This study was designed to examine the reproducibility and the learning effects of incremental maximal cycle ergometer testing in patients with obstructive lung diseases who are untrained and have never done such a test before. Eleven patients were studied on 2 consecutive days with an interval of 24 h. The reproducibility (expressed in relative duplicate error) of most of the parameters tested was good: workload (watt) 4.5%, maximal oxygen consumption (VO2max 3.5%, gas exchange ratio (R) 3.4%, and heart rate 3.7%. Moderate reproducibility was found in maximal ventilation (VEmax) 6.6%, carbon dioxide production (VCO2max 6.0%, and respiration frequency (RF) 8.3%. The reproducibility of the time between the start of the test and the minimum value of VE/VO2 and VE/VCO2 was low. We concluded that in these patients the results of incremental cycle ergometer testing showed good reproducibility in a test-retest procedure. There were no learning effects.