seen crossed by dark spaces which give to it the appearance of a string of brilliant beads. The phenomenon, first noticed by Baily, is supposed to be caused by the sun's light shining through hollows between projections on the moon's edge, as those projections approach the outer edge of the solar disc (Lesson 9).

13. Q. What is the appearance of the Corona?

A. The Corona is a luminous circle or "glory" which surrounds the dark edge of the moon while the eclipse continues total. It appears white or of the colour of tarnished silver, and it is now believed to form the sun's atmosphere in which several of the metals exist in the state of vapour.

14. Q. What are the Rose-coloured Protuberances? A. They are coloured masses of light, of various shapes and sizes, irregularly distributed round the sun's edge, and only seen, till recently, during total solar eclipses. By means of the Spectroscope

they are now seen and examined while the sun is shining in full splendour, and are proved to consist of masses of glowing gas, chiefly hydrogen, thrown up with immense velocity to vast heights above the photosphere (Lessons 7, 40).

LESSON 12.-NUMBER OF ECLIPSES: RECESSION OF THE NODES OF THE MOON'S ORBIT.

1. Q. Why are total solar eclipses so seldom seen?

A. Solar eclipses are really more numerous than lunar eclipses, but total eclipses of the sun seldom occur at any given place, because the moon's shadow is comparatively small and many circumstances must concur to produce them. During 575 years (1140-1715) London did not witness one total solar eclipse; indeed according to Mr. Hind the eclipse of 1140 was not total at London, nor will another central and total eclipse of the sun occur in England till August 1999.

2. Q. Why are eclipses of the sun more numerous, but less frequently seen at any place, than those of the moon?

A. They are more numerous, because the solar exceeds the lunar ecliptic limit; and they are more rarely seen at any place, because lunar eclipses are visible at all places having the moon above their horizon, while solar eclipses may only be visible at a few places where the sun is above the horizon at the time. Five eclipses of the sun or three of the moon may occur in a year; and there cannot be less than two-both solar-or more of both than seven-five solar and two lunar.

3. Q. Why must two solar eclipses occur in a year? A. Once a year the sun is near each node, and

at each time there must be one new moon and one eclipse of the sun, because he takes more than a lunar month to move through the solar ecliptic limit-17° on each side of the node. But there may not be a single eclipse of the moon, because the sun is less than a lunar month in passing through the lunar ecliptic limit-12° on each side of the node (Lessons 10, 11).

4. Q. How may seven eclipses occur in a year?

A. Three eclipses, two solar, and, between them, one lunar, may occur when the sun is near each node-in all six eclipses. If the first solar eclipse happens at the beginning of the year, the sun, in consequence of the recession of the node on the ecliptic, may come a second time near enough to the node at new moon to be eclipsed a fifth time within the year.

5. Q. How were eclipses foretold by the ancients?

A. Whatever eclipses occur during a period of 223 Lunations, as the moon's synodical revolutions are called, occur again in nearly the same order, and of the same magnitude, in every similar succeeding period. From a knowledge of this period the ancients were able to predict lunar eclipses, but solar eclipses, were not then known to recur within the same cycle, and could not be foretold.

6. Q. Why do similar eclipses happen in every period of 223 lunations?

A. Eclipses depend on the situation of the nodes of the moon's orbit with respect to the sun, at the time of new and full moon-similar eclipses occurring when the situation of the nodes is the same relatively to the sun. The nodes have nearly

the same position with respect to the sun and moon at the end of each period of 223 lunations, as they had at the beginning of that period, and hence like eclipses recur (Lesson 10).

7. Q. How is this periodical adjustment of the node's position, with respect to the sun and moon, occasioned?

A. It is occasioned by a remarkable relation existing between the periods of the moon's synodic revolution and of the synodic Revolution of the Nodes of her orbit; for the nodes have a retrograde motion and traverse the circumference of the ecliptic in about 186 years. The node, thus moving backwards, meets the sun in his course, and the two arrive at the same relative positions once in a period of about 346-6 days, and, consequently, 19 times in 19 such periods.

8. Q. What have 19 such periods to do with 223 lunations?

A. The two periods are very nearly equal-19 times 346.6 days being 6,585 78 and 223 lunations occupying 6,585 32 days--so that while the former period restores the node exactly, the latter restores it almost exactly, to the same position relatively both to the sun and moon, and the series of eclipses recommences. This series usually includes 70 eclipses-41 solar and 29 lunar.

CHAPTER III.

ON MEASURES OF TIME AND CHANGES OF SEASONS.

THE year of 365 days was early known in China and Chaldea, and was introduced into Greece by Eudoxus of Cnidus (flourished about 370 B.C.); two hundred years later Hipparchus showed that the year was not so long by at least four or five minutes; and its true length was determined to within a few seconds by the Arabian astronomers, one of whom, Al-Batani (880 A.D.), discovered the Motion of the Solar Apogee. The Chinese divided their Circle into 365 degrees, to correspond to the number of days in the year, and the Chaldeans are said to have originated the Duo-decimal Division of the Day and to have used Clepsydra-instruments not unlike hour-glasses, in which water instead of sand was used-to measure the flow of time. The Sun-dial scems to have been early used among the Jews (B.c. 742) and Spartans (B.C. 545), but it was useless during the night, and even in the daytime when the sun was hidden by clouds. All methods of measuring time, even the best, continued very imperfect till Huyghens (A.D. 1629-1695) furnished astronomers, and mankind in general, with the means of securing far greater accuracy by applying the Pendulum to Clocks.

Hipparchus discovered the Precession of the Equinoxes; found that the Sun's motion in the ecliptic is not uniform; and showed that, in consequence of this variation in his motion, the Sun takes a longer period to describe the northern than the southern half of his course-thus causing the Summer to be longer than the Winter. The Nutation of the Earth's Axis was discovered by Dr. Bradley.

Different nations have reckoned their Time in different ways. The Greeks used the Lunar Year consisting of 354 days, and as this differed from the Solar Year, the arrangement of the Calendar occupied their astronomers during several centuries. The object aimed at was to find some period which would always begin when the Month and Year began together, and terminate when next the Month and Year ended together. Meton, an Athenian mathematician, discovered (B.c. 432) that a period of 6,940 days would very nearly answer the purpose, and the public honours with which he was rewarded, attested the importance of the discovery. This period, known as the Metonic Cycle, was adopted by all Grecce. It consists of 19 Solar

E