3^4 iJew Phenomena of the Tranftriijfton and ReJleB'ton of Light, 



\Fhich from their immobility feemed as if attached to the lower glafs. Thefc fcarcely 

 moveable colours, however, were quickly moved; and brought nearer to each other by raifing 

 one edge of the glafs with the hand ; and they continued vifible when the glafs was lifted 

 as high as one-fortieth or thirtieth of an inch, by flipping a knife under its edge. When 

 I plate of metal was placed beneath the lower glafs, and eleftricity (which from the difpofi- 

 tion of the machine happened at that time to be negative) was communicated to the upper 

 plate, the vivid colours appeared, and the adhefion of the glafles was increafed ; but the 

 fcarcely moveable ranges remained little if at all afFe£ted. When the glafles were taken from 

 the machine towards the window, the vivid colours flowly receded as the eleilricity was 

 difperfed, and in this ftate they were very advantageoufly and evidently feen crofling the 

 ether lefs moveable ranges. 



Thefe fads appeared at that time to lead to no other conclufion than that the moveable 

 ranges had been difturbed by fome other circumftance befides that of the mere diftancc of 

 the plates : for, if the diftance had indeed been varied, it feems reafonable to fuppofe that 

 both fets of ranges would have been affe£ted. 



Soon afterwards, upon making fome obfervations with the artificial horizon, and a 

 fextant conftruded by Troughton, I obferved a feriesof colours in the horizon-,glaf3, when 

 the pofition of zero was to be afcertained. They appeared both in the filvered and the 

 clear part. The pofition of the glafles, when the colours were feen, was very nearly 

 parallel as to the vertical fituation of the planes ; but it admitted of the index being moved 

 through nearly forty minutes before they difappeared. Thefe glafles, namely the index-" 

 glafs, and the horizon-glafs, were 3! inches afunder. 



This laft fa£b appears to juftify the inference of Newton, who confidered the colours of 

 thick and thin plates as depending on the fame caufe ; which dodrine was applied by Dr. 

 Pemberton, to account for the numerous ranges of colours fometimcs obferved beneath the 



common rainbow *. 



As one of the greateft difliculties in Newton's dodrine feems to have been that the fits 

 of refledion and tranfmiflion are fuppofed to extend to vaft diftances, it feemed defirable to 

 repeat this experiment with as great an interval between the glafles as could conveniently be 

 had. With this view I placed one plate of glafs on the furfacc of a veflel of mercury, and 

 held anotTier in my hand parallel to it, in fuch a pofition that the refleded light from the 

 firfl: pafled through the fecond glafs to my eye. The white clouded fky and the chimney 

 of an oppofite houfe were feen by refledion in both glafles, and it was eafy to move the 

 Mpper glafs till both images of the chimney coincided. In this pofition the glafles would 

 have been parallel if the objed had been indefinitely dift;ant ; but in the prefent cafe, the 

 parallelifm could not be obtained but by a flight elevation of the moveable image. By this 

 difpofition the colours were made to appear when the plates were four feet afunder, and I 

 have no doubt but that the fame would have happened at much greater diftances if the im- 

 perfed method of adjufting the parallelifm could have been applied. 



The fame eff*ed, as might naturally be expeded, took place when the eye was fo placed 

 as to receive the refleded light from the lower glafs, after it had been tranfmitted through th« 

 vpper. 



♦ Philof. Tranf. abridged, vol. vii. or Prieflley's Optics, p. j9«, , 



In 



