Impulse noises have long been a matter of interest and concern because of its effects on the cochlea. Of all impulse noise sources, M-16, the standard army rifle, may most often be involved in hearing loss because of its widespread use among army personnel and sporting members. Despite the fact little is known about precise auditory hazard from the exposure to M-16 impulse noise. The past decades there have been many studies on inner ear histopathology following the impulse noise exposure(Schuknecht, 1953 ; Engström & Ades, 1960 ; Beagly, 1965 ; Stockwell et al, 1969 ; Spoendlin, 1971 ; Lim & Dunn, 1979 ; Hunter-Duvar et al, 1982 ; Engström & Borg, 1983). Three damage risk criteria have been explored for impulse noises that are currently used throughout the world(CHABA, 1968 ; Pfander, 1975 ; Smoorenberg, 1977). There are, however, several questions which still unresolved : 1) Mechanisms of inner ear damage, 2) discrepancies between histopathological changes and audiological findings, 3) more accurate damage risk criteria, 4) individual susceptibilities, etc. The purpose of this study is to learn about histopathological changes in the inner ear at different peak levels(140dB, 145dB, 150dB, 155dB SPL) of the M-16 impulse noise, with the hope that precise damage risk criteria of the impulse noise can be developed. Twenty six healthy adult cats were used for the study. Twenty one cats were divided into 4 groups and each group was exposed to 60 impulses at 140, 145, 150, 155dB SPL(peak level) and five cats were used as control. For the noise, a M-16 was mounted on the ground and fired 60 rounds with 2-5 second intervals at the peak levels of 140, 145, 150, 155dB SPL. All cats were sacrificed 30 days after the noise exposure. Both temporal bones were removed and the cochleas were processed for scanning electron microscopic examinations as follows : 1) The temporal bones were trimmed and holes were made in the oval, round windows and apical turn of cochlea to instill the fixative. 2) The cochleas were fixed in 3% glutaraldehyde for 24 hrs and refixed in 1% osmium tetroxide for 2 hrs. 3) The cochleas were dissected in 70% ethanol and divided into apical and remained(middle & basal) turns. 4) The cochleas were further processed for the SEM study and the entire length of each cochlea was serially photographed at magnifications of 20-5,000X with Hitachi HCP-2 SEM. For the analysis of the result the raw data from the serial microphotographs of each cochlea were fed into the computer and hair cell losses were counted in every 2.5% distance form the apex to the base. And the hair cell losses were counted to evaluate the overall effect on the organ of Corti, and the cochleagrams were made in each group to identify the lesion-site. The wipeout areas(regions with all hair cells loss) and areas of over 67% loss of outer hair cells were observed regarding as severe lesion-site. The obtained data were analyzed statistically to assess the significance of the hair cell losses between groups and between distance intervals(every 2.5% distance from the apex to the base). Finally, the histological damage risk criteria was caculated based on the SPSS regression curve of the OHC loss and the following conclusions were made. 1) There has been noted a great individual variability in the sensitivity of the cochlea to M-16 impulse noise. 2) As the peak level of the M-16 impulse noise increased, the extent of the hair cell pathology became wider and the severe lesion-site was expanded to the basal end. 3) As the peak level of the M-16 impulse noise increased, the relative ratio of IHC loss to the control became higher than that of OHC loss to the control. 4) In general, the lesion-sites of IHCs and OHCs were corresponded but the extent of IHC pathology was narrower than that of OHC pathology. 5) The lesion-sites of OHCs have been observed in the corresponding areas of impulse noise frequencies and the areas of 1-2 octave higher frequencies. In addition the lesions of higher frequency-regions were more severe and wider than those of lower frequency regions. 6) As the peak level of M-16 impulse noise increased, the OHC-loss rates at the 1st & 2nd rows became higher and in contrast, the OHC-loss rates of 3rd row became lower. 7) The mechanism of cochlear damage resulted from the impulse of peak intensities greater than 150dB SPL was different from that which occurred from intensities below 145dB. 8) Histological damage risk criteria calculated by SPSS regression was the peak level of 141dB SPL.