1867.] Colouring-matters by the Spectrum Microscope. 439 



same in a different condition ; whereas, if two spectra agree, they may be 

 the same substance, or two distinct substances, whose different actions are 

 made equal by particular circumstances. As an illustration, we may refer 

 to a short string, which may give the same note as a longer whose tension 

 is greater. For this reason we should be careful not to rely too much on 

 one spectrum. If, however, we can produce some great physical change 

 in both substances, and still their spectra remain the same under equal 

 conditions, and if this occur uniformly in several different changes, we 

 may conclude that they are identical. Hence the value of the various 

 reagents named below. Many excellent illustrations of these principles 

 could easily be given. 



6. General Method of Experiments. 



Since the spectrum-microscope enables us to use very small quantities, 

 it appeared desirable to adopt such a method of research as would enable 

 us to take full advantage of this circumstance, and to avoid as much as 

 possible previous chemical manipulations. On this account I shall say 

 nothing about modified chemical methods, which may, of course, be also 

 employed when sufficient material is at command. My aim has been to 

 contrive a special system of qualitative analysis of coloured substances 

 applicable to minute quantities, and as independent of general chemistry as 

 the blowpipe method is in the case of minerals. I may here say that in some 

 very important practical applications to the detection of blood-stains 

 not above of a grain was at disposal, and yet perfectly satisfactory 

 results were obtained. 



I was led to study the colouring-matters of flowers, leaves, fruits, woods, 

 and roots, because it appeared a most admirable field of inquiry to teach 

 the general principles of the subject. The colours being so various, and 

 occurring under such complicated conditions, I thought that if methods 

 could be devised to distinguish those that are dissimilar and to prove the 

 identity of those that are alike, even when mixed with coloured impurities, 

 such principles could easily be applied to other inquiries. If the question 

 were merely to distinguish or compare absolutely pure colouring- m.atters, 

 there would be little or no difficulty ; but it appeared to me that one great 

 value of the method would be to be able to apply it at once to very im.pure 

 and mixed materials. In such cases mere colour is of very secondary im- 

 portance, since that may be totally changed by a very small amount of 

 impurity. 



7. Preparation of Colours. 



If the petals, leaves, &c. of plants be crushed in water, it very com- 

 monly happens that the colour is rapidly decomposed and no clear 

 solution can be obtained; but if crushed in a moderate quantity of 

 spirits of wine, and the solution squeezed out, filtered, and evaporated to 

 dryness at a gentle heat, the colouring-matter does not decompose, even 



