Table of contents

When soluble matter is introduced into a solvent flowing slowly through a capillary tube it is dispersed longitudinally by a process which involves both the variation in fluid velocity over the cross section of the tube and radial diffusion by molecular agitation. Measurements of longitudinal dispersion provide a new means for measuring diffusion coefficients. Results obtained by this method will be given.

The stability of solutions contained in vertical tubes when the density increases upwards also depends on radial diffusion. Measurements of the equilibrium density gradient can be used as another new method for determining diffusion coefficients.

The mechanics of dispersion in turbulent flow through a pipe can be discussed by a method which is analogous to that used for streamline flow. The results of this calculation are compared with experiments in which brine was injected into water flowing in 3/8 inch and 40 inch pipes, and the subsequent dispersion along the pipe was measured. Similar comparisons are made with American measurements in long pipe lines.

The effect of intense magnetic fields on electroluminescent powder phosphors is investigated to furnish information which should be helpful in developing a theory of the mechanism of electroluminescence. No quenching effect was observed even for magnetic field intensities as high as 1.3 × 10⁵ oersteds. The significance of this result is discussed.

A gaussian distribution function for the variation of the axial ratio m of powder particles is suggested as the explanation for the measured values of the coercivity of γ-ferric oxide powders being less than the theoretical values for particles of the observed mean shape factor m. Superposition of the theoretical hysteresis loops for 20 packets of equal volume, each assumed to consist of particles with one value only of m, approximating to such a distribution gives values of bulk coercivity in very close agreement with measured values for two particular examples of widely different mean shape factors. In the light of these results explanations are offered of some other observed values for this oxide, and it is suggested that this mechanism may account in part for results obtained by Weil some years ago for precipitated ferronickels.

The following theoretical results on the abrasion of rubber have been deduced from a few simple assumptions concerning the initiation of the surface damage. The abrasion is proportional to the normal load, independent of the particle size of the abrasive if the particles are polyhedral, and proportional to their mean radius of curvature if they can be approximated to hemispheres. The spacing of the abrasion pattern is proportional to the cube root of the normal load, proportional to the two-thirds power of the particle size on an abrasive with polyhedral particles, and directly proportional to the particle size on an abrasive with hemispherical particles.

These predictions have been reasonably well confirmed by experiment.