MOLECULES, CORPUSCLES AND IONS 281 



the turbid liquid. When the ultra-microscope was brought 

 out in Jena, the study of this curiosity assumed a direct 

 scientific interest, and the impression gained ground that 

 the observer really watched molecular movements akin to 

 those which the particles of gases describe according to the 

 kinetic theory of gases. The idea originated, we believe, 

 with Einstein ; he certainly worked out the mathematics of 

 the problem. During the past few years J. Perrin, Gouy, 

 Svedberg, and others have supplied apparent experimental 

 proofs for the molecular character of the movements. 

 Perrin counted the number of gamboge granules or parti- 

 cles, in a portion of his colloidal solution, measured their 

 diameters, masses, and paths, and calculated their average 

 kinetic energy. He concluded that the granules had the 

 same average energy of movement as the molecules of the 

 liquid in which they were suspended, and that they behaved 

 thus like molecules of a very high molecular weight. 



Now molecules of a high molecular weight in other 

 words, molecules consisting of a great number of atoms of 

 different elements are nothing strange to the chemist. 

 Emil Fischer, in his famous researches on the albuminoids, 

 has come to very high molecular weights indeed. In his 

 four series of experiments Perrin dealt with granules 

 whose masses varied as 1 : 3 : 8 : 27. Allowing for the 

 coarseness of his granules and the friction in his medium 

 (water), Perrin deduced for the number N of molecules 

 per cubic centimeter very nearly the same figure, 3 X 10 19 , 

 to which other researches have led us. Estimates of this 

 N, we should add, have been made by the most varied and 

 entirely independent methods. Some of the methods give 

 results which, it may be foreseen, should be considered as 

 upper limits, others will yield lower limits. The average 

 accepted value for N was, a few years ago, probably 

 6 X 10 19 ; at present scientists incline to half that value, 

 3 X 10 19 . 



The experiments of Ehrenhaft confirm those of Perrin. 

 Ehrenhaft volatilized silver electrodes in air; the fine-dust 



