45 



two divisions clearly effect a double longitudinal fission of the original 

 bilobed chromosomes, D. H. Tennent ('06) describes the second 

 maturation mitosis as effecting a transverse division, but the fact that 

 the subjected the eggs to CO, in sea water (which retarded the pro- 

 cess about 40 min.) may account for this discrepancy of results. My 

 sections show a characteristic dumb-bell shape of the chromosomes at 

 all stages previous to the late telophase of the second mitosis. Here 

 they become stubby or globular and subsequently give rise to five or 

 six vesicles which fuse to form the female pronucleus. The latter has 

 about the size of the nucleolus of the immature oocyte, and always 

 contains a plastin nucleolus (plasmosome) which frequently becomes 

 chromatic. 



The reduced number of chromosomes is 18. This agrees with the 

 count given by Tennent and substantially also with the number re- 

 ported by Mathews (17?). The early segmentation stages yield a 

 chromosome count of approximately 36. At the time when the eighteen 

 chromosomes enter the first maturation spindle many appear double. 

 In some cases the two arms have separated so far as to give the 

 chromosomes a V-shape. From these facts it is apparent that many 

 (possibly all) of the pre- reduction chromosomes are bivalents. However, 

 since synapsis could not be observed — due to the minute size of the 

 oögonia and the complete absence of mitotic figures among them 

 — the only positive evidence on this point is wanting. Lack of evi- 

 dence as to the manner of synapsis also precludes all positive state- 

 ments regarding the reduction of the chromosomes as to whether it 

 is a true qualitative reduction or quantitative merely. If the chro- 

 mosomes fused in pairs side by side in synapsis and the products 

 condensed into bilobed bodies so that in each compound chromosome 

 one of the globes may be represented by a and the other by h then 

 the chromosomes resulting from the double longitudinal fission must 

 be represented by ab. If the condensation occurred so that both globes 

 must be represented by ah then the resulting chromosomes are two 

 of a and two of h. If synapsis occurred endwise, a double longi- 

 tudinal fission will result in similar products according as the globes 

 represent whole chromosomes or the fusion of respective halves of two 

 chromosomes. 



More detailed studies are in progress concerning the chromosome- 

 nucleolus problem in Asterias forbesii and the work will shortly be 

 extended to cover a wider range of Echinoderm forms. 



I take this occasion to acknowledge my indebtedness to Prof. E. 

 B. Wilson, who suggested the investigation; to Dr. F. B. Sumner, 



