THE LONDON, EDINBURGH, AND DUBLIN PHILOSOPHICAL MAGAZINE AND JOURNAL OF SCIENCE. [SEVENTH SERIES.] SUPPLEMENT, NOVEMBER 1928. LXXI. Ionization Measurements of y-Rays. ME 1. Introduction. EASUREMENTS on y-rays can only be carried out by measuring some property of the secondary B-rays they release from matter. Various properties are available, and, in particular, the most direct information is yielded by measurements of the number (1), (2) or the energy (3) of the B-rays; but such experiments, owing to their length and difficulty, are not available for a great deal of the work on y-rays. The work has largely been carried out by measurement of the ionization produced in the gas in a chamber, and thus it is of great importance to consider what meaning may be attached to such measurements. Since the ionization is a tertiary effect, being produced by the secondary B-rays, it is clear that ionization measurements cannot be expected to give direct information about the primary y-rays. Though it has long been recognized, e. g. by Bragg (4) in 1910, that ionization from y-rays is a complex phenomenon, yet many workers have, in default of other data, been forced to assume that ionizations measure either number of quanta [e. g. Ahmad (5), Kohlrausch (6), etc.] * Communicated by Sir E. Rutherford, P.R.S. Phil. Mag. S. 7. Vol. 6. No. 38. Suppl. Nov. 1928. 3 C or y-ray energy [e. g. Miss Szmidt, Ellis and Wooster (3), and others in connexion with the RaB/RaC y-ray energy ratio]. The variation with frequency of the ionization per quantum is conveniently termed the "ionization function," and clearly might depend on the conditions used, e.g. on the shape, size, and material of the chamber, on the gas contained, on the relative positions of the source and chamber, etc. Ionization measurements of y-rays may be divided into two main classes: (a) the comparison of sources having the same spectral distribution of y-rays, as in measurements of standards and of decay periods; and (b) the comparison of y-rays of different qualities; this includes the standardization of thorium, and other, sources in terms of radium, e. g. (8), and also all absorption measurements, as frequency distribution must be altered by absorption. Measurements of class (a) cannot be affected by the conditions used, but those of class (6) may be so affected. The simplest case to consider is that of the comparison of two homogeneous 7-ray beams, and this can easily be extended to more complex cases; what we can actually measure are J1, J2, the ionizations produced by N1, N2 quanta respectively of the two beams; if I,, Ï1⁄2 are the values of the ionization function of the electroscope for the two frequencies of the y-rays, then J1 = N1I1, J2 = N2I2. If we keep the external conditions unaltered, the relative N1 number of quanta will be unaltered, i. e. is constant; J1 I1 and hence we can investigate the change in J2 I, 2 SO I1 the ratio of the values of the ionization function for two frequencies. 2. Objects and Nature of Experiments. The objects of the experiments carried out were to demonstrate as simply and unambiguously as possible the effect of the ionization function on y-ray measurements, to obtain an idea of the magnitude and to see how it must influence past and future y-ray ionization work. The source used was a radium standard of 7.65 mgm. giving y-rays almost entirely from RaB and RaC; this is particularly suitable as there are two distinct groups of y-rays, from RaB and RaC respectively, of considerable frequency difference, while each group is itself heterogeneous. Thus, if we can measure the relative ionizations of the two groups, or of two parts of the same group, we should expect the results to be dependent on the conditions if the ionization function has any appreciable effect. For measuring the ionizations a gold-leaf electroscope of a normal type was used, the leaf being suspended inside the chamber investigated. To vary the relative intensities of rays of different frequencies in the beam, absorption in lead sheets up to 3 cm. in thickness was used; the results can be plotted as the logarithm of the ionization against the thickness of absorber. It was found, as is well known, e. g. (9), that beyond about 1-8 cm. the points lie well on a straight line, i. e. that the absorption is exponential. The slope of the line gives the absorption coefficient of the RaC rays, and the intercept on the axis of zero absorption gives, vide (7), the fractions of the total ionization due to the RaB and RaC y-rays. For the complete description of the absorption curve, it is necessary to find also the absorption coefficient of the RaB rays by subtracting the effect of the RaC rays. It is not possible to make use of a change in ionization function due to altering the shape or size of the chamber or the relative positions of source and chamber, for, as has been pointed out by Keetman (10) and others, the absorption curves are here altered owing to differences in the y-rays entering the chamber, because of scattering. We are here concerned with the difference in the effects of similar beams. of y-rays, so we must keep all conditions outside the chamber constant. The changes in conditions that remain available are alterations of the gas or pressure inside the electroscope, and change of lining of the electroscope. The first experiments carried out were concerned with the effect of pressures above atmospheric; but, to an accuracy of about 4 per cent. in relative ionizations, the absorption curves showed no definite alteration for pressures up to 21 atmospheres, so description of the experimental arrangement is needless. The variation of the gas in the chamber was carried out in one instance by using CO2 in place of air, and the effect was inappreciable; but there are considerable features of interest in other work on the subject; Rutherford and Richardson (11) and Richardson (12) used methyl iodide in the chamber, and increased the effect of the RaB y-rays to 85 per cent. of the total effect, as compared with 10-20 per cent. for air; they also noticed that the absorption coefficient for the RaC rays is different |