§ 262. Refraction is not the only change light experiences in our atmosphere. Reflection does not influence the distant luminaries, but acts at all hours and on every body within our atmosphere.

It is not easy to separate the effects of refraction from those of reflection. Twilight, for instance, is the effect of reflection following refraction; reflection prolongs it more than refraction, but to refraction and absorption we owe the infinite variety of the morning and evening sky. Without their influence there would be a sudden transition from splendor to darkness, from day to perfect night. As refraction brings up the sun's disc when actually below the horizon, so it afterward brings his rays up higher, and makes them visible longer than they would otherwise be.

Reflection causes the rays from a body below the horizon after rising above the horizon and striking against the vapors and clouds, and perhaps the atoms of air, to be sent downward to the earth. If the sky is clear these rays are reflected to us of a pure yellow light. If it is loaded with clouds and vapors at different heights, different colored rays struggle through these, and we have a sky varying in color every instant as the rays strike the clouds more or less obliquely. The red rays have most momentum and therefore pass through a misty sky where no others would. The other colored rays are absorbed. The sun's rays rise sufficiently high in our atmosphere to be reflected until he is 18° below the horizon.

§ 263. The usual duration of twilight in the temperate zone is an hour and a fifth long. Duration of twilight is increased even more than we should expect on high mountains. De Saussure passed several nights on the high Alps, and saw the whole horizon surrounded with pale but distinct light which lasted from sunset to sunrise, although the sun must in the middle of the night have been 45° below the horizon. This reflection did not come from the layers of air where the observers stood, for on such heights. they are so near that their reflection is very feeble, but from the thick and deep mass of air which borders the horizon on all sides.

Analogous phenomena are sometimes seen during an

eclipse of the moon. Her disc is not in different eclipses always of the same color. It has been supposed that when that portion of the sun's atmosphere which is so situated as to reflect the sun's rays upon the moon is laden with vapor, it gives to the moon the peculiar light and color which are sometimes observed.

§ 264. Reflected light is not seen only in twilight. Almost all the light which falls upon our eyes has been again and again reflected. The light which comes from a bright luminous body is too brilliant to be agreeable; it is painful to the eye. If it fell upon bodies and were only once reflected to the eye, we should see on every object a round brilliant image of the sun, such as is reflected from polished steel and from water. The moon would send us only a reduced image of the sun. Every body in the direct rays of the sun would be painfully brilliant, and all other bodies would be in the deepest darkness; every room into which the sun was not shining would be as dark as in the night. But most objects which the sun shines on are too rough to reflect his image; they break his rays into innumerable smaller rays, and the greater or less brightness of these and the angle at which they touch our eyes, teach us the form of the body. In very distant bodies we lose the difference of shade and the form consequently, Thus the sun and the moon from their great distance show a flat disc. Besides we see no bodies by the direct light of the sun only, but also by an infinity of cross lights, which coming to us from all parts of the body show us its whole shape. These cross lights on the surface of the earth arise partly from reflection from large objects, but more from reflection from vapors and small particles floating in the air, and perhaps also from the particles of air themselves. The power of very small particles to reflect light is shown by the path of a sunbeam across the room, or across a moist atmosphere, when the sun is improperly said to draw up water. Particles of dust and vapor, before invisible, become perfectly luminous, that is break and reflect the rays in all directions, so that they are themselves seen. Undoubtedly much smaller particles have this power, and by their unseen action produce the soft generally diffused

light of day. The rays are broken and sent as messengers in all directions, crossing and recrossing and wrapping us in a perfect web of light, till we almost forget that the little disc which our two hands can shut out from our sight is the source of it all.

§ 265. The indirect light which is reflected from the sky is often very considerable. In Edinburg it amounts perhaps to 30° or 40° of the photometer in summer, and 10° or 15° in winter. This secondary light is most pow erful when the sky is overspread with thin fleecy clouds. It is feeblest in two very different conditions, either when the sun's rays are obstructed by thick clouds, or when the atmosphere is quite clear and of a pure azure tint. In higher regions, the direct rays of the sun, not being impaired by a long passage through the atmosphere, are more vigorous than at the surface of the earth. But the diffuse, indirect light of the sky, being reflected from a rarer mass of air, unstained by vapors, is proportionately feeble. The silvery hue of the sky changes to a dark hue, slightly tinged with blue in the day time, and at night serving as a transparent black ground for the multitude of stars. As this feeble diffused light does not interfere with vision, large stars and planets are visible from the shade even in the day time.

Reflection from the ground and other opaque objects makes no inconsiderable addition to the amount of light. From a sandy beach, the reflected equals one third of the incident light. From a wide surface of snow, it amounts to five sixths of the direct light; the numerous facets of the bright snowy flakes, which are presented in every possible position, detaining only one sixth of the incident rays, and scattering the rest in all directions.

The laws and facts thus far studied concern the globe as a whole. Before entering on those which take place in portions of its surface, we will see what effect the position of an observer has on the appearance of the heavenly bodies.

CHAPTER XII.

PHENOMENA WHICH

DIFFER IN DIFFERENT PARTS OF
THE EARTH.

Day and Night. Circle of Illumination. Twilight. The Seasons. Curve traced by the Sun's combined daily and yearly Motions. Portion of the Heavens visible in different Latitudes. Length of Day and Night. Equinoxes. Effect of Twilight. Amount of Light and Heat received in a given place. Equality of the distribution of Heat in the Northern and Southern Hemispheres. Difference in their respective Seasons.

§ 266. Before we enter on those those celestial phenomena which appear different when viewed from different parts of the earth, we must inquire what effect our position on the surface of the earth has on external phenomena. The utmost distance by which two observers on the earth's surface can be separated is 8,000 miles; and 8,000 miles is an appreciable distance to bodies no more distant than the sun, moon and planets. It is sufficiently large to be distinguished from them, and causes them to change their places as viewed from the earth. This displacement of a body from its true place, as seen from the earth's centre, is called parallax. Owing to it we never see a heavenly body in its true place unless when the line joining it and our eye passes through the centre of the earth. The displacement is greatest when the body is in the horizon of the observer, because the observer is then distant from the line joining the body to the centre of the earth by the greatest possible amount, the earth's radius. It is then called the horizontal parallax. Its amount may be observed for the sun, moon and planets; it is greatest for near bodies, less for those more distant, and inappreciable for the fixed stars. It is less for each body according to its height above the horizon; it is only in the case of a body in the zenith that it becomes nothing, and in high latitudes this can never take place with any member of the solar system. It always acts in a vertical circle, and always depresses the body, thus partially counteract ing the effects of refraction.

§ 267. The laws we have hitherto investigated have concerned the globe as a whole. We will now study some phenomena which are unlike in different parts of the globe, the variations of seasons and in the length of days. An astronomical day includes twenty-four hours, a natural day may be of any length between nothing and six months. The various modes of reckoning days and years will be mentioned hereafter; at present we have only to account for the length of days at different latitudes, and in the same latitudes at different parts of the year.

To do this we must return to the position of our earth and its two motions; and we must imagine the equatorial and the ecliptic to be marked out on the concave sphere of the heavens; and must remember that while the earth by its rotation causes the sun to appear to move from east to west in the equator or parallel to it, by its motion of revolution it causes the sun to appear to move in the ecliptic from west to east.

One half of this earth's surface is always illuminated, the other half in darkness. This illuminated hemisphere has its edges bounded by a great circle called the circle of illumination. In the spring and autumnal equinoxes it is bounded by a meridian, called the solstitial colure. At no other season of the year is it bounded by a meridian. The illuminated hemisphere extends 90° in every direction from the point to which the sun is at each moment vertical. This point is always in the ecliptic. The illuminated hemisphere therefore may extend 23° beyond either pole, or fall 23° short of it. Since rotation never allows the same spot to remain beneath the sun's vertical rays a moment, the earth each moment turns up a different hemisphere to be illuminated.

§ 268. Let us suppose a concave hemisphere of light to be fastened directly between the earth and the sun in the plane of the earth's revolution, and to move round in the course of the year so as to represent the light falling from the sun. Then let us imagine the earth performing her rotation in a plane 23° inclined to this hemisphere of light, and we shall see that the space for 23° round the pole will rotate sometimes entirely in light, sometimes entirely in darkness.