CYTOKINESIS. 113 



mitotic figure, which precedes unequal cleavage, is itself a result, rather than a cause, 

 of cytoplasmic structure and activity. This is shown especially well in the forma- 

 tion of the polar bodies in Crepidula. The two centrosomes and asters are here 

 absolutely equal until one pole of the spindle comes into contact with the egg mem- 

 brane. Then a lobe of cytoplasm is formed over this pole and the peripheral move- 

 ment of the spindle continues until the centrosome and chromosomes at the peri- 

 pheral pole are thrust clear through this lobe into contact with its distal cell wall. 

 Then the whole spindle becomes shorter and stouter, the connecting fibres being 

 curved and bent, showing that the shortening of the spindle is due to some force 

 entirely outside of the spindle which is propelling it against the cell wall. Such a 

 case shows in the clearest possible manner that the eccentric position of the spindle 

 is the result of cytoplasmic activities. 



In the cleavage the same fact is apparent in every unequal division. Such 

 divisions are preceded by an eccentricity of the mitotic figure, but this in turn is 

 caused by active movements of the cytoplasm. In fact one may frequently be able 

 to determine that a given cleavage will be unequal long before the spindle is formed, 

 by the position of the nucleus during the rest. In the first quartette cells of Crepi- 

 dula the nuclei lie in the inner angles of the cells during the rest, figs. 91, 93, and 

 when the next spindle is formed and the nuclear membrane- dissolved the entire 

 mitotic figure moves away from the animal pole until the peripheral pole of the 

 spindle come into contact with the cell membrane at the outer side of the cell and 

 here the small peripheral trochoblast is separated from the large apical cephaloblast, 

 figs. 93-96. In short the nucleus in the resting period preceding division lay 

 entirely within the area of the future larger cell. The same phenomenon is shown 

 in the division of the cephaloblasts [cf. figs. 96, 97). 



The conclusion that the eccentricity of the spindle is caused by the activity of 

 the cytoplasm is supported by the observations of other authors, particularly by 

 Lillie's ('99, '01) work on Unto. The first cleavage in this animal is quite unequal 

 yet "the spindle forms in the center of the egg in the plane already indicated by the 

 elongation of the sphere substance. . . The entire spindle then moves directly along 

 the prolongation of its axis, and thus parallel to the direction of elongation of the 

 sphere substance, to one side of the egg, until the centrosome of one end comes 

 almost into contact with the peripheral layer of protoplasm." Then the spindle 

 again moves back toward the center of the egg, and then again toward the periphery 

 until it finally comes to rest with its equator in the plane of the future cleavage. 

 These movements of the spindle Lillie attributed to the orientation of the cytoplasm. 

 Somewhat similar oscillatory movements of the spindle have been observed by 

 Ziegler ('95) in the living eggs of nematodes, though he did not connect them directly 

 with equality or inequality of division. In my first paper on this subject ('94) I 

 also called attention to the oscillatory movements of nuclei and cytoplasm during 

 cleavage and pointed out the relation of such movements to the direction and 

 equality of division. 



In conclusion, it is obvious that in Crepidula and Unio the place of cell divi- 



15 JOUEN. A. N. S. PHILA., VOL. XII. 



