mesons via the u channel. 5 E. A. Paschos, private communication. Paschos hasElectron elastic-scattering experiments to date have been interpreted using the Rosenbluth formula based on the single-photon-exchange model. A measurement of R is a test of this model because a deviation of R from 1 is an indication of the size of the real part of the two-photon-exchange amplitude. 1 Because the interference between the single-photon amplitude and the twophoton amplitude occurs with opposite sign for electrons and positrons, one may write R » 1 + 4 XReB/A, where ReB/A is the ratio of the real part of the two-photon amplitude (ReJ9) to the single-photon amplitude (A). Earlier measurements of R by other experimenters 2 for the most part gave fial. Past theoretical estimates 3 ' 4 either make no definite prediction as to the size of 1/2-11, or predict it to be $0.02. A summary of previous investigations of R has recently been given by Pine. 5 Results.-The ratio R was measured for the laboratory scattering-angle regions 12.5°^ 9 ^ 35.0° and 2.6°^ 6 ^ 15.0° with incident electron (and positron) energies of 4.0 and 10.0 GeV, respectively. The high-? 2 data extend to higher q 2 than earlier experiments, and the moderate-? 2 data include measurements at smaller angles than previously explored.The results are displayed in Table I, and a comparison with previous measurements is giv-analyzed our results in terms of A-trajectory exchange and obtained good agreement with theory. en in Fig. 1. In the table, R is the corrected experimental ratio with its uncertainty. The uncertainty in R is the square root of the sum of the squares of the statistical uncertainty and the estimated uncertainty due to systematic errors, both of which are given in the table. The systematic error is dominated by the beam monitor uncertainty.The difference in radiative corrections for e + and e~ scattering was calculated using the results of Meister and Yennie, 6 with exponentation. The column labeled "Rad Corr" is the net correction to R from radiative effects. No uncertainty is assigned to the radiative corrections. The column labeled "ReB/A" in the table gives the 95% confidence limits for the quantity ReB/A defined earlier.As can be seen in the table, all the elastic data are consistent with R-l.This result is in agreement with estimates by Drell, Ruderman, and others, 3 and supports the one-photon approximation over an enlarged kinematical region.The inelastic measurements in the table, labeled "iV*(1238)," give R for all scattered events in which the missing mass of the final-state particles other than the recoiling electron lay between 1110 and 1370 MeV. About 70% of the cross section leads to iV*(1238) production. The remainder of the scattering in this region can be

No abstract