42 



INFRA-RED ABSORPTION SPECTRA. 



^^h 



I 



3 



3 



once was absorption found for residual gas; this was ethyl ether at 

 8.7 ii, the cell having been washed but three times with air. 



To fill the cell, previously exhausted to 0.02 mm., with gas from a 

 gasometer, the capillary tube, c, was inserted in the rubber connection 

 (A fig"- 9) ^^ the latter, which contained a drop of mercury. On inser- 

 tion, this drop of mercury would displace the air, except the trace con- 

 tained in the bore of the capillary tube. The capillary bore was 0.024 

 cm. in diameter, 3 cm. in length, and had a volume of 0.013 cc, which, 

 compared to the total volume of the cell, would introduce one part of 

 air in 2,500 parts of pure gas. This trace of air could also have been 

 prevented from entering the cell by filling the capillary with mercury. 



Since the presence of this small quantity 

 of air diminishes the total length of gas 

 by only 0.003 cm., the effect on the total 

 transmission can not be detected, as was 

 found on trial, and in practice the capil- 

 lary was not filled with mercury. 



On opening the pinch-cock at f (fig. 9) 

 and the stop-cock of the cell, the gas en- 

 tered the latter and was brought to atmos- 

 pheric pressure, which, with the tempera- 

 ture of the gas (room temperature), fur- 

 nished data for computing the equivalent 

 thickness for the liquid state. 



A helix of iron wire inserted in the 

 glass cell served as a diaphragm to prevent 

 reflection from the walls, and seemed more 

 serviceable than a black paint, since it also 

 prevented occlusion of gases. 



The cell was mounted in the usual man- 

 ner, in a wooden carrier (fig. 11), which 

 worked in vertical ways between the source 

 of energy and the spectrometer slit. The 

 transmission through the exhausted cell was found each time before 

 filling with a gas. As mentioned before, this also served as a test for 

 residual gas, which could have been detected, if present, by setting on 

 one of the absorption bands. 



The ratio of the transmission of the cell filled with gas to that of the 

 empty cell gave the transmission of the gas itself. The transmission of 

 the empty cell changed but little in two months, and, as a typical exam- 

 ple, transmitted 65 per cent at 1.7 /a, 75 per cent at 4^1, decreasing to 

 70 per cent at 6 \i, then gradually increasing to 74 per cent at 10 )U. and 



Fig. 1 1 . — Top view of gas-cell holder with 

 cell c in place. «, top view, 6, side view. 



