The purpose of this study was to assess the reliability and validity of a portable metabolic system (TEEM 100) during submaximal and maximal (VO2max) exercise using a computer-based metabolic system as the reference system (REF). Between repeated trials of submaximal exercise at three constant loads, differences in ventilation (Ve) and oxygen consumption (VO2) were 0.2 +/- 4.9 L . min-1 and 0.03 +/- 0.10 L . min-1 for REF, and 1.9 +/- 0.7 L . min-1 and 0.00 +/- 0.17 L . min-1 for TEEM 100. Pooled intraclass reliability coefficients for Ve and VO2 calculated from the repeated submaximal trials were r = .89 and r = .94 for REF, and r = .86 and r = .94 for the TEEM 100. Respiratory exchange ratio (RER) measured by the TEEM 100 was significantly higher (p = .01) at only the lowest workload. At VO2max, the TEEM 100 recorded significantly higher values for FeO2 (p = .01) and RER (p < .001). These results suggest that the TEEM 100 provides reliable and valid measurements of VO2 during submaximal and maximal exercise.