3. S. MAXWELL 161 



forceps and gently moved to the right or left, forward or backward 

 at will. I quote again from my notes: 



"July 23, 1920. Large shovel-nosed ray (Rhinobalus). 



Removed ampuUce from left ear. 



Exposed small otolith (of recessus utriculi) and placed on it the pellet of cotton. 



Movement of pellet to left caused depression of left eye and elevation of right 

 eye. 



Movement of pellet to right caused depression of right eye and elevation of 

 left eye. 



Movement of pellet forward caused both eyes to roll forward on their axes 

 (anterior pole of each eye depressed and posterior pole elevated). 



Movement of pellet backward caused both eyes to roll backward on their 

 axes. 



When pellet was moved to one side eyes moved in same sense. 



When pellet was held to any side, the eye position was retained. 



Removed the three ampullae of the right ear. 



Repeated the experiment on the right ear with exactly the same results. 



Repeated a score or more of times with no noticeable diminution of the response. 



Holding the pellet to any side held the eyes in the corresponding position." 



I have repeated these experiments on dogfish, leopard sharks, and 

 rays. The experiments on the ray {Rhinohatus) were particularly 

 striking. This fish is broad and flat and usually remains on or near 

 the bottom of the water. It is not apparently used to much turning 

 over or tilting of the head up or down. Taken out of the water, or 

 rotated in the water it does not show any of the compensatory move- 

 ments in so marked a degree as does the dogfish. When, however, 

 the stimulation was applied to the recessus as described above, the 

 eye movements were extraordinarily vigorous, much more so than 

 in response to rotation of the body of the uninjured animal. The 

 eyes rolled right or left, forward or backward as if on actual mechani- 

 cal axes manipulated by cords. 



It will be seen from the above that pressure on the right side of the 

 otolith of either ear produces the same eye movement which results 

 as the compensatory motion to rotation of the body to the left around 

 the longitudinal body axis; and that pressure on the anterior side of 

 the otolith gives the same effect as tilting the head upward. In each 

 case the response is precisely opposite to that which would be expected 

 if the stimulation were produced by the pressure due to the weight of 

 the otolith; for when the body is tilted to the right the weight of the 



