Abstract
There exists some conflicting experimental data concerning the distribution of energy in the β-ray spectrum of radium E. The earliest work, carried out by Schmidt in 1907, using both the absorption and magnetic deflection methods, showed that the spectrum was apparently continuous but had a well-marked upper limit at about Hρ 5500 (energy about 10 6 volts). In 1911, O. v. Baeyer, Hahn and Meitner confirmed this result and Gray came to a similar conclusion in the following year. Danysz, however, using the method of magnetic focussing, while agreeing to the main distribution including the existence of the well-marked upper limit at 5500 Hρ, made a guarded statement concerning the possibility of the existence of β-rays of far higher velocities. “Sur ces mêmes clichés, il semble y avoir des indices d’un faisceau beaucoup plus rapide, de vitesse variable comme le premier (comprise entre les limites approximatives β = 0.94-0.99). Toutefois ces indices sont tellement faibles qu’il est nécessaire de recourir a une autre methode pour être sûr de son existence. Ce faisceau est beaucoup moins visible que le faisceau le plus rapide du radium C . . . ” In 1916, Kovarik and McKeehan, using the ionisation method, considered their results agreed with those of O. v. Baeyer, Hahn and Meitner. Curie and D’Espine, in 1925, using the method of direct magnetic deviation and a photographic plate, agreed to the sharp upper limit of the main distribution, but found also a very weak band between the limits of 7000-10,000 Hρ (energy 1.5 x 10 8 — 3.0 x 10 6 volts). It appears that there was no evidence of any direct connection between the two parts of the spectrum. The subject was considered again in 1927 by Yovanovitch and D’Espine and the previous results of the latter were confirmed. The same year, Madgwick, using the ionisation method, confirmed the existence of a sharp upper limit at 5000 Hρ and stated “there is no reason to doubt the accuracy of the endpoint.” In a short letter to 'Nature' in 1929, Gray and O’Leary mention an experiment, using the magnetic deflection and ionisation method, and conclude that less than one particle in 25,000 is emitted with Hρ greater than 8000. Recently, however, F. R. Terroux, using an expansion chamber, finds no indication of the sharp upper limit of the main spectrum at Hρ 5000, but a gradual tailing off to very high velocities in a manner which, he suggests, indicates a Maxwellian distribution. This would be a point of fundamental importance in attempting any theory of the primary disintegration process, for although it does not give any picture of the nucleus before emission, it imposes a strict criterion which any proposed mechanism must satisfy. In particular, he estimates the number of particles with Hρ greater than 5000 to be as high as 4 per cent, of the total number of particles emitted. This result is in direct contradiction to all previous work and it is clearly important to repeat the investigation using the same method. The advantages of the expansion chamber method are well known; in the problem in hand, the almost equal sensitivity to particles of all velocities is the point of greatest importance. The writer has recently obtained a large number of photographs of β-ray tracks in an automatic expansion chamber. The source was an old radon bulb and the experimental disposition such as to render the results suitable for the investigation of the upper half of the β-ray spectrum of radium E. The existence of the sharp upper limit at about 5000 Hρ has been fully confirmed. No evidence of any gradual tailing off was found and it is estimated that less than one particle in 2000 is emitted with Hρ greater than 5500.

This publication has 0 references indexed in Scilit: