- IS2 — 
fibres of this kind directed towards the nerve-roots, especiall}^ the 
dorsal ones, and penetrating into the nerves. 
It is obvious that the fibres, radiating from the epithelial cells 
of the central groove, and running between the longitudinal nerve- 
cord, have just the same significance as the neuroglia-fibres issuing 
from the neuroglia-cells of higher vertebrates. No such neuroglia- 
cells are found in Amphioxus as are, for instance, found in Myxine 
(vide sequel). We may, therefore, assume that, these epithelial cells 
are the real neuroglia-cells of Amphioxus, and that this neuroglia 
represent the most primary state found amongst the Vertehrata. In 
the description of the neuroglia of Myxine we will have an oppor- 
tunity to return to this subject, 
Fibres are often seen crossing the central groove; these fibres 
seem partly to be neuroglia-fibres, partly nerve-tubes and, as we 
have seen, partly also processes from ganglion cells. 
As to the origin of the nerve-tubes of the ventral and dorsal 
nerve-roots, my investigations are yet in this respect very incomplete, 
and must be reserved for the furture paper. 
Myxine glutinosa. 
It is only the nervous elements of the spinal cord we will, here, 
describe. 
The white siihstance — On examination of transverse and 
longitudinal sections of the spinal nerve-cord of Myxine it is easily 
seen that the white substance chiefly consists of longitudinal nerve- 
tubes, which vary a great deal in size. The large tubes, the so- 
called »Muller's nerve-fibres«, are especially situated on the ventral 
side. Between the large and largish tubes and through the whole 
white substance a multitude of very slender longitudinal nerve-tubes 
occur; the white substance has thus in well preserved transverse 
sections a reticular appearance (vide fig. loo) quite similar to that of 
a section through the dotted substance of the ventral nerve-cord of 
an invertebrate (e. g. Homarus and Nereis). Between each of these 
tubes a multitude of extremely slender tubes (neuroglia-fibres and 
branches from the protoplasmic processes of the ganglion cells) run 
transversal ly. These transverse fibres are not very distinctly seen 
in so well preserved sections as fig. lOO; they are, however, very 
conspicuous in sections of preparations where the contents of the 
longitudinal tubes have shrunk in so much that the transsected tubes 
have the appearance of empty vacuoles, the white substance has 
here become very transparent (vide fig. loi). They run from the 
