282 Mr. T. S. Thomson on the Diffusion of Gases. 



immersion was made very slowly, whilst it occasioned one 

 measured by 40° of sudden deviation when the immersion took 

 place rapidly. 



XLV. Addition to a former communication on the Diffusion 

 of Gases. By Thomas S. Thomson, Esq. 

 To Richard Taylor, Esq. 

 Dear Sir, 



IN following up the ideas contained in my paper on the 

 Diffusion of Gases, which you had the kindness to lay 

 before the public (pres. vol. p. 51), a subsequent consideration 

 has occurred to me which I now send you for publication. 

 The chemical constitution of the atmosphere is twenty-one 

 parts by measure of oxygen and seventy-nine parts by mea- 

 sure of nitrogen. Now supposing a tube twenty-one miles in 

 length and another seventy-nine miles, separated from each 

 other by a perfectly moveable diaphragm, the former filled 

 with oxygen and the latter with nitrogen gas, sound would 

 travel along this tube in a certain appreciable space of time 

 which can be calculated as follows. The data I extract from 

 Poisson's Mecaniqiie. 



a = velocity of sound 



g = 9"-80896 



h = O'^'IG 



■? s 10-462 



u = 0-00375 

 fl = 15°-9 (centigrade) 

 y= 1-3748. 

 The velocity of sound is expressed by the following formula : 



a=V^(^+^^)- 



A 



I mean now to assert that if, by the removal of the dia- 

 phragm, the gases be permitted to diffuse themselves through 

 each other, sound would still travel in precisely the same time 

 from one end of the tube to the other. 



It is a necessary consequence of D'Alembert's principle, 

 which 1 extract from Poisson, 



AmP + A'm'F + A" m" P" + &c. = 0. 



The direction which molecular chemistry is taking at the 

 present day renders it necessary to recall to the minds of some 

 of your readers, that the repulsion of gaseous particles does 

 not obey the law of the inverse of the square of the distance, 

 but the inverse of the distance itself. It forms the subject of 



