— 169 — 
this respect, and it is well known that a few moments want of nutrition 
for the brain is enough to disorder the nerve-system and produce 
unconsciousness in man. The nutrition is certainly not a function of 
too insufficient importance to explain the extension the ganglion 
cells have in the central nervous system. 
That the ganglion cells of the spinal ganglia have a 
nutritive function is in my opinion vere probable. As the nerve- 
tubes of the dorsal (posterior) nerve-roots have got no special nu- 
tritive centres in the central nerve-system (as they do not originate 
directly from cells) they have got them in special ganglia outside 
the central nerve-system. In Myxine the spinal nerve-cells are uni- 
polar or bipolar. The processes of the imipolar cells suhdivide di- 
chotomically in a T and send one branch peripherically and one 
centrally (cf. Freud's investigations 1. c. 1878). In my opinion the 
cell is the nutritive centre of both these branches, but is of no di- 
rect importance for their sensitive function ; the irritation of the peri- 
pheral branch can not pass through the cell (as it then must pass 
and return through the same process) but passes directly into the 
centripetal branch at the subdivision in T of the process. In the 
bipolar cells the irritation of the peripheral part of the nerve-tube 
must certainly pass through the cell to come into the centripetal part; 
in so far it is no objection that the bipolar cell may be of direct 
importance for the sesitive function of the nerve-tubes. But the fact 
only that in the same ganglion, imipolar cells are present and are 
united with quite similar nerve-tubes must be enough to make us 
doubtful in this respect. If the unipolar cells have a nutritive func- 
tion only, it must be probable that the bipolar cells have the same 
function; indeed, seeing that in higher verterbrata the spinal ganglia 
consist exclusively of unipolar cells (notwithstanding the slender 
protoplasmic processes described by Fritsch (1. c. 1886) which 
processes I suppose to be nutritive processes, cf. above p. 146, 
note i), I think that this must be quite certain. There are also ap- 
pearances in the structure of the bipolar cells, which, as we have 
before seen, indicates that the contents of the processes to some 
extent pass through the protoplasm of the cells without intermixing 
with it (compare the description of the protoplasm of these cells 
p. 163). 
We arrive thus at the conclusion that all the nerve-cells of the 
spinal ganglia have a nutritive function only. They are probaly of 
a similar importance for the sensitive (centripetal) nerve-tubes as the 
»motoric« ganglion cells (i. e. the cells from which the motoric 
