STUDY OF CELL MECHANICS 515 



This egg is typical of a number of monasters examined at this 

 time. We see the two centrioles connected by a spindle with 

 new rays arising from the new centers. The chromosomes re- 

 main attached to the old fibers which are still very prominent. 

 The centrosphere elongates still more and the new centers can 

 no longer be made out. Up to this time no spiral asters are 

 found in the sectioned material. Following this stage, however, 

 we find the eggs figured in the plates, with the exception, of 

 course, of figure 9. 



The figures of the eggs themselves show further that two cen- 

 ters are involved in the spiral aster eggs. Since these stages fol- 

 low on the division of the centrioles, it can scarcely be doubted 

 that, in the former eggs, the centrioles lie at the two points of the 

 centrosphere where the areas of twisting are greatest, as in fig- 

 ures 2, 3, 4, and 5. 



We now have the facts necessary in order to understand what 

 is taking place in the eggs which show spiral asters. In the liv- 

 ing egg, owing to the happy circumstance that the pigment ring 

 was prominent, it was possible to see that a shifting of the posi- 

 tion of the asters and of the spindle was taking place. From a 

 plane which formed a considerable angle with the pigment band, 

 the spindle shifted until it lay parallel to it. In the preserved 

 material, we find every phase of this shifting; and the majority 

 of the eggs show clearly that two centers are involved, from 

 which we may safely conclude that the division of the centrioles 

 normally precedes the shifting process. There are several 

 phases of the shifting which warrant consideration. 



In the foreground stands the question, How is this shifting of 

 the spindle brought about? Do the new centers, the centro- 

 somes, move first, or, are they carried along by the protoplasm 

 lying outside of the centrosphere? The answer to this question 

 is given by a close study of the astral rays. In all of the figures 

 the point has been emphasized that, in the region of the greatest 

 bending, the rays run forward for a part of their length and 

 then backwards. It is clear from the considerations given above 

 that these areas of twisting are associated with the new division 

 centers. Now, were these centers themselves the first points to 



