Purpose: The purpose of this study was to develop a valid (V)over dotO(2max) testing protocol for the NordicTrack (NT) cross-country ski simulator. Methods: To be considered a valid protocol, it was felt that the maximal values obtained from the cross-country skiing (XC) test should be similar to those obtained from a maximal graded treadmill (TM) test. During pilot testing, cadence (strides.min(-1)), arm resistance, leg resistance, and grade were individually manipulated to study the effect on measured (V)over dotO(2). A protocol for the NT was developed from the findings of the pilot study, and the results were compared with a TM (V)over dotO(2max) test utilizing the Bruce protocol. Thirteen male and 15 female volunteers (aged 22-49 yr) served as subjects and completed both tests in random order. The maximal physiological responses from the XC and TM max tests were compared using paired t-tests. Results: There were no significant (P > 0.05) differences in maximal (V)over dotO(2) (42.5 vs 42.6 mL.kg(-1).min(-1); 3.27 vs 3.31 L.min(-1)), heart rates (HR; 185 vs 185 bpm), or ratings of perceived exertion (RPE; 19.0 vs 19.1) for the XC and TM tests. There were significant (P < 0.05) differences for maximal (V)over dotE (124.5 vs 118.6 L.min(-1)) and respiratory exchange ratio (RER; 1.15 vs 1.21) between the XC and TM tests, respectively. Conclusions: It was concluded that the testing protocol developed for the NT XC simulator is valid and can be expected to elicit similar maximal values when compared with TM testing.