The extraction of the ..pi..NN form factor F/sub piN/N(t) from Reggeized one-pion-exchange (OPE) fits to hadronic cross sections is reexamined. The objective is to explain the Goldberger-Treiman discrepancy ..delta../sub ..pi../ which is determined from the extrapolation of F/sub piN/N to t = 0. Until now Reggeized OPE analyses have led to values of ..delta../sub ..pi../ that disagree with experiment by a factor of 2. It is argued that this failure is likely due to an insensitivity of the OPE amplitudes to heavy-pion (three-pion-resonance) contributions, at small t, when F/sub piN/N is parametrized by standard forms, e.g., the monopole or the dual model. A new functional form is proposed for F/sub piN/N(t) which has the virtue of enhancing these contributions at small momentum transfers, thus leading to OPE amplitudes more sensitive to 3..pi..-resonance effects. This form factor is then used to reanalyze differential cross sections for pp..-->..n..delta../sup + +/, pp..-->..p..delta../sup +/, np..-->..pn, p-barp..-->..n-barn, and ..gamma..p..--> pi../sup +/n, as well as asymmetries for ..gamma..p..--> pi../sup +/n, at high energies (E/sub L/approx. =5--25 GeV) and Vertical BartVertical Bar< or approx. =0.3 (GeV/c)/sup 2/. The results show that a self-consistent fit to all these data, with the correct value of ..delta../sub ..pi../, is indeedmore » possible.« less