ORIGIN OF CELLS. 17 



collects around them so as to form projections from the periphery of the 

 cell, which become more and more constricted until they finally separate 

 (Fig. 5). During division the nuclear membrane disappears, and the 

 nucleus usually divides before the cell-protoplasm, but not directly, by 

 simple cleavage, as was formerly supposed, but indirectly, by karyo- 

 kitiesis (from movement of nuclear fibrils). This is an exceedingly com- 

 plicated process. Its different stages may be divided about as follows : 



1. The nuclear fibrils become very distinct, while the nuclear mem- 

 brane disappears and the fibrils of the nuclear net-work become twisted 

 and bent into a more or less dense convolution, while the entire nucleus 

 enlarges. 



2. The fibrils unravel into loops arranged around the centre as a 

 wreath or rosette. 



3. The peripheral points of the loops become broken and a star- 

 shaped figure of single loops is obtained. This is termed the aster, or 

 star. 



4. The loops separate into two groups or new centres. This is the 

 diaster, or double star. 



5. The two groups of threads become farther apart, as if attracted 

 by opposite poles, but still remain connected by fine pole-threads, which 

 represent the interstitial nuclear substance. In this stage the figure 

 resembles a spindle. 



6. The connection between the two sets of threads is broken. 



7. The threads of each set become convoluted. 



8. A membrane forms for each set, and thus new daughter nuclei 

 result. (See Fig. 6.) 



The cell-protoplasm may commence to divide at any stage between 

 the one when the threads aggregate around the two centres and the one 

 when two distinct nuclei are present ; or the division of the nucleus may 

 be followed by division of the cell, so giving a cell with two nuclei. It 

 is not proved that this is the universal mode of divisions of nuclei, 

 though it has been observed in all kinds of cells in the embryo, and to a 

 limited degree in the adult. On the contrary, it is probable that amoeboid 

 cells divide by the direct method and that other nuclei may also undergo 

 direct division, or Remak's division, by simple cleavage, though all the 

 cases in which constriction of nuclei is observed need not be cases of 

 commencing division, as the change in- shape may be due to pressure of 

 cell-protoplasm, or to nuclear contractions. (Klein). 



The second form of cell-formation is termed by Kolliker " endoge- 

 nous cell-formation," and consists in the formation of cells within the 

 membrane of the parent cell, which ultimately bursts and discharges its 

 progeny of j^oung cells. The division of the mammalian ovum, the 

 growth of cartilage, and many pathological cell-formations are types of 



2 



