96 CYTOKINESIS. 



poles. Rhumbler ('97, '99) has emphasized the important fact that in cytodieresis 

 the total surface layer is increased, and that there must be a corresponding growth 

 of the cell wall; he considers that this growth occurs jarincipally in the equatorial 

 plane. In egg cells, I am convinced that the growth of cell membrane takes place 

 principally at the poles, and that the equatorial constriction is due, as Biitschli holds, 

 to relatively greater tension at the equator, this being due to the growth of the 

 membrane at the poles and the consequent decrease of tension at these points. In 

 favor of this view are not only the observations which I have made as to the growth 

 of membrane at the poles, but also the fact that in cells with pigmented surface 

 layer the pigment moves away from the poles of the spindle, while the pigmented 

 cell wall is carried down into the equatorial constriction and ultimately almost all 

 the way through the division plane (see observations of Nusbaum, Van Bambeke 

 and Rhumbler, mentioned later in section on movements in telokinesis). In such 

 cases, therefore, the so-called "new cell wall" is largely the old cell wall, while 

 the new wall is formed chiefly at the poles. However, the central portion of the 

 division wall, i. e., the portion lying near the mid-body, is a new formation. 



While the factors just described explain the equatorial constriction, the com- 

 plete separation of daughter cells and the constriction of the connective fibres of the 

 spindle are probably due to other additional factors. I believe that the principal 

 additional factor in this constriction may be found in the flow of substance from the 

 poles to the equator, near the surface of the cell, and thence into the spindle axis. 

 As evidence of such a flow, I adduce the movements of the yolk spherules in all the 

 divisions of the macromeres, and particularly in the first and second cleavages (see 

 pp. 79 and 81). In all these cases the poles of the spindle, surrounded by cyto- 

 plasm, move awa}' from the equatorial plane, and at the same time yolk sjjherules 

 at the periphery of the cell move toward the equator, and thence in toward the 

 middle of the cell in the plane of the future cell wall. Such a movement is a true 

 vortex, and it might be expected that the j'olk spherules which are carried in along 

 the plane of the approaching cleavage would then be carried out through the spindle 

 axis to points on the surface o^jposite the poles of the spindle. Such, however, is 

 not the case ; the yolk spherules lie in the plane of the future cell wall, but never 

 move out through the spindle axis. This cannot be held to demonstrate that 

 there is no movement through the spindle axis toward the jJoles since the same 

 forces which crowd the 3"olk spherules and large alveoles out from the centrosomes 

 (see p. 91) would operate to prevent the flow of these large structures through the 

 spindle axis toward the poles. 



In all cells, whether they possess yolk or not, it is readily seen that during and 

 after the metakinesis the cytoplasm in the equatorial region stains more faintlj', and 

 is composed of larger alveoles as division advances. In fact, the earliest indication 

 iu the cytoplasm of the plane where the equatorial constriction Avill occur is the 

 clear, non-staining zone which runs through the cell in the plane of the future cell 

 wall ; this zone is composed of large alveoles with a I'elatively large amount of non- 

 staining enchylemma and a small quantity of the stainable hyaloj)lasm. Soon after 



