McAllister—Cytology and Embryology. 623 
Tlie evidences are strong that in addition to this polyembry- 
onic budding from the nucellus fertilization occurs. Pollen 
tubes are very abundant in the tubular styles and are not in¬ 
frequently found in the vicinity of the micropyle and in sev¬ 
eral cases were seen entering the micropyle. The confused col¬ 
lection of nuclei of various sizes and stages of degeneration at 
the outer end of the nucellar cavity makes it very difficult to 
identify a male nucleus. The lack of any definite easily recog¬ 
nizable egg cell has also been a great obstacle in determining 
whether any nuclear fusion occurs here. 
In order to determine whether pollination was necessary to 
the growth of embryos I castrated, in the spring of 1908, from 
75 to .1*25 flowers on each of thirteen racemes, after which each 
raceme was carefully bagged to prevent pollination from other 
plants. 
All of the flowers thus treated, probably from 1200 to 1400, 
withered and dropped off within three or four days. This 
shows conclusively that, although an embryo is probably rarely 
developed from the egg, still pollination is necessary to initiate 
the vegetative budding of the nucellar cells. 
As will be seen from figures 50, 51, 53, 54, 55, considerable 
variation exists in the form and arrangement of the cells of 
these nucellar proliferations. Some have very broad bases, 
which suggests that they may have originated from more than 
one cell. Others have but one cell in contact with the nucellar 
wall (Fig. 52). In all these embryos the irregularity in 
the arrangement of the cells is conspicuous. It is also to be 
noted that no embryos are to be found which resemble embryos 
commonly figured as developing from an egg. 
These adventitious embryos continue to grow and ultimately 
from one to three develop into mature embryos (Figs. 58 and 
59). Usually all of these embryos appear perfectly normal 
and nearly equal in size. I have been unable thus far to ger¬ 
minate the seeds and so am unable to state whether all of the 
embryos will form plants. 
I have been able, through the kindness of Prof. Osterhout 
who has furnished me with material of Smilacina sessifolia, to 
