A variety of tests have been proposed for evaluation of motor deficits in rat models of hemiparkinsonism. The most widely used test, the "rotational behaviour" test, is based upon motor asymmetry induced by drug administration, leading to problems of sensitization, conditioning and priming. In addition, lesion-induced rotational behaviour is sometimes overcompensated by treatments (grafts) while other deficits are unaffected. However, rotational behaviour is reproducible and easy to quantify. On the contrary, most drug-free tests have proved of limited applicability because of subjective scaling, excessive interaction between the experimenter and animals, motivational effects, excessively complex tasks, etc., making it difficult to compare results from different laboratories. In this paper, we present a rotarod based method for automated drug-free evaluation of overall motor deficits in the rat model of hemiparkinsonism, and for identifying treatment-induced changes. Animals are pre-trained on the rotarod and then tested at a series of increasing rod speeds; from this set of values a unique score for each animal is computed (the overall rotarod performance, ORP) summarizing its performance at different speeds. This value is easy to compute and greatly simplifies statistical comparisons. Major advantages of the test are: (i) it does not require drugs, but is nevertheless highly objective, reproducible and easy to quantify; and (ii) falling of animals from rotating rod seems to depend on a combination of lesion-induced deficits which become more evident when rats are forced to move at faster speeds. Since the test is not based exclusively on motor asymmetry, it may also be useful for characterization of bilaterally lesioned animals, for which drug-induced rotational behaviour tests cannot be used.