overlaps that of the sun a total eclipse ensues.

When the body of the moon hides a portion of the sun's disc a partial eclipse takes place. When the moon is in the most remote part of the orbit her disc is diminished, as viewed from the earth; and if an eclipse of the sun occur under such circumstances, and the moon's centre coincide with the sun's, the disc of the moon is not large enough to cover that of the sun, so that a ring of the sun is seen, while the other part of it is obscured, or occulted. This phenomena is the annular eclipse (annulus, a ring), and is of rare occurrence.

3. "How often would similar eclipses return, if there were no regression of the moon's nodes?"

In

In the absence of regression of the moon's nodes similar eclipses of the sun would occur on each completion of the moon's synodic revolution, or at the interval between the moon's leaving the sun and returning to her again, as viewed from the earth, this period, 29 days 12 hours 44 minutes, the moon passes through all her phases, or completes a lunation. The eclipses of the moon would occur as frequently as those of the sun, but in the middle of the synodic revolution of the moon, the solar eclipses being due to conjunction, and the lunar eclipses to opposition of the moon but this is only correct on the hypothesis that the moon's nodes are in conjunction, or opposition, for one eclipse, after which regression of the nodes is assumed not to occur; for if the moon's nodes were not in conjunction, or opposition, at the time of the cessation of their backward movement an eclipse could never

occur.

SECTION III.

1. "Describe the apparent path of the sun on a summer's day in the arctic circle."

Within the frigid zone the apparent path of the sun, on a summer's day, would be a circle. When the sun attained the most elevated point of this circle it would be noon; when he reached the lowest point, it would be midnight; and sunrise would occur at the same instant with sunset, if we may so term the least visible elevation of the sun. At noon he would be seen in the south, and the place having noon would be directly intermediate between the pole and the sun. But twelve hours afterwards the place which was between the pole and the sun would have the pole between itself and the sun, which, consequently, would be seen in the north at midnight.

2. "Show that more of the sun's rays fall in a given portion of the earth's surface when they are incident vertically, than when obliquely."

If a circular piece of board be lodged on the upper surface of a globe having the same diameter with the board, and a light be placed immediately over the centre of the board, it is evident that the board will interrupt every ray of light and heat that would have fallen upon the globe in the absence of the board. But the shaded surface of the globe is greater than any section of the globe, such as the board. Hence, the rays of heat and light that would be received on the circular board, have, on its removal, to spread over a larger surface, and their intensity is proportionably diminished. And this diminution is in proportion to the distance of the globular surface from the centre of our supposed obstacle; for if we suppose the board to be very small in comparison with its distance from the luminary whose heat it interrupts, then equal portions of the board receive equal quantities of heat. But these equal portions of the board do not shade equal surfaces; and if we suppose it to be cut into concentric rings, with a circle in the centre,

each ring and the central portion being of equal area, the central portion would shade the least, the ring adjoining it more, and the outside ring the greatest portion of globular surface. More of the heat and light, therefore, would face on that part of the globe which was shaded by the central section of our supposed obstacle than on an equal portion of the surface shaded by the outer ring. In the first case the rays would be vertical; in the latter, slanting, or oblique; and the sun being substituted for our imaginary light, and the earth for the globe, the illustration required by the question is afforded.

3. "The extreme summer heat of Moscow is equal to that of Nantes, that of Tobolsk to that of Cherbourg, and that of Astrachan to that of Bordeaux. Account for these and similar facts."

From the reply to the preceding question it will appear that, as a general rule, the temperature of any place on the earth's surface is in proportion to its distance from that part of the earth which receives the direct rays of the sun-i.e., from the equator. But this rule is subject to important modifications, owing to the local circumstances of some situations. The greater or less column of the atmosphere, over any place, is a datum of only inferior importance to that of the variation in the obliquity of the sun's rays; for the denser the air, the greater its capacity for heat. Now the column of air is greater at the sea-level; and the greater the elevation of the land the less is the weight of the superincumbent atmosphere. Hence, we should expect for the same latitude a higher temperature in countries near the sea than in others situated inland, and having a greater elevation above the sea-level. And for the average annual temperature this is the case; but many causes operate in producing excep

tions. Nantes, Cherbourg, and Bordeaux, on account of their proximity to the sea, and free exposure to the moist westerly winds from the Atlantic, have a lower summer temperature than is due to their latitude. On the other hand, Moscow, Tobolsk, Astrachan, have a higher summer temperature than their latitude would lead us to expect, and for the following reasons:- -They are surrounded by vast tracts of land, which, in the long days, and under the powerful sun of summer, acquire a much higher temperature than similar tracts of water would accumulate. The accumulated heat of the land is transferred to the atmosphere, which is further raised in temperature by the influx of vast volumes of heated air supplied by the easterly winds from the great deserts of central Asia. The soil of these places, too, is itself sandy, and therefore more capable of absorbing and retaining the solar heat than the more compact soils of the localities with which they are contrasted in our question.

The mild and genial climates of southern Europe, as compared with the same latitudes in America, afford a striking contrast between the effect of proximity in the first case to the great arid deserts of Africa; and in the second, to the ice-bound ocean to the north of the American continent.

The free egress afforded to the ice of the Antarctic ocean, by the absence of a belt of land, such as to a great extent confines that of the Arctic, causes a considerable diminution of the temperature of some southern latitudes, as contrasted with those equally far north of the equator. The greater preponderance of land in the northern hemisphere affords another reason for the higher temperature that prevails in it than in the southern; for the land reflects better than water the solar heat which it receives to the air above it.

SECTION IV.

1. "Account for the apparently retrograde motions of the planets."

From our station, the earth, we observe a section of the movements of the planets, in contradistinction to what may be termed a plan of the same movements.

Sometimes they seem to advance-then to relax their rate of progress-then to cease moving entirely-then to reverse their motion; but, with all their irregularities, to have a greater amount of direct than of retrograde motion. These anomalies will be in some measure accounted for by reference to the above diagram. To explain all the peculiarities of apparent planetary motions would require many complicated diagrams, for which the reader is referred to the works of Herschell, Dick, &c.

In the above figure S is the sun; E F G, three positions of the earth in its orbit; M N O, three positions of a planet in its orbit. From the earth the position M of the planet would be referred to m in x y, the concave of the heavens. When the earth advanced to F and the planet to N, the latter would be referred to the position n in the heavens, and would consequently have appeared to have advanced from m to n. When