moon admirably serves the purpose. Suppose it is previously ascertained, by calculationjthat the moment the ball falls, the moon, in its progress through the sky, will touch or occult a certain star; then, though the mariner cannot see the ball, it is enough if he can watch the moment when the moon comes in contact with the star. The nautical almanac tells when he may expect the contact, and if he succeeds in observing it, it is as serviceable to him as if he had the time-signal itself. It is not necessary, however, that the moon should actually come in contact with some star. It may happen that the moon is still some little distance from the star at the moment the ball falls, but this serves the purpose equally. If it is known from the almanac, that the moon will be so many degrees from the star when the ball falls, it is only necessary to watch the instant when the moon comes within the precise distance, in order to know when it is one o'clock at Greenwich. But there is no need to limit the signal to one o'clock. It is only necessary to know from the almanac the time by the Greenwich clock when the moon will arrive at any point i'n her track across the heavens, in order to ascertain what the time at Greenwich is. To people living along a line of railway, the trains are so many time-signals. The time-table shews the Greenwich time when the trains should arrive at each station, and if they be regular, they serve the purpose of a clock. If a person wishes to know the longitude of
