352 DYNAMIC FUNCTIONS OF LABYRINTH 



of the head downward) excites them. So also the crista of the right 

 horizontal canal is on its right or outer side and the stimulus for it is 

 rotation to the right. Of course a similar relation exists for the left 

 ampulla. Examination of the extensive series of drawings by Ret- 

 zius^ shows that the dogfish is not a special case but that the arrange- 

 ment is general. 



A second fact which is significant is that the mouths of the ampullae 

 are continuous with the utriculus, an elongated, thin walled sac, 

 stretched across the cavity of the vestibule and occupying only a por- 

 tion, in the dogfish a not relatively large portion, of the vestibular 

 space. Furthermore the utriculus is so attached by means of the sinus 

 superior and other structures that a movement of the liquid contents of 

 the vestibule might readily press it upwards towards the dorsal side 

 of the cavity, but could have little effect to move it downwards. 

 The relations as far as the ampulla of the anterior vertical canal is 

 concerned are shown diagrammatically in Fig. 1. Rotation of the 

 head downwards, that is, in the direction of the outer arrow, would 

 tend by inertia to produce the same effect as if, with the head station- 

 ary, the perilymph was rotated in the opposite direction, as indicated 

 by the small arrow within. This would put pressure and tension on 

 the under side of the anterior end of the utriculus ; this tension would 

 be communicated to the ampulla and especially to its lower side which 

 bears the crista. 



In order to convince myself of the correctness or incorrectness of 

 the above reasoning, I constructed a model by carving cavities and 

 channels corresponding to the relations shown in Fig. 1. In these I 

 placed a thin rubber model of the two canals shown in the figure. 

 The canals and utriculus as well as the perilymphatic space were 

 filled with mercury. On rotating the apparatus it could be seen that 

 movement in one direction gave a very perceptible pull on the am- 

 pulla ; movement in the opposite direction was almost without effect. 

 It is possible that the rotation which puts the ampulla under mechani- 

 cal strain would also tend to produce an increased liquid pressure 

 within it, but this I could not determine in my present model. 



* Retzius, G., Das Gehororgan der Wirbelthiere. I. Das Gehororgan der Fische 

 und Amphibien, Stockholm, 1882. 



