﻿BOTANICAL GAZETTE 



disintegrating masses of tissue represent the remains of only 2 mega- 

 spores. Even though it may be true that the number of cells in 

 the axial row varies in closely related species of the same genus, this 

 does not seem to be the case for the 3 species of Trillium under dis- 

 cussion. The close resemblance of the series of figures given for 

 each of the 3 species and the fact that Chamberlain does not 

 figure a 4-celled axial row make it highly probable that T. recur- 

 vation, like T. grandijlorum and T. cernuum, has an axial row of 2 

 binucleate cells, and that the first 2 cells of its embryo sac represent 

 2 megaspores. During a summer session which I spent at the 

 University of Chicago in 191 2, Dr. Chamberlain very kindly 

 examined my slides covering these stages and agreed with this 

 interpretation. 



After a period of growth of the young embryo sac, the 2 nuclei 

 divide rapidly to form a 4-nucleate, and then an 8-nucleate, embryo 

 sac (figs. 17-19). These phenomena and those attending the 

 maturation of the embryo sac agree so closely with the account 

 already given by Ernst for T. grandijlorum as to make further 

 comment on them unnecessary. 



From the foregoing account it may be seen that T. grandijlorum 

 and T. cernuum, the 2 species to be used in the work of hybridiz- 

 ing, agree in all important respects regarding the origin and the 

 development of the embryo sac. The chief points of difference in 

 the two accounts are as follows: (1) one row of primary parietal 

 tissue is formed in the ovules of T. cernuum, while in T. grandiflorum 

 primary parietal tissue is entirely wanting; (2) in T. cernuum the 

 2 nuclei resulting from the heterotypic division do not form resting 



Wellesley, Mass. 



EXPLANATION OF PLATE XXVII 



An Abbe camera lucida was used for all of the drawings. The figures are 

 oriented with the micropylar end toward the upper edge of the page. The 

 abbreviations used in labeling the drawings are as follows: m.m.c, megaspore 

 mother cell; p.p.c, primary parietal cell. 



Fig. i . — Longitudinal section of a young ovule; X 73- 



Fig. 2. — Same as fig. 1 ; X43°- 



Fig. 3.— Megaspore mother cell in synapsis; X43°- 



