CONGRESSES AND CONFERENCES

FIFTH ALL-UNION CONFERENCE ON THE ATOMIC NUCLEUS

On November 20–26, 1940, the Fifth All-Union Conference on the Atomic Nucleus was held in Moscow. Physicists from Moscow, Leningrad, Kharkov, and other cities took part in the work of the Conference.

In all, more than forty reports were presented at the Conference, devoted mainly to the following questions:

1. Cosmic rays (12 reports).
2. Properties of fast electrons and hard photons (6 reports).
3. Isomerism of nuclei (3 reports).
4. Fission of heavy nuclei (4 reports).
5. Neutrons and the structure of nuclei (7 reports).
6. Nuclear reactions inside stars (1 report).
7. Application of nuclear physics to chemistry, biology, and medicine (5 reports).
8. Techniques for obtaining fast particles (3 reports).

Even from this list it is clear that at the present Conference all the most urgent problems of the physics of the atomic nucleus and of cosmic rays were discussed.

A distinctive feature of the last Conference may be considered the exceptionally large percentage of theoretical papers—over 30% of all reports. This percentage undoubtedly reflects the present state of the physics of the atomic nucleus and cosmic rays.

The Conference opened with a large report by I. E. Tamm, in which the speaker gave a very complete survey of the modern theory of mesotrons and nuclear forces1.

At the same session L. D. Landau presented his work “On the radius of elementary particles.”

By the radius of an elementary particle \(r\) the following is understood: if a particle interacts with electromagnetic radiation of wavelength \(\lambda\), then the particle may be regarded as pointlike only when \(\lambda \gg r\).

As the radius of the electron in classical electrodynamics there serves, as is known, the quantity \(\dfrac{e^2}{mc^2}\). This means that in classical electrodynamics one may consider the interaction of a point electron only with such electromagnetic waves whose wavelength is greater than the electron radius \(\dfrac{e^2}{mc^2}\). This radius may be obtained, independently of considerations connected with the proper mass of the electron, from the condition of the smallness of the back action of the electromagnetic field of the electron on the electron itself.

As is known, the equation of motion of a classical electron in an external electric field, taking into account the Lorentz force of friction, has the form

\[ m\ddot{\mathbf r}=e\mathbf E+\frac{2e^2}{3c^3}\dddot{\mathbf r}, \]

where $\mathbf E$ is the intensity of the incident field, and the second term represents the Lorentz force acting on the electron from the electromagnetic field created by the electron itself. In order that one may use classical electrodynamics, it is necessary that the Lorentz force constitute a small correction to the equations of motion, i.e.

\[ m\ddot{\mathbf r}\gg {2e^2\over 3c^3}\,\dddot{\mathbf r}. \]

If the incident field is electromagnetic waves of frequency $\omega$, then it is easy to find that the length of the incident wave must satisfy the inequality

\[ m\gg {e^2\over c^3}\omega, \]

whence it follows directly that the role of the electron radius in classical electrodynamics is played by the quantity

\[ {e^2\over mc^2}. \]

Thus, the limit of applicability of classical electrodynamics can be obtained without the aid of any additional physical assumptions, from the condition that the back action of the electron on itself be small.

In reality, however, the limit of applicability of classical electrodynamics lies considerably higher, at $\lambda\sim {\hbar\over mc}$, i.e. at the Compton wavelength, since at smaller wavelengths quantum effects begin to play an essential role.

Recently there have appeared a number of works, in particular by Heisenberg and Weizsäcker, in which the question of the limits of applicability of quantum mechanics was discussed. However, all these works proceeded from certain, quite arbitrary, physical assumptions.

L. Landau showed that in quantum theory a method can be developed for finding the limit of its applicability, i.e. the limit beyond which the theory comes into contradiction with itself without any arbitrary assumptions, but proceeding from considerations of the smallness of the back action of the field analogous to those made in finding the limit of applicability of classical electrodynamics.

Neglecting the back action of the field on a particle means that we regard the particle as moving under the influence of an external field, without taking into account the change in the latter produced by the particle itself. This corresponds to a certain weak interaction of the particle with the field. Formally, the weakness of this interaction means that the interaction of the particle with the electromagnetic field can be considered as a weak perturbation. In this case the matrix elements of the interaction of the particle with a photon are calculated with the aid of the wave functions of “free” photons and of the particle.

Thus, the smallness of the back action of the field on a particle in quantum mechanics corresponds to the applicability of perturbation theory to the consideration of the interaction of a charged particle with a photon.

Let us consider the effective cross section $\sigma_l$ for the scattering of a particle with angular momentum $l$. As is shown in the general theory of scattering\(^1\), in a reference system moving with the center of inertia,

\[ \sigma_l=4(2l+1)\pi\lambda^2\sin^2\delta_l, \]

where $\lambda$ is the wavelength of the particle, and $\delta_l$ is the phase of its wave function at infinity.

The condition for applicability of perturbation theory has the form $\delta_l\ll 1$, or

\[ \sigma_l\ll 4(2l+1)\pi\lambda^2. \]

\(^1\) See, for example, Mott, Theory of Atomic Collisions, ONTI, 1936.

Since scattering decreases with increasing angular momentum \(l\), it is sufficient that this condition be satisfied, for example, for particles with \(l \sim 1\), i.e., that

\[ \sigma_l \ll \lambda^2 . \]

If in the last formula \(\sigma_l\) denotes the effective cross section for the scattering of a particle with wavelength \(\lambda\) by a photon (the Compton effect), then the validity of this formula also represents the condition that the back action of the field be small, i.e., the criterion for the applicability of quantum theory. If this criterion is applied to the electron, then, with the aid of the Klein—Nishina formula transformed accordingly (to the coordinate system moving with the center of inertia), it can be written in the form

\[ e^2 \ll \hbar c . \]

The latter inequality is always satisfied. This means that in quantum mechanics the electron does not follow from the very limits of its applicability for short wavelengths. The electron may be “illuminated” by photons of arbitrarily small wavelength and at the same time be regarded as pointlike. By decreasing the wavelength, one can localize the electron in an ever smaller spatial interval. In this sense the radius of the electron in quantum theory is equal to zero.

Application of the same criterion to a particle with spin one (the “mesotron”) leads to quite different results. Namely, it turns out that the radius of a particle with spin 1 is equal to

\[ r_0=\frac{e^2}{\mu c^2}, \]

where \(\mu\) is the mass of the particle. Thus, in this case the radius coincides with the classical radius of the particle. The mesotron, like the electron, can classically be regarded as pointlike only if it interacts with a field of wavelength \(\lambda \gg r_0\).

This result explains the origin of the difficulties connected with the fact that the various effective cross sections for the scattering of mesotrons increase rapidly with increasing energy, which is at variance with experiment. Namely, in the case of high mesotron energies, perturbation theory proves inapplicable.

L. Landau developed a method for finding effective cross sections also in the region in which perturbation theory is already inapplicable. This method is based on the fact that, irrespective of the applicability of perturbation theory, the inequality must always hold

\[ \sigma_l < 4(2l+1)\pi \lambda'^2, \quad \text{where } \lambda'=\frac{\lambda}{2\pi} \]

(i.e. \(\delta_l < 1\)).

Failure of this condition would mean that more particles with angular momentum \(l\) are scattered than are incident, which is clearly absurd.

The quantity \(4(2l+1)\pi \lambda'^2\) gives an upper bound for the scattering cross section.

One may suppose that if the theory gives, for the effective cross section, a value greater than \(4(2l+1)\pi \lambda'^2\), this means that the exact theory would instead give

\[ \sigma_l = 4(2l+1)\pi \lambda'^2 . \]

This makes it possible to estimate the effective cross section even in the region where the theory is already certainly inapplicable.

The papers of Berestetskii, “Formation of Mesotrons under the Action of \(\gamma\)-Rays on Nuclei,” Smorodinskii, “Radiation Effect of Particles with Spin 1,” and Pomeranchuk, “Interaction of Cosmic Mesotrons with Matter,” were devoted to the application of Landau’s method to obtaining effective cross sections for various processes involving fast mesotrons.

In the discussion, not only the very great fundamental importance of Landau’s work was emphasized, but also its great practical impor—

…ness, since the method he developed is at present the only known way of considering effects with fast mesotrons.

Two experimental papers presented at the Conference were devoted to the study of the properties of mesotrons.

In the paper by V. I. Veksler and N. A. Dobrotin, “On Secondary Mesotrons,” the formation of secondary mesotrons by cosmic radiation at an altitude of \(4\,200\) m above sea level was studied. The dependence of the number of secondary mesotrons on the thickness of a filter placed between Geiger–Müller counters and proportional counters was measured. The decrease in the number of recorded mesotrons with increasing filter thickness showed that the recorded mesotrons are indeed secondary particles and not the ends of the ranges of fast particles of the penetrating component.

The measurements showed that the bulk of the secondary mesotrons have a total range of less than \(2\)—\(3\) g/cm\(^2\). The equilibrium intensity of the secondary mesotrons amounts to about \(1\%\) of the intensity of the hard component.

In the paper by S. Ya. Nikitin and N. O. Fedorenko (LPhTI, Academy of Sciences of the USSR), “On the Question of Mesotron Decay,” the half-life of mesotrons was measured.

With the aid of an arrangement of three Geiger–Müller counters, operating by the coincidence method, the vertical intensity of mesotrons at sea level was compared with the intensity at altitude and at an angle determined by \(\theta\) with the vertical. Measurements of the intensity of mesotrons were made at an altitude of \(3\,000\) m. The ratio of the intensity of mesotrons at this altitude (at an angle of \(45^\circ\) to the vertical) to the vertical intensity at sea level proved to be equal to \(0.81\).

Taking the mean energy of the mesotrons to be \(2.8\cdot 10^9\) eV and assuming that the formation of the greater part of the mesotrons occurs at an altitude corresponding to \(1/10\) of the entire atmosphere, from the observed ratio of intensities one can find the lifetime of the mesotrons. It turns out to be equal to
\[ \tau=(2.8\pm0.4)\cdot10^{-6}\ \text{sec}. \]

Measurements of the ratio of the intensities of the soft and hard components at different angles, made in this same work, showed that this ratio remains constant. This means that at an altitude of \(3\,000\) m the soft component is in equilibrium with the hard one. The authors also measured the increase in the intensity of the soft and hard components with altitude from sea level to \(3\,000\) m. In this case it proved that the relative increase in the intensity of the soft component in comparison with the hard component was \(1.37\). This increase in the relative intensity of the soft component, in the opinion of the authors, can be attributed entirely to an increase in the number of decay electrons with altitude.

Assuming that the number of decay electrons increases with altitude inversely proportional to the pressure, one can from these data estimate the lifetime of the mesotron, which proves to be equal to
\[ \tau=(2\pm0.4)\cdot10^{-6}\ \text{sec}. \]

Both values of \(\tau\), obtained independently, are in good agreement with one another.

In the discussion of the work of Nikitin and Fedorenko, D. V. Skobeltsyn expressed doubt as to the validity of the results of the authors named, pointing out that their measurements contradict the literature data and the preliminary experiments of Skobeltsyn himself and his collaborators. In turn, A. I. Alikhanov pointed out that the experiments of Skobeltsyn and his collaborators are not sufficiently clean and that no conclusions of any kind can be drawn from them, in particular conclusions about the absence of decay electrons in the proper amount. As a result of the discussion, the Conference did not arrive at a unanimous conclusion.

Undoubtedly, however, if the measurements of Nikitin and Fedorenko, indicating the existence of equilibrium between the soft and hard components, are sufficiently accurate, then they bring great clarity to the question

about the passage of cosmic rays through the atmosphere. Namely, in this case all phenomena connected with the passage of cosmic rays through the atmosphere fit into the following general scheme: primary particles arriving from world space, mainly electrons and positrons and $\gamma$-rays, form in the upper layers of the atmosphere mesotrons with high energy and are absorbed by means of the shower mechanism. The mesotrons, in turn, passing through the atmosphere, create along their path $\delta$-electrons, i.e., fast electrons knocked out of atoms directly by the impact of mesotrons, and spontaneously decay into electrons and neutrinos.

The soft component of cosmic rays at sea level is entirely of tertiary origin and is in equilibrium with the hard component.

A number of papers reported at the conference were devoted to investigations of showers caused by cosmic rays.

In the paper by O. N. Vavilov, S. N. Vernov, and N. L. Grigorov, measurements were made of the transition effect in lead, taking account of electron scattering.

The measurements were carried out by means of a thin-walled ionization chamber surrounded on all sides by lead, at an altitude of 4,200 m and in a substratostat rising to an altitude of up to 9 km. It turned out that, when the scattered radiation is taken into account, the transition curve has a maximum exactly coinciding with the maximum predicted by cascade theory. The ionization at the maximum proved to be twice as great as the ionization caused by the soft component in air at the same altitude.

The measurements showed that the scattering of the soft component in lead is very large. For example, the introduction of a layer of lead beneath the ionization chamber showed that about 40% of the soft component is scattered from below upward.

This result is also in complete agreement with the cascade theory of showers, as was indicated by Landau in his report “On the cascade theory of showers.”

In his report L. D. Landau described a number of new results obtained by him in the shower theory of cascades. It became clear that the mean angles of scattering of particles in a shower are very large already at electron energies $E \sim 30$ kV and depend almost not at all on the substance in which the shower propagates. Calculations also showed that the horizontal spread of a shower is extremely large and amounts, for example, in air to about 250 m.

High appraisal at the Conference was given to the work of Spivak (LFTI), who investigated the transition effect in lead under large thicknesses. A quite perfected technique was employed for eliminating the harmful background caused by showers formed at the edges of a lead shield (by particles that do not pass through the whole thickness of the shield, but only part of it), and strongly exaggerating the number of recorded showers.

In Spivak’s experiments, protective counters were installed along the edge of the lead shield and were connected in an “anticoincidence” circuit, so that showers passing through these counters were not recorded at all.

As a result of measurements made at sea level and at an altitude of 2,900 m, it was shown that the increase in the number of showers above large layers of lead with altitude corresponds to the increase in the intensity of the hard component, and that these showers are produced by penetrating ionizing particles.

As regards the general situation now existing in the physics of cosmic rays, it must be acknowledged that in this field experiment still cannot answer a number of questions posed by theory and is in a far from brilliant state. There exists a large number of experimental works contradicting one another, and the methodology of many works is still very imperfect. At the present time, the most urgent question is improvement of the measurement technique and the obtaining of cleaner results, not distorted by the many interferences encountered in this field.

A considerable number of the papers reported at the Conference were devoted to the study of the properties of fast electrons and hard rays.

A large part of these works had as its aim the general clarification of the real nature of those discrepancies between the theory of scattering of fast particles and ex-

…by experiment, the existence of which had been indicated by many authors and which had repeatedly been discussed at previous conferences on the nucleus.

In the work of Petukhov and Vyshinsky, the scattering of electrons of comparatively low energies (40–120 KeV) in thin aluminum films was investigated. The thickness of the films ranged from \(5\cdot 10^{-6}\) to \(3\cdot 10^{-5}\) cm.

Electrons, focused by a longitudinal homogeneous magnetic field, were passed through the films, and scattering through large angles \((119^\circ\text{—}122^\circ)\) was studied. The accuracy of the measurements was so great that the effective scattering cross sections were measured with an accuracy up to \(10\%\). At the same time it turned out that the measured scattering cross sections are in very good agreement with the theoretical values of the effective cross sections obtained by Mott in the nonrelativistic region of velocities.

The experiments of Petukhov and Vyshinsky make it possible to state quite unambiguously that no anomalous inelastic scattering of electrons of these energies by nuclei, of which various authors had repeatedly reported, exists.

A. I. Alikhanov, A. I. Alikhanyan, and A. O. Weisenberg investigated single scattering of monochromatic beams of electrons by nuclei through large angles in the relativistic region of velocities. The energy of the electrons was 600–1200 kV. They were scattered in very thin films of Al, Ni, Ag, and Au. Scattering was considered single at such film thicknesses for which a direct proportionality was observed between the scattering intensity and the thickness of the plate (i.e., the number of scattering atoms).

For electrons of energy 1 MeV this occurred at thicknesses of the order of \(3\text{—}6\ \mathrm{mg}/\mathrm{cm}^2\) Al and \(1\text{—}2\ \mathrm{mg}/\mathrm{cm}^2\) Ag.

If one takes as the unit of cross section the scattering cross section in Al, where according to all data one may expect good agreement between theory and experiment, then for Ni and Ag the cross sections are in good agreement with Mott’s formula in the relativistic region. However, for Au the experimental cross section proves to be somewhat smaller than is required by Mott’s exact formula obtained by numerical calculation. The course of the scattering intensity with energy is also in agreement with Mott’s theory.

L. A. Kul’chitskii and G. D. Latyshev observed multiple scattering of fast electrons in Al and Pb plates. The electron energy was 2000 kV. The obtained curves of the angular distribution of the scattered electrons were compared with the theoretical distribution of electrons over angles in the case of multiple scattering, obtained by Williams. For the case of Al, good agreement with theory was obtained; for the case of lead, however, a somewhat smaller value of the mean scattering angle was observed than follows from Williams’ theory.

L. V. Groshev (FIAN USSR) reported on his work “Formation of pairs by \(\gamma\)-rays in gases.” In this work the formation of pairs by \(\gamma\)-rays in nitrogen, krypton, and xenon was studied. The variation of the effective cross section as a function of the atomic number of the substance in which pair production takes place was found. In addition, the mean angles of emission of the electron and positron with respect to the direction of flight of the \(\gamma\)-quantum were studied. The course of all these quantities as a function of the atomic number \(z\) proved to be in complete agreement with the exact theory of Heitler and Eger.

In other reported works as well, general agreement was observed between the experimental data and the predictions of theory. Thus, at the Conference it became clear that the predictions of theory are certainly precisely justified for \(\gamma\)-rays and nonrelativistic electrons. In the case of relativistic electrons, individual deviations from theory are observed. However, on the whole, the theoretical formulas are sufficiently well justified here too. Therefore one may apparently think that the observed slight discrepancies between experimental data and theory are due to secondary causes, and that in general the theory of scattering of both slow and fast electrons is in complete agreement with experiment.

CONGRESSES AND CONFERENCES

The phenomenon of nuclear isomerism was the subject of a review report by Rusinov, published in this issue of Uspekhi fizicheskikh nauk1.

G. Zavelevich reported on the results of calculations he had carried out of the relative probability of internal conversion in the \(K\)- and \(L\)-shells for radiation of various multipolarities. The calculations were made in the nonrelativistic approximation. Magnetic dipole radiation was also not considered. In particular, for bromine \((Z = 35)\), at an energy of the converted \(\gamma\)-quanta \(h\nu = 49\ \mathrm{KeV}\), the ratio of the number of electrons ejected from the \(L\)-shell to the number of electrons ejected from the \(K\)-shell,

\[ \frac{N_L}{N_K}, \]

is \(8,\ 17,\ 52,\ 160,\) and \(310\%\), for dipole, quadrupole, octupole, etc. radiation. The experimental results obtained by Rusinov and Iozefovich indicate the octupole character of the radiation of Br nuclei.

A review report on uranium fission was given by I. V. Kurchatov. Its contents are presented in the present issue2.

G. N. Flerov and K. A. Petrzhak reported on the spontaneous fission of uranium discovered by them3.

Experiments on the fission of uranium by \(\gamma\)-rays showed that, in order for fission to take place, a comparatively small excitation energy is necessary (fission is caused by \(\gamma\)-quanta with an energy of \(6\ \mathrm{MV}\)). This circumstance led to the supposition that spontaneous fission of uranium nuclei exists.

Libby’s measurements showed that, if such spontaneous decay does exist, then the half-life is \(T > 10^{14}\) years. The apparatus available to Libby did not permit the registration of such rare processes.

Flerov and Petrzhak constructed an ionization chamber with an area of \(1{,}000\ \mathrm{cm}^2\), which was coated with uranium oxide \(\mathrm{U_3O_8}\). The increase in area was achieved by making the chamber multi-layered. On the outside the chamber was covered with a layer of cadmium, which was intended to absorb the cosmic neutrons present in the air. The ionization chamber was connected to an amplifier.

Since fragments from the fission of a uranium nucleus produced large pulses in the amplifier, whereas \(\alpha\)-particles gave a background of weak pulses.

When the chamber was placed under sufficiently clean conditions, the amplifier recorded six large pulses per minute.

All possible causes of the occurrence of large pulses were analyzed and eliminated: gas amplification of the ionization of \(\alpha\)-particles; the overlap of ionization pulses from several \(\alpha\)-particles; reactions caused by \(\alpha\)-particles and having neutrons as their products. These were ruled out by the fact that, when the chamber was filled with radon—an obvious source of \(\alpha\)-particles—the number of large pulses did not increase.

Cosmic neutrons were excluded by the cadmium shielding. Ionization pulses of Hoffmann in the chamber were eliminated by the fact that in an empty chamber coated with a layer of Cd they were absent. Thus it was established that the observed pulses were due to fission of uranium nuclei. To confirm this supposition, a neutron source was brought up to the chamber, and the distribution of the number of pulses in the amplifier caused by fission was observed according to their magnitude. Comparison showed that this distribution coincides with that which was observed without the neutron source.

The observed fission of uranium nuclei could have a twofold origin: either it could be caused by cosmic particles, for example mesotrons absorbed by nuclei, or it was spontaneous fission. Experiments with the chamber, carried out at a depth equivalent to \(100\ \mathrm{m}\ \mathrm{H_2O}\) underground, made it possible finally to settle the question in favor of spontaneous fission. Namely, although at such a depth the intensity of cosmic rays decreased by several tens of times, the intensity of the large pulses did not decrease.

If spontaneous decay is ascribed to the uranium isotope \(U_{92}^{238}\), then its lifetime \(T\) turns out to be

\[ T=(4\pm 1)\cdot 10^{16}\ \text{years}. \]

Similar experiments were also carried out with Th; however, spontaneous fission of thorium nuclei was not observed. This shows that for thorium \(T>5\cdot 10^{18}\) years.

V. Berestetskii and A. B. Migdal subjected to criticism the quantitative theory of fission of heavy nuclei developed by N. Bohr and Wheeler. The initial assumption of the Bohr–Wheeler theory is the supposition that the deformations of the nucleus that lead to its fission are small deformations. However, as Berestetskii and Migdal emphasized, this assumption is not fulfilled for actually existing heavy nuclei.

This leads, for example, to the fact that applying the Bohr–Wheeler theory to the phenomenon of spontaneous fission gives a lifetime \(\tau \sim 10^3\) sec, which is plainly in contradiction with experiment. In reality, the deformations leading to the fission of real heavy nuclei are large. Therefore a quantitative theory of nuclear fission cannot at present be constructed.

T. A. Goloborodko and A. I. Leipunskii reported on their investigation of the scattering of monochromatic neutrons by various nuclei. The neutron energies were 130, 220, and about 900 kV. In the experiment the total scattering cross section was measured. For the fastest neutrons, with energy 900 eV, the total cross section also included the inelastic scattering cross section. A large number of nuclei were studied. In a number of light nuclei new resonance levels were found. For example, in Mg a resonance level was found at an energy of 130 kV, in Si—at 200 kV, and so on. The measurements showed that for heavy nuclei the total cross section falls with energy.

Of great interest is the fact that the measured effective cross sections for the scattering of fast neutrons in some cases turned out to be larger than the scattering cross sections of thermal neutrons. This fact does not fit within the framework of the Bohr liquid-drop theory of the nucleus.

A. I. Leipunskii also drew attention to another fact that contradicts Bohr’s theory. Namely, it turns out that the cross sections of reactions involving neutron capture with subsequent emission of \(\gamma\)-rays vary chaotically from element to element. In some elements, such as bismuth, for example, neutron capture is not observed at all.

At the session of the Conference devoted to the question of stellar energy, a review report by I. Ya. Pomeranchuk was heard. At present one can point to a number of nuclear processes capable of serving as sources of stellar energy. Such sources are, first, a large number of exothermic nuclear reactions with light nuclei and, second, the mechanism of neutron-core formation proposed by L. Landau. Therefore it may be asserted that the fundamental possibility of liberating, as a result of nuclear reactions, an amount of energy sufficient for this purpose has been established.

However, at present we are still very far from a factual resolution of the question of the internal structure of stars and of the sources of stellar energy. In studying the physical conditions in stars one has to reckon with a number of difficulties. First among these difficulties is the difficulty connected with the question of the stability of stars of sufficiently large mass: as L. Landau showed, if the mass of a star is sufficiently large, it must continuously contract (if the star were at absolute-zero temperature, the critical mass after which the star would have to contract continuously would be \(3/2\) the mass of the Sun). At present the reasons that prevent a star from contracting to negligibly small dimensions are unknown. The second fundamental difficulty encountered in studying the question of the sources of stellar energy is that at present the physical conditions inside stars—the distribution of temperatures, densities, pressures, etc.—are completely unknown. However, the very mechanism of the source of stellar energy may depend to the greatest degree on these conditions,

Thus, for example, the energy output of nuclear reactions depends exponentially on the temperature prevailing in the region in which the reaction takes place. The character of nuclear reactions may also depend to the greatest degree on the percentage content of various nuclei in the region where the reaction is taking place, on the conditions of transport of matter to this region, etc. A change in all these factors leads to entirely different sources of stellar energy.

The situation is made still more complicated by the possibility of the formation of neutron nuclei in a star. It may therefore be asserted that the solution of the question of the sources of stellar reactions comes up against the necessity of clarifying the physical conditions inside stars.

Only after it proves possible to determine the physical conditions inside stars will a complete solution become possible to the question of the mechanism by which stellar energy is released. At present, however, this question remains open.

By assigning to the temperatures inside stars certain, to a considerable extent arbitrary, values, one may indicate a number of nuclear reactions that could serve as sources of stellar energy, at least in stars of the main sequence. However, in the present state of affairs it is impossible to settle on any one of them. This will become possible only in connection with the clarification of the general physical conditions inside stars.

One of the sessions of the conference was devoted to applications of nuclear physics.

A major report on the application of artificial radioactive isotopes in biology and medicine was delivered by G. M. Frank. G. M. Frank’s report is printed in this issue of Uspekhi fizicheskikh nauk.1

A. E. Braunstein delivered a survey report on the application of stable isotopes in biochemistry. The use of artificial radioactive nuclei as indicators of metabolism in the organism is limited by the fact that there are no sufficiently convenient radioactive isotopes of the elements most important in biochemical processes—hydrogen, carbon, and oxygen. Therefore, in studying the processes of metabolism in the organism, the method of stable isotopes is of great importance. Stable isotopes of the indicated elements are introduced into various organic compounds. For example, deuterium is introduced in place of hydrogen, etc. Atoms labeled in this way serve for the study of metabolism in the organism. Despite the fact that the stable-isotope method has lower sensitivity in comparison with the radioactive-isotope method, it will undoubtedly be of great importance in biochemistry.

The concluding session of the Conference was devoted to the technique of producing fast particles.

V. G. Levich, Moscow

1. See p. 179. ↩↩↩

2. See p. 159. ↩

3. See p. 171. ↩