154- Jmpe}-fec}ii)Hs from Colour In Speculum!, 



fourth propofition is merely a finipJe cafe of flexion. And the two laft require no illuftra- 

 tion, I (hall now relate how 1 inquired into the truth of thefe things a pofteriori. 



Ohftyvat'ion i.— Locking at a plane glafs mirror expofed to the fun's light, I obfervcd that 

 up and down its furhce there were minute fcratches (called hairs by workmen), and that 

 each of thefe reflecled a bright colour, fome red, others green, and others blue. On mov- 

 ing the mirror to a difll rent inclination, or my eye to a different pofition with refpeil to the 

 mirror, I faw the fpecics of the colours change ; the red, for inftance, became oreen, and 

 the green blue. I applied my eye clofe to the mirror, and received on it the li"-ht refle£ted 

 from one hair. I obferved feveral diftinft images of the fun much diftended and regularly 

 coloured, juft like thofc defcribed above ; the fame appearances were obfcrvable in all fpe- 

 cula, metal and glafs, which had thefe hairs, and 1 ne\er fa?/ any metal one without fome : 

 their fize is exceedingly fmall, not above -^-J-^^ of an inch. Rubbing a minute particle of 

 greafe on the furface of the fpeculum, images were fc-en on the fibrous furface; and they 

 always lay at right angles to that direction in which the greafe was difpofed by drawing the 

 hand along it. 



Ohfervatioti 2. — Befides thefe polifhed hairs, many fpecula have fewer or more fmall fpecks 

 and threads, rough and black. Perhaps every polifhed furface is fludded with a number of 

 fmall ones, invifible to the naked eye from the quantity of regular light which it reflefls. 

 I took, from a reflecting telefcope, a fmall concave fpeculum not very well finifhed j its 

 furface (hewed feveral fpecks to the naked eye, and many with a microfcopc. Its diameter 

 was |-Jofan inch, its focal diflance two inches, and the fphere to which it was ground 

 eight inches diameter. I placed it at right angles to the rays of the fun, coming throuo-h a 

 fmall hole ^of an inch diameter, into a very well darkened room ; I then moved it verti- 

 cally, fo that the rays might be refleiSled to a chart 12 inches from the fpeculum, and con- 

 fequently 10 from the focus: and though the focus appeared v/hite and bright,^ yet on the 

 chart, the broad image was very different. It was mottled with a vaft number of dark fpots ;. 

 thefe were of two forts chiefly, circular and oblong. Of the former a cunfiJcrable number 

 were diflinft and large, the reft fmallcr and more confufed, but fo numerous that they 

 feemed to fill the whole image. None were quite black, but rather of a blucifli grey, and, 

 the oblong ones had a line of faint light in the middle, juft as is the cafe in fhadows of fmall 

 bodies. But the chief thing which I remarked was the colours. Each oblono- and round 

 fpot was bordered by a gleam of white, and feveral coloured fringes feparated by fmall. 

 dark fpaccs. The fringes were exactly like thofc furrounding the (hadows of bodies, of the 

 fame (hape with the dark fpace, having the colours in the order, red on the outfide, blue or 

 violet in the infide — the innermoft fringe was broadefl, the others decreafin* in order from 

 the firft. I could fometimes fee four of them, and, when made at the edge of the large 

 image, I could indiitin£lly difcern the lineaments of a fifth:, when two of the fpots were 

 very near one another, their rings or fringes ran into one another, crofling. 



Ohfervation 3. — When the chart was removed to a greater diflance, as fix feet the 

 fringes were very diflin£l and large in proportion ; alfo the fmaller fpots became more plain 

 and their rings were feen, though confufedly, from mixing with one another. When the 

 fpeculum was turned round horizontally, fo that its inclmation to the incident rays might 

 be greater, the diftance of the chart remaining the fame (by being drawn round in a circle), 



the 



