of dark Heat-rays by Gases and Vapours. 27 



certainly made up of many portions, possibly compensating 

 each other. On this account the results obtained in physical 

 experiments are not applicable to the explanation of any mo- 

 lecular condition of matter whatever, and in future the inves- 

 tigation of absorption-spectra must be extended to the domain 

 of heat-radiation. 



These considerations induce us to attach but little value to 

 the connexion between chemical composition and absorption 

 which we have pointed out in this investigation. 



Hence it follows that a comparison of the absorption-coeffi- 

 cients of different substances is only possible at equal pressures. 

 Since, as already said, in no one of the cases examined was 

 there any regular connexion between pressure and absorption, 

 we chose a graphical method, in which we represented the 

 pressures as abscissas and the absorptions as ordinates. The 

 absorption-coefficients of the substances examined, which be- 

 longed to the fatty series (containing nearly equal numbers of 

 molecules) were, for the radiation of a source of heat at 100°, 

 approximately in the following order: — 



I. Methyl alcohol, formic acid, carbonic oxide, carbonic acid, 

 chloroform. 



II. Ethyl alcohol, acetic acid, ethyl ether, ethylene. 



III. Butyl alcohol. 



IV. Amyl alcohol. 



Whilst the absorption of the substances placed in any one 

 line appeared to us nearly the same (with exception of the 

 probably not quite pure acetic acid), the absorption appears to 

 increase rapidly with increasing percentage of carbon ; and, 

 further, the absorption appears to be essentially dependent on 

 the number of carbon atoms directly combined with each other 

 by simple linking. Thus, for example, benzol, in spite of its 

 six carbon atoms, exerts only a very small absorption, pro- 

 bably in consequence of the mode in which the carbon atoms 

 are combined. 



Yet all these relationships may be purely accidental ; not 

 until we have a spectroscopic investigation of the ratios of 

 absorption shall we be in a position to draw conclusions on the 

 modes of vibration of the atoms. The extraordinary difficulty 

 of investigations of this sort would be richly repaid by the 

 attainment of quantitative results; whilst the corresponding 

 optical investigations (immeasurably easier) will always remain 

 more of a qualitative nature. 



Physical Institute of Vienna. 



