A method of studying magnetic properties in alternating fields and its application to the intermediate state of a superconducting sphere are described. It is shown that the behaviour of the sphere in alternating fields of frequencies between 10 and 100 is different from that in very slowly varying fields, the difference becoming less for lower frequencies and for smaller radius. When the specimen is in the intermediate state, energy losses are produced in it by the alternating field, showing that the induced currents are out of phase with the alternating field. The induced currents do not, however, flow in the same way as in a homogeneous isotropic conductor, and this is probably on account of an anisotropic structure of the specimen when in the intermediate state. This structure depends strongly on the amplitude of the alternating field in such a way that the effective average conductivity is increased with decreasing amplitude, and this amplitude effect depends on the size of the specimen. Evidence is given for a time lag in the magnetocaloric effect, and this time lag decreases with decrease of the size of the specimen, being of order sec. for a sphere of about 1 cm. diameter.

In conclusion I wish to thank Prof. Lord Rutherford and Dr Cockcroft for their interest in this work; Dr Peierls for suggesting the experiment and for much valuable advice and assistance; Mr Shire for advice on some technical points; and various members of the staff of the Royal Society Mond Laboratory for help in the measurements.

Measurements have been made on the proportion of the effect induced in silver which is due to neutrons strongly absorbed by cadmium, using cylinders of water and boric acid of different sizes to slow down the neutrons. In addition to furnishing data for use in the foregoing paper, the results show that practically all the neutrons of groups A and B must have energies greater than any of those of group C.

The author desires to thank Lord Rutherford for his interest in the work and for the use of a Ra + Be source, and Dr Oliphant for discussion and advice. He is also indebted to the Department of Scientific and Industrial Research for a Senior Research Award.

Some applications of an alternating field method to various problems of superconductivity are described. It is shown how the method can be used to measure the resistance and self inductance of a metal ring, and this was applied to show how the resistance of a superconducting tin ring is restored by a magnetic field. The method is also convenient for a rapid determination of the order of magnitude of the specific resistance of a substance, and to see whether or not it becomes superconducting; several elements were examined in this way, but no new superconductors were found.

Measurements of the reflection coefficient at normal incidence on lead and tantalum mirrors in the supraconducting and non-supraconducting state did not reveal any change within the limits of accuracy (0·2−0·5% in R). This does not agree with a suggestion of R. de L. Kronig, according to which an increase in reflectivity of about 35% could have been expected.

A method of growing single-crystal resistance specimens free, to a large extent, from strains and impurities is described. The magneto-resistance effect in cadmium single crystals has been studied in some detail at the temperature of liquid nitrogen, using sufficiently high magnetic fields to observe the linear effect found by Kapitza.

Although it proved impossible, in general, to determine the orientation of individual crystals, the experiments suggest that the “critical field” of the linear effect is dependent on the orientation of the crystal with respect to the current and magnetic field, and not on the perfection of the crystal lattice.

It is concluded that the “overshoot phenomenon” in superconductivity described by F. B. Silsbee was due to his method of mounting the specimen.

Absorption experiments have been carried out on the β- and γ-rays of RaC”. The range of the β-rays in aluminium corresponds to an energy limit of (1·95 ± 0·15) × 106 e.v.; the absorption coefficient of the γ-rays in copper corresponds to a fairly homogeneous radiation of quantum energy equal to (5 ± 1) × 106 e.v. The energy of disintegration is, within the limits of error, in agreement with the value 6·1 × 106 e.v. to be expected from the known energies of the other disintegrations in the RaC branches.

The actual distribution of a homogeneous group of electrons by semicircular magnetic focusing is calculated for different types of source.

The dependence of maximum intensity on the radius of curvature of the path of electrons is investigated and the approximate power laws assigned to different cases. The results can be used in the measurement of the intensity of β-ray line spectra.

Tables of numerical values of effective solid angle subtended by the photographic plate at the source are given for some special cases.

The results of the above calculations have been verified experimentally.

According to recent experiments by Cosyns(1), in which the probability that a cosmic-ray electron actuates a Geiger-Müller Counter is observed, the rate of production of primary ions is practically the same in helium as in hydrogen. The value obtained for hydrogen (which agrees very closely with that previously obtained by Danforth and Ramsey (2) using the same method) is moreover little different from that found for fast β-particles in expansion-chamber experiments.