The rate of sputtering of magnesium in oxygen was observed with an apparatus which permitted the field immediately in front of the target to be reversed. Change in the sign of the field did not affect the rate of sputtering which remained abnormally low. It is therefore concluded that, while the low rate is in some way due to the presence of an oxide film, it is not due to charged particles being returned to the cathode by the electric field, as had been suggested by other workers.

Photometrical measurements of the photographic images of cloud tracks of the disintegration of boron by slow neutrons,

have enabled the ranges of both the He and the Li particle to be determined. We find that the He particle has a range of 7·0 ± 0·3 mm. and the Li particle a range of 4·3 ± 0·2 mm. in standard air.

We deduce that the 7Li nucleus is usually formed in an excited state, with energy of excitation of either 0·5 or 0·8 M.e.V. The wide latitude in excitation energy is due to the uncertain state of our knowledge of the energy-range relation for α-particles.

The disintegrations by which Ac B passes into the inactive AcPb are accompanied by a γ-radiation very weak compared with the intense γ-emission in the case of Ra or Th-active deposit. The analysis of the secondary β-ray spectrum of actinium-active deposit has revealed the existence of at least five γ-rays (1) (see Table I), of which the ray with energy 0·349 × 106 e.V. definitely belongs to the disintegration Ac C—C″ and is associated with the fine-structure of α-particles of Ac C. According to the measurements of Surugue the two rays of 0·4038 × 106 and 0·4257 × 106 e.V. energy are to be attributed to the disintegration AcB—C, whereas the origin of the 0·829 × 106 e.V. ray is less definite. The fit with experiments is best if this ray is assumed to be emitted from Ac B—C; but it may also (within experimental error) be attributed to the disintegration Ac C″—Pb. Experiments on the absorption coefficient of the γ-radiation of RaAc and its disintegration products prove that the 0·829 × 106 e.V. ray is the hardest γ-ray emitted by the active deposit of actinium(2). The value for the absorption coefficient between 4·6 and 10·6 cm. of lead was found to be μ/ρ = 0·76, in good agreement with the value found in previous experiments at smaller absorptions in aluminium (3). There is, on the whole, excellent agreement concerning the energies of the γ-components as measured according to different methods by different observers.

No ionization was found in hydrogen leaving palladium, either when the gas diffused through the metal or was evolved by warming palladium previously charged with hydrogen by electrolysis in dilute sulphuric acid.

The method of coincidence counting has been applied to an investigation of the possible production of positron-electron pairs by the high energy β-particles from a source of uranium X in absorbers of aluminium, brass and lead. The results are not inconsistent with the high values recently reported for atomic cross-sections for the effect, nor with the suggestion that the atomic cross-section is proportional to the first power of the atomic number rather than the second. Suggestions are made for the use of the β-particles from artificially radioactive substances in an attempt to increase the sensitivity of the method.

This note describes measurements of the velocity of sound in liquid nitrogen by a method which has previously been used in the case of liquid oxygen(1). The method depends on the diffraction of light by suprasonic waves.

The magnetic anisotropy of crystals of calcite and sodium nitrate does not alter between room temperature and liquid nitrogen temperature. By using such diamagnetic crystals, it is easy to compare strong magnetic fields to 1 part in 1000. The effect is large enough for the measurement of fields of the order of 100 gauss, but in small fields the accuracy is limited by ferromagnetic contamination.

The radiations from Na24 have been examined by the absorption and coincidence methods and evidence has been obtained for the conclusion that the energy spectrum of the β-particles is a simple continuous spectrum with an upper limit at 1·40 × 106 e.V., and that the emission of γ-radiation after the particle disintegration takes place almost always as a cascade process in which at least two quanta are involved, and very rarely by a single transition to the ground state.

The previous measurements made on the surface tension of liquid helium by Onnes(1) were carried only as far down in temperature as 1·5° K. Since we had apparatus set up suitable for such measurements and capable of reaching temperatures nearly half a degree lower, we considered it worth while to repeat the measurements. Not only was this done for the purpose of extending them to lower temperatures but also to investigate carefully the variation of surface tension in the region of the λ-point (2·186° K.), the transition temperature between Hei and Henii.

Photometrical determinations of the variation of density along photographs of expansion champer tracks have enabled the positions of maximum ionizing efficiency for single α-particles, protons, deuterons, and 3H particles to be determinated.

The maximum rate of ionization for an α-particle in air occure when it has a residual range of 3·7 mm. in standard air.

The distance of the position of maximum ionizing efficiency from the end of the particle range is shown to be approximately in the ratio of the particle mass in the case of the particles 1H, 2H, and 3H.

These experiments were carried out in the Cavendish High Tension Laboratory.

By measuring the resistance of a phosphor bronze wire in thermal equilibrium via various substances with crystals of iron ammonium alum it is shown that the time for thermal equilibrium between the ionic magnets of the salt and its lattice vibrations is less than 0·5 sec. for all temperatures above 0·025° T*. When liquid helium or a german silver tube forms part of the cooled portion of the apparatus, the time for equilibrium is increased to a few seconds for temperatures below 0·4° K. It appears possible that the thermal conductivity of german silver is less than 10−8 cal. cm.−1 sec.−1 degree−1 below 0·05° T*, and there are indications that the thermal conductivity of liquid helium at temperatures below 0·3° K. is small compared with its value at 2° K.

The purpose of this note is to draw attention to a certain correspondence between the melting-points of normal paraffins and of fatty acids and to indicate a simple interpretation of this phenomenon. If the number of carbon atoms in a normal paraffin is plotted against the corresponding melting temperature, all the points in the diagram lie very close to a smooth curve drawn among them. If a similar diagram is made for the fatty acids, the points corresponding to an even number of carbon atoms lie on a curve of the same character as that found for paraffins. The points for an odd number of carbon atoms lie on a separate but similar curve, exemplifying the well-known alternation property.

In a previous publication (1) the author described experiments in which the “overshoot phenomenon” reported by Silsbee, Scott and Brickwedde (2) might have been expected but was not observed. The specimens used were straight wires mounted so as to be completely surrounded by the liquid helium bath. Silsbee and his co-workers used wires coiled on paper or glass cylinders. It was concluded that the “overshoot phenomenon” was due to this method of mounting the specimen and was therefore a secondary effect.