54 INFRA-RED ABSORPTION SPECTRA. 



It thus appears that there are three subordinate series, but how are 

 we to estabUsh the vaHdity of these observations ? More pairs of bands 

 are necessary, but, as we approach the visible spectrum, the pairs of 

 maxima must lie closer together, and, hence, can only be resolved with 

 a larger dispersion. In addition to this, the regions at 2 /u, and 4 fi are 

 transparent. This, however, can probably be remedied by using longer 

 columns of the gas. In fig. 25 curve b is for a pressure of 24 cm. of 

 mercury, while the curve thus, x x x x , is for the purified NH4CI. 



Absorption Spectra of Liquids and Souds. 



(Tables II, VI, and VIII a to f.) 

 HAIvOGEN SUBSTITUTION PRODUCTS. 



iChIvOroform. CHCI3. (Fig. 26.) 



This is one of the first compounds investigated, and is of interest 

 because it is one of the simplest combinations of carbon and hydrogen. 



Numerous small absorption bands are to be noticed throughout the 

 whole curve, and one is reminded of the line spectrum found by Abney 

 and Festing at 1.2/*. The wide band from 6.8 /a to 7.2 /a is complex. 

 The 5.9 ju. band is very small here, but is to be noticed, since it occurs 

 in so many other compounds having CH-groups. Since chloroform is 

 one of the simplest compounds having a CH-group, one would expect 

 to find absorption bands due to this group, provided it has the power 

 to act alone. Turning to benzene (CgHe), which has a ring of 

 CH-groups, we find that none of the conspicuous absorption bands are 

 in common. The largest band of CHCI3 at 8.3 fi is shifted to 8.7 fi in 

 benzene, while the 10.82 yu, band has disappeared entirely in the latter. 

 If this disappearance of the 10.82 ju, band is to be attributed to the lack 

 of CI in benzene, one would look for it in monochlorbenzene (CgHgCl). 

 It occurs there (fig. 85) as a slight depression in the transmission curve. 

 It also occurs in S and in benzaldehyde (CoHgCHO), which brings us 

 back to the previous conclusion of the nonentity of the atom in the 

 molecule. 



Iodoform. CHIs. (Fig. 27.) 



This is a crystalline solid, and hence difficult to obtain in a thin film. 

 The film used was 0.05 mm, thickness, but, not being homogeneous, it 

 increased the opacity by scattering the incident energy. As a result, 

 the general transmission is reduced to about 40 per cent, and only two 

 strong bands are resolved, at 8.6 [x and 9.3 fi, respectively. In chloro- 

 form these bands are quite obliterated by others of greater depth. The 

 9.4 fi band is to be found in numerous other compounds. On melting, 

 iodine was set free, thus coloring the film red, but this evaporated in a 



