Page:Astronomy for Everybody.djvu/168

is added that Thales, the Milesian, had predicted to the Ionians this change of day, even the very year in which it should occur. Our astronomical tables show that there actually was a total eclipse of the sun in the year B. C. 585, which was near enough to the time of the battle to be the one alluded to, but it is now known that the path of the shadow did not quite reach the seat of hostilities till after sunset. Some doubt therefore still rests on the subject.

There is a curious law of the recurrence of eclipses which has been known from ancient times. It is based on the fact that the sun and moon return to nearly the same positions, relative to the node and perigee of the moon's orbit, after a period of six thousand five hundred and eighty-five days eight hours, or eighteen years and twelve days. This period is called the Saros. Eclipses of every sort repeat themselves at the end of a Saros. For example, the eclipse of May, 1900, may be regarded as a repetition of those which occurred in the years 1846, 1864, and 1882. But when such an eclipse recurs it is not visible in the same part of the earth, because of the excess of eight hours in the period. During this eight hours the earth performs one third of a rotation on its axis, which brings a different region under the sun. Each eclipse is visible in a region about one third of the way round the world, or one hundred and twenty degrees of longitude, west of where it occurred before. Only after three periods will the recurrence be near the same region. But in the meantime the moon's line of motion will have