all times when it is seen; indeed, many years it is altogether invisible, especially in our latitude. It would appear to have some connexion with the solar atmosphere; but that this atmosphere should extend beyond the boundary of the orbit of Venus seems unaccountable. Sir John Herschel's opinion is, "that this phenomenon may be no other than the denser part of that medium which, as we have reason to believe, resists the motions of comets, loaded, perhaps, with the actual materials of the tails of millions of these bodies, of which they have been stripped in their perihelion passages, and which may be slowly subsiding into the sun.”

19. The following account of the appearance of the zodiacal light within the tropics, from the pen of Cap‐ tain W. H. Smyth, R.N., whose authority as an observer stands very high, will be gratifying to those who have not had an opportunity of witnessing it under equally favourable circumstances: "At first, it seems a faint, whitish zone of light, less intense than the milky way, with ill-defined borders, scarcely to be distinguished from the twilight, being then but little elevated, and its figure nearly agreeing with that of a spheroid seen in profile. As it rises above the horizon it becomes brighter and larger, till it resembles a lenticular beam of light, somewhat analogous to the tail of a comet, rounded at the vertex, with its base towards the sun, and its axis in the direction of the zodiac."

The zodiacal light may frequently be seen in this country most advantageously in March and September, when the sun is near the equator.

18

SECTION III.

ON THE EARTH.

THE EARTH OF A GLOBULAR FORM-DETERMINATION OF THE EARTH'S DIAMETER BY VARIOUS METHODS-DECREASE OF GRAVITY AT THE EQUATOR ARISING FROM ITS SPHEROIDAL FORM-METHODS OF DETERMINING THE RATIO OF THE TWO DIAMETERS

FOUCAULT'S

PENDULUM EXPERIMENT-DENSITY OF THE EARTH-CONNEXION OF NAVIGATION WITH ASTRONOMY-THE ATMOSPHERE-REFRACTION-TWILIGHT.

20. As the earth which we inhabit is that particular body of our system with whose form and size we have the most favourable opportunities of becoming acquainted, we shall, before proceeding to describe the other planets, notice those particulars which science has contributed respecting it. Nor shall we, in doing so, be deviating from the particular province of astronomyseeing that the earth itself is a planet, subject to the same laws as the others, and occupying a definite place in the celestial sphere.

From an intimate acquaintance with her physical constitution, we may hope to be able, by analogical reasoning, to learn somewhat of her sister planets-their condition, size, and form. Moreover, the correct knowledge of the earth's diameter has supplied us with a scale for the measurement of distances throughout our system, as will be clearly shewn in a future section: viewed then in this light, it must be acknowledged that the measurement of the size, and determination of the form

of the earth, are most intimately connected with the science of astronomy.

21. The most inattentive observer must have arrived at the conclusion, that the earth is not an undeviating plane or flat surface. Standing on the shore of the ocean, it will be seen that, on the approach of vessels, their masts, sails, or rigging will be in sight long before the hull or body of the vessel is seen; while the first objects which meet the sailor's eye are the summits of lofty buildings or elevated mountains. Now if the earth were an extended plane, when a vessel first came in sight, she would be a mere speck in the horizon; but, as she approached, we should not see one part sooner than another, but the whole of her at once.

The surface of the water, then, which we are apt to think perfectly level, is not so ; and hence, in constructing canals, an allowance of eight inches in a mile, below the horizontal plane, is necessary, to compensate for the curvature of water when at rest. Let us suppose, for the sake of illustration, that a tolerably large portion of water were solidified, and had become capable of being cut off; the slice, so to speak, which should measure two miles across, would rise eight inches in the middle. This degree of curvature, or nearly this, is found to be the same on every part of the earth, and in every direction hence we conclude, without further examination, that its form is globular, or nearly so.

22. Still further. If the earth were an extended plane, a ship starting from any point on its surface, and sailing constantly in the same direction, either east or west, would never return to the place from which she started, but continue to increase her distance

from it. But navigators constantly are leaving England, as well as other countries, and following the same course, making only such bends as the obstruction of the land obliges them to do, eventually return to their own country after having circumnavigated the globe.

23. The fact of its sphericity is, however, put beyond a doubt by the last proof which we shall adduce. An eclipse of the moon (§ 63) is caused by her passing through the shadow of the earth; now, whatever part of the shadow she enters, or whatever part of the earth projects this shadow (which at no two consecutive eclipses is exactly the same, nor, indeed, during the entire duration of one eclipse), the shadow thrown on the moon's disc is invariably observed to be a segment of a circle. Now no other form than that of a sphere or globe will, in every direction, cast a circular shadow; the earth therefore must be globular.

24. It may be supposed that the rotundity of our earth is rendered irregular by the elevations and depressions of the mountains and valleys on its surface: this is by no means the case; the greatest elevation of the highest mountains above the level of the sea is not more than five miles, and their mass takes off no more from the earth's rotundity, than particles of dust do from that of a small terrestrial globe.

25. The determination of the size of the earth may appear a problem incapable of solution to those who have not much attended to mathematical pursuits. In order to remove the impression of its impossibility, we shall refer to and explain a few of the methods adopted, premising that all the more minute circumstances will be entirely left out of the consideration; these are nume

rous, and in practical surveys, are of great importance; so much so, that of late years, since the great improvements in mathematical instruments and methods of measurement, the determination simply of the form of the globe, or of large portions of its surface, has been constituted a distinct science, under the name of Geodesy; the province of which extends over all those exceedingly nice corrections, arising, amongst other causes, from the unequal curvature of the earth, the state of the atmosphere causing unequal refractions, the height of stations above the horizontal plane, &c. &c. ; the effects of which, though minute individually, would, when accumulated in an extensive survey, produce a serious error in the result. So accurately has the diameter of the earth been measured by the greatest mathematicians of the principal countries in Europe, that, as Sir John Herschel declares, the error, if any does exist in the estimate of the earth's diameter, a space of 7912 miles, cannot be so much as 200 yards.

26. One method of determining the circumference of the globe is by measuring the distance between two towers, or any other high buildings, and taking, at the same time, the angle subtended by a line from the top of either, and a plumb-line directed towards the earth's