At the distance of two or three leagues from the Mission, we find, in these plains intersected by granitic hills, a vegetation no less rich than varied. On comparing the site of Carichana with that of all the villages above the Great Cataracts, we are surprised at the facility with which we traverse the country, without following the banks of the rivers, or being stopped by the thickness of the forests. M. Bonpland made several excursions on horseback, which furnished him with a rich harvest of plants. I shall mention only the paraguatan, a magnificent species of the macrocnemum, the bark of which yields a red dye;* (* Macrocnemum tinctorium.) the guaricamo, with a poisonous root;* (* Ityania coccidea.) the Jacaranda obtusifolia; and the serrape, or jape* (* Dipterix odorata, Willd. or Baryosma tongo of Gaertner. The jape furnishes Carichana with excellent timber.) of the Salive Indians, which is the Coumarouna of Aublet, so celebrated throughout Terra Firma for its aromatic fruit. This fruit, which at Caracas is placed among linen, as in Europe it is in snuff, under the name of tonca, or Tonquin bean, is regarded as poisonous. It is a false notion, very general in the province of Cumana, that the excellent liqueur fabricated at Martinique owes its peculiar flavour to the jape. In the Missions it is called simaruba; a name that may occasion serious mistakes, the true simaruba being a febrifuge species of the Quassia genus, found in Spanish Guiana only in the valley of Rio Caura, where the Paudacot Indians give it the name of achecchari.

I found the dip of the magnetic needle, in the great square at Carichana, 33.7 degrees (new division). The intensity of the magnetic action was expressed by two hundred and twenty-seven oscillations in ten minutes of time; an increase of force that would seem to indicate some local attraction. Yet the blocks of the granite, blackened by the waters of the Orinoco, have no perceptible action upon the needle.

The river had risen several inches during the day on the 10th of April; this phenomenon surprised the natives so much the more, as the first swellings are almost imperceptible, and are usually followed in the month of April by a fall for some days. The Orinoco was already three feet higher than the level of the lowest waters. The natives showed us on a granite wall the traces of the great rise of the waters of late years. We found them to be forty-two feet high, which is double the mean rise of the Nile. But this measure was taken in a place where the bed of the Orinoco is singularly hemmed in by rocks, and I could only notice the marks shown me by the natives. It may easily be conceived that the effect and the height of the increase differs according to the profile of the river, the nature of the banks more or less elevated, the number of rivers flowing in that collect the pluvial waters, and the length of ground passed over. It is an unquestionable fact that at Carichana, at San Borja, at Atures, and at Maypures, wherever the river has forced its way through the mountains, you see at a hundred, sometimes at a hundred and thirty feet, above the highest present swell of the river, black bands and erosions, that indicate the ancient levels of the waters. Is then this river, which appears to us so grand and so majestic, only the feeble remains of those immense currents of fresh water which heretofore traversed the country at the east of the Andes, like arms of inland seas? What must have been the state of those low countries of Guiana that now undergo the effects of annual inundations? What immense numbers of crocodiles, manatees, and boas must have inhabited these vast spaces of land, converted alternately into marshes of stagnant water, and into barren and fissured plains! The more peaceful world which we inhabit has then succeeded to a world of tumult. The bones of mastodons and American elephants are found dispersed on the table-lands of the Andes. The megatherium inhabited the plains of Uruguay. On digging deep into the ground, in high valleys, where neither palm-trees nor arborescent ferns can grow, strata of coal are discovered, that still show vestiges of gigantic monocotyledonous plants.

There was a remote period then, in which the classes of plants were otherwise distributed, when the animals were larger, and the rivers broader and of greater depth. There end those records of nature, that it is in our power to consult. We are ignorant whether the human race, which at the time of the discovery of America scarcely formed a few feeble tribes on the east of the Cordilleras, had already descended into the plains; or whether the ancient tradition of the great waters, which is found among the nations of the Orinoco, the Erevato, and the Caura, belong to other climates, whence it has been propagated to this part of the New Continent.

On the 11th of April, we left Carichana at two in the afternoon, and found the course of the river more and more encumbered by blocks of granite rocks. We passed on the west the Cano Orupe, and then the great rock known by the name of Piedra del Tigre. The river is there so deep, that no bottom can be found with a line of twenty-two fathoms. Towards evening the weather became cloudy and gloomy. The proximity of the storm was marked by squalls alternating with dead calms. The rain was violent, and the roof of foliage, under which we lay, afforded but little shelter. Happily these showers drove away the mosquitos, at least for some time. We found ourselves before the cataract of Cariven, and the impulse of the waters was so strong, that we had great difficulty in gaining the land. We were continually driven back to the middle of the current. At length two Salive Indians, excellent swimmers, leaped into the water, and having drawn the boat to shore by means of a rope, made it fast to the Piedra de Carichana Vieja, a shelf of bare rock, on which we passed the night. The thunder continued to roll during a part of the night; the swell of the river became considerable; and we were several times afraid that our frail bark would be driven from the shore by the impetuosity of the waves.

The granitic rock on which we lay is one of those, where travellers on the Orinoco have heard from time to time, towards sunrise, subterraneous sounds, resembling those of the organ. The missionaries call these stones laxas de musica. "It is witchcraft (cosa de bruxas)," said our young Indian pilot, who could speak Spanish. We never ourselves heard these mysterious sounds, either at Carichana Vieja, or in the Upper Orinoco; but from information given us by witnesses worthy of belief, the existence of a phenomenon that seems to depend on a certain state of the atmosphere, cannot be denied. The shelves of rock are full of very narrow and deep crevices. They are heated during the day to 48 or 50 degrees. I several times found their temperature at the surface, during the night, at 39 degrees, the surrounding atmosphere being at 28 degrees. It may easily be conceived, that the difference of temperature between the subterranean and the external air attains its maximum about sunrise, or at that moment which is at the same time farthest from the period of the maximum of the heat of the preceding day. May not these organ-like sounds, which are heard when a person lays his ear in contact with the stone, be the effect of a current of air that issues out through the crevices? Does not the impulse of the air against the elastic spangles of mica that intercept the crevices, contribute to modify the sounds? May we not admit that the ancient inhabitants of Egypt, in passing incessantly up and down the Nile, had made the same observation on some rock of the Thebaid; and that the music of the rocks there led to the jugglery of the priests in the statue of Memnon? Perhaps, when, "the rosy-fingered Aurora rendered her son, the glorious Memnon, vocal,"* (* These are the words of an inscription, which attests that sounds were heard on the 13th of the month Pachon, in the tenth year of the reign of Antoninus. See Monuments de l'Egypte Ancienne.) the voice was that of a man hidden beneath the pedestal of the statue; but the observation of the natives of the Orinoco, which we relate, seems to explain in a natural manner what gave rise to the Egyptian belief of a stone that poured forth sounds at sunrise.

Almost at the same period at which I communicated these conjectures to some of the learned of Europe, three French travellers, MM. Jomard, Jollois, and Devilliers, were led to analogous ideas. They heard, at sunrise, in a monument of granite, at the centre of the spot on which stands the palace of Karnak, a noise resembling that of a string breaking. Now this comparison is precisely that which the ancients employed in speaking of the voice of Memnon. The French travellers thought, like me, that the passage of rarefied air through the fissures of a sonorous stone might have suggested to the Egyptian priests the invention of the juggleries of the Memnomium.

We left the rock at four in the morning. The missionary had told us that we should have great difficulty in passing the rapids and the mouth of the Meta. The Indians rowed twelve hours and a half without intermission, and during all that time, they took no other nourishment than cassava and plantains. When we consider the difficulty of overcoming the force of the current, and of passing the cataracts; when we reflect on the constant employment of the muscular powers during a navigation of two months; we are equally surprised at the constitutional vigour and the abstinence of the Indians of the Orinoco and the Amazon. Amylaceous and saccharine substances, sometimes fish and the fat of turtles' eggs, supply the place of food drawn from the first two classes of the animal kingdom, those of quadrupeds and birds.

We found the bed of the river, to the length of six hundred toises, full of granite rocks. Here is what is called the Raudal de Cariven. We passed through channels that were not five feet broad. Our canoe was sometimes jammed between two blocks of granite. We sought to avoid these passages, into which the waters rushed with a fearful noise; but there is really little danger, in a canoe steered by a good Indian pilot. When the current is too violent to be resisted the rowers leap into the water, and fasten a rope to the point of a rock, to warp the boat along. This manoeuvre is very tedious; and we sometimes availed ourselves of it, to climb the rocks among which we were entangled. They are of all dimensions, rounded, very black, glossy like lead, and destitute of vegetation. It is an extraordinary phenomenon to see the waters of one of the largest rivers on the globe in some sort disappear. We perceived, even far from the shore, those immense blocks of granite, rising from the ground, and leaning one against another. The intervening channels in the rapids are more than twenty-five fathoms deep; and are the more difficult to be observed, as the rocks are often narrow toward their bases, and form vaults suspended over the surface of the river. We perceived no crocodiles in the raudal; these animals seem to shun the noise of cataracts.

From Cabruta to the mouth of the Rio Sinaruco, a distance of nearly two degrees of latitude, the left bank of the Orinoco is entirely uninhabited; but to the west of the Raudal de Cariven an enterprising man, Don Felix Relinchon, had assembled some Jaruro and Ottomac Indians in a small village. It is an attempt at civilization, on which the monks have had no direct influence. It is superfluous to add, that Don Felix lives at open war with the missionaries on the right bank of the Orinoco.

Proceeding up the river we arrived, at nine in the morning, before the mouth of the Meta, opposite the spot where the Mission of Santa Teresa, founded by the Jesuits, was heretofore situated.

Next to the Guaviare, the Meta is the most considerable river that flows into the Orinoco. It may be compared to the Danube, not for the length of its course, but for the volume of its waters. Its mean depth is thirty-six feet, and it sometimes reaches eighty-four. The union of these two rivers presents a very impressive spectacle. Lonely rocks rise on the eastern bank. Blocks of granite, piled upon one another, appear from afar like castles in ruins. Vast sandy shores keep the skirting of the forest at a distance from the river; but we discover amid them, in the horizon, solitary palm-trees, backed by the sky, and crowning the tops of the mountains. We passed two hours on a large rock, standing in the middle of the Orinoco, and called the Piedra de la Paciencia, or the Stone of Patience, because the canoes, in going up, are sometimes detained there two days, to extricate themselves from the whirlpool caused by this rock.

The Rio Meta, which traverses the vast plains of Casanare, and which is navigable as far as the foot of the Andes of New Grenada, will one day be of great political importance to the inhabitants of Guiana and Venezuela. From the Golfo Triste and the Boca del Drago a small fleet may go up the Orinoco and the Meta to within fifteen or twenty leagues of Santa Fe de Bogota. The flour of New Grenada may be conveyed the same way. The Meta is like a canal of communication between countries placed in the same latitude, but differing in their productions as much as France and Senegal. The Meta has its source in the union of two rivers which descend from the paramos of Chingasa and Suma Paz. The first is the Rio Negro, which, lower down, receives the Pachaquiaro; the second is the Rio de Aguas Blancas, or Umadea. The junction takes place near the port of Marayal. It is only eight or ten leagues from the Passo de la Cabulla, where you quit the Rio Negro, to the capital of Santa Fe. From the villages of Xiramena and Cabullaro to those of Guanapalo and Santa Rosalia de Cabapuna, a distance of sixty leagues, the banks of the Meta are more inhabited than those of the Orinoco. We find in this space fourteen Christian settlements, in part very populous; but from the mouths of the rivers Pauto and Casanare, for a space of more than fifty leagues, the Meta is infested by the Guahibos, a race of savages.* (* I find the word written Guajibos, Guahivos, and Guagivos. They call themselves Gua-iva.)

The navigation of this river was much more active in the time of the Jesuits, and particularly during the expedition of Iturriaga, in 1756, than it is at present. Missionaries of the same order then governed the banks of the Meta and of the Orinoco. The villages of Macuco, Zurimena, and Casimena, were founded by the Jesuits, as well as those of Uruana, Encaramada, and Carichana.

These Fathers had conceived the project of forming a series of Missions from the junction of the Casanare with the Meta to that of the Meta with the Orinoco. A narrow zone of cultivated land would have crossed the vast steppes that separate the forests of Guiana from the Andes of New Grenada.

At the period of the harvest of turtles' eggs, not only the flour of Santa Fe descended the river, but the salt of Chita,* (* East of Labranza Grande, and the north-west of Pore, now the capital of the province of Casanare.) the cotton cloth of San Gil, and the printed counterpanes of Socorro. To give some security to the little traders who devoted themselves to this inland commerce, attacks were made from time to time from the castillo or fort of Carichana, on the Guahibos.

To keep these Guahibos in awe, the Capuchin missionaries, who succeeded the Jesuits in the government of the Missions of the Orinoco, formed the project of founding a city at the mouth of the Meta, under the name of the Villa de San Carlos. Indolence, and the dread of tertian fevers, have prevented the execution of this project; and all that has ever existed of the city of San Carlos, is a coat of arms painted on fine parchment, with an enormous cross erected on the bank of the Meta. The Guahibos, who, it is said, are some thousands in number, have become so insolent, that, at the time of our passage by Carichana, they sent word to the missionary that they would come on rafts, and burn his village. These rafts (valzas), which we had an opportunity of seeing, are scarcely three feet broad, and twelve feet long. They carry only two or three Indians; but fifteen or sixteen of these rafts are fastened to each other with the stems of the paullinia, the dolichos, and other creeping plants. It is difficult to conceive how these small craft remain tied together in passing the rapids. Many fugitives from the villages of the Casanare and the Apure have joined the Guahibos, and taught them the practice of eating beef, and preparing hides. The farms of San Vicente, Rubio, and San Antonio, have lost great numbers of their horned cattle by the incursions of the Indians, who also prevent travellers, as far as the junction of the Casanare, from sleeping on the shore in going up the Meta. It often happens, while the waters are low, that the traders of New Grenada, some of whom still visit the encampment of Pararuma, are killed by the poisoned arrows of the Guahibos.

From the mouth of the Meta, the Orinoco appeared to us to be freer of shoals and rocks. We navigated in a channel five hundred toises broad. The Indians remained rowing in the boat, without towing or pushing it forward with their arms, and wearying us with their wild cries. We passed the Canos of Uita and Endava on the west. It was night when we reached the Raudal de Tabaje. The Indians would not hazard passing the cataract; and we slept on a very incommodious spot, on the shelf of a rock, with a slope of more than eighteen degrees, and of which the crevices sheltered a swarm of bats. We heard the cries of the jaguar very near us during the whole night. They were answered by our great dog in lengthened howlings. I waited the appearance of the stars in vain: the sky was exceedingly black; and the hoarse sounds of the cascades of the Orinoco mingled with the rolling of the distant thunder.

Early in the morning of the 13th April we passed the rapids of Tabaje, and again disembarked. Father Zea, who accompanied us, desired to perform mass in the new Mission of San Borja, established two years before. We there found six houses inhabited by uncatechised Guahibos. They differ in nothing from the wild Indians. Their eyes, which are large and black, have more vivacity than those of the Indians who inhabit the ancient missions. We in vain offered them brandy; they would not even taste it. The faces of all the young girls were marked with round black spots; like the patches by which the ladies of Europe formerly imagined they set off the whiteness of their skins. The bodies of the Guahibos were not painted. Several of them had beards, of which they seemed proud; and, taking us by the chin, showed us by signs, that they were made like us. Their shape was in general slender. I was again struck, as I had been among the Salives and the Macos, with the little uniformity of features to be found among the Indians of the Orinoco. Their look is sad and gloomy; but neither stern nor ferocious. Without having any notion of the practices of the Christian religion, they behaved with the utmost decency at church. The Indians love to exhibit themselves; and will submit temporarily to any restraint or subjection, provided they are sure of drawing attention. At the moment of the consecration, they made signs to one another, to indicate beforehand that the priest was going to raise the chalice to his lips. With the exception of this gesture, they remained motionless and in imperturbable apathy.

The interest with which we examined these poor savages became perhaps the cause of the destruction of the mission. Some among them, who preferred a wandering life to the labours of agriculture, persuaded the rest to return to the plains of the Meta. They told them, that the white men would come back to San Borja, to take them away in the boats, and sell them as poitos, or slaves, at Angostura. The Guahibos awaited the news of our return from the Rio Negro by the Cassiquiare; and when they heard that we were arrived at the first great cataract, that of Atures, they all deserted, and fled to the savannahs that border the Orinoco on the west. The Jesuit Fathers had already formed a mission on this spot, and bearing the same name. No tribe is more difficult to fix to the soil than the Guahibos. They would rather feed on stale fish, scolopendras, and worms, than cultivate a little spot of ground. The other Indians say, that a Guahibo eats everything that exists, both on and under the ground.

In ascending the Orinoco more to the south, the heat, far from increasing, became more bearable. The air in the day was at 26 or 27.5 degrees; and at night, at 23.7. The water of the Orinoco retained its habitual temperature of 27.7 degrees. The torment of the mosquitos augmented severely, notwithstanding the decrease of heat. We never suffered so much from them as at San Borja. We could neither speak nor uncover our faces without having our mouths and noses filled with insects. We were surprised not to find the thermometer at 35 or 36 degrees; the extreme irritation of the skin made us believe that the air was scorching. We passed the night on the beach of Guaripo. The fear of the little caribe fish prevented us from bathing. The crocodiles we had met with this day were all of an extraordinary size, from twenty-two to twenty-four feet.

Our sufferings from the zancudos made us depart at five o'clock on the morning of the 14th. There are fewer insects in the strata of air lying immediately on the river, than near the edge of the forests. We stopped to breakfast at the island of Guachaco, or Vachaco, where the granite is immediately covered by a formation of sandstone, or conglomerate. This sandstone contains fragments of quartz, and even of feldspar, cemented by indurated clay. It exhibits little veins of brown iron-ore, which separate in laminae, or plates, of one line in thickness. We had already found these plates on the shores between Encaramada and Baraguan, where the missionaries had sometimes taken them for an ore of gold, and sometimes for tin. It is probable, that this secondary formation occupied formerly a larger space. Having passed the mouth of the Rio Parueni, beyond which the Maco Indians dwell, we spent the night on the island of Panumana. I could with difficulty take the altitudes of Canopus, in order to fix the longitude of the point, near which the river suddenly turns towards the west. The island of Panumana is rich in plants. We there again found those shelves of bare rock, those tufts of melastomas, those thickets of small shrubs, the blended scenery of which had charmed us in the plains of Carichana. The mountains of the Great Cataracts bounded the horizon towards the south-east. In proportion as we advanced, the shores of the Orinoco exhibited a more imposing and picturesque aspect.

CHAPTER 2.20.

THE MOUTH OF THE RIO ANAVENI. PEAK OF UNIANA. MISSION OF ATURES. CATARACT, OR RAUDAL OF MAPARA. ISLETS OF SURUPAMANA AND UIRAPURI.

The river of the Orinoco, in running from south to north, is crossed by a chain of granitic mountains. Twice confined in its course, it turbulently breaks on the rocks, that form steps and transverse dykes. Nothing can be grander than the aspect of this spot. Neither the fall of the Tequendama, near Santa Fe de Bogota, nor the magnificent scenes of the Cordilleras, could weaken the impression produced upon my mind by the first view of the rapids of Atures and of Maypures. When the spectator is so stationed that the eye can at once take in the long succession of cataracts, the immense sheet of foam and vapours illumined by the rays of the setting sun, the whole river seems as it were suspended over its bed.

Scenes so astonishing must for ages have fixed the attention of the inhabitants of the New World. When Diego de Todaz, Alfonzo de Herrera, and the intrepid Raleigh, anchored at the mouth of the Orinoco, they were informed by the Indians of the Great Cataracts, which they themselves had never visited, and which they even confounded with cascades farther to the east. Whatever obstacles the force of vegetation under the torrid zone may throw in the way of intercourse among nations, all that relates to the course of great rivers acquires a celebrity which extends to vast distances. The Orinoco, the Amazon, and the Uruguay, traverse, like inland arms of seas, in different directions, a land covered with forests, and inhabited by tribes, part of whom are cannibals. It is not yet two hundred years since civilization and the light of a more humane religion have pursued their way along the banks of these ancient canals traced by the hand of nature; long, however, before the introduction of agriculture, before communications for the purposes of barter were established among these scattered and often hostile tribes, the knowledge of extraordinary phenomena, of falls of water, of volcanic fires, and of snows resisting all the ardent heat of summer, was propagated by a thousand fortuitous circumstances. Three hundred leagues from the coast, in the centre of South America, among nations whose excursions do not extend to three days' journey, we find an idea of the ocean, and words that denote a mass of salt water extending as far as the eye can discern. Various events, which repeatedly occur in savage life, contribute to enlarge these conceptions. In consequence of the petty wars between neighbouring tribes, a prisoner is brought into a strange country, and treated as a poito or mero, that is to say, as a slave. After being often sold, he is dragged to new wars, escapes, and returns home; he relates what he has seen, and what he has heard from those whose tongue he has been compelled to learn. As on discovering a coast, we hear of great inland animals, so, on entering the valley of a vast river, we are surprised to find that savages, who are strangers to navigation, have acquired a knowledge of distant things. In the infant state of society, the exchange of ideas precedes, to a certain point, the exchange of productions.

The two great cataracts of the Orinoco, the celebrity of which is so far-spread and so ancient, are formed by the passage of the river across the mountains of Parima. They are called by the natives Mapara and Quittuna; but the missionaries have substituted for these names those of Atures and Maypures, after the names of the tribes which were first assembled together in the nearest villages. On the coast of Caracas, the two Great Cataracts are denoted by the simple appellation of the two Raudales, or rapids; a denomination which implies that the other falls of water, even the rapids of Camiseta and of Carichana, are not considered as worthy of attention when compared with the cataracts of Atures and Maypures.

These last, situated between five and six degrees of north latitude, and a hundred leagues west of the Cordilleras of New Grenada, in the meridian of Porto Cabello, are only twelve leagues distant from each other. It is surprising that their existence was not known to D'Anville, who, in his fine map of South America, marks the inconsiderable cascades of Marimara and San Borja, by the names of the rapids of Carichana and Tabaje. The Great Cataracts divide the Christian establishments of Spanish Guiana into two unequal parts. Those situated between the Raudal of Atures and the mouth of the river are called the Missions of the Lower Orinoco; the Missions of the Upper Orinoco comprehend the villages between the Raudal of Maypures and the mountains of Duida. The course of the Lower Orinoco, if we estimate the sinuosities at one-third of the distance in a direct line, is two hundred and sixty nautical leagues: the course of the Upper Orinoco, supposing its sources to be three degrees east of Duida, includes one hundred and sixty-seven leagues.

Beyond the Great Cataracts an unknown land begins. The country is partly mountainous and partly flat, receiving at once the confluents of the Amazon and the Orinoco. From the facility of its communications with the Rio Negro and Grand Para, it appears to belong still more to Brazil than to the Spanish colonies. None of the missionaries who have described the Orinoco before me, neither Father Gumilla, Gili, nor Caulin, had passed the Raudal of Maypures. We found but three Christian establishments above the Great Cataracts, along the shores of the Orinoco, in an extent of more than a hundred leagues; and these three establishments contained scarcely six or eight white persons, that is to say, persons of European race. We cannot be surprised that such a desert region should have been at all times the land of fable and fairy visions. There, according to the statements of certain missionaries, are found races of men, some of whom have an eye in the centre of the forehead, whilst others have dogs' heads, and mouths below their stomachs. There they pretend to have found all that the ancients relate of the Garamantes, of the Arimaspes, and of the Hyperboreans. It would be an error to suppose that these simple and often rustic missionaries had themselves invented all these exaggerated fictions; they derived them in great part from the recitals of the Indians. A fondness for narration prevails in the Missions, as it does at sea, in the East, and in every place where the mind seeks amusement. A missionary, from his vocation, is not inclined to scepticism; he imprints on his memory what the natives have so often repeated to him; and, when returned to Europe, and restored to the civilized world, he finds a pleasure in creating astonishment by a recital of facts which he thinks he has collected, and by an animated description of remote things. These stories, which the Spanish colonists call tales of travellers and of monks (cuentos de viageros y frailes), increase in improbability in proportion as you increase your distance from the forests of the Orinoco, and approach the coasts inhabited by the whites. When, at Cumana, Nueva Barcelona, and other seaports which have frequent communication with the Missions, you betray any sign of incredulity, you are reduced to silence by these few words: The fathers have seen it, but far above the Great Cataracts (mas arriba de los Raudales).

On the 15th of April, we left the island of Panumana at four in the morning, two hours before sunrise. The sky was in great part obscured, and lightnings flashed over dense clouds at more than forty degrees of elevation. We were surprised at not hearing thunder; but possibly this was owing to the prodigious height of the storm? It appears to us, that in Europe the electric flashes without thunder, vaguely called heat-lightning, are seen generally nearer the horizon. Under a cloudy sky, that sent back the radiant caloric of the soil, the heat was stifling; not a breath of wind agitated the foliage of the trees. The jaguars, as usual, had crossed the arm of the Orinoco by which we were separated from the shore, and we heard their cries extremely near. During the night the Indians had advised us to quit our station in the open air, and retire to a deserted hut belonging to the conucos of the inhabitants of Atures. They had taken care to barricade the opening with planks, a precaution which seemed to us superfluous; but near the Cataracts tigers are very numerous, and two years before, in these very conucos of Panumana, an Indian returning to his hut, towards the close of the rainy season, found a tigress settled in it with her two young. These animals had inhabited the dwelling for several months; they were dislodged from it with difficulty, and it was only after an obstinate combat that the former master regained possession of his dwelling. The jaguars are fond of retiring to deserted ruins, and I believe it is more prudent in general for a solitary traveller to encamp in the open air, between two fires, than to seek shelter in uninhabited huts.

On quitting the island of Panumana, we perceived on the western bank of the river the fires of an encampment of Guahibo savages. The missionary who accompanied us caused a few musket-shots to be fired in the air, which he said would intimidate them, and shew that we were in a state to defend ourselves. The savages most likely had no canoes, and were not desirous of troubling us in the middle of the river. We passed at sunrise the mouth of the Rio Anaveni, which descends from the eastern mountains. On its banks, now deserted, Father Olmos had established, in the time of the Jesuits, a small village of Japuins or Jaruros. The heat was so excessive that we rested a long time in a woody spot, to fish with a hook and line, and it was not without some trouble that we carried away all the fish we had caught. We did not arrive till very late at the foot of the Great Cataract, in a bay called the lower harbour (puerto de abaxo); and we followed, not without difficulty, in a dark night, the narrow path that leads to the Mission of Atures, a league distant from the river. We crossed a plain covered with large blocks of granite.

The little village of San Juan Nepomuceno de los Atures was founded by the Jesuit Francisco Gonzales, in 1748. In going up the river this is the last of the Christian missions that owe their origin to the order of St. Ignatius. The more southern establishments, those of Atabapo, of Cassiquiare, and of Rio Negro, were formed by the fathers of the Observance of St. Francis. The Orinoco appears to have flowed heretofore where the village of Atures now stands, and the flat savannah that surrounds the village no doubt formed part of the river. I saw to the east of the mission a succession of rocks, which seemed to have been the ancient shore of the Orinoco. In the lapse of ages the river has been impelled westward, in consequence of the accumulations of earth, which occur more frequently on the side of the eastern mountains, that are furrowed by torrents. The cataract bears the name of Mapara,* as we have mentioned above (* I am ignorant of the etymology of this word, which I believe means only a fall of water. Gili translates into Maypure a small cascade (raudalito) by uccamatisi mapara canacapatirri. Should we not spell this word matpara? mat being a radical of the Maypure tongue, and meaning bad (Hervas, Saggio N. 29). The radical par (para) is found among American tribes more than five hundred leagues distant from each other, the Caribs, Maypures, Brazilians, and Peruvians, in the words sea, rain, water, lake. We must not confound mapara with mapaja; this last word signifies, in Maypure and Tamanac, the papaw or melon-tree, no doubt on account of the sweetness of its fruit, for mapa means in the Maypure, as well as in the Peruvian and Omagua tongues, the honey of bees. The Tamanacs call a cascade, or raudal, in general uatapurutpe; the Maypures, uca.); while the name of the village is derived from that of the nation of Atures, now believed to be extinct. I find on the maps of the seventeenth century, Island and Cataract of Athule; which is the word Atures written according to the pronunciation of the Tamanacs, who confound, like so many other people, the consonants l and r. This mountainous region was so little known in Europe, even in the middle of the eighteenth century, that D'Anville, in the first edition of his South America, makes a branch issue from the Orinoco, near Salto de los Atures, and fall into the Amazon, to which branch he gives the name of Rio Negro.

Early maps, as well as Father Gumilla's work, place the Mission in latitude 1 degree 30 minutes. Abbe Gili gives it 3 degrees 50 minutes. I found, by meridian altitudes of Canopus and a of the Southern Cross, 5 degrees 38 minutes 4 seconds for the latitude; and by the chronometer 4 hours 41 minutes 17 seconds of longitude west of the meridian of Paris.

We found this small Mission in the most deplorable state. It contained, even at the time of the expedition of Solano, commonly called the expedition of the boundaries, three hundred and twenty Indians. This number had diminished, at the time of our passage by the Cataracts, to forty-seven; and the missionary assured us that this diminution became from year to year more sensible. He showed us, that in the space of thirty-two months only one marriage had been entered in the registers of the parish church. Two others had been contracted by uncatechised natives, and celebrated before the Indian Gobernador. At the first foundation of the Mission, the Atures, Maypures, Meyepures, Abanis, and Quirupas, had been assembled together. Instead of these tribes we found only Guahibos, and a few families of the nation of Macos. The Atures have almost entirely disappeared; they are no longer known, except by the tombs in the cavern of Ataruipe, which recall to mind the sepulchres of the Guanches at Teneriffe. We learned on the spot, that the Atures, as well as the Quaquas, and the Macos or Piaroas, belong to the great stock of the Salive nations; while the Maypures, the Abanis, the Parenis, and the Guaypunaves, are of the same race as the Cabres or Caveres, celebrated for their long wars with the Caribs. In this labyrinth of petty nations, divided from one another as the nations of Latium, Asia Minor, and Sogdiana, formerly were, we can trace no general relations but by following the analogy of tongues. These are the only monuments that have reached us from the early ages of the world; the only monuments, which, not being fixed to the soil, are at once moveable and lasting, and have as it were traversed time and space. They owe their duration, and the extent they occupy, much less to conquering and polished nations, than to those wandering and half-savage tribes, who, fleeing before a powerful enemy, carried along with them in their extreme wretchedness only their wives, their children, and the languages of their fathers.

Between the latitudes of 4 and 8 degrees, the Orinoco not only separates the great forest of the Parime from the bare savannahs of the Apure, Meta, and Guaviare, but also forms the boundary between tribes of very different manners. To the westward, over treeless plains, wander the Guahibos, the Chiricoas, and the Guamos; nations, proud of their savage independence, whom it is difficult to fix to the soil, or habituate to regular labour. The Spanish missionaries characterise them well by the name of Indios andantes (errant or vagabond Indians), because they are perpetually moving from place to place. To the east of the Orinoco, between the neighbouring sources of the Caura, Cataniapo, and Ventuari, live the Macos, the Salives, the Curacicanas, Parecas, and Maquiritares, mild, tranquil tribes, addicted to agriculture, and easily subjected to the discipline of the Missions. The Indian of the plains differs from the Indian of the forests in language as well as manners and mental disposition; both have an idiom abounding in spirited and bold terms; but the language of the former is harsher, more concise, and more impassioned; that of the latter, softer, more diffuse, and fuller of ambiguous expressions.

The Mission of Atures, like most of the Missions of the Orinoco, situated between the mouths of the Apure and the Atabapo, is composed of both the classes of tribes we have just described. We there find the Indians of the forests, and the Indians heretofore nomadic* (Indios monteros and Indios llaneros, or andantes). (* I employ the word nomadic as synonymous with wandering, and not in its primitive signification. The wandering nations of America (those of the indigenous tribes, it is to be understood) are never shepherds; they live by fishing and hunting, on the fruit of a few trees, the farinaceous pith of palm-trees, etc.) We visited with the missionary the huts of Macos, whom the Spaniards call Piraoas, and those of the Guahibos. The first indicated more love of order, cleanliness, and ease. The independent Macos (I do not designate them by the name of savages) have their rochelas, or fixed dwellings, two or three days' journey east of Atures, toward the sources of the little river Cataniapo. They are very numerous. Like most of the natives of the woods, they cultivate, not maize, but cassava; and they live in great harmony with the Christian Indians of the mission. The harmony was established and wisely cultivated by the Franciscan monk, Bernardo Zea. This alcalde of the reduced Macos quitted the village of Atures for a few months every year, to live in the plantations which he possessed in the midst of the forests near the hamlet of the independent Macos. In consequence of this peaceful intercourse, many of the Indios monteros came and established themselves some time ago in the mission. They asked eagerly for knives, fishing hooks, and those coloured glass beads, which, notwithstanding the positive prohibition of the priests, were employed not as necklaces, but as ornaments of the guayuco (perizoma). Having obtained what they sought, they returned to the woods, weary of the regulations of the mission. Epidemic fevers, which prevailed with violence at the entrance of the rainy season, contributed greatly to this unexpected flight. In 1799 the mortality was very considerable at Carichana, on the banks of the Meta, and at the Raudal of Atures. The Indian of the forest conceives a horror of the life of the civilized man, when, I will not say any misfortune befalls his family settled in the mission, but merely any disagreeable or unforeseen accident. Natives, who were neophytes, have been known to desert for ever the Christian establishments, on account of a great drought; as if this calamity would not have reached them equally in their plantations, had they remained in their primitive independence.

The fevers which prevail during a great part of the year in the villages of Atures and Maypures, around the two Great Cataracts of the Orinoco, render these spots highly dangerous to European travellers. They are caused by violent heats, in combination with the excessive humidity of the air, bad nutriment, and, if we may believe the natives, the pestilent exhalations rising from the bare rocks of the Raudales. These fevers of the Orinoco appeared to us to resemble those which prevail every year between New Barcelona, La Guayra, and Porto Cabello, in the vicinity of the sea; and which often degenerate into adynamic fevers. "I have had my little fever (mi calenturita) only eight months," said the good missionary of the Atures, who accompanied us to the Rio Negro; speaking of it as of an habitual evil, easy to be borne. The fits were violent, but of short duration. He was sometimes seized with them when lying along in the boat under a shelter of branches of trees, sometimes when exposed to the burning rays of the sun on an open beach. These tertian agues are attended with great debility of the muscular system; yet we find poor ecclesiastics on the Orinoco, who endure for several years these calenturitas, or tercianas: their effects are not so fatal as those which are experienced from fevers of much shorter duration in temperate climates.

I have just alluded to the noxious influence on the salubrity of the atmosphere, which is attributed by the natives, and even the missionaries, to the bare rocks. This opinion is the more worthy of attention, as it is connected with a physical phenomenon lately observed in different parts of the globe, and not yet sufficiently explained. Among the cataracts, and wherever the Orinoco, between the Missions of Carichana and of Santa Barbara, periodically washes the granitic rocks, they become smooth, black, and as if coated with plumbago. The colouring matter does not penetrate the stone, which is coarse-grained granite, containing a few solitary crystals of hornblende. Taking a general view of the primitive formation of Atures, we perceive, that, like the granite of Syene in Egypt, it is a granite with hornblende, and not a real syenite formation. Many of the layers are entirely destitute of hornblende. The black crust is 0.3 of a line in thickness; it is found chiefly on the quartzose parts. The crystals of feldspar sometimes preserve externally their reddish-white colour, and rise above the black crust. On breaking the stone with a hammer, the inside is found to be white, and without any trace of decomposition. These enormous stony masses appear sometimes in rhombs, sometimes under those hemispheric forms, peculiar to granitic rocks when they separate in blocks. They give the landscape a singularly gloomy aspect; their colour being in strong contrast with that of the foam of the river which covers them, and of the vegetation by which they are surrounded. The Indians say, that the rocks are burnt (or carbonized) by the rays of the sun. We saw them not only in the bed of the Orinoco, but in some spots as far as five hundred toises from its present shore, on heights which the waters now never reach even in their greatest swellings.

What is this brownish black crust, which gives these rocks, when they have a globular form, the appearance of meteoric stones? What idea can we form of the action of the water, which produces a deposit, or a change of colour, so extraordinary? We must observe, in the first place, that this phenomenon does not belong to the cataracts of the Orinoco alone, but is found in both hemispheres. At my return from Mexico in 1807, when I showed the granites of Atures and Maypures to M. Roziere, who had travelled over the valley of Egypt, the coasts of the Red Sea, and Mount Sinai, this learned geologist pointed out to me that the primitive rocks of the little cataracts of Syene display, like the rocks of the Orinoco, a glossy surface, of a blackish-grey, or almost leaden colour, and of which some of the fragments seem coated with tar. Recently, in the unfortunate expedition of Captain Tuckey, the English naturalists were struck with the same appearance in the yellalas (rapids and shoals) that obstruct the river Congo or Zaire. Dr. Koenig has placed in the British Museum, beside the syenites of the Congo, the granites of Atures, taken from a series of rocks which were presented by M. Bonpland and myself to the illustrious president of the Royal Society of London. "These fragments," says Mr. Koenig, "alike resemble meteoric stones; in both rocks, those of the Orinoco and of Africa, the black crust is composed, according to the analysis of Mr. Children, of the oxide of iron and manganese." Some experiments made at Mexico, conjointly with Senor del Rio, led me to think that the rocks of Atures, which blacken the paper in which they are wrapped,* contain, besides oxide of manganese, carbon, and supercarburetted iron. (* I remarked the same phenomenon from spongy grains of platina one or two lines in length, collected at the stream-works of Taddo, in the province of Choco. Having been wrapped up in white paper during a journey of several months, they left a black stain, like that of plumbago or supercarburetted iron.) At the Orinoco, granitic masses of forty or fifty feet thick are uniformly coated with these oxides; and, however thin these crusts may appear, they must nevertheless contain pretty considerable quantities of iron and manganese, since they occupy a space of above a league square.

It must be observed that all these phenomena of coloration have hitherto appeared in the torrid zone only, in rivers that have periodical overflowings, of which the habitual temperature is from twenty-four to twenty-eight centesimal degrees, and which flow, not over gritstone or calcareous rocks, but over granite, gneiss, and hornblende rocks. Quartz and feldspar scarcely contain five or six thousandths of oxide of iron and of manganese; but in mica and hornblende these oxides, and particularly that of iron, amount, according to Klaproth and Herrmann, to fifteen or twenty parts in a hundred. The hornblende contains also some carbon, like the Lydian stone and kieselschiefer. Now, if these black crusts were formed by a slow decomposition of the granitic rock, under the double influence of humidity and the tropical sun, how is it to be conceived that these oxides are spread so uniformly over the whole surface of the stony masses, and are not more abundant round a crystal of mica or hornblende than on the feldspar and milky quartz? The ferruginous sandstones, granites, and marbles, that become cinereous and sometimes brown in damp air, have an aspect altogether different. In reflecting upon the lustre and equal thickness of the crusts, we are rather inclined to think that this matter is deposited by the Orinoco, and that the water has penetrated even into the clefts of the rocks. Adopting this hypothesis, it may be asked whether the river holds the oxides suspended like sand and other earthy substances, or whether they are found in a state of chemical solution. The first supposition is less admissible, on account of the homogeneity of the crusts, which contain neither grains of sand, nor spangles of mica, mixed with the oxides. We must then recur to the idea of a chemical solution; and this idea is no way at variance with the phenomena daily observable in our laboratories. The waters of great rivers contain carbonic acid; and, were they even entirely pure, they would still be capable, in very great volumes, of dissolving some portions of oxide, or those metallic hydrates which are regarded as the least soluble. The mud of the Nile, which is the sediment of the matters which the river holds suspended, is destitute of manganese; but it contains, according to the analysis of M. Regnault, six parts in a hundred of oxide of iron; and its colour, at first black, changes to yellowish brown by desiccation and the contact of air. The mud consequently is not the cause of the black crusts on the rocks of Syene. Berzelius, who, at my request, examined these crusts, recognized in them, as in those of the granites of the Orinoco and River Congo, the union of iron and manganese. That celebrated chemist was of opinion that the rivers do not take up these oxides from the soil over which they flow, but that they derive them from their subterranean sources, and deposit them on the rocks in the manner of cementation, by the action of particular affinities, perhaps by that of the potash of the feldspar. A long residence at the cataracts of the Orinoco, the Nile, and the Rio Congo, and an examination of the circumstances attendant on this phenomenon of coloration, could alone lead to the complete solution of the problem we have discussed. Is this phenomenon independent of the nature of the rocks? I shall content myself with observing, in general, that neither the granitic masses remote from the ancient bed of the Orinoco, but exposed during the rainy season to the alternations of heat and moisture, nor the granitic rocks bathed by the brownish waters of the Rio Negro, assume the appearance of meteoric stones. The Indians say, that the rocks are black only where the waters are white. They ought, perhaps, to add, where the waters acquire great swiftness, and strike with force against the rocks of the banks. Cementation seems to explain why the crusts augment so little in thickness.

I know not whether it be an error, but in the Missions of the Orinoco, the neighbourhood of bare rocks, and especially of the masses that have crusts of carbon, oxide of iron, and manganese, are considered injurious to health. In the torrid zone, still more than in others, the people multiply pathogenic causes at will. They are afraid to sleep in the open air, if forced to expose the face to the rays of the full moon. They also think it dangerous to sleep on granite near the river; and many examples are cited of persons, who, after having passed the night on these black and naked rocks, have awakened in the morning with a strong paroxysm of fever. Without entirely lending faith to the assertions of the missionaries and natives, we generally avoided the laxas negras, and stretched ourselves on the beach covered with white sand, when we found no tree from which to suspend our hammocks. At Carichana, the village is intended to be destroyed, and its place changed, merely to remove it from the black rocks, or from a site where, for a space of more than ten thousand square toises, banks of bare granite form the surface. From similar motives, which must appear very chimerical to the naturalists of Europe, the Jesuits Olmo, Forneri, and Mellis, removed a village of Jaruros to three different spots, between the Raudal of Tabaje and the Rio Anaveni. I merely state these facts as they were related to me, because we are almost wholly ignorant of the nature of the gaseous mixtures which cause the insalubrity of the atmosphere. Can it be admitted that, under the influence of excessive heat and of constant humidity, the black crusts of the granitic rocks are capable of acting upon the ambient air, and producing miasmata with a triple basis of carbon, azote, and hydrogen? This I doubt. The granites of the Orinoco, it is true, often contain hornblende; and those who are accustomed to practical labour in mines are not ignorant that the most noxious exhalations rise from galleries wrought in syenitic and hornblende rocks: but in an atmosphere renewed every instant by the action of little currents of air, the effect cannot be the same as in a mine.

It is probably dangerous to sleep on the laxas negras, only because these rocks retain a very elevated temperature during the night. I have found their temperature in the day at 48 degrees, the air in the shade being at 29.7 degrees; during the night the thermometer on the rock indicated 36 degrees, the air being at 26 degrees. When the accumulation of heat in the stony masses has reached a stationary degree, these masses become at the same hours nearly of the same temperature. What they have acquired more in the day they lose at night by radiation, the force of which depends on the state of the surface of the radiating body, the interior arrangement of its particles, and, above all, on the clearness of the sky, that is, on the transparency of the atmosphere and the absence of clouds. When the declination of the sun varies very little, this luminary adds daily nearly the same quantities of heat, and the rocks are not hotter at the end than in the middle of summer. There is a certain maximum which they cannot pass, because they do not change the state of their surface, their density, or their capacity for caloric. On the shores of the Orinoco, on getting out of one's hammock during the night, and touching with the bare feet the rocky surface of the ground, the sensation of heat experienced is very remarkable. I observed pretty constantly, in putting the bulb of the thermometer in contact with the ledges of bare rocks, that the laxas negras are hotter during the day than the reddish-white granites at a distance from the river; but the latter cool during the night less rapidly than the former. It may be easily conceived that the emission and loss of caloric is more rapid in masses with black crusts than in those which abound in laminae of silvery mica. When walking between the hours of one and three in the afternoon, at Carichana, Atures, or Maypures, among those blocks of stone destitute of vegetable mould, and piled up to great heights, one feels a sensation of suffocation, as if standing before the opening of a furnace. The winds, if ever felt in those woody regions, far from bringing coolness, appear more heated when they have passed over beds of stone, and heaps of rounded blocks of granite. This augmentation of heat adds to the insalubrity of the climate.

Among the causes of the depopulation of the Raudales, I have not reckoned the small-pox, that malady which in other parts of America makes such cruel ravages that the natives, seized with dismay, burn their huts, kill their children, and renounce every kind of society. This scourge is almost unknown on the banks of the Orinoco, and should it penetrate thither, it is to be hoped that its effects may be immediately counteracted by vaccination, the blessings of which are daily felt along the coasts of Terra Firma. The causes which depopulate the Christian settlements are, the repugnance of the Indians for the regulations of the missions, insalubrity of climate, bad nourishment, want of care in the diseases of children, and the guilty practice of preventing pregnancy by the use of deleterious herbs. Among the barbarous people of Guiana, as well as those of the half-civilized islands of the South Sea, young wives are fearful of becoming mothers. If they have children, their offspring are exposed not only to the dangers of savage life, but also to other dangers arising from the strangest popular prejudices. When twins are born, false notions of propriety and family honour require that one of them should be destroyed. To bring twins into the world, say the Indians, is to be exposed to public scorn; it is to resemble rats, opossums, and the vilest animals, which bring forth a great number of young at a time. Nay, more, they affirm that two children born at the same time cannot belong to the same father. This is an axiom of physiology among the Salives; and in every zone, and in different states of society, when the vulgar seize upon an axiom, they adhere to it with more stedfastness than the better-informed men by whom it was first hazarded. To avoid the disturbance of conjugal tranquillity, the old female relations of the mother take care, that when twins are born one of them shall disappear. If a new-born infant, though not a twin, have any physical deformity, the father instantly puts it to death. They will have none but robust and well-made children, for deformities indicate some influence of the evil spirit Ioloquiamo, or the bird Tikitiki, the enemy of the human race. Sometimes children of a feeble constitution undergo the same fate. When the father is asked what is become of one of his sons, he will pretend that he has lost him by a natural death. He will disavow an action that appears to him blameable, but not criminal. "The poor boy," he will tell you, "could not follow us; we must have waited for him every moment; he has not been seen again; he did not come to sleep where we passed the night." Such is the candour and simplicity of manners—such the boasted happiness—of man in the state of nature! He kills his son to escape the ridicule of having twins, or to avoid journeying more slowly; in fact, to avoid a little inconvenience.

These acts of cruelty, I confess, are less frequent than they are believed to be; yet they occur even in the Missions, during the time when the Indians leave the village, to retire to the conucos of the neighbouring forests. It would be erroneous to attribute these actions to the state of polygamy in which the uncatechized Indians live. Polygamy no doubt diminishes the domestic happiness and internal union of families; but this practice, sanctioned by Ismaelism, does not prevent the people of the east from loving their children with tenderness. Among the Indians of the Orinoco, the father returns home only to eat, or to sleep in his hammock; he lavishes no caresses on his infants, or on his wives, whose office it is to serve him. Parental affection begins to display itself only when the son has become strong enough to take a part in hunting, fishing, and the agricultural labours of the plantations.

While our boat was unloading, we examined closely, wherever the shore could be approached, the terrific spectacle of a great river narrowed and reduced as it were to foam. I shall endeavour to paint, not the sensations we felt, but the aspect of a spot so celebrated among the scenes of the New World. The more imposing and majestic the objects we describe, the more essential it becomes to seize them in their smallest details, to fix the outline of the picture we would present to the imagination of the reader, and to describe with simplicity what characterises the great and imperishable monuments of nature.

The navigation of the Orinoco from its mouth as far as the confluence of the Anaveni, an extent of 260 leagues, is not impeded. There are shoals and eddies near Muitaco, in a cove that bears the name of the Mouth of Hell (Boca del Infierno); and there are rapids (raudalitos) near Carichana and San Borja; but in all these places the river is never entirely barred, as a channel is left by which boats can pass up and down.

In all this navigation of the Lower Orinoco travellers experience no other danger than that of the natural rafts formed by trees, which are uprooted by the river, and swept along in its great floods. Woe to the canoes that during the night strike against these rafts of wood interwoven with lianas! Covered with aquatic plants, they resemble here, as in the Mississippi, floating meadows, the chinampas or floating gardens of the Mexican lakes. The Indians, when they wish to surprise a tribe of their enemies, bring together several canoes, fasten them to each other with cords, and cover them with grass and branches, to imitate this assemblage of trunks of trees, which the Orinoco sweeps along in its middle current. The Caribs are accused of having heretofore excelled in the use of this artifice; at present the Spanish smugglers in the neighbourhood of Angostura have recourse to the same expedient to escape the vigilance of the custom-house officers.

After proceeding up the Orinoco beyond the Rio Anaveni, we find, between the mountains of Uniana and Sipapu, the Great Cataracts of Mapara and Quittuna, or, as they are more commonly called by the missionaries, the Raudales of Atures and Maypures. These bars, which extend from one bank to the other, present in general a similar aspect: they are composed of innumerable islands, dikes of rock, and blocks of granite piled on one another and covered with palm-trees. But, notwithstanding a uniformity of aspect, each of these cataracts preserves an individual character. The first, the Atures, is most easily passable when the waters are low. The Indians prefer crossing the second, the Maypures, at the time of great floods. Beyond the Maypures and the mouth of the Cano Cameji, the Orinoco is again unobstructed for the length of more than one hundred and sixty-seven leagues, or nearly to its source; that is to say, as far as the Raudalito of Guaharibos, east of the Cano Chiguire and the lofty mountains of Yumariquin.

Having visited the basins of the two rivers Orinoco and Amazon, I was singularly struck by the differences they display in their course of unequal extent. The falls of the Amazon, which is nearly nine hundred and eighty nautical leagues (twenty to a degree) in length, are pretty near its source in the first sixth of its total length, and five-sixths of its course are entirely free. We find the great falls of the Orinoco on a point far more unfavourable to navigation; if not at the half, at least much beyond the first third of its length. In both rivers it is neither the mountains, nor the different stages of flat lands lying over one another, whence they take their origin, that cause the cataracts; they are produced by other mountains, other ledges which, after a long and tranquil course, the rivers have to pass over, precipitating themselves from step to step.

The Amazon does not pierce its way through the principal chain of the Andes, as was affirmed at a period when it was gratuitously supposed that, wherever mountains are divided into parallel chains, the intermedial or central ridge must be more elevated than the others. This great river rises (and this is a point of some importance to geology) eastward of the western chain, which alone in this latitude merits the denomination of the high chain of the Andes. It is formed by the junction of the river Aguamiros with the Rio Chavinillo, which issues from the lake Llauricocha in a longitudinal valley bounded by the western and the intermedial chain of the Andes. To form an accurate idea of these hydrographical relations, it must be borne in mind that a division into three chains takes place in the colossal group or knot of the mountains of Pasco and Huanuco. The western chain, which is the loftiest, and takes the name of the Cordillera Real de Nieve, directs its course (between Huary and Caxatamba, Guamachuco and Luema, Micuipampa and Guangamarca) by the Nevados of Viuda, Pelagatos, Moyopata, and Huaylillas, and by the Paramos of Guamani and Guaringa, towards the town of Loxa. The intermedial chain separates the waters of the Upper Maranon from those of the Guallaga, and over a long space reaches only the small elevation of a thousand toises; it enters the region of perpetual snow to the south of Huanuco in the Cordillera of Sasaguanca. It stretches at first northward by Huacrachuco, Chachapoyas, Moyobamba, and the Paramo of Piscoguannuna; then it progressively lowers toward Peca, Copallin, and the Mission of Santiago, at the eastern extremity of the province of Jaen de Bracamoros. The third, or easternmost chain, skirts the right bank of the Rio Guallaga, and loses itself in the seventh degree of latitude. So long as the Amazon flows from south to north in the longitudinal valley, between two chains of unequal height (that is, from the farms of Quivilla and Guancaybamba, where the river is crossed on wooden bridges, as far as the confluence of the Rio Chinchipe), there are neither bars, nor any obstacle whatever to the navigation of boats. The falls of water begin only where the Amazon turns toward the east, crossing the intermedial chain of the Andes, which widens considerably toward the north. It meets with the first rocks of red sandstone, or ancient conglomerate, between Tambillo and the Pongo of Rentema (near which I measured the breadth, depth, and swiftness of the waters), and it leaves the rocks of red sandstone east of the famous strait of Manseriche, near the Pongo of Tayuchuc, where the hills rise no higher than forty or fifty toises above the level of its waters. The river does not reach the most easterly chain, which bounds the Pampas del Sacramento. From the hills of Tayuchuc as far as Grand Para, during a course of more than seven hundred and fifty leagues, the navigation is free from obstacles. It results from this rapid sketch, that, if the Maranon had not to pass over the hilly country between Santiago and Tomependa (which belongs to the central chain of the Andes) it would be navigable from its mouth as far as Pumpo, near Piscobamba in the province of Conchucos, forty-three leagues north of its source.

We have just seen that, in the Orinoco, as in the Amazon, the great cataracts are not found near the sources of the rivers. After a tranquil course of more than one hundred and sixty leagues from the little Raudal of Guaharibos, east of Esmeralda, as far as the mountains of Sipapu, the river, augmented by the waters of the Jao, the Ventuari, the Atabapo, and the Guaviare, suddenly changes its primitive direction from east to west, and runs from south to north: then, in crossing the land-strait* in the plains of Meta, (* This strait, which I have several times mentioned, is formed by the Cordilleras of the Andes of New Granada, and the Cordillera of Parima.) meets the advanced buttresses of the Cordillera of Parima. This obstacle causes cataracts far more considerable, and presents greater impediments to navigation, than all the Pongos of the Upper Maranon, because they are proportionally nearer to the mouth of the river. These geographical details serve to prove, in the instances of the two greatest rivers of the New World, first, that it cannot be ascertained in an absolute manner that, beyond a certain number of toises, or a certain height above the level of the sea, rivers are not navigable; secondly, that the rapids are not always occasioned, as several treatises of general topography affirm, by the height of the first obstacles, by the first lines of ridges which the waters have to surmount near their sources.

The most northern of the great cataracts of the Orinoco is the only one bounded on each side by lofty mountains. The left bank of the river is generally lower, but it makes part of a plane which rises again west of Atures, towards the Peak of Uniana, a pyramid nearly three thousand feet high, and placed on a wall of rock with steep slopes. The situation of this solitary peak in the plain contributes to render its aspect more imposing and majestic. Near the Mission, in the country which surrounds the cataract, the aspect of the landscape varies at every step. Within a small space we find all that is most rude and gloomy in nature, united with an open country and lovely pastoral scenery. In the physical, as in the moral world, the contrast of effects, the comparison of what is powerful and menacing with what is soft and peaceful, is a never-failing source of our pleasures and our emotions.

I shall here repeat some scattered features of a picture which I traced in another work shortly after my return to Europe.* (* Views of Nature page 153 Bohn's edition.) The savannahs of Atures, covered with slender plants and grasses, are really meadows resembling those of Europe. They are never inundated by the rivers, and seem as if waiting to be ploughed by the hand of man. Notwithstanding their extent, these savannahs do not exhibit the monotony of our plains; they surround groups of rocks and blocks of granite piled on one another. On the very borders of these plains and this open country, glens are seen scarcely lighted by the rays of the setting sun, and hollows where the humid soil, loaded with arums, heliconias, and lianas, manifests at every step the wild fecundity of nature. Everywhere, just rising above the earth, appear those shelves of granite completely bare, which we saw at Carichana, and which I have already described. Where springs gush from the bosom of these rocks, verrucarias, psoras, and lichens are fixed on the decomposed granite, and have there accumulated mould. Little euphorbias, peperomias, and other succulent plants, have taken the place of the cryptogamous tribes; and evergreen shrubs, rhexias, and purple-flowered melastomas, form verdant isles amid desert and rocky plains. The distribution of these spots, the clusters of small trees with coriaceous and shining leaves scattered in the savannahs, the limpid rills that dig channels across the rocks, and wind alternately through fertile places and over bare shelves of granite, all call to mind the most lovely and picturesque plantations and pleasure-grounds of Europe. We seem to recognise the industry of man, and the traces of cultivation, amid this wild scenery.

The lofty mountains that bound the horizon on every side, contribute also, by their forms and the nature of their vegetation, to give an extraordinary character to the landscape. The average height of these mountains is not more than seven or eight hundred feet above the surrounding plains. Their summits are rounded, as for the most part in granitic mountains, and covered with thick forests of the laurel-tribe. Clusters of palm-trees,* (* El cucurito.) the leaves of which, curled like feathers, rise majestically at an angle of seventy degrees, are dispersed amid trees with horizontal branches; and their bare trunks, like columns of a hundred or a hundred and twenty feet high, shoot up into the air, and when seen in distinct relief against the azure vault of the sky, they resemble a forest planted upon another forest. When, as the moon was going down behind the mountains of Uniana, her reddish disc was hidden behind the pinnated foliage of the palm-trees, and again appeared in the aerial zone that separates the two forests, I thought myself transported for a few moments to the hermitage which Bernardin de Saint-Pierre has described as one of the most delicious scenes of the Isle of Bourbon, and I felt how much the aspect of the plants and their groupings resembled each other in the two worlds. In describing a small spot of land in an island of the Indian Ocean, the inimitable author of Paul and Virginia has sketched the vast picture of the landscape of the tropics. He knew how to paint nature, not because he had studied it scientifically, but because he felt it in all its harmonious analogies of forms, colours, and interior powers.

East of the Atures, near these rounded mountains crowned, as it were, by two superimposed forests of laurels and palms, other mountains of a very different aspect arise. Their ridge is bristled with pointed rocks, towering like pillars above the summits of the trees and shrubs. These effects are common to all granitic table-lands, at the Harz, in the metalliferous mountains of Bohemia, in Galicia, on the limit of the two Castiles, or wherever a granite of new formation appears above the ground. The rocks, which are at distances from each other, are composed of blocks piled together, or divided into regular and horizontal beds. On the summits of those situated near the Orinoco, flamingos, soldados,* (* The soldado (soldier) is a large species of heron.) and other fishing-birds perch, and look like men posted as sentinels. This resemblance is so striking, that the inhabitants of Angostura, soon after the foundation of their city, were one day alarmed by the sudden appearance of soldados and garzas, on a mountain towards the south. They believed they were menaced with an attack of Indios monteros (wild Indians called mountaineers); and the people were not perfectly tranquilized, till they saw the birds soaring in the air, and continuing their migration towards the mouths of the Orinoco.

The fine vegetation of the mountains spreads over the plains, wherever the rock is covered with mould, We generally find that this black mould, mixed with fibrous vegetable matter, is separated from the granitic rock by a layer of white sand. The missionary assured us that verdure of perpetual freshness prevails in the vicinity of the cataracts, produced by the quantity of vapour which the river, broken into torrents and cascades for the length of three or four thousand toises, diffuses in the air.

We had not heard thunder more than once or twice at Atures, and the vegetation everywhere displayed that vigorous aspect, that brilliancy of colour, seen on the coast only at the end of the rainy season. The old trees were decorated with beautiful orchideas,* (* Cymbidium violaceum, Habenaria angustifolia, etc.) yellow bannisterias, blue-flowered bignonias, peperomias, arums, and pothoses. A single trunk displays a greater variety of vegetable forms than are contained within an extensive space of ground in our countries. Close to the parasite plants peculiar to very hot climates we observed, not without surprise, in the centre of the torrid zone, and near the level of the sea, mosses resembling in every respect those of Europe. We gathered, near the Great Cataract of Atures, that fine specimen of Grimmia* with fontinalis leaves, which has so much fixed the attention of botanists. (* Grimmia fontinaloides. See Hooker's Musci Exotici, 1818 tab. 2. The learned author of the Monography of the Jungermanniae (Mr. Jackson Hooker), with noble disinterestedness, published at his own expense, in London, the whole collection of cryptogamous plants, brought by Bonpland and Humboldt from the equinoctial regions of America.) It is suspended to the branches of the loftiest trees. Of the phaenerogamous plants, those which prevail in the woody spots are the mimosa, ficus, and laurinea. This fact is the more characteristic as, according to the observations of Mr. Brown, the laurineae appear to be almost entirely wanting on the opposite continent, in the equinoctial part of Africa. Plants that love humidity adorn the scenery surrounding the cataracts. We there find in the plains groups of heliconias and other scitamineae with large and glossy leaves, bamboos, and the three palm-trees, the murichi, jagua, and vadgiai, each of which forms a separate group. The murichi, or mauritia with scaly fruits, is the celebrated sago-tree of the Guaraon Indians. It has palmate leaves, and has no relation to the palm-trees with pinnate and curled leaves; to the jagua, which appears to be a species of the cocoa-tree; or to the vadgiai or cucurito, which may be assimilated to the fine species Oreodoxa. The cucurito, which is the palm most prevalent around the cataracts of the Atures and Maypures, is remarkable for its stateliness. Its leaves, or rather its palms, crown a trunk of eighty or one hundred feet high; their direction is almost perpendicular when young, as well as at their full growth, the points only being incurvated. They look like plumes of the most soft and verdant green. The cucurito, the pirijao, the fruit of which resembles the apricot, the Oreodoxa regia or palma real of the island of Cuba, and the ceroxylon of the high Andes, are the most majestic of all the palm-trees we saw in the New World. As we advance toward the temperate zone, the plants of this family decrease in size and beauty. What a difference between the species we have just mentioned, and the date-tree of the East, which unfortunately has become to the landscape painters of Europe the type of a group of palm-trees!

It is not suprising that persons who have travelled only in the north of Africa, in Sicily, or in Spain, cannot conceive that, of all large trees, the palm is the most grand and beautiful in form. Incomplete analogies prevent Europeans from having a just idea of the aspect of the torrid zone. All the world knows, for instance, that this zone is embellished by the contrasts exhibited in the foliage of the trees, and particularly by the great number of those with pinnate leaves. The ash, the service-tree, the inga, the acacia of the United States, the gleditsia, the tamarind, the mimosa, the desmanthus, have all pinnate leaves, with foliolae more or less long, slender, tough, and shining. But can a group of ash-trees, of service-trees, or of sumach, recall the picturesque effect of tamarinds or mimosas, when the azure of the sky appears through their small, slender, and delicately pinnated leaves? These considerations are more important than they may at first seem. The forms of plants determine the physiognomy of nature; and this physiognomy influences the moral dispositions of nations. Every type comprehends species, which, while exhibiting the same general appearance, differ in the varied development of the similar organs. The palm-trees, the scitamineae, the malvaceae, the trees with pinnate leaves, do not all display the same picturesque beauties; and generally the most beautiful species of each type, in plants as in animals, belong to the equinoctial zone.

The proteaceae,* (* Rhopalas, which characterise the vegetation of the Llanos.) crotons, agaves, and the great tribe of the cactuses, which inhabit exclusively the New World, disappear gradually, as we ascend the Orinoco above the Apure and the Meta. It is, however, the shade and humidity, rather than the distance from the coast, which oppose the migration of the cactuses southward. We found forests of them mingled with crotons, covering a great space of arid land to the east of the Andes, in the province of Bracamoros, towards the Upper Maranon. The arborescent ferns seem to fail entirely near the cataracts of the Orinoco; we found no species as far as San Fernando de Atabapo, that is, to the confluence of the Orinoco and the Guaviare.

Having now examined the vicinity of the Atures, it remains for me to speak of the rapids themselves, which occur in a part of the valley where the bed of the river, deeply ingulfed, has almost inaccessible banks. It was only in a very few spots that we could enter the Orinoco to bathe, between the two cataracts, in coves where the waters have eddies of little velocity. Persons who have dwelt in the Alps, the Pyrenees, or even the Cordilleras, so celebrated for the fractures and the vestiges of destruction which they display at every step, can scarcely picture to themselves, from a mere narration, the state of the bed of the river. It is traversed, in an extent of more than five miles, by innumerable dikes of rock, forming so many natural dams, so many barriers resembling those of the Dnieper, which the ancients designated by the name of phragmoi. The space between the rocky dikes of the Orinoco is filled with islands of different dimensions; some hilly, divided into several peaks, and two or three hundred toises in length, others small, low, and like mere shoals. These islands divide the river into a number of torrents, which boil up as they break against the rocks. The jaguas and cucuritos with plumy leaves, with which all the islands are covered, seem like groves of palm-trees rising from the foamy surface of the waters. The Indians, whose task it is to pass the boats empty over the raudales, distinguish every shelf, and every rock, by a particular name. On entering from the south you find first the Leap of the Toucan (Salto del Piapoco); and between the islands of Avaguri and Javariveni is the Raudal of Javariveni, where, on our return from Rio Negro, we passed some hours amid the rapids, waiting for our boat. A great part of the river appeared dry. Blocks of granite are heaped together, as in the moraines which the glaciers of Switzerland drive before them. The river is ingulfed in caverns; and in one of these caverns we heard the water roll at once over our heads and beneath our feet. The Orinoco seems divided into a multitude of arms or torrents, each of which seeks to force a passage through the rocks. We were struck with the little water to be seen in the bed of the river, the frequency of subterraneous falls, and the tumult of the waters breaking on the rocks in foam.

Cuncta fremunt undis; ac multo murmure montis Spumeus invictis canescit fluctibus amnis.* (* Lucan, Pharsalia lib 10 v 132.)

Having passed the Raudal of Javariveni (I name here only the principal falls) we come to the Raudal of Canucari, formed by a ledge of rocks uniting the islands of Surupamana and Uirapuri. When the dikes, or natural dams, are only two or three feet high, the Indians venture to descend them in boats. In going up the river, they swim on before, and if, after many vain efforts, they succeed in fixing a rope to one of the points of rock that crown the dike, they then, by means of that rope, draw the bark to the top of the raudal. The bark, during this arduous task, often fills with water; at other times it is stove against the rocks, and the Indians, their bodies bruised and bleeding, extricate themselves with difficulty from the whirlpools, and reach, by swimming, the nearest island. When the steps or rocky barriers are very high, and entirely bar the river, light boats are carried on shore, and with the help of branches of trees placed under them to serve as rollers, they are drawn as far as the place where the river again becomes navigable. This operation is seldom necessary when the water is high. We cannot speak of the cataracts of the Orinoco without recalling to mind the manner heretofore employed for descending the cataracts of the Nile, of which Seneca has left us a description probably more poetical than accurate. I shall cite the passage, which traces with fidelity what may be seen every day at Atures, Maypures, and in some pongos of the Amazon. "Two men embark in a small boat; one steers, and the other empties it as it fills with water. Long buffeted by the rapids, the whirlpools, and the contrary currents, they pass through the narrowest channels, avoid the shoals, and rush down the whole river, guiding the course of the boat in its accelerated fall." (Nat. Quaest. lib 4 cap 2 edit. Elzev. tome 2 page 609.)

In hydrographic descriptions of countries, the vague names of cataracts, cascades, falls, and rapids,* (* The corresponding terms in use among the people of South America, are saltos, chorros, pongos, cachoeiras, and raudales.) denoting those tumultuous movements of water which arise from very different circumstances, are generally confounded with one another. Sometimes a whole river precipitating itself from a great height, and by one single fall, renders navigation impossible. Such is the majestic fall of the Rio Tequendama, which I have represented in my Views of the Cordilleras; such are the falls of Niagara and of the Rhine, much less remarkable for their elevation, than for the mass of water they contain. Sometimes stony dikes of small height succeed each other at great distances, and form distinct falls; such are the cachoeiras of the Rio Negro and the Rio Madeira, the saltos of the Rio Cauca, and the greater part of the pongos that are found in the Upper Maranon, from the confluence of the Chinchipe to the village of San Borja. The highest and most formidable of these pongos, which are descended on rafts, that of Mayasi, is however only three feet in height. Sometimes small rocky dikes are so near each other that they form for several miles an uninterrupted succession of cascades and whirlpools (chorros and remolinos); these are properly what are called rapids (raudales). Such are the yellalas, or rapids of the River Zaire,* or Congo, which Captain Tuckey has recently made known to us (* Voyage to explore the River Zaire, 1818, pages 152, 327, 340. What the inhabitants of Upper Egypt and Nubia call chellal in the Nile, is called yellala in the River Congo. This analogy between words signifying rapids is remarkable, on account of the enormous distance of the yellalas of the Congo from the chellal and djenadel of the Nile. Did the word chellal penetrate with the Moors into the west of Africa? If, with Burckhardt, we consider the origin of this word as Arabic (Travels in Nubia, 1819), it must be derived from the root challa, to disperse, which forms chelil, water falling through a narrow channel.); the rapids of the Orange River in Africa, above Pella; and the falls of the Missouri, which are four leagues in length, where the river issues from the Rocky Mountains. Such also are the cataracts of Atures and Maypures; the only cataracts which, situated in the equinoctial region of the New World, are adorned with the noble growth of palm-trees. At all seasons they exhibit the aspect of cascades, and present the greatest obstacles to the navigation of the Orinoco, while the rapids of the Ohio and of Upper Egypt are scarcely visible at the period of floods. A solitary cataract, like Niagara, or the cascade of Terni, affords a grand but single picture, varying only as the observer changes his place. Rapids, on the contrary, especially when adorned with large trees, embellish a landscape during a length of several leagues. Sometimes the tumultuous movement of the waters is caused only by extraordinary contractions of the beds of the rivers. Such is the angostura of Carare, in the river Magdalena, a strait that impedes communication between Santa Fe de Bogota and the coast of Carthagena; and such is the pongo of Manseriche, in the Upper Maranon.

The Orinoco, the Rio Negro, and almost all the confluents of the Amazon and the Maranon, have falls or rapids, either because they cross the mountains where they take rise, or because they meet other mountains in their course. If the Amazon, from the pongo of Manseriche (or, to speak with more precision, from the pongo of Tayuchuc) as far as its mouth, a space of more than seven hundred and fifty leagues, exhibit no tumultuous movement of the waters, the river owes this advantage to the uniform direction of its course. It flows from west to east in a vast plain, forming a longitudinal valley between the mountains of Parima and the great mass of the mountains of Brazil.

I was surprised to find by actual measurement that the rapids of the Orinoco, the roar of which is heard at the distance of more than a league, and which are so eminently picturesque from the varied appearance of the waters, the palm-trees and the rocks, have not probably, on their whole length, a height of more than twenty-eight feet perpendicular. In reflecting on this, we find that it is a great deal for rapids, while it would be very little for a single cataract. The Yellalas of the Rio Congo, in the contracted part of the river from Banza Noki as far as Banza Inga, furnish, between the upper and lower levels, a much more considerable difference; but Mr. Barrow observes, that among the great number of these rapids there is one fall, which alone is thirty feet high. On the other hand, the famous pongos of the river Amazon, so dangerous to go up, the falls of Rentema, of Escurrebragas, and of Mayasi, are but a few feet in perpendicular height. Those who are engaged in hydraulic works know the effect that a bar of eighteen or twenty inches' height produces in a great river. The whirling and tumultuous movement of the water does not depend solely on the greatness of partial falls; what determines the force and impetuosity is the nearness of these falls, the steepness of the rocky ledges, the returning sheets of water which strike against and surmount each other, the form of the islands and shoals, the direction of the counter-currents, and the contraction and sinuosity of the channels through which the waters force a passage between two adjacent levels. In two rivers equally large, that of which the falls have least height may sometimes present the greatest dangers and the most impetuous movements.

It is probable that the river Orinoco loses part of its waters in the cataracts, not only by increased evaporation, caused by the dispersion of minute drops in the atmosphere, but still more by filtrations into the subterraneous cavities. These losses, however, are not very perceptible when we compare the mass of waters entering into the raudal with that which issues out near the mouth of the Rio Anaveni. It was by a similar comparison that the existence of subterraneous cavities in the yellalas or rapids of the river Congo was discovered. The pongo of Manseriche, which ought rather to be called a strait than a fall, ingulfs, in a manner not yet sufficiently explored, a part of the waters and all the floating wood of the Upper Maranon.

The spectator, seated on the bank of the Orinoco, with his eyes fixed on those rocky dikes, is naturally led to inquire whether, in the lapse of ages, the falls change their form or height. I am not much inclined to believe in such effects of the shock of water against blocks of granite, and in the erosion of siliceous matter. The holes narrowed toward the bottom, the funnels that are discovered in the raudales, as well as near so many other cascades in Europe, are owing only to the friction of the sand, and the movement of quartz pebbles. We saw many such, whirled perpetually by the current at the bottom of the funnels, and contributing to enlarge them in every direction. The pongos of the river Amazon are easily destroyed, because the rocky dikes are not granite, but a conglomerate, or red sandstone with large fragments. A part of the pongo of Rentama was broken down eighty years ago, and the course of the waters being interrupted by a new bar, the bed of the river remained dry for some hours, to the great astonishment of the inhabitants of the village of Payaya, seven leagues below the pongo. The Indians of Atures assert (and in this their testimony is contrary to the opinion of Caulin) that the rocks of the raudal preserve the same aspect; but that the partial torrents into which the great river divides itself as it passes through the heaped blocks of granite, change their direction, and carry sometimes more, sometimes less water towards one or the other bank; but the causes of these changes may be very remote from the cataracts, for in the rivers that spread life over the surface of the globe, as in the arteries by which it is diffused through organized bodies, all the movements are propagated to great distances. Oscillations, that at first seem partial, react on the whole liquid mass contained in the trunk as well as in its numerous ramifications.

Some of the Missionaries in their writings have alleged that the inhabitants of Atures and Maypures have been struck with deafness by the noise of the Great Cataracts, but this is untrue. When the noise is heard in the plain that surrounds the mission, at the distance of more than a league, you seem to be near a coast skirted by reefs and breakers. The noise is three times as loud by night as by day, and gives an inexpressible charm to these solitary scenes. What can be the cause of this increased intensity of sound, in a desert where nothing seems to interrupt the silence of nature? The velocity of the propagation of sound, far from augmenting, decreases with the lowering of the temperature. The intensity diminishes in air agitated by a wind which is contrary to the direction of the sound; it diminishes also by dilatation of the air, and is weaker in the higher than in the lower regions of the atmosphere, where the number of particles of air in motion is greater in the same radius. The intensity is the same in dry air, and in air mingled with vapours; but it is feebler in carbonic acid gas than in mixtures of azote and oxygen. From these facts, which are all we know with any certainty, it is difficult to explain a phenomenon observed near every cascade in Europe, and which, long before our arrival in the village of Atures, had struck the missionary and the Indians.

It may be thought that, even in places not inhabited by man, the hum of insects, the song of birds, the rustling of leaves agitated by the feeblest winds, occasion during the day a confused noise, which we perceive the less because it is uniform, and constantly strikes the ear. Now this noise, however slightly perceptible it may be, may diminish the intensity of a louder noise; and this diminution may cease if during the calm of the night the song of birds, the hum of insects, and the action of the wind upon the leaves be interrupted. But this reasoning, even admitting its justness, can scarcely be applied to the forests of the Orinoco, where the air is constantly filled by an innumerable quantity of mosquitos, where the hum of insects is much louder by night than by day, and where the breeze, if ever it be felt, blows only after sunset.

I rather think that the presence of the sun acts upon the propagation and intensity of sound by the obstacles met in currents of air of different density, and by the partial undulations of the atmosphere arising from the unequal heating of different parts of the soil. In calm air, whether dry or mingled with vesicular vapours equally distributed, sound-waves are propagated without difficulty. But when the air is crossed in every direction by small currents of hotter air, the sonorous undulation is divided into two undulations where the density of the medium changes abruptly; partial echoes are formed that weaken the sound, because one of the streams comes back upon itself; and those divisions of undulations take place of which M. Poisson has developed the theory with great sagacity.* (* Annales de Chimie tome 7 page 293.) It is not therefore the movement of the particles of air from below to above in the ascending current, or the small oblique currents that we consider as opposing by a shock the propagation of the sonorous undulations. A shock given to the surface of a liquid will form circles around the centre of percussion, even when the liquid is agitated. Several kinds of undulations may cross each other in water, as in air, without being disturbed in their propagation: little movements may, as it were, ride over each other, and the real cause of the less intensity of sound during the day appears to be the interpretation of homogeneity in the elastic medium. During the day there is a sudden interruption of density wherever small streamlets of air of a high temperature rise over parts of the soil unequally heated. The sonorous undulations are divided, as the rays of light are refracted and form the mirage wherever strata of air of unequal density are contiguous. The propagation of sound is altered when a stratum of hydrogen gas is made to rise in a tube closed at one end above a stratum of atmospheric air; and M. Biot has well explained, by the interposition of bubbles of carbonic acid gas, why a glass filled with champagne is not sonorous so long as that gas is evolved, and passing through the strata of the liquid.