... Wilson chamber, the appearance of positrons under the action of $\gamma$-radiation on matter. In their experiments, an active thorium deposit, applied to an aluminum wire, was sealed in a thin-walled $(0.1\ \mathrm{mm})$ glass tube, through whose walls most of the $\gamma$-rays of the source could pass. The tube was placed in a Wilson chamber; in this case 24 positrons were observed for approximately 2,000 negative electrons. It might have been thought that the positrons were the result of the action of $\gamma$-rays on the aluminum wire, but replacing the aluminum by platinum changed nothing.

The question of the emission of positrons by radioactive elements was investigated in more detail in the work of Alikhanov and Kozodaev (Alichanow, Kosodaew, Z. Phys. 90, 249, 1934). They used for the study of positrons the method of magnetic spectroscopy, developed long ago for $\beta$-rays. In this case the positron indicators were two Geiger counters, in which simultaneous discharges caused by the passage of one and the same positron were observed. They investigated a radium emanation preparation enclosed in a thin-walled glass ampoule. The preparation gave a large number of positrons.

A distinctive feature of the positrons emitted by a radioactive source is their energy spectrum. Whereas the spectrum of positrons emitted under the action of $\gamma$-radiation on heavy elements has a number of sharply expressed maxima corresponding to definite lines of the $\gamma$-spectrum of the element under investigation, the spectrum of positrons emitted by the radioactive element itself is continuous, with small maxima, due, in all probability, to the interaction of the $\gamma$-radiation with the nuclei of the radioactive element. The continuous positron spectrum is in general similar to the continuous spectrum of $\beta$-rays.

To explain the positron emission of radioactive elements there are two possibilities: 1) positrons are emitted by the nuclei of the element, 2) the emission of positrons is a secondary effect caused by the interaction of $\alpha$-, $\beta$-, and $\gamma$-rays with extranuclear electrons. At present the second possibility must be considered the more probable.

L. Groshev

$\gamma$-QUANTA OF HIGH ENERGY ARISING IN THE BOMBARDMENT OF LITHIUM AND FLUORINE BY FAST PROTONS¹

When fast protons or deuterons act on certain light elements, $\gamma$-quanta arise in the latter. To estimate the energies of these quanta one usually uses data obtained from analysis of the absorption curves of the given $\gamma$-radiation in some substance. However, this method not only has low sensitivity, but also does not make possible an unambiguous determination of the energy of the $\gamma$-quantum in the case when the absorption curve has been obtained in only one substance. The reason for this is that for quanta of high energy, in addition to the usual absorption mechanisms, which weaken with increasing frequency of the $\gamma$-quanta (Compton scattering, photoelectric absorption), an ever greater role is played by a new type of absorption—the creation of pairs (electron—positron); in this process the part of the absorbed energy due to it increases strongly with increasing frequency of the $\gamma$-quantum and depends to a great extent on the atomic number of the absorbing element. Fig. 1 gives theoretical curves for the dependence of the absorption coefficient (with pair creation taken into account) on the energy of the $\gamma$-quantum, calculated by Oppenheimer for Pb and Cu. From the curves it is evident how strongly their form changes in passing from Cu to Pb. They also show why, from the absorption coefficient for a single element, it is impossible to judge unambiguously the energy of the $\gamma$-quantum. Therefore, in order to determine the energy of $\gamma$-radiation by this method, it is necessary to measure simultaneously the absorption coefficients at least for two substances.

Lauritsen et al. determined the absorption coefficients in lead and copper

for the γ-radiation arising in Li (LiCl) and F (CaF₂) when they are bombarded with protons of energy 800 MeV. Their data and the values for the γ-radiation of radium are plotted in Fig. 1 as horizontal bars. From the curves it can be seen that the γ-quanta from F have an energy of approximately 5.5 MeV and those from Li, 6.3 MeV. However, the correction for scattered radiation (10%), taken from the data for Ra and extended to all the values obtained by the authors, does not give complete confidence in the correctness of the results obtained. Therefore, in order to estimate the energy of the γ-quanta the authors also used another, more direct method. With a Wilson chamber they measured the energy spectrum of electrons and positrons knocked out of a lead plate by the γ-radiation under study; in this case the energy was measured from the curvature of the tracks in a magnetic field of strength 1200 gauss. Fig. 2 gives the energy spectrum for electrons (circles) and positrons (dots) knocked out of a lead plate (3 mm) by the γ-radiation arising in CaF₂ when it is bombarded with protons of energy 800 MeV. A total of 641 tracks were measured, of which 434 belonged to electrons and 207 to positrons. Fig. 3 gives the same for LiCl. Here 956 tracks were measured: 521 electron tracks and 435 positron tracks.

Making a correction in the distribution curve caused by errors in measuring the curvature of the tracks (according to Chadwick, Blackett, and Occhialini), the authors find for the energy of the γ-radiation from F 5.4 MeV (in agreement with the preceding data). For Li the γ-radiation is complex: besides the component at 12 MeV, there is a component of approximately 4 MeV. The true ratio of the intensities of these components cannot be estimated. It should be noted, however, that, in view of the different thicknesses of the effective layer for electrons and positrons, the component in

Fig. 1.

Fig. 2.

Fig. 3.

4 MeV in Fig. 3 is reduced. Until the true ratio of the intensities of these components is known, it is difficult to say whether the results for the $\gamma$-radiation of Li obtained by one method or another agree or do not agree. One can only state that there is no gross contradiction in these data.

Assuming that the initial energy distribution for electrons and positrons is the same, and taking into account the different thicknesses of the effective layers for paired and single electrons, the authors find that in $1\ \mathrm{cm}^3$ of lead 1.8 pairs are formed per 1 single electron for a $\gamma$-quantum energy of 5.4 MeV, and 10 pairs per 1 electron for 12 MeV. These data are in good agreement with Oppenheimer’s theory.

The reported results show what an important role, in the process of absorption of high-energy $\gamma$-quanta, is played by pair production, which until quite recently had entirely escaped observation.

L. Gorshev.

ARE THE ATOMS OF A METAL DEFORMED DURING COLD WORKING

A number of authors (Fan-Limp, Tammann) have expressed the opinion that the atoms of a metal after plastic deformation are in a special state and have a charge distribution different from the normal one. The latter can be determined by measuring the brightness of X-rays reflected from the face of a crystal. Taking into account the dependence of the intensity of the reflected X-rays on various factors (the distribution of atoms—the structure factor, thermal motion—the Debye factor, etc.), we find among them also the factor due to the distribution of charge inside the atom, the atomic factor $F$:

\[ F = Z \int_{d/2}^{d} p(z)\cos \frac{4\pi z \sin \vartheta}{\lambda}\,dz, \]

where $p(z)$ is the distribution of charge density between neighboring atomic planes of the reflecting face of the crystal. The defined integral is a function of $\vartheta$. When $p(z)$ changes, the angular dependence of $F$ also changes.

J. Boyd attempts, by measuring the atomic factor for deformed (powdered) and undeformed Be, to decide the question of deformation of Be atoms. The latter is chosen because of the small total number of electrons in it, so that the influence of the outer electron is quite clearly manifested. The results lead the author to the opinion that the atomic factor does not change as a result of the work-hardening of the metal. Attention is drawn, however, to the fact that the lattice-disorder factor due to deformation was not taken into account; this factor should act analogously to the temperature factor. Nevertheless, the work is interesting as an attempt to solve a question of considerable importance for the theory of plastic deformation (James E. Boyd, Scattering of X-Rays by cold worked and annealed Berillium. Phys. Rev., 45, 832, 1934).

S. Konobeevsky