100 CYTOKINESIS. 



Rhumbler attributes the phenomena of cytodieresis to at least five factors : — 

 (1) The pull of the astral rays, (2) the pull of the central spindle, (3) the rounding 

 of the cell, (4) the growth of the membrane, (5) the decrease of the nuclear lumen. 

 According to my view the most important factors of all are omitted from this cate- 

 gory, viz. : (1) the decrease of surface tension at the poles and the consequent 

 elongation of the cell in the spindle axis, and (2) the vortical flow of cell substance. 



2. Movements during Telokmesis. — The cell movements during telokinesis 

 are of a rotary character, the spindle axis and cell contents in each daughter cell 

 moving through an angle var3dng from 30° to 180°. In my former paper ('99) on 

 these movements I did not sufficiently distinguish between the vortical movements 

 of karyokinesis and the rotary ones of telokinesis. There is abundant evidence, 

 however, that the movements of telokinesis are in the main of a rotarv and not of 

 a vortical character. The halves of the spindle shift their positions so that they 

 come to lie close to each other on opposite sides of the new cell wall ; in general, 

 there is no flow of one portion of the cell contents through another, but all parts 

 rotate in a given plane around some point which serves as a center. In the first 

 and second cleavages the center of rotation is approximately the center. of each 

 daughter cell. In the later cleavages the center of rotation varies in different cell 

 generations, but is usually above the center of the cell, t. e., nearer the animal pole, 

 in the case of the micromeres and below the center or nearer the vegetal pole in the 

 case of the macromeres. 



(1.) In this rotation the entire cell contents take part; there is not merely a 

 bending of the spindle axis but also a movement of the cytoplasm and yolk. During 

 this rotation the half of the spindle axis in each daughter cell is preserved as a 

 straight line, the bend in this axis occurring only at the mid-body. Throughout 

 the telokinesis the spindle fibres may be recognized connecting the mid-body and 

 nucleus and in some cases passing around the nucleus to the centrosome and 

 sphere, figs. 61, 73. The bending of the spindle axis on itself in the two daughter 

 cells, rather than its rotation with the cytoplasm and yolk, is thus explained by 

 the persistence of the spindle fibres, which attach the structures of the spindle 

 axis to the mid-body. Throughout this rotation the nucleus preserves its polarity, 

 its grooved side (central pole) being turned toward the centrosome, though this 

 general rule maj- be departed from to a limited extent in cases where the move- 

 ments of the nucleus or centrosome are interfered with. In the cytojilasni, radia- 

 ting rows of alveoles are present during the whole of this rotation ; in the first and 

 second cleavages they become curved, as shown in figs. 61 and 69. They are 

 entirely lacking on the side of the spindle axis next the new cell wall, where the 

 cytoplasm is clear, non-stainable and shoAvs no traces of alveolar structure. 



(2.) The movements in the two daughter cells are always in opposite dii'cc- 

 tions and are always toward the animal pole ; consequently, if the rotation is dexio- 

 tropic in one cell it is la^otropic in the other. In all spiral cleavages the move- 

 ments in the upper cell are in the direction of the cleavage by which that cell was 

 formed ; thus the first quartette is formed by a dexiotropic cleavage, and the rota- 



