126 George Lefevre 



plate, is shown in Fig. 29. The chromosomes divide typically, 

 pass to the poles, and in the telophase are converted into vesicles 

 from v^hich the daughter nuclei arise by fusion (Fig. 30). The 

 division of the egg into two equal blastomeres now takes place. 

 During the anaphase, the centrosomes at each pole begin to sepa- 

 rate, and soon delicate radiations begin to arise around each as a 

 new center before the original rays have disappeared (Fig. 30). 

 The difficulties in the way of tracing the cell lineage of the 

 parthenogenetic eggs are so great that a detailed comparison with 

 the normal cleavage and a determination of the origin and fate of 

 the constituent cells of the embryo are rendeed practically im- 

 possible. Owing to irregularity in the rate of division and the 

 possible presence of greater or less abnormalities at all times in 

 development, the arranging of stages that appear to be normal in 

 their proper sequence can only be attended at best by uncertain 

 results. I shall, therefore, not attempt a continuous account of 

 the development as far as I have been able to follow it, as any 

 reconstruction of stages would be quite arbitrary and would rest 

 almost entirely on the identifiction of individual embryos with 

 normal ones of a corresponding degree of differentiation. Al- 

 though in favorable experiments, where optimum solutions were 

 used, I have met with no such "carnival of development" as many 

 other experimenters in artificial parthenogenesis have observed, 

 and although the proportion of embryos that are either normal or 

 nearly so is very large, nevertheless abnormalities of all kinds are 

 of not infrequent occurrence, and the determination in a given 

 case as to whether an embryo is normal or not is by no means easy. 

 But in spite of these difficulties, I am convinced that the partheno- 

 genetic eggs in a great many cases undergo a normal development 

 and exhibit the usual processes of differentiation leading up to the 

 formation of the swimming larva. The only differences in such 

 cases between the organisms which are produced parthenogen- 

 etically and those arising from fertilized eggs are to be found in the 

 numerical relations of the chromosomes, as I shall point out be- 

 yond, in the rate of development and in the fact that the partheno- 

 genetic larvae do not rise to the surface of the water when they 

 begin to swim. 



