205 
appear to commence at the epoch named in my Lepidosteus 
paper; i. e. with the formation of the permanent central canal of the 
cord. On this point it is necessary to correct an error in the paper 
just mentioned. The processes of the ganglion cells there spoken of 
have turned out to be either nerves (axis-cylinders) or the remains of 
sucht). These are not „cut off“, but, like the macro- 
ganglion cells themselves, they undergo a gradual 
simple atrophy, and traces of them can be seen fora 
long time, especially in the form of shrunken pro- 
cesses from poles of the ganglion cells. The latter persist 
for a much longer period, and, indeed, have not entirely disappeared 
in newly hatched R. batis of some 20 centimetres in length. The 
lengthened period, which complete atrophy would appear to require 
in this case, may be judged of, when it is stated that the process 
began during the 5 month of life within the egg, while traces of 
the cells could still be found in skate, which hatched out some 17 months 
after the fresh egg was taken from the parent. 
8) In addition to the motor, there is sometimes present, but as 
yet it has not been shown to exist in every embryo, a subepithelial 
(subepiblastic) system of nerves, possibly of a sensory nature. It 
arises in connection with ganglion cells of the cord (as an outgrowth 
of cells of the latter) slightly ventrad of the motor macro-ganglion 
cells. This system has also been seen in embryos of 45 mm. in length, 
but its complete history can be at present only a matter of surmise. 
Further investigation is necessary, there can, however, be no doubt 
of its larval and transient nature ®). 
1) See note 2 of preceding page as to poles of ganglion cells. More 
than one axis cylinder would occasionally appear to be connected with 
any one cell. 
2) There seems to be no likelihood of its homology with the plexus 
recently described by Rerztvs in Myxine. At first sight, this system in 
Raja would seem to afford evidence in favour of the Professors HErtwie’s 
hypothesis of the origin of the nerves of sensation in the Vertebrata from 
a subepithelial plexus such as would appear to exist in many Invertebrates 
(or as quite recently found by Lenossex in the earthworm). Vide Hxrt- 
wie, O., „Entwiekelungsgeschichte des Menschen etc.“, 2. Aufl. 1888, 
p- 356. The transient nature of the apparatus appears to me to militate 
against such a view. Renewed studies, undertaken while this paper was 
in the press, lead me to seriously doubt a sensory character of these 
subepiblastic nerves. My observations are not yet complete, but so far 
they seem to lead in the direction of a recognition of the motor nature 
of these nerves. 
