570 Darkness of Shadows. Mr. Crookes has lately applied his radiometer to photometric purposes* (Art. 563). The operation here depends on the deflection of a pith-bar under exposure to light. The bar is suspended in a vacuum globe, and the degrees of deflection produced by light admit of accurate measurement. A candle on each side of the apparatus keeps the index ray of light at zero. By balancing a standard candle on one side against any source of light on the other, the value of the latter, in terms of a candle, is readily shown. 800. The real darkness of a shadow, depends much on the extent and nature of the light-reflecting surfaces around it. Thus, shadows are less deep opposite to any white surface, than in other situations. An invalid lying in bed with back to the light may be saved the trouble of turning to show the countenance, by a sheet of white paper held up to act as a mirror. The reason why the moon when eclipsed—that is, as will be afterwards explained, when passing behind the earth, through the shadow cast by the earth in a direction away from the sun-becomes almost, if not quite invisible, is, that there are no other moons or bodies near our moon to reflect their light towards it. And the reason why moonless nights on earth are darker than the shadows behind a house or rock in the sunshine of day, is merely that there are then no other great bodies near the earth to reflect light into its shadow, as there are other houses and rocks near to illuminate the day-shadow of these. The moon is the only light-reflecting body of magnitude which the earth has near it; and we perceive how much less dark the earth's nightshadow is, when the moon is in a position to bear upon it. Many persons have doubted whether the light of the moon could be, as astronomy teaches, altogether reflected light of the sun; the moon appearing to them so much more luminous than any opaque body on earth merely exposed to sunshine. They judged the socalled "lamp of night" to be as naturally self-luminous as "the lamp of day." Their error arises from contrasting the appearance of the moon while receiving direct sunshine, with that of objects on the surface of this earth, then in the darkness of the earth's shadow, which we call night. The moon when above our horizon in the day-time, is always visible from the earth to those who look for it, and is then throwing towards the earth just as much light as it does during the night; but the day-moon does not appear more luminous to human sight than any small white cloud of the same apparent *For a description cf his process the reader is referred to Nature, March 16, 1876, p. 391. breadth, so that, although visible in some part of every day except near what is called the change of moon, many persons have passed their lives without noticing it, as might happen with respect to a candle burning in the open air at noon. The full moon gives to the earth only about a one-hundred-thousandth part as much light as the sun. That it is not self-luminous, but that it reflects the light which it receives from the sun is proved by the fact that the light is polarized. (See Polarization of Light.) "Light moves with extraordinary velocity, which, however, has been calculated." 801. The great precision with which the astronomical skill of modern times, enables men to foretell the instants of remarkable occurrences or changes among the heavenly bodies, has served for the detection of the fact, that light is not an instantaneous communication between distant objects and the eye, as was formerly believed, but is a messenger which requires time to travel: and the rate of the travelling has been ascertained. If the interval between two successive eclipses of the satellites or moons of the planet Jupiter be observed when the planet is on the same side of the sun with the earth, and all the subsequent instants of eclipse calculated therefrom, it is found that when Jupiter is on the opposite side of the sun from our earth, the eclipses occur about 16 minutes later than the calculated times. At intermediate situations, the difference between the observed and the calculated times diminishes proportionally, and when Jupiter comes once more into conjunction with the earth, the times agree exactly. This discrepancy can only be explained on the hypothesis that light requires 16 minutes to travel across the earth's orbit, or 8 minutes to come to us from the sun. This immense velocity may perhaps be better appreciated by stating that light would traverse a space equal to the whole circumference of our globe in the eighth part of a second'! The sun's light reflected from the moon is estimated to reach the earth in a second and a quarter. As the planet Jupiter is five times as distant from the sun as the earth, the light, by which he is seen from the earth, must have left the planet at least forty minutes before, and it would require four hours for light to reach us from the remote planet Neptune. In spite of this great velocity, it is calculated that the light of the nearest fixed star would require at least three years to reach the earth, and it might occupy centuries in coming from the nearest nebula, so that we should see the star 572 Proofs of the Velocity of Light. and nebula by the lights which emanated from them years and centuries ago.* 802. The velocity of light, ascertained by the method above described, is 30 marvellous, about 192,000 miles in a second, that the philosophic Dr. Hooke, when the assertion was first made by the Danish astronomer Roemer, in 1676, said he could more easily believe the passage to be absolutely instantaneous, even for any distance, than that there should be a progressive movement so prodigiously rapid. The truth, however, is now put beyond doubt, not only by numerous other facts bearing upon the subject, but also by direct experiment. This feat was accomplished, in 1849, by the French philosopher M. Fizeau, who determined the time taken by a beam of light to travel from Suresnes to Montmartre and back again by reflection, a distance of eleven miles. His contrivance consisted essentially of a wheel, with 720 teeth, which could be rotated with a very high velocity. When a sufficient speed of rotation had been attained he found that a beam of light which passed through one interval of the toothed wheel returned after reflection at the distant station, not through the same interval but through the next one. Before the light had travelled eleven miles, therefore, the wheel had turned through the space of one tooth. The velocity of the wheel being shown by an indicator, the velocity of light was easily found. Fizeau thus calculated it to be about 194,000 miles in a second. All ordinary phenomena upon earth may be considered as happening at the very instant when the eye perceives them, the difference of time being too small to be appreciated. It is therefore not sensibly incorrect, when we are measuring the velocity of sound, by observing the time between the flash and the report of a cannon fired, to assume that the event takes place at the very moment when the eye notes it. The speed of light and of the electric impulse are found to be nearly the same, and the public mind, now familiar with the prodigies of the electric telegraph, is less astonished than it was two centuries ago. * Pouillet, in considering this subject, says that the nearest star cannot be at a less distance than 200,000 times the distance of the earth from the sun. Hence, light in issuing from this star, would require for its passage 200,000 times 8' 13", i.e., 1141 days or three years and forty-five days. It is not unreasonable to suppose that we see stars a hundred thousand times more remote, and that many centuries might elapse before the light from thesc reached our earth. Transparent and Opaque Substances. “Light passes readily through some bodies-which are therefore called transparent; but when it enters or leaves their surfaces obliquely, its course is bent." 573 803. It is very remarkable that light is able to dari readily and in every direction through great masses of solid matter, such as thick plates of glass, blocks of rock crystal, and mountains of ice; while a mere film of another substance may suffice to obstruct it. The reason we cannot yet explain, but we perceive that the arrange. ment of the particles of the mass, has more influence than their peculiar nature. Nothing is more opaque than masses of the metals, but all these become quite transparent when held suspended in liquids, or when forming part of a coloured glass. Thus gold is opaque in substance, but when separated from its solution by phosphorus the metal is so finely reduced that it becomes quite transparent, forming a clear, ruby-coloured liquid with water (Art. 2, p. 2). When diffused through glass by fusion, metallic gold imparts to it that splendid transparent ruby colour, which is so much admired in some varieties of Bohemian glass. Bodies which allow objects to be seen through them are called transparent. This property is possessed by glass, water, and air. Those which cut off the light completely and prevent substances from being seen through them are called opaque. Others which allow light to be transmitted, but which prevent the forms of bodies from being distinctly seen through them, are called translucent— such as ground glass and oiled paper. No substances are, however, absolutely opaque, and none perfectly transparent. Gold itself, which is one of the densest metals, when hammered into very thin leaves, allows a feeble green light to pass through it, and when alloyed with a small proportion of silver, the light which traverses the metal is of a rich purple colour. As the thickness of the leaves is only the 300,000th of an inch, it follows that light penetrates gold at least to this depth. The purest water and air arrest a part of the light which enters them. Miller states that a column of the clearest water seven feet in depth arrests one half of the light which falls upon it. The light which traverses it has a bluish-green tint. This has been employed as a test of purity: the water is examined by a good light through a glass tube of this length enclosed in a metal tube, the two ends being closed with plate glass. If the water is impure very little light passes, and it differs in colour. Pure air also arrests light. Thus, when the light of the sun traverses a great extent of 574 Refraction of Light. atmosphere in a slanting direction-as shortly after sunrise or before sunset-a considerable portion is absorbed and lost. Dr. Young estimated that the light of the sun passing horizontally through about two hundred miles of air possessed only the twothousandth part of its original intensity. REFRACTION OF LIGHT. 804. Light having once entered any transparent mass of uniform or homogeneous nature, passes forward in it as straightly as in a vacuum; but at the surface, whether on entering or leaving it, if the passage be oblique, and if the mass be of a different density or nature from the other transparent medium in contact with it, a very curious and most important phenomenon occurs, namely, the light suffers a degree of bending from its previous direction proportioned to the obliquity. Such bending is technically called REFRACTION. Light passing from air directly or perpendicularly into water, glass, or any such transparent body, suffers not the least bending or k f Fig. 186. deviation from its course. A ray, for a' instance, passing from air in the di rection, a c, into a piece of glass, gh fig. 186), would reach directly across to the points, o and b; but if the ray fall slantingly or obliquely, as along dc, then, instead of continuing along cik, in its first direction, it is at the moment of entering the new medium bent into a path, ce, nearer to co, the perpendicular to the surface at the point of entrance. Moving straightly while in the substance of the glass, it is bent, on passing out again at e, just as much as at first, but in the contrary direction, or away from the perpendicular at that surface, viz., into the line, ef, instead of e n. A ray, therefore, passing obliquely through a transparent body with parallel surfaces, has its course shifted a little to one side of the original course, but still proceeds in the same direction, or in a line parallel to the first—as here shown in the line, eƒ, parallel and near to the line, ik. If the surfaces of the transparent body are not parallel to each other, the ray is otherwise bent, as will be explained in coming pages. The degree of this bending or refraction of light is measured by comparing the obliquity of its approach to the surface, called the angle of incidence, d ca, with the obliquity of its departure beyond, |