Given integers 0 =< r < k, is there a standard way to relate the series Z(s; k,r) := Sum{n=1..oo} 1/(kn+r)^s to Zeta(s) := Sum{n=1..oo} 1/n^s ? (Of course for r = 0 this is trivial.) --Dan _____________________________________________________________________ "It don't mean a thing if it ain't got that certain je ne sais quoi." --Peter Schickele