AN'DES, the great mountain chain of South America, extending in nearly parallel with the Pacific, along almost the whole length of the continent, The chain falls short of the isthmus of Darien; between which and the Atrato—a river falling into the Caribbean sea—it gradually subsides into a merely undulating country. It appears, also, to fall still further short of the strait of Magellan, so far as 'the mainland is concerned. But, on geological groundS, it has been traced, first along the islands that breast Pata gonia to the w., and next along thoSe that form the Fujian archipelago. Thus may the chain be said to stretch from the neighborhood of the mouth of the Atrato, not merely to cape Horn, but even to the rocks of Diego Ramirez, which lie about 60 m.' to the s.w. of that promontory. The extreme length, therefore, is from lat. 8° 15' n. to lat. 50' s.—comprising, of course, • 45', or, Without any allowance for windings or devia tions, about 4500 En,glisli miles. But to mark the scale on which nature has molded the new world, the A. may be regarded• as merely a part of the sufficiently continuous chain of about 9000 in. which loses itself near the mouth of the Mackenzie, towards the shores of the Arctic ocean. In this respect, the old continent can bring nothing into comparison.
Position.—The A., besides being generally in a direction nearly parallel with the Pacific, verge closely On that ocean. From the rockS, indeed of Diego Ramirez to about lat. 40' s., the mountains, whether they are found on glands or on the mainland, are almost literally washed by the surf; while northward from that parallel, there spreads out. between the chain itself and the sea, a belt of land not exceeding, in average breadth, 70 or 80 miles. Within the limits of Peru, the belt in question is narrowest, while above and below it is, in general, somewhat more extensive. The position of the A. with respect to the Atlantic ocean presents a striking contrast. To illustrate this, a passage is subjoined from Herndon, the explorer of the Amazon in behalf of the United States. Crossing. from Lima to the head-waters of the Amazon, by the pass of Antarangra, he writes thus: " Yanacoto, on the western slope Of the A., at the height of 2337 ft. above the sea-level, is only 28 tn. from the ocean that washes the base of 'the slope on which it is situated; while fort San Ramon. at nearly the same elevation on the opposite side,
cannot be much less than 4000 m. from its ocean by the windings of the river, and in the river's direct course is at least 2500 miles." Further to compare the two areas respectively to the w. and e. of the dividing ridge, the former has been estimated at 180,000 sq.m., and the latter at twenty times as much.
Hydrography.--This interesting feature of the A. has been already anticipated to a considerable extent, under the heads of the AMAZON and AMERICA. It only remains to observe that from one end of the continent to the other, the true and only watershed, wherever there are two ranges, is the range nearer to the Pacific. Not only is the water shed in question obviously far closer to the w. than to the e., but, beyond this, it is, apparently without .a single exception, pushed as far to the westward as possible; it thus affords the most conspicuous and most decisive example of an almost universal law in the hydrography of the earth. Throughout both continents, almost every leading water-shed presents a longer descent towards the c. than towards the w., or, in other words, sends off larger streams in the former direction than in the latter. To cite a few instances: compare, in North America, the Missouri with the Columbia; in Europe, the Volga with the Neva; in Asia, the Iloang-ho of China with the Oxus of the sea of Aral; and even in Africa, where, as also in Arabia, hydrographical traces have been largely overlaid by deserts of sand, the plateau of the Sahara and the chain of the Atlas gradually incline, both of them, towards the c. But, if the water-shed be inva riably found as far as possible to the westward, it necessarily follows, that, wherever there are two ranges, the more easterly range cannot also be a continuous water-shed unless, indeed, it may be regarded as such with respect to the landlocked basin of the connected lakes, Titicaca and liroz, already mentioned under the head of AMERICA With this exception, all the gatherings between the two ranges, whether the interme diate space be plateau or sierra, have found or formed channels of escape—narrow, •deep, and dark as they often are—only to that sea which is 30 or 40 times, more distant than the one at their back.