NEARBY STARS IN PHOTOGRAPHIC PLATES EXPOSED IN THE STRATOSPHERE

In work¹, stars produced by cosmic radiation were observed in photographic plates lifted by balloon pilots to an altitude of about 27 km. The authors measured the number of stars \(P\) located from one another at distances smaller than \(r\), and compared it with the expected number of such stars \((Q)\), obtained under the assumption that the stars are distributed in the emulsion at random. A total of 1077 stars in 19 photographic plates were subjected to such investigation. The results of the measurements are shown in the figure, on whose abscissa axis the distance between the centers of the stars is plotted, and on the ordinate axis the difference \(P - Q\). If the distribution of distances

between stars were in fact Poissonian, then the measured differences \(P - Q\) would be grouped around the straight line \(P - Q = 0\). It is evident from the figure that the observed distribution at small distances (less than 1500 microns) differs strongly from the Poisson distribution, showing a maximum for distances \(\sim 1000\) microns. The statistical accuracy of this result, as is evident from the figure, is very high. An analogous result was obtained also in work², whose authors observed 1723 stars in plates exposed at an altitude of about 25 km. They found that the number of nearby stars separated by distances smaller than 1500 microns was three times greater than the expected number of such stars. According to their estimate, the probability that such a deviation from the Poisson distribution is random is close to 0.001. Similar phenomena were noted in works³,⁴,⁵; moreover, in work⁵ it is indicated that in plates exposed at sea level the difference \(P - Q\) is equal to 0, while upon ascent to the altitude of mountains the difference \(P - Q\) becomes greater than 0, revealing a flat maximum in the region of distances of about 1500 microns. Several explanations were proposed for this phenomenon of paired stars:

a) One of the stars of a pair emits a neutral particle, which produces the second star in the immediate vicinity of the first.

b) Nearby pairs of stars are produced by particles passing through the emulsion whose nature is unknown.

c) Pairs of stars are produced by a collimated shower of particles of unknown nature.

Hypothesis a) can be checked if one attempts to estimate the product \(q\sigma\), where \(q\) is the mean number of such neutral particles in a star, and \(\sigma\) is the cross section for the production of stars by neutral particles. Obviously, the quantity \(q\sigma\) depends on \(P - Q\), and the authors estimated the mean value of \(q\sigma\), knowing \(P - Q\) from their own data and from the data of other authors.

In doing so, they obtained that \(q\sigma\) is equal to \(28 \cdot 10^{-24}\ \text{cm}^2\). The mean effective cross section of the nuclei of the emulsion is \(0.67 \cdot 10^{-24}\ \text{cm}^2\), whence it follows that the number of hypothetical neutral nuclear-active particles arising in stars must be close to 40, which, of course, cannot correspond to reality. If, however, one assumes that 2–3 such mesons are produced, then the effective cross section for the formation of stars by these mesons should be 15–20 times greater than the geometrical cross section of the emulsion nucleus, which is also impossible. Thus hypothesis a), at least in its simplest form, contradicts the experimental data. Hypotheses b) and c) likewise cannot be accepted: if these particles interact with the nuclei of the emulsion with an effective cross section equal to the geometrical one, then the probability of the formation of one particle and two close stars separated by a distance of 1 mm is extremely small.

Thus, from the works considered it follows that the existence of an anomalous number of nearby stars in photographic plates is apparently proved, although at present there is no reasonable explanation of this phenomenon.

A. V.

References Cited

1. Davis, Marion, Delord, Kind, Phys. Rev. 88, No. 2, 368 (1952).
2. Brown, Masket, Phys. Rev. 88, No. 5, 1204 (1952).
3. Leprince-Ringuet, Heidemann, Nature 161, 844 (1948).
4. Li, Perkins, Nature 161, 844 (1948).
5. Li, Phil. Mag. 41, 1152 (1950).