404 Dynamic Theory. 



Spectrum Analysis. The manner in which the spectrum is produced 

 by a pr{sm was mentioned in the last chapter. There is another way of 

 producing a spectrum, and that is by diffraction of light by means of 

 gratings ruled on glass or on speculum metal. They may be ruled at 

 the rate of 400 to the inch, and from that up to 20,000 per inch. The 

 diffraction of the light analyzes it and produces the colors the same as 

 the prism. They are not so bright as the prismatic colors, because the 

 gratings produce several spectra which overlap each other. But they 

 separate the colors in proportion to their several wave lengths, and they 

 show the greatest heating qualities to be in the yellow and orange rays. 

 The prism shows them in the invisible part of the spectrum below the 

 red. It is thought that the prism gives too much dispersion to the 

 violet and extra-violet rays, and too little to the red and infra-red rays. 

 According to the prismatic dispersion, if the whole spectrum be divided 

 into 320 parts, about 120 will belong to the visible portion, while the 

 other 200 are divided between the infra-red and the extra-violet, giving 

 the largest moiety to the latter. But by the diffraction spectrum a 

 much longer proportion represents invisible heat and a much shorter 

 proportion is above the visible violet. According to Prof. Roscoe, the 

 total length of the observed spectrum is between three and four octaves, 

 the visible part being but one. 



As said in the last chapter, all incandescent liquid or solid bodies form 

 continuous spectra, while incandescent vapors or gases form spectra 

 made up of disconnected bands of color, scattered along in different 

 parts of the space which would form the spectrum of a solid. 



These discontinuous spectra are the signatures of the various alkalies, 

 metals, gases, &c. , when they are raised to an incandescent vapor. And 

 no two of them make the same sign. The signature of some consists 

 of only a few bands across the spectrum, others have many iron has 

 six hundred lines across the spectrum. If a number of substances, say 

 a dozen or twenty, be mixed together and all vaporized by intense heat, 

 each one will write its peculiar marks upon the same spectrum, and the 

 spectroscopist can pick them all out and identify them with perfect ease 

 and accuracy. There are, in the aggregate, some thousands of these 

 lines, and yet only six of them are found to belong to more than one 

 substance. These six belong to two each unless better instruments 

 with more dispersive power shall show them to be separated by a small 

 space. Next must be noticed the extreme delicacy of the action of the 

 spectroscope. Often a quantity of a substance so small that it cannot be 

 perceived in any other way, will make its mark on the spectrum. The 

 18 millionth part of a grain of salt if vaporized will mark the D line 

 across the spectrum. Indeed it is difficult to get a spectrum without 

 having the D line ( at first ), because there is salt enough in the air 



