On the Ising problem and Mayer's cluster sums