CENTRAL NERVOUS SYSTEM. 637 



cord and the spinal ganglion. Study of its development has shown that in 

 the first instance the spinal ganglion becomes connected with the cord by the 

 outgrowth from the cells of the ganglion of those fibres which form the dorsal 

 root. It would follow that as the cells of the spinal ganglion can regenerate 

 the fibres which pass toward the periphery, they should also be able to regen- 

 erate those which form the dorsal root, but as yet there have not been reported 

 any cases where a dorsal root has been thus re-formed. 



That the regeneration is due to an outgrowth of the central stump has 

 been clearly shown by Huber, 1 who inserted a bone tube between the two ends 

 of the sciatic nerve of the dog, and obtained regeneration of the nerve with 

 a return of function although the initial interval between the two parts of the 

 nerve was more than three centimeters. The rate of growth from the central 

 end has been specially studied by Vanlair. 2 In the facial nerve of the rabbit, 

 function was restored in eight months after section, and in the pneumogastric 

 and ischiadic nerves of the dog in about eleven months. In the latter case, 

 this gives an average rate of growth of about 1 millimeter a day. In the 

 scar-tissue between the two parts of the nerve the rate is not more than 0.25 

 millimeter a day, and hence the return of function tends to be delayed by any 

 increase in the distance between the cut ends of the nerve. It appears also 

 that the return of the cutaneous sensibility is more rapid than the return of 

 motion (Howell and Huber). 



On testing the capacity of the sciatic nerve for repeated regeneration 

 Vanlair found that in a dog, when it was cut a second time, it not only regen- 

 erated but did so more rapidly than in the first case. 



Much interest has always attached to the exact course taken by the regen- 

 erating fibres. They appear in a general way to be guided by the old sheaths 

 of the peripheral portion. But the peripheral nerves contain both afferent and 

 efferent fibres, and it would appear most probable that in the process of re-for- 

 mation these should undergo much rearrangement. Since the peripheral por- 

 tion of the nerve acts as a guide to the growing fibres, the experiment has 

 been tried of cross-suturing. Thus Howell and Huber 3 having cut both the 

 median and ulnar nerves in dogs, sutured the central end of one nerve to the 

 peripheral end of the other, and obtained reunion with extensive return of 

 sensation and movement, and without inco-ordination to be attributed to the 

 unusual arrangement of the nerve-fibres. Such a rearrangement without 

 inco-ordination is not easy to explain in view of the association of certain 

 functions, such as the control of a given set of muscles, with a special cell- 

 group in the cord. The most remarkable observation, however, on the regen- 

 eration of nerve-trunks has recently been reported by Langley. 4 The pre- 

 ganglionic nerve going to the superior cervical ganglion of the cat is composed 

 of fibres with several functions. These fibres are derived from the first 

 thoracic nerve, which mainly controls those cells in the ganglion that are 



1 Journal of Morphology, 1895, vol. xi. 



1 Archiv de Physiologic normale et patholoffique, 1894. 



3 Loc. cit. * Journal of Physiology, 1895, vol. xviii. 



