74 proceedings: philosophical society 



the wire and is a linear function of the energy appHed for all gases. 

 In hydrogen the critical voltages are much lower than in air and its 

 characteristics are such that almost complete I'ectification of alter- 

 nating currents is possible. When corona is present the potential dis- 

 tribution curves are much distorted from the theoretical position and 

 indicate an accumulation of charges near the electrodes. Roughness or 

 oxidation of the wire lowers the critical point and shifts the characteristic 

 I-V curves. Different metj^ls also show different characteristics. 



A short series spark destroys the bead formation on the negative wire, 

 produces divergent pencils, brushes or streamers on the positive wire, 

 and reverses the position of the characteristic I-V curves. A special 

 hot-lime-cathode Braun tuloe showed the currents through the spark 

 and the gas to be unidirectional pulses. 



The uniform positive glow is explained as ionization by collision of gas 

 particles near the wire. The negative beads may be due to a combina- 

 tion of ionization by collision and electron discharge. The series 

 spark impulses may cause a disruptive action in the surface of the wire 

 shooting off high velocity positive ions which give rise to the positive 

 sti'eamers. The apparent destruction of the negative beads is probably 

 a superimposed building up and decay of the observed negative dis- 

 charge forms for each spark impulse. 



Discussion: The paper was discussed by Messrs. Agnew and L. J. 

 Briggs. 



The 794th meeting was held at the Cosmos Club November 10, 1917; 

 President Buckingham in the chair; 38 persons present. 



By invitation Mr. L. W. McKeehan gave an illustrated paper on 

 Diffusion of, and recoil from, actinium emanation. Previous work on 

 the diffusion of actinium emanation and on the distribution of its 

 active deposit between charged metal plates was reviewed, and the 

 causes of some uncertainties in the interpretation of the results were 

 pointed out. Apparatus was designed to avoid the defects thus dis- 

 covered, and measurements made under a variety of conditions. The 

 value of the diffusion coefficient of the emanation into air at the mean 

 temperature of the experiments, 20.7°C., was found to be 0.109 

 (plus or minus 2 per cent). The distribution of the active deposit 

 between positive and negative plates at different pressures agreed with 

 that to be expected on the basis of radioactive recoil from the dis- 

 integrating emanation, and gave as the range of recoil in air at atmo- 

 spheric pressure and at 20.7°C. the value 0.0092 cm. A closer theo- 

 retical study made it seem probable that the recoil paths are riot straight, 

 and preliminary work by the author and another, using C. T. R. Wil- 

 son's photographic method for studying the initial portions of a-ray 

 trails, is now in progress. 



Mr. N. E. Dorset gave an illustrated paper on Radium luminous 

 materials. No abstract. 



Discussion: The paper was discussed by Messrs. Bauer and E. B. 

 Stevenson. 



