measuring any distant object by means of a base line. For, from our knowledge of the form and dimensions of the earth (Figure 40), we know the length and position of the line GC; and the observations made at Greenwich and at the Cape of Good Hope give us the angles MGC and MCG; and thus we have the elements for computing the lines GM and CM, and then the distance EM can be found with little trouble.
I then added, that there is one cause of uncertainty, which is refraction, and which produces its effect in this way: we refer the observation of the moon at Greenwich to the North Pole of the heavens; and we refer the observation made at the Cape of Good Hope to the South Pole of the heavens. In deducing the real places of the moon from the apparent places, it is necessary to take into account the quantity of refraction which enters in these two cases. There is one calculation of refraction amounting to a great many seconds, or, perhaps a minute or two, to be taken into account in the observations made at Greenwich; and another calculation of refraction, perhaps amounting to a like quantity, to be taken into account at the Cape of Good Hope. As I said before, refraction is the plague of astronomers, and owing to it, there is always a little uncertainty in the measurement of large angles on the celestial meridian. On that account it is desirable, if possible, to diminish that refraction. If we suppose that there is a star S, Figure 40, at a distance so great that its position is sensibly the same when seen from any part of the earth, and if two observers select the star by previous concert; and if the person at G observes with his instrument the place of the moon as well as that of the star, he finds how many degrees, minutes, and seconds,
