[ 249 ] 



XII. Researches towards establishing a Theory of the Dispersion of Light. By the 

 Rev, Baden Powell, M.A. F.R.S. Savilian Professor of Geometry in the Uni- 

 versity of Oxford. 



Received February 19, — Read March 12, 1835. 



Introductory Remarks. 

 JL HE phenomena of prismatic dispersion, as originally discovered by Newton, and 

 since examined throughout a vast range of transparent bodies by succeeding philo- 

 sophers, especially Sir David Brewster, were principally considered with reference 

 only to successive parts, or spaces, of the coloured spectrum, designated generally as 

 the red, or violet, or mean rays. 



The increasing precision of modern science has been evinced in the elaborate and 

 justly celebrated researches of M. Fraunhofer, who, availing himself of the dark and 

 bright lines to mark and designate distinct points of the spectrum, by prismatic ob- 

 servations for ten different media, solid and liquid, has determined in each the re- 

 fractive indices for seven principal rays, thus always absolutely identifiable. We will 

 use the term "definite rays" to signify the specific parts of the spectrum thus defined. 

 As to the law of the phenomena, the first notion of a simple proportionality was soon 

 disproved. The refrangibility was seen to vary considerably and irregularly for each 

 ray and each medium ; and when Fraunhofer had assigned serieses of numbers as 

 the accurate expressions of the varying refractive powers throughout the several 

 spectra, the apparent absence of any law connecting these numbers was only ren- 

 dered more palpable. All that could be said was, that the numbers increased from 

 the red to the blue end of the scale, and in a different way in each medium. 

 ' The first object of inquiry in the search after such a law, would be some other cha- 

 racteristic of the same definite rays, equally well determined ; between which and 

 the refractive index some connexion might possibly be found to subsist. 



The only such characteristic, perhaps, is the length of the interval for each ray, 

 the Newtonian fit, or the undulatory wave, which (by whatever name it be called,) 

 has demonstrably a real existence in the nature of light ; and the value of which, for 

 each of the definite rays, has also been determined by Fraunhofer, with his usual 

 accuracy, from phenomena totally independent of refraction, viz., his very remark- 

 able experiments, in which a spectrum absolutely pure and perfect is obtained with- 

 out the intervention of any prism, from the interferences produced by a fine grating 

 of parallel wires covering an object-glass. The positions assumed by the successive 



MDCCCXXXV. 2 K 



