- Email: [email protected]
Cloud photographs of cosmic-ray stösse
Journal of
The Franklin Instilule Devoted to Science and the Mechanic Arts Vol. 216
DECEMBER, 1933
No. 6
CLOUD PHOTOGRAPHS OF COSMIC-RAY STOSSE.* BY G O R D O N L. L O C H E R , Ph.D., National Research Fellow.
INTRODUCTION.
Large instantaneous increments in the FOUNDATION ionization by cosmic rays are found in io~fizacommunication~o.S0, tion chambers that are surrounded by, or in the vicinity of, massive material. These large fluctuations are generally called "St6sse," "bursts." or "spurts of ionization" ; their magnitudes sometimes correspond to the passage of several thousands of electron tracks throught the measuring apparatus. The St6sse have mostly been studied by ionization methods; 1, 2, 3, 4, 5, 6, 7, 8 S w a n n and Montgomery 9 c o m b i n e d an ionization chamber with a coincidence set of GeigerM filler counters, finding that large increments in the ionization were frequently accompanied by the simultaneous discharge of
BARTOL RESEARCH
* Part of this paper has been printed in the Physical Review. ' Hoffmann and Pforte, Phys. Zeits., 3 I, 347 (I93o). 2 Steinke, Phys. Zeits., 3x, lOI9 (I93O). Schindler, Zeits. fiir Phys., 72, 625 (1931). 4 Steinke and Schindler, Zeits. fiir Phys., 75, 115 (I932). Messerschmidt, Zeits. fiir Phys., 78, 668 (I932). n Swann and Street, Phys. Rev., 40, Io49A (1932). 7 Messerschmidt, Naturwiss., 21, 285 (1933). 8 Steinke, Gastell and Nie, Naturwiss., 21,560 (1933). 0 Swann and Montgomery, Phys. Rev., 43, 782A (1933); Phys. Rev., 44, 52 (I933). (Note.--The Franklin Institute is not responsible for the statements and opinions advanced by contributorsto the JOURNAL.)
VOL. 216, NO. I296--46
673
674
GORDON L. LOCHER.
[J. F. I.
the counters, which were about a meter apart, outside the chamber. T h e infrequency with which St6sse occur makes the accumulation of d a t a on t h e m slow. (Relatively large bursts are produced, roughly, at 3 per hr. per IOO,OOO cc. of absorber.) At the present time it can be said t h a t the rate of their production is nearly proportional to the number of atoms in the absorbing material, for dense m a t t e r ; t h a t m a n y of the particles have ranges of more t h a n I0 cm. in lead (some 2o cm.), and t h a t the ionization per Stoss depends on the atomic n u m b e r of the absorber in which it is generated. F r o m a m a x i m u m ionization j u m p corresponding to about 4ooo electron tracks or 26o proton tracks, observed in one instance, 8 the bursts appear to taper off to sizes t h a t are too small to be recognized, with no sharply defined classes of size. F r o m the information now available, it seems impossible to tell whether the ejected particles derive their energy wholly from the primary cosmic ray, or draw part of it from the nuclei of disrupted atoms, nor can we say, in view of the discovery of the apparent synthesis of particles from photons, whether the particles ejected from St6sse all previously existed in the disrupted atoms; and, finally, we cannot be certain t h a t a Stoss is always produced by a single primary cosmic ray, rather t h a n by a shower of p r i m a r y disrupting agents from outside the earth's atmosphere. T h e m e t h o d of photographing cloud tracks of the disintegration products of St6sse promises to be a fruitful means of a t t a c k on the problem. But for a Stoss rate of I per hr., the chance of getting a significant photograph is about I/36,000, if the chamber is operated in the customary manner. Anderson 10 did accidentally get one burst, b u t the solution lay in a simple modification of the m e t h o d of Blackett and Occhialini, 11who automatically set off the expansion of a cloud machine immediately after the ionizing particle discharged a coincidence pair of cosmic-ray counters in line with the cloud chamber. In the course of their experiments, they not only verified the observation t h a t the particles discharging the lo Anderson, Phys. Rev., 43, 368 (1933). n Blackett and Occhialini, Proc. Roy. Soc. A, I39, 699 (I933).
Dec., 1933.]
CLOUD PHOTOGRAPHS OF STOSSE.
675
counters produce t r a c k s y but obtained photographs of several interesting bursts, as well. EXPERIMENTAL ARRANGEMENTS.
A cloud chamber 15 cm. in diameter has been arranged to be set off immediately after the simultaneous discharge of three Geiger-Mfiller counting tubes, each IOO cm3 in crosssectional area, placed two above and one below the chamber, but as far out of line as possible; no single particle, nor downward-directed branched pair originating near the chamber can set off the counters. The smallest sphere containing the counters, which have a maximum separation of 80 cm., has a volume about 500 times t h a t of the cloud chamber, so that the fraction of the tracks from the St6sse which pass through the cloud chamber may be small, if the point of origin is remote. Of the triple-coincidence counting rate, 1.26 per hr., it is estimated from the photographs taken that about one-third of the discharges are from St6sse, and two-thirds from branched tracks and small groups of tracks. Surrounding the chamber are about 135o lbs. of massive material, consisting of 750 lbs. of lead, 400 Ibs. of copper and brass, and 2o0 lbs. of iron. A cloud atmosphere of argon at 64. 5 cm. pressure is used. The specific ionization of tracks in argon is high ; the relatively high viscosity makes the tracks persist better than in air; the expansion ratio is small (1.18, as compared with 1.3 for air), and the probability of obtaining elastic-collision recoil atoms from any neutroas present is relatively high. Since the ionizing particles traverse the chamber about o.oi sec. before the expansion, the tracks are somewhat diffused by thermal agitation and by the expansion; equality of diffuseness of tracks of like densities serves to establish approximate simultaneity of their formation, and makes it easy to distinguish between pre- and post-expansion tracks. Stereophotographs are taken with two cameras whose optic axes intersect in the chamber at an angle of 2 2 ° . The connecting link between the counter set and the cloud machine is a thyratron (G.E. Co. Type FG 67), which acts through a quick-acting electromagnetic trigger to release a 12Mott-Smith and Locher, -Phys. Rev., 38, I399 (I931).
676
GORDON L. LOCHER.
[J. F. I.
mechanical drive mechanism that is normally held in readiness by vacuum force and spring tension. RESULTS AND
DISCUSSION.
Figures I, 2, and 3 are stereophotographs of interesting St6sse; Figs. 4, 5, and 6 are single photographs of three F I G . I.
Stoss with two nucleus tracks (N), and S electron tracks (E). E~, E2, and N~ converge to a point in the lead, above and behind the chamber. E3, E4, and E~ converge to another point. N~ is believed to be the track of an argon atom recoiling after elastic collision with a neutron. The heavy nucleus track, Nt, emerges from one of the centers of disintegration and ends in the chamber, with a sharp bend. T h e range of this particle m u s t have been > 5 ram. in lead (Mag, o.54). F I G . 2.
Stoss with 4 electron tracks, 2 neutron-recoil tracks, N1 and Ns, and a long, diffuse nucleus track, N~, near the bottom of the chamber (Mag. o.54).
D e c . , ~933.]
CLOUD
PHOTOGRAPI[S
Ol."
ST6SSE.
677
FIG. 3.
Stoss with x3 tracks of the same "age." Eleven of these converge in a lead block 40 cm. above and in front of the cloud chamber; the other two appear to be secondary branches originating in the glass lid of the cloud chamber (Mag. o.s7).
F I G . 4.
Stoss with z5 tracks: II simultaneous pre-expansion electron tracks, 3 neutron*recoil argontracks, NI, NI, a n d N,, and a post-expansion electron track. Most of the electron tracks traverse the chamber in nearly vertical directions, b u t could not have come from less than 4 points of convergence (Mag. o.83).
678
GORDON L. LOCHER.
[J. F. I.
others. 13 Conspicuous features found in nearly every burst p h o t o g r a p h e d are the non-convergence to a single point of the (simultaneous) tracks of each burst, and the presence of very short nucleus-tracks, originating and ending in the gas, which closely resemble the recoil-atom tracks formed by beryllium neutrons in the same atmosphere. Blackett and Occhialini n FIG. 5.
Four electron tracks of the same age, and'a neutron-recoil argon atom, N. Et and E2 do not intersect; Ea and E* seem to intersect in the iron of the arc used for illumination (Mag. o.83).
m e n t i o n cases of showers e m a n a t i n g from more than one point, and the burst found by Anderson showed the same characteristic. If we postulate t h a t each burst derives its energy from a single primary entity, we must suppose that the primary entity generates secondary ones, which, in turn, are very e~cacious in 1~ I n p r e p a r i n g these p h o t o g r a p h s for reproduction, some faint t r a c k s a n d o t h e r detail h a v e become lost; t h e density relations between t h e t r a c k s are less faithful t h a n on t h e original negatives, f r o m which t h e n u m b e r s , kinds, a n d points of convergence were, of course, t a k e n .
D e c . , 1933. ]
CLOUD
PHOTOGRAPHS
OF
STOSSE.
679
setting up tertiary centers of disruption, since the (tertiary) track-loci are relatively close together. It seems very unlikely to the writer t h a t a single primary cosmic ray establishes the observed points of disintegration directly, w i t h o u t the aid of a secondary agent (that is presumedly non-ionizing), since in one case, at least, the several track-loci are not collinear. On the basis of the evidence so far available, it seems improbable to the writer t h a t the secondary entities are photons. T h e FIG. 6.
Stoss with at least 9 electron tracks and one nucleus track in the cloud chamber. Et and E2 nearly horizontal and in line with the arc; 1/3, E4, and E~ are nearly parallel and vertical; ET, Ea and E. diverge upward from a point near the bottom of the chamber. N is believed to be a neutron-recoil nucleus; it is somewhat out of focus, and appears larger than it really is (Mag. o.8). are
necessarily high energies of such photons would be incompatible with the observed facility of their absorption by m a t t e r near the cloud chamber, unless their numbers were so very great as to impose excessive requirements on the energy drawn from the primary entity. (On the other hand, it seems reasonable to expect t h a t some low-energy quanta, at least, are set free from St6sse.) Against the possibility t h a t the secondary agents are ionizing particles, is only the fact t h a t on no picture was there a track observed which linked two track
680
GORDON L. LOCHER.
[J. F. I.
loci together. F u r t h e r cloud pictures of bursts should decide this question. Figure 7 shows photomicrographs of some of the short " r e c o i l - a t o m " tracks from St6sse, also some recoil-atom tracks from Be neutrons, in the same cloud atmosphere, for cornFro. 7.
A through J, short nucleus tracks from cosmlc-ray StiSsse photographs, attributed to argon atoms recoilingfrom elasticcollisionswith neutrons emitted from the St6sse. K through P, recoilatom tracks formed in the same atmosphere, but with beryllium neutrons from a Po-Be source. T h e energies of the atoms of the two groups are of the same order of magnitude (Mag. 3.9).
parison of the ionization density and energy. The similarity of the two groups is very evident. T h e writer believes, accordingly, t h a t neutrons are generated, or liberated, in cosmic-ray St6sse. Tracks of this kind are recongizable with considerable certainty because of the enormous density of
Dec., I933.]
CLOUD PHOTOGRAPHS OF STOSSE.
68I
their ionizatioL1. T h e use of argon probably facilitates the detection of the neutrons considerably; Bonner 14 found from ionization m e a s u r e m e n t s t h a t the target area of the argon a t o m for Be neutrons is about 17 times t h a t of hydrogen, or 4.85 times t h a t of nitrogen. Since the tracks of the St6sse do not converge to single points, it is impossible to tell from what material the neutrons arise ; but the infrequency of appearance of recoil-atom tracks on pictures other t h a n those of St6sse, indicates t h a t they somehow arise from disintegration processes, so t h a t their presence here need n o t be interpreted as contradictory to the results of Curie and Joliot, 15 who failed to detect neutrons in primary cosmic rays. The numbers of short recoil-atom tracks from St/~sse is about the same as the n u m b e r of Be neutron recoil-atom tracks from a Po-Be source of a b o u t 0.05 millicurie radium equivalent, placed on top of the cloud chamber, or I to IO millicuries at the Stoss track-loci. But the energy and ionization characteristics of the Stoss neutrons are unknown, hence a comparison of their n u m b e r with those of Be neutrons, from the relative n u m b e r s of recoil atoms, is a dubious extrapolation of ideas. T h e (secondary) agents responsible for ejecting the convergent groups of tracks found in St6sse are still unidentified. If the low-energy recoil tracks observed are produced by highenergy neutrons, the latter might satisfy the difficult requirem e n t s of an agent t h a t has extremely high energy, b u t produces n u m e r o u s subsidiary points of disintegration within a small region. But this hypothesis is based rather on our lack of knowledge of the processes by which such neutrons lose their energy, than on our knowledge of them. These experiments are being continued with masses of a l u m i n u m about the cloud chamber, in place of lead, in order to determine whether the more intense ionization produced by A1 St6sse s results from the generation of m a n y nucleus tracks, or from more a b u n d a n t electron tracks. Meanwhile, an a p p a r a t u s now under construction for another experiment will be used for examining the remote possibility t h a t the St6sse are manifestations of the arrival of 14 Bonner, Phys. Rev., 43, 871 (1933). 16Curie and Joliot, .Tour. Phys. et Rad., 4, 492 (1933).
682
GORDON L. LOCHER.
[J. F. I.
infrequent, instantaneous, very intense showers of primary cosmic-ray photons or particles from interstellar space, perhaps covering large areas, and possessing the power to disintegrate heavy nuclei. However enormously improbable such an occurrence may seem, because of the energy involved in such a shower, it would explain the closeness of the centers of disintegration indicated by Stoss photographs, and seems not incompatible with any other cosmic-ray experiment so far performed. The apparatus consists of eight triple-coincidence counter sets, operating independently, but recording the discharges on the same moving film. The sets, mounted vertically, and capped with blocks of massive material will be operated several meters apart. Their coincidence records will be examined for simultaneous discharges of two or more sets. with especial regard to the geographical positions of the sets concerned. Information should also be forthcoming from this experiment in regard to the branching of high-energy corpuscular tracks at high altitudes. The geometry of the sets and the distances between them will determine the lowest point from which such branched tracks could produce simultaneous discharges. The writer is indebted to Dr. W. F. G. Swarm for suggesting the cloud-photography of StGsse, for discussion of the subject, and for criticisms of this paper.
The Franklin Instilule Devoted to Science and the Mechanic Arts Vol. 216
DECEMBER, 1933
No. 6
CLOUD PHOTOGRAPHS OF COSMIC-RAY STOSSE.* BY G O R D O N L. L O C H E R , Ph.D., National Research Fellow.
INTRODUCTION.
Large instantaneous increments in the FOUNDATION ionization by cosmic rays are found in io~fizacommunication~o.S0, tion chambers that are surrounded by, or in the vicinity of, massive material. These large fluctuations are generally called "St6sse," "bursts." or "spurts of ionization" ; their magnitudes sometimes correspond to the passage of several thousands of electron tracks throught the measuring apparatus. The St6sse have mostly been studied by ionization methods; 1, 2, 3, 4, 5, 6, 7, 8 S w a n n and Montgomery 9 c o m b i n e d an ionization chamber with a coincidence set of GeigerM filler counters, finding that large increments in the ionization were frequently accompanied by the simultaneous discharge of
BARTOL RESEARCH
* Part of this paper has been printed in the Physical Review. ' Hoffmann and Pforte, Phys. Zeits., 3 I, 347 (I93o). 2 Steinke, Phys. Zeits., 3x, lOI9 (I93O). Schindler, Zeits. fiir Phys., 72, 625 (1931). 4 Steinke and Schindler, Zeits. fiir Phys., 75, 115 (I932). Messerschmidt, Zeits. fiir Phys., 78, 668 (I932). n Swann and Street, Phys. Rev., 40, Io49A (1932). 7 Messerschmidt, Naturwiss., 21, 285 (1933). 8 Steinke, Gastell and Nie, Naturwiss., 21,560 (1933). 0 Swann and Montgomery, Phys. Rev., 43, 782A (1933); Phys. Rev., 44, 52 (I933). (Note.--The Franklin Institute is not responsible for the statements and opinions advanced by contributorsto the JOURNAL.)
VOL. 216, NO. I296--46
673
674
GORDON L. LOCHER.
[J. F. I.
the counters, which were about a meter apart, outside the chamber. T h e infrequency with which St6sse occur makes the accumulation of d a t a on t h e m slow. (Relatively large bursts are produced, roughly, at 3 per hr. per IOO,OOO cc. of absorber.) At the present time it can be said t h a t the rate of their production is nearly proportional to the number of atoms in the absorbing material, for dense m a t t e r ; t h a t m a n y of the particles have ranges of more t h a n I0 cm. in lead (some 2o cm.), and t h a t the ionization per Stoss depends on the atomic n u m b e r of the absorber in which it is generated. F r o m a m a x i m u m ionization j u m p corresponding to about 4ooo electron tracks or 26o proton tracks, observed in one instance, 8 the bursts appear to taper off to sizes t h a t are too small to be recognized, with no sharply defined classes of size. F r o m the information now available, it seems impossible to tell whether the ejected particles derive their energy wholly from the primary cosmic ray, or draw part of it from the nuclei of disrupted atoms, nor can we say, in view of the discovery of the apparent synthesis of particles from photons, whether the particles ejected from St6sse all previously existed in the disrupted atoms; and, finally, we cannot be certain t h a t a Stoss is always produced by a single primary cosmic ray, rather t h a n by a shower of p r i m a r y disrupting agents from outside the earth's atmosphere. T h e m e t h o d of photographing cloud tracks of the disintegration products of St6sse promises to be a fruitful means of a t t a c k on the problem. But for a Stoss rate of I per hr., the chance of getting a significant photograph is about I/36,000, if the chamber is operated in the customary manner. Anderson 10 did accidentally get one burst, b u t the solution lay in a simple modification of the m e t h o d of Blackett and Occhialini, 11who automatically set off the expansion of a cloud machine immediately after the ionizing particle discharged a coincidence pair of cosmic-ray counters in line with the cloud chamber. In the course of their experiments, they not only verified the observation t h a t the particles discharging the lo Anderson, Phys. Rev., 43, 368 (1933). n Blackett and Occhialini, Proc. Roy. Soc. A, I39, 699 (I933).
Dec., 1933.]
CLOUD PHOTOGRAPHS OF STOSSE.
675
counters produce t r a c k s y but obtained photographs of several interesting bursts, as well. EXPERIMENTAL ARRANGEMENTS.
A cloud chamber 15 cm. in diameter has been arranged to be set off immediately after the simultaneous discharge of three Geiger-Mfiller counting tubes, each IOO cm3 in crosssectional area, placed two above and one below the chamber, but as far out of line as possible; no single particle, nor downward-directed branched pair originating near the chamber can set off the counters. The smallest sphere containing the counters, which have a maximum separation of 80 cm., has a volume about 500 times t h a t of the cloud chamber, so that the fraction of the tracks from the St6sse which pass through the cloud chamber may be small, if the point of origin is remote. Of the triple-coincidence counting rate, 1.26 per hr., it is estimated from the photographs taken that about one-third of the discharges are from St6sse, and two-thirds from branched tracks and small groups of tracks. Surrounding the chamber are about 135o lbs. of massive material, consisting of 750 lbs. of lead, 400 Ibs. of copper and brass, and 2o0 lbs. of iron. A cloud atmosphere of argon at 64. 5 cm. pressure is used. The specific ionization of tracks in argon is high ; the relatively high viscosity makes the tracks persist better than in air; the expansion ratio is small (1.18, as compared with 1.3 for air), and the probability of obtaining elastic-collision recoil atoms from any neutroas present is relatively high. Since the ionizing particles traverse the chamber about o.oi sec. before the expansion, the tracks are somewhat diffused by thermal agitation and by the expansion; equality of diffuseness of tracks of like densities serves to establish approximate simultaneity of their formation, and makes it easy to distinguish between pre- and post-expansion tracks. Stereophotographs are taken with two cameras whose optic axes intersect in the chamber at an angle of 2 2 ° . The connecting link between the counter set and the cloud machine is a thyratron (G.E. Co. Type FG 67), which acts through a quick-acting electromagnetic trigger to release a 12Mott-Smith and Locher, -Phys. Rev., 38, I399 (I931).
676
GORDON L. LOCHER.
[J. F. I.
mechanical drive mechanism that is normally held in readiness by vacuum force and spring tension. RESULTS AND
DISCUSSION.
Figures I, 2, and 3 are stereophotographs of interesting St6sse; Figs. 4, 5, and 6 are single photographs of three F I G . I.
Stoss with two nucleus tracks (N), and S electron tracks (E). E~, E2, and N~ converge to a point in the lead, above and behind the chamber. E3, E4, and E~ converge to another point. N~ is believed to be the track of an argon atom recoiling after elastic collision with a neutron. The heavy nucleus track, Nt, emerges from one of the centers of disintegration and ends in the chamber, with a sharp bend. T h e range of this particle m u s t have been > 5 ram. in lead (Mag, o.54). F I G . 2.
Stoss with 4 electron tracks, 2 neutron-recoil tracks, N1 and Ns, and a long, diffuse nucleus track, N~, near the bottom of the chamber (Mag. o.54).
D e c . , ~933.]
CLOUD
PHOTOGRAPI[S
Ol."
ST6SSE.
677
FIG. 3.
Stoss with x3 tracks of the same "age." Eleven of these converge in a lead block 40 cm. above and in front of the cloud chamber; the other two appear to be secondary branches originating in the glass lid of the cloud chamber (Mag. o.s7).
F I G . 4.
Stoss with z5 tracks: II simultaneous pre-expansion electron tracks, 3 neutron*recoil argontracks, NI, NI, a n d N,, and a post-expansion electron track. Most of the electron tracks traverse the chamber in nearly vertical directions, b u t could not have come from less than 4 points of convergence (Mag. o.83).
678
GORDON L. LOCHER.
[J. F. I.
others. 13 Conspicuous features found in nearly every burst p h o t o g r a p h e d are the non-convergence to a single point of the (simultaneous) tracks of each burst, and the presence of very short nucleus-tracks, originating and ending in the gas, which closely resemble the recoil-atom tracks formed by beryllium neutrons in the same atmosphere. Blackett and Occhialini n FIG. 5.
Four electron tracks of the same age, and'a neutron-recoil argon atom, N. Et and E2 do not intersect; Ea and E* seem to intersect in the iron of the arc used for illumination (Mag. o.83).
m e n t i o n cases of showers e m a n a t i n g from more than one point, and the burst found by Anderson showed the same characteristic. If we postulate t h a t each burst derives its energy from a single primary entity, we must suppose that the primary entity generates secondary ones, which, in turn, are very e~cacious in 1~ I n p r e p a r i n g these p h o t o g r a p h s for reproduction, some faint t r a c k s a n d o t h e r detail h a v e become lost; t h e density relations between t h e t r a c k s are less faithful t h a n on t h e original negatives, f r o m which t h e n u m b e r s , kinds, a n d points of convergence were, of course, t a k e n .
D e c . , 1933. ]
CLOUD
PHOTOGRAPHS
OF
STOSSE.
679
setting up tertiary centers of disruption, since the (tertiary) track-loci are relatively close together. It seems very unlikely to the writer t h a t a single primary cosmic ray establishes the observed points of disintegration directly, w i t h o u t the aid of a secondary agent (that is presumedly non-ionizing), since in one case, at least, the several track-loci are not collinear. On the basis of the evidence so far available, it seems improbable to the writer t h a t the secondary entities are photons. T h e FIG. 6.
Stoss with at least 9 electron tracks and one nucleus track in the cloud chamber. Et and E2 nearly horizontal and in line with the arc; 1/3, E4, and E~ are nearly parallel and vertical; ET, Ea and E. diverge upward from a point near the bottom of the chamber. N is believed to be a neutron-recoil nucleus; it is somewhat out of focus, and appears larger than it really is (Mag. o.8). are
necessarily high energies of such photons would be incompatible with the observed facility of their absorption by m a t t e r near the cloud chamber, unless their numbers were so very great as to impose excessive requirements on the energy drawn from the primary entity. (On the other hand, it seems reasonable to expect t h a t some low-energy quanta, at least, are set free from St6sse.) Against the possibility t h a t the secondary agents are ionizing particles, is only the fact t h a t on no picture was there a track observed which linked two track
680
GORDON L. LOCHER.
[J. F. I.
loci together. F u r t h e r cloud pictures of bursts should decide this question. Figure 7 shows photomicrographs of some of the short " r e c o i l - a t o m " tracks from St6sse, also some recoil-atom tracks from Be neutrons, in the same cloud atmosphere, for cornFro. 7.
A through J, short nucleus tracks from cosmlc-ray StiSsse photographs, attributed to argon atoms recoilingfrom elasticcollisionswith neutrons emitted from the St6sse. K through P, recoilatom tracks formed in the same atmosphere, but with beryllium neutrons from a Po-Be source. T h e energies of the atoms of the two groups are of the same order of magnitude (Mag. 3.9).
parison of the ionization density and energy. The similarity of the two groups is very evident. T h e writer believes, accordingly, t h a t neutrons are generated, or liberated, in cosmic-ray St6sse. Tracks of this kind are recongizable with considerable certainty because of the enormous density of
Dec., I933.]
CLOUD PHOTOGRAPHS OF STOSSE.
68I
their ionizatioL1. T h e use of argon probably facilitates the detection of the neutrons considerably; Bonner 14 found from ionization m e a s u r e m e n t s t h a t the target area of the argon a t o m for Be neutrons is about 17 times t h a t of hydrogen, or 4.85 times t h a t of nitrogen. Since the tracks of the St6sse do not converge to single points, it is impossible to tell from what material the neutrons arise ; but the infrequency of appearance of recoil-atom tracks on pictures other t h a n those of St6sse, indicates t h a t they somehow arise from disintegration processes, so t h a t their presence here need n o t be interpreted as contradictory to the results of Curie and Joliot, 15 who failed to detect neutrons in primary cosmic rays. The numbers of short recoil-atom tracks from St/~sse is about the same as the n u m b e r of Be neutron recoil-atom tracks from a Po-Be source of a b o u t 0.05 millicurie radium equivalent, placed on top of the cloud chamber, or I to IO millicuries at the Stoss track-loci. But the energy and ionization characteristics of the Stoss neutrons are unknown, hence a comparison of their n u m b e r with those of Be neutrons, from the relative n u m b e r s of recoil atoms, is a dubious extrapolation of ideas. T h e (secondary) agents responsible for ejecting the convergent groups of tracks found in St6sse are still unidentified. If the low-energy recoil tracks observed are produced by highenergy neutrons, the latter might satisfy the difficult requirem e n t s of an agent t h a t has extremely high energy, b u t produces n u m e r o u s subsidiary points of disintegration within a small region. But this hypothesis is based rather on our lack of knowledge of the processes by which such neutrons lose their energy, than on our knowledge of them. These experiments are being continued with masses of a l u m i n u m about the cloud chamber, in place of lead, in order to determine whether the more intense ionization produced by A1 St6sse s results from the generation of m a n y nucleus tracks, or from more a b u n d a n t electron tracks. Meanwhile, an a p p a r a t u s now under construction for another experiment will be used for examining the remote possibility t h a t the St6sse are manifestations of the arrival of 14 Bonner, Phys. Rev., 43, 871 (1933). 16Curie and Joliot, .Tour. Phys. et Rad., 4, 492 (1933).
682
GORDON L. LOCHER.
[J. F. I.
infrequent, instantaneous, very intense showers of primary cosmic-ray photons or particles from interstellar space, perhaps covering large areas, and possessing the power to disintegrate heavy nuclei. However enormously improbable such an occurrence may seem, because of the energy involved in such a shower, it would explain the closeness of the centers of disintegration indicated by Stoss photographs, and seems not incompatible with any other cosmic-ray experiment so far performed. The apparatus consists of eight triple-coincidence counter sets, operating independently, but recording the discharges on the same moving film. The sets, mounted vertically, and capped with blocks of massive material will be operated several meters apart. Their coincidence records will be examined for simultaneous discharges of two or more sets. with especial regard to the geographical positions of the sets concerned. Information should also be forthcoming from this experiment in regard to the branching of high-energy corpuscular tracks at high altitudes. The geometry of the sets and the distances between them will determine the lowest point from which such branched tracks could produce simultaneous discharges. The writer is indebted to Dr. W. F. G. Swarm for suggesting the cloud-photography of StGsse, for discussion of the subject, and for criticisms of this paper.