114 



INFRA-RED ABSORPTION SPECTRA. 



groups. In discussing the sources of errors it was shown that the 

 6.86 /t band is quite accurately known. 



Beyond 13.6 /x there is a band of frequent recurrence. In this region, 

 however, it is difficult to locate the bands with great accuracy, because 

 of the weakness of the radiation. As a whole, however, the bands at 

 1.71, 3.43, 6.86, and 13.6 to 13.8^ are closely harmonic, e. g., in the 

 petroleum distillates, and, taken with the 0.867 ii band of Abney and 

 Festing (0.83 to 0.86 |it in present work), would seem to indicate a 

 vibration about a fixed point. Even if in the future this relation should 

 ultimately be found false, the constant recurrence of these bands in so 

 many compounds can not be without meaning. These bands are so 

 sharp and symmetrical that it is difficult to conceive how, with greater 

 dispersion, they can be resolved into lines which are very unsymmetric- 

 ally placed about the present centers of gravity. 



In carbon tetrachloride and tetrachlor-ethylene it was shown that this 

 group of compounds is conspicuous for its absence of absorption bands 

 except at 6.5 and 13 ix (harmonics), where there are large bands, each 

 of which is evidently complex. The complete table of wave-lengths of 

 absorption bands contains still further illustrations. An explanation 

 of the significance of these relations is not attempted, and it will be suf- 

 ficient to add that any such harmonic relation would seem to indicate 

 the resonance of a definite group of atoms, or " ions," to which these 

 lines are solely due. But to attribute a given line to a certain group 

 of chemical atoms is dangerous, for it has already been shown in the 

 case of the mustard oils that the manner of grouping of the atoms is not 

 the only characteristic of this group of compounds. Thus the physical 

 properties of benzene and thiophene are so similar that these two com- 

 pounds are readily confounded,^ yet their absorption spectra are entirely 

 different. It might be added that in the gases, where one would natur- 

 ally expect such harmonic relations, only acetylene at 3.7 and 7.4 /x, and 

 methyl ether at 3.45 and 6.9 /x have bands satisfying this condition. 

 Whether this relation will ultimately be proved absolutely true remains 

 to be seen. To determine this question a very much larger dispersion 

 will have to be employed than has yet been available. This means a 

 far more sensitive recording apparatus than has yet been devised. 



In dismissing this question it will be sufficient to add that after a 

 year's struggle with it, to prove or disprove it, the result has been a 

 closer agreement in the values first obtained, especially for the bands 

 at 3.43 /x and 6.86 fx. 



^Smith's (Richter) Organic Chem., vol. 2, p. 47. 



