Our work of 1936 on the anomalous scattering of protons by protons in the energy region 600 to 900 kv was repeated during 1938 with an entirely new scattering apparatus designed for better angular measurements. Particular care was taken in order to know the energy (velocity) of the protons at the scattering volume with accuracy. The number of protons in the primary beam during each observation was determined by counting the protons scattered backward from a gold foil placed in the path of the proton-beam beyond the hydrogen-scattering volume. The scattering of protons by spectroscopically pure argon for several voltages and various angles was found to obey the Rutherford-Darwin formula. These measurements with argon provided a calibration of the voltage-scale in absolute units which was found to agree with our standard voltage-scale within about one percent. On the latter scale the midpoint of the gamma-ray resonance for protons on lithium is at 440 kv, the molecular ions thus giving a calibration point at 880 kv. A strong resonance for protons on fluorine occurs at 867 kv. The scattering anomaly arises from the failure of the Coulomb law of repulsion for very close distances of approach, because of the existence of a strong attraction of very short range which is the nuclear force between two protons. The newer measurements are in good agreement with our results in 1936, which were shown by Breit and his colleagues to correspond in the wave mechanics to a simple $s$ wave scattering. They also agree well with the data of Herb and his colleagues where the measurements overlap at about 860 kv. No clear-cut evidence for higher order scattering is found in our newer data, although the Mott ratios for angles from 20° to 30° are higher than would be predicted from the values for angles from 40° to 45° on the basis of $s$ wave scattering only. The departures are from five to ten percent, whereas from the internal consistency of the data one would hardly expect statistical errors of this magnitude. The discussion indicates that systematic errors affecting the scattering at 20° to 30°, although unlikely, cannot definitely be excluded. These observed systematic deviations are of the type which would be contributed by a small amount of $p$ wave scattering superposed on the spherically symmetrical $s$ wave scattering.

• Received 9 October 1939

DOI:https://doi.org/10.1103/PhysRev.56.1078