REFRACTION. 



that fi)l!o\vs from an liypothefis fo arbi- taiice witliout llieir lurlacc 



coulfquence 

 trary. 



Dechalcs, in order to explain tlic law of rcfradtion, fup- 

 pofesthat every ray of liglit is compofed of fevcral fnrialJer 

 rays, whicli adhere to one another ; and that they are re- 

 frafted towards the perpendicular, in pafiing into a dcnfer 

 medium, becaufe one part of the ray meets with more rc- 

 fiftance than another part ; fo that the former traverfes a 

 fraaller fpace than the latter ; in confcquence of which the 

 raymuft neceflarily bend a little towards the perpendicular. 

 This hypothefis was adopted by the famous Dr. Barrow, 

 ■who, as fome fay, was the author of it. On this hypothefis, 

 it is plain that mediums of a greater refradlive power muil 

 give greater refiftance to the pail'age of the rays of light than 

 mediums of a lefs refractive power, which is contrary to 



ha. 



The Bernouillis, both father and fon, have attempted to 

 explain the caufe of refraftion on mechanical principles ; 

 the former on the equilibrium of forces, and the latter 

 • on the fame principles with the fuppofition of etherial 

 vortices ; but neither of thefe hypotheles have gained much 

 credit. M. Mairan fuppofes a fubtle fluid, filling the pores 

 of all bodies, and extending, like an atmofphcre, to a fmall 

 diftance beyond their furfacos ; and then he fuppofes that 

 the refraftion of light is nothing more than a neceflary and 

 mechanical effeft of the incidence of a fmall body in thofe cir- 

 cumftances. There is more, he fays, of the refrafting fluid 

 in water than in air, lefs in water than in glafs, and in gene- 

 ral lefs in a denfe medium than in one that is rarer. M. de 

 Maupertuis fuppofes that the courfe which every ray takes, 

 in pafling from one medium into another, is that which re- 

 quires the leail quantity of aftion, which depends upon the 

 velocity of the body, and the fpace it pafles over ; fo that 

 it is in proportion to the fum of the fpaces, multiplied by 

 the velocity with which bodies pafs over them. From this 

 principle he deduces the necefnty of the fine of the angle of 

 incidence being in a conitant proportion to that of refraftion : 

 and alfo all the other laws relating to the propagation and 

 , refledtion of light. 



Dr. Smith (in his Optics, Remarks, p. 70.) obferves, that 

 all other theories for explaining the reflection and refrac- 

 tions of light, except fir Ifaac Newtan's, fuppofe that it 

 ftrikes upon bodies, and is refiftcd by them, which has never 

 been proved by any deduftion from experience. On the 

 contrary, it appears by various confideratiotis, and might be 

 fhewn by Mr. Molyneiix's and profeflor Bradley's obferva- 

 tions on the parallax of the fixed ftars, that their rays are 

 not at all impelled by the rapid motion of the earth's atmo- 

 fphcre, nor by the objeft-glafs of the telcfcope through 

 which they pafs. And by fir Ifaac Newton's theory of re- 

 fraftion, which is grounded on experience only, it appears, 

 that light is fo far from being refifted and retarded by re- 

 fraftion into any denfe medium, that it is fwifter there than 

 in vacuo, in the ratio of the fine of incidence in vacuo to 

 the fine of refraftion into the denfe medium. From fome 

 cafes of the produftion of colours, we are led to luppofe, 

 fays Dr. Young (Left. Phil. vol. i. p. 460.) that the velocity 

 of light muft be fmaller in a denfer than in a rarer medium ; 

 and fuppofing this faft to be eftablifhed, the exiftence of 

 fuch an attraftive force as fir Ifaac Newton fuggeits could 

 no longer be allowed ; nor, he fays, could the fyftem of the 

 emanation of light, in oppofition to that of the undulation 

 of an etiierial medium, be maintained by any one. Prieft- 

 ky's Hid. of Light, Sec. p. 102, &c. p. 333, &c. 



The refraftion of light, fir Ifaac Newtcn (hews, is not 

 performed by the very rays falling on the furface of bodies ; 

 but it is done without any contaft, by the aftion of fome 

 power belonging to bodies, and extending to a cei'tain dif- 



VoL. XXIX.' 



by which fame po\\er, ading 

 in other circumilaiiccs, they arc alfo emitted and reflcfted. 



Admitting that certain powers of attraftion and repulfion 

 belong to bodies, and extend to a certain dilbnce beyond 

 their furfaces, and fuppofing alfo that light confifts of par- 

 ticks emitted from luminous bodies, fir Ifaac Newton dc- 

 monflrates, in his " Principia," that the fine of the angle 

 of incidence mufl. be always to the fine of the angle of 

 refraftion in fome certain ratio. For, as he proves geo- 

 metrically, if two fimilar mediums be fcparatcd from each 

 other by a fpace terminated on both fides by parallel planes, 

 and a body, in its pallage through that fpace, be attrafted 

 or impelled perpendicularly towards either of thofe mediums, 

 and not agitated or hindered by any other force ; and if the' 

 attraftion be every where the fame, at equal diftanccs from 

 either plane, taken towards the fame hand of the plane, the 

 fine of incidence upon either plane will be to the fine of 

 emergence from the other plane in a given ratio. 



In a corollary to this propofition he alfo (hews, that if, 

 infteadof one parallel fpace, bodies be furrounded with feve- 

 ral, or an infinite number of them, the attraftive power of 

 each of which difi'ers from the next, increafing or decreafing, 

 Hill, the ratio refpefting any two contiguous ones being 

 given, the ratio of the extremes will be given. He alfo de- 

 monftrates, that, if the caufe of refraftion be the attrac- 

 tion of the refrafting medium, the velocity of light before 

 its incidence will be to its velocity afterwards, as the fine of 

 the angle of refraftion is to the fine of the angle of inci- 

 dence ; fo that light, in paflfing from a rarer medium into a 

 dcnfer, receives an increafe of velocity. The fame things 

 being fuppofed, he (bews, that if the velocity before inci- 

 dence be greater than afterwards, fo that the angle of 

 emergence muft always increafe, in confequence of the re- 

 pulfive power prevailing over the attraftive one, the body 

 will at length be reflefted ; and that, in this cafe, the angle 

 of refleftion muil neceflarily be equal to the angle of 

 incidence. 



That light does not confiit of any mode of aftion upon a 

 fluid medium, he concludes from having demonftrated that 

 preflurc cannot be propagated through a fluid in reftilinear 

 direftions, unlefs where the particles of the fluid lie in a 

 right line, but muft diverge into the unmoved fpaces ; and 

 that every tremulous body, in an elaftic medium, propagates 

 the motion of the pulfes on every fide ftraight forwards. 

 Admitting this, there is hardly any hypothefis that will 

 agree with the phenomena of light, but that of particles 

 emitted from the luminous body, affefted by the attradtiona 

 and repulfions of other bodies. 



The fame arguments by which we have proved, that re- 

 Jtettion (which fee) is performed without immediate contaft, 

 will go a great way towards demonftrating the fame of re- 

 fraftion : to which may be added the following ones. 



1. Becaufe, if when light falls out of glafs into air with 

 the utmoft obliquity it will be tranfmitted at, it be then 

 made to fall a little more obliquely, it becomes wholly re- 

 flefted : for the power of the glafs, after it has refrafted 

 light emerging as obliquely as poflible, fuppofing the rays 

 to fall ftill more obliquely, will be too ftrong to let any of 

 the rays pafs ; confequently, inftead of being refrafted, they 

 will all be reflefted. 



2. Becaufe in thin laniellx, or plates of glafs, light is re- 

 flefted and tranfmitted feveral times alternately, as the thick- 

 nefs of the lamellae increafes in arithmetical progrefllon ; for 

 here it depends on the thicknefs of the lamina which of the 

 two the glafs ftiall do, whether refleft it, or let it be tranf- 

 mitted. 



3. Becaufe, whereas the power of other bodies both to 

 refleft and rcfraft light are ver^' nearly proportional to their 



4 E dcnfities ; 



