202 Original Articles. [April, 



the instrument may be applied. Of course, a number of tbese could 

 be studied by means of an ordinary spectroscope ; but then, not only 

 are they in many cases so intimately connected with microscopical 

 inquiries, that a microscope as such can be most usefully enrployed ; 

 but moreover, to say nothing about the advantage of making one instru- 

 ment serve instead of two, with the very simple additions I have 

 described, a spectrum microscope answers in some respects better than 

 an ordinary spectroscope. In the construction of that instrument the 

 aim is to have a wide spectrum, so as to distinctly separate well- 

 marked lines, whereas the absorption bands seen in the spectra of 

 many coloured bodies are often so faint, that they are quite invisible 

 when spread out in a wide spectrum, though readily seen when narrow 

 and more concentrated. This being the aim in constructing the spec- 

 trum microscope, it is necessarily inferior to an ordinary spectroscope 

 when applied to the study of the lines in coloured flames, though, at 

 the same time, it is by no means to be despised for rough experiments of 

 that kind. The microscope should then be made horizontal, and the 

 prism and slit vertical, so that the flame may pass straight in front for 

 some distance. 



3. Spectra of Crystalline Salts and of their Solutions. 



The action of various coloured solutions and crystals has been 

 already studied by many authors ; but, so far as I am aware, not much 

 attention has been directed to the relation existing between the light 

 transmitted by solid or powdered crystals and by their solutions, in 

 order to assist in arriving at a definite conclusion as to their condition 

 when dissolved. This subject can be worked out in a very satisfactory 

 manner by means of the instrument I have described ; for the spectra 

 of solutions and of very minute crystals can be compared side by side, 

 and very slight differences easily detected. I have already collected 

 a large amount of material, and, at first, intended to describe many of 

 the facts in this paper, but I find that so many still remain to be 

 determined, and the conclusions to which they lead are so intimately 

 connected with a general theory of the connection between chemical 

 and mechanical force, that they ought rather to form part of a memoir 

 on that subject. I shall therefore now chiefly confine myself to a de- 

 scription of the requisite apparatus and the manner in which it may 

 be used. 



Highly-coloured solutions may be very conveniently kept in tubes 

 about 3 inches long and J to £ of an inch in diameter, made with one 

 end flat and the other drawn out to a point, which can be sealed up 

 after having introduced the solution by means of an air-pump. These 

 are put lengthwise in front of the slit, as already described ; and when 

 all is properly arranged, the spectrum of the solution is seen extend- 

 ing over half the field of the microscope, divided from the natural 

 spectrum by a narrow dark band, due to the flat end of the tube. I 

 have also often mounted solutions in cells made from thick glass tube, 

 similar to those used to mount objects in liquid, and have examined 

 very faintly-coloiired solutions in tubes varying in length up to 20 



