220 Papers from the Department of Marine Biology. 
where Bouin’s fixative containing 5 per cent potassium bichromate 
followed by a hard stain in iron hematoxylin has brought them out 
black. In every case where the decolorization was carried far enough 
to show the chromatin nucleoli the bodies in question were decolorized 
and could not be made out by the best lenses and refraction conditions. 
The alcohols took out any brown or yellow color that they possessed. 
I shall follow Held and Garten in calling them neurosomes. 
These neurosomes were found scattered all through the cell-body 
(cytoplasm) and appeared to lie in much more intimate relation to the 
fibrillar substance than the Nissl bodies did. Thus they appeared to 
lie in spaces between the Nissl bodies. While scattered sparingly 
through the whole cytoplasm, they were gathered in large masses at 
the pole of the cell from which the axis-cylinder took its origin. Here 
they formed a thick ring around the axis-cylinder process and are to 
be seen in figures 10 to 18, plate 3, where 9 cells from two torpedoes are 
shown after bits of the electric lobes had been macerated and teased 
out on the slide. In this case the neurosomes were a rich golden brown. 
In figure 3, plate 2, a cell is shown as it appeared in the living state 
under closed substage diaphragm, thus showing the parts of the cyto- 
plasm by diffraction. 'The neurosomes can be very easily seen under 
this condition apart from any color that they may or may not possess. 
The neurosomes are also shown by the strongest osmic-acid fixations, 
but are not a fat of the ordinary kind (see Dogiel, 128). They are found 
in other nerve-cells and can be best seen in living tissue, pressed between 
the thin cover and a slide and examined with the highest powers and 
with the small diaphragm. 
These ‘“‘neurosomes”’ are the bodies that have often been described 
by all writers on the structures of the nerve-cell as “‘ pigment granules,”’ 
or collectively as the ‘‘pigment body.’ They have, perhaps, been 
best studied in the spinal ganglion nerve-cells of vertebrates, especially 
the mammals. Dogiel, in a review of the structures of these cells, 
in a book of comparatively recent date (12 B), describes them and cites 
a valuable list of writers on the subject who have discussed the struc- 
tures from a physiological and chemical side. This summary, while 
not giving the substance a definite name, or associating it with any 
known function of the cell, does show the structures to be one and the 
same with the structures described above by the present writer and, 
further, clearly shows them to be a constant and specific organ of 
practically all nerve-cells rather than the products of senile or patho- 
logical nerve-cells. That they are closely associated with the func- 
tional polarity of the cell, as a nerve-cell, is clear from their concentra- 
tion near the origin of the efferent process, but that they are not in any 
way concerned with the polarity of the nucleus under discussion is 
equally clear, and they will not be further discussed in this connection. 
A slightly different and thin layer of homogeneous cytoplasm, the 
perinuclear cytoplasm, is found around the nucleus. It is marked 
