MUSEUM OF COMPARATIVE ZOOLOGY. 281 



ance, which increases in size, and about which a stellate circle is formed 

 in the yolk. The knobs continue to separate, while the connecting por- 

 tion becomes reduced in thickness to a mere thread. This thread of 

 connection remains some time, but finally during segmentation its halves 

 are slowly contracted toward the nuclei to form, close to them, a new 

 knoblike swelling. The radiate figures in the yolk now become less 

 distinct ; the outlines of the nuclei more definite. During the whole 

 process of division the contour is confused, and, in addition to the radial 

 figures of the yolk at the poles of the lemon-shaped nucleus, Biitschli 

 seemed to see raylike processes stretch out from the nucleus into the 

 substance of the yolk, which served to strengthen his conviction that the 

 nucleus possesses at times a considerable degree of mobility. Whether, 

 however, this phenomenon has anything to do with the radial arrange- 

 ment of the yolk granules, he did not venture to decide. 



The failure to see arched rays joining the two centres of attraction 

 may readily be understood when it is remembered that his studies were 

 made exclusively on living eggs. 



The subsequent segmentations presented essentially the same phe- 

 nomena. There is no such thing in his opinion as a disappearance of 

 the nucleus, although before the division it becomes quite indistinct, a 

 fact which he is inclined to connect with its mobility. 



The conclusion seems to me unavoidable that the knoblike swellings at 

 the poles of the lemon-shaped nucleus correspond to the centres of the stel- 

 late figures which Fol saw, and are by Biitschli connected too intimately 

 (as parts of the lemon-shaped figure) with the nucleus. The knoblike 

 swellings which are formed at the close of the segmentation out of the 

 contracting thread are really the new nuclei, for which the centres of 

 the stellate figures were mistaken. The radiate structure was also ob- 

 served (p. 35) by Biitschli in the formation of the sperm cells. 



In the study of a much less favorable object, Anodonta, Flemming ('74, 

 pp. 286 - 290, Taf XVI. Figs. 22 - 29) was also fortunate enough to see 

 some of the phases already noticed by Fol and Biitschli. He found that 

 many of the segmentation spheres under gentle pressure presented near 

 their centres one or two clear spots without granules, and stretching out 

 from these toward the periphery in an almost strictly radial direction 

 rays of clear protoplasm (kornerloser Substanz), so that the granules 

 which lay between these rays were likewise arranged in diverging lines. 

 Subsequent to this condition followed a stage in which two nuclei were 

 found in the undivided cleavage spheres. In no case were radial struc- 

 ture and nucleus found to be present at the same time. 



