Although much has been reported on air-conduction auditory brainstem evokedresponse(ACABR), little ettention gas been directed to bone-conduction ABR(BC ABR) owing to limitations in clinical settings. Recently the use of BC ABR has been emphasized in evaluating the hearing level in children who cannot be assessed by pure tone audiometry. It may be employed in evaluating sensorineural hearing loss in children with middle ear disease. The authors aimed to evaluate the reliability of BC ABR by performing AC & BC ABR's on 10 normal-hearing adults. Hearing thresholds and latencies were recorded in open and plugged ears, to mimic normal hearing and conductive hearing loss. In our study, the threshold, peak latency, and same-latency intensity in normal-hearing ears (open ears) did not show significant difference between AC & BC ABR's, and those of the BC ABR in artificially induced conductive hearing loss(pulgged ears) showed similar results. The results of AC ABR in plugged ears showed delayed latency and elevated threshold as compared with those of the open ears, and the threshold shift was similar to that of pure tone averages.