11. In like manner, that the voyage from Amisus to Colchis, and the route to the Caspian, and thence on to Bactra, are both due east, is proved by the winds, the seasons, the fruits, and even the sun-risings. Frequently evidence such as this, and general agreement, are more to be relied on than the measurement taken by means of instruments. Hipparchus himself was not wholly indebted to instruments and geometrical calculations for his statement that the Pillars and Cilicia lie in a direct line due east. For that part of it included between the Pillars and the Strait of Sicily he rests entirely on the assertion of sailors. It is therefore incorrect to say that, because we cannot exactly determine the duration of the longest and shortest days, nor the degree of shadow of the gnomon throughout the mountainous region between Cilicia and India, that therefore we are unable to decide whether the line traced obliquely on the ancient charts should or should not be parallel, and consequently must leave it unreformed, keeping it oblique as the ancient charts have it. For in the first place, not to determine any thing is to leave it undetermined; and to leave a thing undetermined, is neither to take one view of the matter nor the other: but to agree to leave it as the ancients have, that is to take a view of the case. It would have been more consistent with his reasoning, if he had told us to leave Geography alone altogether, since we are similarly unable to determine the position of the Alps, the Pyrenees, and the mountains of Thrace,[495] Illyria,[496] and Germany. Wherefore should we give more credit to the ancient writers than to the modern, when we call to mind the numerous errors of their charts which have been pointed out by Eratosthenes, and which Hipparchus has not attempted to defend. 12. But the system of Hipparchus altogether teems with difficulties. Reflect for an instant on the following absurdity; after admitting that the southern extremity of India is under the same degree of latitude as Meroe, and that the distance from Meroe to the Strait of Byzantium is about 18,000[497] stadia, he then makes the distance from the southern extremity of India to the mountains 30,000 stadia. Since Byzantium and Marseilles are under the same parallel of latitude, as Hipparchus tells us they are, on the authority of Pytheas, and since Byzantium and the Dnieper[498] have also the same meridian, as Hipparchus equally assures us, if we take his assertion that there is a distance of 3700[499] stadia between Byzantium and the Dnieper, there will of course be a like difference between the latitude of Marseilles and the Dnieper. This would make the latitude of the Dnieper identical with that of Keltica next the Ocean; for on proceeding 3700 stadia [north of Marseilles], we reach the ocean.[500]