BOT.— Vol. I.] CAMPBELL— NAIAS AND ZANNICHELLLA. 21 



noticeable difference in the size of the polar nuclei, such as 

 Schaffner (1. c, p. 126) mentions for Alisma. 



In the upper part of the embryo-sac the changes are more 

 marked. Two of the nuclei enlarge but little and show but 

 little chromatin and a small nucleolus. The third one, how- 

 ever, enlarges a good deal, and has a large nucleolus and much 

 more chromatin than the others. The smaller nuclei are 

 those of the synergid^, the other the nucleus of the ^%^, 

 which extends below the two synergidge (figs. 50-52). All 

 these cells are bounded by a clearly defined protoplasmic 

 membrane, and the protoplasm within shows a somewhat 

 vacuolate structure, especially in the ^^g. Here, too, the 

 demonstration of the centrosphere was not satisfactory, 

 but it may have been on account of defective methods. 



The cells covering the apex of the embryo-sac divide but 

 little, but are not encroached upon by its growth; while the 

 lateral cells of the nucellus are much compressed and the 

 inner layers completely destroyed, so that at the time of fer- 

 tilization there is but a single complete layer of cells bound- 

 ing the sides of the embryo-sac (fig. 57). 



In the course of these investigations a number of striking 

 departures from the normal type of embryo-sac were ob- 

 served. In the one shown in fig. 54 there was apparently 

 no absorption of the cells above- the embryo-sac, and in the 

 embryo-sac itself the usual course of development had not 

 gone on in a normal way. There were three nuclei, one 

 somewhat larger than the others, at the apex of the sac, and 

 it looked as if there might be an imperfect septation by very 

 thin walls; but the latter may have been simply sections of 

 the walls of the nucellar cells surrounding the embryo-sac. 

 The ^gg apparatus was not differentiated, but the polar 

 nuclei were formed as usual. The antipodal cells were, 

 however, six instead of three (fig. 55), presumably due to 

 an extra division in each, subsequent to the formation of the 

 polar nuclei. Another striking instance is shown in fig. 56. 

 Here, apparently, fertilization had not been effected and the 

 embryo-sac had become filled with parenchymatous tissue, 

 composed of elongated, somewhat thick-walled cells. As 



