(, Uiieonwtm Effeils of RefraBion. 



If a pencil of compounded light be imagined to pafs obliqiieiy out of the plane furface of 

 any denfe medium into a vacuum, it is well known tliat the refra£lion will be made 

 from the perpendicular, the violet rays being moft refracted and the red lead, fo that thefe 

 T.ivs will be inclined to each other. But fuppofc, again, that inflead of the rays being 

 fuiTcrcd to pafs into the vacuum, another medium were applied, having a plane furface in 

 contacl with that of the former, and that its power of difpcrfion were fuch as prccifely to 

 counteract the difpe'five power exerted on the rays by the former medium : in this cafe it is 

 evident that the pencil of light would pafs forward colburlefs. Nevcrthclcfs, as the dif- 

 perfive power may be equal in two mediums which differ in mean refra<5live power, it is no 

 Icfs clear, that refra<5Vion will take place wherever thefe two powers arc not precifcly equal ; 

 and that either towards the perpendicular, if the power, or, as it is called, rcfra£livc denfity 

 of the fecond medium, be greater than that of the firfl, or from it if lefs. 



In tlie next place, fuppofe the two mediums not to agree in difperfive power, but that the 

 fecond medium difperfes more than the other. The violet ray will not only be prevented 

 from fcparating on the one fide of the red, but, by the excefs of difperfive power, it will 

 be drawn fo far as to make an angle with that ray on the other fide, and the difperfion will 

 be made the contrary way. If die mean refraftion of the fecond medium exceed that of the 

 firft, the whole pencil will be bended towards the perpendicular, and towards the fame region 

 as the difperfion carried the violet ray ; fo that the efte£l will be fimilar (though lefs in 

 power) to that of a common refra£lion and difperfion by a fingle denfe medium in vacuo. 

 This happens in the tr.infition from crown-glafs into flint-glafs. But if the mean refraiflivc 

 power of the fecond medium be Icaft, the pencil will be bended from the perpendicular, 

 though the violet ray will notwithftanding continue neareft the fame perpendicular. The 

 red ray wilt therefore be the moft defleded, and the violet leaft. This eftccl takes place 

 ■when light is refrafted in the confine of crown-glafs and oil of turpentine, and alfo of many 

 other fluids. If the re fraflive powers of the two mediums upon the mean ray be equal, the 

 direi^ion of this ray wull continue unaltered at its tranfition from the one to the other. But 

 if the difperfive power differs, the extreme rays will form an angle, of which the red ray« 

 will be either at the one or the other limit, according to circumftanccs. For example, if the 

 difperfive power of the firft medium be greatcft, the angle will merely be diminiflied by the 

 aflion of the fecond medium ; fo that the violet ray will be the moft, and the red the leaft, 

 refra£led from the perpendicular ; as would have happened (though more llrongly) in the 

 tranfition into a vacuum. But if, on the contrary, the difperfive power of the fecond me- 

 dium were greateft, the violet ray will be carried more towards the perpendicular than the 

 red by the difference of the two powers. This cafe of refra£lion is found to take place in 

 the confine of crown-glafs and butter of antimony, when the latter is fo far diluted as that 

 both mediums equally refraft the mean ray under equal angles of incidence.' 



In the cafe of refraclion without colour, as before explained, the power of difperfion 

 in the fecond medium was alTumed to be fuch as prerifely to counteraft the effect of the 

 firft ; that is to fay, not only with regard to the extreme red and violet rays, but alfo with 

 regard to the mean and all the intermediate rays. But Doctor Blair's Experiments flicw, 

 thatj'in mediums which difperfe the light but little, the green is the mean refrangible ray; 

 that in by far the greateft number of more difperfive mediums, including flint-glafs, metallic 

 folutions, and eflential oils, the green light is not the mean refrangible order, but forms one 

 6 of 



