REGENERATION OF SYMPATHETIC FIBRES. 653 



stimulating the cut nerve ; and in the subsequent observations of Schiff", v. 

 Beke Callenfels, and others, on the duration of the effects of excision of 

 the sympathetic, it was pointed out that a considerable length of nerve 

 must be excised to prevent regeneration. I have found 1 a return, 

 though not to the full extent, of all the functions of the sympathetic in 

 the cat twenty-four days after section. The return of function probably 

 begins in about a fortnight. Eegeneration, then, is fairly rapid in 

 the cervical sympathetic, and hence probably in other pre-ganglionic 

 sympathetic fibres. 



We have seen that in the cat seven thoracic nerves send fibres to 

 the superior cervical ganglion, and that each of these produces its own 

 peculiar effects. Hence, no two thoracic nerves send fibres to exactly 

 the same cells in the superior cervical ganglion. Moreover, since the 

 highest of the series of thoracic nerves do not produce any of the effects 

 of the lowest of the series, the highest and the lowest must run to a 

 completely different set of cells in the ganglion. 



Now, when the several thoracic nerves in the cat are stimulated 

 after regeneration of the cut cervical sympathetic, 2 each produces, in 

 the majority of cases, its normal effects. 



Thus (usually), the first thoracic nerve causes good dilatation of the 

 pupil, opening of the eye, and retraction of the nictitatiug membrane, but 

 causes little or no contraction of the arteries of the ear, and no effect on 

 the erectile hairs ; the third thoracic nerve causes very little dilatation 

 of the pupil, moderate opening of the eye, and retraction of the nicti- 

 tating membrane, good contraction of the arteries of the ear, but 

 (usually), like the first thoracic nerve, it has no effect upon the erectile 

 hairs ; the sixth thoracic nerve produces none of the effects mentioned 

 above, but causes more or less marked movement of the erectile 

 hairs. 



It is conceivable that in regeneration the nerve fibres should grow 

 through the superior cervical ganglion and become directly connected with 

 peripheral structures. But since nicotin, both when applied locally to 

 the ganglion and when injected into the blood, has its normal effect, we 

 must conclude that the nerve fibres have formed new terminations in 

 connection with the nerve cells. 



We arrive, then, at the remarkable fact that, in most cases, each 

 spinal nerve becomes connected with nerve cells similar to, if not the 

 same as, those with which it was previously connected, and with no 

 others. It is difficult to account for this, except on the supposition 

 that the regenerating fibres are guided to their appropriate nerve cells 

 by some chemiotactic attraction. The guiding influence might either 

 lie in the remains of the old fibres or in the characters of the nerve 

 cells. 



The selective connection of nerve fibres with nerve cells occurs also 

 if a regenerated nerve be cut ; and the second regeneration takes place 

 as readily as the first. 



It may be mentioned that not infrequently, though by no means 

 always, an upper thoracic nerve, after regeneration of the cervical 

 sympathetic, has not quite so strong an action as normally, as if the 

 regeneration were not quite complete. This is probably due to a 

 failure of some of the nerve fibres to make their way through the 



1 Journ. PhysioL, Cambridge and London, 1897, vol. xxii. p. 228. 

 2 Langley, ibid., 1895, vol. xviii. p. 280 ; and 1897, vol. xxii. p. 215. 



