262 ON LIGHT. 



occupy its whole area and suffer no ray to enter it which 

 does not come from some part of the coloured surface ; 

 the spectrum will be seen deficient in all those rays 

 which the object does not reflect, and which belong to 

 its complementary colour. The use of this little instru- 

 ment, at once simple, portable, and inexpensive, will be 

 found to afford an inexhaustible source of amusement 

 and interest. To the florist, on a bright sunny morning, 

 the analysis of the tints of flowers and leaves, or the 

 hues of a butterfly's wing, and of every variety of coloured 

 object; to the water-colour painter, the study of the 

 prismatic composition of his (so fancied) simple washes 

 of colour and the effects of their mixture and super- 

 position ; to the oil painter, that of the various bril- 

 liantly coloured powders which mixed with oil form the 

 material of his artistic creations, are all replete with 

 interest and instruction. 



(46.) If instead of a reflected colour we would exa- 

 mine a transmitted one, as in the case of a coloured 

 glass, or some natural transparent coloured product, if 

 in the form of a plate or lamina, it may be laid over the 

 slity^and when directed to any bright white light (as that 

 of a white cloud), its spectrum will be exhibited if a 

 coloured flame, the slit may be placed close to it, but if 

 a liquid, it will be preferable to make it its own prism 

 by enclosing it in a hollow prism formed of plates of 

 glass cemented together, when the differences arising 

 from difference of the thickness of the medium traversed 

 by the refracted rays will be more easily studied. 



(47.) The colours of transparent media such as 



