Bat predation on eared moths: a test of the allotonic frequency hypothesis

Many species of moths (Lepidoptera) possess ears which enable them to hear the echolocation calls of bats and avoid predation. The range of best hearing of eared moths is to frequencies usually between 20 and 50 kHz; therefore, bats with calls above or below these frequencies are predicted to capture eared moths more frequently than other bats. This prediction is the basis of the allotonic frequency hypothesis. We conducted a field test of the hypothesis using three species of flutter-detecting bats which capture flying insects. The species were diadem leafnosed bat Hipposideros diadema (call frequency 55-58 kHz), eastern horseshoe bat Rhinolophus megaphyllus (67-71 kHz) and dusky leafnosed bat H. ater (160-164 kHz). fared moths represented only 7.2% of individuals among prey remains of H. diadema. Moths dominated the diet of both R. megaphyllus and H. ater, being present in almost all faeces and representing the vast majority of prey remains. However, R. megaphyllus captured predominantly non-eared moths (Anthelidae, Lasiocampidae, Hepialidae) or Gelechioid moths, a superfamily which has not been tested for hearing. By contrast, H. ater fed mostly on eared moths, predominantly members of the family Noctuidae. fared moths made up 91.7% and 82.5% of all moths captured by H. ater at two sites. The data show that the incorporation of eared moths in the diet was much higher in the bat species, H. ater, with a call frequency well above the range of best hearing of moths. This result supports the allotonic frequency hypothesis. Our study indicates that flutter-detecting bats could have imposed selective pressures on moths during the evolution of moth hearing.