398 



MECHANICAL AND ACOUSTICAL SENSES 



attraction (Hobson 1963, Nelson and Gruber 1963, Wisby et al. 1964, 

 personal observation). Banner (1972) demonstrated in young lemon sharks 

 that the rapidity of pulsing is directly related to the relative attractiveness of 

 various biological sounds produced by their prey. This same correlation 

 existed in silky sharks for synthesized sounds so long as their spectral 

 content was appropriate (Myrberg et al. 1972). Although extremely low 

 frequency signals cannot be pulsed rapidly and still maintain their spectral 

 integrity, even a constant pulse rate of 1/s appears to be slightly attractive 

 (Myrberg et al. 1972) (Table 3). 



Since natural sounds often show irregularity in their pulse structures 

 (e.g., erratic movements during feeding, flight, and stress), attention to this 

 type of signal should be highly adaptive to any predator. Accordingly, it is 

 not surprising that the most attractive sounds have irregular pulses (Myrberg 

 et al. 1972, Nelson and Johnson 1972). Only two studies have compared the 

 effects of natural and synthesized sounds. Nelson and Johnson (1970) 

 believed that the sound produced by a "stampeded" school of bonefish 

 (Albula uulpes) elicited slightly stronger response from their subjects than 

 similarly structured synthesized sounds. In contrast, Banner (1972) showed 

 that a synthesized sound possessing characteristics most conducive to attrac- 

 tion (i.e., rapid pulsing, low frequency, sufficient loudness) was generally 

 as effective as natural sounds. 



Unfortunately, we know little about the degree to which sharks recognize 

 differences between extremely brief intervals within given sounds. Yet, there 

 is no a priori reason why their ability should be less than the recognition 

 afforded remarkably small intervals (less than 10 ms) by various teleosts 



Table 3. Differential attraction of silky sharks to various instrumental 

 signals, each having a different pulse character; all signals had the frequency 

 spectrum of 25-50 Hz (Tongue of the Ocean, Bahamas).* 



*From Myrberg et al. 1972. 



'Distribution of sightings among the signals is significantly different from random distri- 

 bution: -0.05 (Kolmogrov-Smirnoff, one sample test). 



