Woo wrote: > So I done it by sum(B<A^2), but I'm not sure if there is anther way to > calculate the number of cases in the distribution or what I have done is > fine What you have done is fine. -- Karl Ove Hufthammer