To the Editor.Ìý —Markham and Diamond¹ drew attention to the controversy concerning the early start or postponement of levodopa therapy in Parkinson's disease. We should like to point to an inaccuracy in their article. They use two disability scores (Hoehn and Yahr and UCLA scores), but because these are so-called ordinal scales, average values cannot be calculated. As Wulff² said: "To measure on an ordinal scale is like using an unevenly stretched tape measure."In our opinion, figures like the one shown in their article cannot be made for either the UCLA or the Hoehn and Yahr scale, and, as a consequence, their figure has no meaning.