OBSERVATIONS WITH ELECTRON-SENSITIVE PLATES EXPOSED TO COSMIC RADIATION^*

By Miss R. BROWN, U. CAMERINI, P. H. FOWLER, H. MUIRHEAD and PROF. C. F. POWELL

H. H. Wills Physical Laboratory, University of Bristol and D. M. RITSON Clarendon Laboratory, Oxford

PART 2. FURTHER EVIDENCE FOR THE EXISTENCE OF UNSTABLE CHARGED PARTICLES, OF MASS ~ 1,000 m_e, AND OBSERVATIONS ON THEIR MODE OF DECAY

ONE of the first events found in the examination of electron-sensitive plates exposed at the Jungfraujoch is represented in the mosaic of photomicrographs shown in Fig. 8. There are two centres, A and B, from which the tracks of charged particles diverge, and these are joined by a common track, t. Because of the short duration of the exposure, and the small number of disintegrations occurring in the plate, the chance that the observation corresponds to a fortuitous juxtaposition of the tracks of unrelated events is very small—of the order 1 in 10⁷. It is therefore reasonable to exclude it as a serious possibility. Further observations in support of this assumption are presented in a later paragraph.

An inspection of the track k shows that the particle producing it approached the centre of disintegration A. The range of the particle in the emulsion exceeds 3,000 m, and there is continuous increase in the grain-density along the track in approaching A. Near A, the grain-density is indistinguishable from that of particles of charge e, recorded in the same plate, near the end of their range.

The evidence for the direction of motion of the particle based on grain-counts is supported by observations on the small-angle deviations in the track due to Coulomb scattering. These deviations are most frequent near A, and the scattering is less marked at points remote from it.

From these observations, it is reasonable to conclude that the particle k approached the point A; that it carried the elementary electronic charge and that it had reached, or was near, the end of its range at the point A. We therefore assume that the particle k initiated the train of events represented by the tracks radiating from A and B. It follows that the particle producing track t originated in star A, and produced the disintegration B. In order to analyse the event, we first attempted to determine the mass of the particle k.

Mass Determinations by Grain-Counts

About a year ago, experiments were made in this Laboratory to determine the ratio, m_p/m_m, of the masses of p- and m-mesons, by the method of grain-counting⁵, and by studying the small-angle scattering of the particles in their passage through the emulsion⁴. The values obtained by the two methods were m_p/m_m = 1·65 ± 0·11, and m_p/m_m = 1·35 ± 0·10^*, respectively. Recent experiments at Berkeley⁶ suggest that the true value is 1·33 ± 0·02, a result which throws serious doubt on the reliability of the method based on grain-counts. Because of the advantage of this method, and of the important conclusions which have been based on it, experiments were made to determine the conditions in which reliable results can be obtained.

In the first experiments⁵, the two most serious experimental difficulties arose from the fading of the latent image and from the variation of the degree of development with depth. This made it necessary to work only with tracks formed contemporaneously; to compare the grain-density along the tracks of the p- and m-mesons of the same pair. As a result, the tracks of the p-mesons available for measurement were, in most cases, shorter than 400 m. In continuing the experiments, much more favourable conditions were obtained by using short exposures, so that the effects of fading were negligible; and by developing the plates by the method employed by Dilworth, Occhialini and Payne⁷, which gives a nearly uniform degree of development with depth.

In the plates obtained by these methods, it is legitimate to compare the grain-density in the tracks of unrelated particles. Further, it is now known that at least the majority, and possibly all, the mesons which produce 'stars' are p⁻-particles^6,8; and that most of the r-mesons are m⁺- and m⁻-particles. In determining m_p and m_m, we have therefore made measurements on the tracks of p⁺- and p⁻-, m⁺-and m⁻-particles, of length greater than 1,000 m, comparing the results with those of similar measurements made on the tracks of protons. In these conditions, we have found m_p/m_m = 1·33 ± 0·05. A detailed account of the observations will be published elsewhere; but, for the purpose of the present paper, it is sufficient to note that the results appear to be in good accord with those obtained by other methods. We conclude that, using the Ilford C2 emulsion in the new conditions, reliable information can be obtained.

We have seen that the conditions of uniform development and absence of fading have been achieved in the present experiments with the new Kodak emulsions, and we therefore attempted to measure the mass of particles by similar methods to those employed with the Ilford plates. The results obtained in observations on the tracks of four protons and four m-particles, occurring in the same plate, are represented in Fig. 9. In this figure, the number of grains per unit length in. the tracks is plotted for different values of the residual range; and the mean values, for tracks of the same type, are indicated by the full lines. The ratio of the masses of the two types of particles can be deduced by making a comparison of the values of the residual range at which the grain densities have the same value. The result thus obtained is m_m = 220 ± 20 m_e.

Using similar methods, we have made estimates of the mass of the particle, k, and the measurements are represented in Fig. 10. This figure shows the mean values of the grain-density in the tracks of the four m-mesons and four protons, together with the corresponding results for the particle k. All the tracks under consideration occurred in the same plate.

Table 1 shows the values of the mass of the particle, k, as determined from these results, by making a comparison of the grain-density in the track of the particle with the mean curve for protons. The values thus obtained are all independent and the mean is m_k = 1,080 ± 160 m_e.

The limits of error given above have been deduced in the following manner: We have compared the grain-density in the tracks of the four individual protons with the mean curve for the same particles—(see Fig. 9)—and have thus obtained a number of independent values for the apparent mass of each of these particles. The distribution in these values allows us to calculate the 'probable error' associated with the mass as determined from the observations on any one track, expressed as a percentage of the apparent mass of the particle. It is then assumed that the 'probable' percentage error in the calculated mass of the particle k has the same value.

We have also determined the mass m_k by studying the small-angle scattering of the particle, by the methods recently described⁴, and the result thus obtained is m_k = 1,800 ± 400 m_e. If the true mass of the particle is 1,080 m_e, the chance that the value obtained by observations on scattering shall be equal to, or greater than, 1,800 m_e is one in four. Because of the large statistical fluctuations associated with the observations in the scattering experiments, we give more weight to the measurements by grain-counting. It appears certain, from these observations, that the true value of m_k lies between 700 and 1,800 m_e, and we think it highly probable that it is substantially less than that of the proton. Thus every individual point representing the grain-density in the track k, at a particular value of the residual range, lies below the corresponding points for each of the four protons.

Disintegration 'B'

The tracks, c and d, of the two particles emitted from point B are characteristic of protons or heavier particles, and we regard them as due to a disintegration produced by the particle t. This particle was frequently scattered in passing through the emulsion and was therefore of low velocity; and the evidence is consistent with the assumption that it had reached the end of its range at the point B.

The only known slow charged particle which is capable of producing a disintegration of the type represented by star B is a p⁻-particle^6,8. We therefore assume that a negative meson of mass 286 m_e was created at the point A, and reached the end of its range to produce the disintegration B.

Transmutation 'A'

In order to interpret the transmutation A, we first made a detailed examination of the tracks of the emitted particles. Of the two tracks a and b, the former has a length in the emulsion of more than 2,000 m, and ends in the surface, whereas b ends in the glass and is 116 m. long. The grain-densities in the two tracks are equal to within the limits defined by the statistical fluctuations. The average grain-density in the long track a is 49·0 grains per 100 m; that is, 2·17 times the value characteristic of minimum ionization for a particle of charge e. Unless we admit the existence of fractional values of the electronic charge, we must conclude that the particles producing the tracks a and b both carried charges of magnitude e.

In order to determine the possible values for the energy of the particles producing tracks a and b, we have calculated the variation with energy of the specific ionization of a particle of charge e, from the formula of Halpern and Hall⁹, assuming the atomic composition of the emulsion to be identical with that of the Ilford C2 plates. This formula is a modification of that of Bloch¹⁰; it applies to particles moving in a solid medium and gives results in good agreement with experiment for particles of low energy. The results are shown in Fig. 11, where the specific ionization is plotted as a function of the quantity E/m, where E is the energy and m the mass of the particle, both quantities being measured in MeV. From Fig. 11, we have determined the possible values of the energy of the particles, a, b, corresponding to the observed grain-density in the tracks, assuming them to be protons, p-mesons, m -mesons or electrons. The resulting values are tabulated in Table 2.

There are two possible interpretations of the transmutation produced at A by the particle k. We can assume, either that the particle was captured by a nucleus, or that it decayed spontaneously. From the measured values of the mass of the particle, it would be possible, from the point of view of the conservation of mass and energy, to admit that, at the end of its range in the emulsion, it was captured by a nucleus and led to the ejection of two energetic protons and a p⁻-particle. It appears almost certain, however, that the release in a nucleus of such a large amount of energy would lead to the 'evaporation' of many nucleons, a process commonly observed in plates exposed to the cosmic radiation; and that two protons of great energy would be only two components of a 'many-pronged' star. (It may be noticed that we cannot assume that the particle k was captured by one of the rare nuclei of heavy hydrogen, present in the gelatine. In such an interaction, the algebraic sum of the charges on the two initial particles is 0 or 2e, whereas that of the product particles is e or 3e.) We shall see later that there are other objections to the hypothesis that the trades a and b were produced by protons, or heavier nuclei of charge e.

It follows from the above considerations that if we are to describe the transmutation in terms of particles of which the existence is already established, we must attribute the tracks a and b either to electrons, to m-mesons or to p-mesons. Considering the first of these possibilities, we must assume the electrons to have had an energy value greater than that corresponding to minimum ionization, namely, greater than 1,000 MeV.; for with the alternative lower value corresponding to the observed ionization, 300 keV., the particle would have had a range in the emulsion of only about 100 m, and would have been frequently scattered. The assumption that the particles a and b were electrons is therefore inconsistent with the conservation of energy and can be rejected. We are left with the alternatives that the tracks were produced either by p- or by m-mesons.

If the particles a and b were mesons, we must assume, in order to conserve mass-energy, that their kinetic energies were 27 MeV. or 37 MeV., respectively, in the case of m- or p-mesons (see Fig. 11). In either case, it appears to be very difficult to reconcile the observations with the assumption that the particles were emitted as a consequence of the liberation in a nucleus of the energy corresponding to the rest-mass of particle k. We are therefore led to examine the possibility of explaining the observations in terms of a spontaneous decay of this particle.

Assumption of a Spontaneous Decay of the k-Particle

In examining the possibility that the transmutation A corresponds to a spontaneous decay of the particle k, we require to know the relative directions of motion of the three ejected particles. For this purpose it is necessary to determine the shrinkage of the emulsion; the ratio, S, of the thickness of the emulsion during exposure to that after it had been developed, fixed and dried. We have measured this quantity by examining the tracks of a-particles, produced in the emulsion by uncontrolled radioactive contamination. Among such 'stars', it is possible to identify some, due to an original atom, of radiothorium, from which an a-particle of thorium C′ was emitted. The shrinkage has been measured by determining the lengths of the projection of the corresponding tracks on the surface of the emulsion, and their apparent angles of 'dip'. The value of the 'shrinkage' thus found is S = 2·7 ± 0·1. Knowing the value of S, the original orientation of a track in the emulsion, before processing, can be determined, in favourable cases, with a precision of the order of 1°, by observing the apparent angle of 'dip' of the particle, and the direction of its projection on the plane defined by the surface of the emulsion. Using these methods, the original directions of motion of the three particles a, b and t were found to be coplanar. The departure of the direction of motion of any one particle from the plane defined by the other two is less than 4°. The error in this determination is largely due to the fact that track t is of short range, and the particle producing it was of low velocity, and frequently scattered.

The values of the angles between the directions of motion of the particles in the common plane are shown in Fig. 12. The observed coplanarity makes it legitimate to assume that the three particles arise as a result of the spontaneous decay of the k-particle at the end of its range in the emulsion, and that they are the only product of its disintegration; that no neutral particles, which would escape observation, are emitted. It follows that the vector sum of the momenta of the three particles must be assumed to be equal to zero.

If we are correct in attributing the track t to a p⁻-particle, it follows from the observed range, 45 m, that the kinetic energy of ejection was 1·04 MeV. The corresponding value of the momentum of the particle is 17·5 MeV./c. From the observed directions of motion, the momenta of the particles giving tracks a and b are then found to be 98 ± 5 and 104 ± 5 MeV./c, respectively. Those values are to be compared with those corresponding to electrons or mesons listed in Table 2, which have been deduced from the observed grain-density in the tracks. We have seen that the values given in Table 2 for the momenta of the two particles, if they are assumed to be electrons, are many times too large. It follows that there is a wide departure from a momentum balance if the tracks a and b are assumed to be due to either electrons or protons. Further, the values of the momenta, as deduced from observations on the scattering of the particles, are inconsistent with those obtained from grain-counts, if the particles are assumed to have been either electrons or protons (see Table 2).

The agreement between the sets of values for mesons, however, is most remarkable, and gives strong support for the assumption of a spontaneous decay of the k-particle. Only a very rare combination of unrelated features, including the co-planarity of the trades and the directions of motion of the particles in the common plane, the range of the particle t, and the specific ionization of the particles producing tracks a and b, could produce such an agreement between the estimated values of the momenta, if the result is fortuitous.

The values of the momenta of the particles producing tracks a and b, as determined by the three different methods, are consistent, within the errors of measurements, with the assumption of a spontaneous decay of the k-particle whether the product particles are assumed to be m-mesons or p-mesons. We can apply a further test by calculating the values of the rest-mass of the particle k which corresponds to the two different assumptions, and the results are tabulated in Table 3.

It will be seen from Table 3 that the assumption of two m-mesons corresponds to a rest mass of the k-particle of 869 m_e; and for two p-mesons, 985 m_e. The assumption of different particles, one p- and one m-meson, gives an intermediate value of approximately 925 m_e. In view of the error in the direct determination of m_k, the results are not decisive.

If the transmutation is to be interpreted in terms of particles of which the existence is already established, we are left with four possibilities for the nature of the particles producing tracks a and b. These are indicated schematically in Table 4.

For the following reasons, case 3, Table 4, is the most improbable. If track a is that of a p-meson, we can calculate the momentum and the grain-density to be expected in track b. We thus obtain the value of 34 grains per micron instead of 51·0 ± 6·0 as observed. For case 4, on the other hand, if a is a m-meson, the calculated grain-density for track b is 64, a value which differs from that observed by an amount only twice that corresponding to the standard deviations. The observed grain-densities agree best with the assumption that the two particles are of the same type.

Observations on the scattering of the particle producing track a are in better accord with the assumption that it is a p-meson rather than a m-meson (see Table 2); but the results are again indecisive. We may sum up this evidence, and that provided by the mass determinations by grain-counting, by saying that there is some support for the view that the three product-particles are p-mesons; but that the alternative possibilities of one p- and two m-, or two p- and one m-meson cannot be excluded.

Chance Juxtaposition of Unrelated Events

In the light of the analysis made in the preceding sections, we can now return to the original assumption that the event is not to be regarded as a fortuitous juxtaposition of tracks. The accuracy of the determination of the mass of the particle k does not allow us to exclude the possibility that it has a mass as great as that of a proton, although the observations by grain-counts render it very improbable. Suppose then that a proton, unrelated to the particles producing the other tracks, came to the end of its range at A. Even with this assumption, the event is still difficult to explain in conventional terms. Many examples of p⁻-particles ejected from stars have been observed in this Laboratory⁸, but in the present instance the existence of a nuclear interaction in which two protons of great energy are emitted, unaccompanied by slow protons and a-particles, would remain to be explained. A similar difficulty is met if we assume that a particle producing one of the tracks, a or b, approached A and produced the transmutation.

If, alternatively, the tracks c and d, diverging from star B, represent an unrelated disintegration—produced, for example, by a g-ray—we could then assume track t to be that of a proton. We are then left with the difficulties associated with the features peculiar to star A, which must now be assumed to have been produced by a slow, charged particle; difficulties which have already been discussed in a previous paragraph. These considerations give further support to the original assumption, that all the tracks shown in the mosaic represent a succession of associated processes.

Relation of the Present Results to Other Observations

If a particle with the elementary electronic charge suffers a spontaneous decay, the law of the conservation of charge demands that the number of emitted particles of charge e shall be odd. From this point of view, the sign of the charge of the original particle can have been either positive or negative. If the particles producing tracks a and b form a pair of opposite sign, then the original k-particle was negative. The only other alternative is that they were both positively charged, in which case the k-particle was also positive. It is therefore possible that our observations correspond to a mode of decay of positive particles of mass approximately 900 m_e, and that the observation by Leprince-Ringuet² demonstrates the fate of the corresponding negative particles—nuclear capture with the production of a 'star' and the ejection of a p⁻-particle.

Rochester and Butler² have published an expansion-chamber photograph which appears to be due to the spontaneous decay of a neutral particle of mass approximately 900 m_e into a pair of oppositely charged particles of rest-mass approximately 300 m_e. We have therefore considered the possibility that the decay process suggested by the present results can be regarded as taking place in two stages: the emission of a p⁻-particle of low energy, followed by the spontaneous decay of the resulting neutral particle. On this view, however, it would be necessary to assume that the neutral particle has a life-time of the order of 10⁻¹⁴ sec. Otherwise, in recoiling from the p-particle, it would move away from the original point of decay, and the two charged particles into which it became transformed would originate from a point separated from the beginning of the track of the p⁻-particle. It follows that we cannot identify such a postulated unstable neutral particle with that for which evidence is provided in the experiments of Rochester and Butler.

Finally, we have considered the possible relations of the present results to the particles of mass approximately 800 m_e referred to as t-mesons, evidence for which has been recently reported by Bradt and Peters². It is a remarkable feature of their experiments that their t-mesons give rise to no recorded secondary particles at the end of their range. It appears to be possible that these particles also decay with the emission of three fast mesons, but that the transmutation usually takes place with a more equal partition of kinetic energy than in the case we have observed. It would then follow that in the Ilford C2 emulsion the disintegration products would commonly escape observation. If this view is correct, we must regard the event we have observed as representing a rare example of a common mode of decay of these mesons; an example which, by chance, has allowed a detailed analysis to be carried out. If so, the t-meson of Bradt and Peters, when recorded by electron-sensitive emulsions, should show the tracks of three particles, of low specific ionization, and of which the directions of motion are co-planar.

We have pleasure in thanking Prof, von Muralt and members of the staff of the Jungfraujoch Forschungsstation for hospitality and assistance in obtaining the exposures; Dr. E. R. Davies and Dr. W. E. Berriman, of Messrs. Kodak, Ltd., for special photographic plates; Miss C. Dilworth and Dr. G. P. S. Occhialini for advice on development; Mr. W. O. Lock and Mr. J. H. Davies for assistance in making observations on the scattering of particles in the emulsion; and to the team of microscope observers of this Laboratory. We are indebted to Prof. N. F. Mott and other colleagues for a number of discussions on the processes associated with the capture of negative mesons by nuclei.

Note added in proof. Since completing this article, we have been informed by Dr. Peters that, in Ilford C2 emulsions exposed at 90,000 feet, ho and Dr. Bradt have observed three events with the following characteristics. A particle, which they judge to be similar in mass to their t-mesons, appears to come to rest and to lead to the emission of a particle of smaller mass, which, at the end of its range, produces a nuclear disintegration. The ranges of the secondary particles, in the three cases, are 20, 25 and 45 m, respectively. The authors were not aware of our results when they suggested to us that their observations may correspond to the spontaneous decay of heavy mesons. According to their description, these events are precisely similar to those we should expect to observe in C2 emulsions as a result of the spontaneous decay of heavy particles of the type we have postulated; for any particles of low specific ionization will not be recorded by the Ilford plates. The observations of Peters and Bradt appear, therefore, to give further support for the assumption that the present observations are not due to a chance juxtaposition of tracks; and they suggest that it will be possible, in the near future, to find similar examples suitable for making a detailed analysis.

1. Berriman, Nature, 162, 992 (1948).
2. Leprince-Ringuet, C.R., 226, 1897 (1948). Rochester and Butler, Nature, 160, 855 (1947). Bradt and Peters, Report to the Bristol Symposium, 1948 (in the press). Alichanian. Alichanov and Weissenberg, J. Exp. and Theoret. Phys., U.S.S.R., 18, 301 (1948); and other references.
3. Camerini, Muirhead, Powell and Rltson, Nature, 162, 433 (1948).
4. Goldschmidt-Clormont, King, Muirhead and Ritson, Proc. Phys. Soc., 61, 138 (1948).
5. Lattes, Occhialini and Powell, Proc. Phys. Soc., 61, 173 (1948).
6. Serber, Report of Solvay Conference for 1948.
7. Dilworth, Occhialini and Payne, Nature, 162, 102 (1948).
8. Occhialini and Powell, Nature, 162, 168 (1948).
9. Halpern and Hall, Phys. Rev., 73, 477 (1948).
10. Livingston and Bethc, Rev. Mod. Phys., 9, 263 (1937).
11. Camerini and Lattes (private communication); see also Powell and Occhialini, “Nuclear Physics in Photographs”, 112 (Oxford, 1947).

^* For the following reasons, the limits of error quoted above, in the determination of m_p/m_m by observations on scattering, are less than those given in ref. 4. Previously, values for the mass of the different types of mesons, classified phenomenologically, were given separately. It is now known, however, that at least the majority of the s-mesons are p⁻-particles; and the r-mesons, m⁺- and m⁻-particles. The different results can therefore be combined to give a value for m_p/m_m with a greater statistical weight.

DECAY OF V-PARTICLES

By R. ARMENTEROS, K. H. BARKER, C. C. BUTLER, A. CACHON and A. H. CHAPMAN

Physical Laboratories, University of Manchester

IN 1947, Rochester and Butler¹, using a Wilson cloud chamber in a magnetic field, observed two V-shaped tracks associated with high-energy penetrating showers. The first event consisted of two tracks at minimum ionization starting from a point in the gas, and the angle of the resulting fork was 67°. The second event consisted of a minimum ionization track, which was apparently scattered through 20° in the gas of the chamber, but no recoil track was produced. The two events were interpreted respectively as the spontaneous decay of a neutral and of a charged particle, each with a minimum mass of about 1,000 m_e.

From 1947 to 1949, Butler et al.² and Barker and Butler³ operated the same cloud chamber at sea-level, but no additional V-shaped tracks were observed. Anderson and his colleagues⁴, using a similar apparatus, both at sea-level and at an altitude of 3,200 m., found thirty-four examples of the V-shaped tracks, thirty neutral and four charged. They concluded that the interpretation suggested by Rochester and Butler was correct. After consultation with Prof. Carl Anderson, Prof. P. M. S. Blackett has suggested that the particles responsible for the V-shaped tracks should provisionally be called 'V-particles', so postponing a more precise nomenclature until further knowledge of their properties is available.

The original cloud chamber and electromagnet used by Rochester and Butler have been re-assembled at the Observatoire du Pic-du-Midi de Bagnères-de-Bigorre (2,867 m.) in the Pyrenees. During the first six months of running the apparatus (that is, up to January 15, 1951), forty-three V-tracks were observed, of which thirty-six were neutral decays and seven charged decays. No decay processes involving three charged secondaries were found. In the present article some of the main features of these now results will be described.

The cloud chamber was operated in a field of 7,000 gauss and triggered by a set of Geiger counters sensitive to showers of particles. The maximum detectable momentum was found to be 8 × 10² eV./c for tracks of at least 6 cm. length. A total of 7,500 photographs has been obtained, but not all can be identified as penetrating showers. We define the term 'penetrating shower' for the Pic-du-Midi photographs as one in which two or more contemporary particles penetrate a 2.2-cm. lead plate in the chamber without multiplication, or alternatively one in which there is one penetrating particle and one or more heavily-ionizing particles. Most of the V-tracks are associated with identified penetrating showers.

The results obtained on the Pic-du-Midi are compared with all the sea-level results in Table 1, and a preliminary discussion of the V-track follows.

There are a number of well-known processes which, under special conditions, may appear on the photographs as V-shaped tracks; for example, (a) certain types of nuclear interactions and scatterings in the gas, and (b) the decay of the p-meson and the m-meson.

Events of type (a) are expected to be more numerous in the lead plate than in the gas by a factor of about 400. In fact, very few wide-angle pairs from the plate and only 120 scatterings in the plate of greater than 5° have been observed. We have observed, however, four stars in the gas and 150 nuclear interactions in the plate. Many events of the latter type do not produce visible prongs owing to absorption in the lead; correcting for this loss of events, we estimate the actual number of stars in the plate to be about 350. The observed ratio of stars in the gas to stars in the plate is therefore about 1 : 90, instead of the expected ratio of 1 : 400. We conclude that the number of observed stars in the gas is already large by comparison with the predicted number, and consequently it is scarcely possible to explain an additional 26 V-events as stars in the gas. This conclusion is supported by making an estimate of the flux of nuclear interacting particles through the chamber. From this flux, we predict the occurrence of about three stars in the gas; moreover, these stars would have much lower energies than the observed V-tracks. Nuclear interacting particles, with momenta greater than 10⁹ eV./c, can produce stars in the plate which are usually visible in the chamber. If we make an estimate of the flux of these particles, then the predicted number of interactions in the plate is about 200, in agreement with the observed number.

Two of the charged V-tracks can be explained as events of type (b); one is a p → m decay and one a m → e decay.

We conclude that not more than 10 per cent of the observed V-tracks can be explained by well-known processes, and the rest must therefore be explained by the spontaneous decay of V-particles.

Whenever possible, momentum measurements and ionization estimates have been made of the products of V-particle decays, and, using the techniques described by Butler et al.², some of the particles have been identified as protons and mesons with considerable certainty. The measurements on the secondaries of neutral V-decays are the most numerous and are summarized in Table 2.

Four of the secondary particles of uncharged V-decays are observed to penetrate the lead plate without interaction, and one is scattered through 15°; this is probably a nuclear scattering. In addition, one positive particle is catastrophically absorbed in the plate. Since no electron cascades have been produced in the plate, we conclude, tentatively, that the secondaries consist of protons and mesons.

No neutral V-event has been found on which both secondaries can be identified. Anderson et al.⁴ found one example of an identifiable meson; the four protons in Table 2 appear to be the first identified in the cloud chamber as secondaries of neutral V-particles. Two of these protons are shown in Figs. 1 and 2. In each case the negative particle is approximately at minimum ionization and has momentum about 10⁸ eV./c; they must therefore be mesons, probably p-mesons. Anderson et al. find that in twelve cases the origin of the main nuclear interaction lies in the plane of the V-track; therefore they conclude that the decay scheme involves only two secondary particles. If this conclusion is correct, we can analyse the data obtained from the four cases with proton secondaries according to the decay scheme:

The results are given in Table 3.

Many of the remaining neutral V-tracks may also be explained by employing decay scheme (1). Consideration of the dynamics of this decay scheme shows that, in general, the mesons will have lower momenta than the protons. This is observed in many cases, as can be seen from Table 2, since the number of identified negative mesons is much larger than the number of positive mesons.

Hopper and Biswas⁵ have observed an unusual two-pronged star in a photographic emulsion exposed to cosmic rays at an altitude of 70,000 ft. This event occurred close to, but not coplanar with, a very large star, and consists of an identified proton and a particle at minimum ionization with a momentum of 329 MeV./c, which may have been a meson or an electron. Hopper and Biswas consider the fork to arise from the decay of a neutral V-particle and, using decay scheme (1), find the mass to be (2,370 ± 60)m_e. They also conclude that the event is a two-particle decay, which is not associated with the large star. If it is associated with the star, there must be at least one neutral secondary particle as well as the proton and meson.

A possible alternative explanation to the one adopted by Hopper and Biswas is that the fork may be due to a nuclear collision of the type:

in which a neutron makes a glancing collision with a large nucleus. It is interesting to consider whether our four events with proton secondaries observed in the chamber could be explained as nuclear collisions of the above type. For this to be possible the estimated flux of neutrons must be appreciably in error; however, we have already shown that our estimate is in agreement with the observed number of interactions in the lead plate. In order to reach a definite conclusion about the event in the photographic emulsion, a complete analysis of two-pronged stars would have to be made.

Three of the positive secondary particles from neutral V-particle decays cannot be protons and are likely to be mesons. Unfortunately, the corresponding negative particles cannot be identified. Since the existence of the negative proton has not yet been established, we suggest, provisionally, that these negative particles may also be mesons. The decay scheme may be of the type:

The neutral V-particle found by Rochester and Butler¹ may be of this type. A reconsideration of the measurements on their positive particle, which is at minimum ionization, shows that the momentum is within the range (2–3) × 10⁸ eV./c, and so is unlikely to be a proton. The negative track is short and it is difficult to assess an upper limit for its momentum. If the V-particle originated in the main interaction, then the geometry of the two-particle decay requires the momentum of the negative particle to be about 10⁹ eV./c. Using these values, the mass of the neutral V-particle is within the range (1,000–1,400) m_e.

The second event with a low-momentum positive particle is shown in Fig. 3. The apex of the V-track is close to the edge of the chamber (near A) and lies behind the heavily-ionizing proton (track 3). The angle of the fork is 37·5°. The positive particle (track) has a momentum of 2·0 × 10⁸ eV./c. It may be compared with track (3), which is a proton of momentum 2·2 × 10⁸ eV./c. The negative particle (track 2) is close to the edge of the chamber, where distortions are often severe. Its momentum is probably greater than 10⁹ eV./c. If the measured momentum values are used and both particles are p-mesons, the mass of the neutral V-particle is about 950 m_e.

The third event of the same type consists of a positive particle at minimum ionization with momentum 1·5 × 10⁸ eV./c and a short unmeasurable negative particle.

We conclude that the photographs provide strong evidence for the existence of two types of unstable neutral particles, one of greater and one of less than the proton mass, probably decaying according to schemes (1) and (3). This conclusion can only be avoided by assuming the existence of the negative proton.

The data on the charged V-decays are very meagre, but some provisional conclusions and a detailed discussion of one event will be given. Six charged V-tracks are due to the decay of new unstable particles, but measurements can be made only on four of the events. The four charged secondary tracks can be measured; one is positive and could be a proton, and three are negative and are probably mesons. The tracks of three of the unstable particles are too short for measurement, but one negative particle traversed the top half of the chamber and decayed below the plate, producing a meson and a neutral particle, which afterwards decayed, forming a neutral V-track. This event will be described in detail in a later paper; provisionally, we conclude that the neutral particle from this charged decay may be a neutral V-particle.

One of the charged V-particles decayed when travelling slowly in the chamber; it is shown in Fig. 4. A high-energy nuclear interaction occurred in the plate and a short heavy track (1) was produced, the ionization of which is more than four times minimum. Thus the velocity of the initial particle must be less than 0·45 c. The single charged decay product (track 2, on right) is at minimum ionization. The decay point is not seen easily on a single photograph owing to the presence of many other tracks, but is clearly visible stereoscopically. Track (2) has a momentum of −(1·4 ± 0·1) × 10⁸ eV./c; it makes an angle of 100° with track (1) and is probably a negative p- or m-meson. The decay process cannot be either a p → m decay or a m → e decay. The unknown neutral particle may be a V⁰-particle, a neutron or a p⁰-meson. The mass is found by assuming the decaying particle to have the maximum momentum consistent with the observed specific ionization. The results of these calculations are shown in Table 4.

A more detailed discussion of the measurements on all the V-tracks will be published in due course; meanwhile, we can summarize the main results of the present investigations as follows.

(a) The occurrence of V-tracks in the chamber has again been confirmed.

(b) Probably not more than 10 per cent of the V-tracks can be explained by processes other than the spontaneous decay of new unstable particles.

(c) Momentum measurements on fifty-four secondary particles from neutral V-decays have been made and, together with ionization estimates, make it possible to identify with certainty both protons and mesons.

(d) Assuming that only two particles are produced by the neutral decays, two schemes are suggested to explain the photographs: V⁰ → p⁺ + p⁻, with the V⁰ mass in the range (2,000–2,500) × m_e V⁰ → p⁺ + p⁻, with the V⁰ mass about 1,000 m_e. The relative frequency of those two processes will be discussed in a later paper, but the evidence is already strong for the existence of two types of V⁰-particle.

(e) Six charged decays have been observed, but no definite conclusions have been reached about the mass of the charged V-particles.

We wish to thank Prof. P. M. S. Blackett for his great interest and help in this work and for the excellent laboratory facilities he has given us. We are also much indebted to Prof. J. Rösch, director of the Observatoire du Pic-du-Midi de Bagnères-de-Bigorre, who gave permission for, and carried out with extreme efficiency, the installation of the 11-ton magnet in the mountain observatory. In addition, Prof. Rösch and his staff have been of the greatest assistance during the running of the apparatus. Our thanks are due to the Nuffield Trustees and the Department of Scientific and Industrial Research for financial assistance. We also wish to thank Dr. G. D. Rochester and Dr. J. G. Wilson for many helpful discussions.

1. Rochester, G. D., and Butler, C. C., Nature, 160, 855 (1947).
2. Butler, C. C., Rosser, W. G. V., and Barker, K. H., Proc. Phys. Soc., A, 63, 145 (1950).
3. Barker, K. H., and Butler, C. C., Proc. Phys. Soc., A, 64, 4 (1951).
4. Seriff, A. J., Leighton, R. B., Hsiao. C., Cowan, E. W., and Anderson, C. D., Phys. Rev., 78, 290 (1950).
5. Hopper, V. D., and Biswas, S., Phys. Rev., 80, 1099 (1950).

| return to looking back index page |