No. 448.] 



CARASSIUS AURATUS L. 



281 



operation of cutting the eighth nerve, severe as it is, is not 

 sufficiently so to account for. even a small part of the reduc- 

 tion in the number of responses which follows the elimination of 

 both ears as sense organs. 



In another instance, a fish which before any operation 

 responded vigorously to sound by movements of the pectoral 

 fins, was prepared by cutting two holes in the top of the skull, 

 through which the eighth nerves could conveniently be reached, 

 and the following parts were severed : the spinal cord, the lateral 

 line branches of the tenth nerves, and the superficial portions 

 of the fifth and seventh nerves of both sides. After recovery, 

 the fish gave ten vigorous pectoral fin responses to as many 

 trials with the sounding apparatus. The eighth nerv^es were 

 then cut, and in twelve tests only one response, and that of 

 a doubtful character, was observed. It, therefore, seems incred- 

 ible that nervous shock can account for the almost complete loss 

 of response to sound, after cutting the eighth nerves, and I am 

 firmly convinced from the foregoing experiments, that the ear in 

 the goldfish is an organ of hearing, and that it is the loss of 

 this which is accountable for the difference of reaction between 

 fishes in which the ears were intact, and those in which the 

 eighth nerves had been cut. 



VI. Discussion of Results. 



The results of my experiments differ so essentially from 

 those obtained by Kreidl ('95) that a further discussion of 

 these differences is necessary. So far as reactions to sounds 

 were concerned, Kreidl was unable to distinguish between 

 normal goldfishes and individuals from which the semicircular 

 canals with the attached parts of the ears had been removed. 

 In my experiments, however, while normal fishes responded to 

 sound in about 78 of the trials, those in which the eighth 

 ner\-es had been cut, scarcely responded at all. The difference 

 between these two sets of results was so great that I determined 

 to repeat, with as much precision of detail as possible, Kreidl's 

 experiments. 



I tested several goldfishes, and, having found that they 



