GEROULD: CAUDINA. 
27 
the body were not under tension, and the presence of a layer of 
Hat epithelial cells lining the epineural spaces everywhere, except 
on the side occupied by the nerve band, prove conclusively, it 
seems to me, that these are normal cavities. The fibers which 
pass across the radial epineural canal are similar to the transverse 
fibers of the covering epithelium which pass through the nerve 
bands; and it is probable that the former, as well as the latter, 
serve as a support for the nerve band. 
The radial epineural canals are continuous in front with a 
circular canal which lies anterior to the nerve ring. Cross sections 
through the nerve ring uniformly show this circular canal ( dn . crc., 
Plate 3 , fig. 42; Plate 4 , fig. 44). Although most observers 
have overlooked it entirely, J. Muller (’50) and Teuscher (’76) 
apparently observed it, although their interpretation of the relations 
of the neural canals to the radial nerves and to the nerve ring: 
are inaccurate. Herouard (’89) found in Cucumaria a similar 
space, and states expressly that “ Idespace extra-nercien qui existe 
sur toute la longueur de chaque bande nerveuse externe se continue 
aussi au-dessus de Vanneau nerveux Cu6not (’91), on the 
other hand, states that there is no circular epineural canal, and that 
the radial epineural space terminates blindly at the two extremities 
of the radial nerve. He presents no evidence, however, in support 
of this assertion. Ludwig (’91 b ) has followed the process of the 
formation of the epineural space in Cucumaria Planci. The nerve 
bands appear at the end of the fourth day as a ring-shaped 
thickening of the superficial ectoderm around the mouth, with 
five radial ectodermic thickening's running out from it. This funda- 
ment of the nervous system, composed of the deeper cells of the 
ectoderm, later cuts itself off from the more superficial portion 
of this layer, and by sinking down away from the surface gives 
rise between the nervous layer and the permanent ectoderm to 
a fissure-like space, which is destined to become the epineural 
cavity. This condition he finds both along the radial bands 
and along the nerve ring. The process is clearly not one of 
invagination, and hence the epineural canals of holotlmrians can 
hardly be regarded as homologous to those of echinoids and 
opliiurans, which owe their origin, as is believed, to a real 
invagination. 
Function of the neural canals. — The most probable view 
hitherto advanced as to the function of the neural canals seems 
