360 Mr. Yngve Bjornstahl on Magnetic 



prevailing view as to the shape and structure o£ the particles. 

 As is generally known, one cannot get any idea of the shape 

 of the particles in a colloidal solution with the aid of the 

 microscope, on account of the resolving limit. The ultra- 

 microscope only exhibits the existence of a discontinuity of 

 the refractive index. Thus it was necessary to resort to 

 macroscopic phenomena : — there are two subjects the study 

 of which has decidedly supported our opinion on these 

 questions : investigations as to light absorption and the 

 sedimentation of the colloids. 



Several authors have dealt with the theory of light ab- 

 sorption by colloidal metal solutions. Maxwell -Garnett * 

 calculated the absorption spectra of the gold hydrosols and 

 found in several cases close agreement. The same problem 

 has been more fully treated by Mie f , who also supposes the 

 particles to be massive and of a spherical shape. Steubing J 

 undertook to verify the calculations made by Mie and 

 measured the light absorption of several gold solutions. He 

 states that the maximum of the absorption curve moves in 

 the direction of greater wave-lengths with increasing size 

 of the particles, in accordance with Mie's theory. In a 

 theoretical paper Gans § discussed in detail the light absorp- 

 tion of particles, whose form could be represented by ellipsoids 

 of revolution. He found that even at moderate eccentricities 

 the effect differs considerably from that calculated for a 

 sphere. Gans is of the opinion that the particles in Steubing's 

 colloidal solutions did not deviate very much from the 

 spherical shape. The first systematic quantitative investi- 

 gations as to the siz^ of the particles were made by The 

 Svedberg ||, and later by Pihlblad % The latter obtained a 

 rather good agreement between Mie's theory and the 

 measurements of the light absorption. From this one may 

 conclude that the particles have a spherical shape. 



The rate of sedimentation of colloidal gold solutions has 

 been investigated by A. Westgren**. He obtained a very 

 good agreement between the size obtained by counting the 

 number of particles in a fixed volume and that which is 

 calculated from the rate of fall in accordance with Stokes's 



* Phil. Trans. A. 203. p. 385 (1904) ; 205. p. 237 (190G). 



t Ann. der Phys. (4) xxv. p. 377 (1908). 



t Loc. cit. xxvi. p. 329 (1908). 



§ Ann. der Phys. (4) xxxvii. p. 881 (1912). 



f| Koll. Zeit. vi. p. L38 (1910). 

 51 Inaug. Dissertation. Upsala, 1918. 

 ** Dissertation, Tab. 132, Upsala, 1915. 



