Gland-Cells of Internal Secretion in Spinal Cord of Skates. 11 
tioned. However, a few points of interest about some of these forms 
will be taken up. 
In all cases the cells are present in the caudal portion of the spinal 
cord, and they are even more restricted in anterior-posterior distribu- 
tion than in the case of the skate. For example, in the dog-fish, Mus- 
telus, they occur only from the level of the second dorsal fin (the caudal- 
most one) to the tip of the tail. In Torpedo they are present only in 
the region of the two little tail-fins. In the sting-ray and goose-fish the 
anterior limit was not determined exactly, but in the sting-ray the cells 
were found in the whip-like prolongation of the tail, and in the goose- 
fish they are about 2 inches from the caudal end of the spinal cord. 
In the dog-fish, shark, torpedo, and sting-ray the cells are very much 
the same in general appearance as in the skate, but are much smaller, 
only about one-half as great in diameter as that of the skate in the case 
of the dog-fish. Indeed, the cells of the dog-fish almost exactly resem- 
ble cells of half-grown skates. One of the largest of these dog-fish cells 
is shown in figure 18. In the sting-ray the cells are smaller than in the 
dog-fish. The nuclei are somewhat different. They are not so highly 
lobular, but are concentrated into a more spherical mass. Several 
distinct nuclear membranes may often be made out, as if the nucleus 
had been formed from a union of several nuclei which had not quite 
completely fused. No cell vacuolation or granular material has been 
observed in either the dog-fish, torpedo, shark, or sting-ray. 
The cells of the goose-fish, however, are the most interesting for this 
study. They differ greatly from all of the others. A photograph of one 
rather highly magnified is shown in figure 4. It resembles a nerve-cell 
in many respects. The nucleus is lobular and amceboid in shape, but 
much less so than in the case of the other fishes examined. Further- 
more, it usually contains one or more plasmosomes, the first of these 
to be seen in any of these large and peculiar cells. The cell represented 
in the figure has three plasmosomes, one for each lobe of the nucleus. 
Only one plasmosome was in the particular section photographed, the 
other two being in adjacent sections. In an iron-hematoxylin van 
Gieson preparation these plasmosomes stain red with the acid fuchsin. 
The large chromatic granules characteristic of the nucleus in homolo- 
gous cells of other fishes are not present. Instead, the chromatin is 
very finely granular and is scattered throughout on the reticulum that 
extends to all parts of the nucleus. The cytoplasm gives off processes 
which look as if they might contain some nerve-fibrils. No special 
nerve-stains have yet been tried. No Nissl granules have been made 
out. Occasionally the cytoplasm shows some vacuolation—very 
slight, however, compared with that shown by the cells of the skate. 
No granules are present. The cells do not seem to be limited strictly 
to the anterior horn, but in some cases are found very close to the outer 
membranes of the spinal cord. Moreover, in some cases as many as 
five or six of these cells were present in the same section. 
