Mabch 19, 1915] 



SCIENCE 



411 



appropriate stimuli to the successive move- 

 ments, which eventually come back to the 

 starting point and so lead to the repetition 

 of the series. To avoid misunderstanding, 

 it may be well to say at the outset that no 

 one denies that afferent impulses play an 

 important role in locomotion. The phe- 

 nomena of locomotor ataxia are conclusive 

 evidence on that point; but so do afferent 

 impulses over the pulmonary fibers of the 

 vagus nerve play an important role in 

 regulating the fundamentally automatic 

 discharge from the respiratory center, with- 

 out being in any way its exciting cause. 

 The work of T. Graham Brown^ suggests 

 that the same thing is true of the rhythmic 

 movements of locomotion. Brown shows 

 that in a certain stage of ether narcosis in 

 the decerebrate animal, when reflexes can 

 no longer be elicited from the afferent 

 nerves, rhythmic movements of flexion and 

 extension occur in the hind legs; and 

 furthermore, that these movements occur 

 after the afferent nerves from the moving 

 limbs are cut. In other words, these move- 

 ments which suggest the basis of the move- 

 ments of locomotion, involving as they do 

 the alternate rhythmic action of antagon- 

 istic groups of muscles, are executed by 

 efferent neurones without any stimulation 

 from afferent neurones. They constitute 

 an "endless chain," but not an endless 

 chain of reflexes. 



This discovery seems to me to be of suffi- 

 cient importance to justify dwelling upon 

 it at some length ; and in order to obtain a 

 clear picture of the possibilities, we may 

 give briefly Brown's very plausible hypoth- 

 esis of the nature of the nervous mechan- 

 ism involved. A movement of this kind 

 consists fundamentally in the alternate 

 contraction of antagonistic groups of 

 muscles. "We may denominate the nerve 



2 T. Graham Brown, Journal of Physiology, 1914, 

 XLVII., p. 18. 



cells innervating each antagonistic group 

 as a half-center, the two together making 

 the entire nerve center for the given move- 

 ment. Brown supposes that each half-cen- 

 ter sends inhibiting collaterals to its an- 

 tagonist (reciprocal innervation of Sher- 

 rington), so that when the flexors, for ex- 

 ample, are being excited, the extensor 

 neuro-museular mechanism is inhibited. 

 He then assumes that the efficiency of this 

 inhibition rapidly diminishes — somewhat 

 as the heart escapes from vagus inhibition 

 — either by the fatigue of the inhibitory 

 mechanism or by the increase of the dis- 

 charging power of the inhibited cells. The 

 result is that in a short time the inhibited 

 center breaks through its inhibition, ex- 

 cites its muscles to contract and at the 

 same time inhibits the previously active 

 antagonistic half -center.^ The repetition of 

 these processes leads, of course, to the 

 rhythmic movements referred to. 



Brown further raises the very interest- 

 ing question whether these automatic actions 

 of locomotion do not present a more primi- 

 tive form of nervous activity than the 

 reflex. He points out the difficulty of 

 imagining the origin of a reflex are by nat- 

 ural selection, since neither the afferent 

 nor the efferent limb would be of any use 

 to the animal without the other; and it is 

 almost impossible to conceive of both aris- 

 ing at the same time by any assumed proc- 

 ess of evolution. It is far more easy to 



3 This theory assumes that the cells of both 

 half -centers are automatic and subject to the same 

 environmental conditions (e. g., tension of carbon 

 dioxid) governing their discharge. If both cen- 

 ters were in exactly the same physiological condi- 

 tion and subject to the same environmental con- 

 ditions, they would discharge simultaneously and 

 alternate rhythmic contractions of antagonists 

 ■would be impossible. This condition of equal irri- 

 tability, however, ds rarely realized. When it does 

 not obtain, one half -center wall discharge first and, 

 as explained, inhibit for the time being the dis- 

 charge of the other. 



