Ennius, writing towards the end of the second century B.C., describes an eclipse which was said to have happened nearly two hundred years before (404, B.C.), in the following remarkable words:--"On the nones of July the moon passed over the sun, and there was night." Aristarchus, three centuries before Christ, understood and explained the nature of eclipses; but the chief of the ancient authors upon this subject was Hipparchus. He and his disciples were able to predict eclipses with considerable accuracy, both as to their time and duration. Geminus and Cleomedes were two other writers, somewhat later, who explained and predicted eclipses. In later times regular tables were drawn up, showing when the eclipses would happen. One that Ptolemy was the author of was founded on data derived from ancient observers--Callipus, Democritus, Eudoxus, Hipparchus--aided by his own calculations. After the days of Ptolemy the knowledge of the eclipses advanced _pari passu_ with the advance of astronomy generally. So long as astronomy itself was empirical, the time of the return of an eclipse was only reckoned by the intervals that had elapsed during the same portion of previous cycles; but after the discovery of elliptic orbits and the force of gravitation the whole motion of the moon could be calculated with as great accuracy as any other astronomical phenomenon. In point of fact, if the new moon is in the plane of the ecliptic there must be an eclipse of the sun; if the full moon is there, there must be an eclipse of the moon; and if it should in these cases be only partially in that plane, the eclipses also will be partial. The cycle of changes that the position of the moon can undergo when new and full occupies a period of eighteen years and eleven days, in which period there are forty-one eclipses of the sun and twenty-nine of the moon. Each year there are at most seven and at least two eclipses; if only two, they are eclipses of the sun. Although more numerous in reality for the whole earth, eclipses of the sun are more rarely observed in any particular place, because they are not seen everywhere, but only where the shadow of the moon passes; while all that part of the earth that sees the moon at all at the time sees it eclipsed. We now come to comets.