﻿±82 Prof. C. A. Skinner on the Evolution of Gas 



to helium, therefore, argon suggested itself as a gas which 

 might fulfil this condition. This having proved true, the 

 experiments described below were performed with the above- 

 mentioned intent. 



Experimental Arrangement. 



The discharge-tube used was one which has served in the 

 other investigations (I. <?.). It was devised for making com- 

 parative tests of several different electrodes without in the 

 meantime opening to the atmosphere. These electrodes could 

 be removed for polishing or be replaced by new ones. 



Helium and argon were obtained in sealed bulbs from 

 Messrs. Thomas Tyrer & Co., London. Each gas was trans- 

 ferred from the bulb containing it to a storage-chamber 

 connected by a stopcock to the system containing the dis- 

 charge-tube, manometer, drying-chamber, and pump con- 

 nexion. 



The quantity of gas evolved being determined from the 

 increase in pressure, it was necessary to measure the pressure 

 very carefully. This was accomplished by means of a 

 MacLeod gauge, which multiplied it one hundred times, and 

 with which single observations were seldom more than one 

 half per cenL off the mean of several. 



The volume of the chamber occupied by the gas was 

 obtained by attaching to the system a chamber of known 

 volume, measuring the pressure when a given quantity of 

 gas occupied the unknown volume, then the known in addition, 

 and from these data calculating the desired value. This was 

 found to be 455 c.c. 



The electric current through the discharge-tube was 

 furnished by a battery of small accumulators, measured by a 

 Weston milliammeter, and regulated by a resistance of cadmium 

 iodide in amylic alcohol. 



Customary precautions were taken to avoid the presence 

 of moisture and hydrocarbons. 



The electrodes were given in all experiments a mirror 

 polish, thereafter carefully cleaned with a dry cloth, then 

 mounted in the discharge-tube, and this allowed to stand 

 evacuated in connexion with the drying-chamber, at least 

 twenty-four hours before making a test. To avoid the pos- 

 sibility of time changes in the metal, the plan was followed 

 of testing one sample of it in one kind of gas, then following 

 this with a test of another sample of the same metal (which 

 had been prepared in a similar manner) in the second gas. 

 This method proved to be wholly satisfactory. 



Earlier experiments (I. c.) having shown that both 



