TOWER, PRODUCTION OF SOUND IN CERTAIN FISHES 163 



2. Clapping together of pharyngeal teeth (Gunther). 



3. Vibrations of air in air-bladder and of the walls of the latter when 



set in motion by certain muscles (Sorensen). 



4. Forcing of air from air-bladder into one of lateral horns (Holbrook, 



Jordan and Evermann). 



Experiments to Determine Cause of Sound. 



That the explanation given by Gunther is wrong can be very easily 

 seen from the following experiments, in all of v/hich the animals were kept 

 alive by artificial respiration, i. e. by irrigating the gills with a stream of 

 fresh water. 



Experiment I. The air was drawn from the swim-bladder of a squeteague by- 

 means of a trochar and the drumming immediately ceased. 



a. The stomach was then filled with water. The drumming returned but not 

 as loud as normal. 



Expenment. II. — An incision one inch long was made in the mid-ventral line. 

 Through this a portion of the air-bladder was pvilled out but the drumming con- 

 tinued. 



a. The bladder was now amputated and the drumming ceased. 



b. A collapsed rubber balloon was then inserted into the abdominal cavity, 

 and, as soon as it was inflated, the drumming returned with apparently normal 

 intensity and pitch. 



c. The rubber balloon was filled with salt water instead of air. The drum- 

 ming continued until the water was allowed to escape; then it ceased. The tone 

 is low and apparently changed but little under the different conditions. 



Experiment III. — The air-bladder of a male squeteague was removed. The 

 drumming ceased. The air-bladder from a female squeteague, which can produce 

 no noise, was inserted into the abdominal cavity of the male, and the drumming 

 immediately returned. 



These three experiments show conclusively that the "clapping together 

 of the pharyngeal teeth" has nothing to do with the production of the 

 drumming noise in the squeteague. It is also shown in experiment III 

 that there is no difference between the function of the bladder in the male, 

 which drums, and that of the female, which does not drum, as far as the 

 noise-production is concerned. 



Experiment IV. — The entire viscera (intestines, spleen, liver, reproductive or- 

 gans and air-bladder) were removed from a male squeteague. The drumming 

 stopped, and the sonificus contracted as usual, but there was no noise. A rubber 

 balloon filled with air was then inserted into the abdominal cavity. The drumming 

 again returned, but was not of normal character. 



a. The balloon was filled with water and the drumming continued, but was 

 weaker and of apparently different pitch. 



