The Chinese Magic Mirror.

TO THE COMMITTEE ON PUBLICATIONS OF THE JOURNAL OF THE FRANKLIN INSTITUTE.

I notice, in your last number, an attempt to solve the mysterious phenomenon of the Chinese magic mirror: as that explanation seems to me erroneous, I am induced to offer another solution, which, if I am correct, can easily be verified by some of your ingenious readers. My attention was called to this curious piece of Chinese ingenuity by the intelligent proprietor of the Chinese Museum, in Boston, some time since. I examined it with attention, and found that the mirror consisted of a thin, flat plate of copper, with Chinese characters and figures on the back, projecting nearly the eighth of an inch from its surface; the face of the plate was silvered, and highly polished, and, on the closest inspection, no difference appeared on any part of the surface; the plate seemed to be of one casting, and the silver plating put on in the ordinary way, and equally polished. If this mirror was placed in the sun-light, so as to cast its reflexion on the wall, the characters that were raised on the back of the plate appeared in a brighter reflexion than the other parts of the mirror, which would seem to prove that those parts could not be made of a coarser metal. I therefore came to the conclusion that, in consequence of those parts over the characters being rigid, and the other parts elastic, a slight inequality of polish was produced on the different parts of the surface; so small, indeed, as not to be detected in any way but by testing its powers of reflexion.

Washington City, D. C., Nov. 30, 1847.

On the Theory of Dew.

J. J. GREENOUGH.

We find, in the "Annales de Chimie et de Physique," for October, 1847, two letters from M. Melloni to M. Arago, containing an account of some interesting experiments made by the former upon the production of dew, and the opinions and conclusions resulting from his researches relative to the subject. Not being able to afford space for the whole, we translate such portions as appear to us sufficient to explain the views of the author.

After the experiments of Wells, he thinks it might have been surely admitted, that dew neither rises from the earth nor falls from the sky, but is formed from an invisible and elastic fluid diffused throughout the space which surrounds bodies, and that the precipitation of this aqueous vapor should be attributed to the cold resulting from the calorific radiation of bodies towards the clear sky. According to this view, the leaves of vegetables, wood, glass, varnish, lampblack, &c., will be covered with dew, because they easily part with their heat and are considerably cooled when exposed to the sky: while the metals will remain dry, on account of their feeble power of radiation towards the upper regions of the atmosphere. If a polished metallic

vessel, filled with boiling water, be presented to a thermoscopic apparatus-and afterwards one exactly similar, with its sides covered with varnish, or lampblack-the radiation of heat indicated in the latter case will be much greater than in the former. But however clear may be the deduction, it would seem that it has not been so considered by all. Indeed, Benedict Prevost, and Saussure before him, attributed the absence of dew upon metals to an electric influence; Leshe explained this phenomenon by a particular repulsion exercised by metallic surfaces upon aqueous vapor; and the supporters of the theory founded upon the supposed rising of dew, account for it by the heat and electricity disengaged in the chemical action of metals upon the molecules of this same vapor, at the moment of their passage into a liquid state.

In order to show that these different hypotheses could not be sustained, M. Melloni took three thermometers, graduated upon the stem, and passed each through a cork, which remained fixed at a distance of about a quarter of an inch from the bulb. This cork served as a support to the two parts of the metallic casing with which the thermometers were enveloped; the first being a little vessel of silver, or copper, very thin, similar to a thimble, with its surface smoothed and polished, of sufficient size to contain the bulb of the instrument; the other, a cylinder of tin, closed at one end and open at the other, used to cover the graduated tube of the thermometer. These two metallic pieces, which could be very easily removed and replaced, were readily kept in their proper positions by the friction and elasticity of the cork. Three tin vessels were then prepared, supported by slender metallic tubes, and having lids, or covers, of the same material; an opening being made in the side of each, near the bottom, into which were introduced the covered bulbs of the three thermometers, leaving the tubes and their envelopes horizontally disposed on the outside. The casing of one of the thermometers was blackened, and the others left in their bright state, with a view of comparing the nocturnal radiation of silver with that of lampblack. The whole was exposed to the free air in a calm and still night, the tin vessels being sometimes open and sometimes closed by their covers. When closed, the three thermometers indicated the same temperature; but on leaving one of the vessels containing one of the bright-cased thermometers closed, and opening the two others, the falling of the mercury in the bright-cased thermometer, thus opened to the sky, was so slow that it required a very delicate instrument and very exact comparisons to observe and measure its movement; while the blackened thermometer fell so rapidly as to be readily seen, and in a few minutes it marked three or four degrees less than the one in the closed vessel-thus proving, evidently, that this difference was due to the radiation, towards the sky, of caloric from the blackened instrument, and by no means to the contact of the exterior air, which acted equally upon it and upon the polished case of the other uncovered thermometer.

Thermoscopic apparatus, similar to those just described, having the cases covered with varnish, plumbago, fish glue, saw dust, sand, earth, and the leaves of plants, have constantly indicated a very sensible di

minution of the temperature before being covered by dew. The interval has been sometimes for several hours; and there is often, also, an abatement of the temperature without the precipitation of vapor at any period of the night. The latter phenomenon occurs the more frequently as the thermometers are placed at a greater distance from the ground. Operating at a certain elevation, the deposit of dew upon bodies might, therefore, be suppressed, or retarded, and the fact be incontestably established that the formation of dew always follows, and never precedes, the production of cold. M. Melloni remarks that he has never seen the polished metallic cases of his thermometers covered by a condensation of aqueous vapor, in very damp nights, when there was no appearance of fog, or mist, in the atmosphere.

From all this, he concludes that dew, properly so called, always requires a certain cooling of the bodies upon which it is found; and that metals, exposed to a clear sky, are not covered with dew, because of their possessing the quality of being thus cooled but in a very slight degree. In order to determine whether other causes existed to prevent the accumulation of dew upon metals, and whether their feeble radiation is really the only and true cause of their being never covered with dew, he undertook the following experiment, the result of which he considers conclusive, and that it proves, at the same time, the falsity of the hypotheses of the rising and the falling of dew, as well as the truth of the principle asserted by Wells.

Upon a disk composed of a single plate of tin, as large and as thin as could be obtained, he drew a concentric circle with a radius equal to one-third of that of the disk, and covered it with a thick coat of varnish. From another plate of tin, he then cut a second disk, with a diameter about half an inch less than that of the varnished circle, and soldered to its centre, perpendicularly to the surface, the end of a piece of iron wire, 08 of an inch in thickness, and eight or ten inches in length, pointed at the other end, which was inserted into a small hole made in the centre of the larger disk. Being inserted on the varnished side of the larger disk, the wire was pushed through until the distance between the two disks was reduced to about two-tenths of an inch, and then fastened in that position by a few drops of solder. The two plates, thus united, being taken, in a calm and clear evening, into an open field, and left in a horizontal position for a short time, removed from contact with other bodies, showed on the surface of the larger disk, phenomena of dew which may be readily anticipated.

The smaller disk being uppermost, and its radius being a quarter of an inch less than that of the varnished circle upon the larger disk below, it follows that an annular band of that circle, a quarter of an inch in breadth, extended around the vertical projection of the roof formed above it by the little metallic disk. Now it is apparent that this band would radiate towards the sky, lose a portion of its temperature, become covered with dew, and communicate, little by little, both cold and dew, consecutively, towards the centre and also towards the circumference. In the latter case, however, this propagation would be carried much further than in the former; for the portion cooled by communication would be further cooled by radiation when covered

with dew-while the varnished portion, under the smaller disk, would be scarcely cooled except by contact. In effect, the central part of the varnished circle remained always dry, and the metallic zone which surrounded it became wet, even to the extremity, if the atmosphere was very moist.

But what could not, perhaps, have been anticipated at first, was the exact reproduction of the same appearances upon that side of the larger disk which was towards the ground. The dew began to show itself upon this side at points opposite to the little exterior ring of varnish, and a slight whitish circle was perceived, which, appearing all at once upon the blank field of polished metal, seemed like the formation of a Daguerrian image. This circle, becoming stronger and spreading by degrees, sometimes extended to the edge of the plate, but never reached the central part, which always remained dry and maintained its metallic brilliancy, like the corresponding portion on the other side of the disk, and the little circular roof, which covered without touching it.

This simple and easy experiment may be regarded as furnishing conclusive evidence upon several points connected with the theory of dew. It shows that dew does not fall from the sky-for the upper disk always remained dry, while the greater part of the lower one is moist. Neither does dew rise from the ground-for if a portion of the under side of the lower plate is covered with dew, there is always a space around the centre which remains dry and shining. The theory that metals repel the aqueous vapor which forms dew, or that they cause its evaporation as rapidly as it may be precipitated upon them, is disproved by the fact that we here see some parts of the metallic surfaces quite wet, while other parts remain perfectly dry.

The first appearance of dew upon the uncovered band of varnish, and its successive propagation to adjacent and opposite portions of the larger disk, associated with the increase of temperature developed by the varnished cases of thermometers exposed to the open air, prove, in short, that dew is a mere consequent of nocturnal radiation, which produces, in bodies endued with great emissive power, the degree of cold necessary for the condensation of the elastic aqueous vapor diffused in the atmosphere.

In his second letter, M. Melloni proceeds to show that, if the principle of Wells concerning the origin of dew were established beyond doubt, it would still be impossible to account for all the thermometric and hygrometric phenomena which are produced during a clear and still night, without taking into consideration a new circumstance, until now entirely neglected, notwithstanding its exceeding importance in the nocturnal cooling of bodies. But before proceeding to give an account of this, M. Melloni makes some observations upon two series of experiments which the supporters of the theory that dew rises from the earth have put forward with the pretension that these alone are sufficient to overturn all theories of dew founded upon the principle of calorific radiation.

They took a certain number of thermometers, as nearly equal as possible in volume and in sensibility; some of which were coated with

lampblack, varnish, Indian ink; others gilded, silvered, covered with leaves of plants, tin, copper, &c. These instruments, thus prepared, being exposed in the fields during a clear and still night, at first indicated temperatures with some little difference; but after a certain space of time they were found to agree very nearly. The experiment was then varied in the following manner. Cylinders of glass were planted in the ground, on the tops of which were placed plates of zinc, copper, and glass, in each of which was made an opening to receive the bulb of a thermometer, whose tube, supported by an iron wire, was elevated above the upper surface. A thermometer was freely suspended between the plates, for the purpose of showing the temperature of the air. In this case, as in the other, the apparatus being exposed early at night, indicated differences of temperature which afterwards disappeared, insomuch that by daybreak next morning all the thermometers were found at nearly an equal height.

The opponents of the principle established by Wells, regarded these results as decisive, and they consequently stigmatized the principle that the nocturnal cold of bodies is indispensable in the formation of dew, as a mere chimerical illusion. M. Melloni, on the contrary, thinks that if there is anything imaginary, or fantastic, in the matter, it is in the reasoning which led to such a conclusion; inasmuch as the experiments of these gentlemen were made near the ground, in an atmosphere charged with moisture-all the tubes of the thermometers being uncovered-and in the last experiment the bulbs communicated, by means of the plates, with the cylinders which served to support the apparatus. Now the glass of which these tubes and cylinders were composed radiated considerably, its temperature was reduced, and the cold acquired was communicated to the bodies which touched it; these, being surrounded by a very moist atmosphere, then condensed the aqueous vapor; and we know that water radiates and cools itself with as much energy as glass, varnish, and lampblack. It is not, therefore, surprising that the thermometers in contact with the sheets, or plates, of inetal, should indicate, after some time, the same temperature as the thermometers surrounded by substances which radiate most; and because the metallic surfaces, which were found to be covered with dew, were as cold as the glassy, or the blackened, surfaces, it might well follow that water, glass, and lampblack, are substances of nearly equal emissive power. But without a resort to poetic fiction, it would never be logically inferred, from this experiinent, that metals are cooled, during calm and clear nights, as much. as lampblack and glass.

In order, therefore, to understand the true condition of things, glass should not be used for the supports, but slender tubes of tin, which have scarcely any radiation, and maintain the thermometers suthiciently isolated from the warmth of the ground; it is necessary, moreover, to enclose every part of the thermometer with metallic laminæ. Then, these laminæ being polished, the thermometer will give a temperature very nearly approximate to that of the air, and when the metallic enclosure is varnished, blackened, or in contact with vegetable leaves, or any other substance, we can ascertain, by a simple com