288 Cardiff : Synapsis and reduction 



* 



Later there is a transverse segmentation (figure 48^) to form the 

 chromosomes of the immediate prophase. Several of these still 

 show their bivalent character though not to such a marked extent 

 as in the case of Acer (figure 49). 



What may be the significance of the small spherical bodies in 

 the nucleus at this stage (figure 49^) was not definitely determined, 

 though it seems probable that they are a disorganized nucleolus, as 

 there appears to be no regularity in their number. The number of 

 the chromosomes is probably seven or eight though not enough 

 counts were made in this case to determine definitely. That there 

 are individual differences as to size, shape, etc., which are constant 

 in the different cells, seems evident. As the chromosomes are 

 arranged in the plate at metaphase they have the short, thick ap- 

 pearance of the usual heterotypic chromosomes (figures 50 and 

 5 1). But few of them show a bivalent character and it is impos- 

 sible to tell whether they are divided along the line of their origi- 

 nal conjugation or not. 



One point of interest in the metaphase of the heterotypic divis- 

 ion is that in a very large number of the cells one of the chromo- 

 somes appears to pass undivided to one pole considerably sooner 

 than the others (figures 50 and 51), seeming, generally, to reach 

 the pole about the time that the split is complete in the other 

 chromosomes (figure 51). When first noticed it was thought the 

 position of this chromosome was due to faulty sectioning. Upon 

 closer examination, however, it was seen that this was not the case, 

 but that the behavior of this chromosome is undoubtedly different 

 from that of the others. What the significance of this is I hope 

 to find by later work. The fact undoubtedly suggests a compari- 

 son of this chromosome with the "accessory" or heterotropic 

 chromosome found in the spermatogenesis of Orthoptera, Hem- 

 iptera, and other insects, but in view of Wilson's ('050, '05^ '06) 

 results regarding the relation of this chromosome to sex produc- 

 tion it seems improbable that it should occur in an organism which 

 is essentially hermaphrodite. No chromosomes exactly similar to 

 this were noted in the other forms studied, though in many of them 

 could be found a chromosome that divided earlier than the others. 

 These may possibly correspond to the idiochromosomes of Wilson 

 but there is no direct evidence of this. If the heterotropic chromo- 



