LONDON. Royal Society, March 5.?W. L. Garstang and C. N. Hinshelwood: The kinetics of the combination of hydrogen and oxygen: the influence of iodine. The surface reaction in vessels of silica and porcelain is accelerated by iodine. The amounts of iodine required are small. The efficiency of collisions with iodine in breaking chains is, however, less than about 10-4. Water vapour lowers the upper critical limit of the low pressure explosion region and, in sufficient amount, inhibits the explosion altogether. Commercial hydrogen appears to contain no inhibiting substances.?D. R. Hartree: Optical and equivalent paths in a stratified medium, treated from a wave standpoint. Wave treatment is essential for the interpretation of some of the phenomena of reflection from the Heaviside layer. Expressions for the optical and equivalent paths in terms of the solutions of the equations of wave propagation are obtained, both for normal and oblique incidence, and exact expressions are obtained for the optical path for normal incidence on stratified media with certain simple variations of refractive index. Approximate expressions are obtained for the optical and equivalent paths in a totally reflecting medium.?H. J. Phelps: The adsorption of substances by fuller' s earth. The adsorption of weak solutions of various organic acids and bases has been studied; in particular, attention is directed to a study of the adsorption of such substances from solutions of various hydrogen-ion concentrations. The adsorption of the simple amines is molecular, or 'apolar', in solutions of reaction more alkaline than pH 9, while it is ionic, or 'polar', in solutions more acid than about pH. 8. The adsorption of bases and also of oxalic acid falls to zero in solutions more acid than pH. 3, due to the fact that the calcium salts in fuller' s earth, the presence of which is essential for adsorption to take place, are soluble under those conditions.?L. J. Mordell: The arithmetically reduced indefinite quadratic form in n variables. A simple method is given for finding inequalities satisfied by the coefficients of a reduced indefinite form. When the coefficients are integers a simple proof is given that the class number for forms with given determinant is finite. A normal type is also found for the reduced forms.?A. E. Moelwyn-Hughes and C. N. Hinshelwood: The kinetics of reactions in solution (1, 2). The decom position of chlorine monoxide in carbon tetrachloride solution proceeds at the same rate, possesses the same heat of activation, and apparently takes place by the same mechanism as in the gaseous state. The retardation of bimolecular reactions by solvents thus appears to depend upon a specific influence of particular solvents, and need not occur in an 'ideal' solution. The aim of the experiments was to determine whether constancy of rate and heat of activation with change of solvent can be regarded as the usual characteristic of a unimolecular reaction. It is concluded that it is only true of certain' ideal' solvents.?F. R. Terroux: The upper limit of energy in the spectrum of radium E. The velocities of the fastest s-rays from radium E were determined from the curvature of the tracks in a magnetic field and the distribution of the rays obtained in the region extending from 4000 to 10,000 H?. The complete spectrum of radium E was obtained by combining the present observations with previous work, and the resulting curve shows no trace of the hitherto accepted end point at about 5000 H?. The spectrum appears to tail off very gradually and to extend at least to 12,000 H?, which corresponds to an energy of 3,000,000 electron volts. The number of particles above 5000 H? is about 4 per cent of the total number emitted. An energy distribution curve was obtained, and from this the average energy per disintegration is estimated at about 473,000 volts, with a probable error of 20 per cent, which is in fair agreement with the values obtained from heating measurements. It appears probable that the?-particles from radium E are emitted from the nuclei according to a simple distribution law which resembles the Max-wellian form, and that there is a finite probability for the emission of a particle with any velocity.? K. R. Rao and J. S. Badami: Investigations on the spectrum of selenium?(1). The spectrum of selenium has been investigated from?7000 to?650, mainly using different intensities of discharge through capillary tubes containing vapour of selenium. Between?1400 and?400, vacuum spark spectra have been taken with and without inductance. With the aid of these data the doublet system of trebly-ionised selenium has been identified. A few singlets and inter-combination lines in Se V have been added to the triplet system that was already known. The largest term, 4s 1S0, is found to be 589,781. cm.-1, leading to an approximate ionisation potential of 72?8 volts.--J. K. L. Macdonald: Stark effect in molecular hydrogen in the range 4100-4770 A. Certain complex structures reported by Kinti are resolved into independent lines. Displacements of line components are measured and the observations are considered from a theoretical point of view. Certain groups of P- and R-branch lines with common initial levels are found to be adequately described, as regards number and polarisation of components, by a theory which is briefly discussed.?T. Alty: The reflection of vapour molecules at a liquid surface. The rate of evaporation from a water surface is measured as a function of the vapour pressure above the evaporating surface. By extrapolation to zero pressure the rate of evaporation into a vacuum is found. On comparing this experimental result with the formula of the kinetic theory of gases for the number of vapour molecules striking a water surface per second from the saturated vapour, it appears that only about 1 per cent of the molecules incident on the surface are able to enter the liquid.