120 Chronicles of Science. [Jan., 
flame simultaneously, the green lines of the bromide predominate 
at the first imstant; the first of the double indigo line of the 
chloride is visible ; the superposition of the more refrangible rays 
of the two spectra give rise to new appearances. ‘The presence of 
the iodide of copper produces no change of importance ; and thus 
it is easy to recognize at least a chloride and a bromide in a mixture 
of the three salts. 
Iodide of bismuth gives the clearest indication of iodine; the 
spectrum of the salt up to 130 is often ill-defined and difficult to 
distinguish from the same part of the bromide and chloride spec- 
trum; but the beautiful violet band which terminates that of the 
iodide is a convincing proof of the presence of iodine. 
Fluoride of calcium must always be used to show the spectrum 
of fluorme. ‘The green line situated about 121 is very brillant 
when a very high temperature is employed, and may be considered 
characteristic of this metalloid. 
In some cases the author recommends precipitation with nitrate 
of silver, but instead of using the silver precipitated in the manner 
directed by Mitscherlich (see ante), he treats it with sulphuretted 
hydrogen, saturates one part of the acid liquor with oxide of copper, 
and the other with freshly precipitated oxide of bismuth. The 
liquid or the dry residue of evaporation may be tested directly in 
the flame, the one for chlorine and bromine, the other for iodine, as 
indicated above. 
Lastly, M. Diacon repeats that the spectra given by Kirchhoff 
and Bunsen for the alkaline-earthy metals are a mixture of 
the spectra of the oxide and the chloride, as are also the spectra 
given by Mitscherlich as those of the chlorides. In the former, he 
says, the spectrum of the oxide predominates; in the second, that 
of the chlorides. A metal, he states, may give different systems 
of lines, according to the experimental conditions or the nature of 
the compound experimented upon, and no absolutely specific charac- 
ter can be attached to the spectra given by Kirchhoff and Bunsen ; 
they can only be considered characteristic of the conditions under 
which they were observed. 
The nature of the invisible photographic image has been a 
noever-ceasing subject for discussion amongst scientific photo- 
graphers. Lately, Mr. Carey Lea,* an experimentalist who stands 
in the first rank as chemist-photographer, has tried some experi- 
ments, which to his mind seem to finally settle the long-contested 
question as to the nature of the invisible photographic image. 
The view that the change which takes place in iodo-bromized plates 
in the camera is a purely physical one, that no chemical decomposi- 
tion takes place, and neither liberation of iodine nor reduction of 
silver, has obtained a pretty general acceptance. But latterly it 
* « American Journal of Science and Art.’ vol. xi., No. 118. 
