MUSEUM OF COMPARATIVE ZOOLOGY. 411 



I should not have occasion to call attention here to a paper by 

 Semper ('75, pp. 4 - 13), in which he discusses the origin and nature of 

 the so-called Testazellen of ascidians, were it not for his attempt to 

 identify these structures with the polar globules of other animals, — the 

 less occasion, as his observations relate principally to artificial produc- 

 tions which do not contain nuclei (although the same is true in his 

 opinion for the normally produced " Testazellen ") and are only entitled 

 to the name " Testatropfen.'' The reasons given to establish the iden- 

 tity alluded to have proved to be unfortunate, as Whitman ('78") has 

 already pointed out. Nor can Semper's claim be maintained, that the 

 polar globules " first make their appearance at the moment of segmenta- 

 tion," * or that they " are enuclear " ; nor is there any evidence to prove 

 that they are capable of moving themselves in an amoeboid fashion 

 around the embryo. Even though a change in the contour of these 

 globules has often been observed, and the existence of rigid processes 

 which resemble pseudopodia has been demonstrated by Flemming, it is 

 far from proving the polar globules endowed with the power of mak- 

 ing excursions on their own account about the embryo. As Kupffer 

 ('70, pp. 123, 124, and '72, p. 366) has shown, the "Testazellen" appear 

 while the germinative vesicle is still intact, and this is not objected to 

 by Semper, who admits that they arise, not from the nucleus, but from 

 the yolk. The genetic connection of the polar globules with the germi- 

 native vesicle therefore forbids the comparison which he has instituted. 

 But when Semper, partially recognizing the possibilities of such a ge- 

 netic relation, in a foot-note substitutes for an " Uebereinstimmung in 

 fast jeder Beziehung " a j^ki/siological comparison between these two 

 sorts of bodies, he is no longer defending the view already promulgated, 

 — a view, it is to be observed, which he endeavored to establish with 

 morphological arguments, — but is really supporting new ideas. The 

 physiological role which Semper discovers in these bodies is " the de- 

 tachment of a hitherto integral component of the egg-cell, in some 

 manner a defecation of the same, an elimination of substance apparently 

 i useless for the approaching processes." 



In a preliminary paper on early stages in the development of the 

 rabbit, Ed. van Beneden ('75, pp. 690 - 693, 695 - 700) signalizes the 

 existence of two or three small round bodies (pseicdo-nucleoles) and a 



* P. S, — The recent statement by Brooks ('80, p. 79, PI, I. Figs. 3, 4) that 

 the polar globules in Physa arise dicring segmentation will also prohablj'' be found to be 

 I inaccurate, for it is not consistent with what is known of the nature of these globules 

 I in all other investigated animals. 



