732 SEMICIRCULAR CANALS. [BOOK in. 



section of the eighth nerve or of its vestibular portion (the 

 eighth nerve consists really of two distinct nerves, the vestibu- 

 lar nerve and the cochlear nerve) the semicircular canals being 

 left intact ; and in cases where the experiment has been possible, 

 the effect of dividing one semicircular canal has been reproduced 

 by the section of the nerve branch distributed to its ampulla, 

 without any injury to the canal itself. 



Now in all animals the three canals are placed in the three 

 planes of space in relation to each other, though not necessarily 

 so that one of them lies exactly in the median plane of the head. 

 Hence whenever the head is turned the cristae of the ampullae 

 are unequally affected by the changes of pressure or of flow 

 which the turning brings about in the endolymph of the canals, 

 and when the head is at rest the relations of the endolymph to 

 the several canals are different in the different positions of 

 the head. And these facts naturally suggest the view that 

 according to the relations of the endolymph to the ampullae, 

 impulses are generated in the cristae, which impulses pass- 

 ing up to the brain supply the data by which the animal 

 becomes aware of the position of his head and so of his 

 body, and enter into the coordination of his movements ; these 

 impulses in fact are the special afferent impulses spoken of 

 a little while back. This view is further supported by the 

 following experimental results. Characteristic movements of 

 the head or eyes may be obtained by carefully laying bare a 

 canal and gently blowing over the contained endolymph with 

 a fine glass cannula, the movements differing according as the 

 current of air drives the endolymph towards or away from the 

 ampulla. Similar characteristic movements may be brought 

 about by stimulating the cristse in various ways, as by passing 

 a fine hair into the ampulla, or by suddenly heating or cooling 

 the canal, or, in certain cases at least, by passing an electric cur- 

 rent through the ampulla, the last two operations not necessi- 

 tating the opening of the bony canal. 



Without entering into any discussion as to the exact way in 

 which the impulses are generated, as to whether the results for 

 instance of the section of a canal are due to the lack of normal 

 impulses or to the introduction of abnormal ones, we may say 

 that the evidence seems to shew beyond doubt that the cristae 

 of the ampullae are organs through which are generated, accord- 

 ing to the position of the head, afferent impulses which form 

 the basis of the sense of equilibrium and enter into the coordi- 

 nation of movements affecting that equilibrium. 



The three canals however are only affected by a turning of 

 the head ; when the head otherwise unmoved is carried directly 

 forwards or backwards, or directly upwards or downwards, the 

 effect on all the ampullae is the same. Yet we are as aware of 

 these movements as of turning movements, and we may con- 



