C C' (Fig. 2.) is a box of sheet iron about 30 inches in length and 6 or 8 in width and depth. Being filled with ignited charcoal, it is suspended nearly on a level with the eye, and by a visual ray, grazing the edges of the box, you look at a mark some distance off, as M for instance. Then you see a direct image of the mark in the direction P M, and an inverted image in the direction P M'. This second image is analagous to the inverted images of the mirage; it is obviously produced by reflection of the light from the strata of heated air which surround the box, and not by reflection from the sides themselves. It is not material to the success of the experiment, whether the visual ray approach the side or the top of the box.

Wollaston, moreover, imagined another experiment, by which mirage may be produced in a liquid. Take a small glass vial, either round or square; put in it two liquids of unequal density, one above the other, with great care, so that they may gradually combine where they meet. Water and sulphuric acid, water and alcohol, water and strong syrup, will do very well for the experiment. When the combination has been effected very evenly in a stratum of sufficient thickness, place the eye directly opposite to this stratum and look at a small object upon the opposite side, and you will see an erect and an inverted image of this object.

MIRAGE OBSERVED IN DIFFERENT PLACES AND UNDER VARIOUS

CIRCUMSTANCES.

A singular effect of mirage was observed at Ramsgate by Dr. Vince. In looking from Ramsgate, in the direction of Dover, if the weather is favorable, the tops of the four highest towers of Dover Castle are visible; the rest of the edifice is hidden by a hill at twelve miles' distance from the observer, half the space between being occupied by the surface of the sea. Dr. Vince, when at Ramsgate, near 70 feet above the surface of the sea, was much surprised, on the 6th of August, 1806, when looking towards Dover, at 7 o'clock in the evening, to see not only the four towers of the castle, as usual, but the entire building even to its base. "It was seen, says he, "as perfectly as if it had been suddenly transported to the Ramsgate side of the hill."

The same gentleman has since published other observations, taken on the same spot, and particularly some made with a good telescope upon ves

sels which approached or left Ramsgate. We shall quote the two following. One day he perceived a vessel exactly in the horizon; he distinguished it perfectly; but at the same time, he saw its image inverted, very regular, and disposed vertically above the object, so that the top of the real mast and that of the inverted image met in the same point.

Again, in the same month of August, and towards evening, he saw a variation of the phenomenon; the image of the vessel was still inverted, but below the object.

Captain Scoresby observed many similar phenomena in the Greenland seas. When the sun appears in these latitudes, the strata of air next the surface of the earth and sea rapidly attain a much higher temper ature than those some feet above, and the most varied and fantastic appearances are produced by the extraordinary refractions consequent upon this.

Messrs. Biot and Mathieu made similar observations at Dunkirk, on the sandy beach which stretches out at the foot of fort Risban. M. Biot has given the theory in detail, in the Memoirs of the Institute for 1809; he has shown that, beginning from a certain point T, at some distance in front of the observer O (Fig. 3), we may suppose a curve T C B, such that all the points below it remain invisible, while all the points above, up to a certain height, present two images, one ordinary and direct, the other extraordinary, beneath its stratum, and inverted. Thus a person leaving the point T and going from the observer, would present the successive appearances represented in figure 3.

Figure 3.

B

B

T

Messrs. Soret and Jurine observed on the lake of Geneva, in September, 1818, at ten o'clock in the morning, the remarkable phenomenon represented in figure 4. The curve A B C represents the east bank of the lake; a boat laden with casks, having all its sails unfurled, was at P, opposite the point of Belle-Rive, directing her course towards Geneva. It was seen by the observers with a telescope in the direction GP, from the shore of the lake, in the second story of Jurine's house, at about two leagues' distance. As the boat occupied in succession the positions Q, R, S, a lateral image was distinctly seen in Q, R', S', advancing with the boat, but seeming to pass off to the left of G P, while the boat moved to the right. When the sun shone on the sails, this image was perceptible to the naked eye.

The direction of the sun's rays at the moment of observation is indicated by L Y.

From the position of the places, it is obvious that this is a case of lateral mirage; the air to the right of G P had remained in the shade all the morning; to the left, on the contrary, it had been heated by the sun, and the surface of separation between the warm and the cold air must have been nearly vertical to a sinall extent above the water; on one side and the other of this stratum a mixture of different densities must have taken place, increasing from the left to the right; and thus was produced in vertical strata what is usually seen on the ground in horizontal strata.

These examples will suffice to give some idea of the varied and singular appearances which may result from those extraordinary refractions which light experiences in strata of air whose density changes suddenly. We have supposed these changes to take place in plain, regular strata ; but it is obvious that they must frequently, from inumerable causes, occur in curved irregular strata, when the images produced must be distorted every way, sometimes enlarged, sometimes indefinitely elongated, and sometimes dispersed as if the objects were broken into a thousand pieces. Doubtless, the phenomenon known under the name of Fata Morgana is a species of mirage. This has been observed at Naples, at Reggio, and on the seacoast of Sicily. Sometimes the people crowd to the seashore to enjoy this singular spectacle far off in the air are seen ruins, columns, castles, palaces, and a multitude of objects which seem to displace one another and to change their aspect continually. All this fairy work is but the representation of

terrestial objects, which, invisible in the usual state of the atmosphere, become apparent and movable, when the rays of light proceeding from them move in curved lines through strata of air of unequal density.

V. HALOS.

HALOS are bright and usually colored circles which are sometimes seen to surround the disk of the sun and of the moon. They are also called coronæ. The luminary is in the centre, and the space comprised between its edges and the interior of the luminous circles forms the area of the halo. This space is of a deeper grey than the sky about it, if the atmosphere is foggy, and of a deeper blue, if it is clear.

The diameter of halos has been measured frequently, at different times, and in different places, and always makes to the eye of the observer an angle comprehended between 45° and 46°. The halo round the moon is a white luminous circle without decided color, except when a pale red borders the interior of the circle. The colors of that which surrounds the sun are usually very distinct, though less vivid than those of the rainbow. The red is innermost, marking decidedly the area of the halo; the indigo and violet are outermost, always very indistinct, and gradually blending with the tints of the sky.

Under some circumstances, a second halo is seen, much larger than the first, but concentric with it; its diameter usually appears to be of 90°, or nearly that; its colors are very faint, and its whole light much less than that of the interior halo.

Descartes, Huygens, Mariotte, and many other philosophers have attempted an explanation of this phenomenon.

Descartes attributes it to rays transmitted through certain small stars, which are observed in snow, and which may become sufficiently transparent when the heat begins to melt them.. "These stars," he says, "are always swelled out towards the middle, and their usual convexity doubtless determines the diameter of 45° exhibited by halos." According to this hy、 pothesis, the exterior coronæ would be produced by rays traversing two rows of small convex stars.

Huygens imagines, that halos are produced by small transparent globules having an opaque nucleus, these globules being water or ice, and the opaque nucleus compressed snow, like that in fine hail. But, in order to give the constant diameter of 45°, there must always be a determinate ratio between the thickness of the opaque part and that of the transparent part. Mariotte finds a cause for halos, in the form of those small, transparent, prismatic needles of which snow is composed. This last hypothesis is the most probable, and we shall attempt its developement.

In congealing, water assumes very regular crystalline shapes, among which we often meet with those whose faces make angles of 60°, thus constituting prisms of ice whose refracting angle is 60°. These prisms are doubtless turned in the air all possible ways, and are thus receiving the solar rays under all possible inclinations. It is well known that, in certain positions of prisms, light experiences in passing through them a minimum deviation; this position is determined by the condition that the refracted ray makes an isosceles triangle with the two sides of the prism, or, what comes to the same thing, that the angle of refraction is equal to half the refracting angle. As the refracting angle is here 60°, the angle of refraction would be 30°, and, consequently, the angle of incidence about 41°. In this case, the deviation is equal to twice the angle of incidence, diminished by the refracting angle, which gives here 2x 41 — 60—82—60 = 220, nearly half the diameter of the halo.

Figure 5.

B

We may then conceive, that, the observer being placed at P (Fig. 5) when the direct rays arrive in the direction SP, all the small prisms of 60° floating in the higher regions of the atmosphere, and properly turned like the prism A C B, will refract towards the eye a small but very bright beam, because this will be composed of rays sensibly parallel, on account of the condition of a minimum, and the same phenomenon reproducing itself in a conical surface making an angle of 22° about the line S P drawn to the centre of the sun, the observer will see a corona of 44° in diameter.

The ratio of refraction in violet light being greater than that in red light, we shall have for this kind of rays a greater deviation, and consequently a somewhat larger corona.

Finally, the sun's diameter, being 30', will contribute to increase the breadth of the colored bands.