586 Mason tions in the whole page and the whole sheet concerned, can act- ually determine which of the possibilities here suggested is the true quittance for our quandary, or whether we must seek further for another explanation such as the occurrence of a "cancel." To bibliographers, then, we may leave the problem, but not without profit to ourselves, even though the little investigation has produced no definite solution: for my pur- pose has been to convince myself and to remind other chiefly literary students that textual criticism can afford to generalize and dogmatize on the basis of a single copy of a Renaissance printed book very little more securely than on the basis of a single Mediaeval manuscript. LAWRENCE MASON.
Yale University.
