A general mechanism of airborne hearing in recent and early non-tympanate tetrapods.

Tetrapod tympanic hearing probably emerged in the Triassic with independent origins of middle ear structures in each of the major groups, more than 120 Myr after the origin of tetrapods. During this period, any auditory sensitivity must have been based on non-tympanic mechanisms. We focused on the simplest model for non-tympanic hearing: that sound translates the head, and that this vibration is transduced by the inner ear. This is the mode of human low-frequency bone conduction sensitivity and is also the mode of underwater auditory stimulation for most fishes. The efficiency of translation of an object by sound depends on its density and ka, the product of the acoustic wavenumber (k) and the radius (a) of the head. Analytic and simple finite-element models of translation show that head vibration velocities largely are determined by ka and density (for objects of the same shape and composition), and are almost constant (between 4 and 5 μm s-1 Pa-1; neglecting friction) for objects with ka<1. We compared sensitivity to sound and to head vibrations in animals lacking tympanic middle ears (snakes, salamanders, earless frogs and lungfish) and showed that the low-frequency airborne sound sensitivity in these species is largely consistent with a translation mechanism. Stimulation of the inner ear by sound translation is likely by an inertial system like the otolithic/otoconial ears of fish and early tetrapods, or by fluid inertia in the inner ear generating hydrodynamic waves that stimulate the hair cells, providing a simple mode of sound reception in earless animals.