Potentials evoked by clicks and tone pips were recorded by fine wires inserted extracranially in four West Indian manatees (Trichechus manatus) in air. Sounds were delivered via padded ear phones.

Averaging a few thousand trials at 20/s reveals early peaks at N5.4 (‘vertex’ negativity to a frontal reference, at 5.4 ms), P7.6, N8.8, P9.5 — probably equivalent to waves IV and VII of the typical mammalian auditory brainstem response (ABR). Averaging 100 trials at <4/s suffices to reveal a complex sequence of later peaks including N25, P80, N150 and P190; consistent smaller peaks are visible when several hundred trials are averaged.

Using tone pips with a rise and fall time of 2–5 ms the carrier frequency becomes important. Evoked potential wave forms are not the same at different frequencies, bringing out the fact that frequency is not a scalar that can be compensated for by intensity. Therefore the method was not used to obtain audiograms; however the largest EPs occur in the range of 1–1.5 kHz. EPs are found up to 35 kHz; almost no evoked potential is discernible at 40 kHz but the undistorted intensity available was limited. This is in reasonable agreement with the theoretical expectation for the upper limit of behavioral hearing from Heffner and Masterton based on head size and aquatic medium.

Among several ear phone placements, that over the external auditory meatus was the most effective, but only slightly so. The external canal is presumably fluid or tissue filled and sound enters over a large area.

Comparing data for two species on the most effective range of frequencies and the power spectra of their vocalizations, T. manatus is lower than T. inunguis in both respects.

The results show the utility and limitations of the method of recording extracranial evoked potentials to sounds, especially for large and valuable animals under makeshift conditions.