IR, 
the myotome, but outside of it, a macro-ganglion cell connected either 
by contact or fusion with such a fibril as that in fig. 3 (ma. n.). 
gl.c. 
Fig. 4. Transverse section of a R. 
batis of 20 mm. 
A not infrequent variation is depicted in fig. 4, taken from a 
R. batis of 20 mm. in length. Lying close to the dorsal aspect of 
the spinal ganglion, there is a macro-ganglion cell (w. gl. c.), connected 
by means of an axis-cylinder process with a central cell (gl. c.) on 
the one hand, while on the other it is in contact with one or two 
cells of the myotome (mp.) by means of a similar nerve process from 
its other side. The ganglion cell is here not only intercalated') in 
the course of the fibril, but is to be looked upon as the genetic 
forerunner of the latter. The history of this cell is easily deducible 
from the conditions in earlier embryos, in which the wanderings of 
such macro-ganglion cells from the cord into the mesoderm (as in 
fig. 1) play so prominent a röle. Undoubtedly this cell has migrated 
to its present position — indeed, all stages of its history could be 
illustrated from actual examples. As it wandered, it spun out an axis- 
cylinder process keeping it in touch with the centre, while also 
sending out a corresponding process to the myotome. In many in- 
stances a „free ending“ of such axis cylinders in the myotome has as 
yet been alone made out. 
One of the facts of interest about this transient apparatus is a 
novel kind of ending for a motor nerve fibril. In a certain sense new, 
it throws a flood of light on the meaning of the motorial end-plates 
of muscle, and affords evidence,. stronger than any hitherto adduced, 
for the view of a ganglionic nature of such endings put forward by 
the present writer in more than one paper, and reiterated in the 
communication which succeeds this. 
1) In many cases like the present one the axis-cylinder would appear 
to pass along the ganglion cell, i. e. the latter is closely applied to it. 
