QQ CHEMICAL EFFECTS OF THE INTERFERENCE SPECTRUM. 



212. The fixed lines, used in this way, enable us at once to divide the interference 

 spectrum into any number of parts, and to indicate effects either in space or in time. 

 For chemical purposes, in which mathematical accuracy is scarcely attainable, all that 

 we have to do, in order to determine the lengths of waves producing given effects, or 

 the times of vibration, is to determine, upon the interference spectrum, the point at 

 which the change in question has taken place, and, using the numbers which FRAUN- 

 HOFER measured for the different fixed lines, find what is its relation to them. Thus, 

 for example, suppose it has been found, by experiment, that a certain substance, expo- 

 sed to the interference spectrum, exhibited a maximum point of decomposition exactly 

 midway between the fixed lines F and G, which are both impressed on it, it is required 

 to know what is the length of the wave which brought about that decomposition in 

 parts of a Paris inch. The numbers given by FRAUNHOFER are, 



ForG 0-00001587, 



" F 0.00001794; 



and the point in question, being midway between the two, has for its wave-length 

 0-00001690. 



213. In the frontispiece, I have annexed FRAUNHOFER'S numbers to each ray, omit- 

 ting, for the sake of brevity, the ciphers. 



214. In Jig. 134 is represented the effect of the interference spectrum upon a silver 

 plate, rendered sensitive by exposure to iodine vapour, and then to bromine. The time 

 of its exposure in the camera was half an hour. The point of maximum falls, as may 

 he determined upon the foregoing principles, nearly at the point 0-00001538. The 

 dotted lines, x and y, indicate the beginning and end of the stain. The maximum 

 point does not fall equally between the two, but is nearer to the more refrangible ex- 

 tremity. When this spectrum is compared with the corresponding prismatic one, in 

 Jig. 103, we see how great is the difference ; the effect, which is there carried far be- 

 yond the extra-violet regions, is here compressed down into a narrow space. 



215. Fig. 135 represents a very beautiful result, obtained on a silver plate exposed 

 first to the vapour of iodine, and then for a short time to the vapour of chloride of 

 iodine. In this case the point of maximum falls nearer the line G. The time of ex- 

 posure was ten minutes. The fixed lines, which were discovered by M. BECQUEREL and 

 myself beyond the visible spectrum, are here crowded down into so small a space that 

 the individual groups are all found together, so that they would scarcely be recognised. 



216. Fig. 136 gives a very perfect result which I obtained on a plate exposed first 

 to the vapour of iodine, then to bromine, and then to chloride of iodine ; the point of 

 maximum falls at 0-00001538, as \njig. 134; the time of exposure was one hour. 

 The decomposition extended on one side beyond E, in the green space, to the point 

 0-00002007 ; and beyond the violet space to the point 0-00001257. As in Jig. 134, 

 we here again see the point of maximum is not in the middle of the image, but is to- 

 wards its more refrangible extremity. 



217. As respects this spectrum {fig. 136), although it extends to the yellow space, the 

 line E is not represented in it. I may remark of these spectra, as was formerly re- 

 marked (Ap., 746-747) in the case of the prismatic, that the lines D and E do not ap- 



