MUSEUM OF COMPARATIVE ZOOLOGY. 223 



be seen to be considerably flattened parallel with the plane passing 

 through the animal pole and the two spots. The latter appear in this 

 position transversely oval. 



A little later, the furrow will be seen to have extended around on the 

 sides of the yolk as a shallow depression, reaching something more than 

 half-way toward the vegetative pole. The oval spots have meantime 

 been'increasing in extent, and their axes have now (Fig. 32) become more 

 nearly parallel. In some cases one may see a faintly but unmistakably 

 radiate arrangement of the vitelline substance around these oval spots 

 as centres (Fig. 32). More frequently the yolk is too opaque for that, or 

 even for the detection of the oval spots. Soon after the appearance of 

 the depression at the animal pole, — within four or five minutes at the 

 usual temperature of a summer day, — the furrow has run quite around 

 the yolk, and now appears at the vegetative pole as a very broad, shallow 

 depression (Fig. 35). This annular constriction now deepens on all sides, 

 but most rapidly from that of the animal pole. Thus the axes of the 

 two clear spots become first parallel, and then somewhat divergent toward 

 the animal pole. But when the furrow of the animal half has advanced 

 to near the middle of the yolk, it has become narrowed, by the approach 

 of the opposing faces of the incipient spheres, almost to a fissure (Fig. 62*), 

 whereas the depression from the opposite side is now a broad groove, so 

 that the axes of the two clear spots have by this means become again 

 convergent toward the animal pole. By the further deepening of the 

 constriction on all sides, there are formed two equal, symmetrical, ovoid 

 bodies, which are connected by only a slender thread of protoplasm, 

 situated much nearer the vegetative than the animal surface (Fig. 62*). 

 The long axes of the new spheres are at this moment (before complete 

 separation) directed convergingly toward a point in the plane of division 

 which lies in the prolongation of the animal radius. At this time, too, 

 the blunter end of the new spheroids is the one corresponding to the 

 vegetative half of the unsegmented yolk. Seen along the animal radius, 

 they appear elongated in a direction perpendicular to the line joining 

 their centres, and the surfaces which face are less convex than those 

 which look outward. A plane perpendicular to that of cleavage, and 

 coinciding with the animal radius, would divide each into symmetri- 

 cal halves. There is no other plane which could accomplish a like 

 result. 



At length the slender filament becomes more attenuated, and finally 

 parts. The cleavage is accomplished; but each of the spheroids still 

 continues to undergo further changes of form, and promptly assumes 



