The purpose of this investigation was to assess the reliability and validity of maximal oxygen uptake estimates (ESTmax) from the ACSM submaximal cycle ergometer test. Subjects included 15 men and 15 women aged 21-54 yr who performed two submaximal tests and one maximal cycle ergometer test to determine maximal oxygen uptake (VO2max). During the submaximal tests, heart rates (HR) were recorded from a radio telemetry monitor. ESTmax was predicted for both submaximal trials by extrapolating HR to an age-predicted maximal HR. Correlation coefficient and standard error of measure (SEmeas) for ESTmax between submaximal trials were r = 0.863 and SEmeas = 0.40 l.min-1, while a t-test revealed no significant difference between trials. Although trial means were not significantly different, large variation in individual cases was evident by the high SEmeas (0.40 l.min-1) and by a large SEmeas expressed as a percentage of the mean (13%). The mean of the two ESTmax significantly overestimated measured VO2max with percent error, total error, and mean error equal to 25.7%, 0.89 l.min-1, and 0.63 l.min-1, respectively. The standard error of estimate expressed as a percentage of the mean was equal to 16% and 15% for both ESTmax. In summary, the ACSM protocol failed to be reliable as represented by the large differences found between submaximal trials. Furthermore, the protocol significantly overestimates VO2max and should not be used when an accurate assessment of VO2max is required.