Strabo observes,(288) that the ancients only guessed that the inundations of the Nile were owing to the rains which fall in great abundance in Ethiopia; but adds, that several travellers have since been eye-witnesses of it; Ptolemy Philadelphus, who was very curious in all things relating to arts and sciences, having sent thither able persons, purposely to examine this matter, and to ascertain the cause of so uncommon and remarkable an effect. 4. _The Time and Continuance of the Inundations._—Herodotus, and after him Diodorus Siculus, and several other authors, declare, that the Nile begins to swell in Egypt at the summer solstice, that is, about the end of June, and continues to rise till the end of September; and then decreases gradually during the months of October and November; after which it returns to its channel, and resumes its wonted course.(289) This account agrees very nearly with the relations of all the moderns, and is founded in reality on the natural cause of the inundation, _viz._ the rains which fall in Ethiopia. Now, according to the constant testimony of those who have been on the spot, these rains begin to fall in the month of April, and continue, during five months, till the end of August and beginning of September. The Nile’s increase in Egypt must, consequently, begin three weeks or a month after the rains have begun to fall in Abyssinia; and accordingly travellers observe, that the Nile begins to rise in the month of May, but so slowly at the first, that it probably does not yet overflow its banks. The inundation happens not till about the end of June, and lasts the three following months, according to Herodotus. I must point out to such as consult the originals, a contradiction in this place between Herodotus and Diodorus on one side; and between Strabo, Pliny, and Solinus, on the other. These last shorten very much the continuance of the inundation; and suppose the Nile to draw off from the lands in three months or a hundred days. And what adds to the difficulty, is, that Pliny seems to ground his opinion on the testimony of Herodotus: _In totum autem revocatur Nilus intra ripas in Librá, ut tradit Herodotus, centesimo die._ I leave to the learned the reconciling of this contradiction.