388 Anniversary Meeting. [Nov. 30, 



starting with an examination of the cause of irregularities which had 

 troubled* him in his weighing of thallium ; and, going on to trials 

 for improving Cavendish's gravitational measurement, in the course 

 of which he discovered that the seeming attraction by heat is only 

 found in air of greater than 1/lOOOf of ordinary density ; and that 

 there is repulsion increasing to a maximum when the density is 

 decreased from 1/1000 to 36/1,000,000, and thence diminishing 

 towards zero as the rarefaction is farther extended to density 

 1/20,000,000. From this discovery Crookes came to his radiometer, 

 first without and then with electrification ; and, powerfully aided by 

 Sir George Stokes, J he brought all his work more and more into 

 touch with the kinetic theory of gases; so much so that when 

 he discovered the molecular torrent he immediately gave it its true 

 explanation molecules of residual air, or gas, or vapour pro- 

 jected at great velocities by electric repulsion from the negative 

 electrode. This explanation has been repeatedly and strenuously 

 attacked by many other able investigators, but Crookes has defended|| 

 it, and thoroughly established it by what I believe is irrefragable 

 evidence of experiment. Skilful investigation perse veringly con- 

 tinued brought out more and more of wonderful and valuable results : 

 the non-importance of the position of the positive electrode ; the pro- 

 jection of the torrent perpendicularly from the surface of the negative 

 electrode ; its convergence to a focus and divergence thenceforward 

 when the surface is slightly convex ; the slight but perceptible re- 

 pulsion between two parallel torrents due, according to Crookes, to 

 negative electrifications of their constituent molecules ; the change of 

 direction of the molecular torrent by a neighbouring magnet; the 

 tremendous heating effect of the torrent from a concave electrode 

 when glass, metal, or any ponderable substance is placed in the focus ; 

 the phosphorescence produced on a plate coated with sensitive paint 

 by a molecular torrent skirting along it; the brilliant colours 

 turquoise-blue, emerald, orange, ruby-red with which grey colourless 

 objects and clear colourless crystals glow on their struck faces when 

 lying separately or piled up in a heap in the course of a molecular 

 torrent ; " electrical evaporation " of negatively electrified liquids and 

 solids ;^[ the seemingly red hot glow, but with no heat conducted 

 inwards from the surface, of cool solid silver kept negatively electri- 

 fied in a vacuum of 1/1,000,000 of an atmosphere, and thereby caused 



* Tribulation, not undisturbed progress, gives life and soul, and leads to success 

 when success can be reached, in the struggle for natural knowledge. 



f Crookes, " On the Viscosity of Gases at High Exhaustions,' 655, ' Phil. 

 Trans.,' Feb., 1881, p. 403. 



* ' Phil. Trans.,' vol. 172 (1881), pp. 387, 435. 



Probably, I believe, not greater in any case than 2 or 3 kilometres per second. 

 || Address to the Institute of Telegraphic Engineers, 1891. 

 t ' Roy. Soc. Proc.,' June 11, 1891. 



