spanish guiana. angostura. palm-inhabiting tribes. missions of the capuchins. the laguna parime. el dorado. legendary tales of the early voyagers.

i shall commence this chapter by a description of spanish guiana (provincia de la guyana), which is a part of the ancient capitania general of caracas. since the end of the sixteenth century three towns have successively borne the name of st. thomas of guiana. the first was situated opposite to the island of faxardo, at the confluence of the carony and the orinoco, and was destroyed* by the dutch, under the command of captain adrian janson, in 1579. the second, founded by antonio de berrio in 1591, near twelve leagues east of the mouth of the carony, made a courageous resistance to sir walter raleigh, whom the spanish writers of the conquest know only by the name of the pirate reali. the third town, now the capital of the province, is fifty leagues west of the confluence of the carony. it was begun in 1764, under the governor don joacquin moreno de mendoza, and is distinguished in the public documents from the second town, vulgarly called the fortress (el castillo, las fortalezas), or old guayana (vieja guayana), by the name of santo thome de la nueva guayana. this name being very long, that of angostura* (the strait) has been commonly substituted for it.

[* the first of the voyages undertaken at raleigh’s expense was in 1595; the second, that of laurence keymis, in 1596; the third, described by thomas masham, in 1597; and the fourth, in 1617. the first and last only were performed by raleigh in person. this celebrated man was beheaded on october the 29th, 1618. it is therefore the second town of santo tomas, now called vieja guyana, which existed in the time of raleigh.]

[* europe has learnt the existence of the town of angostura by the trade carried on by the catalonians in the carony bark, which is the beneficial bark of the bonplanda trifoliata. this bark, coming from nueva guiana, was called corteza or cascarilla del angostura (cortex angosturae). botanists so little guessed the origin of this geographical denomination that they began by writing augustura, and then augusta.]

angostura, the longitude and latitude of which i have already indicated from astronomical observations, stands at the foot of a hill of amphibolic schist* bare of vegetation. the streets are regular, and for the most part parallel with the course of the river. several of the houses are built on the bare rock; and here, as at carichana, and in many other parts of the missions, the action of black and strong strata, when strongly heated by the rays of the sun upon the atmosphere, is considered injurious to health. i think the small pools of stagnant water (lagunas y anegadizos), which extend behind the town in the direction of south-east, are more to be feared. the houses of angostura are lofty and convenient; they are for the most part built of stone; which proves that the inhabitants have but little dread of earthquakes. but unhappily this security is not founded on induction from any precise data. it is true that the shore of nueva andalusia sometimes undergoes very violent shocks, without the commotion being propagated across the llanos. the fatal catastrophe of cumana, on the 4th of february, 1797, was not felt at angostura; but in the great earthquake of 1766, which destroyed the same city, the granitic soil of the two banks of the orinoco was agitated as far as the raudales of atures and maypures. south of these raudales shocks are sometimes felt, which are confined to the basin of the upper orinoco and the rio negro. they appear to depend on a volcanic focus distant from that of the caribbee islands. we were told by the missionaries at javita and san fernando de atabapo that in 1798 violent earthquakes took place between the guaviare and the rio negro, which were not propagated on the north towards maypures. we cannot be sufficiently attentive to whatever relates to the simultaneity of the oscillations, and to the independence of the movements in contiguous ground. everything seems to prove that the propagation of the commotion is not superficial, but depends on very deep crevices that terminate in different centres of action.

[* hornblendschiefer.]

the scenery around the town of angostura is little varied; but the view of the river, which forms a vast canal, stretching from south-west to north-east, is singularly majestic.

when the waters are high, the river inundates the quays; and it sometimes happens that, even in the town, imprudent persons become the prey of crocodiles. i shall transcribe from my journal a fact that took place during m. bonpland’s illness. a guaykeri indian, from the island of la margareta, was anchoring his canoe in a cove where there were not three feet of water. a very fierce crocodile, which habitually haunted that spot, seized him by the leg, and withdrew from the shore, remaining on the surface of the water. the cries of the indian drew together a crowd of spectators. this unfortunate man was first seen seeking, with astonishing presence of mind, for a knife which he had in his pocket. not being able to find it, he seized the head of the crocodile and thrust his fingers into its eyes. no man in the hot regions of america is ignorant that this carnivorous reptile, covered with a buckler of hard and dry scales, is extremely sensitive in the only parts of his body which are soft and unprotected, such as the eyes, the hollow underneath the shoulders, the nostrils, and beneath the lower jaw, where there are two glands of musk. the guaykeri indian was less fortunate than the negro of mungo park, and the girl of uritucu, whom i mentioned in a former part of this work, for the crocodile did not open its jaws and lose hold of its prey. the animal, overcome by pain, plunged to the bottom of the river, and, after having drowned the indian, came up to the surface of the water, dragging the dead body to an island opposite the port. a great number of the inhabitants of angostura witnessed this melancholy spectacle.

the crocodile, owing to the structure of its larynx, of the hyoidal bone, and of the folds of its tongue, can seize, though not swallow, its prey under water; thus when a man disappears, the animal is usually perceived some hours after devouring its prey on a neighbouring beach. the number of individuals who perish annually, the victims of their own imprudence and of the ferocity of these reptiles, is much greater than is believed in europe. it is particularly so in villages where the neighbouring grounds are often inundated. the same crocodiles remain long in the same places. they become from year to year more daring, especially, as the indians assert, if they have once tasted of human flesh. these animals are so wary, that they are killed with difficulty. a ball does not pierce their skin; and the shot is only mortal when it penetrates the throat or a part beneath the shoulder. the indians, who know little of the use of fire-arms, attack the crocodile with lances, after the animal has been caught with large pointed iron hooks, baited with pieces of meat, and fastened by a chain to the trunk of a tree. they do not approach the animal till it has struggled a long time to disengage itself from the iron fixed in the upper jaw. there is little probability that a country in which a labyrinth of rivers without number brings every day new bands of crocodiles from the eastern back of the andes, by the meta and the apure, toward the coast of spanish guiana, should ever be delivered from these reptiles. all that will be gained by civilization will be to render them more timid and more easily put to flight.

affecting instances are related of african slaves, who have exposed their lives to save those of their masters, who had fallen into the jaws of a crocodile. a few years ago, between uritucu and the mission de abaxo, a negro, hearing the cries of his master, flew to the spot, armed with a long knife (machete), and plunged into the river. he forced the crocodile, by putting out his eyes, to let go his prey and to plunge under the water. the slave bore his expiring master to the shore; but all succour was unavailing to restore him to life. he had died of suffocation, for his wounds were not deep. the crocodile, like the dog, appears not to close its jaws firmly while swimming.

the inhabitants of the banks of the orinoco and its tributary streams discourse continually on the dangers to which they are exposed. they have marked the manners of the crocodile, as the torero has studied the manners of the bull. when they are assailed, they put in practice, with that presence of mind and that resignation which characterize the indians, the zamboes, and copper-coloured men in general, the counsels they have heard from their infancy. in countries where nature is so powerful and so terrible, man is constantly prepared for danger. we have mentioned before the answer of the young indian girl, who delivered herself from the jaws of the crocodile: “i knew he would let me go if i thrust my fingers into his eyes.” this girl belonged to the indigent class of the people, in whom the habits of physical want augment energy of character; but how can we avoid being surprised to observe in the countries convulsed by terrible earthquakes, on the table-land of the province of quito, women belonging to the highest classes of society display in the moment of peril, the same calm, the same reflecting intrepidity? i shall mention one example only in support of this assertion. on the 4th of february, 1797, when 35,000 indians perished in the space of a few minutes, a young mother saved herself and her children, crying out to them to extend their arms at the moment when the cracked ground was ready to swallow them up. when this courageous woman heard the astonishment that was expressed at a presence of mind so extraordinary, she answered, with great simplicity, “i had been told in my infancy: if the earthquake surprise you in a house, place yourself under a doorway that communicates from one apartment to another; if you be in the open air and feel the ground opening beneath you, extend both your arms, and try to support yourself on the edge of the crevice.” thus, in savage regions or in countries exposed to frequent convulsions, man is prepared to struggle with the beasts of the forest, to deliver himself from the jaws of the crocodile, and to escape from the conflict of the elements.

the town of angostura, in the early years of its foundation, had no direct communication with the mother-country. the inhabitants were contented with carrying on a trifling contraband trade in dried meat and tobacco with the west india islands, and with the dutch colony of essequibo, by the rio carony. neither wine, oil, nor flour, three articles of importation the most sought after, was received directly from spain. some merchants, in 1771, sent the first schooner to cadiz; and since that period a direct exchange of commodities with the ports of andalusia and catalonia has become extremely active. the population of angostura,* after having been a long time languishing, has much increased since 1785. at the time of my abode in guiana, however, it was far from being equal to that of stabroek, the nearest english town. the mouths of the orinoco have an advantage over every other part in terra firma. they afford the most prompt communications with the peninsula. the voyage from cadiz to punta barima is performed sometimes in eighteen or twenty days. the return to europe takes from thirty to thirty-five days. these mouths being placed to windward of all the islands, the vessels of angostura can maintain a more advantageous commerce with the west indies than la guayra and porto cabello. the merchants of caracas, therefore, have been always jealous of the progress of industry in spanish guiana; and caracas having been hitherto the seat of the supreme government, the port of angostura has been treated with still less favour than the ports of cumana and nueva barcelona. with respect to the inland trade, the most active is that of the province of varinas, which sends mules, cacao, indigo, cotton, and sugar to angostura; and in return receives generos, that is, the products of the manufacturing industry of europe. i have seen long boats (lanchas) set off, the cargoes of which were valued at eight or ten thousand piastres. these boats went first up the orinoco to cabruta; then along the apure to san vicente; and finally, on the rio santo domingo, as far as torunos, which is the port of varinas nuevas. the little town of san fernando de apure, of which i have already given a description, is the magazine of this river-trade, which might become more considerable by the introduction of steamboats.

[* angostura, or santo thome de la nueva guayana, in 1768, had only 500 inhabitants. caulin page 63. they were numbered in 1780 and the result was 1513 (455 whites, 449 blacks, 363 mulattoes and zamboes, and 246 indians). the population in the year 1789 rose to 4590; and in 1800 to 6600 souls. official lists manuscript. the capital of the english colony of demerara, the town of stabroek, the name of which is scarcely known in europe, is only fifty leagues distant, south-east of the mouths of the orinoco. it contains, according to bolingbroke, nearly 10,000 inhabitants.]

i have now described the country through which we passed during a voyage of five hundred leagues; it remains for me to make known the small space of three degrees fifty-two minutes of longitude, that separates the present capital from the mouth of the orinoco. exact knowledge of the delta and the course of the rio carony is at once interesting to hydrography and to european commerce.

when a vessel coming from sea would enter the principal mouth of the orinoco, the boca de navios, it should make the land at the punta barima. the right or southern bank is the highest: the granitic rock pierces the marshy soil at a small distance in the interior, between the cano barima, the aquire, and the cuyuni. the left, or northern bank of the orinoco, which stretches along the delta towards the boca de mariusas and the punta baxa, is very low, and is distinguishable at a distance only by the clumps of moriche palm-trees which embellish the passage. this is the sago-tree* of the country the quantity of nutritious matter which the real sago-tree of asia affords (sagus rumphii, or metroxylon sagu, roxb.) exceeds that which is furnished by any other plant useful to man. one trunk of a tree in its fifteenth year sometimes yields six hundred pounds weight of sago, or meal (for the word sago signifies meal in the dialect of amboyna). mr. crawfurd, who resided a long time in the indian archipelago, calculates that an english acre could contain four hundred and thirty-five sago-trees, which would yield one hundred and twenty thousand five hundred pounds avoirdupois of fecula, or more than eight thousand pounds yearly. history of the indian archipelago volume 1 pages 387 and 393. this produce is triple that of corn, and double that of potatoes in france. but the plantain produces, on the same surface of land, still more alimentary substance than the sago-tree.); it yields the flour of which the yuruma bread is made; and far from being a palm-tree of the shore, like the chamaerops humilis, the common cocoa-tree, and the lodoicea of commerson, is found as a palm-tree of the marshes as far as the sources of the orinoco.* in the season of inundations these clumps of mauritia, with their leaves in the form of a fan, have the appearance of a forest rising from the bosom of the waters. the navigator, in proceeding along the channels of the delta of the orinoco at night, sees with surprise the summit of the palm-trees illumined by large fires. these are the habitations of the guaraons (tivitivas and waraweties of raleigh*), which are suspended from the trunks of trees. these tribes hang up mats in the air, which they fill with earth, and kindle, on a layer of moist clay, the fire necessary for their household wants. they have owed their liberty and their political independence for ages to the quaking and swampy soil, which they pass over in the time of drought, and on which they alone know how to walk in security to their solitude in the delta of the orinoco; to their abode on the trees where religious enthusiasm will probably never lead any american stylites.* i have already mentioned in another place that the mauritia palm-tree, the tree of life of the missionaries, not only affords the guaraons a safe dwelling during the risings of the orinoco, but that its shelly fruit, its farinaceous pith, its juice, abounding in saccharine matter, and the fibres of its petioles, furnish them with food, wine,* and thread proper for making cords and weaving hammocks. these customs of the indians of the delta of the orinoco were found formerly in the gulf of darien (uraba), and in the greater part of the inundated lands between the guarapiche and the mouths of the amazon. it is curious to observe in the lowest degree of human civilization the existence of a whole tribe depending on one single species of palm-tree, similar to those insects which feed on one and the same flower, or on one and the same part of a plant.

[* the nutritious fecula or medullary flour of the sago-trees is found principally in a group of palms which m. kunth has distinguished by the name of calameae. it is collected, however, in the indian archipelago, as an article of trade, from the trunks of the cycas revoluta, the phoenix farinifera, the corypha umbraculifera, and the caryota urens. (ainslie, materia medica of hindostan, madras 1813.)]

[* i dwell much on these divisions of the great and fine families of palms according to the distribution of the species: first, in dry places, or inland plains, corypha tectorum; second, on the sea-coast, chamaerops humilis, cocos nucifera, corypha maritima, lodoicea seychellarum, labill.; third, in the fresh-water marshes, sagus rumphii, mauritia flexuosa; and 4th, in the alpine regions, between seven and fifteen hundred toises high, ceroxylon andicola, oreodoxa frigida, kunthia montana. this last group of palmae montanae, which rises in the andes of guanacas nearly to the limit of perpetual snow, was, i believe, entirely unknown before our travels in america. (nov. gen. volume 1 page 317; semanario de santa fe de bogota 1819 number 21 page 163.)]

[* the indian name of the tribe of uaraus (guaraunos of the spaniards) may be recognized in the warawety (ouarauoty) of raleigh, one of the branches of the tivitivas. see discovery of guiana, 1576 page 90 and the sketch of the habitations of the guaraons, in raleghi brevis descrip. guianae, 1594 tab 4.]

[* this sect was founded by simeon sisanites, a native of syria. he passed thirty-seven years in mystic contemplation, on five pillars, the last of which was thirty-six cubits high. the sancti columnares attempted to establish their aerial cloisters in the country of treves, in germany; but the bishops opposed these extravagant and perilous enterprises. mosheim, instit. hist. eccles page 192. see humboldt’s views of nature (bohn) pages 13 and 136.]

[* the use of this moriche wine however is not very common. the guaraons prefer in general a beverage of fermented honey.]

the navigation of the river, whether vessels arrive by the boca de navios, or risk entering the labyrinth of the bocas chicas, requires various precautions, according as the waters are high or low. the regularity of these periodical risings of the orinoco has been long an object of admiration to travellers, as the overflowings of the nile furnished the philosophers of antiquity with a problem difficult to solve. the orinoco and the nile, contrary to the direction of the ganges, the indus, the rio de la plata, and the euphrates, flow alike from the south toward the north; but the sources of the orinoco are five or six degrees nearer to the equator than those of the nile. observing every day the accidental variations of the atmosphere, we find it difficult to persuade ourselves that in a great space of time the effects of these variations mutually compensate each other: that in a long succession of years the averages of the temperature of the humidity, and of the barometric pressure, differ so little from month to month; and that nature, notwithstanding the multitude of partial perturbations, follows a constant type in the series of meteorological phenomena. great rivers unite in one receptacle the waters which a surface of several thousand square leagues receives. however unequal may be the quantity of rain that falls during several successive years, in such or such a valley, the swellings of rivers that have a very long course are little affected by these local variations. the swellings represent the average of the humidity that reigns in the whole basin; they follow annually the same progression because their commencement and their duration depend also on the mean of the periods, apparently extremely variable, of the beginning and end of the rains in the different latitudes through which the principal trunk and its various tributary streams flow. hence it follows that the periodical oscillations of rivers are, like the equality of temperature of caverns and springs, a sensible indication of the regular distribution of humidity and heat, which takes place from year to year on a considerable extent of land. they strike the imagination of the vulgar; as order everywhere astonishes, when we cannot easily ascend to first causes. rivers that belong entirely to the torrid zone display in their periodical movements that wonderful regularity which is peculiar to a region where the same wind brings almost always strata of air of the same temperature; and where the change of the sun in its declination causes every year at the same period a rupture of equilibrium in the electric intensity, in the cessation of the breezes, and the commencement of the season of rains. the orinoco, the rio magdalena, and the congo or zaire are the only great rivers of the equinoctial region of the globe, which, rising near the equator, have their mouths in a much higher latitude, though still within the tropics. the nile and the rio de la plata direct their course, in the two opposite hemispheres, from the torrid zone towards the temperate.*

[* in asia, the ganges, the burrampooter, and the majestic rivers of indo–china direct their course towards the equator. the former flow from the temperate to the torrid zone. this circumstance of courses pursuing opposite directions (towards the equator, and towards the temperate climates) has an influence on the period and the height of the risings, on the nature and variety of the productions on the banks of the rivers, on the less or greater activity of trade; and, i may add, from what we know of the nations of egypt, merce, and india, on the progress of civilization along the valleys of the rivers.]

as long as, confounding the rio paragua of esmeralda with the rio guaviare, the sources of the orinoco were sought towards the south-west, on the eastern back of the andes, the risings of this river were attributed to a periodical melting of the snows. this reasoning was as far from the truth as that in which the nile was formerly supposed to be swelled by the waters of the snows of abyssinia. the cordilleras of new grenada, near which the western tributary streams of the orinoco, the guaviare, the meta, and the apure take their rise, enter no more into the limit of perpetual snows, with the sole exception of the paramos of chita and mucuchies, than the alps of abyssinia. snowy mountains are much more rare in the torrid zone than is generally admitted; and the melting of the snows, which is not copious there at any season, does not at all increase at the time of the inundations of the orinoco.

the cause of the periodical swellings of the orinoco acts equally on all the rivers that take rise in the torrid zone. after the vernal equinox, the cessation of the breezes announces the season of rains. the increase of the rivers (which may be considered as natural pluviometers) is in proportion to the quantity of water that falls in the different regions. this quantity, in the centre of the forests of the upper orinoco and the rio negro, appeared to me to exceed 90 or 100 inches annually. such of the natives, therefore, as have lived beneath the misty sky of the esmeralda and the atabapo, know, without the smallest notion of natural philosophy, what eudoxus and eratosthenes knew heretofore,* that the inundations of the great rivers are owing solely to the equatorial rains. the following is the usual progress of the oscillations of the orinoco. immediately after the vernal equinox (the people say on the 25th of march) the commencement of the rising is perceived. it is at first only an inch in twenty-four hours; sometimes the river again sinks in april; it attains its maximum in july; remains at the same level from the end of july till the 25th of august; and then decreases progressively, but more slowly than it increased. it is at its minimum in january and february. in both worlds the rivers of the northern torrid zone attain the greatest height nearly at the same period. the ganges, the niger, and the gambia reach the maximum, like the orinoco, in the month of august.* the nile is two months later, either on account of some local circumstances in the climate of abyssinia, or of the length of its course, from the country of berber, or 17.5° of latitude, to the bifurcation of the delta. the arabian geographers assert that in sennaar and in abyssinia the nile begins to swell in the month of april (nearly as the orinoco); the rise, however, does not become sensible at cairo till toward the summer solstice; and the water attains its greatest height at the end of the month of september.* the river keeps at the same level till the middle of october; and is at its minimum in april and may, a period when the rivers of guiana begin to swell anew. it may be seen from this rapid statement, that, notwithstanding the retardation caused by the form of the natural channels, and by local climatic circumstances, the great phenomenon of the oscillations of the rivers of the torrid zone is everywhere the same. in the two zodiacs vulgarly called the tartar and chaldean, or egyptian (in the zodiac which contains the sign of the rat, an in that which contains those of the fishes and aquarius), particular constellations are consecrated to the periodical overflowings of the rivers. real cycles, divisions of time, have been gradually transformed into divisions of space; but the generality of the physical phenomena of the risings seems to prove that the zodiac which has been transmitted to us by the greeks, and which, by the precession of the equinoxes, becomes an historical monument of high antiquity, may have taken birth far from thebes, and from the sacred valley of the nile. in the zodiacs of the new world — in the mexican, for instance, of which we discover the vestiges in the signs of the days, and the periodical series which they compose — there are also signs of rain and of inundation corresponding to the chou (rat) of the chinese* and thibetan cycle of tse, and to the fishes and aquarius of the dodecatemorion. these two mexican signs are water (atl) and cipactli, the sea-monster furnished with a horn. this animal is at once the antelope-fish of the hindoos, the capricorn of our zodiac, the deucalion of the greeks, and the noah (coxcox) of the azteks.* thus we find the general results of comparative hydrography in the astrological monuments, the divisions of time and the religious traditions of nations the most remote from each other in their situation and in their degree of intellectual advancement.

[* strabo lib. 17 page 789. diod. sic. lib. l c. 5.]

[* nearly forty or fifty days after the summer solstice.]

[* nearly eighty or ninety days after the summer solstice.]

[* the figure of water itself is often substituted for that of the rat (arvicola) in the tartar zodiac. the rat takes the place of aquarius. gaubil, obs. mathem. volume 3 page 33.]

[* coxcox bears also the denomination of teo–cipactli, in which the root god or divine is added to the name of the sign cipactli. it is the man of the fourth age; who, at the fourth destruction of the world (the last renovation of nature), saved himself with his wife, and reached the mountain of colhuacan. according to the commentator germanicus, deucalion was placed in aquarius; but the three signs of the fishes, aquarius and capricorn (the antelope-fish) were heretofore intimately linked together. the animal, which, after having long inhabited the waters, takes the form of an antelope, and climbs the mountains, reminds people, whose restless imagination seizes the most remote similitudes, of the ancient traditions of menou, of noah, and of those deucalions celebrated among the scythians and the thessalians. as the tartarian and mexican zodiacs contain the signs of the monkey and the tiger, they, no doubt, originated in the torrid zone. with the muyscas, inhabitants of new grenada, the first sign, as in eastern asia, was that of water, figured by a frog. it is also remarkable that the astrological worship of the muyscas came to the table-land of bogota from the eastern side, from the plains of san juan, which extend toward the guaviare and the orinoco.]

as the equatorial rains take place in the flat country when the sun passes through the zenith of the place, that is, when its declination becomes homonymous with the zone comprised between the equator and one of the tropics, the waters of the amazon sink, while those of the orinoco rise perceptibly. in a very judicious discussion on the origin of the rio congo,* the attention of philosophers has been already called to the modifications which the periods of the risings must undergo in the course of a river, the sources and the mouth of which are not on the same side of the equinoctial line.* the hydraulic systems of the orinoco and the amazon furnish a combination of circumstances still more extraordinary. they are united by the rio negro and the cassiquiare, a branch of the orinoco; it is a navigable line, between two great basins of rivers, that is crossed by the equator. the river amazon, according to the information which i obtained on its banks, is much less regular in the periods of its oscillations than the orinoco; it generally begins, however, to increase in december, and attains its maximum of height in march.* it sinks from the month of may, and is at its minimum of height in the months of july and august, at the time when the lower orinoco inundates all the surrounding land. as no river of america can cross the equator from south to north, on account of the general configuration of the ground, the risings of the orinoco have an influence on the amazon; but those of the amazon do not alter the progress of the oscillations of the orinoco. it results from these data, that in the two basins of the amazon and the orinoco, the concave and convex summits of the curve of progressive increase and decrease correspond very regularly with each other, since they exhibit the difference of six months, which results from the situation of the rivers in opposite hemispheres. the commencement of the risings only is less tardy in the orinoco. this river increases sensibly as soon as the sun has crossed the equator; in the amazon, on the contrary, the risings do not commence till two months after the equinox. it is known that in the forests north of the line the rains are earlier than in the less woody plains of the southern torrid zone. to this local cause is joined another, which acts perhaps equally on the tardy swellings of the nile. the amazon receives a great part of its waters from the cordillera of the andes, where the seasons, as everywhere among mountains, follow a peculiar type, most frequently opposite to that of the low regions.

[* voyage to the zaire page 17.]

[* among the rivers of america this is the case with the rio negro, the rio branco, and the jupura.]

[* nearly seventy or eighty days after our winter solstice, which is the summer solstice of the southern hemisphere.]

the law of the increase and decrease of the orinoco is more difficult to determine with respect to space, or to the magnitude of the oscillations, than with regard to time, or the period of the maxima and minima. having been able to measure but imperfectly the risings of the river, i report, not without hesitation, estimates that differ much from each other.* foreign pilots admit ninety feet for the ordinary rise in the lower orinoco. m. depons, who has in general collected very accurate notions during his stay at caracas, fixes it at thirteen fathoms. the heights naturally vary according to the breadth of the bed and the number of tributary streams which the principal trunk receives.

[* tuckey, maritime geogr. volume 4 page 309. hippisley, expedition to the orinoco page 38. gumilla volume 1 pages 56 to 59. depons volume 3 page 301. the greatest height of the rise of the mississippi is, at natchez, fifty-five english feet. this river (the largest perhaps of the whole temperate zone) is at its maximum from february to may; at its minimum in august and september. ellicott, journal of an expedition to the ohio.]

the people believe that every five years the orinoco rises three feet higher than common; but the idea of this cycle does not rest on any precise measures. we know by the testimony of antiquity, that the oscillations of the nile have been sensibly the same with respect to their height and duration for thousands of years; which is a proof, well worthy of attention, that the mean state of the humidity and the temperature does not vary in that vast basin. will this constancy in physical phenomena, this equilibrium of the elements, be preserved in the new world also after some ages of cultivation? i think we may reply in the affirmative; for the united efforts of man cannot fail to have an influence on the general causes on which the climate of guiana depends.

according to the barometric height of san fernando de apure, i find from that town to the boca de navios the slope of the apure and the lower orinoco to be three inches and a quarter to a nautical mile of nine hundred and fifty toises.* we may be surprised at the strength of the current in a slope so little perceptible; but i shall remind the reader on this occasion, that, according to measurements made by order of mr. hastings, the ganges was found, in a course of sixty miles (comprising the windings,) to have also only four inches fall to a mile; that the mean swiftness of this river is, in the seasons of drought, three miles an hour, and in those of rains six or eight miles. the strength of the current, therefore, in the ganges as in the orinoco, depends less on the slope of the bed, than on the accumulation of the higher waters, caused by the abundance of the rains, and the number of tributary streams. european colonists have already been settled for two hundred and fifty years on the banks of the orinoco; and during this long period of time, according to a tradition which has been propagated from generation to generation, the periodical oscillations of the river (the time of the beginning of the rising, and that when it attains its maximum) have never been retarded more than twelve or fifteen days.

[* the apure itself has a slope of thirteen inches to the mile.]

when vessels that draw a good deal of water sail up toward angostura in the months of january and february, by favour of the sea-breeze and the tide, they run the risk of taking the ground. the navigable channel often changes its breadth and direction; no buoy, however, has yet been laid down, to indicate any deposit of earth formed in the bed of the river, where the waters have lost their original velocity. there exists on the south of cape barima, as well by the river of this name as by the rio moroca and several estuaries (esteres) a communication with the english colony of essequibo. small vessels can penetrate into the interior as far as the rio poumaron, on which are the ancient settlements of zealand and middleburg. heretofore this communication interested the government of caracas only on account of the facility it furnished to an illicit trade; but since berbice, demerara, and essequibo have fallen into the hands of a more powerful neighbour, it fixes the attention of the spanish americans as being connected with the security of their frontiers. rivers which have a course parallel to the coast, and are nowhere farther distant from it than five or six nautical miles, characterize the whole of the shore between the orinoco and the amazon.

ten leagues distant from cape barima, the great bed of the orinoco is divided for the first time into two branches of two thousand toises in breadth. they are known by the indian names of zacupana and imataca. the first, which is the northernmost, communicates on the west of the islands congrejos and del burro with the bocas chicas of lauran, nuina, and mariusas. as the isla del burro disappears in the time of great inundations, it is unhappily not suited to fortifications. the southern bank of the brazo imataca is cut by a labyrinth of little channels, into which the rio imataca and the rio aquire flow. a long series of little granitic hills rises in the fertile savannahs between the imataca and the cuyuni; it is a prolongation of the cordilleras of parima, which, bounding the horizon south of angostura, forms the celebrated cataracts of the rio caroni, and approaches the orinoco like a projecting cape near the little fort of vieja guyana. the populous missions of the caribbee and guiana indians, governed by the catalonian capuchins, lie near the sources of the imataca and the aquire. the easternmost of these missions are those of miamu, camamu, and palmar, situate in a hilly country, which extends towards tupuquen, santa maria, and the villa de upata. going up the rio aquire, and directing your course across the pastures towards the south, you reach the mission of belem de tumeremo, and thence the confluence of the curumu with the rio cuyuni, where the spanish post or destacamento de cuyuni was formerly established. i enter into this topographical detail because the rio cuyuni, or cuduvini, runs parallel to the orinoco from west to east, through an extent of 2.5 or 3° of longitude,* and furnishes an excellent natural boundary between the territory of caracas and that of english guiana.

[* including the rio juruam, one of the principal branches of the cuyuni. the dutch military post is five leagues west of the union of cuyuni with the essequibo, where the former river receives the mazuruni.]

the two great branches of the orinoco, the zacupana and the imataca, remain separate for fourteen leagues: on going up farther, the waters of the river are found united* in a single channel extremely broad. this channel is near eight leagues long; at its western extremity a second bifurcation appears; and as the summit of the delta is in the northern branch of the bifurcated river, this part of the orinoco is highly important for the military defence of the country. all the channels* that terminate in the bocas chicas, rise from the same point of the trunk of the orinoco. the branch (cano manamo) that separates from it near the village of san rafael has no ramification till after a course of three or four leagues; and by placing a small fort above the island of chaguanes, angostura might be defended against an enemy that should attempt to penetrate by one of the bocas chicas. in my time the station of the gun-boats was east of san rafael, near the northern bank of the orinoco. this is the point which vessels must pass in sailing up toward angostura by the northern channel, that of san rafael, which is the broadest but the most shallow.

[* at this point of union are found two villages of guaraons. they also bear the names of imataca and zacupana.]

[* cano de manamo grande, cano de manamo chico, cano pedernales, cano macareo, cano cutupiti, cano macuona, cano grande de mariusas, etc. the last three branches form by their union the sinuous channel called the vuelta del torno.]

six leagues above the point where the orinoco sends off a branch to the bocas chicas is placed an ancient fort (los castillos de la vieja or antigua guayana,) the first construction of which goes back to the sixteenth century. in this spot the bed of the river is studded with rocky islands; and it is asserted that its breadth is nearly six hundred and fifty toises. the town is almost destroyed, but the fortifications subsist, and are well worthy the attention of the government of terra firma. there is a magnificent view from the battery established on a bluff north-west of the ancient town, which, at the period of great inundations, is entirely surrounded with water. pools that communicate with the orinoco form natural basins, adapted for the reception of vessels that want repairs.

after having passed the little forts of vieja guayana, the bed of the orinoco again widens. the state of cultivation of the country on the two banks affords a striking contrast. on the north is seen the desert part of the province of cumana, steppes (llanos) destitute of habitations, and extending beyond the sources of the rio mamo, toward the tableland or mesa of guanipa. on the south we find three populous villages belonging to the missions of carony, namely, san miguel de uriala, san felix and san joaquin. the last of these villages, situate on the banks of the carony, immediately below the great cataract, is considered as the embarcadero of the catalonian missions. on navigating more to the east, between the mouth of the carony and angostura, the pilot should avoid the rocks of guarampo, the sandbank of mamo, and the piedra del rosario. from the numerous materials which i brought home, and from astronomical discussions, the principal results of which i have indicated above, i have constructed a map of the country bounded by the delta of the orinoco, the carony, and the cuyuni. this part of guiana, from its proximity to the coast, will some day offer the greatest attraction to european settlers.

the whole population of this vast province in its present state is, with the exception of a few spanish parishes, scattered on the banks of the lower orinoco, and subject to two monastic governments. estimating the number of the inhabitants of guiana, who do not live in savage independence, at thirty-five thousand, we find nearly twenty-four thousand settled in the missions, and thus withdrawn as it were from the direct influence of the secular arm. at the period of my voyage, the territory of the observantin monks of st. francis contained seven thousand three hundred inhabitants, and that of the capuchinos catalanes seventeen thousand; an astonishing disproportion, when we reflect on the smallness of the latter territory compared to the vast banks of the upper orinoco, the atabapo, the cassiquiare and the rio negro. it results from these statements that nearly two-thirds of the population of a province of sixteen thousand eight hundred square leagues are found concentrated between the rio imataca and the town of santo thome del angostura, on a space of ground only fifty-five leagues in length, and thirty in breadth. both of these monastic governments are equally inaccessible to whites, and form status in statu. the first, that of the observantins, i have described from my own observations; it remains for me to record here the notions i could procure respecting the second of these governments, that of the catalonian capuchins. fatal civil dissensions and epidemic fevers have of late years diminished the long-increasing prosperity of the missions of the carony; but, notwithstanding these losses, the region which we are going to examine is still highly interesting with respect to political economy.

the missions of the catalonian capuchins, which in 1804 contained at least sixty thousand head of cattle grazing in the savannahs, extend from the eastern banks of the carony and the paragua as far as the banks of the imataca, the curumu, and the cuyuni; at the south-east they border on english guiana, or the colony of essequibo; and toward the south, in going up the desert banks of the paragua and the paraguamasi, and crossing the cordillera of pacaraimo, they touch the portuguese settlements on the rio branco. the whole of this country is open, full of fine savannahs, and no way resembling that through which we passed on the upper orinoco. the forests become impenetrable only on advancing toward the south; on the north are meadows intersected with woody hills. the most picturesque scenes lie near the falls of the carony, and in that chain of mountains, two hundred and fifty toises high, which separates the tributary streams of the orinoco from those of the cuyuni. there are situate the villa de upata,* the capital of the missions, santa maria, and cupapui. small table-lands afford a healthy and temperate climate. cacao, rice, cotton, indigo, and sugar grow in abundance wherever a virgin soil, covered with a thick coat of grasses, is subjected to cultivation. the first christian settlements in those countries are not, i believe, of an earlier date than 1721. the elements of which the present population is composed are the three indian races of the guayanos, the caribs and the guaycas. the last are a people of mountaineers and are far from being so diminutive in size as the guaycas whom we found at esmeralda. it is difficult to fix them to the soil; and the three most modern missions in which they have been collected, those of cura, curucuy, and arechica, are already destroyed. the guayanos, who early in the sixteenth century gave their name to the whole of that vast province, are less intelligent but milder; and more easy, if not to civilize, at least to subjugate, than the caribs. their language appears to belong to the great branch of the caribbee and tamanac tongues. it displays the same analogies of roots and grammatical forms, which are observed between the sanscrit, the persian, the greek, and the german. it is not easy to fix the forms of what is indefinite by its nature; and to agree on the differences which should be admitted between dialects, derivative languages and mother-tongues. the jesuits of paraguay have made known to us another tribe of guayanos* in the southern hemisphere, living in the thick forests of parana. though it cannot be denied in general that in consequence of distant migrations,* the nations that are settled north and south of the amazon have had communications with each other, i will not decide whether the guayanos of parana and of uruguay exhibit any other relation to those of carony, than that of an homonomy, which is perhaps only accidental.

[* founded in 1762. population in 1797, 657 souls; in 1803, 769 souls. the most populous villages of these missions, alta gracia, cupapui, santa rosa de cura, and guri, had between 600 and 900 inhabitants in 1797; but in 1818 epidemic fevers diminished the population more than a third. in some missions these diseases have swept away nearly half of the inhabitants.]

[* they are also called guananas, or gualachas.]

[* like the celebrated migrations of the omaguas, or omeguas.]

the most considerable christian settlements are now concentrated between the mountains of santa maria, the mission of san miguel and the eastern bank of the carony, from san buenaventura as far as guri and the embarcadero of san joaquin; a space of ground which has not more than four hundred and sixty square leagues of surface. the savannahs to the east and south are almost uninhabited; we find there only the solitary missions of belem, tumuremo, tupuquen, puedpa, and santa clara. it were to be wished that the spots preferred for cultivation were distant from the rivers where the land is higher and the air more favourable to health. the rio carony, the waters of which, of an admirable clearness, are not well stocked with fish, is free from shoals from the villa de barceloneta, a little above the confluence of the paragua, as far as the village of guri. farther north it winds between innumerable islands and rocks; and only the small boats of the caribs venture to navigate amid these raudales, or rapids of the carony. happily the river is often divided into several branches; and consequently that can be chosen which, according to the height of the waters, presents the fewest whirlpools and shoals. the great fall, celebrated for the picturesque beauty of its situation, is a little above the village of aguacaqua, or carony, which in my time had a population of seven hundred indians. this cascade is said to be from fifteen to twenty feet high; but the bar does not cross the whole bed of the river, which is more than three hundred feet broad. when the population is more extended toward the east, it will avail itself of the course of the small rivers imataca and aquire, the navigation of which is pretty free from danger. the monks, who like to keep themselves isolated, in order to withdraw from the eye of the secular power, have been hitherto unwilling to settle on the banks of the orinoco. it is, however, by this river only, or by the cuyuni and the essequibo, that the missions of carony can export their productions. the latter way has not yet been tried, though several christian settlements* are formed on one of the principal tributary streams of the cuyuni, the rio juruario. this stream furnishes, at the period of the great swellings, the remarkable phenomenon of a bifurcation. it communicates by the juraricuima and the aurapa with the rio carony; so that the land comprised between the orinoco, the sea, the cuyuni, and the carony, becomes a real island. formidable rapids impede the navigation of the upper cuyuni; and hence of late an attempt has been made to open a road to the colony of essequibo much more to the south-east, in order to fall in with the cuyuni much below the mouth of the curumu.

[* guacipati, tupuquen, angel de la custodia, and cura, where the military post of the frontiers was stationed in 1800, which had been anciently placed at the confluence of the cuyuni and the curumu.]

the whole of this southern territory is traversed by hordes of independent caribs; the feeble remains of that warlike people who were so formidable to the missionaries till 1733 and 1735, at which period the respectable bishop gervais de labrid,* canon of the metropolitan chapter of lyon, father lopez, and several other ecclesiastics, perished by the hands of the caribs. these dangers, too frequent formerly, exist no longer, either in the missions of carony, or in those of the orinoco; but the independent caribs continue, on account of their connection with the dutch colonists of essequibo, an object of mistrust and hatred to the government of guiana. these tribes favour the contraband trade along the coast, and by the channels or estuaries that join the rio barima to the rio moroca; they carry off the cattle belonging to the missionaries, and excite the indians recently converted, and living within the sound of the bell, to return to the forests. the free hordes have everywhere a powerful interest in opposing the progress of cultivation and the encroachments of the whites. the caribs and the aruacas procure fire-arms at essequibo and demerara; and when the traffic of american slaves (poitos) was most active, adventurers of dutch origin took part in these incursions on the paragua, the erevato, and the ventuario. man-hunting took place on these banks, as heretofore (and probably still) on those of the senegal and the gambia. in both worlds europeans have employed the same artifices, and committed the same atrocities, to maintain a trade that dishonours humanity. the missionaries of the carony and the orinoco attribute all the evils they suffer from the independent caribs to the hatred of their neighbours, the calvinist preachers of essequibo. their works are therefore filled with complaints of the secta diabolica de calvino y de lutero, and against the heretics of dutch guiana, who also think fit sometimes to go on missions, and spread the germs of social life among the savages.

[* consecrated a bishop for the four parts of the world (obispo para las quatro partes del mundo) by pope benedict xiii.]

of all the vegetable productions of those countries, that which the industry of the catalonian capuchins has rendered the most celebrated is the tree that furnishes the cortex angosturae, which is erroneously designated by the name of cinchona of carony. we were fortunate enough to make it first known as a new genus distinct from the cinchona, and belonging to the family of meliaceae, or of zanthoxylus. this salutary drug of south america was formerly attributed to the brucea ferruginea which grows in abyssinia, to the magnolia glauca, and to the magnolia plumieri. during the dangerous disease of m. bonpland, m. ravago sent a confidential person to the missions of carony, to procure for us, by favour of the capuchins of upata, branches of the tree in flower which we wished to be able to describe. we obtained very fine specimens, the leaves of which, eighteen inches long, diffused an agreeable aromatic smell. we soon perceived that the cuspare (the indigenous name of the cascarilla or corteza del angostura) forms a new genus; and on sending the plants of the orinoco to m. willdenouw, i begged he would dedicate this plant to m. bonpland. the tree, known at present by the name of bonplandia trifoliata, grows at the distance of five or six leagues from the eastern bank of the carony, at the foot of the hills that surround the missions capapui, upata and alta gracia. the caribbee indians make use of an infusion of the bark of the cuspare, which they consider as a strengthening remedy. m. bonpland discovered the same tree west of cumana, in the gulf of santa fe, where it may become one of the articles of exportation from new andalusia.

the catalonian monks prepare an extract of the cortex angosturae which they send to the convents of their province, and which deserves to be better known in the north of europe. it is to be hoped that the febrifuge and anti-dysenteric bark of the bonplandia will continue to be employed, notwithstanding the introduction of another, described by the name of false angostura bark, and often confounded with the former. this false angostura, or angostura pseudo-ferruginea, comes, it is said, from the brucea antidysenterica; it acts powerfully on the nerves, produces violent attacks of tetanus, and contains, according to the experiments of pelletier and caventon, a peculiar alkaline substance* analogous to morphine and strychnine. as the tree which yields the real cortex angosturae does not grow in great abundance, it is to be wished that plantations of it were formed. the catalonian monks are well fitted to spread this kind of cultivation; they are more economical, industrious, and active than the other missionaries. they have already established tan-yards and cotton-spinning in a few villages; and if they suffer the indians henceforth to enjoy the fruit of their labours, they will find great resources in the native population. concentered on a small space of land, these monks have the consciousness of their political importance, and have from time to time resisted the civil authority, and that of their bishop. the governors who reside at angostura have struggled against them with very unequal success, according as the ministry of madrid showed a complaisant deference for the ecclesiastical hierarchy, or sought to limit its power. in 1768 don manuel centurion carried off twenty thousand head of cattle from the missionaries, in order to distribute them among the indigent inhabitants. this liberality, exerted in a manner not very legal, produced very serious consequences. the governor was disgraced on the complaint of the catalonian monks though he had considerably extended the territory of the missions toward the south, and founded the villa de barceloneta, above the confluence of the carony with the rio paragua, and the ciudad de guirior, near the union of the rio paragua and the paraguamusi. from that period the civil administration has carefully avoided all intervention in the affairs of the capuchins, whose opulence has been exaggerated like that of the jesuits of paraguay.

[* brucine. m. pelletier has wisely avoided using the word angosturine, because it might indicate a substance taken from the real cortex angosturae, or bonplandia trifoliata. (annales de chimie volume 12 page 117.) we saw at peru the barks of two new species of weinmannia and wintera mixed with those of cinchona; a mixture less dangerous, but still injurious, on account of the superabundance of tannin and acrid matter contained in the false cascarilla.]

the missions of the carony, by the configuration of their soil* and the mixture of savannahs and arable lands, unite the advantages of the llanos of calabozo and the valleys of aragua. the real wealth of this country is founded on the care of the herds and the cultivation of colonial produce. it were to be wished that here, as in the fine and fertile province of venezuela, the inhabitants, faithful to the labours of the fields, would not addict themselves too hastily to the research of mines. the example of germany and mexico proves, no doubt, that the working of metals is not at all incompatible with a flourishing state of agriculture; but, according to popular traditions, the banks of the carony lead to the lake dorado and the palace of the gilded man*: and this lake, and this palace, being a local fable, it might be dangerous to awaken remembrances which begin gradually to be effaced. i was assured that in 1760, the independent caribs went to cerro de pajarcima, a mountain to the south of vieja guayana, to submit the decomposed rock to the action of washing. the gold-dust collected by this labour was put into calabashes of the crescentia cujete and sold to the dutch at essequibo. still more recently, some mexican miners, who abused the credulity of don jose avalo, the intendant of caracas, undertook a very considerable work in the centre of the missions of the rio carony, near the town of upata, in the cerros del potrero and de chirica. they declared that the whole rock was auriferous; stamping-mills, brocards, and smelting-furnaces were constructed. after having expended very large sums, it was discovered that the pyrites contained no trace whatever of gold. these essays, though fruitless, served to renew the ancient idea that every shining rock in guiana is teeming with gold (una madre del oro). not contented with taking the mica-slate to the furnace, strata of amphibolic slates were shown to me near angostura, without any mixture of heterogeneous substances, which had been worked under the whimsical name of black ore of gold (oro negro).

[* it appears that the little table-lands between the mountains of upata, cumanu, and tupuquen, are more than one hundred and fifty toises above the level of the sea.]

[* el dorado, that is, el rey o hombre dorado. see volume 2.23.]

this is the place to make known, in order to complete the description of the orinoco, the principal results of my researches on el dorado, the white sea, or laguna parime, and the sources of the orinoco, as they are marked in the most recent maps. the idea of an auriferous earth, eminently rich, has been connected, ever since the end of the sixteenth century, with that of a great inland lake, which furnishes at the same time waters to the orinoco, the rio branco and the rio essequibo. i believe, from a more accurate knowledge of the country, a long and laborious study of the spanish authors who treat of el dorado, and, above all, from comparing a great number of ancient maps, arranged in chronological order, i have succeeded in discovering the source of these errors. all fables have some real foundation; that of el dorado resembles those myths of antiquity, which, travelling from country to country, have been successively adapted to different localities. in the sciences, in order to distinguish truth from error, it often suffices to retrace the history of opinions, and to follow their successive developments. the discussion to which i shall devote the end of this chapter is important, not only because it throws light on the events of the conquest, and that long series of disastrous expeditions made in search of el dorado, the last of which was in the year 1775; it also furnishes, in addition to this simply historical interest, another, more substantial and more generally felt, that of rectifying the geography of south america, and of disembarrassing the maps published in our days of those great lakes, and that strange labyrinth of rivers, placed as if by chance between sixty and sixty-six degrees of longitude. no man in europe believes any longer in the wealth of guiana and the empire of the grand patiti. the town of manoa and its palaces covered with plates of massy gold have long since disappeared; but the geographical apparatus serving to adorn the fable of el dorado, the lake parima, which, similar to the lake of mexico, reflected the image of so many sumptuous edifices, has been religiously preserved by geographers. in the space of three centuries, the same traditions have been differently modified; from ignorance of the american languages, rivers have been taken for lakes, and portages for branches of rivers; one lake, the cassipa, has been made to advance five degrees of latitude toward the south, while another, the parima or dorado, has been transported the distance of a hundred leagues from the western to the eastern bank of the rio branco. from these various changes, the problem we are going to solve has become much more complicated than is generally supposed. the number of geographers who discuss the basis of a map, with regard to the three points of measures, of the comparison of descriptive works, and of the etymological study* of names, is extremely small. almost all the maps of south america which have appeared since the year 1775 are, in what regards the interior of the country, comprised between the steppes of venezuela and the river of the amazons, between the eastern back of the andes and the coast of cayenne, a simple copy of the great spanish map of la cruz olmedilla. a line, indicating the extent of country which don jose solano boasted of having discovered and pacified by his troops and emissaries, was taken for the road followed by that officer, who never went beyond san fernando de atabapo, a village one hundred and sixty leagues distant from the pretended lake parima. the study of the work of father caulin, who was the historiographer of the expedition of solano, and who states very clearly, from the testimony of the indians, how the name of the river parima gave rise to the fable of el dorado, and of an inland sea, has been neglected. no use either has been made of a map of the orinoco, three years posterior to that of la cruz, and traced by surville from the collection of true or hypothetical materials preserved in the archives of the despacho universal de indias. the progress of geography, as manifested on our maps, is much slower than might be supposed from the number of useful results which are found scattered in the works of different nations. astronomical observations and topographic information accumulate during a long lapse of years, without being made use of; and from a principle of stability and preservation, in other respects praiseworthy, those who construct maps often choose rather to add nothing, than to sacrifice a lake, a chain of mountains, or an interbranching of rivers, which have figured there during ages.

[* i use this expression, perhaps an improper one, to mark a species of philological examination, to which the names of rivers, lakes, mountains, and tribes, must be subjected, in order to discover their identity in a great number of maps. the apparent diversity of names arises partly from the difference of the dialects spoken by one and the same family of people, partly from the imperfection of our european orthography, and from the extreme negligence with which geographers copy one another. we recognize with difficulty the rio uaupe in the guaupe or guape; the xie, in the guaicia; the raudal de atures, in athule; the caribbees, in the calinas and galibis; the guaraunos or uarau, in the oaraw-its; etc. it is, however, by similar mutations of letters, that the spaniards have made hijo of filius; hambre, of fames; and felipo de urre, and even utre, of the conquistador philip von huten; that the tamanacs in america have substituted choraro for soldado; and the jews in china, ialemeiohang for jeremiah. analogy and a certain etymological tact must guide geographers in researches of this kind, in which they would be exposed to serious errors, if they were not to study at the same time the respective situations of the upper and lower tributary streams of the same river. our maps of america are overloaded with names, for which rivers have been created. this desire of compiling, of filling up vacancies, and of employing, without investigation, heterogeneous materials, has given our maps of countries the least visited an appearance of exactness, the falsity of which is discovered when we arrive on the spot.]

the fabulous traditions of el dorado and the lake parima having been diversely modified according to the aspect of the countries to which they were to be adapted, we must distinguish what they contain that is real from what is merely imaginary. to avoid entering here into minute particulars, i shall begin first to call the attention of the reader to those spots which have been, at various periods, the theatre of the expeditions undertaken for the discovery of el dorado. when we have learnt to know the aspect of the country, and the local circumstances, such as they can now be described, it will be easy to conceive how the different hypotheses recorded on our maps have taken rise by degrees, and have modified each other. to oppose an error, it is sufficient to recall to mind the variable forms in which we have seen it appear at different periods.

till the middle of the eighteenth century, all that vast space of land comprised between the mountains of french guiana and the forests of the upper orinoco, between the sources of the carony and the river amazon (from 0 to 4° of north latitude, and from 57 to 68° of longitude), was so little known that geographers could place in it lakes where they pleased, create communications between rivers, and figure chains of mountains more or less lofty. they have made full use of this liberty; and the situation of lakes, as well as the course and branches of rivers, has been varied in so many ways that it would not be surprising if among the great number of maps some were found that trace the real state of things. the field of hypotheses is now singularly narrowed. i have determined the longitude of esmeralda in the upper orinoco; more to the east amid the plains of parima (a land as unknown as wangara and dar–saley, in africa), a band of twenty leagues broad has been travelled over from north to south along the banks of the rio carony and the rio branco in the longitude of sixty-three degrees. this is the perilous road which was taken by don antonio santos in going from santo thome del angostura to rio negro and the amazon; by this road also the colonists of surinam communicated very recently with the inhabitants of grand para. this road divides the terra incognita of parima into two unequal portions; and fixes limits at the same time to the sources of the orinoco, which it is no longer possible to carry back indefinitely toward the east, without supposing that the bed of the rio branco, which flows from north to south, is crossed by the bed of the upper orinoco, which flows from east to west. if we follow the course of the rio branco, or that strip of cultivated land which is dependent on the capitania general of grand para, we see lakes, partly imaginary and partly enlarged by geographers, forming two distinct groups. the first of these groups includes the lakes which they place between the esmeralda and the rio branco; and to the second belong those that are supposed to lie between the rio branco and the mountains of dutch and french guiana. it results from this sketch that the question whether there exists a lake parima on the east of the rio branco is altogether foreign to the problem of the sources of the orinoco.

beside the country which we have just noticed (the dorado de la parime, traversed by the rio branco), another part of america is found, two hundred and sixty leagues toward the west, near the eastern back of the cordillera of the andes, equally celebrated in the expeditions to el dorado. this is the mesopotamia between the caqueta, the rio negro, the uaupes, and the yurubesh, of which i have already given a particular account; it is the dorado of the omaguas which contains lake manoa of father acunha, the laguna de oro of the guanes and the auriferous land whence father fritz received plates of beaten gold in his mission on the amazon, toward the end of the seventeenth century.

the first and above all the most celebrated enterprises attempted in search of el dorado were directed toward the eastern back of the andes of new grenada. fired with the ideas which an indian of tacunga had given of the wealth of the king or zaque of cundirumarca, sebastian de belalcazar, in 1535, sent his captains anasco and ampudia, to discover the valley of el dorado,* twelve days’ journey from guallabamba, consequently in the mountains between pasto and popayan. the information which pedro de anasco had obtained from the natives, joined to that which was received subsequently (1536) by diaz de pineda, who had discovered the provinces of quixos and canela, between the rio napo and the rio pastaca, gave birth to the idea that on the east of the nevados of tunguragua, cayambe, and popayan, were vast plains, abounding in precious metals, and where the inhabitants were covered with armour of massy gold. gonzales pizarro, in searching for these treasures, discovered accidentally, in 1539, the cinnamon-trees of america (laurus cinnamomoides, mut.); and francisco de orellana went down the napo, to reach the river amazon. since that period expeditions were undertaken at the same time from venezuela, new grenada, quito, peru, and even from brazil and the rio de la plata,* for the conquest of el dorado. those of which the remembrance have been best preserved, and which have most contributed to spread the fable of the riches of the manaos, the omaguas, and the guaypes, as well as the existence of the lagunas de oro, and the town of the gilded king (grand patiti, grand moxo, grand paru, or enim), are the incursions made to the south of the guaviare, the rio fragua, and the caqueta. orellana, having found idols of massy gold, had fixed men’s ideas on an auriferous land between the papamene and the guaviare. his narrative, and those of the voyages of jorge de espira (george von speier), hernan perez de quesada, and felipe de urre (philip von huten), undertaken in 1536, 1542, and 1545, furnish, amid much exaggeration, proofs of very exact local knowledge.* when these are examined merely in a geographical point of view, we perceive the constant desire of the first conquistadores to reach the land comprised between the sources of the rio negro, of the uaupes (guape), and of the jupura or caqueta. this is the land which, in order to distinguish it from el dorado de la parime, we have called el dorado des omaguas.* no doubt the whole country between the amazon and the orinoco was vaguely known by the name of las provincias del dorado; but in this vast extent of forests, savannahs, and mountains, the progress of those who sought the great lake with auriferous banks, and the town of the gilded king, was directed towards two points only, on the north-east and south-west of the rio negro; that is, to parima (or the isthmus between the carony, the essequibo, and the rio branco), and to the ancient abode of the manaos, the inhabitants of the banks of the yurubesh. i have just mentioned the situation of the latter spot, which is celebrated in the history of the conquest from 1535 to 1560; and it remains for me to speak of the configuration of the country between the spanish missions of the rio carony, and the portuguese missions of the rio branco or parima. this is the country lying near the lower orinoco, the esmeralda, and french and dutch guiana, on which, since the end of the sixteenth century, the enterprises and exaggerated narratives of raleigh have shed so bright a splendour.

[* el valle del dorado. pineda relates: que mas adelante de la provincia de la canela se hallan tierras muy ricas, adonde andaban los hombres armados de piecas y joyas de oro, y que no havia sierra, ni montana. [beyond the province of canela there are found very rich countries (though without mountains) in which the natives are adorned with trinkets and plates of gold.] herrera dec. 5 lib. 10 cap. 14 and dec. 6 lib. 8 cap. 6 geogr. blaviana volume 11 page 261. southey tome 1 pages 78 and 373.]

[* nuno de chaves went from the ciudad de la asumpcion, situate on rio paraguay, to discover, in the latitude of 24° south, the vast empire of el dorado, which was everywhere supposed to lie on the eastern back of the andes.]

[* we may be surprised to see, that the expedition of huten is passed over in absolute silence by herrera (dec. 7 lib. 10 cap. 7 volume 4 238). fray pedro simon gives the whole particulars of it, true or fabulous; but he composed his work from materials that were unknown to herrera.]

[* in 1560 pedro de ursua even took the title of governador del dorado y de omagua. fray pedro simon volume 6 chapter 10 page 430.]

from the general disposition of the course of the orinoco, directed successively towards the west, the north, and the east, its mouth lies almost in the same meridian as its sources: so that by proceeding from vieja guyana to the south the traveller passes through the whole of the country in which geographers have successively placed an inland sea (mar blanco), and the different lakes which are connected with the el dorado de la parime. we find first the rio carony, which is formed by the union of two branches of almost equal magnitude, the carony properly so called, and the rio paragua. the missionaries of piritu call the latter river a lake (laguna): it is full of shoals, and little cascades; but, passing through a country entirely flat, it is subject at the same time to great inundations, and its real bed (su verdadera caxa) can scarcely be discovered. the natives have given it the name of paragua or parava, which means in the caribbee language sea, or great lake. these local circumstances and this denomination no doubt have given rise to the idea of transforming the rio paragua, a tributary stream of the carony, into a lake called cassipa, on account of the cassipagotos,* who lived in those countries. raleigh gives this basin forty miles in breadth; and, as all the lakes of parima must have auriferous sands, he does not fail to assert that in summer, when the waters retire, pieces of gold of considerable weight are found there.

[* raleigh pages 64 and 69. i always quote, when the contrary is not expressly said, the original edition of 1596. have these tribes of cassipagtos, epuremei, and orinoqueponi, so often mentioned by raleigh, disappeared? or did some misapprehension give rise to these denominations? i am surprised to find the indian words [of one of the different carib dialects?] ezrabeta cassipuna aquerewana, translated by raleigh, the great princes or greatest commander. since acarwana certainly signifies a chief, or any person who commands (raleigh pages 6 and 7), cassipuna perhaps means great, and lake cassipa is synonymous with great lake. in the same manner cass-iquiare may be a great river, for iquiare, like veni, is, an the north of the amazon, a termination common to all rivers. goto, however, in cassipa-goto, is a caribbee term denoting a tribe.]

the sources of the tributary streams of the carony, the arui, and the caura (caroli, arvi, and caora,* of the ancient geographers) being very near each other, this suggested the idea of making all these rivers take their rise from the pretended lake cassipa.* sanson has so much enlarged this lake, that he gives it forty-two leagues in length, and fifteen in breadth. the ancient geographers placed opposite to each other, with very little hesitation, the tributary streams of the two banks of a river; and they place the mouth of the carony, and lake cassipa, which communicates by the carony with the orinoco, sometimes* above the confluence of the meta. thus it is carried back by hondius as far as the latitudes of 2 and 3°, giving it the form of a rectangle, the longest sides of which run from north to south. this circumstance is worthy of remark, because, in assigning gradually a more southern latitude to lake cassipa, it has been detached from the carony and the arui, and has taken the name of parima. to follow this metamorphosis in its progressive development, we must compare the maps which have appeared since the voyage of raleigh till now. la cruz, who has been copied by all the modern geographers, has preserved the oblong form of the lake cassipa for his lake parima, although this form is entirely different from that of the ancient lake parima, or rupunuwini, of which the great axis was directed from east to west. the ancient lake (that of hondius, sanson, and coronelli) was also surrounded by mountains, and gave birth to no river; while the lake parima of la cruz and the modern geographers communicates with the upper orinoco, as the cassipa with the lower orinoco.

[* d’anville names the rio caura, coari; and the rio arui, aroay. i have not been able hitherto to guess what is meant by the aloica (atoca, atoica of raleigh), which issues from the lake cassipa, between the caura and the arui.]

[* raleigh makes only the carony and the arui issue from it (hondius, nieuwe caerte van het wonderbare landt guiana, besocht door sir walter raleigh, 1594 to 1596): but in later maps, for instance that of sanson, the rio caura issues also from lake cassipa.]

[* sanson, map for the voyage of acunha, 1680. id. south america, 1659. coronelli, indes occidentales, 1689.]

i have stated the origin of the fable of the lake cassipa, and the influence it has had on the opinion that the lake parima is the source of the orinoco. let us now examine what relates to this latter basin, this pretended interior sea, called rupunuwini by the geographers of the sixteenth century. in the latitude of four degrees or four degrees and a half (in which direction unfortunately, south of santo thome del angostura to the extent of eight degrees, no astronomical observation has been made) is a long and narrow cordillera, that of pacaraimo, quimiropaca, and ucucuamo; which, stretching from east to south-west, unites the group of mountains of parima to the mountains of dutch and french guiana. it divides its waters between the carony, the rupunury or rupunuwini, and the rio branco, and consequently between the valleys of the lower orinoco, the essequibo, and the rio negro. on the north-west of the cordillera de pacaraimo, which has been traversed but by a small number of europeans (by the german surgeon, nicolas hortsmann, in 1739; by a spanish officer, don antonio santos, in 1775; by the portuguese colonel, barata, in 1791; and by several english settlers, in 1811), descend the noeapra, the paraguamusi, and the paragua, which fall into the rio carony; on the north-east, the rupunuwini, a tributary stream of the rio essequibo. toward the south, the tacutu and the urariquera form together the famous rio parima, or rio branco.

this isthmus, between the branches of the rio essequibo and the rio branco (that is, between the rupunuwini on one side, and the pirara, the mahu, and the uraricuera or rio parima on the other), may be considered as the classical soil of the dorado of parima. the rivers at the foot of the mountains of pacaraimo are subject to frequent overflowings. above santa rosa, the right bank of the urariapara, a tributary stream of the uraricuera, is called el valle de la inundacion. great pools are also found between the rio parima and the xurumu. these are marked on the maps recently constructed in brazil, which furnish the most ample details of those countries. more to the west, the cano pirara, a tributary stream of the mahu, issues from a lake covered with rushes. this is the lake amucu described by nicolas hortsmann, and respecting which some portuguese of barcelos, who had visited the rio branco (rio parima or rio paravigiana), gave me precise notions during my stay at san carlos del rio negro. the lake amucu is several leagues broad, and contains two small islands, which santos heard called islas ipomucena. the rupunuwini (rupunury), on the banks of which hortsmann discovered rocks covered with hieroglyphical figures, approaches very near this lake, but does not communicate with it. the portage between the rupunuwini and the mahu is farther north, where the mountain of ucucuamo* rises, the natives still call the mountain of gold. they advised hortsmann to seek round the rio mahu for a mine of silver (no doubt mica with large plates), of diamonds, and emeralds. he found nothing but rocky crystals. his account seems to prove that the whole length of the mountains of the upper orinoco (sierra parima) toward the east, is composed of granitic rocks, full of druses and open veins, the peak of duida. near these lands, which still enjoy a great celebrity for their riches, on the western limits of dutch guiana, live the macusis, aturajos, and acuvajos. the traveller santos found them stationed between the rupunuwini, the mahu, and the chain of pacaraimo. it is the appearance of the micaceous rocks of the ucucuamo, the name of the rio parima, the inundations of the rivers urariapara, parima, and xurumu, and more especially the existence of the lake amucu (near the rio rupunuwini, and regarded as the principal source of the rio parima), which have given rise to the fable of the white sea and the dorado of parima. all these circumstances (which have served on this very account to corroborate the general opinion) are found united on a space of ground which is eight or nine leagues broad from north to south, and forty long from east to west. this direction, too, was always assigned to the white sea, by lengthening it in the direction of the latitude, till the beginning of the sixteenth century. now this white sea is nothing but the rio parima, which is called the white river (rio branco, or rio del aguas blancas), and runs through and inundates the whole of this land. the name of rupunuwini is given to the white sea on the most ancient maps, which identifies the place of the fable, since of all the tributary streams of the rio essequibo the rupunuwini is the nearest to the lake amucu. raleigh, in his first voyage (1595), had formed no precise idea of the situation of el dorado and the lake parima, which he believed to be salt, and which he calls another caspian sea. it was not till the second voyage (1596), performed equally at the expense of raleigh, that laurence keymis fixed so well the localities of el dorado, that he appears to me to have no doubt of the identity of the parima de manao with the lake amucu, and with the isthmus between the rupunuwini (a tributary stream of the essequibo) and the rio parima or rio branco. “the indians,” says keymis, “go up the dessekebe [essequibo] in twenty days, towards the south. to mark the greatness of this river, they call it the brother of the orinoco. after twenty days’ navigating they convey their canoes by a portage of one day, from the river dessekebe to a lake, which the jaos call roponowini, and the caribbees parime. this lake is as large as a sea; it is covered with an infinite number of canoes; and i suppose” [the indians then had told him nothing of this] “that this lake is no other than that which contains the town of manoa.”* hondius has given a curious plate of this portage; and, as the mouth of the carony was then supposed to be in latitude 4° (instead of 8° 8′), the portage of parima was placed close to the equator. at the same period the viapoco (oyapoc) and the rio cayenne (maroni?) were made to issue from this lake parima. the same name being given by the caribs to the western branch of the rio branco has perhaps contributed as much to the imaginary enlargement of the lake amucu, as the inundations of the various tributary streams of the uraricuera, from the confluence of the tacutu to the valle de la inundacion.

[* i follow the orthography of the manuscript journal of rodriguez; it is the cerro acuquamo of caulin, or rather of his commentator. hist. corogr. page 176.]

[* cayley’s life of raleigh volume 1 pages 159, 236 and 283. masham in the third voyage of raleigh (1596) repeats these accounts of the lake rupunuwini.]

we have shown above that the spaniards took the rio paragua, or parava, which falls into the carony, for a lake, because the word parava signifies sea, lake, river. parima seems also to denote vaguely great water; for the root par is found in the carib words that designate rivers, pools, lakes, and the ocean.* in arabic and in persian, bahr and deria are also applied at the same time to the sea, to lakes, and to rivers; and this practice, common to many nations in both worlds, has, on our ancient maps, converted lakes into rivers and rivers into lakes. in support of what i here advance, i shall appeal to very respectable testimony, that of father caulin. “when i inquired of the indians,” says this missionary, who sojourned longer than i on the banks of the lower orinoco, “what parima was, they answered that it was nothing more than a river that issued from a chain of mountains, the opposite side of which furnished waters to the essequibo.” caulin, knowing nothing of lake amucu, attributes the erroneous opinion of the existence of an inland sea solely to the inundations of the plains (a las inundaciones dilatadas por los bajos del pais). according to him, the mistakes of geographers arise from the vexatious circumstance of all the rivers of guiana having different names at their mouths and near their sources. “i have no doubt,” he adds, “that one of the upper branches of the rio branco is that very rio parima which the spaniards have taken for a lake (a quien suponian laguna).” such are the opinions which the historiographer of the expedition of the boundaries had formed on the spot. he could not expect that la cruz and surville, mingling old hypotheses with accurate ideas, would reproduce on their maps the mar dorado or mar blanco. thus, notwithstanding the numerous proofs which i have furnished since my return from america, of the non-existence of an inland sea the origin of the orinoco, a map has been published in my name,* on which the laguna parima figures anew.

[* in persian the root water (ab) is found also in lake (abdan). for other etymologies of the words parima and manoa see gili volume 1 pages 81 and 141; and gumilla volume 1 page 403.]

[* carte de l’amerique, dressee sur les observations de m. de humboldt, par fried. vienna 1818.]

from the whole of these statements it follows, first, that the laguna rupunuwini, or parima of the voyage of raleigh and of the maps of hondius, is an imaginary lake, formed by the lake amucu* and the tributary streams of the uraricuera, which often overflow their banks; secondly, that the laguna parime of surville’s map is the lake amucu, which gives rise to the rio pirara and (conjointly with the mahu, the tacutu, the uraricuera, or rio parima, properly so called) to the rio branco; thirdly, that the laguna parime of la cruz is an imaginary swelling of the rio parime (confounded with the orinoco) below the junction of the mahu with the xurumu. the distance from the mouth of the mahu to that of the tacutu is scarcely 0° 40′; la cruz enlarges it to 7° of latitude. he calls the upper part of the rio branco (that which receives the mahu) orinoco or purumu. there can be no doubt of its being the xurumu, one of the tributary streams of the tacutu, which is well known to the inhabitants of the neighbouring fort of san joaquim. all the names that figure in the fable of el dorado are found in the tributary streams of the rio branco. slight local circumstances, joined to the remembrances of the salt lake of mexico, more especially of the celebrated lake manoa in the dorado des omaguas, have served to complete a picture created by the imagination of raleigh and his two lieutenants, keymis and masham. the inundations of the rio branco, i conceive, may be compared at the utmost to those of the red river of louisiana, between nachitoches and cados, but not to the laguna de los xarayes, which is a temporary swelling of the rio paraguay.*

[* this is the lake amaca of surville and la cruz. by a singular mistake, the name of this lake is transformed to a village on arrowsmith’s map.]

[* southey volume 1 page 130. these periodical overflowings of the rio paraguay have long acted the same part in the southern hemisphere, as lake parima has been made to perform in the northern. hondius and sanson have made the rio de la plata, the rio topajos (a tributary stream of the amazon), the rio tocantines, and the rio de san francisco, issue from the laguna de los xarayes.]

we have now examined a white sea,* which the principal of the rio branco is made to traverse; and another,* which is placed on the east of this river, and communicates with it by the cano pirara. a third lake* is figured on the west of the rio branco, respecting which i found recently some curious details in the manuscript journal of the surgeon hortsmann. “at the distance of two days’ journey below the confluence of the mahu (tacutu) with the rio parima (uraricuera) a lake is found on top of a mountain. this lake is stocked with the same fish as the rio parima; but the waters of the former are black, and those of the latter white.” may not surville, from a vague notion of this basin, have imagined, in his map prefixed to father caulin’s work, an alpine lake of ten leagues in length, near which, towards the east, rise at the same time the orinoco, and the rio idapa, a tributary stream of the rio negro? however vague may be the account of the surgeon of hildesheim, it is impossible to admit that the mountain, which has a lake at its summit, is to the north of the parallel of 2° 30′: and this latitude coincides nearly with that of the cerro unturan. hence it follows that the alpine lake of hortsmann, which has escaped the attention of d’anville, and which is perhaps situate amid a group of mountains, lies north-east of the portage from the idapa to the mavaca, and south-east of the orinoco, where it goes up above esmeralda.

[* that of d’anville and la cruz, and of the greater part of the modern maps.]

[* the lake of surville, which takes the place of lake amucu.]

[* the lake which surville calls laguna tenida hasta ahora or la una parime.]

most of the historians who have treated of the first ages of the conquest seem persuaded that the name provincias or pais del dorado denoted originally every region abounding in gold. forgetting the precise etymology of the word el dorado (the gilded), they have not perceived that this tradition is a local fable, as were almost all the ancient fables of the greeks, the hindoos, and the persians. the history of the gilded man belongs originally to the andes of new grenada, and particularly to the plains in the vicinity of their eastern side: we see it progressively advance, as i observed above, three hundred leagues toward the east-north-east, from the sources of the caqueta to those of the rio branco and the essequibo. gold was sought in different parts of south america before 1536, without the word el dorado having been ever pronounced, and without the belief of the existence of any other centre of civilization and wealth, than the empire of the inca of cuzco. countries which now do not furnish commerce with the smallest quantities of the precious metals, the coast of paria, terra firma (castillo del oro), the mountains of santa marta, and the isthmus of darien, then enjoyed the same celebrity which has been more recently acquired by the auriferous lands of sonora, choco, and brazil.

diego de ordaz (1531) and alonzo de herrera (1535) directed their journeys of discovery along the banks of the lower orinoco. the former is the famous conquistador of mexico, who boasted that he had taken sulphur out of the crater of the peak of popocatepetl, and whom the emperor charles v permitted to wear a burning volcano on his armorial bearings. ordaz, named adelantado of all the country which he could conquer between brazil and the coast of venezuela, which was then called the country of the german company of welsers (belzares) of augsburg, began his expedition by the mouth of the maranon. he there saw, in the hands of the natives, “emeralds as big as a man’s fist.” they were, no doubt, pieces of that saussurite jade, or compact feldspar, which we brought home from the orinoco, and which la condamine found in abundance at the mouth of the rio topayos. the indians related to diego de ordaz that on going up during a certain number of suns toward the west, he would find a large rock (pena) of green stone; but before they reached this pretended mountain of emerald (rocks of euphotide?) a shipwreck put an end to all farther discovery. the spaniards saved themselves with difficulty in two small vessels. they hastened to get out of the mouth of the amazon; and the currents, which in those parts run with violence to the north-west, led ordaz to the coast of paria where, in the territory of the cacique yuripari (uriapari, viapari), sedeno had constructed the casa fuerte de paria. this post being very near the mouth of the orinoco, the mexican conquistador resolved to attempt an expedition on this great river. he sojourned first at carao (caroa, carora), a large indian village, which appears to me to have been a little to the east of the confluence of the carony; he then went up the cabruta (cabuta, cabritu), and to the mouth of the meta (metacuyu), where he found great difficulty in passing his boats through the raudal of cariven. the aruacas, whom ordaz employed as guides, advised him to go up the meta; where, on advancing towards the west, they asserted he would find men clothed, and gold in abundance. ordaz pursued in preference the navigation of the orinoco, but the cataracts of tabaje (perhaps even those of the atures) compelled him to terminate his discoveries.

it is worthy of remark that in this voyage, far anterior to that of orellana, and consequently the greatest which the spaniards had then performed on a river of the new world, the name of the orinoco was for the first time heard. ordaz, the leader of the expedition, affirms that the river, from its mouth as far as the confluence of the meta, is called uriaparia, but that above this confluence it bears the name of orinucu. this word (formed analogously with the words tamanacu, otomacu, sinarucu) is, in fact, of the tamanac tongue; and, as the tamanacs dwell south-east of encaramada, it is natural that the conquistadores heard the actual name of the river only on drawing near the rio meta.* on this last tributary stream diego de ordaz received from the natives the first idea of civilized nations who inhabited the table-lands of the andes of new granada; of a very powerful prince with one eye (indio tuerto), and of animals less than stags, but fit for riding like spanish horses. ordaz had no idea that these animals were llamas (ovejas del peru). must we admit that llamas, which were used in the andes to draw the plough and as beasts of burden, but not for riding, were already common on the north and east of quito? i find that orellana saw these animals at the river amazon, above the confluence of the rio negro, consequently in a climate very different from that of the table-land of the andes. the table of an army of omaguas mounted on llamas served to embellish the account given by the fellow-travellers of felipe de urre of their adventurous expedition to the upper caqueta. we cannot be sufficiently attentive to these traditions, which seem to prove that the domestic animals of quito and peru had already begun to descend the cordilleras, and spread themselves by degrees in the eastern regions of south america.

[* gili volume 3 page 381. the following are the most ancient names of the orinoco, known to the natives near its mouth, and which historians give us altered by the double fault of pronunciation and orthography; yuyapari, yjupari, huriaparia, urapari, viapari, rio de paria. the tamanac word orinucu was disfigured by the dutch pilots into worinoque. the otomacs say joga-apurura (great river); the cabres and guaypunabis, paragua, bazagua parava, three words signifying great water, river, sea. that part of the orinoco between the apure and the guaviare is often denoted by the name of baraguan. a famous strait, which we have described above, bears also this name, which is no doubt a corruption of the word paragua. great rivers in every zone are called by the dwellers on their banks the river, without any particular denominations. if other names be added, they change in every province. thus the rio turiva, near the encaramada, has five names in the different parts of its course. the upper orinoco, or paragua, is called by the maquiritares (near esmeralda) maraguaca, on account of the lofty mountains of this name near duida. gili volume 1 pages 22 and 364. caulin page 75. in most of the names of the rivers of america we recognize the root water. thus yacu in the peruvian, and veni in the maypure tongues, signify water and river. in the lule dialect i find fo, water; foyavolto, a river; foysi, a lake; as in persian, ab is water; abi frat, the river euphrates; abdan, a lake. the root water is preserved in the derivatives.]

herrera, the treasurer of the expedition of ordaz, was sent in 1553, by the governor geronimo de ortal, to pursue the discovery of the orinoco and the meta. he lost nearly thirteen months between punta barina and the confluence of the carony in constructing flat-bottomed boats, and making the preparations indispensable for a long voyage. we cannot read without astonishment the narrative of those daring enterprises, in which three or four hundred horses were embarked to be put ashore whenever cavalry could act on one of the banks. we find in the expedition of herrera the same stations which we already knew; the fortress of paria, the indian village of uriaparia (no doubt below imataca, on a point where the inundations of the delta prevented the spaniards from being able to procure firewood), caroa, in the province of carora; the rivers caranaca (caura?) and caxavana (cuchivero?); the village of cabritu (cabruta), and the raudal near the mouth of the meta (probably the raudal of cariven and the piedra de la paciencia). as the rio meta, on account of the proximity of its sources and of its tributary streams to the auriferous cordilleras of new grenada (cundinamarca), enjoyed great celebrity, herrera attempted to go up this river. he there found nations more civilized than those of the orinoco, but that fed on the flesh of mute dogs. herrera was killed in battle by an arrow poisoned with the juice of curare (yierva); and when dying named alvaro de ordaz his lieutenant, who led the remains of the expedition (1535) to the fortress of paria, after having lost the few horses which had resisted a campaign of eighteen months.

confused reports which were circulated of the wealth of the inhabitants of the meta, and the other tributary streams that descend from the eastern side of the cordilleras of new grenada, engaged successively geronimo de ortal, nicolas federmann, and jorge de espira (george von speier), in 1535 and 1536, to undertake expeditions by land towards the south and south-west. from the promontory of paria, as far as cabo de la vela, little figures of molten gold had been found in the hands of the natives, as early as the years 1498 and 1500. the principal markets for these amulets, which the women used as ornaments, were the villages of curiana (coro) and cauchieto (near the rio la hacha). the metal employed by the founders of cauchieto came from a mountainous country more to the south. it may be conceived that the expeditions of ordaz and herrera served to increase the desire of drawing nearer to those auriferous countries. george von speier left coro (1535), and penetrated by the mountains of merida to the banks of the apure and the meta. he passed these two rivers near their sources, where they have but little breadth. the indians told him that, farther on, white men wandered about the plains. speier, who imagined that he was not far from the banks of the amazon, had no doubt that these wandering spaniards were men unfortunately shipwrecked in the expedition of ordaz. he crossed the savannahs of san juan de los llanos, which were said to abound in gold; and made a long stay at an indian village called pueblo de nuestra senora, and afterwards la fragua, south-east of the paramo de la suma paz. i have been on the western back of this group of mountains, at fusagasuga, and there heard that the plains by which they are skirted toward the east still enjoy some celebrity for wealth among the natives. speier found in the populous village of la fragua a casa del sol (temple of the sun), and a convent of virgins similar to those of peru and new granada. were these the consequence of a migration of religious rites towards the east? or must we admit that the plains of san juan were their first cradle? tradition, indeed, records that bochica, the legislator of new granada and high-priest of iraca, had gone up from the plains of the east to the table-land of bogota. but bochica being at once the offspring and the symbol of the sun, his history may contain allegories that are merely astrological. speier, pursuing his way toward the south, and crossing the two branches of the guaviare, which are the ariare and the guayavero (guayare or canicamare), arrived on the banks of the great rio papamene or caqueta. the resistance he met with during a whole year in the province de los choques, put an end, in 1537, to this memorable expedition. nicolas federmann and geronimo de ortal (1536), who went from macarapana and the mouth of the rio neveri, followed (1535) the traces of jorge de espira. the former sought for gold in the rio grande de la magdalena; the latter endeavoured to discover a temple of the sun (casa del sol) on the banks of the meta. ignorant of the idiom of the natives, they seemed to see everywhere, at the foot of the cordilleras, the reflexion of the greatness of the temples of iraca (sogamozo), which was then the centre of the civilization of cundinamarca.

i have now examined, in a geographical point of view, the expeditions on the orinoco, and in a western and southern direction on the eastern back of the andes, before the tradition of el dorado was spread among the conquistadores. this tradition, as we have noticed above, had its origin in the kingdom of quito, where luis daza (1535) met with an indian of new grenada who had been sent by his prince (no doubt the zippa of bogota, or the zaque of tunja), to demand assistance from atahualpa, inca of peru. this ambassador boasted, as is usual, the wealth of his country; but what particularly fixed the attention of the spaniards who were assembled with daza in the town of tacunga (llactacunga), was the history of a lord who, his body covered with powdered gold, went into a lake amid the mountains. this lake may have been the laguna de totta, a little to the east of sogamozo (iraca) and of tunja (hunca, the town of huncahua), where two chiefs, ecclesiastical and secular, of the empire of cundinamarca, or cundirumarca, resided; but no historical remembrance being attached to this mountain lake, i rather suppose that it was the sacred lake of guatavita, on the east of the mines of rock-salt of zipaquira, into which the gilded lord was made to enter. i saw on its banks the remains of a staircase hewn in the rock, and serving for the ceremonies of ablution. the indians said that powder of gold and golden vessels were thrown into this lake, as a sacrifice to the adoratorio de guatavita. vestiges are still found of a breach which was made by the spaniards for the purpose of draining the lake. the temple of the sun at sogamozo being pretty near the northern coasts of terra firma, the notions of the gilded man were soon applied to a high-priest of the sect of bochica, or indacanzas, who every morning, before he performed his sacrifice, caused powder of gold to be stuck upon his hands and face, after they had been smeared with grease. other accounts, preserved in a letter of oviedo addressed to the celebrated cardinal bembo, say that gonzalo pizarro, when he discovered the province of cinnamon-trees, “sought at the same time a great prince, noised in those countries, who was always covered with powdered gold, so that from head to foot he resembled an image of gold fashioned by the hand of a skilful workman (a una figura d’oro lavorato di mano d’un buonissimo orefice). the powdered gold is fixed to the body by means of an odoriferous resin; but, as this kind of garment would be uneasy to him while he slept, the prince washes himself every evening, and is gilded anew in the morning, which proves that the empire of el dorado is infinitely rich in mines.” it seems probable that there was something in the ceremonies of the worship introduced by bochica which gave rise to a tradition so generally spread. the strangest customs are found in the new world. in mexico the sacrificers painted their bodies and wore a kind of cape, with hanging sleeves of tanned human skin.

on the banks of the caura, and in other wild parts of guiana, where painting the body is used instead of tattooing, the nations anoint themselves with turtle-fat, and stick spangles of mica with a metallic lustre, white as silver and red as copper, on their skin, so that at a distance they seem to wear laced clothes. the fable of the gilded man is, perhaps, founded on a similar custom; and, as there were two sovereign princes in new granada, the lama of iraca and the secular chief or zaque of tunja, we cannot be surprised that the same ceremony was attributed sometimes to the prince and sometimes to the high-priest. it is more extraordinary that, as early as the year 1535, the country of el dorado was sought for on the east of the andes. robertson is mistaken in admitting that orellana received the first notions of it (1540) on the banks of the amazon. the history of fray piedro simon, founded on the memoirs of queseda, the conqueror of cundirumarca, proves directly the contrary; and gonzalo diaz de pineda, as early as 1536, sought for the gilded man beyond the plains of the province of quixos. the ambassador of bogota, whom daza met with in the kingdom of quito, had spoken of a country situate toward the east. was this because the table-land of new granada is not on the north, but on the north-east of quito? we may venture to say that the tradition of a naked man covered with powdered gold must have belonged originally to a hot region, and not to the cold table-lands of cundirumarca, where i often saw the thermometer sink below four or five degrees; however, on account of the extraordinary configuration of the country, the climate differs greatly at guatavita, tunja, iraca, and on the banks of the sogamozo. sometimes, also, religious ceremonies are preserved which took rise in another zone; and the muyscas, according to ancient traditions, made bochica, their first legislator and the founder of their worship, arrive from the plains situate to the east of the cordilleras. i shall not decide whether these traditions expressed an historical fact, or merely indicated, as we have already observed in another place, that the first lama, who was the offspring and symbol of the sun, must necessarily have come from the countries of the east. be it as it may, it is not less certain that the celebrity which the expeditions of ordaz, herrera, and speier had already given to the orinoco, the meta, and the province of papamene, situate between the sources of the guaviare and caqueta, contributed to fix the fable of el dorado near to the eastern back of the cordilleras.

the junction of three bodies of troops on the table-land of new granada spread through all that part of america occupied by the spaniards the news of an immensely rich and populous country which remained to be conquered. sebastian de belalcazar marched from quito by way of popayan (1536) to bogota; nicholas federmann, coming from venezuela, arrived from the east by the plains of meta. these two captains found, already settled on the table-land of cundirumarca, the famous adelantado gonzalo ximenez de queseda, one of whose descendants i saw near zipaquira, with bare feet, attending cattle. the fortuitous meeting of the three conquistadores, one of the most extraordinary and dramatic events of the history of the conquest, took place in 1538. belalcazar’s narratives inflamed the imagination of warriors eager for adventurous enterprises; and the notions communicated to luis daza by the indian of tacunga were compared with the confused ideas which ordaz had collected on the meta respecting the treasures of a great king with one eye (indio tuerto), and a people clothed, who rode upon llamas. an old soldier, pedro de limpias, who had accompanied federmann to the table-land of bogota, carried the first news of el dorado to coro, where the remembrance of the expedition of speier (1535 to 1537) to the rio papamene was still fresh. it was from this same town of coro that felipe von huten (urre, utre) undertook his celebrated voyage to the province of the omaguas, while pizarro, orellana, and hernan perez de quesada, brother of the adelantado, sought for the gold country at the rio napo, along the river of the amazons, and on the eastern chain of the andes of new grenada. the natives, in order to get rid of their troublesome guests, continually described dorado as easy to be reached, and situate at no considerable distance. it was like a phantom that seemed to flee before the spaniards, and to call on them unceasingly. it is in the nature of man, wandering on the earth, to figure to himself happiness beyond the region which he knows. el dorado, similar to atlas and the islands of the hesperides, disappeared by degrees from the domain of geography, and entered that of mythological fictions.

i shall not here relate the numerous enterprises which were undertaken for the conquest of this imaginary country. unquestionably we are indebted to them in great part for our knowledge of the interior of america; they have been useful to geography, as errors and daring hypotheses are often to the search of truth: but in the discussion on which we are employed, it is incumbent on me to rest only upon those facts which have had the most direct influence on the construction of ancient and modern maps. hernan perez de quesada, after the departure of his brother the adelantado for europe, sought anew (1539) but this time in the mountainous land north-east of bogota, the temple of the sun (casa del sol), of which geronimo de ortal had heard spoken in 1536 on the banks of the meta. the worship of the sun introduced by bochica, and the celebrity of the sanctuary of iraca, or sogamozo, gave rise to those confused reports of temples and idols of massy gold; but on the mountains as in the plains, the traveller believed himself to be always at a distance from them, because the reality never corresponded with the chimerical dreams of the imagination. francisco de orellana, after having vainly sought el dorado with pizarro in the provincia de los canelos, and on the auriferous banks of the napo, went down (1540) the great river of the amazon. he found there, between the mouths of the javari and the rio de la trinidad (yupura?) a province rich in gold, called machiparo (muchifaro), in the vicinity of that of the aomaguas, or omaguas. these notions contributed to carry el dorado toward the south-east, for the names omaguas (om-aguas, aguas), dit–aguas, and papamene, designated the same country — that which jorge de espira had discovered in his expedition to the caqueta. the omaguas, the manaos or manoas, and the guaypes (uaupes or guayupes) live in the plains on the north of the amazon. they are three powerful nations, the latter of which, stretching toward the west along the banks of the guape or uaupe, had been already mentioned in the voyages of quesada and huten. these two conquistadores, alike celebrated in the history of america, reached by different roads the llanos of san juan, then called valle de nuestra senora. hernan perez de quesada (1541) passed the cordilleras of cundirumarca, probably between the paramos of chingasa and suma paz; while felipe de huten, accompanied by pedro de limpias (the same who had carried to venezuela the first news of dorado from the land of bogota), directed his course from north to south, by the road which speier had taken to the eastern side of the mountains. huten left coro, the principal seat of the german factory or company of welser, when henry remboldt was its director. after having traversed (1541) the plains of casanare, the meta, and the caguan, he arrived at the banks of the upper guaviare (guayuare), a river which was long believed to be the source of the orinoco, and the mouth of which i saw in passing by san fernando de atabapo to the rio negro. not far from the right bank of the guaviare, huten entered macatoa, the city of the guapes. the people there were clothed, the fields appeared well cultivated; everything denoted a degree of civilization unknown in the hot region of america which extends to the east of the cordilleras. speier, in his expedition to the rio caqueta and the province of papamene, had probably crossed the guaviare far above macatoa, before the junction of the two branches of this river, the ariari and the guayavero. huten was told that on advancing more to the south-east he would enter the territory of the great nation of the omaguas, the priest-king of which was called quareca, and which possessed numerous herds of llamas. these traces of cultivation — these ancient resemblances to the table-land of quito — appear to me very remarkable. it has already been said above that orellana saw llamas at the dwelling of an indian chief on the banks of the amazon, and that ordaz had heard mention made of them in the plains of meta.

i pause where ends the domain of geography and shall not follow huten in the description either of that town of immense extent, which he saw from afar; or of the battle of the omaguas, where thirty-nine spaniards (the names of fourteen are recorded in the annals of the time) fought against fifteen thousand indians. these false reports contributed greatly to embellish the fable of el dorado. the name of the town of the omaguas is not found in the narrative of huten; but the manoas, from whom father fritz received, in the seventeenth century, plates of beaten gold, in his mission of yurim–aguas, are neighbours of the omaguas. the name of manoa subsequently passed from the country of the amazons to an imaginary town, placed in el dorado de la parima. the celebrity attached to those countries between the caqueta (papamene) and the guaupe (one of the tributary streams of the rio negro) excited pedro de ursua, in 1560, to that fatal expedition, which ended by the revolt of the tyrant aguirre. ursua, in going down the caqueta to enter the river of the amazons, heard of the province of caricuri. this denomination clearly indicates the country of gold; for i find that this metal is called caricuri in the tamanac, and carucuru in the caribbee. is it a foreign word that denotes gold among the nations of the orinoco, as the words sugar and cotton are in our european languages? this would prove that these nations learned to know the precious metals among the foreign products which came to them from the cordilleras,* or from the plains at the eastern back of the andes.

[* in peruvian or quichua (lengua del inca) gold is called cori, whence are derived chichicori, gold in powder, and corikoya, gold-ore.]

we arrive now at the period when the fable of el dorado was fixed in the eastern part of guiana, first at the pretended lake cassipa (on the banks of the paragua, a tributary stream of the carony), and afterwards between the sources of the rio essequibo and the rio branco. this circumstance has had the greatest influence on the state of geography in those countries. antonio de berrio, son-inlaw* and sole heir of the great adelantado gonzalo ximenez de quesada, passed the cordilleras to the east of tunja,* embarked on the rio casanare, and went down by this river, the meta, and the orinoco, to the island of trinidad. we scarcely know this voyage except by the narrative of raleigh; it appears to have preceded a few years the first foundation of vieja guayana, which was in the year 1591. a few years later (1595) berrio caused his maese de campo, domingo de vera, to prepare in europe an expedition of two thousand men to go up the orinoco, and conquer el dorado, which then began to be called the country of the manoa, and even the laguna de la gran manoa. rich landholders sold their farms, to take part in a crusade, to which twelve observantin monks, and ten secular ecclesiastics were annexed. the tales related by one martinez* (juan martin de albujar?), who said he had been abandoned in the expedition of diego de ordaz, and led from town to town till he reached the capital of el dorado, had inflamed the imagination of berrio. it is difficult to distinguish what this conquistador had himself observed in going down the orinoco from what he said he had collected in a pretended journal of martinez, deposited at porto rico. it appears that in general at that period the same ideas prevailed respecting america as those which we have long entertained in regard to africa; it was imagined that more civilization would be found towards the centre of the continent than on the coasts. already juan gonzalez, whom diego de ordaz had sent in 1531 to explore the banks of the orinoco, announced that “the farther you went up this river the more you saw the population increase.” berrio mentions the often-inundated province of amapaja, between the confluence of the meta and the cuchivero, where he found many little idols of molten gold, similar to those which were fabricated at cauchieto, east of coro. he believed this gold to be a product of the granitic soil that covers the mountainous country between the carichana, uruana, and cuchivero. in fact the natives have recently found a mass of native gold in the quebrada del tigre near the mission of encaramada. berrio mentions on the east of the province of amapaja the rio carony (caroly), which was said to issue from a great lake, because one of the tributary streams of the carony, the rio paragua (river of the great water), had been taken for an inland sea, from ignorance of the indian languages. several of the spanish historians believed that this lake, the source of the carony, was the grand manoa of berrio; but the notions he communicated to raleigh show that the laguna de manoa (del dorado, or de parime) was supposed to be to the south of the rio paragua, transformed into laguna cassipa. “both these basins had auriferous sands; but on the banks of the cassipa was situate macureguarai (margureguaira), the capital of the cacique of aromaja, and the first city of the imaginary empire of guyana.”

[* properly casado con una sobrina. fray pedro simon pages 597 and 608. harris coll. volume 2 page 212. laet page 652. caulin page 175. raleigh calls quesada cemenes de casada. he also confounds the periods of the voyages of ordaz (ordace), orellana (oreliano), and ursua. see empire of guiana pages 13 to 20.]

[* no doubt between the paramos of chita and of zoraca, taking the road of chire and pore. berrio told raleigh that he came from the casanare to the pato, from the pato to the meta, and from the meta to the baraguan (orinoco). we must not confound this rio pato (a name connected no doubt with that of the ancient mission of patuto) with the rio paute.]

[* i believe i can demonstrate that the fable of juan martinez, spread abroad by the narrative of raleigh, was founded on the adventures of juan martin de albujar, well known to the spanish historians of the conquest; and who, in the expedition of pedro de silva (1570), fell into the hands of the caribs of the lower orinoco. this albujar married an indian woman and became a savage himself, as happens sometimes in our own days on the western limits of canada and of the united states. after having long wandered with the caribs, the desire of rejoining the whites led him by the rio essequibo to the island of trinidad. he made several excursions to santa fe de bogota, and at length settled at carora. (simon page 591). i know not whether he died at porto rico; but it cannot be doubted that it was he who learned from the carib traders the name of the manoas [of jurubesh]. as he lived on the banks of the upper carony and reappeared by the rio essequibo, he may have contributed also to place the lake manoa at the isthmus of rupunuwini. raleigh makes his juan martinez embark below morequito, a village at the east of that confluence of the carony with the orinoco. thence he makes him dragged by the caribs from town to town, till he finds at manoa a relation of the inca atabalipa (atahualpa), whom he had known before at caxamarca, and who had fled before the spaniards. it appears that raleigh had forgotten that the voyage of ordaz (1531) was two years anterior to the death of atahualpa and the entire destruction of the empire of peru! he must have confounded the expedition of ordaz with that of silva (1570), in which juan martin de albuzar partook. the latter, who related his tales at santa fe, at venezuela, and perhaps at porto rico, must have combined what he had heard from the caribs with what he had learned from the spaniards respecting the town of the omaguas seen by huten; of the gilded man who sacrificed in a lake, and of the flight of the family of atahualpa into the forests of vilcabamba, and the eastern cordillera of the andes. garcilasso volume 2 page 194.]

as these often-inundated lands have been at all times inhabited by nations of carib race, who carried on a very active inland trade with the most distant regions, we must not be surprised that more gold was found here in the hands of the indians than elsewhere. the natives of the coast did not employ this metal in the form of ornaments or amulets only; but also as a medium of exchange. it is not extraordinary, therefore, that gold has disappeared on the coast of paria, and among the nations of the orinoco, their inland communications have been impeded by the europeans. the natives who have remained independent are in our days, no doubt, more wretched, more indolent, and in a ruder state, than they were before the conquest. the king of morequito, whose son raleigh took to england, had visited cumana in 1594, to exchange a great quantity of images of massy gold for iron tools, and european merchandise. the unexpected appearance of an indian chief augmented the celebrity of the riches of the orinoco. it was supposed that el dorado must be near the country from which the king of morequito came; and as this country was often inundated, and rivers vaguely called great seas, or great basins of water, el dorado must be on the banks of a lake. it was forgotten that the gold brought by the caribs and other trading people was as little the produce of the soil as the diamonds of brazil and india are the produce of the regions of europe, where they are most abundant. the expedition of berrio which had increased in number during the stay of the vessels at cumana, la margareta, and the island of trinidad, proceeded by morequito (near vieja guayana) towards the rio paragua, a tributary stream of the carony; but sickness, the ferocity of the natives, and the want of subsistence, opposed invincible obstacles to the progress of the spaniards. they all perished; except about thirty, who returned in a deplorable state to the post of santo thome.

these disasters did not calm the ardour displayed during the first half of the 17th century in the search of el dorado. the governor of the island of trinidad, antonio de berrio, became the prisoner of sir walter raleigh in the celebrated incursion of that navigator, in 1595, on the coast of venezuela and at the mouths of the orinoco. raleigh collected from berrio, and from other prisoners made by captain preston* at the taking of caracas, all the information which had been obtained at that period on the countries situate to the south of vieya guayana. he lent faith to the fables invented by juan martin de albujar, and entertained no doubt either of the existence of the two lakes cassipa and rupunuwini, or of that of the great empire of the inca, which, after the death of atahualpa, the fugitive princes were supposed to have founded near the sources of the essequibo. we are not in possession of a map that was constructed by raleigh, and which he recommended to lord charles howard to keep secret. the geographer hondius has filled up this void; and has even added to his map a table of longitudes and latitudes, among which figure the laguna del dorado, and the ville imperiale de manoas. raleigh, when at anchor near the punta del gallo* in the island of trinidad, made his lieutenants explore the mouths of the orinoco, principally those of capuri, grand amana (manamo grande), and macureo (macareo). as his ships drew a great deal of water, he found it difficult to enter the bocas chicas, and was obliged to construct flat-bottomed barks. he remarked the fires of the tivitivas (tibitibies), of the race of the guaraon indians, on the tops of the mauritia palm-trees; and appears to have first brought the fruit to europe (fructum squamosum, similem palmae pini). i am surprised, that he scarcely mentions the settlement, which had been made by berrio under the name of santo thome (la vieja guayana.) this settlement however dates from 1591; and though, according to fray pedro simon, “religion and policy prohibited all mercantile connection between christians [spaniards] and heretics [the dutch and english],” there was then carried on at the end of the sixteenth century, as in our days, an active contraband trade by the mouths of the orinoco. raleigh passed the river europa (guarapo), and “the plains of saymas (chaymas), which extend, keeping the same level, as far as cumana and caracas;” he stopped at morequito (perhaps a little to the north of the site of the villa de upata, in the missions of the carony), where an old cacique confirmed to him all the reveries of berrio on the irruption of foreign nations (orejones and epuremei) into guiana. the raudales or cataracts of the caroli (carony), a river which was at that period considered as the shortest way for reaching the towns of macureguarai and manoa, situate on the banks of lake cassipa and of lake rupunuwini or dorado, put an end to this expedition.

[* these prisoners belonged to the expedition of berrio and of hernandez de serpa. the english landed at macuto (then guayca macuto), whence a white man, villalpando, led them by a mountain-path between cumbre and the silla (perhaps passing over the ridge of galipano) to the town of caracas. simon page 594; raleigh page 19. those only who are acquainted with the situation can be sensible how difficult and daring this enterprise was.]

[* the northern part of la punta de icacos, which is the south-east cape of the island of trinidad. christopher columbus cast anchor there on august 3, 1498. a great confusion exists in the denomination of the different capes of the island of trinidad; and as recently, since the expedition of fidalgo and churruca, the spaniards reckon the longitudes in south america west of la punta de la galera (latitude 10° 50′, longitude 63° 20′), it is important to fix the attention of geographers on this point. columbus called the south-east cape of the island punta galera, on account of the form of a rock. from punta de la galera he sailed to the west and landed at a low cape, which he calls punta del arenal; this is our punta de icacos. in this passage, near a place (punta de la playa) where he stopped to take in water (perhaps at the mouth of the rio erin), he saw to the south, for the first time, the continent of america, which he called isla santa. it was, therefore, the eastern coast of the province of cumana, to the east of the cano macareo, near punta redonda, and not the mountainous coast of paria (isla de gracia, of columbus), which was first discovered.]

raleigh went scarcely the distance of sixty leagues along the orinoco; but he names the upper tributary streams, according to the vague notions he had collected; the cari, the pao, the apure (capuri?) the guarico (voari?) the meta,* and even, “in the province of baraguan, the great cataract of athule (atures), which prevents all further navigation.” notwithstanding raleigh’s exaggeration, so little worthy of a statesman, his narrative contains important materials for the history of geography. the orinoco above the confluence of the apure was at that period as little known to europeans, as in our time the course of the niger below sego. the names of several very remote tributary streams were known, but not their situation; and when the same name, differently pronounced, or not properly apprehended by the ear, furnished different sounds, their number was multiplied. other errors had perhaps their source in the little interest which antonio de berrio, the spanish governor, felt in communicating true and precise notions to raleigh, who indeed complains of his prisoner, “as being utterly unlearned, and not knowing the east from the west.” i shall not here discuss the point how far the belief of raleigh, in all he relates of inland seas similar to the caspian sea; on “the imperial and golden city of manoa,” and on the magnificent palaces built by the emperor inga of guyana, in imitation of those of his ancestors at peru, was real or pretended. the learned historian of brazil, mr. southey, and the biographer of raleigh, sir g. cayley, have recently thrown much light on this subject. it seems to me difficult to doubt of the extreme credulity of the chief of the expedition, and of his lieutenants. we see raleigh adapted everything to the hypotheses he had previously formed. he was certainly deceived himself; but when he sought to influence the imagination of queen elizabeth, and execute the projects of his own ambitious policy, he neglected none of the artifices of flattery. he described to the queen “the transports of those barbarous nations at the sight of her picture;” he would have “the name of the august virgin, who knows how to conquer empires, reach as far as the country of the warlike women of the orinoco and the amazon;” he asserts that “at the period when the spaniards overthrew the throne of cuzco, an ancient prophecy was found, which predicted that the dynasty of the incas would one day owe its restoration to great britain;” he advises that “on pretext of defending the territory against external enemies, garrisons of three or four thousand english should be placed in the towns of the inca, obliging this prince to pay a contribution annually to queen elizabeth of three hundred thousand pounds sterling;” finally he adds, like a man who foresees the future, that “all the vast countries of south america will one day belong to the english nation.”*

[* raleigh distinguishes the meta from the beta, which flows into the baraguan (the orinoco) conjointly with the daune, near athule; as he distinguishes the casanare, a tributary stream of the meta, and the casnero, which comes from the south, and appears to be the rio cuchivero. all above the confluence of the apure was then very confusedly known; and streams that flow into the tributary streams of the orinoco were considered as flowing into this river itself. the apure (capuri) and meta appeared long to be the same river on account of their proximity, and the numerous branches by which the arauca and the apure join each other. is the name of beta perchance connected with that of the nation of betoyes, of the plains of the casanare and the meta? hondius and the geographers who have followed him, with the exception of de l’isle (1700), and of sanson (1656), place the province of amapaja erroneously to the east of the orinoco. we see clearly by the narrative of raleigh (pages 26 and 72), that amapaja is the inundated country between the meta and the guarico. where are the rivers dauney and ubarro? the guaviare appears to me to be the goavar of raleigh.]

[* “i showed them her majesty’s picture, which the casigui so admired and honoured, as it had been easy to have brought them idolatrous thereof. and i further remember that berreo confessed to me and others (which i protest before the majesty of god to be true), that there was found among prophecies at peru (at such a time as the empire was reduced to the spanish obedience) in their chiefest temple, among divers others which foreshowed the losse of the said empyre, that from inglatierra those ingas should be again in time to come restored. the inga would yield to her majesty by composition many hundred thousand pounds yearely as to defend him against all enemies abroad and defray the expenses of a garrison of 3000 or 4000 soldiers. it seemeth to me that this empyre of guiana is reserved for the english nation.” (raleigh pages 7, 17, 51 and 100.)]

the four voyages of raleigh to the lower orinoco succeeded each other from 1595 to 1617. after all these useless attempts the ardour of research after el dorado has greatly diminished. no expeditions have since been formed by a numerous band of colonists; but some solitary enterprises have been encouraged by the governors of the provinces. the notions spread by the journeys of father acunha in 1688, and father fritz in 1637, to the auriferous land of the manoas of jurubesh, and to the laguna de ore, contributed to renew the ideas of el dorado in the portuguese and spanish colonies north and south of the equator. at cuenza, in the kingdom of quito, i met with some men, who were employed by the bishop marfil to seek at the east of the cordilleras, in the plains of macas, the ruins of the town of logrono, which was believed to be situate in a country rich in gold. we learn by the journal of hortsmann, which i have often quoted, that it was supposed, in 1740, el dorado might be reached from dutch guiana by going up the rio essequibo. don manuel centurion, the governor of santo thome del angostura, displayed an extreme ardour for reaching the imaginary lake of manoa. arimuicaipi, an indian of the nation of the ipurucotos, went down the rio carony, and by his false narrations inflamed the imagination of the spanish colonists. he showed them in the southern sky the clouds of magellan, the whitish light of which he said was the reflection of the argentiferous rocks situate in the middle of the laguna parima. this was describing in a very poetical manner the splendour of the micaceous and talcy slates of his country! another indian chief, known among the caribs of essequibo by the name el capitan jurado, vainly attempted to undeceive the governor centurion. fruitless attempts were made by the caura and the rio paragua; and several hundred persons perished miserably in these rash enterprises, from which, however, geography has derived some advantages. nicolas rodriguez and antonio santos (1775 to 1780) were employed by the spanish governor. santos, proceeding by the carony, the paragua, the paraguamusi, the anocapra, and the mountains of pacaraymo and quimiropaca, reached the uraricuera and the rio branco. i found some valuable information in the journals of these perilous expeditions.

the maritime charts which the florentine traveller, amerigo vespucci,* constructed in the early years of the sixteenth century, as piloto mayor de la casa de contratacion of seville, and in which he placed, perhaps artfully, the words tierra de amerigo, have not reached our times. the most ancient monument we possess of the geography of the new continent,* is the map of the world by john ruysch, annexed to a roman edition of ptolemy in 1508. we there find yucatan and honduras (the most southern part of mexico)* figured as an island, by the name of culicar. there is no isthmus of panama, but a passage, which permits of a direct navigation from europe to india. the great southern island (south america) bears the name of terra de pareas, bounded by two rivers, the rio lareno and the rio formoso. these pareas are, no doubt, the inhabitants of paria, a name which christopher columbus had already heard in 1498, and which was long applied to a great part of america. bishop geraldini says clearly, in a letter addressed to pope leo x in 1516: insula illa, quae europa et asia est major, quam indocti continentem asiae appellant, et alii americam vel pariam nuncupant [that island, larger than europe and asia joined together, which the unlearned call the continent of asia, and others america or paria].* i find in the map of the world of 1508 no trace whatever of the orinoco. this river appears, for the first time, by the name of rio dolce, on the celebrated map constructed in 1529 by diego ribeyro, cosmographer of the emperor charles v, which was published, with a learned commentary, by m. sprengel, in 1795. neither columbus (1498) nor alonzo de ojeda, accompanied by amerigo vespucci (1499), had seen the real mouth of the orinoco; they confounded it with the northern opening of the gulf of paria, to which they attributed (by an exaggeration so common to the navigators of that time, an immense volume of fresh water. it was vicente yanez pincon, who, after having discovered the mouth of the rio maranon,* first saw, in 1500, that of the orinoco. he called this river rio dolce — a name which, since ribeyro, was long preserved on our maps, and which has sometimes been given erroneously to the maroni and to the essequibo.

[* he died in 1512, as mr. munoz has proved by the documents of the archives of simancas. hist. del nuevo mundo volume 1 page 17. tiraboschi, storia della litteratura.]

[* see the learned researches of m. walckenaer, in the bibliographie universelle volume 6 page 209 article buckinck. on the maps added to ptolemy in 1506 we find no trace of the discoveries of columbus.]

[* no doubt the lands between uucatan, cape gracias a dios, and veragua, discovered by columbus (1502 and 1503), by solis, and by pincon (1506).]

[* alexandri geraldini itinerarium page 250.]

[* the name of maranon was known fifty-nine years before the expedition of lopez de aguirre; the denomination of the river is therefore erroneously attributed to the nickname of maranos (hogs), which this adventurer gave his companions in going down the river amazon. was not this vulgar jest rather an allusion to the indian name of the river?]

the great lake parima did not appear on our maps* till after the first voyage of raleigh. it was jodocus hondius who, as early as the year 1599, fixed the ideas of geographers and figured the interior of spanish guiana as a country well known. he transformed the isthmus between the rio branco and the rio rupunuwini (one of the tributary streams of the essequibo) into the lake rupunuwini, parima, or dorado, two hundred leagues long, and forty broad, and bounded by the latitudes of 1 degree 45 minutes south, and 2° north. this inland sea, larger than the caspian, is sometimes traced in the midst of a mountainous country, without communication with any river;* and sometimes the rio oyapok (waiapago, japoc, viapoco) and the rio de cayana are made to issue from it.* the first of these rivers, confounded in the eighth article of the treaty of utrecht with the rio de vicente pincon (rio calsoene of d’anville), has been, even down to the late congress of vienna, the subject of interminable discussions between the french and portuguese diplomatists.* the second is an imaginary prolongation either of the tonnegrande or of the oyac (wia?). the inland sea (laguna parime) was at first placed in such a manner that its western extremity coincided with the meridian of the confluence of the apure and the orinoco. by degrees it was advanced toward the east,* the western extremity being found to the south of the mouth of the orinoco. this change produced others in the respective situations of the lakes parima and cassipa, as well as in the direction of the course of the orinoco. this great river is represented as running from its delta as far as beyond the meta, from south to north, like the river magdalena. the tributary streams, therefore, which were made to issue from the lake cassipa, the carony, the arui, and the caura, then took the direction of the latitude, while in nature they follow that of a meridian. beside the lakes parima and cassipa, a third was traced upon the maps, from which the aprouague (apurwaca) was made to issue. it was then a general practice among geographers to attach all rivers to great lakes. by this means ortelius joined the nile to the zaire or rio congo, and the vistula to the wolga and the dnieper. north of mexico, in the pretended kingdoms of quivira and cibola, rendered celebrated by the falsehoods of the monk marcos de niza, a great inland sea was imagined, from which the rio colorado of california was made to issue.* a branch of the rio magdalena flowed to the laguna de maracaybo; and the lake of xarayes, near which a southern dorado was placed, communicated with the amazon, the miari* (meary) and the rio de san francisco. these hydrographic reveries have for the most part disappeared; but the lakes cassipa and dorado have been long simultaneously preserved on our maps.

[* i find no trace of it on a very rare map, dedicated to richard hakluyt, and constructed on the meridian of toledo. novus orbis, paris 1587. in this map, published before the voyage of quiros, a group of islands is marked (infortunatae insulae) where the friendly islands actually are. ortelius (1570) already knew them. were they islands seen by magellan?]

[* see, for instance, hondius, nieuwe caerte van het goudrycke landt guiana, 1599; and sanson’s map of america, in 1656 and 1669.]

[* brasilia et caribaua, auct. hondio et huelsen 1599.]

[* i have treated this question in a memoire sur la fixation des limites de la guyane francaise, written at the desire of the portuguese government during the negotiations of paris in 1817. (see schoell, archives polit. or pieces inedites volume 1 pages 48 to 58.) ribeyro, in his celebrated map of the world of 1529, places the rio de vicente pincon south of the amazon, near the gulf of maranhao. this navigator landed at this spot, after having been at cape saint augustin, and before he reached the mouth of the amazon. herrera dec. i page 107. the narrative of gomara, hist. nat. 1553 page 48, is very confused in a geographical point of view.]

[* compare the maps of 1599 with those of sanson (1656) and of blaeuw (1633).]

[* this is the mexican dorado, where it was pretended that vessels had been found on the coasts [of new albion?] loaded with the merchandise of catayo and china (gomara, hist. gen. page 117), and where fray marcos (like huten in the country of the omaguas) had seen from afar the gilded roofs of a great town, one of the siete ciudades. the inhabitants have great dogs, en los quales quando se mudan cargan su menage. (herrera dec. 6 pages 157 and 206.) later discoveries, however, leave no doubt that there existed a centre of civilization in those countries.]

[* as this river flows into the gulf of maranhao (so named because some french colonists, rifault, de vaux, and ravadiere, believed they were opposite the mouth of the maranon or amazon), the ancient maps call the meary maranon, or maranham. see the maps of hondius, and paulo de forlani. perhaps the idea that pincon, to whom the discovery of the real maranon is due, had landed in these parts, since become celebrated by the shipwreck of ayres da cunha, has also contributed to this confusion. the meary appears to me identical with the rio de vicente pincon of diego ribeyro, which is more than one hundred and forty leagues from that of the modern geographers. at present the name of maranon has remained at the same time to the river of the amazons, and to a province much farther eastward, the capital of which is maranhao, or st. louis de maranon.]

in following the history of geography we see the cassipa, figured as a rectangular parallelogram, enlarge by degrees at the expense of el dorado. while the latter is sometimes suppressed, no one ventures to touch the former,* which is the rio paragua (a tributary stream of the caroni) enlarged by temporary inundations. when d’anville learned from the expedition of solano that the sources of the orinoco, far from lying to the west, on the back of the andes of pasto, came from the east, from the mountains of parima, he restored in the second edition of his fine map of america (1760) the laguna parime, and very arbitrarily made it to communicate with three rivers, the orinoco, the rio branco, and the essequibo, by the mazuruni and the cujuni; assigning to it the latitude from 3 to 4° north, which had till then been given to lake cassipa.

[* sanson, course of the amazon, 1680; de l’isle, amerique merid. 1700. d’anville, first edition of his america, 1748.]

i have now stated, as i announced above, the variable forms which geographical errors have assumed at different periods. i have explained what in the configuration of the soil, the course of the rivers, the names of the tributary streams, and the multiplicity of the portages, may have given rise to the hypothesis of an inland sea in the centre of guiana. however dry discussions of this nature may appear, they ought not to be regarded as sterile and fruitless. they show travellers what remains to be discovered; and make known the degree of certainty which long-repeated assertions may claim. it is with maps, as with those tables of astronomical positions which are contained in our ephemerides, designed for the use of navigators: the most heterogeneous materials have been employed in their construction during a long space of time; and, without the aid of the history of geography, we could scarcely hope to discover at some future day on what authority every partial statement rests.

before i resume the thread of my narrative, it remains for me to add a few general reflections on the auriferous lands situate between the amazon and the orinoco. we have just shown that the fable of el dorado, like the most celebrated fables of the nations of the ancient world, has been applied progressively to different spots. we have seen it advance from the south-west to the north-east, from the oriental declivity of the andes towards the plains of rio branco and the essequibo, an identical direction with that in which the caribs for ages conducted their warlike and mercantile expeditions. it may be conceived that the gold of the cordilleras might be conveyed from hand to hand, through an infinite number of tribes, as far as the shore of guiana; since, long before the fur-trade had attracted english, russian, and american vessels to the north-west coast of america, iron tools had been carried from new mexico and canada beyond the rocky mountains. from an error in longitude, the traces of which we find in all the maps of the 16th century, the auriferous mountains of peru and new granada were supposed to be much nearer the mouths of the orinoco and the amazon than they are in fact. geographers have the habit of augmenting and extending beyond measure countries that are recently discovered. in the map of peru, published at verona by paulo di forlani, the town of quito is placed at the distance of 400 leagues from the coast of the south sea, on the meridian of cumana; and the cordillera of the andes there fills almost the whole surface of spanish, french, and dutch guiana. this erroneous opinion of the breadth of the andes has no doubt contributed to give so much importance to the granitic plains that extend on their eastern side. unceasingly confounding the tributary streams of the amazon with those of the orinoco, or (as the lieutenants of raleigh called it, to flatter their chief) the rio raleana, to the latter were attributed all the traditions which had been collected respecting the dorado of quixos, the omaguas, and the manoas.* the geographer hondius supposed that the andes of loxa, celebrated for their forests of cinchona, were only twenty leagues distant from the lake parima, or the banks of the rio branco. this proximity procured credit to the tidings of the flight of the inca into the forests of guiana, and the removal of the treasures of cuzco to the easternmost parts of that country. no doubt in going up towards the east, either by the meta or by the amazon, the civilization of the natives, between the puruz, the jupura, and the iquiari, was observed to increase. they possessed amulets, little idols of molten gold, and chairs, elegantly carved; but these traces of dawning civilization are far distant from those cities and houses of stone described by raleigh and those who followed him. we have made drawings of some ruins of great edifices east of the cordilleras, when going down from loxa towards the amazon, in the province of jaen de bracamoros; and thus far the incas had carried their arms, their religion, and their arts. the inhabitants of the orinoco were also, before the conquest, when abandoned to themselves, somewhat more civilized than the independent hordes of our days. they had populous villages along the river, and a regular trade with more southern nations; but nothing indicates that they ever constructed an edifice of stone. we saw no vestige of any during the course of our journey.

[* the flight of manco–inca, brother of atahualpa, to the east of the cordilleras, no doubt gave rise to the tradition of the new empire of the incas in dorado. it was forgotten that caxamarca and cuzco, two towns where the princes of that unfortunate family were at the time of their emigration, are situate to the south of the amazon, in the latitudes seven degrees eight minutes, and thirteen degrees twenty-one minutes south, and consequently four hundred leagues south-west of the pretended town of manoa on the lake parima (three degrees and a half north latitude). it is probable that, from the extreme difficulty of penetration into the plains east of the andes, covered with forests, the fugitive princes never went beyond the banks of the beni. the following is what i learnt with certainty respecting the emigration of the family of the inca, some sad vestiges of which i saw on passing by caxamarca. manco–inca, acknowledged as the legitimate successor of atahualpa, made war without success against the spaniards. he retired at length into the mountains and thick forests of vilcabamba, which are accessible either by huamanga and antahuaylla, or by the valley of yucay, north of cuzco. of the two sons of manco–inca, the eldest, sayri–tupac, surrendered himself to the spaniards, upon the invitation of the viceroy of peru, hurtado de mendoza. he was received with great pomp at lima, was baptized there, and died peaceably in the fine valley of yucay. the youngest son of manco–inca, tupac–amaru, was carried off by stratagem from the forests of vilcabamba, and beheaded on pretext of a conspiracy formed against the spanish usurpers. at the same period, thirty-five distant relations of the inca atahualpa were seized, and conveyed to lima, in order to remain under the inspection of the audiencia. (garcilasso volume 2 pages 194, 480 and 501.) it is interesting to inquire whether any other princes of the family of manco–capac have remained in the forests of vilcabamba, and if there still exist any descendants of the incas of peru between the apurimac and the beni. this supposition gave rise in 1741 to the famous rebellion of the chuncoes, and to that of the amages and campoes led on by their chief, juan santos, called the false atahualpa. the late political events of spain have liberated from prison the remains of the family of jose gabriel condorcanqui, an artful and intrepid man, who, under the name of the inca tupac–amaru, attempted in 1781 that restoration of the ancient dynasty which raleigh had projected in the time of queen elizabeth.]

though the celebrity of the riches of spanish guiana is chiefly assignable to the geographical situation of the country and the errors of the old maps, we are not justified in denying the existence of any auriferous land in the tract of country of eighty-two thousand square leagues, which stretches between the orinoco and the amazon, on the east of the andes of quito and new granada. what i saw of this country between the second and eighth degrees of latitude, and the sixty-sixth and seventy-first degrees of longitude, is entirely composed of granite, and of a gneiss passing into micaceous and talcous slate. these rocks appear naked in the lofty mountains of parima, as well as in the plains of the atabapo and the cassiquiare. granite predominates there over the other rocks; and though, in both continents, the granite of ancient formation is pretty generally destitute of gold-ore, we cannot thence conclude that the granite of parima contains no vein, no stratum of auriferous quartz. on the east of the cassiquiare towards the sources of the orinoco, we observed that the number of these strata and these veins increased. the granite of these countries, by its structure, its mixture of hornblende, and other geological features alike important, appears to me to belong to a more recent formation, perhaps posterior to the gneiss, and analogous to the stanniferous granites, the hyalomictes, and the pegmatites. now the least ancient granites are also the least destitute of metals; and several auriferous rivers and torrents in the andes, in the salzburg, fichtelgebirge, and the table-land of the two castiles, lead us to believe that these granites sometimes contain native gold, and portions of auriferous pyrites and galena disseminated throughout the whole rock, as is the case with tin and magnetic and micaceous iron. the group of the mountains of parima, several summits of which attain the height of one thousand three hundred toises, was almost entirely unknown before our visit to the orinoco. this group, however, is a hundred leagues long and eighty broad; and though wherever m. bonpland and i traversed this vast group of mountains, its structure seemed to us extremely uniform, it would be wrong to affirm that it may not contain very metalliferous transition rocks and mica-slates superimposed on the granite.

i have already observed that the silvery lustre and frequency of mica have contributed to give guiana great celebrity for metallic wealth. the peak of calitamini, glowing every evening at sunset with a reddish fire, still attracts the attention of the inhabitants of maypures. according to the fabulous stories of the natives, the islets of mica-slate, situate in lake amucu, augment by their reflection the lustre of the nebulae of the southern sky. “every mountain,” says raleigh, “every stone in the forests of the orinoco, shines like the precious metals; if it be not gold, it is madre del oro (mother of gold).” raleigh asserts that he brought back gangues of auriferous white quartz (“harde white sparr”); and to prove the richness of this ore he gives an account of the assays that were made by the officers of the mint at london.* i have no reason to believe that the chemists of that time sought to lead queen elizabeth into error, and i will not insult the memory of raleigh by supposing, like his contemporaries,* that the auriferous quartz which he brought home had not been collected in america. we cannot judge of things from which we are separated by so long an interval of time. the gneiss of the littoral chain* contains traces of the precious metals; and some grains of gold have been found in the mountains of parima, near the mission of encaramada. how can we infer the absolute sterility of the primitive rocks of guiana from testimony merely negative, from the circumstance that during a journey of three months we saw no auriferous vein appearing above the soil?

[* messrs. westewood, dimocke, and bulmar.]

[* see the defence of raleigh in the preface to the discovery of guiana, 1596 pages 2 to 4.]

[* in the southern branch of this chain which passes by yusma, villa de cura and ocumare, particularly near buria, los teques and los marietas.]

in order to bring together whatever may enlighten the government of this country on a subject so long disputed, i will enter upon a few more geological considerations. the mountains of brazil, notwithstanding the numerous traces of embedded ore which they display between saint paul and villa rica, have furnished only stream-works of gold. more than six-sevenths of the seventy-eight thousand marks (52,000 pounds) of this metal, with which at the beginning of the 19th century america annually supplied the commerce of europe, have come, not from the lofty cordilleras of the andes, but from the alluvial lands on the east and west of the cordilleras. these lands are raised but little above the level of the sea, like those of sonora in mexico, and of choco and barbacoas in new granada; or they stretch along in table-lands, as in the interior of brazil.* is it not probable that some other depositions of auriferous earth extend toward the northern hemisphere, as far as the banks of the upper orinoco and the rio negro, two rivers which form but one basin with that of the amazon? i observed, when speaking of el dorado de canelas, the omaguas and the iquiare, that almost all the rivers which flow from the west wash down gold in abundance, and very far from the cordilleras. from loxa to popayan these cordilleras are composed alternately of trachytes and primitive rocks. the plains of ramora, of logrono, and of macas (sevilla del oro), the great rio napo with its tributary streams* (the ansupi and the coca, in the province of quixos), the caqueta de mocoa as far as the mouth of the fragua, in fine, all the country comprised between jaen de bracamoros and the guaviare,* preserve their ancient celebrity for metallic wealth. more to the east, between the sources of the guainia (rio negro), the uaupes, the iquiare, and the yurubesh, we find a soil incontestably auriferous. there acunha and father fritz placed their laguna del oro; and various accounts which i obtained at san carlos from portuguese americans explain perfectly what la condamine has related of the plates of beaten gold found in the hands of the natives. if we pass from the iquiare to the left bank of the rio negro, we enter a country entirely unknown, between the rio branco, the sources of the essequibo, and the mountains of portuguese guiana. acunha speaks of the gold washed down by the northern tributary streams of the lower maranon, such as the rio trombetas (oriximina), the curupatuba, and the ginipape (rio de paru). it appears to me a circumstance worthy of attention that all these rivers descend from the same table-land, the northern slope of which contains the lake amucu, the dorado of raleigh and the dutch, and the isthmus between the rupunuri (rupunuwini) and the rio mahu. there is no reason for denying the existence of auriferous alluvial lands far from the cordilleras of the andes on the north of the amazon; as there are on the south in the mountains of brazil. the caribs of the carony, the cuyuni and the essequibo, have practised on a small scale the washing of alluvial earth from the remotest times.* when we examine the structure of mountains and embrace in one point of view an extensive surface of the globe, distances disappear; and places the most remote insensibly draw near each other. the basin of the upper orinoco, the rio negro, and the amazon is bounded by the mountains of parime on the north, and by those of minas geraes, and matogrosso on the south. the opposite slopes of the same valley often display an analogy in their geological relations.

[* the height of villa rica is six hundred and thirty toises; but the great table-land of the capitania de minas geraes is only three hundred toises in height. see the profile which colonel d’eschwege has published at weimar, with an indication of the rocks, in imitation of my profile of the mexican table-land.]

[* the little rivers cosanga, quixos, and papallacta or maspa, which form the coca, rise on the eastern slope of the nevado de antisana. the rio ansupi brings down the largest grains of gold: it flows into the napo, south of the archidona, above the mouth of the misagualli. between the misagualli and the rio coca, in the province of avila, five other northern tributary streams of the napo (the siguna, munino, suno, guataracu, and pucono) are known as being singularly auriferous. these local details are taken from several manuscript reports of the governor of quixos, from which i traced the map of the countries east of the antisana.]

[* from rio santiago, a tributary stream of the upper maranon, to the llanos of caguan and of san juan.]

[* “on the north of the confluence of the curupatuba and the amazon,” says acunha, “is the mountain of paraguaxo, which, when illumined by the sun, glows with the most beautiful colours; and thence from time to time issues a horrible noise (revienta con grandes struenos).” is there a volcanic phenomenon in this eastern part of the new continent? or is it the love of the marvellous, which has given rise to the tradition of the bellowings (bramidos) of paraguaxo? the lustre emitted from the sides of the mountain recalls to mind what we have mentioned above of the miraculous rocks of calitamini, and the island ipomucena, in the imaginary lake dorado. in one of the spanish letters intercepted at sea by captain george popham, in 1594, it is said, “having inquired of the natives whence they obtained the spangles and powder of gold, which we found in their huts, and which they stick on their skin by means of some greasy substances, they told us that in a certain plain they tore up the grass, and gathered the earth in baskets, to subject it to the process of washing.” raleigh page 109. can this passage be explained by supposing that the indians sought thus laboriously, not for gold, but for spangles of mica, which the natives of rio caura still employ as ornaments, when they paint their bodies?]

i have described in this and the preceding volume the vast provinces of venezuela and spanish guiana. while examining their natural limits, their climate, and their productions, i have discussed the influence produced by the configuration of the soil on agriculture, commerce, and the more or less rapid progress of society. i have successively passed over the three regions that succeed each other from north to south; from the mediterranean of the west indies to the forests of the upper orinoco and of the amazon. the fertile land of the shore, the centre of agricultural riches, is succeeded by the llanos, inhabited by pastoral tribes. these llanos are in their turn bordered by the region of forests, the inhabitants of which enjoy, i will not say liberty, which is always the result of civilization, but a sort of savage independence. on the limit of these two latter zones the struggle now exists which will decide the emancipation and future prosperity of america. the changes which are preparing cannot efface the individual character of each region; but the manners and condition of the inhabitants will assume a more uniform colour. this consideration perhaps adds interest to a tour made in the beginning of the nineteenth century. we like to see, traced in the same picture, the civilized nations of the sea-shore, and the feeble remains of the natives of the orinoco, who know no other worship than that of the powers of nature; and who, like the ancient germans, deify the mysterious object which excites their simple admiration.*

[* deorum nominibus appellant secretum illud, quod sola reverentia vident. tacitus germania 9.]