After a few years of experience with extracorporeal shock wave lithotripsy (ESWL) and other fragmentation techniques, it has become apparent that stone fragility is a significant clinical distinction that should be taken into consideration when selecting a treatment program. In 30 unselected patients, stone mineral content, density and area were measured in vivo by dual-photon absorptiometry prior to perform ESWL treatment. Stone area determinations showed a median of 4.21 with a range of 0.46 to 49.7 cm2. Stone mineral content (g) and stone density (g/cm2) values were 2.47 and 0.46 with ranges of 0.37 to 13.7 and 0.167 to 1.203 respectively. The number of shocks needed for total fragmentation were 2375 with a range of 1200 to 7800. No correlation could be found between the number of shocks needed for fragmentation and the stone area or density. On the other hand, a strong linear correlation (r = 0.81, p < 0.001) (Fig. 1) could be demonstrated between stone mineral content and the number of shocks needed for fragmentation. Our results support the concept that size alone is not always a suitable criterion for selecting a stone as appropiate for ESWL, since no correlation could be found between stone area and the number of shocks needed for total fragmentation. We were also unable to find any correlation between in vivo stone density measured by dual-photon absorptiometry and the number of shocks required for stone fragmentation. Instead, a strong linear correlation between stone mineral content and its resistance to shock wave fragmentation was found. Therefore, calculation of mineral content appears to be the determinant of the amount of energy required for total fragmentation. Our results strongly suggest that in vivo stone mineral content measurement provides helpful information for predicting the fragmentation prospect of a stone.