- University of Toronto
- Carleton University
Summary
The auditory sensitivity characteristics of moths sampled in southeastern Ontario and south-central British Columbia were compared in four separate bandwidths from 5 kHz to 110 kHz. British Columbian moths exhibit a significantly greater sensitivity to frequencies from 30 kHz to 75 kHz (midrange) then do their Ontario counterparts. Measurements taken of echolocating bats at both these sites reveal a greater diversity of acoustic frequencies emitted by British Columbia bats and we suggest sympatric moths have responded to this increased predation pressure by selectively maximizing their frequency sensitivity pertinent to the bats they are exposed to. The possibility exists that the low frequency British Columbia bat, Euderma maculatum, may be foraging on moths as a result of its echolocation signals for which sympatric moths show no particular sensitivity.
Methodology
We analyzed noctuoid moths (particularly from the families Noctuidae and Arctiidae) captured from ultraviolet lamps at two study sites, one in southeastern Ontario (44'34' N, 79'15' W) during 1978 and 1979 at the Queen's University Biology Station and another located in southcentral British Columbia (42'12' N, 1 19'33' W) in the Okanagan Valley in 198 1. Data on the bats of these two areas are based on 15 summers of sampling at the Queen's Station, and on 3 summers of work in he Okanagan Valley (more details in Fenton et al. 1980; Woodsworth 198 1 ; Woodsworth et al. 198 1). Audiograms of the moths were obtained using the techniques described in Fenton and Fullard (1979), to expose and record from the moths' thoracic auditory nerve (IIINl b, Roeder 1967) as the ear was acoustically stimulated at 10 cm. Neurophysiological auditory sensitivity curves were determined using frequency increments of 5 kHz from 5 to 1 10 kHz, and these were then analyzed as a function of the area of the curve beneath the arbitrary value of 90dB (re 20 FP~) (Masterton et al. 1969; Fullard 1982). Total auditory sensitivity in these species was computed as the entire area in Pascals.kilohertz of the audiogram beneath 90 dB (0.63 Pa). This value was divided into four bandwidths to determine sensitivities at low (5 to 25 kHz), low to middle (30 to 50 kHz), middle to high (55 to 75 kHz), and high (80 to IlOkHz) frequencies. The data distributions from both sites were determined to be significantly nonnormal using a D'Agostino (1971) test so comparisons were performed using a Mann- Whitney normal approximation 2-test (Zar 1974). To obtain an indication of the relative abundances and activity of bats in the two study areas, we used three techniques. Captures of individuals in mist nets or Tuttle traps (Tuttle 1974) were combined with two patterns of monitoring of echolocation calls. We monitored bat activity at different frequencies using QMC mini bat detectors (QMC Instruments Ltd., 229 Mile End Road, London, England El 4AA) to obtain an indication of bat traffic at different frequencies. We also monitored the activity of specific species of bats using a period meter (zero-crossing analyzer) display of their orientation calls (Fenton and Bell 1981).