredity coefficient, but is this also true for germinal 

 variation? We believe it is. Variations there are in 

 both asexual and sexual reproduction, but it can not be 

 maintained that they occur more frequently in the latter. 

 There are insects in Oligocene amber apparently identical 

 with those of to-day, proving that constancy of type is 

 possible through long periods of time under sexual repro- 

 duction; yet germinal variations occur to-day in some- 

 what noteworthy numbers, as Morgan's work on Dro- 

 sophila shows, although the proportion of these varia- 

 tions which show possibilities of having an evolutionary 

 value, as evidenced by persistence in natural types, is 

 probably small. On the other hand, the number of varia- 

 tions produced under the dominance of asexual repro- 

 duction can not be said to be less numerous, even among 

 organisms of a relatively high specialization. If there 

 are those who doubt this statement, let them refer to the 

 immense list of bud-variations in the higher plants com- 

 piled by Cramer (1907). 



There would be little reason in pushing the claims 



