The rotarod test, in which animals must balance on a rotating drum, is widely used to assess motor deficit in neurodegenerative disease models in rodents. Performance is measured by the duration that an animal stays on the rod as a function of drum speed. Two different protocols are widely used, incremental fixed speeds or an accelerating protocol, but there is little information on their equivalence or the relative power, reliability and sensitivity of the two protocols. The present study was undertaken to compare the incremental fixed-speed and accelerating rotarod protocols on two different lesions of the ascending forebrain dopamine pathways. Three groups of rats were used, controls, rats with 6-OHDA lesions of nigrostriatal bundle, and rats with terminal 6-OHDA lesions within the striatum. Rats were tested at different time points after the lesion. We report that whereas the incremental protocol is more sensitive to detect the presence of a lesion, the accelerating protocol provides a more discriminative test to correlate motor deficits against lesion size.