26 Papers from the Department of Marine Biology. 
membrane that lines the vacuole. In this case the membrane is shown 
shrunken away from the cell cytoplasm so that it becomes plainly visi- 
ble. <A fully ripe vacuole (fig. 37) consists of deeply staining granules 
of various sizes. No precipitate is to be seen in these. Figure 38 
shows a very large granule with other smaller granules in process of 
fusion with it. As it does not stain uniformly with the hematoxylin, 
it is probable that some of the material has not yet been completely 
built up into the ordinary granular secretion. 
Before the ripe granules are fully formed the vacuoles may become 
separated from the cell that produced them. Figures 39, 40, and 41 rep- 
resent vacuoles that have become separated from the cell at various 
stages in the process of granule formation. Figure 39 shows a vacuole 
with lightly staining precipitate only. In the vacuole of figure 40 are 
small granules with still some indication of the lightly staining precipi- 
tate. This vacuole was evidently separated from the mother cell later 
in its history than the preceding one. Figure 41 shows a vacuole with 
fully ripe granules, many of them in process of fusion to form larger 
ones. A remarkably fine group of vacuoles and granules outside the 
cells is shown in the photograph of figure 22. Some of these vacuoles 
show only the lightly staining precipitate, others show both the pre- 
cipitate and granules, and still others only the fully formed granules. 
A number of granules are also present in the tissues of the spinal cord 
not inclosed by any vacuolar membrane. The ends of two of the large 
cells which gave rise to all of these vacuoles and granules may be seen. 
It has been stated that the granules are probably absorbed by the 
blood, being dissolved as absorption takes place. In sections taken 
through the entire tail, however, granules were also found in the vascu- 
lar connective tissue that is present lateral to each electric organ. 
These sections were taken about a half inch from the posterior tip of the 
tail. In sections farther anterior at the level of the first tail-fin no 
granules were seen. The significance of granules in this vascular con- 
nective tissue has not been determined. 
Mention must also be made of a few morpohlogical variations or 
peculiarities that occur in connection with these cells. It has already 
been stated that vacuoles become separated from the cells. Sometimes 
a cell process, or part of a cell process, will break off entirely from the 
cell proper, losing all connection with it. Inasmuch as the cell is ex- 
ceedingly variable in its position in the spinal cord, sometimes with a 
process that extends ventrally down past the central canal, reaching as 
far as the ventral spinal artery, and sometimes with a lateral process 
extending out to the enveloping membranes of the spinal cord, it is 
clear that these broken-off cell processes may be found in any region of 
the ventral half of the cord. If not enough nuclear material has also 
been broken off, the cytoplasm undergoes degeneration, taking on a 
peculiar dark-blue stain with iron hematoxylin. Many puzzling “dark 
