ALIMENTARY TRACT OF THE KING SALMON 
83 
Stratum granulosum . — There is present throughout the alimentary tract of the king 
salmon a special type of cell forming a distinct layer or coat. Throughout the prelimi- 
nary studies this layer has been designated the granule cell layer, on account of the 
distinctive characteristic of the cells composing it. Oppel has described a stratum 
granulosum for the mucosa of certain animals, and while the homologies are not certain 
in the king salmon, I propose to use this designation as a permanent name for the struc- 
ture in question present in Oncorhynchus . 
The granule cell layer is a part of the mucosa. In the salmon stomach its cells lie 
between the stratum compactum and the muscularis mucosae and in the meshes of the 
stratum compactum itself. The granule cells form quite a dense and well-marked coat, 
not only in the stomach but in the intestine and caeca. However, the boundaries of this 
coat are not sharp on either the inner or outer margin. On the inner margin cells of the 
coat are found scattered among the inner meshes of the network of the stratum com- 
pactum and occasionally in the tunica propria. The outer margin or boundary of the 
zone is sharper. The inner margin of the museularis mueosae marks this boundary. 
Yet a few scattered granule cells are found in the connective tissue of the submucosa 
outside the muscularis proper. 
The arrangement of the cells about the surface of the layer, also the occasional 
displaced granule cells, suggest that these cells may be amoeboid and migratory. This 
I believe to be true to a very limited extent. But I have not yet succeeded in observ- 
ing amoeboid activity in teased living cells. In fasting salmon, however, I find extensive 
migration of the cells into the tunica propria on the inner border and into the muscular 
coats on the outer border of the stratum granulosum. This observation is especially 
striking in the degenerating cseca. The granule cells, however, are never found far 
removed from the proper location of the stratum. They have not been observed in 
the blood vessels, in the liver, or in any skeletal muscle tissue. 
The granule cells are supported by a delicate meshwork of fibrils of the histological 
type and staining properties of the stratum compactum. There is no distinct pattern 
or type of arrangement of the cells. That portion of the layer just within the muscu- 
laris is from four to eight cells thick. Many blood vessels pierce this coat on the way 
to the deeper structures. But ordinary histological methods do not reveal any definite 
vascular supply to this structure. It seems to derive its nutritive materials from the 
capillaries of the adjacent muscles and of the tunica propria. 
The granule cells are irregularly oval in outline, varying in size between 6 by 9 
and 7 by 12 in diameter. The nucleus is always eccentric in position and relatively 
small in size, 2.5 by 4 in diameter. 
The special and characteristic feature of the structure of these cells is the presence 
of the cytoplasmic granules that always crowd them, uniformly filling the cell body. 
The only differentiation noted in this regard is that those cells most free and isolated 
from the stratum proper have somewhat fewer granules. 
The granules stain deeply with a number of dyes, such as eosin, orange G, acid 
fuchsin, etc. The sharpest differentiation is obtained by staining with iron haematoxylin, 
