INVESTIGATION WITH A ROCK-SALT PRISM. 



41 



in a state sufficiently pure for this work. Not being satisfied with the 

 usual washing in bromine water, and liquefaction method, the gas- 

 ometer was placed in series with a wash-pipette of fuming sulphuric 

 acid, and the gas washed back and forth half an hour, after which only 

 a suspicious trace of ethylene remained, as will be noticed at 10.5 ^ in 

 the absorption curves. 



The Absorption Cell. 



The absorption cell for gases is shown in fig. 10, Two glass cells 

 were used, in length 6.3 and 5.7 cm., diameters 2.4 and 2.2 cm., capacity 

 30.8 cc. and 21.5 cc, respectively. 



The rock-salt windows were split from the natural crystal, which 

 gave smooth, plane surfaces which were not attacked by moisture, 



and lasted throughout the work of two months. The windows were 

 attached with Le Page's glue, which became exceedingly hard on 

 drying, and had such a low vapor pressure that no absorption bands 

 could be detected, even after exhausting the cell to 0.02 mm. and per- 

 mitting it to stand four days. One of the vapors to be expected from 

 it would have been acetic acid. Before filling the cell with a new gas 

 It was always washed out thoroughly with air. This was done by means 

 of a water aspirator attached to the pump, the cell being exhausted from 

 five to seven times, each time allowing it to fill with air. The final 

 exhaustion was carried to 0.02 mm., and the absorption of the cell, thus 

 exhausted, was found each time before filling it with a new gas. Only 



