The maturation and fertilization of the egg of Cerebratulus. 443 



fore, no more than one six- thousandth as much substance as is found 

 in the egg. This is in marked distinction from the polar bodies of 

 the eggs of the mouse, where Sobotta (25) finds them "grösser als 

 Oewebszellen (sogar mehrfach so gross, als die meisten), nicht so er- 

 heblich viel kleiner, als das Ei selbst". He even considers it "durch- 

 aus nicht für ausgeschlossen, dass auch mal ein Richtungskörperchen 

 befruchtet wird". Garnault (6) likewise finds that the fii^t polar 

 body in Mollusks "en général de petite taille peut atteindre quelquefois 

 le l environ du volume de l'oeuf". 



The substance of the polar bodies is purely protoplasmic without 

 trace of yolk-globules. 



These little cells are very persistent in Cerebratulus margi- 

 natus and may often be seen attached to some portion of the body 

 of the embryo even after it has been swimming for a considerable time. 



The Second Polar Body. 



As has been seen, the centrosomes at each end of the first polar 

 spindle divide at a very early period and often even before the spindle 

 itself is fully formed. After the separation of the first polar body 

 the centrosphere remaining in the egg lies comparatively near the 

 surface. It now elangates in a direction perpendicular to a radius of 

 the egg and sinks deeper into its substance. The two centrosomes 

 separate more widely, and the centrosphere elongates correspondingly, 

 at the same time becoming less and less distinct. The aster-fibres 

 lose their symmetrical arrangement to some extent but still remain 

 grouped about the elongated centrosphere in most cases, rather than 

 about the two centrosomes. Soon, however, and sometimes even before 

 the first polar body is entirely separated (Fig. 20), the fibres at the 

 ends of the centrosphere increase in size, while those about its middle 

 degenerate. Thus the fibres become grouped about the two centro- 

 somes which lie in the ends of the elongated centrosphere. This 

 centrosphere fades away into the surrounding cytoplasm , and new 

 radiations appear about the two centrosomes as they separate more 

 widely. They appear partially within the boundaries of the old centro- 

 sphere. These new radiations are formed to some extent by the direct 

 transferrence of a portion of the rays of the old aster, but largely 

 by the growth of entirely new fibres. It was impossible to determine 

 whether the old rays joined directly on to the ends of newly formed 

 fibres, or whether they were transferred bodily to the new system. 



For an explanation of the manner in which an aster-fibre is 



