424 
appearance, as though it had buckled up when the pull of the fibre 
ceased to be exerted. It passes insensibly into the basal portion of 
the cone, which is of a fibrous (connective) tissue indistinguishable 
from the wall of the sinus terminalis, and may be looked upon as 
that portion of the wall into which Reıssxer’s fibre is inserted and 
which has been drawn into the sinus terminalis by the tension of 
the fibre. Fig. 2 (Pl. I) shows the terminal plug in Gasterosteus. 
This condition, which I regard as the normal one, was not observed 
in the skate. Of that animal, apart from the material used in the 
experiments, I had but two tails sectioned which had been thoroughly 
fixed before the spinal cord was cut, and of these, one was useless 
for the examination of this particular point owing to its having become 
twisted during the paraffin embedding in such a way that sections, 
sagittal to the greater part of the tail, cut the extremity very obliquely. 
In the second, however, which had been preserved entire in spirit, 
Reıssner’s fibre was found stretching back through the central canal 
in a straight line to end in the sinus terminalis in a mass of coiled 
fibre which formed a heap upon the postero-ventral wall of the terminal 
sinus and probably conceals the terminal plug. That it must have 
existed in this condition in the living animal seems certain, for, although 
the central nervous system was well exposed to the fixing fluid, it 
was not cut or injured in any way until the specimen had been dead 
for many hours. 
Such a coiling of the fibre terminally may not infrequently be met 
with in well preserved material which has been hardened entire, and 
can only be explained by the supposition that the fibre is liable during 
life to accidental breakage and recoil, a supposition which is borne 
out by further observations to be recorded upon Rhina and Scyllium )). 
Of Rhina squatina I have examined sections of three tails, two 
of which (A, B) were thoroughly hardened before they were removed, 
while the third (C) was cut off in the fresh condition. 
In this species, the sinus terminalis extends scarcely at all beyond 
the end ot the vertebral column, upon the upper surface of which it 
lies with its extremity wholly uncovered by neural arches. It is a 
rather narrow chamber (the central canal of the filum terminale dilating 
very slightly), and it is prolonged into a pointed extremity which is 
directed posteriorly and, just a little, dorsally. The walls of the sinus 
1) I have also obtained a considerable amount of evidence upon 
this point from an examination of a quantity of cyclostome material, 
a full account of which will be given in a paper, dealing with Rrıssner’s 
fibre in that class, which is now nearly ready for publication. 
