I shall conclude what further I have to say on this subject, by observing, that nothing can be more inaccurate than all Abyssinian calculations. Besides their absolute ignorance in arithmetic, their excessive idleness and aversion to study, and a number of fanciful, whimsical combinations, by which every particular scribe or monk distinguishes himself, there are obvious reasons why there should be a variation between their chronology and ours. I have already observed, that the beginning of our years are different; ours begin on the 1st of January, and theirs on the 1st day of September, so that there are 8 months difference between us. The last day of August may be the year 1780 with us, and 1779 only with the Abyssinians. And in the reign of their kings they very seldom mention either month or day beyond an even number of years. Supposing, then, it is known that the reign of ten kings extended from such to such a period, where all the months and days are comprehended, when we come to assign to each of these an equal number of years, without the correspondent months and days, it is plain that, when all these separate reigns come to be added together, the one sum-total will not agree with the other, but will be more or less than the just time which that prince reigned. This, indeed, as errors compensate full as frequently as they accumulate, will seldom amount to a difference above three years; a space of time too trivial to be of any consequence in the history of barbarous nations.

However, it will occur that even this agreement is no positive evidence of the exactness of the time, for it may so happen that the sum-totals may agree, and yet every particular sum constituting the whole may be false, that is, if the quantity of errors which are too much exactly correspond with the quantity of errors that are too little; to obviate this as much as possible, I have considered three eclipses of the sun as recorded in the Abyssinian annals. The first was in the reign of David III. the year before the king marched out to his first campaign against Maffudi the Moor, in the unfortunate war with Adel. The year that the king marched into Dawaro was the 1526, after having dispatched the Portuguese ambassador Don Roderigo de Lima, who embarked at Masuah on the 26th of April on board the fleet commanded by Don Hector de Silveyra, who had come from India on purpose to fetch him; and the Abyssinian annals say, that, the year before the king marched, a remarkable eclipse of the sun had happened in the Ethiopic month Ter. Now, in consulting our European accounts, we find that, on the second of January, answering to the 18th day of Ter, there did happen an eclipse of the sun, which, as it was in the time of the year when the sky is cloudless both night and day, must have been visible all the time of its duration. So here our accounts do agree precisely.

The second happened on the 13th year of the reign of Claudius, as the Abyssinian account states it. Claudius succeeded to the crown in the 1540, and the 13th year of his reign will fall to be on the 1553. Now we find this eclipse did happen in the same clear season of the year, that is, on the 24th of January 1553, so in this second instance our chronology is perfectly correct.

The third eclipse of the sun happened in the 7th year of the reign of Yasous II. in Magabit, the seventh month of the Abyssinians. Now Yasous came to the crown in 1729, so that the 7th year of his reign will be in 1736, and on the 4th day of October, answering to the 8th day of the month Tekemt, N. S. in that year, we see this eclipse observed in Europe.

As a further confirmation of this, we have dated the particulars of a comet which, the Abyssinian annals say, appeared at Gondar in the month of November, in the 9th year of the reign of Yasous I. and as this comet was observed in Europe to have come to its perihelion in December 1689, and as that year, according to our account, was really the 9th of that king’s reign, no further proof of the exactness of our chronology can possibly be required. By means of these observations, counting backward to the rime of Icon Amlac, and again forward to the death of Joas, which happened in 1768, and assigning to each prince the number of years that his own historians say he reigned, I have, in the most unexceptionable manner that I can devise, settled the chronology of this country; and the exact agreement it hath with all the remarkable events, regularly and sufficiently vouched, plainly shews the accuracy of this method. If, therefore, in a few cases, I differ two or three years from the Jesuits in their first account of this country, I do not in any shape believe the fault to be mine, because there are, at all these periods, errors in point of fact, both in Alvarez and Tellez, much more material and unaccountable than the mistake of a few years; and these errors have been adopted with great confidence in the Hispania Illustrata, and some of the best books of Portuguese history which have made mention of this country.

TRAVELS
TO DISCOVER
THE SOURCE OF THE NILE.

BOOK VI.
FIRST ATTEMPT TO DISCOVER THE SOURCE OF THE NILE FRUSTRATED—A SUCCESSFUL JOURNEY THITHER, WITH A FULL ACCOUNT OF EVERY THING RELATING TO THAT CELEBRATED RIVER.

CHAP. I.
The Author made Governor of Ras el Feel.

I soon received an instance of kindness from Ayto Confu which gave me great pleasure on several accounts. On the south part of Abyssinia, on the frontiers of Sennaar, is a hot, unwholesome, low stripe of country, inhabited entirely by Mahometans, divided into several small districts, known by the general name of Mazaga. Of this I have often before spoken, and shall have further occasion in the sequel.