152 Royal Society : — Mr. H. C. Sorby on Analysis of Animal 



spectra, and have at length concluded that for general purposes the 

 following are the most convenient. Those which are solid v are best 

 kept in small bottles as coarse powder, and added to the small cells 

 in a solid state, so that the quantity used may be more readily known. 



9. Reagents. 

 Hydrochloric acid. 

 Citric acid. 



Benzoic acid. 



Boracic acid. 



Bicarbonate of ammonia. 



Carbonate of soda. 



Diluted solution of ammonia. 



Caustic potash. 



Sulphite of soda. 



Sulphate of protoxide of iron. 



Alum. 



Iodine dissolved in alcohol. 



Bromine dissolved in water. 



Solution of hypochlorite of soda. 



Permanganate of potash. 

 This list might of course be very much extended, if we were to 

 include such reagents as may be used in separating or decomposing 

 colours by the ordinary chemical methods. In describing the effect 

 of those named in this list, I feel that I could not avoid mentioning 

 some well-known facts without breaking the thread of my argument ; 

 and therefore I trust it may not be thought out of place if I give a 

 general account of the whole from the particular point of view required 

 by the subject more especially before us. 



The action of many reagents is so intimately related to different 

 parts of the spectrum, as to show that there must be some connexion 

 between. so-called chemical reactions and optical phenomena. Not 

 that their effect is absolutely the same in the case of all coloured 

 substances, but generally only the extent differs, whilst the character 

 of the change is uniform, — unless, indeed, decomposition take place ; 

 and even then it has a tendency to conform to a general law. 



10. Solvents. 



Water and alcohol are the most useful solvents, and the spectra 

 of the two solutions of the same substance often differ most stri- 

 kingly ; in fact they often behave in other respects as if they were 

 solutions of different substances. Sometimes the spectra are abso- 

 lutely identical ; but often well-marked narrow absorption-bands are 

 seen in the alcoholic solution, where they are almost, or quite, in- 

 visible in the aqueous. Very commonly the same bands are seen in 

 both, but not exactly in the same place, alcohol sometimes raising 

 them to a higher part of the spectrum, and sometimes depressing 

 them. Occasionally the spectrum of the dry material is like that of 

 the alcoholic solution, and unlike that of the aqueous, as if the 

 difference were due to the presence of water ; but in other cases it 



