MUSEUM OF COMPARATIVE ZOOLOGY. 305 



compact transverse layer. This persists, and as a wall of separation 

 prevents the fusion of the two new nuclei which arise near each other in 

 the handle of the dumb-bell figure. 



(4) . It is not fully, nor even principally, composed of '' Kernsub- 

 stanz," for its principal mass does not enter into the formation of the 

 new nuclei. 



Furthermore, the new nuclei do not arise by the division of a mother 

 nucleus. The substance of the striate body (spindle) does not enter 

 into the formation. of the new nuclei, but the latter are differentiated 

 only at the poles of the spindle body, as two relatively small, spherical, 

 clear, and homogeneous bodies, which, at times, are seen to be formed 

 by the confluence of smaller drops, and therefore evince their origin as 

 collections of previously distributed substance. The greater portion of 

 the spindle structure is not transferred to the new nuclei, but merges 

 with the protoplasm, and even lies in the periphery of the daughter 

 cells, where it helps to form (plants) the cellulose membrane. The 

 striate body is a structure combined out of nuclear substance and cell 

 protoplasm, which latter has made its way in from the sides. 



The evidences of the formation of the spindle within the still per- 

 sisting nuclear membrane (see especially 0. Hertwig's studies) are too 

 numerous and unequivocal to allow any doubt in the cases presented. 

 I cannot but believe, however, that the exact size of the spindle, as 

 compared with that of the nucleus, has little to do with the question of 

 a dissolution of the whole nucleus. A total disappearance of the old 

 nucleus is, of course, the cardinal point. I have never found a stage in 

 which nuclear substance was not demonstrable by staining, and agree 

 with those writers who derive the new nuclei primarily from the halves 

 of the nuclear plate, — I even doubt if the poles of the spindle (or the 

 corpuscles of the central "areas ") take any direct share in the composi- 

 tion of the new nuclei.* Of this latter point, however, I am not fully 

 persuaded. It is possible that the increase in the size of the young nu- 

 cleus may ultimately bring its periphery in contact with these corpus- 

 cles, while the latter are still intact, and that they may then directly 

 contribute to the formation of a nuclear membrane. I am, however, 

 more inclined to think the corpuscles cease to exist as discrete struc- 

 tures before any such event could transpire, and that they contribute 

 less directly, if at all, to the substance of the nuclear mass. 



Another paper by Auerbach, directed principally against views enter- 

 tained by Strasburger, is considered in another connection (p. 370). 



* See also the objections raised by Flemming (*78^, p. 415). 

 VOL. VI. — NO. 12. 20 



