A082812

[franktaw at netscape.net]

"Positions of sign changes in running sum of mu(k)/k: n such that 
sign(Sum(k=1..n; mu(k)/k)) <> sign(Sum(k=1..n+1; mu(k)/k))."

Has the comment: "Sequence is infinite since Sum(k=1..inf; mu(k)/k) 
equals 1/Zeta(1)= 0."

The sequence may well be infinite, but the argument given is not 
sufficient to establish it.  Can anybody do so?

Franklin T. Adams-Watters

________________________________________________________________________
Check Out the new free AIM(R) Mail -- 2 GB of storage and 
industry-leading spam and email virus protection.