In representative elasmobranchs, osteichthyans, amphibians, reptiles and birds, average evoked potentials in response to acoustic clicks and tone bursts were recorded intracranially, but outside the brain, or extracranially. Controls against artifacts and tests after transections show that these potentials conform to criteria for auditory brain stem responses (ABRs). Brief waves in a 10-15 msec sequence originate successively in the eighth nerve, medulla and midbrain; there is little contribution to the latter waves from the lower levels. This response pattern appears to be consistent within each species and is similar to that extensively studied in mammals. Some of its features are remarkably alike in all the vertebrate classes tested, implying a generality in the existence of a subset of auditory neurons at several brain levels that are highly synchronous in activity, even after several synapses, and geometrically oriented to add their macroscopic, open, dipole fields. The intensity, repetition rate and the power spectrum of the click stimuli have little effect on the ABR pattern, except when the peak energy is in the low frequency range. In the range below ca. 700 Hz frequency content has a considerable effect; lower frequencies broaden certain waves. Cooling has marked and differential effects on component processes. Reversing click phase, e.g. from initial compression to initial rarefaction, can show no effect or any of several effects, depending on the species. Tone bursts evoke onset ABRs and in some cases after a transitional period a sustained frequency following response. The ABR resembles a click evoked potential even when stimulus rise time is slow. Background tones of particular frequency are most efficient in masking click evoked ABRs; white noise is less efficient. The ABR should be useful in neuroethology since it can be studied without invading the brain. It can tell that the brain is sensitive to a sound. In an immobilized animal it can be recorded in a single sweep, or it can be averaged from an awake tethered animal. It shows good sensitivity and at least some correspondence with behavioral measures of hearing.