1^6 • On the Properties of Light. — Colours from Sftcuttr. 



been in Obf. lO; but by letting the rays fall on the half next the chart, and inclining that^all 

 very much, I could produce them, though lefs diftin^l^Iy, by a fingle reflexion, I now held a 

 plain metal fpeculum, fo that the rays might be refle£led to form a white image on a chart. 

 On inclining the fpeculum much, 1 faw the image turn red atthe edge ; it then became a 

 little diftended ; and laftly, fringes emerged from it well coloured, and ta regular order, wi;h 

 their dark intervals. This may eafily be tried by candle-light with a piece of looking-glafs ; 

 and thofc who without much trouble would fatisfy themfelves of the truth of the whole experi- 

 nient contained in this and the lafl obfervation, may eafily do it in this way with a concave 

 fpeculum ; but the beauty of the appearance is hereby quite impaired. After this detail it is 

 almofl: fiiperfluous to add, that the fringes at B, fig. 6, are formed by deflexian from the 

 edge of the fpeculum next the fun, and then falling oa it are rcfle(5led to the chart ; that the- 

 images by reflexion are either formed by the light being decompounded at its firil reflexion, 

 and then undergoing a fecond, or, in other inftances, without this fecond reflexion; and that 

 the other fringes are produced exactly as defcribed above, from the neceflary confetiuences 

 cf the theory. I fliall only add, that nothing could have been more pleafing to me than the 

 fuccefs of this experiment; not only becaufe in itfeif it was really beautiful from its variety,. 

 but alfo becaufe it was the moft peremptory confirmation of what followed from the theory 

 a priori, and in that point where the fingularity of its confequences molt inclined me to 

 doubt its truth. 



Let us now attend to feveral eonclufions to which the foregoing obfervations lead, inde- 

 pendently of the propofitions {viz. the five firft) which they were made to examine. 



I. We muft be inmediately flruck with the extreme refemblance between the rino-s fur- 

 rounding the black fpots on the image made by an ill-poli(hed fpeculum, and thofe produced 

 by thin plates obfervcd by Newton ; but perhaps the refemblance is (till more confpicuous in 

 tile colours furrounding the image made by any fpeculum whatever, and fully defcribed in 

 Obf. 10 and ii. The only difl^srence in the circumftances is now to be reconciled. The 

 Zings furrounding the black fpot on the top .of a bubble of water, and thofe alfo furrounding 

 the fpot between two objcfl-glafles*, have dark intervals (exadlly like thofe rings I have 

 juft now defcribed, and the fringes furrounding the fhadows of bodies) j but thefe intervals 

 Iranfmit other fringes of the fame nature, though with colours in the reverfe order; from 

 which Sir Ifaac Newton juflly inferred, that at one thicknefs of a plate the rays were tranf- 

 mltted in rings, and at another refle£led in like rings. Now it is evident, that neither re- 

 fiexibility nor refrangibility will account for either fort of rings ; becaufe the plate is far too 

 thin for feparating the rays by the latter, and becaufe the colours are in the wrong order for. 

 the former ; and alfo becaufe the whole appearance is totally unlike any that refrangibility 

 and reflexibility ever produce. To fay that they are formed by the thicknefs of the plates, is 

 rot explaining the thing at all. It is demanded, in what way? And indeed we fee the like 

 dark intervals and the fame fringes formed at a diflance from bodies by flexion, where there 

 is no plate through which the rays pafs. The ftate of the cafe then feems to be this : " When 



* a phenomenon is produced in a particular combination of circuii.ilaaces, and the fame 



* phenomenon is alfo produced in another combination, where fome of the circumftances 

 " before prcfent arc wanting ; we are entitled to conclude, that tiic latter is the mofl gene- 



* Optics, bopk ii. p. i. 



«ral 



