INVESTIGATION WITH A ROCK-SALT PRISM. 97 



Several other bands are in common with benzene and pyridine. There 

 are an unusual number of bands in common with toluene, from which 

 it differs in having an additional N atom. 



When examined with the large spectrometer the small bands, e. g., 

 2.92 fi, are brought out very prominently. The 6.86 fi band is to be 

 noticed, since it is found in many CH3 compounds. 





PiPERiDiNE. CsHiiN. (Fig. 121.) 



12 Piperidine is of special interest on account of its ring of 



jj C'^^N:H2 CHj-groups and the NH-group of atoms. The petroleum 

 I I distillates are rings or chains of CH^-groups, hence we 



^ ^it" ^ would expect to find some relations among the absorption 

 " bands. But no such relations exist, except at 13.7;^, show- 



ing again the effect of bonding, as well as of the presence of certain 

 groups of atoms like NH. The piperidine spectrum is unlike the pyri- 

 dine, except at 6.95 /x and 9.55 fi and the band at 3.0 /x, which is shifted 

 a little toward the long wave-lengths. The 3.5 fi band is unusual, being 

 found in paraldehyde. Since the NH-group occurs in pyrrol one would 

 expect to find similarities if a specific group of atoms cause absorption. 

 The only band which seems in common is the one at 2.95 fi, and this one 

 does not coincide with that of piperidine at 3.0 /i/. 



These three compounds were fresh and colorless, showing that they 

 had not decomposed, so they were not redistilled before using. Since 

 there was no difference in the absorption bands for the museum and 

 Kahlbaum's preparations, it seems evident that there could not have 

 been a great difference in their purity. 



This compound is of the greatest importance in this work, for it is 

 a ring of CH^-groups, while benzene is a ring of CH-groups. 



Moreover, the C atoms in the piperidine ring are united by a single 

 bond, while in the benzene ring the bonding is alternately double and 

 single. We might, then, conclude from this fact that the difference in 

 their spectra is due solely to the manner of bonding of the atoms. But 

 this is not admissible, because in pyrrol, which is a ring of four CH- 

 groups and an NH-group, one would then expect to find bands in com- 

 mon with benzene. No such bands exist. It has already been men- 

 tioned that in the piperidine ring of CHj-groups and in the methylene 

 ring of CHj-groups of the petroleum distillates no bands are in common. 

 This method of reasoning excludes the idea that simply the bonding 

 of the atoms causes these characteristic bands. Since we have not yet 

 established the fact that a definite group of chemical atoms causes cer- 

 tain absorption bands, and since certain groups of compounds have 



