POSSIBLE AUTO NO MIC FIBRES IN NERVES. 669 



a special kind of autonomic fibre unrepresented in other spinal nerves. And 

 such fibres, if they exist, must be special forms of nerve fibres, distinct from all 

 others. 



We have seen that the fibres running from the central nervous system to 

 the sympathetic, to the cranial and sacral autonomic systems, end in connec- 

 tion with nerve cells. The spinal nerves have 110 nerve cells of the sympathetic 

 type on their peripheral course, so that, if the alleged direct fibres end at all in 

 cells of the sympathetic type, it must be in the posterior root ganglion. Even 

 in this case they would form a class of nerve fibres distinct from other auto- 

 nomic nerve fibres. But there is no evidence of the presence of characteristic 

 sympathetic nerve cells in the posterior root ganglia, and if this be so, the 

 fibres must differ widely from other autonomic fibres, for either they are con- 

 nected with a special form of nerve cell, or they have no nerve cell at all on 

 their course. 



We may turn now to the particular cases which have been brought 

 forward. These are (1) The third cervical nerve (vasomotor fibres) ; 

 (2) the nerves running to the brachial plexus (vasomotor and secretory 

 fibres); (3) the nerves forming the lumbo-sacral plexus (vasomotor 

 and vaso-dilator secretory fibres). 



As we have already said (p. 624), Schiff showed that the great aur- 

 icular nerve in the rabbit contains vaso- constrictor fibres for the ear, 

 and Loven that the vaso-constrictor fibres run chiefly to the distal part 

 of the ear. Schiff considered that these fibres issue from the spinal 

 cord in the roots of the cervical nerves, which give rise to the great 

 auricular. He found l that degeneration of the fibres which run from 

 the superior cervical ganglion did not dimmish the vasomotor action of 

 the auricular nerve. This observation, however, does not touch the 

 question, for it has been shown by Fletcher that in the rabbit vaso- 

 motor fibres run to the great auricular nerve from the ganglion stellatum. 

 In order, then, to show that the cervical nerve roots in the rabbit have 

 vaso-constrictor fibres, the cervical nerves should be stimulated a week 

 or more after extirpation of both the ganglion stellatum and the sup- 

 erior cervical ganglion. This has not been done in the rabbit, but one 

 case has been placed on record by Morat, 2 in which the operation was 

 performed. Morat stimulated the great auricular nerve, after allowing 

 time for degeneration, and found, as a result, dilatation of the vessels of 

 the ear. The experiment, however, can hardly be regarded as a con- 

 clusive one, since normally the great auricular nerve does not cause 

 vascular dilatation. 



I have made a few experiments in the cat and rabbit, and have not found 

 any effect on the vessels of the ear when the cat's upper cervical nerves are 

 stimulated close to the dura mater. Sometimes, when the electrodes were 

 shifted up to the end of the nerve from the vertebral canal, there was con- 

 traction of the artery at the tip of the ear, from the second cervical nerve in 

 the cat, and the third in the rabbit. 



On the whole, I think the evidence is against the view that vasomotor 

 fibres issue in the roots of upper cervical nerves. 



The experiments brought forward to show that direct autonomic 

 fibres occur in the nerves to the limbs, fall into two classes (a) Those 

 which are said to show that vasomotor fibres pass out in the anterior 



1 Schiff, "Lecons sur la physiol. de la digestion," Turin, 1867, Lecon 11. 



2 Arch, de physiol. iwrm. etpath., Paris, 1891, p. 93. 



