Primes classification

[Max A.]

On 1/3/07, Artur <grafix at csl.pl> wrote:

> Wolstenholme in 1862 proof that nominators always are square for each
> prime >=5
> that mean for A127042={5, 7, 17, 19, 29, 31, 37, 41, 97, 127, 131, 211,
> 223, 227, 229,...}
> Sum_{k=1..p-1} 1/k^2} have form a^2/b^2

Something's wrong here.
E.g.: for p=5 we have
1/1^2 + 1/2^2 + 1/3^2 + 1/4^2 = 205/144
but 205 is not a square.

Max