166 ANNALS NEW YORK ACADEMY OF SCIENCES 



lifting a part of the abdominal muscles, the drumming over that part im- 

 mediately ceases. That the most ventral parts of the abdomen are active 

 in drumming is evident from the vibrations of this part of the body of a 

 squeteague. The whole mid-ventral area, from pectoral fins to anus, 

 pulsates in a strong rhythmical manner, which corresponds to the con- 

 traction of the m. sonificus as can be readily seen from the appended kymo- 

 graph tracings. 



Experiment X. — An incision one inch long was made about half way between 

 the pectoral fins and the anus and at right angles to the long axis of the body. Great 

 care was taken in order not to injure the drumming muscles. Between the ventral 

 muscles and air-bladder was inserted a piece of sheet cork about two and one-half 

 inches long and two inches wide. This stretched severely the mid-ventral part of 

 the abdominal muscle and held it rigid, so that it could not be pulled in when the 

 sonificus contracted. No noise was produced, yet the muscle apparently contracted 

 in a perfectly normal manner. This would again show that the dnunming is not a 

 muscular tone intensified by the air-bladder. 



The drumming is undoubtedly a sexual character, for in the squeteague 

 the male only makes this noise. The female not having developed any 

 drumming muscles is not able to produce this sound. In some other 

 scisenoids, as the croaker, both male and female produce the drumming, 

 but the former is said to produce a much more intense noise than the female. 

 I have often observed that the drumming muscles in the male croaker are 

 much thicker and heavier than in the female. 



The conclusion is that by each contraction of the m. sonificus a sudden 

 blow is dealt which throws into vibration the abdominal walls and organs. 

 The physics of this phenomenon is very complex, as undoubtedly all of the 

 abdominal parts play a r61e. But the organ that chiefly participates in 

 the vibration is the swim-bladder with its walls made tense by the pressure 

 of the contained gas. It is well known that in man the chest walls and 

 abdominal walls can be set into irregular vibration by being percussed and 

 that there is here a resonance effect produced by a resonance cavity or 

 semifluid material which is selectively set in resonance vibration. The 

 gas pressure in the air-bladder as well as the character of the muscular 

 contractions which will be immediately described indicate the same con- 

 clusion. 



In all of the above experiments the pitch of the drumming sound was 

 not determined with scientific accuracy. Undoubtedly if the tone could 

 have been determined by physical apparatus the pitch, which to the ear 

 was apparently the same, would have been found to be different in the 

 various experiments. 



