On the Properties of Light.— Colours from Specula. 19.7 



♦'rat cafcy and muft try to refolve the other into it." In the firft place, the order of the 

 colours in the Newtonian rings is juft fuch as flexion would produce-, that is, thofe which 

 are tranfmitted have the red innermoft, thofe which are reflefted have the red outermolt : 

 the former are the colours arranged as they would be by inflexion, the latter as they would 

 be by deflexion ; and here by outermofl and innermoft muft be underftood relative pofuiofi' 

 only, or pofitlon with refpe£t to the thicknefs of the plate, not of the central fpot. Second- 

 ly, the thinneft plate makes the broadeft ring, (the diameter of the rings being in the inverfe 

 fubduplicate ratio of the plate's thicknefs) : juft fo is it with fringes by flexion ; nearer the- 

 body of the fringes are broadeft, and their diameters increafe in the fame ratio with the. 

 diameters of the ringsby plates whofe thicknefs is uniform ; each diftance from the bending 

 body therefore correfponds with a ring or fringe of a particular breadth, and the alternate. 

 diftances correfpond with the dark intervals. The queftion then is, what becomes of the 

 light which falls on or paffes at thefe alternate diftances ? In the cafe of thin plates, this light 

 is tranfmitted in other rings : we fliould therefore be led to think, that in the cafe of the 

 light palling by bodies, it fhould be at one diftance infleiSled, and at another defledled : and. 

 in fa6l the phenomena agree with this ; for fringes are formed by inflexion within the, 

 fliadows of bodies ; they are feparated by dark intervals ; the fringes and the intervals with- 

 out the Ihadow decreafe in breadth according to the fame law, fo that the fringes and inter- 

 vals within the (hadow correfpond with the intervals and fringes without refpedively. Nor ' 

 Hwill this explanation at all affecl the theory formerly laid down ; it will only (if found con— 

 fjftent with farther indudlion). change the definite fpheres of inflexion and deflexion into- 

 alternate fpheres. At. any rate, the fa£ls here being the fame with thofe defcribed by 

 Newton, but in different circumftances, teach us to reconcile the difference, which we have 

 attempted todo, as far as is confiftent with firidnefs; and what we have feen not only en- 

 titles us to concludq that the caufe is the fame, but alfo inclines us to look for farther light' 

 concerning that caufe's general operation: and I truft fome experiments which T. have, 

 l)lanned, with an inftrument contrived for the purpofe of inveftigating the ratio of the bendin* 

 power to the diftances at which it a£ls, will finally fettle this point. 



XL Another conclufion follows from the experiments now related, viz. that we fee the ■ 

 great importance of having fpecula for refle£tors delicately polifhed, not: only becaufe the 

 more dark imperfeftions there are on the furface the more light is loft, and the more colours 

 are produced by flexion (thefe colours would be moftly mixed, and form white in the focus), 

 but alfo becaufe the fmaileft fcratches or hairs being poliftied produce colours by reflexion^, 

 and thefe diverging irregularly from the point of incidence are never colledted into a focus,,, 

 but tend to confufe the image. Indeed it is wonderful that reflcftors do not fuffer more from . 

 this caufe, confidering the almoft impoflrbility of avoiding the hairs we fpeak of: however, . 

 that, they do actually fufl^er is proved by experience. I have tried feveral fpecula from re- 

 flefting telefcopes, and found that though they performed very well from having a good 

 figure, yet from the focus (when they were held in the fun's light) feveral ftreaks diverged, 

 and were never correfted ; others had the hairs fo fmall that it was very difficult to per- 

 ceive the colours produced by them unlefs they fell on the eye. Glafs concaves were freer,- 

 from thefe hairs, but they were much more hurt by dark fpots, &c. In general the hairs . 

 are fo fmall in well wrought metals that they do little hurt ; but v/hen enlarged by any ' 

 length, of . expofure to the light and, heat in foIar.obfervtttions> they produce irregularities, 



round: 



