NERVOUS SYSTEM OF THE VERTEBRATA. 381 



the roots of the seventh nerve ; and it is not visible till much later than 

 the cranial nerves above described. 



In addition to this nerve Marshall has found, both in the third nerve 

 and in the fifth nerve, a series of non-gangliated roots, which arise in a 

 manner not yet satisfactorily elucidated, considerably later than, and in front 

 of, the main roots. These roots join the gangliated roots om the jiroximal 

 side of the ganglion or in the ganglion '; and Marshall believes them to 

 be homologous with the anterior roots of spinal nerves, while he holds the 

 sixth nerve to be an anterior root of the seventh nerve. 



In addition to these nerves Marshall holds certain ventral roots, which 

 occur in Elasniobranchs close to the boundary of the spinal cord and medulla, 

 and which probably form the hypoglossal nerve of higher types, to be 

 anterior roots of the vagus. It is very difficult to prove anything definitely 

 about these nerves, but, for reasons stated in my woi-k on Elasmohranch 

 Fishes, I am inclined to regard them as antei'ior roots of one or more 

 spinal nerves. 



Before attempting to decide how far Marshall's views about the so-called 

 anterior roots of the seventh, the fifth and the third nerves are well founded 

 it will conduce to clearness to state the characters and relations of the two 

 roots of spinal nerves. 



The posterior root is (1) always purely sensory ; (2) it always develops a 

 ganglion. The anterior root is (1) always purely motor; (2) it always joins 

 the postei'ior root belov) the ganglion, except in Petromyzon (though not in 

 Myxine) where the two roots are stated to be independent. 



How far do Marshall's antei-ior and posterior roots of the cranial nerves 

 exhibit these respective peculiarities'? 



With reference to the sixth and seventh nerves he states " we must 

 regard the sixth nerve as having the same relation to the seventh that the 

 anterior root of a spinal nerve has to the posterior root." On this I would 

 i-emark ( 1 ) that the posterior root of this nerve is a mixed sensory and motor 

 nerve and therefore differs in a very fundamental point from that of a 

 spinal nerve ; (2) the sixth nerve though resembling the anterior root of a 

 spinal nerve in being motor and without a ganglion, differs from the nearly 

 universal, arrangement of spinal nerves in not uniting with the seventh. 



With reference to the fifth nerve it is to be observed that it is by 

 no means certain that the whole of the motor fibres are supplied by the 

 so-called anterior roots, and that these roots differ again in the most 

 marked manner from the anterior roots of spinal nerves in joining the 

 main root of the nerve above (nearer the brain), and not as in a spinal 

 nerve below the ganglion. The gangliated root of the third nerve is 

 piu-ely motor^, and its so-called anterior roots again differ from the anterior 

 roots of spinal nerves, in the same manner as those of the fifth nerve. 



With reference to the glossopharyngeal and vagus nerves I would 

 merely remark that no anterior i-oot has even been suggested for the 

 glossopharyngeal nerve and that the posterior roots of both these nerves 

 contain a mixture of sensory and motor fibres. 



1 These non-gaugliated roots of the fifth nerve are not to be confounded with the 

 motor root of the fifth nerve in higher types. They appear to form the anterior root of 

 the adult which gives origin to the ramus ophthalmicus. 



2 If Marshall's view about the ramus ophthalmicus profundus (p. 379) is correct, the 

 third must still be, as it no doubt was primitively, a mixed motor and sensory nerve. 



