GAGER: GLANDS IN THE EMBRYO OF ZEA 131 
The work of Brown and Morris indicates that the endosperm 
of the grass fruit is dead, but the only inference warranted by the 
papers of Green, Krabbe, Hansteen, Pfeffer, and Linz is, as Linz 
definitely states (page 312), that it is alive. 
The most obvious conclusion to be drawn from this review of 
the literature is that there is still need for further careful experi- 
mental investigation of the subject, in which every precaution 
shall be used to exclude bacterial contamination, and other sources 
of error. Such work has been done with the date seed by Pond,” 
whose experiments seem to leave little doubt that the date-endo- 
sperm, at least, is incapable of self-digestion.* 
The facts of teratology have frequently thrown light upon 
normal structure, helping to establish the homology of an organ 
whose interpretation would otherwise remain in doubt. From the 
fact that structure isan expression of function, anatomical variations 
in the direction of a structure whose role is well understood, may 
quite justifiably be taken, in connection, of course, with other facts, 
as evidence of the probable function of the part that varies. It 
was with considerable interest, therefore, in the light of our present 
knowledge of the homology and physiology of the parts of the 
fruit of the Gramineae, that the writer, in an examination of cross- 
sections of the corn grain, observed a variation in the scutellar 
epithelium, the significance of which can scarcely be questioned. 
This tissue, one cell thick, and variously called the “ absorptive 
epithelium ” and the “ glandular epithelium,” is, as is well known, 
clearly defined anatomically from the adjacent tissue on either side. 
The shape of its cells, narrow and oblong in section, their palisade 
arrangement, and the appearance of the protoplasm, granular and 
relatively dense in the resting seed, more vacuolated as germination 
begins, and with a well defined, vigorous nucleus, clearly distin- 
guish it. Normally it forms an unbroken layer over the convex 
surface of the scutellum. 
In the sections examined, this layer was found invaginated in 
places, in such a way as to form small pockets or sacs in the tissue 
of the scutellum. On one side there were two such structures, 
and on the other side one, with a slight suggestion of an unfinished 
* A conclusion contrary to that reached by the same author #! in 1904, when there 
was failure to observe certain necessary precautions of method. 
