The purpose of this study was to determine the coefficients of correlation for intrarater and interrater reliability for producing the strength-duration curve. The intrarater reliability curve was determined by a test-retest by the same examiner within a one-week interval. The interrater reliability was determined by using two examiners, each testing and recording his own results on the same subject within the same hour. The tibialis anterior muscle was tested in 24 healthy men. The interrater coefficients of correlation of 11 pulse durations of the strength-duration curve ranged from r = .979 to r = .771 (mean r = .945). The intrarater reliability coefficients ranged from r = .850 to r = .394 (mean r = .541). Both mean coefficients were statistically significant at p less than .01. The variance interpretation of the correlation coefficient data revealed that intrarater coefficients were not highly consistent (r(2) = .29). Interrater coefficients interpreted by variance estimates tended to be consistent (r(2) = .89). We believe that normal physiologic variation in the human subject was the cause of low intrarater values. The necessity of developing a normalized strength-duration curve is discussed.