﻿Volume, and Temperature of Rarefied Gases. 311 



the siphon which could be sealed by the mercury in the 

 reservoir P. The mercury-level was adjusted by exhausting 

 or filling with air the reservoir P in the same manner as 

 with the reservoir C of the mercury-pump. The amount of 

 gas admitted could, if desired, be measured in the gas 

 volume-meter Q, which was designed on the model of the 

 McLeod gauge. By means of this apparatus as small a 

 quantity of gas as ^-^- Q of a cubic centimetre could be accu- 

 rately measured and admitted. 



The two gauges E and F were placed close together on the 

 stand in order to bring them as much as possible under the 

 same conditions of temperature and exhaustion, and were 

 almost exactly of the same size. They consisted of a large 

 volume-chamber (A) about 90 cub. centim. in capacity, with 

 a pressure-tube (B) at the side and volume-tubes (C) and (D) 

 at the top. The tube (B) was about \ inch internal diameter; 

 (C) consisted of a somewhat large capillary tubing, and (D) 

 of a tube of very small bore. (C) and (D) were each about 

 250 millim. long. The volume and pressure tubes were 

 accurately graduated in millimetres, the divisions on each 

 exactly corresponding. The zero-points of the scales were 

 placed at the top of (D) and at the corresponding place on 

 (B). (B) was graduated 500 millim. below and 250 millim. 

 above the zero-point. The total length of the volume-tube 

 was of course 500 millim. A trap (E) was sealed below the 

 chamber (A) in order to more accurately determine the 

 volume of the gauge. 



The calibration of the gauge was a most important point 

 and was most carefully done. Every centimetre division on 

 the volume-tube was taken as a reading-point, making fifty 

 possible. At each of these points the capillary depression of 

 the mercury had to be determined, L e., the difference in 

 reading between the mercury levels in the pressure and 

 volume tubes when both levels were under the same pressure. 

 In order to measure these differences the extreme top of (D) 

 was drawn out before a blowpipe to a fine point and the tip 

 cut off. The gauge was then placed in an exactly similar 

 position to that which it would occupy when finally fixed on 

 the stand. Mercury was run in from the bottom and readings 

 taken in the pressure-tube, with the mercury standing at each 

 of the 50 reading-points on the volume-tube. After this had 

 been done both with rising and falling mercury, the actual 

 volumes corresponding to each of the 50 reading-points on 

 the volume-tube measured from its upper end were deter- 

 mined. The tip of (D) was sealed, and accurately weighed 



