4 8 
bulletin of the bureau of fisheries. 
with the surface of the skin normally sensitive, and (3) those with the ears intact but 
with the surface of the skin rendered insensitive. 
When a normal dogfish is placed in a large wooden aquarium, it at first swims about 
in a disturbed and irregular manner. After half an hour or so it becomes so far accus- 
tomed to its new quarters as to move about with apparent complacency. If, while the 
dogfish is swimming through the water and is not in contact with the sides or bottom 
of the aquarium, a fairly vigorous blow is struck with a mallet on the wooden wall of the 
aquarium, the dogfish will almost invariably respond with a sudden jump forward. 
This can be repeated many times provided that a few minutes intervene between the 
trials. If the blow is not very vigorous the response may be only a slight waving of 
the fins, best seen on the posterior edges of the pectorals. 
To get some measure of this response, I suspended on a stout cord from the ceiling 
of the room in which the experiments were conducted a large spherical iron weight so 
that it formed the bob of a pendulum which, when at rest, just touched the middle of one 
of the wooden sides of the aquarium. By drawing this iron bob away from its position 
of rest and letting it swing squarely against the wooden side of the aquarium, a noise 
was produced that would be louder or fainter depending upon the distance between the 
bob and the aquarium side when the bob was liberated. The momentum with which 
the blow given by the bob was struck was taken as a rough measure of the noise pro- 
duced. As the whole apparatus was a simple pendulum, it was comparatively easy to 
make the necessary calculations for a scale to be placed next the cord of the pendulum 
to indicate the positions from which the bob must be liberated in order to generate 
given momenta. The length of the pendulum was 260 centimeters and the weight of 
its bob was 3,800 grams. The momenta used in the experiments and expressed in 
centimeter-gram-second units were (1) 83,600, (2) 125,400, (3) 167,200, (4) 250,800, and 
(5) 334,400, or, calling momentum (1) unity, they could be more conveniently designated 
as 1, 1.5, 2, 3, and 4. 
Normal dogfishes when swimming freely in the water of the aquarium occasionally 
responded by pectoral fin movements to the sound generated by the bob of the pendulum 
striking the wall of the aquarium with a momentum of 1, and invariably responded when 
the momentum was 1.5. The range from 1 to 1.5 was therefore taken as the range of 
minimum stimulus for a normal fish. 
Six dogfishes, which had previously been tested to ascertain that they were normally 
responsive, were now subjected to the operation for cutting the eighth nerve, and after 
recovery they were again tried for their responsiveness. None reacted to the sounds 
produced when the ball struck the side of the aquarium with a momentum of less than 
3, and they responded invariably only when the momentum was 4. 
At first sight this considerable reduction in the sensitiveness of the fish might be 
taken to be a final answer to the question of the significance of the ear as a receptive 
organ for sound, but it is possible that its real explanation lies in the reduced physio- 
logical state of the animal as a result of so severe an operation as that of cutting the 
eighth nerve. I therefore repeated these tests on several dogfish in which for other 
purposes the optic nerves had recently been cut, and I found that notwithstanding the 
