THE SEGMENTAL VALUE OP THE CRANIAL NERVES. 131 



of zoologists ; and during the last five or six years our knowledge on 

 these points has received very material additions, additions which 

 have, on the whole, tended strongly to confirm Gegenbaur's views, 

 while causing modification of them in many secondary points. 



The most important of these more recent contributions is un- 

 doubtedly the series of facts brought to light by Balfour concerning 

 the early stages of development of the spinal and cranial nerves in 

 Elasmobranch fishes. Balfour showed that, 1 contrary to the generally 

 accepted theory, the nerves are outgrowths from the central nervous 

 system, and therefore of epiblastic origin, instead of being, as formerly 

 supposed, structures arising independently in the mesoblast and only 

 acquiring a secondary connection with the brain and cord. 



In the case of the spinal nerves, he showed that the two roots, 

 anterior and posterior, arise separately and independently ; that the 

 posterior roots are local outgrowths of a continuous longitudinal band 

 — the neural crest — which grows out along the mid-dorsal line of the 

 spinal cord. By lateral growth of the dorsal summit of the cord the 

 nerve roots of the two sides, which are at first directly continuous with 

 one another across the top of the cord, become separated to a certain 

 extent. The nerve root on either side grows downwards, closely 

 applied to the side of the cord, it then acquires a new or secondary 

 attachment 2 to the side of the cord, some little distance below the 

 primary one. A little later the primary attachment disappears, and 

 the secondary alone remains as the permanent attachment of the 

 posterior root to the cord. 



The anterior roots arise later than the posterior, each as an indepen- 

 dent conical outgrowth from the latero-ventral angle of the cord. 

 The roots grow rapidly, and soon form elongated bands of fusiform 

 cells, which retain their original points of origin from the cord. Each 

 is at first, and for some time, quite distinct from the posterior root, 

 with which, however, it subsequently unites to form the adult spinal 

 nerve. 



Further differences between the anterior and posterior roots 

 are afforded by the fact that the posterior develops at a very 

 early period a large ganglionic swelling — the future spinal ganglion — 



1 Balfour, "On the Development of the Spinal Nerves in Elasmobranch Fishes," Phil. 

 Trans, vol. clxvi. pt. 1, 1S75 ; and A Monograph of the Development of Elasmobranch Fishes, 

 1878, pp. 156-161 and 191-205. 



E The account of this shifting is based on my own observations. Balfour expresses 

 himself as " inclined to adopt this view" (Comparative Embryology, vol, ii. p. 372), but does 

 not definitely do so. 



