126 Oti the unequal Rejlexibilky of Lighi. 



writers have not overlooked this circumftance. Robert Smith, after having explained the 

 9th experiment of Newton's Optics, fays, or makes Newton fay^ " I did not obferve any 

 refraiStion at the fides A C, A B, of the firft prifm, becaufe the light was nearly perpendi- 

 cular to the firft fides, and emerged nearly perpendicular to the fecond ; and, in confequence, 

 underwent none, orfo little, that the angles of incidence at the bafe B C were not fenfibly 

 altered ; particularly if the angles of the prifm at the bafe B C were each about 40°. For 

 the rays F M begin to be totally refleded when the angle C M F is about 50° ; and, 

 confequently, they will then form an angle of 90° with A C." Smith's Optics tranflated 

 into French, by L, P. P. (Pezenas). i. §. 173. p. 190*. 



From all this it follows, that when the angle of a prifm is well chofen, a white ray, falling 

 perpendicularly on its anterior furface A C, may be decompofed ; becaufe it is partly, and not 

 totally, refledted, the violet ray being reflefted while the red is ftiU trarrfmitted. 



I muft here remark, that in order to repeat this experiment, and render it conclufive, it is 

 not neceffary to circumfcl-ibe the angle C, within the limits which render the ray F Al per- 

 pendicular (or nearly fo) to the anterior face A C. All the reafoning we have offered on 

 this head (§ 2.) will be juft, provided that, in the firft pofition of the prifm, the angle 

 A P F be greater than its fupplement F P C. But in order that this circumftance fliould 



take place, when the reflection is made at the point M, it is fufficient that the rcfrading 



angle C be lefs than 40°. 



Interefting refults may be obtained, by varying this experiment ; but the authorities I have 



quoted, leave no doubt with regard to the particular refult, which we have to examine. The 



mofl refrangible rays are reiledted at a lefs angle of incidence. They are more reflexible in 



the Newtonian jenfe. 



Section 3. But Mr. Brougham fupports the contrary opinion, by an experiment which he 



thus announces : " I held a prifm vertically, and let the fpectrum of another prifm be reflected 



by the bafe of the former, fo that the rays had all the fame angle of incidence ; then turning 



round the vertical prifm on its axis, when one fort of rays was tranfmitted or reflected, all 



were tranfmitted or reflected -t^." 



As the complete difcuflion of this experiment would require fome detail, I fhall content 



myfelf with obferving, that the plane of the vertical face cannot be adjufted, fo as to receive 



tntrance and emergence of the rays, with regard to the prifm, might, in fome refpeft, aflift in producing 

 thefe efFefts, it m^y be obfcrved, that the fame effefts are produced, whatever may be the magnitudes of the 



angle AC B ; that is, whatever may be the refraftion of the furface AC It may even be contrived, 



that, when the colours are partly refiefted .... and partly tranfmitted the rays may fall perpendi- 

 cularly on A C, and A B, at their entrance and emergence, and they be refrafted by neither furface, pro- 

 vided the angle A C B be made of about 40°, and yet the fame effefts are produced. N. 



* The paffage here marked as quotation, is retranflated from the French; and, confequently, is not given 

 in the very words of Smith, whofe work I have not. N. 



f The fame experiment, tried by Newton, afforded precifely the contrary refult. « Radii purpuri- 

 formes prime omnium refledluntur et uhimo rubriformcs." Left. Opt. Opufcul. ii. 120. 



all 



