40 



INFRA-RED ABSORPTION SPECTRA. 



part that did not liquefy in e passed out at /. Petroleum thermometers 

 were used to measure the temperature, which, of course, could be deter- 

 mined only approximately, but 

 as close as one can make the 

 fractional liquefactions. The 

 thermometer was placed in con- 

 tact with the side of the bulb, 

 and the whole suspended over 

 the liquid air. By raising or 

 lowering this combination, any 

 desired temperature could be ob- 

 tained. 



From 5 to 10 cc. of the lique- 

 fied gas were collected in e, 

 which was then closed at h, by 

 means of a pinch-cock, and al- 

 lowed to grow warmer. After 

 the air had been thoroughly 

 washed out at /, as shown by the 

 lighted gas, the bulb was per- 

 mitted to become still warmer, 

 and the successive fractions 

 were distilled into the mercury 

 gasometers, g, where they were 

 stored under mercury seals, 

 awaiting analysis and examina- 

 tion. This method of procedure, 

 for the present work, was cer- 

 tainly more desirable and effi- 

 cient than the usual one of join- 

 ing the generator, purifiers, and 

 absorption cell with an exhaust 

 pump. In that case one can not 

 manipulate the gases after once 

 generated. The desirability of 

 handling the stored gases is 

 shown in the purification of 

 ethylene, 

 which is al- 

 most impossi- 

 ble to obtain 



Fig. 9. 



