Modification of the effective-range expansion for the 1 S 0 nucleon-nucleon state to include the effect of the one-pion-exchange contribution (OPEC) by means of the partial-wave dispersion relation 1 ' 2 or the fixed-angle dispersion relation 3 leads to the prediction that the shape parameter, P, in the expansion q cot6 0 = -\/a + \r e q 2 -Pr e s q 4 + • • • is positive. This is also predicted by potential models which include the long-range one-pionexchange potential (OPEP) and either an intermediate-range attraction plus repulsive core 4 or energy-independent boundary condition at intermediate range 5 whose parameters are adjusted to fit the effective range, r e , and scattering length, a. A more quantitative prediction is provided by including the electrostatic repulsion in the partialwave dispersion relation 2 and using two additional parameters to fit observed phase shifts at 95 and 310 MeV as well as a and r e \ this calculation 6 gives P = +0.024. This prediction is of opposite sign to that made by an energy-independent boundary condition at intermediate range, 7 ' 8 an energydependent boundary condition 9 which fits the highenergy (i.e., up to 310 MeV) l S 0 phase shifts, 10 ' 11 or hard-core potentials with intermediate-range attractive tails, 12 which do not include the OPE effect. Since the OPE predictions have been quantitatively confirmed in higher angular momentum states, 13 and since the qualitative features of the phase shifts empirically determined in the 100-to 300-MeV range are in good agreement with models based on the exchange of known bosons and strongly interacting boson systems ("resonances") between the two nucleons (for a brief discussion of these qualitative features and references, see reference 11), it is important to test the consistency of these descriptions with the interaction in the S states as rigorously as possible. This is particularly true since the models in best agreement with the high-energy scattering experiments predict only 2 MeV of the observed 8-MeV binding for the three-nucleon systems, 14 and the latter calculation is more sensitive to the details of the S-state interactions than the high-energy scattering. One of the few tests available is the prediction of the shape parameter. Since the effective-range expansion fails to converge above 10 MeV, 1 this test can only be made by means of very low-energy nucleon-nucleon experiments. Existing n-p data are not of sufficient precision to yield definite conclusions. 11 In this Letter we show that the recently reported experiment on />-/> scattering near the interference minimum at 0.3825 MeV 15 and the p-p differential cross sections measured at 1.397, 1.855, 2.425, and 3.037 MeV 16 can be analyzed to yield a precise value of the shape parameter. This analysis is only possible because the latter experiments also yield a precise value for the J-weighted average of the 3 P phase shifts, and because we claim to have a sufficiently quantitative understanding of the multirange character of the nucleon-nucleon interaction to us...