6 U, S. BUREAU OF FISHERIES. 



from e<>:<?s which are sIoav to cleave. It may be remarked that it is 

 difficult to draw a strict line at times between the abnormal and the 



mm- 





V . ■ ■■-.. ■■•■• • '■■''-:.■'■:■■■ i^ ' . '■'•■ iW--: :\ ■ 



VN'--.-.. .s. ^^ '-^^V ■■■• ■■• Z^'/^' V'w K ^\ 



i » .'• * ■■■ ■■"■ ■ ...:■.,• 





19 



18 



' > -^A', 





A ^ 



'»•< 





20 21 



22 



Figs. IS to 22. — Sections of pike-perch eggs. IMagniflcation, approximately X 1620. 



Fk;. l8o, 1S&. ISr. — Side views of three sections of a plato showing .'^0 chromosomes, 

 the normal number. 



Fig. 19. — Si<le ^^ew of a plate in another cell in the same blastoderm as figure 11, 

 showing 1.") chi'omosouies, the haploid number. 



Fig. 20. — I'olar view of normal plate. 



Fig. 21. — Abnormal multiplication of chromosomes. 



Fig. 22.- — Abnormal elongated nucleus — abortive division. 



unsegmented types in the first few hours. At that time one or a few 

 minute excrescences are occasionally budded from the germinal disk, 



