Page:Travels to Discover the Source of the Nile - In the Years 1768, 1769, 1770, 1771, 1772, and 1773 volume 2.djvu/460

444 perihelion, or had already passed it, and was receding to its aphelion; but by its increasing tail, already at a great length, we may conjecture it was only then going down to its conjunction, and was then near approaching to the sun.

this we should conclude that this comet must have been seen, however rapidly it did move, some time before the 6th of November, or two days before the feast of St Michael. But this depends on the circumstances of the climate; for though the tropical rains cease the first of September, the cloudy weather continues all the month of October; at the end of these fall the latter rains in gentle showers, which allay the fevers in Dembea, and make the country wholesome for the march of the army, and these rains fall mostly in the night. From this it is probable that the comet, having at first little light and no tail, as yet at a distance from the sun, was not very apparent to the naked eye, till by its increased motion and heat it had acquired both tail and brightness, as it approached its perihelion.

we find by our European accounts, that, in the year 1689, there did appear a comet, the orbit of which was calculated by M. Pingre. And this comet arrived at its perihelion on the 1st day of December 1689, so was going down much inflamed, and with a violent motion to the sun, the 6th of November, when it was observed at Gondar, being but 25 days then from its perihelion.

these circumstances are more than sufficient to constitute the identity of the comet, a phenomenon too rare to