Comparing results of various training studies is often confounded by use of different indices of exercise intensity. Two frequently used indices are: 1) the % HR max method (exercise at a HR corresponding to a chosen percentage of maximal HR), and 2) the Karvonen method (KM) which corrects the % HR max method for resting HR. In this study, % net VO2max was employed as the criterion measure of exercise intensity and a comparison was made between the prediction of % net Vo2max by the above two methods. Experiments were performed on nine males (M ± SD age = 23.7 ± 2.6 yrs) and consisted of double determinations of resting VO2 and multiple measurements of resting HR under five different conditions. Duplicate determinations of maximal VO2 and HR preceded a third treadmill test consisting of four, 5-minute workloads representing approx 25,45, 65, and 85% net VO2max. The results demonstrated that the KM's prediction of exercise intensity was not significantly different (P > 0.05) from measured intensity at any of the four workloads. Additionally, the % HR max procedure yielded large, significant overpredictions of exercise intensity (29, 22, 16, and 8% overpredictions for the four submaximal intensities, respectively). These results suggest that the KM yields a training HR that reflects exercise intensity within reasonable limits of accuracy.