78 SCIENCE PROGRESS 



However, this was only an interesting hypothesis, with very 

 little to support it. 



A short time ago, M. Einstein worked out the problem of the 

 motion of a particle due to collisions with the molecules, assum- 

 ing that these collisions were not co-ordinated at all. He 

 arrived at the conclusion that the mean path, which a particle 

 can traverse in a second, is given in centimetres by the 



relation 



/ 3 = RT/37rNr£, 



where r is the radius in cm. of the particle which is supposed 

 to be spherical, T is the absolute temperature, N the number 

 of molecules in one gram molecule (4 x io 23 approx.), k is the 

 coefficient of viscosity (roo x io -2 for water at 20°), and R is 

 the gas constant (8'4 x io 7 ). Substituting these values for the 

 letters, and considering particles of radius 1 /x or io -4 cm, we 

 find that during one second the particle can move over a mean 

 distance of about o'8 //,. It has been mentioned that the particles 

 of the hydrosols of gold and silver may have a radius of only 

 3 to 6 fifi ; for such particles the mean path would be about 

 8 to 10 /x. The amplitude of the oscillations for some particles 

 was seen to be about 10 /*, and thus the order of the values 

 deduced by Einstein is the same as that obtained by experi- 

 ment. Recently Chaudesaigues (November 1908), Henri (1908), 

 and Seddig (1908) have made experiments to test Einstein's 

 formula, and their results seem to support his work. 



Perrirts Work. — It is only within the last year, however, 

 that any direct experiments have been made with a view to 

 throwing some light on the nature of the Brownian movements. 

 In 1908 M. Jean Perrin experimented with colloidal solutions 

 of gamboge, which show the Brownian movements very 

 clearly. 



The purpose of his experiments was to show that the cause 

 of the Brownian movements lies in the molecular agitation 

 of the liquid — Gouy's hypothesis — and in that alone ; and what 

 is perhaps more important, the experiment furnishes a new 

 and more concise method of determining the number of 

 molecules in a gram molecule of a substance. 



If a stick of gamboge is placed in water a liquid is obtained 

 containing grains in suspension, the grains being visible under 

 the microscope with ordinary illumination. By employing a 

 centrifugal force, as one does in the separation of blood cor- 



