82 THE PHYSIOLOGY OF MUSCLE AND NERVE. 



the motor fibers to the voluntary muscles emerge from the spinal 

 cord in the anterior roots, while the fibers that give rise to sensa- 

 tions enter the cord through the posterior roots. These facts have 

 been demonstrated beyond all doubt. Magendie discovered an 

 apparent exception in the phenomenon of recurrent sensibility. 

 When the anterior root is severed and its peripheral end is stimu- 

 lated only motor effects should be obtained. Magendie observed, 

 however, upon dogs that in certain cases the animals showed signs 

 of pain. This apparent exception to the general rule was after- 

 ward explained satisfactorily. It was shown that the fibers in 

 question do not really belong to the anterior root, that is, they do 

 not emerge from the cord with the root fibers; they are, in fact, 

 sensory fibers for the meningeal membranes of the cord which 

 are on their way to the posterior roots and which enter the cord 

 with the fibers of the latter. Since the work of Bell and Magendie 

 it has been a question whether their law applies to all afferent and 

 efferent fibers and not simply to the motor and sensory fibers proper. 

 The experimental evidence upon this point, as far as the mammals 

 are concerned, has accumulated slowly. Various authors have shown 

 that stimulation of the anterior roots of certain spinal nerves may 

 cause a constriction of the blood-vessels, an erection of the hairs 

 (stimulation of the pilomotor fibers), a secretion of sweat, and so 

 on, while stimulation of the posterior roots in the same regions is 

 without effect upon these peripheral tissues. One apparent excep- 

 tion, however, has been noted. A number of observers have found 

 that stimulation of the peripheral end of the divided posterior 

 roots (fifth lumbar to first sacral) causes a vascular dilatation in 

 the hind limb. The matter has been particularly investigated by 

 Bayliss,* who gives undoubted proof of the general fact. At the 

 same 'time he shows that the fibers in question are not efferent 

 fibers from the cord passing out by the posterior instead of the an- 

 terior roots. This is shown by the fact that they do not degenerate 

 when the root is cut between the ganglion and the cord, as they 

 should do if they originated from cells in the cord. Bayliss's own 

 explanation of this curious fact is that the fibers in question are 

 ordinary afferent fibers, but that they are capable of a double ac- 

 tion: they can convey sensory impulses from the blood-vessels to 

 the cord according to the usual type of sensory fibers, but they 

 can also convey efferent impulses, antidromic impulses as he desig- 

 nates them, to the muscles of the blood-vessels. In other words, 

 for this special set of fibers he attempts to re-establish the view 

 held by physiologists before the time of Bell, namely, that one 

 and the same fiber transmits normally both afferent and efferent 

 impulses. An exception so peculiar as this to an otherwise general 

 rule cannot be accepted without hesitation. Some facts are given 

 * Bayliss, "Journal of Physiology," 26, 173, 1901, and 28, 276, 1902. 



