ye 
angular shape, and quite large, having 4—10—12 processes, which 
may be varicose, and piercing the capsule are distributed among the 
unipolar cells of the ganglion. Dogren (5) states that they resemble 
very closely in shape and structure the sympathetic cells of these 
animals. 
As to the function of these multipolar cells, nothing tangible 
has as yet been ascertained. LenaosséK has suggested that these 
cells may have relation to the sympathetic fibres which enter the 
spinal ganglia as described by EHRLICH (6) for the frog and by Ramén 
y Cayat (7) for the rat; that their dentrits may serve to bring about 
better contact with the collateral of said sympathetic fibres. Atten- 
tion needs, however, to be drawn to the observations recently pub- 
lished by DocIEL, to the effect that while he found it difficult to 
determine around which of the several forms of the spinal ganglion 
cells the sympathetic fibres terminated, he is led to believe that they 
end in pericellular baskets surrounding the bodies of cells, to which 
he has given the name, spinal ganglion cells of the second type. These 
hitherto undescribed cells possesses an axis cylinder which after 
repeated branching terminate in pericellular baskets which enclose 
the cell bodies of the unipolar sensory cells. 
That the multipolar cells have some special function may be in- 
ferred from the fact that they seem quite generally distributed 
throughout the vertebrates. What this function is would seem at 
present an open question. 
I have thus briefly summarized the literature dealing with the 
presence of multipolar cells in spinal ganglia, partly to bring together 
our knowledge concerning these structures, and also to emphasize the 
fact that the cells I am about to describe seem thus far to have escaped 
the observations of investigators in this field. 
The observations to be recorded were made on the spinal ganglia 
of the large American bull-frog, Rana Catesbiana, SHaw. The 
preparations were obtained from animals which had been injected through 
the large lateral cutaneous vein, with a 1 to 4 per cent. solution of 
methylene blue in normal salt. Thirty minutes to an hour after the 
injection, the lumbo-sacral ganglia were removed and fixed after 
BrTue’s method. Some of the ganglia were then embedded in paraffin, 
sectioned and as a rule double stained in alum carmine; other ganglia 
were teased. 
The cells about to be described differ from the ordinary unipolar 
cell in that several delicate branches are given off from the prominent 
axis cylinder before it leaves the capsule of the respective cell, these 
30 * 
