62 



Drs. J. P. Kuenen and W. W. Randall. [Nov. 28, 



Since the mercury was always brought to the same point on the 

 stem of the thermometer bulb, the volume of the gas, except for the 

 change produced by the expansion of the glass, was in all the experi- 

 ments the same, while the pressure was, of course, different for each 

 temperature employed. For convenience, the bulb was filled, in the 

 case of each gas, at from two-thirds to three-fourths atmospheric 

 pressure : under these circumstances, the bulb was never subjected to 

 an internal pressure greater than about atmos. 



Corrections. — All pressures were reduced to 0°. The coefficient of 

 expansion of the glass of the bulb was carefully determined, and was 

 found to be 0-00002804 ; its effect was allowed for. The effect of 

 capillarity in depressing the mercury in the narrow tube was deter- 

 mined and taken into account. The volume of that part of the stem 

 of the bulb which was not in the ice or vapom, as the case might bo, 

 was found to be 0*0003 of the whole, and was allowed for in the cal- 

 culations. The change of volume in the bulb due to change of 

 pressure was found to be negligable. 



Gases. — The first gas experimented with was hydrogen. This was 

 prepared from pure zinc, was washed with potassium permanganate 

 solution and then with strong sulphuric acid, and was dried with 

 phosphoric anhydride before it entered the previously exhausted 

 bulb. The thermometer was successively filled and exhausted several 

 times, in order to remove impurities, and was heated while vacuous 

 to dislodge any gas clinging to the surface of the glass. Finally, the 

 purified hydrogen was allowed to enter slowly until the required 

 pressure was obtained. 



Two sets of experiments were made with air. In the first set no 

 effort was made to remove carbon dioxide, although the air was of 

 course carefully dried. The readings were made by one of us alone, and,, 

 on account of the numerous details to be attended to which actually 

 require the attention of two observers to be put beyond question, are 

 probably not as accurate as the other series. The second series had 

 to be brought to a close after the pressures corresponding to 0° and 

 the boiling points of water and qninoline had been determined. In 

 this series care was taken to use air free from carbon dioxide. 



The helium used was some of that prepared and purified by Pro- 

 fessor Ramsay. Its density was 2'13, that of oxygen being taken 

 as 16. 



The argon employed was prepared from atmospheric air by the 

 method of Professor Ramsay. A large gas-holder was filled with air 

 which had been slowly drawn through a long combustion tube filled 

 with red-hot copper. This gas was dried, passed again over the hot 

 copper, and then over red-hot magnesium shavings until absorption 

 of nitrogen ceased. By these processes a gas was obtained consisting 

 of about equal volumes of argon and nitrogen. Passage of this gas 



