THE SEMICIRCULAR CANALS. 1201 



position and of its progressive acceleration in a straight line, and, in 

 the case of rotation, angular acceleration will also be indicated. The 

 vestibule and semicircular canals in Mach's scheme constitute B; the 

 vestibule possibly having to do with the sense of acceleration of movement 

 in a straight line, while the semicircular canals serve for angular accelera- 

 tion. During each angular acceleration around an axis passing through 

 the plane of the canal, there will be a traction or pull in the opposite 

 direction on the nerves of the ampullae, and this will give rise to the 

 sense of angular acceleration. Each excitation produced in this manner, 

 or, as regards the nerves of the ampullae, in accordance with the law of 

 the " specific energies " of nerves, in any other way, will give rise to a 

 sense of rotation. Thus Mach, like Goltz, attaches importance to the 

 fluid in the canals, and by varying pressures of this fluid, produced by 

 angular rotation, and always in a direction opposite to the rotation of 

 the body, sensations of movement are excited, which in turn cause the 

 vertigo of Purkinje and the phenomena of Flourens. 



The hydrodynamical theories advanced by Breuer and Crum Brown 

 differ only in detail from that of Mach. All three observers uphold 

 a dynamical theory, as opposed to the statical theory of Goltz. Suppose 

 the three canals to be full of fluid, a rotation of the head in one direc- 

 tion will cause currents of fluid in the canals in the opposite direction, 

 and the amount of flow in each canal will depend on the plane in which 

 the head is turned and on the rate of rotation. As the auditory hairs 

 exist in the crista of the ampullae, we can conceive that they will be 

 affected by the currents, and produce a sensation of rotation of the 

 head in the plane of the canal in which the flow takes place, and in a 

 direction opposite to it. If the rotation of the head continue, the flow 

 of the endolymph backwards will be arrested by friction, and if the 

 head then suddenly stop, the endolymph will continue to flow in the 

 direction in which the head was previously rotated, and a sensa- 

 tion will be felt of a rotation of the head and body in a direction 

 opposite to that of the first rotation. Crum Brown 1 sums up the 

 matter thus 



" We have two ways in which a relative motion can occur between the 

 endolymph and the walls of the cavity containing it (1) When the head 

 begins to move, here the walls leave the fluid behind; (2) when the head 

 stops, here the fluid flows on. In both cases the sensation of rotation is 

 felt. In the first this sensation corresponds to a real rotation, in the second 

 it does not; but in both it corresponds to a real acceleration (positive or 

 negative) of rotation, using the word acceleration in its technical kinematical 

 sense." 



Mach attributes the effects to variations of pressure in the canals 

 rather than to an actual flow of fluids, while Crum Brown differs from 

 both Mach and Breuer on the following points : (1) In attributing 

 movement or variations of pressure, not merely to the endolymph, but 

 also to the walls of the membranous canals and to the surrounding 

 perilymph ; and (2) in regarding the two labyrinths as one organ, all 

 the six canals being required to form a true conception of the rotatory 

 motion of the head. As already pointed out, the six canals are sensibly 

 parallel, two and two. The two horizontal canals are on the same 

 plane, while the superior canal on one side is nearly parallel with 



1 Nature, London, 1878, vol. xviii. p. 633 ; review of Cyon's work, loc. cit. 

 VOL. II. 76 



