132 EEPORT — 1882. 



'We may say, however, it seems highly probable by this delicate 

 mode of analysis that the hypothetical position of any hydrogen which is 

 replaced may be identified, a point which is of prime importance in 

 organic chemistry. 



' The detection of the presence of chlorine or bromine or iodine in a 

 compound is at present undecided, and it may well be that we may have 

 to look for its effects in a different part of the spectrum. The only trace 

 we can find at present is in ethyl bromide, in which the radical band 

 about 900 is curtailed in one wing. The difference between amyl iodide 

 and amyl bromide is not sufficiently marked to be of any value.' 



If we compare the results obtained by Captain Abney and Colonel 

 Festing with those arrived at by Professor Russell and others in the 

 visible part of the spectrum, we are struck with the great persistency of 

 these infra-red bands and lines. Spectra of different compounds of the 

 same body may resemble each other in the visible part, but wherever the 

 absorption band or line was sufficiently narrow to be looked at under high 

 dispersion some difference made itself apparent, like the one discovered 

 by Bunsen in the case of didymium compounds. These differences are 

 generally sufficiently large to be noticed even with one prism when proper 

 care is taken. It would seem very remarkable if atoms could enter into 

 combination with their vibrations unchanged by the chemical force; though, 

 of course, the change maybe more apparent in some parts of the spectrum 

 than in others. Whether the infra-red bands are not affected at all, or 

 whether only the change is so small that it has not as yet been dis- 

 covered, is an open question. Oar prismatic methods would, of course, 

 discover more easily a shift in the visible part than in the infra-red, but 

 Captain Abney and Colonel Festing used three prisms, and there can be 

 no doubt that if the displacements had been of the same order of magni- 

 tude as they are with the didymium, and especially the cobalt salts, they 

 could not have escaped detection. 



Alexr. Mitscherlich ' was the first to prove that compound bodies, 

 when luminous, have a spectrum of their own, and do not simply show the 

 supposed spectra of the elements. He followed up this important dis- 

 covery by investigating the spectra of different compounds of the same 

 metal, and he could not fail to be struck with the similarity which such 

 spectra often present. 



Many, for instance, will instantly recognise the spectrum of the oxide 

 or chloride of calcium as that of a calcium compound, without being even 

 aware that these spectra present certain well-marked differences. Com- 

 paring together the spectra of the fluoride, chloride, bromide, and iodide 

 of barium, as they appear on Mitscherlich's map, we detect at once a 

 strong similarity ; we seem to have one spectrum shifted towards the red 

 with increasing atomic weight of the metalloid. At the same time the 

 least refracted bands seem to be most affected and, as a consequence, the 

 bands appear nearest together in the fluoride and farthest apart in the 

 iodide. In the calcium and strontium salts we notice the same increase 

 of wave-length in corresponding bands with increasing atomic weight, 

 bat with these two metals the most refracted bands are most affected, so 

 that the bands are the nearer together the higher the atomic weight. 

 Mitscherlich tried to find a numerical expression for these relations, and 

 he expresses the law which, according to his observation, represents the 

 facts, in the following way : — 



' Pogg, Ann. cxvi. p. 499. 



