﻿iqi6] HEATLEY— TRILLIUM CERNUUM 427 



period, the cell plate gradually becomes heavier and extends across 

 the entire cell as a dividing wall (fig. 7) . Every step of this division 

 clearly indicates that it is the heterotypic phase of meiosis. 



Although the daughter cells are at first apparently similar in 

 all respects, the chalazal one gradually shows signs of greater 

 vitality, and by more rapid growth encroaches upon the micropylar 

 cell (figs. 8-13). In no case have I found the outer cell giving 

 evidence of being the more vigorous. It may disintegrate imme- 

 diately (fig. 8); it may pass through one or more phases of the 

 homotypic division (figs. 9-1 1); or it may even complete meiosis 

 before disorganizing (figs. 12-14). A study of many ovules shows 

 that no cell wall is ever laid down, and that even though the spores 

 may be formed they never become functional, but disintegrate early 

 (fig- 15). 



At the completion of the homotypic division in the chalazal cell, 

 no cell wall is formed, and the resulting 2 nuclei undergo a period 

 of rest lasting 2 or 3 days. The cell increases rapidly in size and 

 becomes very vacuolate (fig. 16). The formation of a large, central 

 vacuole soon forces the 2 nuclei to opposite ends of the cell. It is 

 very evident that this structure is a young embryo sac, and thus 

 the female gametophyte is derived from 2 megaspores. Ernst 

 states that 2 megaspores enter into the formation of the embryo 

 sac of T. grandiflorum. Coulter and Chamberlain 4 describe 

 an axial row of 4 megaspores for Trillium and report that in T. 

 recurvation the embryo sac is derived from the chalazal megaspore. 

 A study of Chamberlain's figures for T. recuroatum reveals striking 

 resemblances to those for T. grandiflorum and T. ccrnuiim. As in 

 the last two named species, there is an axial row of 2 binucleate 

 cells which have resulted from the " second division of the nucleus 

 of the mother cell." No sketch of the 4-celled axial row mentioned 

 in his text is given. The next stage illustrated is that of the young 

 2-celled embryo sac capped at the micropylar end by a dense, con- 

 tracted mass, "the remains of the other 3 megaspores." Fig. 15 

 of this paper shows a very similar embryo sac, but an examina- 

 tion of the preceding stages illustrated makes it clear that the 



