Milner's (1963) report of impaired performance on the Wisconsin Card Sorting Test (WCST) in a group of patients with frontal lobe lesions suggested that this test might be a useful one in the investigation of individual patients with suspected brain lesions. However, for many of our older hospital population the WCST was found to be too difficult and distressing, and also the inherent ambiguities associated with certain responses limited the test's usefulness for research purposes. Therefore, a simpler and less ambiguous modification was devised (MCS) and a new method of measuring perseverative errors proposed. In a group of 53 patients with unilateral cerebral lesions, those with frontal lobe lesions performed less well with the MCST and made a higher proportion of perseverative errors than those with lesions elsewhere: there were no laterality effects in either frontal or non-frontal groups. The usefulness of the MCST for detecting frontal lobe lesions in individual patients was established, and the use of cut-off scores briefly discussed.