Amphetamine (AMPH) has been proposed as a treatment for post-stroke motor deficits when coupled with symptom-relevant physical rehabilitation. Whereas a number of experimental studies report improvements in endpoint measures of skilled reaching for food by rats, there has been no assessment of whether beneficial effects extend to overcoming learned non-use of the limb in the acute post-stroke period or to the qualitative deficits in movement in the chronic post-stroke period. In addition to evaluating the effects of AMPH on success, these were the objectives of the present study. In three different reaching experiments, groups of rats were pre-trained in skilled reaching for food prior to receiving a motor cortex stroke via pial removal. Postoperatively the rats received periodic AMPH treatment and daily rehabilitation. In the acute post-stroke period, AMPH failed to prevent the development of learned non-use of the limb, and in the acute and chronic period failed to improve recovery of reaching success, and also failed to improve the qualitative aspects of reaching movements. Nevertheless, AMPH did enhance adjunct non-reaching movements of locomotion, rearing and turning. The results are discussed in relation to the idea that the beneficial effects of post-stroke AMPH treatment do not extend to all movements, especially the movements of a forelimb in retrieving and consuming food.