Effusion of Argon, Helium, and some other Gases. 431 



tap / having been closed, the effusion was started by opening 

 the tap m and a stop-watch set in motioD as soon as the 

 mercury meniscus sank to a certain scale-division ?i 2 a 

 little below %. A smooth and easy motion of the falling 

 mercury meniscus was secured by having a small quantity of 

 glycerine in the gauge, i. e. just sufficient to lubricate the 

 glass wall but not enough to collect over the mercury. The 

 watch was stopped as soon as the meniscus reached a third 

 fixed division, n 3 . Proceeding in this way with each of the 

 gases to be compared, it will be seen that the times of effusion 

 are measured between exactly the same limits of pressure. 

 The pressure in the receiver at the end of the effusion was 

 measured roughly by opening i and reading the closed 

 gauge S. In order to make two consecutive measurements, 

 it was only necessary to open i, pump out the receiver ■ p 9 

 close i, and readjust the pressure in the reservoir to n± by 

 admitting more gas. 



Experimental Results. 



Besides argon and helium the following gases were inves- 

 tigated : — Hydrogen, oxygen, carbon monoxide, nitrogen, 

 carbon dioxide, and cyanogen. The methods employed in 

 their preparation may be briefly indicated here. 



Oxygen. — By heating KMn0 4 . Dried by P 2 5 . 

 Hydrogen. — By heating palladium-hydrogen in vacuo. 



Dried by P 2 5 . 

 Nitrogen. — Prepared in vacuo from sodium hypobromite 

 solution and ammonium chloride. Passed through 

 powdered NaOH, and dried by P 2 5 . 

 .•„•: Carbon .monoxide. — Prepared directly over mercury, by 

 addition of concentrated formic acid to excess of c.oih 

 centra ted sulphuric acid. 

 Carbon dioxide. — By heating NaHC0 3 in vacuo. Dried 



by CaCl 2 and P 2 5 . 

 Cyanogen. — Prepared by heating Hg(CN) 2 in vacuo and 

 dried by P 2 5 . Purified by freezing out in bath of 

 liquid air and pumping off' any admixed nitrogen. 

 The first experiments were made by using very thin-walled 

 glass bulbs in which minute holes had been pierced. This 

 piercing w r as effected by employing the spark from a Wims- 

 hurst machine between terminals of platinum wire. When 

 an aperture of suitable dimensions had been obtained in this 

 way, the stem of the bulb was sealed on close to a capillary 

 tap, and the whole then sealed into the gas-reservoir. Ac- 

 cordingly, in these first measurements, the effusion vessel 



