7905] SHATTUCK—ULMUS AMERICANA 211 
as tapetum contained two or three nuclei and abundance of food 
material which stains deeply. 
_ The tapetum consists generally of a single layer; and is derived 
from the original sporogenous mass. This is clearly shown by the fact 
" that the two layers of cells within the endothecium never contribute 
to the formation of tapetum, but break down early while the endo- 
thecium itself enlarges (fig. 5). That the tapetum is derived from 
the original sporogenous mass is further shown by its extension 
inward, sometimes to the depth of several layers (fig. 4), more or less 
intermingled with the cells which are functioning as spore-forming 
tissue. Bowrr (2) has shown that in Equisetum from one-fourth 
to one-third of the sporogenous cells disorganize and do not form 
spores. He says “their function is that of a diffused tapetum and 
there can be no doubt that their substance contributes to the nutri- 
tion of the survivors.”? This contribution seems to be very evident 
in Ulmus. 
Wy1te (21) has shown in Elodea that there is a regular contribu- 
tion to the tapetum from the sporogenous mass; WEBB (20) shows 
in Astilbe that the tapetum has the same origin as the spore mother 
cells; LAND (14) notes that in Ephedra it is often impossible to dis- 
tinguish the tapetal cells from adjacent mother cells, mentioning that 
the tapetum seems to be potentially sporogenous tissue which has 
become sterile by virtue of its position; and CouLTER (6) states that 
A Ranunculus it seemed as if the whole tapetum were cut off from the 
Periphery of the sporogenous mass. The tapetum in Ulmus is surely 
composed of sterilized sporogenous cells. 
éy 15 the pollen mother cells are just beginning to pass 
in the = winter stage, for on the same slide mother cells were found 
Testing stage (jig. 3), in synapsis (fig. 4), in the first mitosis 
— ih in the second mitosis (fig. 7). Ten days later the tetrads 
: ormed (fig. 5); also there were many cells containing four 
i evidently just preparing to form microspores (fig. 8). 
2 a begins to break down about March 1, and by March 
ea). — absorbed. At this time the tetrads are uninucleate 
ek tows € two inner layers of the sporangium wall also rapidly 
hie « and disappear. At the same time the endothecium 
Its cells take on a cork-like appearance and do not stain 
