CONGRESSES AND CONFERENCES

III ALL-UNION CONFERENCE ON THE ATOMIC NUCLEUS

On October 1–6, 1938, the III All-Union Conference on the Atomic Nucleus, convened by the Division of Mathematical and Natural Sciences of the Academy of Sciences of the USSR, took place in Leningrad. At the conference 29 reports were heard. The topics of the conference were devoted to the following main questions: 1) cosmic rays, 2) the passage of fast particles through matter, 3) the theory of new particles, and 4) the properties of heavy particles and the structure of nuclei.

The first three sessions of the conference were devoted to work in the field of cosmic rays. Research in recent years has established that cosmic rays contain two components: a penetrating component and a soft one. The latter has now been studied comparatively fully, although even here complete clarity is not always available. It has been established that the soft component consists, basically, of electrons and positrons. In addition, it includes a known fraction of $\gamma$-rays. Showers produced by the soft component have also received, at least qualitatively, an explanation in the so-called cascade theory of showers. As for the properties and composition of the penetrating component, here the data are more meager, and therefore their study is of the greatest interest.

Studies of the properties and composition of the penetrating component of cosmic rays were the subject of the work of V. I. Veksler and his collaborators, on which V. I. Veksler reported at the first session. In their experiments the authors used an original method developed by Veksler, consisting in the use of so-called proportional counters, which make it possible to measure not only the number of particles passing through the counters, but also the ionization produced by these particles. A series of experiments was carried out at sea level and on Elbrus, making it possible, from the range in matter and the ionization, to judge that the penetrating component consists of particles with a mass of the order of 200 electron masses and with a charge equal in absolute value to the charge of the electron of both signs. In this way it was proved once more that the penetrating component consists of semi-heavy particles, in agreement with the hypothesis of Anderson and Neddermeyer. Let us note, incidentally, that recently several names have been proposed for the new particles. The most commonly used are “baritron” and “mesotron.” In addition, these experiments established that mesotrons are not primary particles. This conclusion is of very substantial significance. It means that there exists some, as yet unknown, mechanism for the formation of mesotrons by light particles (electrons and positrons). This may be considered established with the same degree of certainty as the very fact of the existence of mesotrons.

S. N. Vernov described his measurements of the intensity of cosmic rays in the stratosphere at various latitudes. The measurements were carried out at the latitudes of Leningrad, Yerevan, and near the magnetic equator, in the region of the Indian Ocean. The study of the intensity of cosmic rays in the stratosphere was made with Geiger–Müller counters raised on sounding balloons. The data from the counters were transmitted by ...

radio. The discharges in the counters were amplified and set in motion a relay, as a result of which a radio signal was transmitted. It should be noted that, when working at low latitudes, one encountered all the essential experimental difficulties. The measurements established the presence of a large latitude effect in the stratosphere, considerably exceeding the latitude effect at sea level. This makes it possible to draw a conclusion concerning the composition of the primary radiation. Namely, the character of the latitude effect at a great altitude shows that no fewer than 75% of the primary rays must be electrons. If one uses the formulas of the cascade theory of showers and compares them with the experimental data, it turns out that charged light particles constitute about 90% of the total primary radiation. In addition, Vernov’s experiments make it possible to judge the absorption of cosmic particles. The data obtained are in qualitative agreement with the cascade theory of showers, especially good for great altitudes.

N. A. Dobrotin made a preliminary communication on very interesting experiments investigating the so-called reverse showers, which were carried out by a group of FIAN collaborators on Elbrus. The phenomenon of reverse showers was discovered by Rossi comparatively recently. It consists in the fact that the number of counter discharges increases not only when lead is placed above it, but also beneath it; i.e., it turns out that there exist showers directed not from top to bottom, but from bottom to top.

The phenomenon of reverse showers appears very puzzling because, according to the cascade theory, all particles of a shower must possess a sharp directionality along the direction of flight of the particle that caused the shower (here the word particle applies equally to photons). Thus, to explain the appearance of particles moving from bottom to top on the basis of the cascade theory of showers is impossible. Nor can one suppose that the reverse shower is created by a $\gamma$-quantum produced as a result of positron annihilation in the substance beneath the counter, since this would lead to an incorrect dependence on $z$ of the substance beneath the counter.

The phenomenon of reverse showers has been investigated by a number of authors, and in complete agreement with one another they assert that the number of reverse showers depends not only on the thickness of the plate beneath the counters, but also on the thickness of the plate placed simultaneously above the counters. Namely, in all authors the number of reverse showers, when a lead plate was placed above the counters, increased by a factor of 2.5–6. The measurements reported by Dobrotin showed the directly opposite result—the number of reverse showers decreased when material was placed above the counter. This discrepancy cannot be attributed either to the altitude (Gilbert’s experiments were conducted at the same altitude of 3,000 m), or to geometrical conditions, on which the effect does not strongly depend. The only path to resolving the contradiction, in Dobrotin’s opinion, consists in accepting the existence of two types of reverse showers—showers consisting of a large number of particles, and of a small number, produced by some different mechanisms. Owing to the low sensitivity of the apparatus used in the experiments described, mainly reverse showers of the first modification were measured. Other authors, however, measured mainly showers consisting of a small number of particles. The authors also proved that reverse showers are caused by the soft component of cosmic rays. Although these data are of a preliminary character, the very fact of discovering a new modification of reverse showers is of great interest.

The reports of Ivanova and Frank-Kamenetsky were devoted to the study of the soft component of cosmic rays near the earth’s surface. Namely, the question investigated was what part of the soft component at the earth’s surface is secondary, i.e. produced by the penetrating component. The soft component was cut off by a layer of lead 10 cm thick, after which the coincidence method was used to study the amount of soft component produced by the penetrating component in thick blocks of substances of various atomic number. It turned out that in equilibrium

intensity of the soft radiation under a block almost coincides with the intensity of the soft component at sea level. It follows from this that the greater part of the soft component at sea level is secondary. In addition, it was established that the equilibrium intensity of the soft component in light elements is greater than in heavy ones. Namely, the equilibrium intensity of the soft component in air is 33%, in Al 11%, and in Pb only 4½%.

A. P. Zhdanov reported on the preliminary results of experiments carried out by him jointly with Gurevich and Filippov, in which, with the aid of thick-layer photographic plates, the splitting of nuclei by cosmic rays was observed. The method was based on the fact that charged particles, passing through a photographic emulsion layer, leave traces in it. Although the data were preliminary in character, the new method proposed by the authors is undoubtedly of interest.

A large review report by I. E. Tamm was devoted to the currently existing theoretical ideas about mesotrons. Possible theoretical interpretations of mesotrons are based on the following experimental facts. 1) Mesotrons possess a mass intermediate between the mass of electrons and the mass of heavy particles; their charge, in absolute value, is equal to the charge of the electron and may be either positive or negative. 2) Mesotrons are formed in the upper layers of the atmosphere by light particles. 3) Mesotrons may in turn transform into electrons or positrons. 4) The penetrating component of cosmic rays, in passing through matter, is accompanied by an equilibrium amount of the soft component; at the same time part of the latter consists of so-called δ-electrons, i.e. electrons knocked out of atoms of the substance by a direct impact of a mesotron.

The question of the production and absorption of mesotrons is closely connected with nuclear forces and is of primary interest. Indeed, the forces of interaction leading to the production and absorption of mesotrons cannot have the character of electrical forces. If these were forces of electrical origin, then a dependence of the probability of mesotron formation on \(Z^2\) would be observed. One may cite a number of other considerations indicating the non-electrical character of the interaction of mesotrons with nuclei. It is therefore necessary to accept that the forces of interaction have an essentially quantum character and are some kind of variety of the so-called exchange forces. Then the question of the spin of mesotrons becomes very substantial.

If the spin of mesotrons is \(s = \tfrac12\), then their production singly is impossible, but only in pairs, so that the total spin is conserved or changes by an integer. On the contrary, if \(s = 0\) or 1, then the formation of single mesotrons is possible.

At present there are no experimental data at all on the spin of mesotrons, and one has to make use of one assumption or another as a certain working hypothesis. The assumption that \(s = \tfrac12\) was expressed by Anderson and Neddermeyer and developed by Tamm. The assumption of an integral spin (\(s = 0\) or 1) was advanced by Yukawa long before the discovery of mesotrons (in 1935) to explain nuclear forces; he proposed introducing a hypothetical particle with a mass of the order of one hundred electron masses and with integral spin.

Adopting one or another value of the spin, one can make fairly definite judgments about the nature of the forces of interaction of mesotrons with heavy particles. This makes it possible, in turn, to construct a theory of the interaction between heavy particles in nuclei. In the case of half-integral spin, the theory is built on the model of the Fermi theory of β-decay. According to this theory, in the field of the nucleus it is possible for a mesotron to transform into an electron and conversely—for an electron to transform into a mesotron. We emphasize that everything said applies equally to positrons as well. Thus it follows from the theory that mesotrons can be created by fast electrons in the upper layers of the atmosphere. The effective cross section for the formation of mesotrons has an acceptable order of magnitude (\(\sim 10^{-25}\)). Further

Proceeding from this theory, one can construct a theory of nuclear forces of interaction between heavy particles. Here, however, substantial difficulties arise, connected with the fact that the forces obtained do not have the character of saturation forces, as is the case in reality.

A large number of theoretical works have recently been devoted to mesotrons with integer spin (Heitler, Fröhlich, Bhabha, Kemmer). Mesotrons with integer spin are described by a wave function having a vector character (i.e., having three components). This wave function satisfies a system of equations very similar to Maxwell’s equations. The reason for this is easy to understand, if one recalls that, according to quantum mechanics, a collection of particles obeying Bose–Einstein statistics, i.e., possessing integer spin, is equivalent to a certain wave field. In particular, from the formal point of view the description of mesotrons is very similar to the description of photons. The theory gives nuclear forces of the correct sign and order of magnitude, acting between a proton and a neutron. However, the interaction forces between identical heavy particles (proton—proton or neutron—neutron) are obtained from this theory only in higher approximations, which in itself is bad, since successive approximations converge poorly. In addition, forces of repulsion are obtained between identical particles, which contradicts experiment. To describe the interaction forces between identical particles one has to introduce a new hypothetical particle, the “neutretto,” representing a mesotron without charge, which is a very weak point in the theory. However, alongside these shortcomings the theory also has a number of virtues. In particular, one can compute the probability of various processes accompanied by the appearance of mesotrons, such as, for example,

$$ h\nu + {}^{1}_{1}u \to {}^{1}_{1}p + M^{-}, $$

where \(M^{-}\) denotes a negative mesotron. An acceptable effective cross section is obtained, although the angular distribution of the mesotrons turns out to be incorrect.

Thus each of the theories has its own advantages and shortcomings, and at present one cannot give preference to any one of them.

It is necessary to note that the application of the laws of quantum mechanics to mesotrons with energies greater than \(10^8\ \mathrm{V}\) is already illegitimate. As Heisenberg has shown, the behavior of mesotrons in the region of high energies may not obey the laws of quantum mechanics, the limits of applicability of which lie, for mesotrons, at energies of the order of \(10^8\ \mathrm{V}\). This is all the more essential since the greater part of mesotrons possess energies of the order of \(10^9\ \mathrm{V}\)^[Already after the close of the conference, a very important paper by Euler and Heisenberg was published, in which it is pointed out that mesotrons may be unstable, capable of decaying into electrons and neutrinos. The lifetime of a mesotron is the greater, the greater its velocity. In passing through matter, the mesotron loses energy by ionization, which leads to its slowing down and to an increase in the probability of decay. In this, the time during which the mesotron moves in matter plays an essential role. Thus, for example, on the path from the upper layers of the atmosphere to the surface of the earth, a considerable fraction of mesotrons will have time to decay, and at the surface of the earth there is a considerable number of secondary electrons. Conversely, in passing through an equivalent layer of lead, only a small fraction of mesotrons will have time to decay.].

A. K. Walter reported to the meeting on work carried out by him jointly with K. D. Sinelnikov concerning the measurement of the energy losses of electrons for ionization when passing through matter. From a Van de Graaff generator, by means of suitable filtration, a beam of monochromatic electrons with energy up to \(2 \tfrac{1}{2}\ \mathrm{MV}\) was obtained at a current of \(80\)—\(100\ \mathrm{mA}\). The electron beam was passed through samples of the substances under investigation of various thicknesses, and a series of measurements was made at different electron energies in the beam. The construction of the apparatus made it possible to—

CONGRESSES AND CONFERENCES

to determine the energy losses indirectly, from the ranges of electrons in matter. It turned out that the observed energy losses are in good agreement with the quantum-mechanical formula for energy losses. For light elements (lithium, carbon), for heavier ones (aluminum and copper) rather considerable discrepancies are already observed, and in the case of lead the discrepancy with the theoretical data is very significant.

The work of A. K. Walter and K. D. Sinelnikov was criticized by a number of persons, who pointed out that, owing to the great thicknesses, multiple scattering inevitably plays a substantial role. An electron, especially in heavy elements, traverses a much longer path than is observed directly. Since, however, no exact theory of multiple scattering exists, a direct comparison of the experimental results with a theory that does not take multiple scattering into account is illegitimate.

The same question of the energy losses of electrons in passing through matter was the subject of the works of Alikhanov and Alikhanian, reported to the conference. However, they used a completely different method, setting themselves the task of excluding as far as possible the influence of multiple scattering. They used monochromatic electrons from a radioactive Th (B + C) preparation with an energy of 2.6 MV, obtained as the result of internal conversion of γ-rays. These electrons were passed through very thin plates of aluminum and lead. The method made it possible to measure the number of electrons that had lost a given amount of energy and, from a series of measurements, to construct the curve of energy losses per unit path length. The curves obtained proved to be in agreement with the theoretical formulae of Bloch and Bethe–Heitler for aluminum within the limits of experimental accuracy. In the case of lead, too, the results proved, in a first approximation, to be in agreement with theory, but differed sharply from the data of most other authors. Apparently this discrepancy should be attributed to the fact that the other experimenters worked with rather thick plates, in which multiple scattering played a substantial role.

In the discussion that developed concerning losses, opinions were divided in the assessment of the work of a number of American experimenters (Crane, Laslett, and others), who had obtained loss values exceeding the theoretical ones. By the end of the discussion no agreement had been reached, but the opinion of the majority tended to be that the more exact value of the losses in these experiments had been due to multiple scattering.

In connection with this same question of the nature of the energy losses of fast electrons in passing through matter, the experiments of Artsimovich and Khramov were presented; they investigated that part of the losses which is due to bremsstrahlung radiation. For heavy elements, at energies of about 10 kV, the energy losses to radiation amount to 10–15% of the ionization losses. At still higher energies practically all the energy losses are connected with bremsstrahlung. In the experiments of Artsimovich and Khramov, very monochromatic beams of electrons were obtained with the aid of a magnetic lens, which separated out a monochromatic beam from the fast electrons of a radioactive preparation. These monochromatic electron beams were passed through plates of copper and aluminum. The observed energy losses to radiation proved to be in very good agreement with the Bethe–Heitler theory, both as a function of energy and as a function of atomic number.

D. V. Skobeltsyn delivered a report at the conference on normal and anomalous δ-electrons. As mentioned above, δ-electrons are the name given to fast electrons knocked out of the atoms of a substance. D. V. Skobeltsyn, observing in a Wilson chamber the tracks of δ-electrons with energies of 0.1–2 MV, checked the fulfillment of the law of conservation of momentum in collisions of δ-electrons with the electrons of matter and the validity of Møller’s formula for the scattering of electrons in the region of relativistic velocities. It turned out that in the majority of cases no anomalies were observed, and the theory proved to be in good agreement with experiment. However, bes—

CONGRESSES AND CONFERENCES

In addition to such normal δ-electrons, δ-electrons were discovered whose behavior is utterly puzzling and unexpected. Namely, in light gases relative to matter, such as neon and argon, δ-electrons were observed with an anomalously large energy loss, without any visible causes. In some cases the energy losses were so great that the tracks suddenly curled up without branching. In other cases anomalous forks were observed, in which the opening angle did not depend on the magnitude of the transmitted impulse. Thus a sharp divergence from theory was observed. The phenomenon of anomalous losses, discovered by D. V. Skobeltsyn and his collaborators, is especially unexpected because in these regions of energy the laws of quantum mechanics would seem to have had to be obeyed very well. At present it is difficult to say anything about the nature of these anomalous losses, since the number of observations is insufficient, although they have already been noted by many experimenters. It is only certain that the probability of these losses, if it does increase with the atomic number \(z\) of the substance, does not do so very rapidly.

I. E. Tamm reported to the conference on the calculations he had carried out of the effect of so-called isotope shift, i.e., the relative displacement of spectral lines in different isotopes of the same substance. One of the causes of isotope shift is the change in the volume of the nucleus with an increase in the number of neutrons in it. The change in volume entails a change in the potential acting on the external electron, which leads to the appearance of isotope shift. Breit’s calculations show that the spectral term of the heavier isotope should lie higher, which is indeed observed in heavy elements. In light elements, however, the directly opposite picture is observed.

I. E. Tamm, allowing for the existence of new weak attractive forces between neutrons and electrons at short distances, calculated the isotope shift as the result of superposing the effect of growth of the nuclear radius and the effect caused by attraction between neutrons and electrons. The formula obtained qualitatively correctly describes the character of the isotope shift both for light and for heavy elements. To speak quantitatively, however, about this region of phenomena, in which intranuclear forces play a role, as the speaker emphasized, is at present premature.

Of great interest is the remark made in his report by I. I. Gurevich concerning the character of the absorption of slow neutrons in nuclei. According to N. Bohr’s statistical theory of the nucleus, the nuclear levels in the region of thermal neutrons are spaced very densely, and, moreover, the number of levels in a unit energy interval must increase with increasing nuclear mass. Therefore one may expect that the probability of resonance absorption of slow neutrons should increase toward the heavier elements. I. I. Gurevich noted, however, the fact that strong absorption by neutrons is found among the group of rare earths, where a rather sharp maximum of the probability of absorption is observed. It is as yet impossible to explain this phenomenon.

M. A. Markov reported on his work “Inelastic scattering of quanta with the formation of pairs.” By Williams’s method the probability was calculated for the inelastic scattering of \(\gamma\)-quanta on nuclei with the formation of pairs. In this method all processes are considered in such a coordinate system in which the photon is at rest while the nucleus moves toward it with enormous velocity. Then the field of the nucleus can be represented in the form of a superposition of a system of plane waves, and the calculation of the effect is reduced to the probability of pair production by two photons. Although the effect calculated by Markov is a third-order effect, its probability turns out to be greater than the probability of second-order effects.

L. I. Rusinov reported on investigations of bromine isomers carried out by him jointly with Iuzefovich. As is known, bromine possesses three isomers with periods of 18 min., 4.2 hours, and 36 hours.

The authors discovered in one of the isomers of bromine a soft electronic radiation, which should be attributed to internal-conversion electrons. This fact may be regarded as some qualitative indication in favor of the theory of isomerism proposed by Weizsäcker. According to this theory, the occurrence of isomerism is connected with the fact that, upon excitation, a nucleus may enter a metastable state. From the metastable state it cannot pass into the normal state for a sufficiently long interval of time because of the selection rule; however, it can then pass into the normal state by transferring the excitation energy to one of the shell electrons. This phenomenon is apparently what is observed in the experiments of Rusinov and Iosefovich. From the energy of the electron one can estimate the order of magnitude of the excitation energy. It proves to be only 40 kV, and not several hundred thousand, as had been supposed earlier.

L. V. Groshev reported on the continuation of work on the study of the process of pair formation in light substances, carried out by him jointly with I. M. Frank. The process of pair formation by γ-rays in nitrogen was studied by the Wilson-chamber method. In contrast to krypton, where pair formation had been investigated earlier and where the measurement of the energy of both components of the pair was complicated by strong scattering, in nitrogen it was possible to measure the energy of the electron and positron rather accurately.

The small scatter of the experimental values of the energies of the components of the pair made it possible to conclude that the law of conservation of energy is fulfilled in pair formation. For the effective cross section of pair formation a value was obtained that coincides, within the limits of error, with the theoretical one. An unexpected distribution of energy between the positron and the electron was, however, observed. Since the latter is attracted to the nucleus while the positron is repelled, the energy of the electron should always be somewhat smaller. Calculations by Jaeger and others show that in nitrogen this difference should be of the order of 20 kV. The experimentally found value of the difference of the energies of the positron and the electron reaches 200 kV.

Alikhanov reported on the results of precise measurements of \(e/m\) for the β-electrons of RaC, the aim of which was to establish the presence in β-rays of particles with different masses.

The results of the experiments amount to the conclusion that the β-spectrum contains no particles with mass greater than that of the electron.

A. S. Papkov reported on his obtaining powerful ion currents with voltages of the order of hundreds of kilovolts, with the aid of which he proposes to obtain intense sources of neutrons. Namely, it is known that when deuterons of such energies collide with one another, a reaction occurs with the emission of neutrons. This very economical method of producing neutrons is of undoubted interest.

In addition to those mentioned above, the following reports were heard at the meeting: Khurgin, “On the theory of the cyclotron”; a communication by Rukavishnikov on the construction of the cyclotron of the Radium Institute; Latysheva, “On the spectrum of positrons of radium C”; Alikhanian and Nikitin, “The β-spectrum of RaC”; Walter, “On the boundary of the photoeffect of beryllium”; Meshcheryakov, “Absorption of slow neutrons”; Cherenkov, “The absolute yield and the energy distribution in the spectrum of the radiation of fast electrons moving in matter,” in which good agreement was reported between the theory of the radiation he had discovered and the experiments; Flerov, “Absorption of slow neutrons”; Lange, “Construction and operation of a pulse generator and a discharge tube at 4 million volts.”

At the end of the meeting, organizational questions connected with the arrangement of research on the atomic nucleus in the USSR were discussed.

V. Levich, Moscow