Are we really measuring the phase of the nuclear scattering amplitude?

The interference at small \t\ of the Coulomb scattering amplitude f(c) and the nuclear amplitude f(n) is used to measure the phase of the nuclear scattering amplitude, and, hence, the rho value, where rho = (Ref(n)/Imf(n))(t=0). The ''normal'' analysis of <(p)over bar p> and pp elastic scattering uses a ''spinless'' Coulomb amplitude, i.e., a Rutherford amplitude (2 root pi alpha/t) multiplied by a Coulomb form factor G(2)(t), an ansatz that pretends that the nucleon does not have any magnetic scattering. In this article, we investigate the role of the anomalous magnetic moment of the nucleon, kappa approximate to 1.79. Given the method of analysis currently used by most published experiments, we conclude that the current experimentally inferred values of rho for <(p)over bar p> should be systematically lowered by approximate to 0.005-0.0100 and, correspondingly, the rho values for pp should be systematically raised by the same amount. We discuss the theoretical uncertainties and a method of experimentally minimizing them.