MUSEUM OF COMPARATIVE ZOOLOGY. 551 



mates the outer wall of its cell. This is a peculiarity in the formation 

 of polar cells which deserves more attention than it has received. It 

 may perhaps be urged as an indication, that the production of polar 

 globules is not accomplished by cell division ; but it appears to me 

 that it does not present any fundamental obstacle to that conception. 



In an examination into the nature of the forces which result either 

 in the production of the polar globule or in cell division, this peculi- 

 arity may furnish some means of extending or correcting conclusions to 

 be drawn from less modified forms of division. 



No theory of a mutual repulsion between the stars of the amphiaster 

 is able to explain either the "orientation" of the spindle, or its migra- 

 tion to the surface, nor is the existence of such a repulsion at this stage 

 certain, since the length of the spindle (with a single exception, no- 

 ticed later) remains during this period practically unchanged.* Neither 

 does it seem possible to explain the migration as due to the attrac- 

 tion which the centres are supposed to exercise on the vitelline proto- 

 plasm, even if such an attraction were capable of putting the spindle 

 in a central position in reference to the active constituents of the vi- 

 tellus, or of causing it to occupy the primitive axis of the yolk. No 

 such attraction could urge both asters into such close approximation 

 to the animal pole. There may be complicated chemical and physical 

 processes underlying all the movements connected with the formation 

 of polar globules which are at present as unintelligible as are all other 

 spontaneous movements of protoplasm. Still it is not useless to inquire 

 whether there is any possible explanation of the movements of the 

 archiamphiaster which, without dealing with the nature of protoplasmic 

 motion in general, is capable of rendering these changes less obscure. 

 An assumed a,ttraction, exerted by the protaplasm upon certain con- 

 stituents of the spindle (the inner half), and a repulsion of other con- 

 stituents (external half), would be sufficient to cause the " orientation " 

 of the spindle in the primitive axis ; but a migration which carries its 

 internal end beyond the centre of the active protoplasm toward the 

 primary pole could only be accomplished by the repulsion preponderat- 

 ing over the attraction, and even with that assumption a lengthening 

 of the spindle should result from the repulsion of one of its ends and 

 the attraction of the other. The final separation of the repelled portion, 

 accomplished by the formation of polar globules, would then leave the 



* In the early stages of its formation a mutual recession of the asters has been 

 observed, and may perhaps be attributed to the development of mutually repulsive 

 properties, but after the formation of the spindle there is no important separation. 



I 



