Eight wheelchair sportsmen conducted eight wheelchair exercise tests on a treadmill. Two workload strategies were followed: strategy 1--increments in speed at a constant slope and strategy 2--increments in slope at constant velocity. Thus, data on cardio-respiratory and propulsion technique parameters were obtained on two identical series of 16 speed and slope combinations. Between each two identical speed and slope combinations of strategies 1 and 2, a different workload history is apparent. A four-factor analysis of variance with repeated measures on the factors "strategy" (workload history), "speed," and "slope" was applied (P less than 0.05). No "strategy" effect was seen in the cardio-respiratory parameters (gross mechanical efficiency, ventilation, oxygen consumption, and heart rate), work/cycle, and cycle time. Thus, within the experimental set-up, workload history did not affect the parameters studied and 3-min workload periods appeared sufficiently long for experienced wheelchair users to adapt to the requirements of a given speed and slope combination. Significant effects were found on "speed," "slope," and their interaction in all parameters tested. Moreover, a comparison of two equal levels of power output, but different speed and slope, led to a significantly higher efficiency, cycle time, and work per cycle for the "low speed and high slope" combination. Push time and recovery time appeared highly dependent on speed and slope, respectively. The findings indicate that propulsion technique and cardio-respiratory parameters should not merely be studied in relation to power output, but also with respect to its constituents, speed, and slope/resistance.