DEVELOPMENTS IN ELECTROMAGNETISM BLOCH. 235 



ionizing force, surrounds itself with a cortege of neutral molecules; 

 the residual positive atomic ion does likewise. Thus originate the 

 ordinary positive and negative ions. They are »characterized by 

 their mobility A"", coefficient of recombination a, and diffusion Z>. At 

 very low pressures and at high temperatures these assemblages are 

 dissociated little by little to the primitive charged center., We will 

 see that some modification of these ideas will be necessary. 



(1) Along the line of theory since the fundamental work of 

 Langevin (Annales Ch. Phys., 1905) several new attempts have been 

 made to explain the order of magnitude of the mobilities and their 

 variations. Among these we should specially mention those of 

 Sutherland/ of Wellischj^ and of Reinganum.^ Sutherland, es- 

 pecially, departing from the hypothesis of molecular agglomera- 

 tion, supposes that an ion is identical with the electron or the 

 primitive atom-ion; its velocity is modified and retarded by the 

 electric action exercised upon the neighboring ions or the molecules 

 polarized by its approach. An apparent viscosity is thus created 

 which explains very well the results of Phillips (see further on) upon 

 the variation of the mobility with the temperature. The actual 

 theory is not unlike that which led Sutherland to his well-known 

 formula for the variation with the temperature of the viscosity of a 

 gas. 



It will be perhaps convenient to use the conventions of the older 

 theory, considering the ions as assemblages in perpetual process of 

 formation and disintegration in a kind of dj'namical equilibrium; 

 the charged center will then be in turn free and loaded with neutral 

 molecules. We will see that a greater part of the experimental data 

 makes such a convention almost necessary. 



(2) With a view to furnishing useful material for the theoretical 

 developments, many measures have been made upon the mobility, 

 the rate of recombination, and the dift'usion at various temperatures 

 and pressures. We will mention the measures of Phillips* (varia- 

 tion of h and a with the temperature), Kovorik,^ Tood,*' Dempster^ 

 (variation at K at high and low pressures), Sales ^ (variation of D 

 with the pressure). These measures show that ionic agglomerations 

 disintegrate faster at low pressures and high temperatures in the case 

 of negative ions and tend for both positive and negative ions to 

 revert to the primitive state. This is in accord with the measures 



1 Sutherland, Phil. Mag., vol. 18, p. 341, 1909. 



2 Wellisch, Phil. Trans., vol. 209, p. 249, 1909. 

 sReinganum, Phys. Zeitschr., vol. 12, pp. 575 and 666, 1911. 

 < Phillips, Proc. Roy. Soc, 1906, and vol. S3, p. 246, 1910. 



sKovorik, Phys. Rev., vol. 30, p. 415, 1910; Proc, vol. 86, p. 154, 1912. 

 "Tood, Radium, p. 113, 1911; p. 465, 1911. 

 •^ Dempster, Phys. Rev., vol. 34, p. 53, 1912. 

 s Sales, Radium, p. 59, 1911. 



