evinces. But when the parts are much denser than the ambient medium, this variation is not fo confiderable; and therefore, the rays which are reflected leaft obliquely may predominate over the reft, fo much as to cause a heap of fuch particles to appear very intenfely of their colour.

M And hence the magnitude of the component parts of natural bodies may be conjectured by their

N

colours.

For, fince the parts of these bodies are of about the fame density as water or glass, as by many circumstances is obvious to collect, it is highly probable that they exhibit the fame colours with a plate of equal thickness. That colour being known, the thickness may be easily found by the preceding obfervations.

CHA P. IX.

OF THE INFLECTIONS OF THE RAYS OF LIGHT WHICH PASS IN THE VICINITIES OF BODIES.

It is obfervable, that if a beam of the Sun's light be let into a dark room through a very small hole, the shadows of things in this light will be larger than they ought to be if the rays went on by the bodies in ftraight lines, and that these shadows have three parallel fringes, bands, or ranks of colours adjacent to them. The principal circumstances of the phenomenon are as follow:

If a beam of the Sun's light be admitted into a darkened chamber through a hole of the breadth

of

of the forty-fecond part of an inch, or thereabouts, the shadows of hairs, thread, ftraws, and other small bodies, appear confiderably broader than they would be if the light paffed by them in straight lines. For example; a hair, whofe breadth was the 280th part of an inch, being held in this light at the diftance of about twelve feet from the hole, did cast a fhadow which, at the distance of four inches from the hair, was the fixtieth part of an inch broad, that is, above four times broader than the hair; and at the distance of ten feet, was the eighth part of an inch broad, that is, thirty-five times broader.

Nor is the effect altered by an alteration in the denfity of the medium contiguous to the hair, for its fhadow at like diftances was equal, whether it was in the open air, or inclosed between two plates of wet glafs, care being had that the incidence and emergence of the ray was perpendicular to the glaffes. Scratches on the furface or veins in the body of polished glaffes did alfo caft the like broadfhadows. And therefore the great breadth of these shadows must proceed from fome other cause than the usual refraction which might arife from any action of the ambient medium.

Let the circle x (fig. 84) represent the middle R of the hair; A DG, BEH, CFI, three rays paffing by one fide of the hair at feveral diftances; KN Q, LOR, MPS, three other rays paffing by the other fide of the hair at the like diftances; D, E, F, and N, O, P, the places where the rays are bent in their paffage by the hair; G, H, I, and Q, R S,

the

the places where the rays fall on a paper, GQ; IS the breadth of the fhadow of the hair caft on the paper; and TI, vs, two rays which fall on the points 1 and s, without being at all deflected by the action of the hair. Then it is manifeft, that all the rays between TI and v s are bent in paffing by the hair, and turned aside from the shadow is, because if any part of the light were not bent it would fall within the fhadow, and there illuminate the paper, contrary to experience. And becaufe, when the paper is at a great distance from the hair, the shadow is broad, and therefore the rays TI and vs are at a great diftance from each other, it follows that the hair acts upon the rays of light at a confiderable distance in their paffing by it. But because the shadow of the hair is much broader in proportion to the distance of the paper from the hair when the paper is nearer to the hair than when it is at a great distance from it, it is evident that the action is stronger on the rays which pass by at least distances, and grows weaker and weaker accordingly as the rays pafs by at diftances greater and greater, as is reprefented in the fcheme.

S The fhadows of all bodies in this light are bor

dered with three parallel fringes or bands of coloured light, of which that contiguous to the fhadow is broadest and most luminous, and that moft remote from it is narroweft, and fo faint as fcarcely to be vifible. If the light be received very obliquely on paper, or any other smooth white body, the colours may be plainly distinguished in this order, viz. the

firft or innermoft fringe is violet, and deep blue next the fhadow, and then light blue, green and yellow in the middle, and red without. The fecond fringe is almost contiguous to the first, and the third to the fecond, and both are blue within, and yellow and red without, but their colours are very faint, especially those of the third. The colours therefore proceed in this order from the fhadow, violet, indigo, pale blue, green, yellow, red; blue, yellow, red; pale blue, pale yellow, and red.

If a larger beam of the Sun's light be admitted T into a dark chamber, and part of it received on the blade of a sharp knife, whose plane interfects the direction of the beam at right angles, while the other part is fuffered to pafs by the edge of the knife, and received on a paper at the distance of about three feet; this laft light will appear to shoot out or fend forth two faint luminous ftreams both ways into the fhadow, fomewhat like the tails of comets. Thefe ftreams being very faint, are fo much obfcured by the light of the principal direct rays, that it is neceffary, in order to fee them with any degree of distinctness, to let the direct rays pass through a hole in the paper on to a piece of black cloth. The light of the streams is then perceptible on the paper to the distance of fix or eight inches from the Sun's direct light each way, and in all the progrefs from that direct light decreases gradually till it becomes infenfible.

If two knife-blades, with ftraight edges, be fou fixed

fixed or fet in a frame, that they may both be fituated in the fame plane, their edges parallel, and facing each other, and one of the blades moveable towards or from the other by means of a screw, fo that their parallelifm may be always preferved, a beam of light may be fuffered to pass between their edges, and the appearances are the following: when the knives are at a confiderable distance, so that the intromitted beam is broad, the streams of light which shoot both ways into the fhadow are fcarce vifible, for the reafon already mentioned, and the edges of the fhadows are not bordered with coloured fringes, they becoming fo broad that they run into each other, and by joining, form one continued light or whitenefs at the beginning of the ftreams. As the knives approach each other the fringes of colour appear on the confine of each shadow, becoming diftincter and larger' until they vanish, which happens when the edges are diftant somewhat more than the 400th part of an inch. After the fringes have disappeared, the line of light, which was in the middle between them, grows very broad, enlarging itself both ways into the streams of light afore-mentioned; and when the knives are diftant above the 400th part of an inch, the light parts in the middle, and leaves a fhadow between the two parts. And as the knives still approach each other, the fhadow grows broader, and the ftreams shorter at their inward ends, which are contiguous to the shadow, till upon the contact of the knives