valuable performances, has charged himself with this task in a memoir which he has latterly communicated to us.

The subject is delicate and contestible. We shall give in succession the explanations of the phenomenon in question given by the two authors, announcing at the same time that we do no hesitate to adopt, at least in its fundamental parts, the opinion of Mr Trechsel.

1

M. Zschokke, at the beginning of his memoir, gives an account of the authors who have observed coloured shadows, and attempted to explain them. It will not be uninteresting to go over this ground. Shadows coloured in blue are those which have been most frequently remarked, because in fact nature presents them oftenest to us. Priestley, in his History of Optics, states that this phenomenon was for the first time observed and described, about the middle of the seventeenth century, by Otto Guerick, the celebrated inventor of the air-pump; but he is wrong, for Leonardo da Vinci speaks of it in his Treatise on Painting, written in the fifteenth century. This able artist sought to discover, with all the interest excited by a subject of so much importance to his art, to what was owing the colouring of shadows in blue. He only saw in it a reflection of the colour of the sky, or rather of the atmosphere, having recourse for this phenomenon to the same explanation as for the purple tints, which colour rocks and buildings, before the rising and after the setting of the sun, or for the greenish reflection which diffuses itself upon the sides of a vessel, or upon the piles of a bridge above a deep and limpid body of water. Bouguer, in his Optics (1729), Buffon, in his Memoirs of the Academie des Sciences for 1743; Begnelin in those of the Berlin Academy for 1767; Monge in 1789, and other natural philosophers, have more or less adopted the opinion of the celebrated painter.

Buffon had the merit of contributing powerfully to direct the attention of observers toward the coloured shadows that form in the solar light. "I observed," says he, "during the summer of the year 1743, more than thirty sun risings and as many set tings. All the shadows that fell upon white, as upon a white wall, were sometimes green, but most commonly blue, and of a blue, as lively as the most beautiful azure. I shewed the phenomenon to several persons, who were as much surprised as my OCTOBER-DECEMBER 1826.

C

2

self. Difference of season has no effect upon it, for there are not eight hours (15th November 1743) since I have seen blue shadows; and whoever will give himself the trouble of looking to the shadow of his fingers at sun-rise and sun-set, upon a bit of white paper, will see like me this blue shadow," &c. The illustrious naturalist also cites a letter of the Abbe Millot, in which he announces to him that at noon, with a cloudy sky, in which some openings were seen here and there in the clouds, he had observed shadows of a beautiful blue upon white paper; and further, that, under certain particular circumstances, he had remarked green, violet, or yellowish shadows, or shadows surrounded with a coloured margin of these different tints. Buffon, recapitulating these various observations made in 1743, adds in 1773: "This blue colour of shadows is nothing else than the colour of the air itself *."

M. de Schrank, in the Memoirs of the Academy of Munich for 1812, brought forward again the opinion proposed in 1783 by Opoix, a French naturalist little known, supporting it by new arguments. The blue shadows, according to him, come from the inflection of the rays' tangent to the edges of the solid, from which the shadow proceeds. As the blue rays are very refrangible, they are more strongly attracted than the others by bodies, and thus come to colour the interior of their shadows. Opoix, as well as M. de Schrank, knew well that the violet rays are more refrangible than the blue rays; and they reply to the objection which arises from this circumstance, the one that, in the shadows of thin bodies, the violet rays are sufficiently deflected to pass beyond the opposite edge of the shadow, and enter into the open light; the other, that, in the case where the body has a sufficient breadth to prevent the application of such an explanation, the rays fall, it is true, into the interior of the shadow, but that the tint which they carry there is too obscure to be perceived.

1

Rumford observed not only the coloured shadows formed in the pure solar light, but also the various shadows resulting from several sorts of coloured lights combined with each other and with the solar light; and thinking that he had remarked that,

Buffon, Hist. Nat. d'Min., Memoire viii.

when seen through a tube, which excluded all comparison of one shadow with another, all these shadows appeared black, he concluded from thence that all these effects are mere optical decep

tions *.

M. de Grotthuss arrived at nearly the same conclusion, but by a different process †. He knew the phenomenon of the blue and yellow shadows, which are produced by the concurrence of the light of a candle and that of the twilight; he also knew the impression which the long continued observation of coloured plates produces upon the retina; an impression which af terwards reproduces in the organ spots tinged with colours exactly complementary, in the scale of the spectrum, to those on which the eye has been fixed; and he in like manner considers the phenomenon of coloured shadows as a physiological deception, as the result of the fatigue caused by an effort of the organ in the same direction, and of the disturbance of an equilibrium of sensibility in it.

After giving this historical narration, Mr Zschokke remarks, that none of the hypotheses explains all the cases in which shadows appear coloured, and he proceeds to the exposition of a new theory. Let us first give an account of the fundamental phenomena, the causes of which form the subject of inquiry.

Coloured shadows are produced in the solar light, when it is refracted by the vapours of the lower strata of the atmosphere, or reflected by the clouds. Thus, 1st, the colouring is perceptible chiefly at sunrise and sunset, when the sun is not higher than from ten to twenty degrees above the horizon. 2d, In winter, the shadows are sometimes coloured at noon, because at that season in our latitude the sun scarcely rises to the height of twenty degrees. In summer, they are only coloured in full day when the sky is overcast, and the clouds reflect a strongly coloured light. 3d, The more deeply the rays penetrate into the lower strata, the more strongly are the shadows coloured,

* See Philosophical Transactions 1794; or in the Biblioth. Britann. vol. i. p. 339, an extract of Count Rumford's paper, terminated by a note (p. 372) upon coloured shadows, and the authors who have treated of them.

+ See in Schweigger's Beytrag. zur Chemie und Physik, vol. iii. p. 148. an abridgment of M. de Grotthus's Paper on the Accidental Colours of Shadows, and on Newton's Theory of Colours.

and the farther may the opaque body which projects them, the hand, for instance, be removed from the white surface which receives them. The distance, therefore, cannot be assigned at which the opaque body should be from the surface. According to the greater or less intensity of the light, this distance may vary from five or eight feet to as many lines. At the moment of twilight, or in very dark days, the end of the finger from which the shadow projects, requires to be held at the most two or three lines from the white surface. 4th, The same opaque body projects shadows variously coloured, according as the surrounding surfaces, such as the walls of the chamber, or the clouds, if it be in the open air, reflect one colour or another.

Coloured shadows also form in a light, coloured by refraction or by reflection. This colouring of the shadows, however, does not take place, if the light so modified penetrates into a chamber otherwise perfectly dark, for in that case the shadows are black. The more intensely the light is coloured, the more distinct is the tint of the colour.

Lastly, the artificial light of a candle, combined with that of the sun, gives rise to coloured shadows. Thus, according to Rumford's experiments, if, in the day-time, the shutter of a dark room be opened about half an inch, and there be placed upon a table a lighted candle (situated in such a manner, that its rays falling upon a piece of white paper, which is presented to it, as well as to the opening of the shutter, make with those coming from this opening an angle of about forty degrees), and the finger be then held at the distance of two or three inches before the paper, this this opaque body will project two shadows, of which that proceeding from the day-light will be yellow, and that from the candle-light of a very beautiful blue. In proportion as the finger is carried nearer the candle, the blue will become deeper, and the yellow fainter, and the contrary will take place if it be removed from the light.

Such being the facts to be accounted for, Mr Z. proposes to establish, a priori, that the shadows produced by the interception of a coloured light must also be coloured." It is known," says he, "that, in the solar spectrum, the white light of the sun is decomposed into coloured rays; on the other hand, the shadow produced by the interception of white and undecomposed

light, is black; if only one of the coloured rays of which it is composed is intercepted, the part cannot produce the same effect as the whole; the coloured ray cannot therefore project a black shadow; this shadow must itself be coloured."

Now, what will be the colour of the shadow projected by a ray of a given colour? To find this, M. Zschokke made the solar rays pass through disks of glass variously coloured; and receiving the light by this procedure upon a white surface, he presented before this surface an opaque body, in order to form shadows with it. He took care to make the experiment when the sun was at a great height upon the horizon, to prevent any natural colouring of the shadows mingling with that which he produced artificially. He then found, that, in the

and that thus to each colour of the ray there corresponds, in the shadow which it projects, a colour which would itself project a shadow of the same tint as the ray.

Such is Mr Zschokke's theory in brief; we regret that we cannot follow him in developments from which his style, always animated and descriptive, takes away the dryness of a scientific dissertation.

66

"The hypothesis of Mr Zschokke," says Mr Trechsel, recommends itself at first sight by its precision, and, if one may so speak, by its paradoxical nature. One fancies he sees in it the great law of polarity, which appears to manifest itself in almost all the branches of natural philosophy. Besides, the most important discoveries have been in fact but gleams of light emitted by geniuses superior to their age, hypotheses imagined a priori, which have been recognised as true by observations and researches made afterwards." These considerations, which ought to recommend the hypothesis in question to the attention of natural philosophers, have engaged Mr Trechsel the younger, to repeat with his father the experiments of Mr Zschokke, and to add