212 The Ohio Naturalist. [Vol. VIII, No. 2, 



hypodermal, were of April 14 (Fig. 3). No integuments have 

 as yet appeared. From this time on development and differen- 

 tiation is rapid. Fig. 4 (April 21) shows the archesporial cell 

 divided into parietal cell and megasporocyte. The parietal cell 

 has divided into two by a vertical wall, and the incipient integu- 

 ments are now visible as may be seen in Fig. 5. This figure also 

 shows a further division of the parietal layer. By April 28 (Fig. G) 

 the megasporocyte has divided into four degaspores which arr 

 arranged in a row, the three upper of which at once begin dis- 

 solving. An extensive parietal tissue is formed by this time; 

 (Fig. 7). In Figs. 7 and 8 the outer megaspore is divided by a 

 vertical wall, while in Fig. 6, the division is horizontal. This 

 latter condition was the more commonly observed. The ar- 

 rangement of the megaspores in a more or less perfect tetrad 

 indicates a rather primitive position of the plant under considera- 

 tion. The same condition was noted by Surface* for Sanguinaria 

 canadensis and by Shreve^ for Sarracenia purpurea. 



Embryo sac. The functional megaspore divides about 

 April 28, forming the two-celled embryo sac !,Fig. 9). The non- 

 functional megaspores are gradually dissolving. This date also 

 shows some ovules having embr^^o sacs with conjugating polar 

 nuclei, the three antipodals, synergids and egg (Fig. 10). 

 The synergids are arranged in such a manner that together with 

 the egg they form a sort of tripod (See Figs. 10, 12 and 13). 

 The embryo sac elongates very greatly during the next week or 

 two (Fig. 11) and the antipodals come to lie close together in the 

 base of the sac and are still plainly visible in sections of May 19 

 (Fig. 15), though beginning to degenerate. This evanescent 

 condition of the antipodals is very different from what would be 

 expected in Ranunculacae. 



The polar nuclei occupy a characteristic position but remain 

 side by side for an unusually long time, apparently about three 

 weeks, before conjugating (Figs. 10 and 11). The various struc- 

 tures of the embr^'o sac are all clearly differentiated by staining, 

 so that interpretation is not difficult. 



Together with the long period during which the polar nuclei 

 remain in contact without fusion, should be noted the equally 

 long time that the oospore remains undivided after the beginning 

 of the endosperm formation. May 19 (Fig. 16) shows the 

 oospore still undivided, but there are already formed a dozen 

 or more endosperm cells. 



Endosperm and embryo. In sections of May 19 (Fig. 12) 

 the first endosperm wall was observed. This wall is transverse 

 and divides the sac into two equal parts. The endosperm nuclei 



1. Surface, Frank M., '05. — Contribution to the Life History of Sanguinaria cana- 

 densis. Ohio Nat. 6:1, 1905. 



2. Shrove, FoiTest, '06. — The Development and Anatomy of Sarracenia purpurea. 

 Bot. Gaz. 42 : 107, 1906. 



