and the Thracian Chersonesus613 and Sunium614 form the Gulf of Melas,615 and likewise those of Macedonia.616 Added to this, it is manifest that the majority of the distances are falsely stated, thus arguing an ignorance of geography scarcely credible, and so far from requiring geometrical demonstration that it stands out prominent on the very face of the statements. For example, the distance from Epidamnus617 to the Thermaic Gulf618 is above 2000 stadia; Eratosthenes gives it at 900. So too he states the distance from Alexandria to Carthage at 13,000619 stadia; it is not more than 9000, that is, if, as he himself tells us, Caria and Rhodes are under the same meridian as Alexandria,620 and the Strait of Messina under the same as Carthage,621 for every one is agreed that the voyage from Caria to the Strait of Sicily does not exceed 9000 stadia.
It is doubtless permissible in very great distances to consider as under one and the same meridian places which are not more east and west of each other than Carthage is west of the Strait;622 but an error of 3000 stadia is too much; and when he places Rome under the same meridian as Carthage, notwithstanding its being so far west of that city, it is but the crowning proof of his extreme ignorance both of these places, and likewise of the other countries farther west as far as the Pillars of Hercules.
41. Since Hipparchus does not furnish a Geography of his own, but merely reviews what is said in that of Eratosthenes, he ought to have gone farther, and corrected the whole of that writer’s mistakes. As for ourselves, it is only in those particulars where Eratosthenes is correct (and we acknowledge that he frequently errs) that we have thought it our duty to quote his own words, in order to reinstate them in their position, and to defend him when he could be acquitted of the charges of Hipparchus; never failing to break a lance with the latter writer whenever his objections seemed to be the result of a mere propensity to find fault. But when Eratosthenes is grossly mistaken, and the animadversions of Hipparchus are just, we have thought it sufficient in our Geography to set him (Eratosthenes) right by merely stating facts as they are. As the mistakes were so continual and numerous, it was better not to mention them except in a sparse and general manner. This principle in the details we shall strive to carry out. In the present instance we shall only remark, that Timosthenes, Eratosthenes, and those who preceded them, were but ill acquainted with Iberia and Keltica,623 and a thousand times less with Germany, Britain, and the land of the Getæ and Bastarnæ.624 Their want of knowledge is also great in regard to Italy, the Adriatic, the Euxine, and the countries north of these. Possibly this last remark may be regarded as captious, since Eratosthenes states, that as to distant countries, he has merely given the admeasurements as he finds them supplied by others, without vouching for their accuracy, although he sometimes adds whether the route indicated is more or less in a right line. We should not therefore subject to a too rigorous examination distances as to which no one is agreed, after the manner Hipparchus does, both in regard to the places already mentioned, and also to those of which Eratosthenes has given the distance from Hyrcania to Bactria and the countries beyond, and those from Colchis to the Sea of Hyrcania. These are points where we should not scrutinize him so narrowly as [when he describes] places situated in the heart of our continent,625 or others equally well known; and even these should be regarded from a geographical rather than a geometrical point of view. Hipparchus, at the end of the second book of his Commentaries on the Geography of Eratosthenes, having found fault with certain statements relative to Ethiopia, tells us at the commencement of the third, that his strictures, though to a certain point geographical, will be mathematical for the most part. As for myself, I cannot find any geography there. To me it seems entirely mathematical; but Eratosthenes himself set the example; for he frequently runs into scientific speculations, having little to do with the subject in hand, and which result in vague and inexact conclusions. Thus he is a mathematician in geography, and in mathematics a geographer; and so lies open to the attacks of both parties. In this third book, both he and Timosthenes get such severe justice, that there seems nothing left for us to do; Hipparchus is quite enough.
CHAPTER II.
1. We will now proceed to examine the statements made by Posidonius in his Treatise on the Ocean. This Treatise contains much geographical information, sometimes given in a manner conformable to the subject, at others too mathematical. It will not, therefore, be amiss to look into some of his statements, both now and afterwards, as opportunity occurs, taking care to confine ourselves within bounds. He deals simply with geography, when he tells us that the earth is spheroidal and the universe too, and admits the necessary consequences of this hypothesis, one of which is, that the earth contains five zones.
2. Posidonius informs us that Parmenides was the first to make this division of the earth into five zones, but that he almost doubled the size of the torrid zone, which is situated between the tropics, by bringing it beyond these into the temperate zones.626 But according to Aristotle the torrid zone is contained between the tropics, the temperate zones occupying the whole space between the tropics and the arctic circles.627 Both of these divisions Posidonius justly condemns, for the torrid zone is properly the space rendered uninhabitable by the heat. Whereas more than half of the space between the tropics is inhabited, as we may judge by the Ethiopians who dwell above Egypt. The equator divides the whole of this space into two equal parts. Now from Syene, which is the limit of the summer tropic, to Meroe, there are 5000 stadia, and thence to the parallel of the Cinnamon region, where the torrid zone commences, 3000 stadia. The whole of this distance has been measured, and it may be gone over either by sea or land; the remaining portion to the equator is, if we adopt the measure of the earth supplied by Eratosthenes, 8800 stadia. Therefore, as 16,800 is to 8800, so is the space comprised between the tropics to the breadth of the torrid zone.
If of the more recent measurements we prefer those which diminish the size of the earth, such as that adopted by Posidonius, which is about 180,000 stadia,628 the torrid zone will still only occupy half, or rather more than half, of the space comprised between the tropics; but never an equal space. [Respecting the system of Aristotle, Posidonius farther says,] “Since it is not every latitude which has Arctic Circles,629 and even those which do possess them have not the same, how can any one determine by them the bounds of the temperate zones, which are immutable?” Nothing however is proved [against Aristotle] from the fact that there are not Arctic Circles for every latitude, since they exist for all the inhabitants of the temperate zone, on whose account alone the zone receives its name of temperate. But the objection that the Arctic Circles do not remain the same for every latitude, but shift their places, is excellent.630
3. Posidonius, who himself divides the earth into zones, tells us that “five is the number best suited for the explanation of the celestial appearances, two of these are periscii,631 which reach from the poles to the point where the tropics serve for Arctic Circles; two more are heteroscii,632 which extend from the former to the inhabitants of the tropics, and one between the tropics, which is called amphiscius,633 but for matters relative to the earth, it is convenient to suppose two other narrow zones placed under the tropics, and divided by them into two halves, over which [every year] for the space of a fortnight, the sun is vertical.”