PURPOSE: The purpose of this study was to determine whether in vitro plain radiography and ultrasonography(US) could predict the outcome of gallstone dissolution with methyl tert-butyl ether(MTBE). METHODS AND MATERIALS: The plain radiography and ultrasonography of 75 gallstones removed at surgery from 75 patients were obtained. The plaih radiographic findings were grouped by their calcification pattern as group l:lucent, group 2:central, group 3:diffuse and group 4: laminated. The ultrasonographic findings were grouped by their echo pattern as group 1 :arc shaped, clearly defined surface echo followed by distinct acoustic shadow, group 2:strong surface echo with gradually attenuating, meniscus shaped inner echo, group 3:strong surface echo with fill-in pattern of internal echo, group 4:strong surface echo with irregular inner echo and group 5:entirely discerning circumsference, homogeneous internal echotexture. After imaging, the gallstones underwent MTBE dissolution for 6 hours. The residual weight of each stone was measured every 2 hours. The correlation between imaging findings and dissolution rate was retrospectively analysed. RESULTS: The radiolucent and central calcification stones were dissolved rapidly as compared with the diffuse and laminated calcification stones(p<.05). There was no statistically significant difference among the groups based on ultrasonographic findings alone. However, after exclusion of the certain calcification groups that were resistent to dissolution-diffuse or laminated calcification stones-all US groups except one that shows wholly circumscribed margin and homogeneous fill-in pattern of internal echo dissolved well to average 20% or less of the original weight(p<.05). CONCLUSION: MTBE dissolution can be tried in stones that meet both ultrasonographic (not a stone with homogeneous fill-in pattern of internal echo) and plain radiographic (radiolucent or cental calcification stone) criteria.