Abstract
The experimental value of a(s)γ, where a(s) is the contact part of the magnetic-dipole hyperfine interaction and γ=μII, is considered as a function of the number N of 3d electrons present. The effects of intermediate coupling and configuration interaction (including core polarization) are allowed for wherever possible in the extraction of a(s) from the available experimental data. The resulting curve, which should more accurately describe the variation of the contact hfs with N, is considerably smoother than when these refinements are omitted. New experimental results for Ni61 are included.