let, and finally a beautiful blue of the second order. Hence, as M. Biot remarks, we observe that the extinction to which the intensity is accidentally subject, does not prevent the tints from following the same order as that of the rings.

A great number of the metallic oxides exhibit a momentary change of tint by being heated, and resume their primitive tint by cooling. This arises from the increase of size in the particles, and consequently the new colours thus developed should rise in the order of colours. A phenomenon of the opposite kind was observed by M. Chevreul in volatilizing indigo spread upon paper. During vaporization the indigo colour passes into a poppy red, highly brilliant, which seems to prove that the particles have become less in the act of evaporation. The same eminent chemist noticed an analogous fact in the new substance, which he calls hæmatine. This substance, when pure and solid, has a greyish tint. When dissolved in water, containing some drops of acetic acid, it produces a fluid, whose colour is a slightly greenish yellow of the second order. If the fluid in this state is introduced into a tube filled with mercury, and heated by surrounding it with a hot iron, it becomes successively yellow, brilliant orange, brilliant red, purple, and bluish purple; and, what is very remarkable, if it is afterwards left to cool, it returns gradually to its primitive tint, which it requires some days to do, if the quantity used is about the onethird of a cubic inch.

The progressive steps by which bodies attain their definite tints, are well seen in the crystallization of a saturated solution of super-oxygenated muriate of potash during its slow cooling. As the temperature falls, the salt is precipitated in thin rectangular scales which unite to one another, and whose thinness is such, that they are differently coloured, according to the obliquity of the incident light, or the thickness of the scales. The thickest are of an uniform white colour, and the thinnest, by uniting themselves to others, become white in their turn. Sometimes they do not apply themselves exactly to one another, and then they do not cease to reflect the tints which they exhibit individually, even though they form part of a plate too thick to produce these colours. Similar variations are seen in the small

The tints described by M. Gay Lussac in the Ann. de Chimie, follow the order of the rings.

scales of acidulous tartrite of potash precipitated from a warm and saturated solution of this salt.*

5. The vegetable kingdom presents many curious illustrations of Newton's theory, as he himself observed, and as we have noticed in Prop. 7. M. Biot is of opinion, that the colours descend+ in the order of rings, as the force of vegetation developes itself, and ascend during its decay. The young buds of the oak and of the poplar, for example, are at first of a red colour, bordering on orange; from this they pass to a reddish orange, and soon to a green, through a kind of reddish yellow, extremely fugitive. When the flower of the honeysuckle blows, its colour is a pure white of the first order, and in decaying it passes into pale yellow, yellow, orange, and deep orange. The flower of the geranium sanguineum, whose colour is a violet red, intermediate between the first and second order, becomes blue in withering. Pinks of a bright red of the second order pass as they decay into a poppy red, and a violet purple. The same thing happens to certain species of roses, but there are others whose colour appears to be red of the third order. While these grow old upon their stalk, they lose by degrees the brilliancy of their red, and the blue and violet of the fourth order, acquiring a greater influence over their tints, they rise to a bluish red. The tigridia, which blows and withers in a few hours, appears, even when it is not quite open, of a bright reddish orange, from which it rises to a deep red of the first order, and in withering it rises to the violet red of the second order. The cobaa when it opens is at first of a pale and imperfect yellowish green of the second order; but it is soon spotted with violet, and in a few hours it becomes wholly violet, without passing through the intermediate blue. In withering, however, it descends from violet to blue. M. Decandolle ascribes the sudden change of colour at the first period to the fecundation, which he considers as the cause which modifies rapidly the colour of a great many flowers.

See Biot's Traité de Physique, tom. iv. p. 135. In quoting the opinions of this eminent philoso pher, it is necessary to state, that when he uses the word ascend in the order of rings, we use descend, because the colours fall from a higher to a lower order. M. Biot's term ascend, indicates a local ascent in the printed table, the first order being at the top of the table, and the last order at the bottom, Biot's Traité de Physique, tom. iv. p. 133,

6. The animal kingdom also contributes its aid in support of the same theory. The choroid coat of the dog and other animals, which produces the blue, green, and red reflexions from the eye of the living animal, retains the same faculty after death. When the choroid coat dries, it becomes black, and the colours disappear. We have found, however, that after remaining dry for nearly ten years, their colours could still be developed by moisture. The black passed instantly into a brilliant blue, the blue into green, and the green into greenish yellow.

5. Colours of the atmosphere.-As the earth is surrounded with an atmosphere varying in density from the surface of the globe, where it is a maximum, to the height of about 45 miles, where it is extremely rare, and just able to reflect the rays of the setting sun, the rays of the sun, moon, and stars are refracted into curve lines, unless when they are incident upon it perpendicularly. Hence the apparent altitude of the celestial bodies is always greater than their real altitude, and they appear above the horizon when they are actually below it.

But while the solar rays traverse the earth's atmosphere, they suffer another change from the resisting medium which they encounter. When the sun, or any of the heavenly bodies, are considerably elevated above the horizon, their light is transmitted to the earth without any perceptible change; but when these bodies are near the horizon, their light must pass through a long tract of air, and is considerably modified before it reaches the eye of the observer. The momentum of the red, or greatest refrangible rays, being greater than the momentum of the violet, or least refrangible rays, the former will force their way through the resisting medium, while the latter will be either reflected or absorbed. A white beam of light, therefore, will be deprived of a portion of its blue rays by its horizontal passage through the atmosphere, and the resulting colour will be either orange or red, according to the quantity of the least refrangible rays that have been stopt in their course. Hence the rich and brilliant hue with which nature is gilded by the setting sun; hence the glowing red which tinges the morning and evening clouds; and hence the sober purple of twilight which they assume when their ruddy glare is tempered by the reflected azure of the sky.

We have already seen that the red rays penetrate through the atmosphere, while the blue rays, less able to surmount the resistance which they meet, are reflected or absorbed in their passage. It is to this cause that we must ascribe the blue colour of the sky, and the bright azure which tinges the mountains of the distant landscape.

As we ascend in the atmosphere, the deepness of the blue tinge gradually dies away; and to the aeronaut who has soared above the denser strata, or to the traveller who has ascended the Alps or the Andes, the sky appears of a deep black, while the blue rays find a ready passage through the attenuated strata of the atmosphere. It is owing to the same cause, that the diver at the bottom of the sea is surrounded with the red light which has pierced through the superincumbent fluid, and that the blue rays are reflected from the surface of the ocean. Were it not for the reflecting power of the air, and of the clouds which float in the lower regions of the atmosphere, we should be involved in total darkness by the setting of the sun, and all the objects around us would suffer a total eclipse by every cloud that passed over his disk. It is to the multiplied reflections which the light of the sun suffers in the atmosphere, that we are indebted for the light of day, when the earth is enveloped with impenetrable clouds.

From the same cause arises the sober hue of the morning and evening twilight which increases as we recede from the equator, till it blesses with perpetual day the inhabitants of the polar regions.

The cause which we have assigned for the blue light of the sky, and which was, we believe, first given by Bouguer, though a very probable one, still required the evidence of demonstration. In examining this light, Dr. Brewster found that a great portion of it was polarized; and hence it follows, that it has suffered reflexion. M. Saussure found that the intensity of the blue colour increased with the height of the observer above the sea; and it has been observed by others, that the intensity diminishes as the quantity of aqueous vapour is increased. In order to measure this intensity, M. Saussure contrived an instrument called a Cyanometer.* A circular band of thick paper or pasteboard is divided into 51 parts, each of

• From two Greek words signifying a measure and blueness.

F

which is painted with a different shade of blue, decreasing gradually from the deepest blue, formed by a mixture of black, to the lightest, formed by a mixture of white. This coloured zone is held in the hand of the observer, who notices the particular tint which corresponds to the colour of the sky. The number of this tint, reckoned from the greatest, is the intensity at the time of observation. Saussure, Humboldt, Depons, and other travellers, have made observations with this instrument. The following are some of their results:

6. Coloured Shadows.-The shadows of bodies placed only in one light, and at a distance from all other bodies capable of reflecting light, must necessarily be black. In a summer morning or evening, however, the shadows of bodies formed either by the light of the sun, or by that of a candle, have been observed to be blue; this obviously arises from the shadows being illuminated with the light of the blue sky. The colours thus produced vary in different countries, and at different seasons of the year, from a pale blue to a violet black; and when there are yellow vapours in the horizon, or yellow light reflected from the lower part of the sky, either at sunrise or sunset, the shadows have a tinge of green arising from the union of these accidental rays with the blue tint of the shadow.

If the light of the sun, or of the candle, be faint, then the shadow of the body formed by the light of the sky will be visible also, and the two shades will be the one blue and the other a pale yellow, two colours which are complementary to each other. This fact has been ascribed to the circumstance of the light of the candle, and that of the rising and setting sun, being of a yellowish tinge; but though this will increase the effect it is not the main cause of it, as one of the shadows would be yellow, even if the light of the sun and the candle had been perfectly white.

The phenomena of coloured shadows are sometimes finely seen in the interior of a room; the source of one of the colours being sometimes the blue sky, and the other the green window blinds, the painted walls, or the coloured furniture. The best method of observing and studying this class of phenomena is to

use two candles, and to hold before one of them a piece of coloured glass, taking care to remove to a greater distance the candle before which the coloured glass is not placed, in order to equalize the darkness of the two shadows. If we use a piece of green glass, one of the shadows will be green, and the other a fine red; if we use blue glass, one of the shadows will be blue, and the other a pale yellow, and so on; the one colour being always complementary to the other, as explained in page 46. The light from the candle with the green glass obviously illuminates the shadow formed by the other candle, and hence it is easy to understand why that shadow is green; but as the other shadow is illuminated only by the common light of the candle which is not red, it appears difficult to discover the origin of the red light. The explanation of this must be sought not among optical, but among physiological principles. We have already seen, when treating of accidental colours, that when a portion of the retina was strongly impressed with any one colour, such as red, that same portion tinged green the images of white objects that fell upon it. In like manner, when nearly the whole retina is impressed with any one colour, such as red, a portion of it not impressed with that colour will tinge white objects green, or, to speak more generally, every excitation of the retina by one colour is accompanied by an excitation of its accidental colour, just as in Acoustics every fundamental sound is actually accompanied by its harmonic sound. Hence, when we see red we at the same time see green, but its impression is less forcible, and the tendency of this double vision of colours is to weaken the original impression, viz. the red; because the union of complementary colours produces whiteness. This may be proved by looking for a considerable time at a red wafer, which will appear less and less red the longer we view it; because the green which the retina is seeing at the same time, produces a whiteness which dilutes the red. This we conceive to be the true theory of accidental colours. Its application to coloured shadows is very obvious: When the eye is impressed with the green colour of the light transmitted through the green glass, it at the same time sees red, which, of course, appears only on the shadow upon which a green light falls.

7. Converging and diverging beams. -When the sun is descending in the west, through masses of open clouds, the divergency of his beams, rendered visible by their passage through numerous openings, forms frequently a very beautiful phenomenon. It is sometimes accompanied, however, with one of an opposite kind, viz. the convergency of beams to a point in the eastern horizon opposite to the sun, and as far beneath the horizon as the sun is above it, as if another sun, throwing out divergent beams, were about to rise in the east. This phenomenon is rarely seen in perfection. Dr. Smith, who observes that he once saw this phenomenon on Lincoln heath, describes it as an apparent convergence of long whitish beams towards a point diametrically opposite to the sun, and, as nearly as he could estimate, as much below the horizon as the sun was then elevated above the opposite point of it.'

On the 9th of October, 1824, we had the satisfaction of seeing this curious appearance in unusual splendour. The sun was considerably elevated, and was throwing out his diverging beams in great beauty through the interstices of the broken masses of clouds which floated in the west. The eastern portion of the horizon, where the converging lines were seen, was occupied with a black cloud, which seems necessary as a ground for rendering visible, by its contrast, such feeble radiations. The converging beams were very much fainter than the diverging ones, and their point of convergence was as far below the horizon as the sun was above it. About ten minutes after the phenomenon was first seen, the convergent lines were black, or very dark. This arose from the real beams having become broad, and of unequal intensity, so that the eye took up, as it were, the spaces between the beams more readily

than the beams themselves.*

In order to explain this phenomenon, which is a case of perspective, let us

This disposition of the eye is a very curious one, and has, we believe, never been observed. When we look steadily at a carpet having figures of one colour, green for example, upon a ground of another colour, suppose red, we shall, sometimes, see the whole of the green pattern, as if the red one were oblitethe red pattern, as if the green one were obliterated. The former effect takes place when the eye is stea dily fixed on the green part, and the latter, when it is steadily fixed on the red portion. It is easy to conceive that when the retina is in a state of irritation or excitation with red light, it will more easily take up, as it were, the vision of a red object than of ar.y other.

rated; and, at other times, we shall see the whole of

suppose a line to join the eye of the observer and the sun. Let beams issue from the sun in all possible directions, and let us suppose that planes pass through these beams, and through the line joining the eye of the observer and the sun, which will be their common intersection, like the axis of an orange, or the axis of the earth, through which there pass all the septa of the former, and all the planes passing through the meridians of the latter. An eye, therefore, situated in this line, or common intersection of all the planes, will, when looking at a concave sky, apparently spherical, see them diverging from the sun on one side, and converging towards the opposite point, just as an eye in the axis of a large globe would perceive all the planes passing through the meridians diverging on one side, and converging on another.*

FROM the phenomena described in the preceding chapters, the reader must have observed, that when light falls upon the most transparent bodies, such as water, glass, &c. a certain portion of it is reflected from their surfaces. When we measure the quantity reflected and the quantity transmitted, we invariably find that the sum of these quantities is less than the light which falls upon the body. Hence it follows, and the fact is a very important one to remember, that light is always lost in passing through is lost in two ways; a portion of it is abthe most transparent bodies. This light heat, and another portion is scattered sorbed or stopped by the body and forms in all directions by irregular reflexion. When light falls on metallic bodies, such as polished silver, or speculum metal, about one half of it is reflected, and the other half lost. The part lost consists, as in the former case, of two portions; one of which, and by far the largest, being absorbed, and the other scattered by irregular reflexion.

No complete set of experiments has yet been made from which the laws of

+ See Smith's Optics, vol. ii. Remarks, p. 57, 56 and Edinburgh Journal of Science, No. iii. p. 136