The first part of this author's inquiries relates to the size of the shadows of objects when the light of the sun passes through a small hole, and is afterwards intercepted by an opaque body. Our author, in repeating the experiment, found the light much more divergent than if it had passed from the hole in strait lines, and more dilated than it ought to be by the penumbral light of the solar disc. This he seems to ascribe to the attraction of the edges of the aperture: but the penumbral light of the sun itself requires some explanation, and must be attributed to the mutual repulsion of the particles of light, when not hindered by the lateral pressure of light around. We must however follow our author, and shall describe the other changes in the condition of the ray in his own words, with his explanation.

The light of this beam was very equally diffused, and almost uniformly dense in all its internal central parts, being also tolerably well defined on its external parts or edges, in a manner very different from what it ought to have been by the penumbral light alone. But the appearances were somewhat different, with holes of different dis mensions.

With a hole one-tenth of an inch wide, or wider, the centre of the beam was composed of the dense direct light of the sun, unchanged in its passage; but farther therefrom, towards the borders of the beam, this light began to decrease in density, and gradually decayed more and more in the approaches nearer and nearer to the borders, becoming at last considerably diluted and evanescent, and rendering the edge of the beam ill-defined and indistinct.

With a smaller hole than the last, the central dense light entirely disappeared; and, with a hole yet smaller than this, of the dimensions of the first before-mentioned hole, the external edges of the beam became more condensed and better defined, and the whole beam of light became, as before described, of more uniform density in all its parts. With a hole smaller than any of the foregoing, about the hundredth part of an inch wide, various colours began to appear in the beam, the central parts of which became now, in their turn, more diluted than the rest, the external parts denser than these, and bordered with tinges of yellow and red light on the very edge or margin of the beam.

All these appearances are to be ascribed to the same attractions of the edges of the holes, and of the different parts of the edges. These, when the hole is large, affect only the parts of the light passing nearest to them; when the hole is reduced, they attract and dilate the whole of the passing light; when the hole is yet more considerably diminished, as in the first part of this observation, they act not only each part upon the light passing nearest to each, but each part also upon the light passing nearest to each opposite part of the edge, condensing by diminishing the attraction and diffusion of the light on the edges of the beam, and rendering the whole more equably and uniformly divergent; and these at last, when the hole is in its most reduced state of about a hundredth part of an inch wide, by their various actions produce colours in the passing light in the manner and upon the principles hereafter to be explained.

In the beam of solar light passed through the small hole, onefiftieth part of an inch wide, I observed the shadows of very slender bodies, pins, needles, straws, hairs, to be considerably broader, as they ought to be in this divergent light, than the bodies themselves; but as each of these bodies exercises upon the light passing by it the same attractions by which the light is bent in passing through the hole, I concluded that a part of the light would be in every case bent, in passing by, towards the body into the shadow, and illuminate it and diminish its breadth. In this conclusion I was confirmed by the variations of dimensions of the shadows measured by Newton, and by the differences of proportion he observed to prevail in their breadths with respect to the distances of observation from the hole.' P. 8.

These appearances are, we know, related with accuracy; but we have always found it very difficult to explain them. In the smallest hole of one-hundredth part of an inch, the light seems to have been wholly attracted by the edges, so as to leave a darker part in the centre-and we shall find some collateral facts to support this idea; and it is this very attenuated light that becomes coloured when seen against the darker air. In general, it must be observed that the opaque body attracts the light, and turns it in a direction towards itself: thus the light becomes more divergent when passing through the hole, and shadows become smaller than from the diverging rays they otherwise would be.

Our author's account of the fringes of colours differs greatly from that of sir Isaac Newton. We shall select the compara

tive view.

On this subject we shall only remark, with our author, that the order is variable, and that the fringes are chiefly conspicuous when the light is thrown obliquely, and of course greatly expanded. The proportions of their breadth also differ considerably in different circumstances; so that the regular progression pointed out by sir Isaac Newton is not so constant as to support any analogy. Our author's observation on the shadow also merits particular notice.

When the distance of observation, from the hair under the cir cumstances of the experiment in observation ß, was very small, and before the first bright streak of light began to appear, the shadow of the hair was distinct and well defined, and of intense blackness. At a greater distance, this shadow appeared to be divided by a parallel line of light, throughout its whole length, into two parts, and resembled a double shadow, or the shadows of two hairs, but was by no means of the same degree of blackness as was the single shadow observed close to the hair. At still greater distances it increased in breadth and diminished in blackness, whilst the transverse dimensions of the dividing line of light increased at the same time, until, at a considerable distance from the hair, this intermediate band or line of light began to put on the appearance of colours on its edges, and to assume, on both sides externally, casts of yellowish and reddish light. By further increase of distance, this apparent shadow, these dark intervals, became more diluted, and of nearly the same colour throughout, the line of light more and more diffused, and was at last extinguished by the extreme diffusion and ultimate invisibility of the light that produced it.' P. 32.

The theory of these appearances, which may in a great degree be imitated by a candle, is not very difficult. In the first instance only, near to the hair, is the shadow truly such. At a greater distance the inflected rays enlighten a little the part darkened, by intercepting the direct light; and when these inflected rays meet in the centre, that part becomes so luminous as to give the appearance of real light by comparison with the darker margins. We have however found that this is only comparatively real light.

Close to the body, the shadow of the hair, and the whole of the passing light, apparently unchanged, may be observed. Further therefrom, that part of the light which constitutes the first fringe begins to put on a considerable brightness, and to be separated from the rest of the light by a dark line of shade. Further, and further still from the hair, the second and third fringes begin also to appear, together with the first, in the form of white streaks of light, separated by dark intervals. At greater distances, these breadths of white light begin to exhibit colours at their edges; and at distances yet greater than these, or by considerable obliquities of reception, they are decomposed and distributed into the three first fringes of the preceding observation S.

Adjacent to these three, in a state of great dilution, approaching

to evanescence, are the subsequent fringes of the same observation ; and adjacent to these last fringes, to some distance from them, the nearest light is divided into alternate streaks of white light and shade, or luminous and dark intervals, which, at the first and nearest distances being under the circumstances of this observation too small and too little separated for perception, and afterwards running into one another and mixing their colours, put on the appearance of white light, and can no longer be distinguished therefrom, up to the point where, by increased distance, the action of the hair ceases altogether on the passing light.' P. 34.

With respect to the production of these fringes, we have as yet made no advances. The language which attributes them to the decomposition of light, or rather to the separation of bundles of luminous rays, must be rejected, as too strictly mechanical. We see them closely connected with the attenuation of light. The breadths of these fringes, and their dark intervals, we have said, are not regular and constant. Our author supposes, that, in measuring these, Newton had the coloured rings, observed in other experiments, in view; but that he has not pursued the analogies, as the result would have militated against his principles for explaining the phænomena of light.

The author next proceeds to the observations on the fringes of colours produced by light passing over the edge of a knife. These are connected intimately with the former subject; but the particular explanation so closely depends on the diagrams, as to be scarcely intelligible without them. We shall add, however, the principle on which all these changes are produced, in our author's own words.

The white rays of light are bent in passing by the side, edge, or angle of any thin plate or body, into lines of direction, deviating from their original course towards the body.

The rays of light are not only so bent, they are also distributed or divided into different rays of different colours, in angles of dispersion, greater as the distances are less, and less as the distances are greater, in such a manner, that, of different colours at the same distance, the purples, blues, greens, yellows, and reds, are bent towards the body; the purples most, each of the others in due succession less, and the reds least, according to the order of their statement; and of colours of the same sorts at different distances, the nearer more than the more remote, and the more remote less than the nearer. So various however are the bendings of different colours at different distances, that in certain distinct portions of light, and at different distances of observation, the more remote and the nearer rays of different colours contained within each of these portions or divisions of the light become variously intermingled with each other, and, by their various intermixtures form each of these divisions into parti-coloured fringes; whilst the rays of different divisions never mixing with those of other divisions, the intervals of the divisions are preserved, and become the dark intervals which separate the fringes.

• Wherever the divergency of a beam of light is such, that by inflection a nearer red is less bent than all the other equidistant or nearer adjoining coloured rays, and more bent than all the other more remote but adjacent rays; and also a further purple is more bent than all the other equidistant or more remote coloured rays, and less bent than all the other nearer adjoining rays; there a dark interval will be produced, on the side of which, nearest to the body, the reds will appear; and on the side most remote, the purples and blues; and wherever two or more such intervals are produced, they include between them a fringe or fringes of various breadths and colours, depending upon the relations between the dispersions and bendings of the rays by the inflecting body and the original divergency of the beam, the purples and blues nearest, the reds furthest from the body.' P. 47.

The experiments of light passing through an aperture formed by the parallel edges of two knives afforded appearances very analogous to those observed by successively diminishing the aperture through which the beam was transmitted in the first experiments; but differing in some respects, as may easily be supposed, by the variation of circumstances and situation. The appearances are detailed with great minuteness, and elude every attempt to describe them in shorter words than those of the author. The experiments on the coloured fringes formed by the edges of two knives meeting at an angle are still more difficult to abridge. The appearances, we have reason to think, are described very correctly; but the author's account of them is so concise, and so closely compacted, that to those only who, like ourselves, have been engaged in similar observations, will it perhaps be intelligible.

Our author next notices the fringes produced in shadows of light, coloured in consequence of passing through a prism, and finds that the same circumstances and the same laws influence the coloured light, which affected the ray previous to its becoming coloured.

It is plain therefore from the circumstances of this, and of all the foregoing observations, that the plienomena of the inflections of light are the consequences of the various bendings, separations, and changes produced in each of the separate portions of a beam of white light passing between the parts and particles of a transparent medium, by the action or attraction of the edge, angle, or side of an approaching body; each of these portions being variously bent aud separated into other portions of differently-coloured lights, each of which being changed from and receiving a constitution or modifica tion different from that of all the others, and from that of the ori ginal portion out of which it was derived, retains that constitution or modification permanently, until it be again re-united into the same, or a similar portion of white light, among the properties of which constitution or modification it is one; that passing at the same distance with rays of other colours from and by the edge of an inCRIT. REV. Vol. 34. April, 1802.

2 H.