Relation between the direction of a recoil electron and that of the scattered x-ray quantum.—It has been shown by cloud expansion experiments previously described, that for each recoil electron produced, an average of one quantum of x-ray energy is scattered by the air in the chamber. If the quantum of scattered x-rays produces a $\beta$-ray in the chamber, then a line drawn from the beginning of the recoil track to the beginning of the $\beta$-track gives the direction of the ray after scattering. Using a chamber 18 cm in diam. and 4 cm deep traversed by a carefully shielded narrow beam of homogeneous x-rays, with exploded tungsten wires as sources of light, nearly 1300 stereoscopic cloud expansion photographs were taken. Of the last 850 plates, 38 show both recoil tracks and $\beta$-tracks. The angles projected on the plane of the photographs were measured and it was found that in 18 cases, the direction of scattering is within 20° of that to be expected if the x-ray is scattered as a quantum so that energy and momentum are conserved during the interaction between the radiation and the recoil electron. This number 18 is four times the number which would have been observed if the energy of the scattered x-rays proceeded in spreading waves, that is if the direction of production of a $\beta$-ray was unrelated to the direction of the recoil track. The chance that this agreement with theory is accidental is about 1/250. The other 20 $\beta$-rays are ascribed to stray x-rays and to radioactivity. This evidence seems a direct and conclusive proof that at least a large proportion of the scattered x-rays proceed in directed quanta of radiant energy.

• Received 23 June 1925

DOI:https://doi.org/10.1103/PhysRev.26.289