An extensive investigation was performed about the effects of high temperature prestraining on the notch toughness of a killed mild steel by using the standard V-notch, 2 mm U-notch and press-notched Charpy impact specimens. It was suggested after examining the test results that the characteristics of notch toughness in steel thus treated could generally be comprehended by selecting three criteria in the standard V-notch Charpy test, the maximum energy absorbed ( E _max) and 50% shear transition temperature ( T_γs ) as the measures representing the behavior of notch toughness in the high temperature range, and 15 ft-lbs transition temperature ( T_γ15 ) as to the ductility transition phenomenon in the low temperature. The notch toughness was scrutinized by using those three criteria, and it was found that it decreased remarkably as the amount of prestraining increased up to about 40% (in natural strain), over which the embrittlement became nearly saturated. The harmful effect of prestraining was more remarkable in T_γ15 than T_γs and it was conspicuously marked after the prestraining at 200°C. The value of T_γ15 became about 200°C higher than that of the base metal in the worst case. Less adverse effect was recognized by the prestraining in higher temperature, and in 600°C the embrittlement was hardly seen after prestraining. Whereas the effect was nearly unrelated to prestraining temperatures in the behavior of E _max value. Moreover, it was proved that the compressive prestraining affected the notch toughness in nearly the same way as tensile prestraining. As described in the Appendix, the role of aging was investigated by comparing the effects of high temperature straining followed by heating at the prestraining temperature with that of aging after room temperature prestraining under various heating times from 2 minutes to 24 hours. As the results it was found out that the greater part of the embrittlement, caused in excess by high temperature prestraining as compared with room temperature treatment, was of the nonagingtype and the remainder was the one due to strain-aging. It was also found that the speed of strain-aging in mild steel was unexpectedly rapid, and the aging was presumed to conclude almost in only a few minutes.