On a Result by M. Rosenblatt Concerning the Von Mises-Smirnov Test