ee 
lu. 7 millimetres caudad of this, or 1 mm from the tip of the tail, 
the lumen has a diameter of 4 « and the fibre a diameter of 0,4 u. 
In a younger specimen, 1,2 cm long, the lumen of the canal half way 
between the anterior and posterior ends is 15 « in diameter, the fibre 
0,9 u. 1 millimetre from the posterior end the lumen is 10 w in dia- 
meter, the fibre 0,9 «. The lumen narrows rapidly from this point 
caudad, having, at a distance of 0,5 mm from the end, a diameter of 
3 u, while the diameter of the fibre at this point is 0,6 u. 
The conditions in larval Lepidosteus are much the same. In a 
specimen 1,5 cm long the fibre had a diameter of 1,2 « through the brain 
ventricles and canalis centralis to near the posterior end. 3 millimetres 
from the end the diameter was 1 mw, the lumen 20 u; 1 mm from the 
end the whole cord had a diameter of 45 u, the lumen 1,5 «, the fibre 
0,7 u; 0,6 mm from the end the lumen was 9 w and the fibre 0,6 u. 
At the posterior end of the cord the lumen of the canalis centralis 
diminishes rapidly, at the same time losing the definite outline formed 
by its epithelial cells, until it becomes merely a Series of continuous 
intercellular spaces. This extreme posterior portion of the cord has 
a very primitive structure, and embryonic conditions prevail here at 
a late stage of development. Rerrtssner’s fibre slowly diminishes in 
size, but not as rapidly as the lumen. It can be followed until the 
lumen can no longer be distinguished and then disappears in the loose 
mass of cells forming the primitive terminal part of the cord. 
Throughout the course of REıssner’s fibre one may occasionally 
see fine fibrille coming off from it and running outward toward the 
periphery of the canalis centralis. These have been seen both in cross 
and longitudinal sections, but as they are very small, usually having a 
diameter of 0,1 « or less, it is difficult to make much out of them. They 
are largest and most numerous in the posterior part, and at the extreme 
posterior end can be made out most clearly (Fig. 5). The diminution 
in the size of Reıssner’s fibre at its posterior end is probably due to 
the giving off of numerous fine processes of this sort. In one or two 
instances these processes have been seen to pass between the epithelial 
cells lining the canalis centralis and into the tissue of the cord. 
In its normal condition the course of Reıssner’s fibre through 
the canalis centralis is perfectly straight. Occasionally the fibre is seen 
to have an undulating course (Fig. 4), as ifit had been under a tension 
sudden release from which allowed it to recoil. This undulating course 
has been occasionally observed in both the canal and ventricle in Amia, 
Cynoscion, Anolis, mouse and other animals. In one series of the 
