Abstract. Research concerning the physiological and biomechanical parameters of wheelchair propulsion requires the use of an accurate and reproducible protocol. Although some research comparing different ergometers has been conducted, it is not easy to fulfil the requirements of the realistic simulation of propulsion together with the careful analysis of metabolic and biomechanical parameters of performance. The VP100H ergometer has been validated for this purpose by comparing the values of power output ($\dot W$
) and force exerted to accelerate the wheels (F) with the same measures obtained using a two-dimensional force transducer (platform). The reproducibility of the power was verified during a test re-test procedure. Ten sportsmen performed an incremental exercise. Maximal aerobic power ($\dot W_{{\rm aer,max}} $
), maximum oxygen uptake ($\dot V{\rm O}_{{\rm 2max}} $
), maximum heart rate (fcmax), $\% \dot W_{{\rm aer,max}} /\% \dot V{\rm O}_{{\rm 2max}} $
relationship (aV) and $\% \dot W_{{\rm aer,max}} /\% f_{{\rm cmax}} $
relationship (aH) were calculated. Results indicated no significant differences (P>0.05) in VP100H versus platform measurements for F and $\dot W$
. Errors of measured Fand $\dot W$
ranged from 0.89% to 7.56% and from 0.41% to 6.74%, respectively, depending upon the trunk muscles that participate in the propulsion. This corresponded to a maximum error of 4.9 W for $\dot W$
. No significant differences (P>0.05) were observed during the test re-test procedure for $\dot W_{{\rm aer,max}} $
, $\dot V{\rm O}_{{\rm 2max}} $
, fcmax, aH and aV. The coefficient of variation of these values ranged from 1.4 to 9.5, and the correlation coefficient from 0.68 to 0.98. We conclude that the VP100H is valid for measuring $\dot W$
, and F, and that these values are reproducible (when tested 10 days later).