Mean stress effects in pressurized steel blocks were examined under constant amplitude fatigue loading. The tests were performed to provide experimental data needed to study the effect of mean stress on fatigue lives of subject specimen, and to substantiate the use of analytical expressions to account for the mean stress. The mean stress was the result of subjecting the specimens to an autofrettage pressure which induced compressive residual stresses at the crossbore intersection of the specimens. Fatigue tests were carried out under both tensile and compressive mean stress levels. Test results were compared to several mean stress accounting relationships such as the Smith-Watson Topper, Bergmann and Seeger, modified Goodman, Gerber and Soderberg. Test results indicated that the modified Goodman equation is favorable in accounting for the effect of both tensile and compressive mean stresses on fatigue life (up to a compressive mean stress to ultimate stress ratio of −0.2). The behavior under compressive mean stress to ultimate stress ratio of less than −0.2 indicated that a linear correction relationship was required.