GENERAL DISCUSSION OF THE SPECTRA. IO9 



only a slight depression in the absorption curve at 2.95 fi. The fatty 

 acids, e. g., caproic, stearic, etc., are of interest because they have no 

 band at 2.95 fi. In electrolysis the alcohols are separated into ethyl and 

 OH ions, while in the fatty acids, instead of the OH ion, we have simply 

 an H ion. Hence, reasoning from this analog>', one would not expect 

 a band at 2.95 /a for the fatty acids. In the other compounds having 

 an OH-group, e. g., eugenol, caymol, menthol, and phenol, strong bands 

 are to be found shifting from 2.87 fi to 3.0 fx. They show no bands at 

 6.0 fi. Can we assume, then, that the bands at 2.9 /i to 3 /it are due to 

 OH? The evidence is not very favorable. Considering the bands of 

 ammonia and of the compounds containing NH2, or certain ones con- 

 taining nitrogen, the coincidence appears to be somewhat accidental. 

 Farther in the infra-red we have numerous cases of the coincidence of 

 absorption bands. 



As a whole, the most definite conclusion we can draw at present is 

 that the alcohols have a characteristic band at about 2.p^ fi, just as the 

 band at 4.78 ii is characteristic of the mustard oils. 



The CH3-group of atoms is probably the most important to be con- 

 sidered, but only a few cases can be noticed here. The most noticeable 

 eflFect is in benzene derivatives. It was shown under the discussion of 

 the eflFect of structure that the benzene group (CgHe), although it 

 appears as a series (CnHgn-e)* is entirely diflferent from the chain com- 

 pounds, like CnH,n-2. But a substitution of several CHa-groups com- 

 pletely absorbs the 3.25 /n benzene band, and the 3.43 /a band, charac- 

 teristic of all compounds containing CH3, takes its place. Whether 

 the 3.25 fi band has actually disappeared is an open question. In mesi- 

 tylene there is still a trace of the 6.75 /a band of benzene, showing that 

 the benzene " ion " has not been destroyed by the substitution of three 

 CHg-groups. In the xylenes the 6.75 /* band is least aflfected, while the 

 3.25 IX suflfers the most, and the whole strengthens the belief mentioned 

 in the beginning, that certain vibrating ions always seem to be present, 

 but that their eflFect in absorbing heat waves seems to depend upon their 

 surroundings. Thus the eflFect of substituting an NH.-group for an 

 H atom, forming aniline, has the least eflFect on the benzene, 3.25 ju,, 

 band, while those from 6 to 7 /a have disappeared entirely. In benzal- 

 dehyde (CgHsCHO) the 3.25 ju. band is not seriously inflenced by a more 

 intense absorption band at 3.55 {i, while in benzonitrile (CeHgCN) and 

 in monobrombenzene (CgHgBr) the 6.25 fx band suflFers no change. 



As a whole, the substitution of a CH3 or NH2 group has a great 

 eflFect on the resulting absorption spectrum. In the benzene derivatives 

 these groups form new bands, which occur beside the benzene bands. 



