1914] BURLINGAME—ARAUCARIA BRASILIENSIS 499 
In the preceding paragraph I have described the appearance of 
the nucellar tissues around the growing gametophyte. I wish now 
to describe some of the anomalous conditions found that have led 
me to suspect the validity of the current accounts of the effects of 
the gametophyte on the nucellus. In fig. 26 is shown an apparently 
enlarging hole in the nucellus surrounded by the two usual bands of 
differentiated cells. Such ovules are fairly common. In some cases 
the megaspore membrane appears to be present. One might infer 
in such cases that the hole is the work of a gametophyte that for 
some reason or other has died. In most cases the megaspore mem- 
brane cannot be demonstrated; but since it is not well developed 
in any case, this would not appear to be an insuperable difficulty. 
Fig. 27 shows a less common condition. There is no game- 
tophyte here, nor is there any place for one; yet there is a remark- 
able correspondence in nucellar structure. In the center there is 
an enlarging mass of cells whose nuclei and cytoplasm are under- 
going degeneration. The central cells are almost completely 
crushed; they have very little or no living contents. Outwardly 
the cells grade off through less and less crushed cells to a band of 
normal shape and size, but with densely staining contents, just as in 
the normal ovules. If there is no gametophyte and even no place 
for one, then the effects cannot be due to the presence of one. It 
seems that these facts admit of but one of two possible explanations: 
either the cells of the spongy tissue, which are possibly potential 
megaspore mother cells, are capable of producing the observed 
effects, or it is a quality of the nucellar cells themselves. to behave 
in this fashion, regardless of the presence of a gametophyte or its 
antecedent archesporial tissue. A noticeable peculiarity of these 
cells in all cases is the thickening of their walls accompanying the 
death and disappearance of the protoplasm. 
The development of such sterile ovules has nothing to do with 
pollination, apparently, for they occur regardless of whether the 
pollen has or has not sent tubes into the nucellus. They are rela- 
tively common and develop to advanced stages. Cones on unpol- 
linated trees on the grounds of Stanford University develop to 
nearly normal size, though the gametophytes do not. 
The megaspore membrane is usually thin and poorly organized. 
