The main purpose of this study was to assess the validity of the Cosmed Fitmate (FM) for the prediction of maximal oxygen consumption (VO(2)max). In addition, this study examined whether measuring submaximal VO(2), rather than predicting it, can improve upon the prediction of VO(2)max. Participants for the study were 48 young to middle-age adults (32 men, 16 women), with a mean age of 31 yr. Each participant completed a submaximal and maximal treadmill test on 2 separate occasions. During the submaximal test, VO(2)max was predicted using the FM. This device extrapolates the linear regression relating heart rate (HR) and measured VO(2) at submaximal work rates to age-predicted maximum HR (HR = 220 - age). The criterion measure was obtained using a graded, maximal treadmill test, with VO(2) measured by the Douglas bag (DB) method. There was no significant difference between VO(2)max predicted by the FM and VO(2)max measured by the DB method. The results of this study showed that a strong positive correlation (r = 0.897) existed between VO(2)max predicted by the FM and VO(2)max measured by the DB method, with a standard error of the estimate (SEE) = 3.97 ml·kg(-1)·min(-1). There was a significant difference in VO(2)max predicted by the American College of Sports Medicine (ACSM) metabolic equations and VO(2)max measured by the DB method (p = 0.01). The correlation between these variables was r = 0.758 (SEE = 5.26 ml·kg(-1)·min(-1)). These findings indicate that a small, portable, and easy-to-use metabolic system provides valid estimates of VO(2)max, and improves upon predictive accuracy, compared to using generalized ACSM metabolic equations.