This review paper briefly considers how stimulus, noise masking and recording parameters affect the frequency and place specificity of auditory brainstem responses (ABRs) to air- and bone-conducted stimuli. Issues concerning the use of clicks for ABR threshold estimates will first be presented, followed by results for tone-evoked ABR thresholds and how well they predict the pure-tone behavioral audiogram. Noise-masking options (e.g. high-pass noise, notched noise and white noise) to improve the frequency specificity of tone-evoked ABRs, which are now available on clinical ABR units, will also be discussed. The goal of this article is to demonstrate that ABRs to tonal stimuli can be successfully recorded in most clinical environments and can provide reasonably accurate estimates of 500- to 4000-Hz pure-tone behavioral thresholds in infants, children and adults. Specific parameters and protocols for obtaining frequency-specific ABR threshold responses are provided.