I2O 



J. B. JOHNSTON. 



One fact makes it seem probable that there is a movement 

 peripherally of the spinal ganglion cells in AnipJtioxus after the 

 period in the ontogeny when the longitudinal fibers of the cord 

 are formed. Many of the fibers adjacent to the roots are bent 

 out far into the root where they recurve and pass on in their 

 former course within the cord. A slight case of this is shown in 

 Fig. 4. In many cases such recurved fibers extend much 

 further out into the nerve roots, and the number of fibers affected 

 in this way is so great that the bulging laterally of the cord 

 toward each root is very striking in horizontal sections. Oc- 

 casionally some of the giant fibers are carried toward a root until 



FIG. 4. Horizontal section of dorsal 

 root showing ganglion cells in the trunk 

 and rami. Combined from two sections. 



FIG. 5. Two cells whose neurites 

 ramify at once on the opposite side of the 

 cord. 



they seem as if they were about to enter it. I can think of no 

 other cause for this curving of fibers out into the roots except 

 the possible active migration of the ganglion cells. 



The place of branching of the dorsal roots into dorsal and 

 ventral rami is of some interest. The branching seldom takes 

 place close beneath the dermis as Hatschek describes it. On the 

 other hand, the division of the root near the cord and the separate 

 origin of dorsal and ventral rami directly from the cord are of 

 more frequent occurrence than Rohde states, and are not con- 

 fined to the anterior end of the body. The separate origin of 

 the rami may be seen in the case of one or several roots in a 

 considerable majority of specimens. Every possible gradation 



