L-dopa therapy reverses some but not all of the motor deficits in human Parkinson patients. Although a number rat analogues of human Parkinson's disease have been developed for evaluating the efficacy of drug therapies, it is not known whether L-dopa has a similar selective action on the motor symptoms in the rat models. To examine the effectiveness of L-dopa in reversing the motor deficits in rats, we administered 6-OHDA unilaterally to produce hemi-Parkinson rats, which were then trained to reach for food using either their impaired (contralateral to the lesion) limb or their good (ipsilateral to the lesion) limb. To assess the skill, accuracy and range of limb movement, rats reached for pasta from a horizontal array of 260 vertically orientated pieces of pasta. The number and location of pasta pieces taken from this matrix was calculated and the qualitative aspects of the reaching movements were rated. The quantitative data on pasta sticks retrieved indicated that forelimb extension and movement radius around the shoulder joint was reduced by 6-OHDA treatment and did not improve after chronic L-dopa treatment. The qualitative analysis showed that grasping patterns, paw movements and body movements impaired by the lesion were also not improved by L-dopa treatment. These findings are the first in the rat to suggest that whereas L-dopa has a general activating effect on the rat's whole-body movements, as displayed in contralateral rotation, its effectiveness does not extend to skilled forelimb movements. The results are discussed in relationship to the idea that the restoration of some skilled movements may require normal synaptic function.