Remarks on exponential congruences and powerful numbers (Q1106255)

This paper considers a number of questions relating the solvability of such congruences as \(a^{p-1}\equiv 1 \pmod{p^2}\) to properties of powerful numbers. In particular, various deductions from, and partial results towards, the conjecture of Erdős, that there do not exist 3 consecutive powerful numbers, are given. In conclusion it is asked, among other things, whether \(m^4-1\) fails to be powerful for ``almost all'' even \(m\). It seems to the reviewer that much more can be proved, by a simple sieve argument. Namely, if \(f(n)\) is an integer polynomial, not identically powerful as a polynomial, then \(f(n)\) fails to be powerful for ``almost all'' \(n\).