Despite the fact that the pressure reduction is acknowledged to be the single most cogent factor in producing decompression sickness, little has been done to define accurately the allowable limits beyond 2 ATA. This study provides some theoretical guidelines for future manned dives related to this problem. There were 324 albino mice used to define the relationship between saturation exposure pressure and the safe abrupt pressure reduction. The results from both the helium-oxygen and nitrogen-oxygen exposures support the idea of a linear, depth-dependent relationship between the saturation depth and the allowable pressure reduction. Support is presented for the use of a modified decompression ratio P1/P2 (P1 equals saturation pressure and P2 equals pressure following decompression) to account for the observed incidence of decompression sickness. An attempt is made, using the existing human data to relate this empirical relationship to the operational dive setting.