It has long been well-known that the curability of cases of insanity diminishes rapidly as the duration of the disease increases. In this respect it follows a parallel course to all other diseases, and indeed to every other evil under the sun. I do not propose in any way to object to this fact, or to dispute the accepted measure of the relation between duration and curability. The object I have in view is to call attention to one or two very grave fallacies of almost equally wide acceptance in the supposed application of this well-known relation. One meets everywhere with the statement that a large proportion of the patients sent to our County Asylums suffer from mental disorder of considerable duration, say over twelvemonths—so far true—but it is added that these cases ought to have been sent to the asylums when recent, in which case a large proportion of them would have recovered, and the accumulation of chronic cases in asylums would be largely or completely checked. All this sounds very logical, but as I propose to show, it is quite untrue, and to those who are well acquainted with the fact, must be so obviously incorrect, that one is surprised to meet this statement in various forms in a very large proportion of the reports of Medical Superintendents of Asylums, in Medical Journals, and elsewhere; nor do I recollect seeing anywhere a hint as to the erroneous use of statistics on which it rests. I cannot, perhaps, give a better view of the remarkable and erroneous opinions to which this error leads than by giving in full a recent instance from a leading Medical Journal of the deductions drawn from it.