Insect disturbance stridulation: Characterization of airborne and vibrational components of the sound

Some insects stridulate when attacked by a predator. This behavior has been interpreted as a defensive response, the sound being a warning to predators of the insect's noxiousness. Since to humans many such disturbance sounds are audibly similar, it is possible that they may in fact be mutually mimetic. This idea was investigated through analysis of the temporal and spectral characteristics of the disturbance sounds of a variety of insects that stridulate by a file- and -scraper device. Properties of both the airborne sound and the underlying cuticular vibration (detected by a special vibration measuring instrument) were examined, and four characteristic features found:

    1. The temporal pattern is simple. Bursts of toothstrike impulses are about 80 ms long, and are separated by pauses about 90 ms long. Bursts occur at a rate of about 5 to 10/s.
    
    2. The temporal pattern is irregular. For toothstrike interval, burst duration, pause duration and interburst interval, the standard deviation is usually >30% of the mean. Much of the irregularity is presumably caused by the insect struggling at the same time it stridulates. Some insects show less variability, and these appear to lack tight coupling between stridulatory movements and struggling movements, so struggling does not interfere with stridulation.
    
    3. The airborne sound pressure waveform is impulsive. The frequency coverage of the sounds is quite broad with an average 10-dB bandwidth of about 40 kHz centered at 25 kHz. The sounds are not intense, ranging from about 10 to 60 dB (re 20×10−6 Pa) at 10 cm.
    
    4. The cuticular vibration waveform is sharply peaked and contains maximum energy at a frequency determined by the tooth-strike rate, usually about 1 kHz. The average decrease in power above this frequency is about 12 dB/octave. The maximum peak-to-peak amplitude of cuticular motion is about 1 to 10 μm.
    

These common characteristics may lead predators to treat insects producing disturbance sounds similarly, although this possibility should be tested empirically. If acoustic mimicry exists, the communicatory interchange between predator and prey may be subtler than is commonly appreciated.

Supported in part by an NSF predoctoral fellowship and Bache Fund stipend from the National Academy of Sciences (Masters), and NIH grant AI-02908 and NSF grant PCM77-25807 (T. Eisner). I thank the insect identification service of the U.S. Department of Agriculture and E.R. Hoebeke for identification of insects. I am grateful to the Director and staff of the Archbold Biological Station, Lake Placid, Florida, for their hospitality while I was there. To my chairman, Dr. Thomas Eisner, I wish to acknowledge my great debt for his original proposal that I study insect disturbance sounds and the suggestion that the sounds might be mutually mimetic, and for his unfailing support throughout my graduate career. I also thank Drs. Eisner, R.R. Capranica, D.J. Aneshansley, W.L. Brown and J. Camhi for helpful criticism of the manuscript, Drs. Capranica and G. Hausfater for loan of equipment, and Dr. A. George for discussions of acoustics. Finally, for invaluable help with all phases of this research, I thank my wife, Anne Moffat.