a 


1901 | THE LIFE HISTORY OF ERYTHRONIUM 381 
and female gametes. /zg. 74 shows an embryo sac with the 
three antipodals, the dividing definitive nucleus, the oospore, 
two synergids, with an extra nucleus which may be a sperm 
cell, and a nucleus in the pollen tube, probably the tube 
nucleus. During the first few divisions of the embryo, the for- 
mation of the endosperm proceeds very rapidly (jig.’76). The 
oospore divides first by a transverse wall ( figs. 75, 76), and then 
each resulting cell divides by a vertical wall, forming a four- 
celled embryo (figs. 77, 78). These divisions are almost simul- 
taneous, although the upper one usually leads. Sometimes, 
however, the divisions are more irregular (fig. 87)» 
The young embryo lies free in the endosperm some distance 
from the upper end of the sac, and the synergids disappear very 
early. After the first few divisions of the embryo the sac 
enlarges greatly below, while the upper part remains narrow 
and may even contract (jigs. 76, 77, 79). Fig. 79 represents a 
five-celled embryo, one cell being cut from the upper tier and 
one from the lower. Fig. 80 is a six-celled embryo in which 
the two lower cells have each divided by a transverse wall, 
while fig. 8r represents a six-celled embryo with two cells of the 
lowest tier cut away. fig. 82 is a twelve- to fifteen-celled 
embryo, and Jig. 83 about a twelve-celled one. These examples 
will show how very irregularly the development proceeds. Up 
to this time and later there is usually a distinct difference 
between the cells which came from the upper and lower cells of 
the first division. This difference is shown not only by a differ- 
ence in the contents of the cell, but especially by the staining 
reaction. Thus in jig. 83 the upper or suspensor cells have a 
bright yellowish cytoplasm, while the embryo cells are very 
granular and deep red. This is also present in Lelium Philadel- 
phicum, sometimes being very prominent in the older embryos 
(9). Figs. 84 and 85 represent later stages of the embryo of 
E. albidum. The suspensor region is rather large and often 
very irregular in shape and much lobed, but the whole structure 
ives rise to only a single embryo, as was verified by numerous 
examples. The embryos were very badly shrunken, however, 
