Eclipses of the Solar eclipses can occur only at the time of new moon. If the sun be more than 19 days from a node of the moon's orbit an eclipse is impossible, if less than 13 days, a central solar eclipse is in evitable. These are the solar ecliptic limits. The sun causes the moon to project toward the earth at new moon, umbral and penumbral shadow cones differing in the main only in size from those which the earth projects toward and beyond the moon at full moon. The shadow cone of the moon may strike the earth's surface, but on the average it does not. Two distinct cases may arise. First, the axis of the shadow cone may strike the earth's surface, and upon it the right line joining the centres of the solar and lunar discs traces the curve of centrality. Here three cases result : (1) The apex of the shadow cone reaches the earth's surface and a central total eclipse occurs. (2) The apex of the shadow cone does not reach the earth's surface and there occurs a central annular eclipse. (3) The apex of the shadow cone reaches the earth's surface only at a middle stretch of the curve of centrality and there occurs a central annular-total eclipse. Second, the axis of the shadow cone may not strike the earth's surface but the earth may still enter the shadow cone, or shadow cone produced, and there occur total, annular and annular-total eclipses, all of which are of a non-central char acter. The path of a total eclipse on the earth may traverse a length from west to east of several thousand miles. An annular eclipse has also a upath of the annulus') in which the annular phase is visible. Outside of both of these paths the sun is only partially eclipsed. The least number of eclipses, solar and lunar, which can occur in a year, is two, and these must be central eclipses of the sun, while the maximum number occurring is seven, two of the moon and five of the sun or four of the sun and three of the moon. A simple record of the date and character of eclipses revealed to early astronomers the fact that after a period of about 18 years eclipses again recur in the same order and kind whether solar, lunar, total or partial. This cycle, called the SAROS, is one of 223 lunar months, each of a little more than 29.5 days, and is precisely 6,585.32 days in length. Depending on the number of leap years, 5 or 4 intervening, this is equal to 18 years 10.32 days or 11.32 days. At the end of the Saros the relative positions of the sun, moon and node of the moon's orbit are nearly the same as at the beginning, and other favoring factors of the motions of the sun and moon also recur, as Newcomb first pointed out, to nearly original values. The excess of one-third of a day in the cycle causes the eclipse to be thrown about eight hours of longitude farther west on the earth's surface. Hence only after three cycles does the eclipse again recur in approximately the same longitude. If the inter vening leap years be regarded, the application of the Saros to various series of solar eclipses of the following table may readily be made.
Ancient eclipses are chiefly of value in his torical and chronological investigation (see TIME). The records of the supposed eclipses of the ancient historians are usually very un certain as to place, time and character of the occurrence, and hence have proved of question able value to the astronomer. The precise cir cumstances of an eclipse occurring centuries ago can readily be computed from the modern tables, but the comparison with the vague his torical references has apparently contributed but little to science. Ginzel has, in his
by the older astronomers and carefully recorded are of great value. And in later times the description of eclipses has been so accurately interwoven with even historical accounts as to furnish unquestionable data for chronology. The terror and apprehensive uncertainty asso ciated with an eclipse in ancient times has given place to a lofty appreciation of the grand occa sion and to a careful and inspiring scientific study of the phenomena.
The annexed table is a complete list of all the eclipses of the century having a total phase, however brief. The first column gives the date for the point on the earth where at noon the eclipse is central. The second column gives the Greenwich civil time of the conjunction of the sun and moon in longitude. The next column states the greatest interval in minutes during which the most favored place may ex perience the total phase. The fourth column gives the latitude and longitude of the place where the eclipse is central at noon. The last column indicates the areas traversed by the moon's shadow. Several of the eclipses, occur ring in the polar regions, have no noon-point and hence, for these, the place of the beginning or end of the eclipse, lowest in latitude, is placed in the fourth, and the corresponding time, sun rise or sunset, in the last column. The eight annular-total eclipses of the century are indi cated by a *, and in these the duration of total phase may be but a few seconds. Two of the list, namely 1928 and 1967, are non-central in character, the moon's shadow just grazing the pole of the earth.
The Phenomena Observed in Total Solar Eclipses.— As totality approaches, a pale purple coloring spreads over the landscape. Within a minute of the total phase the phenomena begin to succeed each other so rapidly that no single observer can note them all. By those glancing at the landscape in the direction of the ap proaching shadow a majestic darkness will be seen to sweep forward with a swiftness truly impressive. Those looking at the earth, and away from the eclipsing sun, will see a succes sion of flitting bands, alternately dark and bright, known as the shadow bands, which for many decades remained an unexplained puzzle. About 15 seconds before the total phase, the moon, with its sharp irregular outline, due to the elevations and depressions in its surface, will have diminished the solar crescent to such an extent that the remnant of it usually begins to be crossed by black lines,— the mountains on the moon's edge connecting the limbs of moon and sun. Then it rather suddenly presents the appearance of a string of detached brilliant points, irregular in size and spacing. The ap pearance is due to irradiation which alike ex aggerates the length and thinness of the pro jected lunar mountains and the size of the resulting bright points of the disappearing solar crescent. Totality, the appearance of the prominences, whether red or white, along the moon's edge, and the presence of the radiant, outspreading corona, are simultaneous. The phenomena of the earlier eclipses were recorded in the most general manner. Hasty drawings of the streamers of the corona, accompanied by descriptions reflecting but the dawn of scientific method, are the chief results. It was not until the application of the spectroscope and the photographic camera that the astronomer attained real insight into the stupendous phenomena.