Otoacoustic emissions from the cochlea of the 'constant frequency' bats, Pteronotus parnellii and Rhinolophus rouxi.

During stimulation with continuous pure tones, the cochlea of each individual of the mustached bat, Pteronotus parnellii, produces a strong evoked stimulus-frequency otoacoustic emission (SFOAE) at about 62 kHz. The SFOAEs were on average 480 Hz above the dominant constant frequency component of the echolocation call (resting frequency, RF). In two out of nine individuals of Pteronotus the SFOAEs changed into spontaneous otoacoustic emissions of 25-40 dB SPL. In the rufuous horseshoe bat, Rhinolophus rouxi spontaneous emissions were not detected and only in two out of seven animals were there weak SFOAEs about 300 Hz above the RF of 78 kHz. This difference may be due to a stronger damping of underlying resonant processes in Rhinolophus (Henson et al., 1985a). Acoustic distortion products behaved quite similar in both species. The first lower sideband distortion 2f1-f2 was measurable over a wide frequency range between 10 and 100 kHz. The optimum frequency separation delta f of the two primary tones to evoke maximum 2f1-f2 distortion was 0.8 to 5.8 kHz in Pteronotus and 1 to 7 kHz in Rhinolophus for frequencies outside the range of the constant frequency components of the call. This corresponds to ratios f2/f1 of about 1.03 to 1.2. At the frequency of the SFOAE in Pteronotus (480 Hz above the RF) and about 300 Hz above the RF in Rhinolophus the optimum delta f decreased sharply to values of 31-63 Hz in Pteronotus (ratio f2/f1 of 1.0005-1.001), and to 39-590 Hz in Rhinolophus (ratio f2/f1 of 1.0005-1.007). In Pteronotus a second minimum of delta f was found at about 90 kHz (values of 180-620 Hz, ratios f2/f1 of 1.002-1.007). In both bat species, the respective minima of delta f are located at or close to frequencies where neuronal tuning sharpness is exceptionally high. This indicates a mechanical origin of enhanced tuning. After adjusting the frequency of f2 to match the optimum delta fs, 2f1-f2 threshold curves were obtained. The distortion product threshold approximately parallels neuronal data and is in both species characterized by a pronounced insensitivity at the RF followed by a steep threshold minimum at frequencies 0.3-3 kHz above the RF. These features may be involved in reducing the cochlear response to the call such that the bats are able to focus on the Doppler-shifted echos which are slightly higher in frequency and thus within the range of the threshold minimum.