444 L. V. HEILBRUNN 



thought that gelation was due to an abstraction of water and 

 solation to an addition of water. In an earUer paper some e\'i- 

 dence for the first view was presented. If solation were due to 

 an addition of water, it would naturally be expected that the 

 breakdown of the nucleus during mitosis would be the primary 

 cause of solation. This is, however, not the case in Cumingia. 

 In the curve shown in figure 1, after the second polar body has 

 formed the viscosity remains high for the formation of the cleav- 

 age spindle. The viscosity then drops, but this drop occurs 

 not after the breakdown of the pronuclei, but before the pro- 

 nuclei have shown any visible indication of breakdown. Thus 

 the return of the protoplasm to a more fluid state is apparently 

 not due to a passage of water from nucleus to cytoplasm. As 

 far as can be judged from the published figures of Morgan ('10), 

 Jordan ('10), the time of liquefaction is approximately simul- 

 taneous with the appearance of the fully formed cleavage spindle. 

 The above authors show the spindle appearing at a time when 

 the pronuclei have grown to a large size. It is at this time 

 that the drop in viscosity occurs. Further evidence that the 

 liquefaction of the protoplasm is not due to a breakdown of the 

 nucleus is indicated in Cumingia by the fact that in this egg the 

 protoplasm becomes more fluid immediately after the second mat- 

 uration spindle is formed. In this case no nuclear breakdown 

 occurs. 



It appears that whenever a spindle forms there is a return to 

 a fluid condition. It is as though the spindle were coagulated 

 out of the cytoplasm. Thus the curve of viscosity at the time 

 of spindle formation is like the viscosity curve of a coagulating 

 albumen solution. As coagulation occurs in such a solution 

 the viscosity suddenly increases greatly, after coagulation has 

 been completed there is a sharp drop in viscosity. The first 

 part of the Cumingia viscosity curve resembles the curve of a 

 coagulating albumen solution given in Dunstan and Thole's 

 'Viscosity of hquids' (fig. 9). Of course, this resemblance may 

 be superficial. At present, however, the idea that the spindle 

 coagulates out probably offers the best interpretation of the 

 viscosity changes which precede and follow spindle formation. 



