284 KEPORT— 1880. 



light which is slightly more refrangible than that which would be sent 

 out by a quiescent molecule or one moving at right angles to the line of 

 sight. On the other hand the molecules which are moving away from us 

 will have the wave-length increased. The lines as they appear to us, and 

 as they come from molecules moving in all directions, must have a certain 

 width. Lippich has pointed out how this limit of sharpness which cannot 

 be surpassed may be made use of to determine which lines in a mixture of 

 gases are due to each component ; for the heavier gas will have its lines 

 narrower than the lighter gas. As a rule, however, the lines of a spec- 

 trum are wider than the limit given, and especially the widening of lines 

 which we have beea discussing in this chapter is of a much higher order 

 of magnitude. 



III. Spectra of Different Orders. 



Spectra may be classified according to their general appearance. The 

 different classes have been called orders by Pliicker and Hittorf. We 

 have first, appearing at the highest temperature, the line spectra which 

 are best known and need no further description ; we have next produced, 

 generally at lower temperatures, the spectra of channelled spaces, the ap- 

 pearance of which was described in the introduction to the report on the 

 spectra of metalloids. Continuous spectra, which need not necessarily 

 stretch through the whole range of the spectrum, form a third order. 

 Pliicker and Hittorf have shown that one and the same element may pos- 

 sess at difierent temperatures specti-a of different orders. Their results 

 have been confirmed in the case of a great many different elements. A 

 discussion has naturally arisen on the cause which can produce such a 

 remarkable change of spectra. We first quote Pliicker and Hittorf's' 

 opinion on the subject : — 



' Certainly in the present state of science we have not the least indica- 

 tion of the connection of the molecular constitution of the gas with the 

 kind of light emitted by it ; but we may assert with confidence, that if 

 one spectrum of a given gas be replaced by quite a different one, there 

 must be an analogous change of the constitution of the ether, indicating 

 a new arrangement of the gaseous molecules. Consequently we must 

 admit either a chemical decomposition or an allotropic state of the gas. 

 Conclusions derived from the whole series of our researches lead us 

 finally to reject the first alternative and to adopt the other.' 



The idea that different spectra of one and the same element are due 

 to differences of molecular structure has fou^nd considerable favour with 

 spectroscopists. It has formed the leading idea of a large part of Lock- 

 yer's ^ work, who gave the following five stages through which spectra 

 often pass, each stage being in his opinion due to a different molecular 

 structure : — 



1. Line-spectrum. 



2. Channelled-space spectrum. 



3. Continuous absorption in the blue. 



4. Continuous absorption in the red. 



5. Continuous absorption throughout. 



Salet^ also adheres to the opinion that different spectra are due to 

 different allotropic states. Helmholtz, as quoted by Moser,"* has suggested 

 that the line-spectra may be due to atoms, the band-spectra to molecules. 



' PMl. Trans, civ. p. 1 (1865). ^ Ann. CJiim. Phja. xxviii. p. I (1873). 



* Proc. Roy. Soc. xxii. p. 372 (1874). * Pogg. Ann. clx. p. 177 (1877). 



