lites by the shadow of the planet, the bursting of a meteor, and the appearance or disappearance of a shooting star. The difficulty of identifying these last mentioned objects and the impossibility of foretelling their occurrence prevents the extended use of this method.
Terrestrial signals may be used and among these can be included those sent by the electric telegraph. But when two stations are near together a signal may be made at either or at an intermediate station, which can be observed at both, the time may be noted at each of the stations and the difference found directly. These signals may be made by flashes of gunpowder, or the appearance and disappearance of a strong light, or preconcerted movement of any object easily seen. The heliotrope reflecting the image of the sun from one station to the other with an arrangement for suddenly eclipsing it, is a useful and efficient apparatus.
Various truly astronomical methods have been employed with good results, of these may be mentioned moon-culminations, azimuths of the moon, lunar distances, etc.
Coming now to the use of the electric telegraph for this purpose the following is a rough outline of the methods employed. Suppose two stations A and B connected by wire, and provided with clocks, chronographs and transit instruments. A list of suitable fixed stars is compiled and each observer furnished with a copy. The observer at A the eastern station, selects a star from his list and sets his transit instrument upon it. He is furnished with a key by which he can send telegraphic signals over the line and also mark the time on his own chronograph. The instant he observes the star crossing the spider line which represents the meridian, he taps his key, thus registering the time on his own chronograph and on that at station B and this operation he repeats with as many stars as necessary. B has his instrument set for the first star, and when it crosses his meridian, he taps his key marking the time on his own chronograph and also on A’s. Then, disregarding instrumental and personal errors and the rate of the clock, A has a record of the times at which the star passed both meridians. The difference of these times is the difference of longitude sought, except for an error due to the time occupied in the transmission of the signal over the wire between the stations. B also has a record of the same difference of time with the same error affecting it in the opposite way. A mean of these
