Fission of Heavy Nuclei1

I. V. Kurchatov, Leningrad

The problem of the fission of heavy nuclei and the related question of the possibility of carrying out a nuclear chain reaction were discussed in detail at the Conference on the Physics of the Atomic Nucleus in 1939 in Kharkov1. I shall therefore confine myself in my report to a presentation and analysis of the principal works that have been carried out during the intervening period[^2].

During the past year no substantially new theoretical ideas about the mechanism of fission have been published. The hypothesis of the instability of heavy nuclei with respect to changes in their shape, put forward earlier by Meitner and Frisch, still remains the basic explanation of the phenomenon of fission. However, a number of quantitative relations that followed from the calculations of Bohr and Wheeler, which proceeded from the same hypothesis, are now being called into question. These questions are dealt with in a separate report by Berestetsky and Migdal, and therefore I shall not dwell on them.

During the past year new possibilities have been discovered for inducing nuclear fission.

Gant[^3] carried out a series of preliminary experiments in which, using Joliot’s method—that is, by the radioactivity of the fragments—he observed the fission of uranium nuclei under the action of deuterons with energies of 8–9 MeV. Wells, Haxby, and others[^4] established the fission of uranium under the action of 6 MeV $\gamma$ rays produced in the disintegration of fluorine by protons. Finally, Petrzhak and Flerov discovered spontaneous fission of uranium nuclei.

In May 1940 Jentschke, Prankl, and Hernegger[^5] published a communication on the fission of thorium nuclei. They showed that thermal neutrons do not cause fission and that it is produced by fast neutrons from a $(d, d)$ source.

Thus, at present the phenomena of fission may be regarded as established for the following nuclei:

\[ {}_{92}\mathrm{U}^{235},\quad {}_{92}\mathrm{U}^{236},\quad {}_{92}\mathrm{U}^{238},\quad {}_{92}\mathrm{U}^{239},\quad {}_{90}\mathrm{Th}^{233},\quad {}_{90}\mathrm{Io}^{231},\quad {}_{91}\mathrm{Pa}^{232}. \]

During this year progress has been made on the question of the thresholds and cross sections of fission for some of the nuclei listed above. Tha—

thanks to the work of Nier et al.,6 who achieved the separation of uranium isotopes, it was established with certainty that thermal neutrons produce fission only of \( \mathrm{U}_{92}^{235} \). The experiments were carried out with quantities of \( \mathrm{U}^{238} \) and \( \mathrm{U}^{235} \) equal, at best, to 4 and 0.03 micrograms, respectively. Attempts were made in these same experiments to clarify the possibility of fission of \( \mathrm{U}^{234} \) by thermal neutrons, but owing to the smallness of the amounts obtained of this isotope and the closeness of its mass to the mass of \( \mathrm{U}^{235} \), this question did not receive a definitive solution.

A number of investigations were carried out with the aim of determining the minimum energy of neutrons causing fission of \( \mathrm{U}^{238} \) and \( \mathrm{Th}^{232} \). Petrzhak and Flerov,7 on the basis of experiments with photoneutrons of beryllium excited by \(\gamma\)-rays of radium and thorium and of their decay products, came to the conclusion that the limiting energy of neutrons causing fission of \( \mathrm{U}^{238} \) is of the order of 1 MeV. Haxby, Wells et al.,8 obtaining neutrons in the reaction \( \mathrm{Li}^{7}(p,n) \), found that the fission threshold of thorium lies at 1.1 MeV. Thus, from a comparison of these results it follows that the limiting neutron energies in the cases considered are identical. This conclusion, however, may prove to be incorrect, since the determination of the threshold depends strongly on the course of the cross section in the adjoining energy regions, on the sensitivity of the method, and on the number of neutrons in the source. It should be noted that matters are even worse with the determination of the fission thresholds of ionium and protactinium; for these elements it is known only that their fission is not caused by thermal neutrons and, in any case, occurs at neutron energies of 2 MeV.

The order of the thresholds could be established from the magnitudes of the fission cross sections for fast neutrons, if one adopts the Bohr and Wheeler interpretation for the energy dependence of the probabilities of inelastic scattering and fission. In their opinion, in regions not very close to the threshold, both probabilities vary with energy according to one and the same law, and since inelastic scattering is the only process that can practically compete with fission, the fission cross section must remain constant as the neutron energy changes. Its values will be the smaller, the higher the threshold, since inelastic scattering will already have a large probability at those energies at which fission is only just becoming possible. The fission cross sections of \( \mathrm{U}^{238} \), \( \mathrm{Pa}^{231} \), \( \mathrm{Th}^{232} \), and \( \mathrm{Io}^{230} \) by fast neutrons are respectively equal to \(5 \cdot 10^{-25}\), \(3 \cdot 10^{-24}\), \(1 \cdot 10^{-25}\), and \(3 \cdot 10^{-25}\ \mathrm{cm}^{2}\), and, consequently, the elements undergoing fission should, according to these considerations, be arranged in order of increasing thresholds as follows: \( \mathrm{U}^{235} \), \( \mathrm{Pa}^{231} \), \( \mathrm{U}^{238} \), \( \mathrm{Io}^{230} \), and \( \mathrm{Th}^{232} \).

The experimental material shows that at least part of the assertions of Bohr and Wheeler agrees with experiment, and in fact the fission cross sections of uranium and thorium in regions not adjacent to the threshold do not depend on the neutron energy. Ladenburg, Kanner, Barshall, and Van Voorhis9 obtained neutrons monochromatic in velocity in the \((d,d)\) reaction. Studying fission under the action of neutrons emitted at different angles with respect to the proto-

clei under incident deuterons, they showed that in the interval of neutron-energy variation from 2.1 to 3.1 MeV the fission cross sections of uranium and thorium remain constant. Ageno, Amaldi, Bocciarelli, and Trabacchi[^10] came to the same conclusion for the energy interval from 2 to 10 MeV, studying the fission of uranium and thorium under the action of neutrons produced in \((d, d)\), \((d, B)\), \((d, Be)\), and \((d, Li)\) reactions. In the same investigation it was shown that the fission cross sections of uranium begin to increase at neutron energies greater than 11 MeV.

According to Bohr, this increase is not connected with a change, at very high excitations, in the ratio of the fission probability to the probability of neutron emission by the \(U^{238}\) nucleus. After the neutron leaves the nucleus, it generally remains excited. It may happen that the excitation energy will be greater than the fission threshold; then, following the evaporation of a neutron from \(U^{239}\), fission of \(U^{238}\) will occur. The addition of this type of fission at large excitations is, according to Bohr, the cause of the increase in the cross section.

The second part of the assertions of Bohr and Wheeler, connecting the magnitude of the cross section with the threshold, has not yet been tested experimentally.

Let us now consider works on the study of the energy and the nature of the fragments arising in the fission of heavy nuclei. The most careful determinations of the energy were made by Kanner and Barschall[^11] with the aid of an ionization chamber connected to a linear amplifier, on one of whose electrodes uranium was deposited. They found in the distribution curve of pulses by magnitude two maxima at energies of 65 and 97 MeV. Kanner and Barschall, in addition, directly measured the total energy of the fragments by placing in the middle of the ionization chamber an aluminum foil onto which, by cathode sputtering, they deposited a very thin layer of metallic uranium. The total energy of the fragments proved to be on the average equal to 159 MeV, which agrees well with the sum of the separate energies of each fragment determined by them. This energy belongs to the most common type of fission into fragments with mass numbers of the order of 100 and 140. The half-width of the distribution, according to the data of Kanner and Barschall, is \(\sim 30\) MeV; the highest energy released in fission is 200 MeV. The calculation of the energy was made from the number of ions formed by the fragment, as usual under the assumption that the mean energy expended by the fragment in forming one ion is the same as for \(\alpha\)-particles. In view of the fact that in this way a certain arbitrariness is introduced, the work of Henderson[^12], who determined the energy of the fragments in the fission of uranium by a calorimetric method, acquires special value. In his experiments 13 g of metallic uranium were irradiated by an intense beam of slow neutrons from the cyclotron at Berkeley. The temperature of the uranium was measured with the aid of a resistance thermometer; the number of nuclear fissions was determined by special, simultaneously performed experiments with a thin layer of uranium in an ionization chamber. Henderson found that the mean energy of the fragments is 175 MeV. This value is greater than the kinetic energy

fragments, since in the mass of uranium and in the surrounding copper shield a considerable part of the soft radiations accompanying the β-decay of the fragments was absorbed.

Physical methods of investigation, as we have seen, give some indications as to the nature of the fragments, but radiochemical investigations are of decisive importance here. In this field a large number of works have been carried out, which to a considerable extent have clarified and extended the results of last year’s investigations.

In most cases the charge of an element was determined on the basis of its chemical properties; for iodine, tellurium, and antimony this was done from the characteristic $K$-rays of the X-ray spectrum. Studies of the change of activity with time after the corresponding chemical separations made it possible to establish with certainty several chains of successive β-transformations. The decay periods and the properties of the radioactive radiation of some fragments coincided with the same quantities for radioactive nuclei obtained by irradiation, with deuterons, protons, and neutrons, of elements in this part of the periodic system. This made it possible to determine the mass numbers of such fragments. The material now known on the chains of successive β-transformations of fragments obtained in the fission of uranium is collected in Table 11.

It must be said that, despite the great labor invested in the chemical investigations of radioactive fragments, the material obtained does not provide substantial data for analysis of the fission process. At present, strictly speaking, it is impossible to establish a single branch for which one could confidently indicate the charge of the initial fragments and their mass numbers. Meanwhile, if this were known, it would be possible by an independent method—by the difference between the mass number of the compound uranium nucleus and the sum of the mass numbers of the initial fragments of the branch—to judge the number of neutrons accompanying nuclear fission.

It is curious to note that the ratio of the number of neutrons and protons in the initial light and heavy fragments turns out to be different. Whereas in the heavy fragments this ratio is very close to that characteristic of uranium, in the light fragments it is considerably smaller. As should be expected, the nuclear substance has time to redistribute itself in uranium before the actual process of fission occurs. Several cases are already known in which light fragments are very close in their mass to stable nuclei. In connection with this, I. P. Selinov expressed the supposition of the possibility of uranium fission with the formation of a stable light fragment.

At the previous conference, in connection with the works of Petrzhak, Eintsche, and Frankl, the question was discussed of the possibility of the formation of different kinds of fragments in the interaction of uranium with slow and fast neutrons. From the distribution of ionization pulses by magnitude it followed that, under the action of fast neutrons, alongside asym-

metrical fission, fragments close in mass are formed. In some agreement with this conclusion are the experiments of Japanese investigators \(^{13}\), indicating the formation of radioactive silver and cadmium (see Table 1). According to their data these nuclei arise only under the action of fast neutrons.

Table 1

In concluding the review of work on fission phenomena, it is also necessary to make several remarks about the neutron radiation associated with these reactions.

Our knowledge of the neutrons emitted in the fission of uranium has now become more reliable, chiefly thanks to the valuable work of Zinn and Szilard. They registered neutrons by the recoil atoms of hydrogen and helium in an ionization chamber, using as the source not very fast neutrons (upper energy limit 130 KeV), produced by the action of the γ-rays of radium on beryllium. From the effect due to the neutrons arising in fission, it was easy to separate the effect caused by the slower neutrons of the source.

Zinn and Szilard established that, for each act of fission caused by a thermal neutron, 2.3 neutrons with energies from 1 to 3 MeV are emitted. The method adopted in their work did not make it possible to say anything about the interval of time separating the emission of neutrons from the moment of fission of the nucleus, but a number of other experiments show beyond doubt that the bulk of the neutrons observed by Zinn and Szilard cannot be connected with processes of delayed emission accompanying the β-decay of the fragments.

It has now become known that delayed emission of neutrons occurs not only for β-decay with a half-period of 12.5 sec. Boos, Dunning, and Slack ^14 found delayed emission with a period of 45 sec., and Brostrom, Koch, and Lauritsen ^15 with periods of 0.1—0.3 and 3 sec. On the average, however, the number of these neutrons accompanying the β-decay of the fragments does not exceed a few percent of the number of fissions. This was shown especially clearly by Gibbs and Thomson ^16, who worked with a pulsed flux of \((d, d)\) neutrons and established that the main mass of neutrons is emitted earlier than 0.001 sec. after the fission event.

Zinn and Szilard are inclined to think that the neutrons they observed are emitted from the fragments very soon after their formation and that the broadening of their spectrum is connected with this. Under this assumption the energy of the neutron with respect to the fragment is equal to \(\sim 2\) MeV. The maximum energy corresponds to the emission of the neutron in the direction of motion of the fragment, the minimum—to its emission in the opposite direction.

Up to the present time it still cannot be considered experimentally established that the neutrons are emitted from excited fragments, and not at the very moment of fission. This question could be resolved by setting up special experiments.

The data of Zinn and Szilard refer to the case of the splitting of uranium by thermal neutrons, i.e., to the fission of U\(^{235}\). We have no reliable data on the number of neutrons accompanying the fission of other nuclei.

Let us now turn to the consideration of the question of the chain nuclear reaction. After it had become clear that each fission event is accompanied by the emission of at least two neutrons, it became possible to think of carrying out a chain reaction. It could be realized in the case where, out of \(\nu\) neutrons accompanying fission, at least one in turn produced further fission. It is necessary, therefore, that \(\nu(1-\gamma)>1\), if \(\gamma\) rep—

constitutes the probability of such processes of interaction of neutrons with the nucleus, as a result of which the neutrons lose the ability to cause fission.

At the 1939 Conference on the Atomic Nucleus, the questions of carrying out a chain reaction for pure uranium and for a mixture of uranium with water were discussed in detail.

Let us first consider the second system. As is known, the cross section for the fission of uranium, depending on the energy of the neutrons, has two regions of large values, indicated in the figure by the lines \(AB\) and \(CD\). The cross section for the absorption of neutrons not leading to fission, for the given system, is represented by the line \(EF\), varying according to the same law as \(AB\), and by the line \(KM\). The segment \(EF\) is due to some extent to the absorption of slow neutrons by hydrogen; the segment \(KM\), however, to absorption only by uranium, leading to the formation of a transuranic radioactive isotope, but not causing fission.

Neutrons accompanying fission have, as we have seen, an energy of several MeV. Slowing down to thermal velocities, they pass through the dangerous region \(KM\) and are partly absorbed there. If \(\nu\) neutrons arise in fission, then only \(\nu p\) neutrons will reach thermal velocities, where \(p\) is the probability of their passage through the region \(KM\). Thus, the fission phenomenon will give rise to \(\nu p \varkappa\) neutrons; \(\varkappa\), equal to

\[ \frac{AL}{AL+EL}, \]

gives the ratio of the fission cross section to the entire absorption cross section of thermal neutrons. In order for a chain reaction to be possible, it is necessary that \(\nu p \varkappa > 1\).

It is not difficult to see that the most favorable conditions for carrying out a chain reaction will occur for a quite definite ratio of the number of hydrogen and uranium atoms in the mixture.

At very large concentrations of hydrogen the coefficient \(p\) will be large, since the neutrons will be slowed down more intensively in the mixture, but the coefficient \(\varkappa\) will be small, since the probability of absorption of thermal neutrons by hydrogen increases. On the other hand, very large concentrations of uranium, although they will lead to larger values of \(\varkappa\), will make the coefficient \(p\) small. At the 1939 Conference, a very carefully performed investigation by Zeldovich and Khariton was reported, devoted to the analysis of this question \(^{2}\). The authors came to the conclusion that even at the most advantageous ratio of the components of the mixture (4 atoms of hydrogen per 1 atom of uranium)—

3 Advances in Physical Sciences, vol. XXV, no. 2.

the product \(\upsilon\chi\) is equal to 0.82, i.e., a chain reaction in the uranium–water mixture is impossible.

Already last year Bohr pointed out, on the basis of simple theoretical considerations, that the segment \(AB\) in the cross-section curves belongs to the rare isotope \(\mathrm{U}^{235}\), while the segments \(EF\), \(CD\), and \(KM\) belong to the main isotope \(\mathrm{U}^{238}\). Thus there arose the possibility, noted by Zeldovich and Khariton, of carrying out a chain reaction in the water–uranium system by enriching the latter with the isotope \(\mathrm{U}^{235}\). Without changing the coefficient \(p\), it would be possible to increase \(\chi\) and bring the product \(\upsilon\chi\) to a value greater than unity.

Although Bohr’s considerations seemed convincing, it was only in 1940, when experiments on separated isotopes were carried out, that final confidence was obtained in the correctness of the assumption concerning the distribution of the individual segments of the curve \(ABKMCD\) among the different isotopes, and at the same time the fundamental solution of the problem of using intranuclear energy in the process of chain fission of uranium.

The practical solution of the problem by this route naturally presents enormous difficulties, in view of the fact that it is connected with a twofold change in the content of the light isotope in large masses of uranium.

Before proceeding to the discussion of chain processes in other systems, it is necessary to make one further remark concerning the conclusions of Zeldovich and Khariton about reactions in the unenriched uranium–water system. In view of the fact that the exact course and position of the resonance absorption band \(KM\) have not yet been established (Anderson \(^{17}\) this year, for example, gave for the resonance energy a value of 5 eV instead of the 25 eV which had earlier been adopted from the work of Hahn and Strassmann), Zeldovich and Khariton, in determining the coefficient \(p\) for different concentrations of the components of the mixture, used the following method:

According to their calculations, the coefficient \(p\) is equal to

\[ p=\exp\left(-a\sqrt{\frac{C_{\mathrm{U}}}{C_{\mathrm{H}}}}\right), \]

where \(C_{\mathrm{U}}\) and \(C_{\mathrm{H}}\) are the concentrations of uranium and hydrogen in the mixture, and \(a\) is a certain constant. It was determined by Zeldovich and Khariton from the experiments of Halban, Kowarski, and Savitch, in which, for a number of pairs of values \(C_{\mathrm{U}}\), \(C_{\mathrm{H}}\), the quantity \(p\) was measured. Knowing the value of \(p\), one can obviously, from the preceding formula, calculate \(p\) for mixtures of different composition, and the result of the calculations will not depend on the form or position of the resonance level \(KM\).

One may attempt to carry out a chain reaction of fission of the isotope \(\mathrm{U}^{235}\) by using, for slowing down, not only protons but also other light nuclei. In view of the fact that the relative concentration of slowing-down nuclei in the mixture must be large, in order to ensure small absorption of neutrons in the dangerous region, only very small values of absorption of thermal neutrons by these nuclei are permissible; consequently,

of the order of \(10^{-27}\)—\(10^{-28}\ \mathrm{cm}^2\). Most light elements absorb neutrons only weakly; but the exact values of the cross sections have been determined unreliably (because of the difficulty of measuring such small interactions of neutrons with matter), and even last year it was not possible, for any of the moderating nuclei except the proton, to make a confident analysis of the conditions for the development of a chain reaction.

In May of this year a report was published by Burst and Harkins \(^{18}\), who measured the cross section for the capture of neutrons by deuterons from the number of decays formed in such absorption of nuclei of the hydrogen isotope with mass 3. This absorption cross section was found to be \(3\cdot10^{-28}\ \mathrm{cm}^2\), i.e., considerably smaller than the critical value \((3\cdot10^{-27}\ \mathrm{cm}^2)\) that would suffice for the development of chains. The realization of a chain disintegration of \(\mathrm{U}^{235}\) in an unenriched uranium—heavy-hydrogen system is therefore possible.

In this system it is possible to avoid the separation of uranium isotopes, but instead there arises the necessity of separating hydrogen isotopes in large quantities, so that carrying out the experiment in this case too is associated with enormous practical difficulties. Calculations made by Zel’dovich and Khariton show that the amount of heavy water required for carrying out a chain reaction is approximately \(15\ t\). In this case the reaction could be realized with this amount only if the cross section for the absorption of thermal neutrons by oxygen is not higher than \(10^{-27}\ \mathrm{cm}^2\). If it is larger, then the reaction could be carried out only in chemically pure hydrogen; the necessary amounts of gas depend strongly on its pressure and could be obtained from \(15\ t\) of heavy water only if the hydrogen were compressed to pressures of several thousand atmospheres \(^{1}\).

The question of the suitability of \(\mathrm{He}^{4}\), \(\mathrm{C}^{12}\), and \(\mathrm{O}^{16}\) as moderating nuclei has not yet been fully clarified, but the requirements on the neutron-capture cross sections by these nuclei, which must respectively be less than \(3\cdot10^{-27}\), \(1.5\cdot10^{-27}\), and \(1.2\cdot10^{-27}\ \mathrm{cm}^2\), make the possibility of using them for carrying out a nuclear chain reaction unlikely.

Let us now consider the conditions for the development of chains in the mass of a pure element undergoing fission under the action of fast neutrons. In doing so, because of the absence of experimental data, we shall have to assume that both the number and the energy of the secondary neutrons are the same as in the fission of \(\mathrm{U}^{235}\).

The main cause of chain termination here is no longer the absorption of neutrons in side processes, which is small at the high velocities of these particles, but their loss of energy in inelastic scattering.

\(^{1}\) Note added in proof. It is now becoming clear that chain development in a mixture of unenriched uranium—heavy hydrogen is impossible.

Recent measurements by Hill and Goldhaber have shown that the half-life of \(\mathrm{H}^{3}\) is not 150 days, as assumed according to the data of Alvarez, Burst, and Harkins, but 30 years. In this connection, for the capture cross section of slow neutrons one obtains the value \(2\cdot10^{-26}\ \mathrm{cm}^2\), i.e., larger than the critical value \((3\cdot10^{-27}\ \mathrm{cm}^2)\).

The value of the cross sections for inelastic scattering of neutrons is now known to us better than last year, thanks to the work of Nikitinskaya and Flerov. They showed for a number of elements that the cross sections of such processes of inelastic scattering of neutrons from an \((\mathrm{Rn} + \mathrm{Be})\) source, after which these neutrons can no longer cause fission of uranium and thorium, are expressed by the formula

\[ \sigma = \pi \left(1.3 \cdot A^{\frac{1}{3}} \cdot 10^{-13}\right)^2, \]

where \(A\) is the mass number of the nucleus. It must be assumed that practically every fast neutron entering a nucleus undergoes inelastic scattering with a large loss of energy.

It follows from this that the cross section for inelastic scattering for the elements in which fission occurs (direct measurements could not be made in this case) will supplement the fission cross section up to the geometrical cross section of the nucleus. Since the fission cross sections of uranium, protactinium, ionium, and thorium are known for neutrons from an \((\mathrm{Rn} + \mathrm{Be})\) source, the inelastic-scattering cross sections for the above-listed nuclei can also easily be indicated for these same neutrons. They are given in Table 2. In the third row of the table the coefficient \(\gamma\) is given, and in the fourth—the quantity \(\gamma(1-\gamma)\), characterizing the possibility of the occurrence of a chain reaction. We see that only for protactinium

Table 2

\(\nu(1-\gamma)\) is greater than unity and, consequently, only in this case is the realization of a chain reaction for fast neutrons possible.

These conclusions apparently are quite plausible. The circumstance that the cross sections were determined for neutrons from an \((\mathrm{Rn} + \mathrm{Be})\) source, and not for the spectrum of neutrons accompanying fission, cannot, it seems to me, have especially substantial significance, since the ratio of the fission and inelastic-scattering cross sections, as has been established experimentally, at least for uranium, depends little on the neutron energy in regions not very close to the boundary

...of fission. It is hardly possible to think, likewise, that the number of neutrons accompanying fission will differ very greatly for different nuclei.

In conclusion I would like once more to emphasize that, although in principle the question of carrying out a nuclear chain disintegration has been resolved in the affirmative, on the path to its practical realization in the systems now investigated enormous difficulties arise. This is clearly seen from Table 3, in the second column of which are indicated the minimum quantities of materials required for a chain reaction, in the third column their stocks in all the laboratories of the world, and in the fourth column the ratios of the two quantities.

Table 3

Perhaps the coming years will bring us other ways of solving the problem, but if this does not happen, then only new, very effective methods of separating the isotopes of uranium or hydrogen will ensure the realization of a nuclear chain reaction.

LITERATURE

1. Leipunskii, Izv. Akad. nauk SSSR, ser. fizich., 4, 291, 1939.
2. Ya. Zel’dovich and Yu. Khariton, Uspekhi fizich. nauk, 23, 329, 1940.
3. Gant, Nature, 144, 707, 1939.
4. Haxby, Shoupp, Stephens, Wells, Phys. Rev., 58, 92, 1940.
5. Jentschke, Prankl u. Hernegger, Naturwiss., 28, 315, 1940.
6. Nier, Booth, Dunning and Grosse, Phys. Rev., 57, 546, 748, 1940; Kington, Pollock, Booth and Dunning, Phys. Rev., 57, 749, 1940.
7. Petrzhak and Flerov, ZhETF, 10, 1012, 1940; Uspekhi fizich. nauk, 25, 178, 1940.
8. Haxby, Shoupp, Stephens, Wells and Goldhaber, Bull. Am. Phys. Soc., 15, 41, 1940.
9. Ladenburg, Kanner and Barschall, Phys. Rev., 56, 168, 1939.

1. Ageno, Amaldi, Bocciarelli, Trabacchi, Ric. Sci., 13, 302, 1940; 11, 413, 1940.
2. Kunner and Barshall, Phys. Rev., 57, 372, 1940.
3. Henderson, Phys. Rev., 56, 703, 1939.
4. Nischina, Jasaki, Kimura and Ikawa, Phys. Rev., 58, 660, 1940; Nature, 146, 24, 1940.
5. Booth, Dunning and Slack, Phys. Rev., 55, 981, 1939.
6. Brostrom, Koch and Lauritsen, Nature, 144, 212, 1939.
7. Gibbs and Thomsen, Nature, 144, 212, 1939.
8. Anderson, Phys. Rev., 57, 567, 1940.
9. Borst and Harkins, Phys. Rev., 57, 619, 1940.

1. The table was compiled by research associate of the LPTI I. P. Selinov.
   * denotes a metastable state of the nucleus. ↩↩↩