. . "It is quite typical of many mathematical theorems that they go in one direction and not the other, and Euclid's theorem on perfect numbers is an example of that." . .