— l62 — 
The cells situated in the grey substance just outside the epi- 
thelial cells, are quite like the latter. They are elongated at their 
interior extremities, and penetrate between the epitheUal cells to 
the central canal (if they cary cillia, like the latter, I can not 
decide); at their exterior extremities they have processes wich pe- 
netrate to the surface of the spinal cord, like those from the epi- 
thelial cells. These cells represent, consequently, a kind of tran- 
sition-stage between the common neuroglia cells and the epithelial 
cells surrounding the central canal; indeed, in the many layers of 
such cells SLirrounding the epithelium of the canal, we can find 
all possible transition-stages. When we, now, consider that in 
Amphioxus the neuroglia principally consists of epithelial cells (we do 
not, here, refer to the sheaths surrounding the nerve-tubes etc.) and 
there is every reason to believe that this is a primary state from 
which the conditions found in Myxine are developed, I think we 
are fully entitled to say, that the neuroglia cells have an ectodermal 
origin, and spring from the epithelial cells of the central canal or we 
may rather say the central groove.^) 
The nerve-cells of the spinal ganglia, — Before we leave 
the nervous system of Myxine I will make some remarks on the 
structure of the spinal ganglion cells, as I think, they exhibit relations 
which in several respects are of great interest for the subjects treated 
of in this paper. Upon the whole the spinal ganglion cells reminds 
in their structure very much of the ganglion cells of the central 
nervous system of many invertebrates, and, amongst those we have 
here examined, especialy the large unipolar cells of Homarus vulgaris. 
They represent obviously a less developed stage than that of the 
ganglion cells of the spinal cord. 
traversing the white substance (vide 1. c. p. 179). Regarding these processes 
in higher Vertebrata I vill also refer the reader to the various memoirs by Ainslie 
Hollis (1. c. 1883), Leydig (Zelle und Gewebe I885 p. 189), Rabl-Buckhard 
(1. c. 1883), Beissner (1. c. 1864), John Dean [\. c 1863') and others. 
^) I think it is most probable that the neuroglia cells of the invertebrates 
have, also, an ectodermal origin. This may, for instance, easily be seen in Nereis, 
where the ventral nerve-cord is not yet fully separated from the ectoderm. It is, 
here, even difficult to see any difference between the neuroglia and the ectodermal 
cells, and it is in many cases impossible to draw any distinct line between the 
cells enveloping the nerve-cord, and the ectodermal cells situated outside them 
(vide fig. 14). I have, here, only considered what we may infer from the nervous 
system of full grown specimens. In examination of the embryological development 
of the nervous system, the ectodermal origin of the neuroglia will be still more 
evident; on that subject the reader may be referred to the recent p aY> ev hy Klein en- 
herg on the embryologie of the Annelides (Zeitschi-. f. wiss. Zool. 1886). 
