On 8/3/07, Fred lunnon <fred.lunnon@gmail.com> wrote:
...
Since s(m, 0) = m!, it's reasonable to enquire whether Legendre's
classic result might generalise.
Harrumph! should have read
s(m, 1) = (-)^(m+1)*(m-1)!
Just testing you. Everybody failed. WFL
