The evolution of hearing in terrestrial animals has resulted in remarkable adaptations enabling exquisitely sensitive sound detection by the ear and sophisticated sound analysis by the brain. In this review, we examine several such characteristics, using examples from insects and vertebrates. We focus on two strong and interdependent forces that have been shaping the auditory systems across taxa: the physical environment of auditory transducers on the small, subcellular scale, and the sensory-ecological environment within which hearing happens, on a larger, evolutionary scale. We briefly discuss acoustical feature selectivity and invariance in the central auditory system, highlighting a major difference between insects and vertebrates as well as a major similarity. Through such comparisons within a sensory ecological framework, we aim to emphasize general principles underlying acute sensitivity to airborne sounds.Introduction Auditory physiology offers a distinctive perspective on the interaction between a sensory system and its environment. On the one hand, auditory systems in vertebrates and insects with sensitive hearing are capable of remarkable performances on multiple levels, both within the sensory periphery where the minute energies associated with sound are converted into electrical signals, as well as within higher-order brain areas where complex natural stimuli such as human speech are processed. For example, displacements caused by acoustical stimuli in the inner ear at the threshold of hearing are sub-nanometer and comparable to the distance between atoms in molecules [1]. Furthermore, thermal fluctuations in the ear's mechanotransduction apparatus not only are significant, but also can be larger than the faintest audible signals, making signal detection a challenging task [2]. Ascending the sensory hierarchy, one encounters other marvels of evolution, such as the ability of individual neurons to encode -using millisecond-long action potentials -inter-aural time differences of only about ten microseconds, and to use this information to localize the source of the sound [3,4]. No engineered system has yet been designed that could understand distorted speech in a noisy and reverberating environment with multiple speakers. That our auditory system achieves this feat is testament to its remarkable performance.On the other hand, an engineer could argue that the auditory system's performance is objectively poor, even in animals with sensitive hearing: at the very first step, during the mechanoelectrical transduction in the inner ear, external sounds are distorted, or even completely suppressed, while new tones are generated by the ear itself [5]. Forward and backward masking, illusory percepts of nonexistent tones (such as the Zwicker illusion [6]), perceptual merging of separate auditory streams, the precedence effect suppressing the perception of echoes that has been demonstrated in insects and vertebrates [7,8] (and which some blind people can unsuppress), auditory hallucinations, and a frustrating inab...