passage from trinidad de cuba to rio sinu. carthagena. air volcanoes of turbaco. canal of mahates.

on the morning of the 17th of march, we came within sight of the most eastern island of the group of the lesser caymans. comparing the reckoning with the chronometric longitude, i ascertained that the currents had borne us in seventeen hours twenty miles westward. the island is called by the english pilots cayman-brack, and by the spanish pilots, cayman chico oriental. it forms a rocky wall, bare and steep towards the south and south-east. the north and north-west part is low, sandy, and scantily covered with vegetation. the rock is broken into narrow horizontal ledges. from its whiteness and its proximity to the island of cuba, i supposed it to be of jura limestone. we approached the eastern extremity of cayman-brack within the distance of 400 toises. the neighbouring coast is not entirely free from danger and breakers; yet the temperature of the sea had not sensibly diminished at its surface. the chronometer of louis berthoud gave me 82° 7′ 37′ for the longitude of the eastern cape of cayman-brack. the latitude reduced by the reckoning on the rhumbs of wind at the meridian observation, appeared to me to be 19° 40′ 50′.

as long as we were within sight of the rock of cayman-brack sea-turtles of extraordinary dimensions swam round our vessel. the abundance of these animals led columbus to give the whole group of the caymans the name of penascales de las tortugas (rocks of the turtles.) our sailors would have thrown themselves into the water to catch some of these animals; but the numerous sharks that accompany them rendered the attempt too perilous. the sharks fixed their jaws on great iron hooks which were flung to them; these hooks were very sharp and (for want of anzuelos encandenados*) they were tied to cords: the sharks were in this manner drawn up half the length of their bodies; and we were surprised to see that those which had their mouths wounded and bleeding continued to seize the bait over and over again during several hours.* at the sight of these voracious fish the sailors in a spanish vessel always recollect the local fable of the coast of venezuela, which describes the benediction of a bishop as having softened the habits of the sharks, which are everywhere else the dread of mariners. do these wild sharks of the port of la guayra specifically differ from those which are so formidable in the port of the havannah? and do the former belong to the group of emissoles with small sharp teeth, which cuvier distinguishes from the melandres, by the name of musteli?

[* fish-hooks with chains.]

[* vidimus quoque squales, quotiescunque, hamo icti, dimidia parte corporis e fluctibus extrahebantur, cito alvo stercus emittere haud absimile excrementis caninis. commovebat intestina (ut arbitramur) subitus pavor. although the form and number of teeth change with age, and the teeth appear successively in the shark genus, i doubt whether don antonio ulloa be correct in stating that the young sharks have two, and the old ones four rows of grinders. these, like many other sea-fish, are easily accustomed to live in fresh water, or in water slightly briny. it is observed that sharks (tiburones) abound of late in the laguna of maracaybo, whither they have been attracted by the dead bodies thrown into the water after the frequent battles between the spanish royalists and the columbian republicans.]

the wind freshened more and more from the south-east, as we advanced in the direction of cape negril and the western extremity of the great bank of la vibora. we were often forced to diverge from our course; and, on account of the extreme smallness of our vessel, we were almost constantly under water. on the 18th of march at noon we found ourselves in latitude 18° 17′ 40′, and in 81° 50′ longitude. the horizon, to the height of 50°, was covered with those reddish vapours so common within the tropics, and which never seem to affect the hygrometer at the surface of the globe. we passed fifty miles west of cape negril on the south, nearly at the point where several charts indicate an insulated flat of which the position is similar to that of sancho pardo, opposite to cape san antonio de cuba. we saw no change in the bottom. it appears that the rocky shoal at a depth of four fathoms, near cape negril, has no more existence than the rock (cascabel) itself, long believed to mark the western extremity of la vibora (pedro bank, portland rock or la sola), marking the eastern extremity. on the 19th of march, at four in the afternoon, the muddy colour of the sea denoted that we had reached that part of the bank of la vibora where we no longer find fifteen, and indeed scarcely nine or ten, fathoms of water. our chronometric longitude was 81° 3′; and our latitude probably below 17°. i was surprised that, at the noon observation, at 17° 7′ of latitude, we yet perceived no change in the colour of the water. spanish vessels going from batabano or trinidad de cuba to carthagena, usually pass over the bank of la vibora, on its western side, at between fifteen and sixteen fathoms water. the dangers of the breakers begin only beyond the meridian 80° 45′ west longitude. in passing along the bank on its southern limit, as pilots often do in proceeding from cumana or other parts of the mainland, to the great caymnan or cape san antonio, they need not ascend along the rocks, above 16° 47′ latitude. fortunately the currents run on the whole bank to south-west.

considering la vibora not as a submerged land, but as a heaved-up part of the surface of the globe, which has not reached the level of the sea, we are struck at finding on this great submarine island, as on the neighbouring land of jamaica and cuba, the loftiest heights towards its eastern boundary. in that direction are situated portland rock, pedro keys and south key, all surrounded by dangerous breakers. the depth is six or eight fathoms; but, in advancing to the middle of the bank, along the line of the summit, first towards the west and then towards the north-west, the depth becomes successively ten, twelve, sixteen and nineteen fathoms. when we survey on the map the proximity of the high lands of san domingo, cuba and jamaica, in the neighbourhood of the windward channel, the position of the island of navaza and the bank of hormigas, between capes tiburon and morant; when we trace that chain of successive breakers, from the vibora, by baxo nuevo, serranilla, and quita sueno, as far as the mosquito sound, we cannot but recognize in this system of islands and shoals the almost-continued line of a heaved-up ridge running from north-east to south-west. this ridge, and the old dyke, which link, by the rock of sancho pardo, cape san antonio to the peninsula of yucatan, divide the great sea of the west indies into three partial basins, similar to those observed in the mediterranean.

the colour of the troubled waters on the shoal of la vibora has not a milky appearance like the waters in the jardinillos and on the bank of bahama; but it is of a dirty grey colour. the striking differences of tint on the bank of newfoundland, in the archipelago of the bahama islands and on la vibora, the variable quantities of earthy matter suspended in the more or less troubled waters of the soundings, may all be the effects of the variable absorption of the rays of light, contributing to modify to a certain point the temperature of the sea. where the shoals are 8 to 10° colder at their surface than the surrounding sea, it cannot be surprising that they should produce a local change of climate. a great mass of very cold water, as on the bank of newfoundland, in the current of the peruvian shore (between the port of callao and punta parina*), or in the african current near cape verd, have necessarily an influence on the atmosphere that covers the sea, and on the climate of the neighbouring land; but it is less easy to conceive that those slight changes of temperature (for instance, a centesimal degree on the bank of la vibora) can impart a peculiar character to the atmosphere of the shoals. may not these submarine islands act upon the formation and accumulation of the vesicular vapours in some other way than by cooling the waters of the surface?

[* i found the surface of the pacific ocean, in the month of october 1802 on the coast of truxillo, 15.8° centigrade; in the port of callao, in november, 15.5; between the parallel of callao and punta parina, in december, 19°; and progressively, when the current advanced towards the equator and receded towards the west-north-west, 20.5 and 22.3°]

quitting the bank of la vibora, we passed between the baxo nuevo and the light-house of camboy; and on the 22nd march we passed more than thirty leagues to westward of el roncador (the snorer), a name which this shoal has received from the pilots who assert, on the authority of ancient traditions, that a sound like snoring is heard from afar. if such a sound be really heard, it arises, no doubt, from a periodical issuing of air compressed by the waters in a rocky cavern. i have observed the same phenomenon on several coasts, for instance, on the promontories of teneriffe, in the limestones of the havannah,* and in the granite of lower peru between truxillo and lima. a project was formed at the canary islands for placing a machine at the issue of the compressed air and allowing the sea to act as an impelling force. while the autumnal equinox is everywhere dreaded in the sea of the west indies (except on the coast of cumana and caracas), the spring equinox produces no effect on the tranquillity of those tropical regions: a phenomenon almost the inverse of that observable in high latitudes. since we had quitted la vibora the weather had been remarkably fine; the colour of the sea was indigo-blue and sometimes violet, owing to the quantity of medusae and eggs of fish (purga de mar) which covered it. its surface was gently agitated. the thermometer kept up, in the shade, from 26 to 27°; not a cloud arose on the horizon although the wind was constantly north, or north-north-west. i know not whether to attribute to this wind, which cools the higher layers of the atmosphere, and there produces icy crystals, the halos which were formed round the moon two nights successively. the halos were of small dimensions, 45° diameter. i never had an opportunity of seeing and measuring any* of which the diameter had attained 90°. the disappearance of one of those lunar halos was followed by the formation of a great black cloud, from which fell some drops of rain; but the sky soon resumed its fixed serenity, and we saw a long series of falling-stars and bolides which moved in one direction and contrary to that of the wind of the lower strata.

[* called by the spanish sailors el cordonazo de san francisco.]

[* in captain parry’s first voyage halos were measured round the sun and moon, of which the rays were 22 1/2°; 22° 52′; 38°; 46°. north-west passage, 1821.]

on the 23rd march, a comparison of the reckoning with the chronometric longitude, indicated the force of a current bearing towards west-south-west. its swiftness, in the parallel of 17°, was twenty to twenty-two miles in twenty-four hours. i found the temperature of the sea somewhat diminished; in latitude 12° 35′ it was only 25.9° (air 27.0°). during the whole day the firmament exhibited a spectacle which was thought remarkable even by the sailors and which i had observed on a previous occasion (june 13th, 1799). there was a total absence of clouds, even of those light vapours called dry; yet the sun coloured, with a fine rosy tint, the air and the horizon of the sea. towards night the sea was covered with great bluish clouds; and when they disappeared we saw, at an immense height, fleecy clouds in regular spaces, and ranged in convergent bands. their direction was from north-north-west to south-south-east, or more exactly, north 20° west, consequently contrary to the direction of the magnetic meridian.

on the 24th march we entered the gulf which is bounded on the east by the coast of santa marta, and on the west by costa rica; for the mouth of the magdalena and that of the rio san juan de nicaragua are on the same parallel, nearly 11° latitude. the proximity of the pacific ocean, the configuration of the neighbouring lands, the smallness of the isthmus of panama, the lowering of the soil between the gulf of papagayo and the port of san juan de nicaragua, the vicinity of the snowy mountains of santa marta, and many other circumstances too numerous to mention, combine to create a peculiar climate in this gulf. the atmosphere is agitated by violent gales known in winter by the name of the brizotes de santa marta. when the wind abates, the currents bear to north-east, and the conflict between the slight breezes (from east and north-east) and the current renders the sea rough and agitated. in calm weather, the vessels going from carthagena to rio sinu, at the mouth of the atrato and at portobello, are impeded in their course by the currents of the coast. the heavy or brizote winds, on the contrary, govern the movement of the waters, which they impel in an opposite direction, towards west-south-west. it is the latter movement which major rennell, in his great hydrographic work, calls drift; and he distinguishes it from real currents, which are not owing to the local action of the wind, but to differences of level in the surface of the ocean; to the rising and accumulation of waters in very distant latitudes. the observations which i have collected on the force and direction of the winds, on the temperature and rapidity of the currents, on the influence of the seasons, or the variable declination of the sun, have thrown some light on the complicated system of those pelagic floods that furrow the surface of the ocean: but it is less easy to conceive the causes of the change in the movement of the waters at the same season and with the same wind. why is the gulf-stream sometimes borne on the coast of florida, sometimes on the border of the shoal of bahama? why do the waters flow, for the space of whole weeks, from the havannah to matanzas, and (to cite an example of the corriente por arriba, which is sometimes observed in the most eastern part of the main land during the prevalence of gentle winds) from la guayra to cape codera and cumana?

as we advanced, on the 25th of march, towards the coast of darien, the north-east wind increased with violence. we might have imagined ourselves transported to another climate. the sea became very rough during the night yet the temperature of the water kept up (from latitude 10° 30′, to 9° 47′) at 25.8°. we perceived at sunrise a part of the archipelago* of saint bernard, which closes the gulf of morrosquillo on the north. a clear spot between the clouds enabled me to take the horary angles. the chronometer, at the little island of mucara, gave longitude 78° 13′ 54′. we passed on the southern extremity of the placer de san bernardo. the waters were milky, although a sounding of twenty-five fathoms did not indicate the bottom; the cooling of the water was not felt, doubtless owing to the rapidity of the current. above the archipelago of saint bernard and cape boqueron we saw in the distance the mountains of tigua. the stormy weather and the difficulty of going up against the wind induced the captain of our frail vessel to seek shelter in the rio sinu, or rather, near the punta del zapote, situated on the eastern bank of the ensenada de cispata, into which flows the river sinu or the zenu of the early conquistadores. it rained with violence, and i availed myself of that occasion to measure the temperature of the rain-water: it was 26.3°, while the thermometer in the air kept up, in a place where the bulb was not wet, at 24.8°. this result differed much from that we had obtained at cumana, where the rain-water was often a degree colder than the air.*

[* it is composed of the islands mucara, ceycen, maravilla, tintipan, panda, palma, mangles, and salamanquilla, which rise little above the sea. several of them have the form of a bastion. there are two passages in the middle of this archipelago, from seventeen to twenty fathoms. large vessels can pass between the isla panda and tintipan, and between the isla de mangles and palma.]

[* as, within the tropics, it takes but little time to collect some inches of water in a vase having a wide opening, and narrowing towards the bottom, i do not think there can be any error in the observation, when the heat of the rain-water differs from that of the air. if the heat of the rain-water be less than that of the air it may be presumed that only a part of the total effect is observed. i often found at mexico at the end of june, the rain at 19.2 or 19.4°, when the air was at 17.8 and 18°. in general it appeared to me that, within the torrid zone, either at the level of the sea, or on table-lands from 1200 to 1500 toises high, there is no rain but that during storms, which falls in large drops very distant from each other, and is sensibly colder than the air. these drops bring with them, no doubt, the low temperature of the high regions. in the rain which i found hotter than the air, two causes may act simultaneously. great clouds heat by the absorption of the rays of the sun which strike their surface; and the drops of water in falling cause an evaporation and produce cold in the air. the temperature of rain-water, to which i devoted much attention during my travels, has become a more important problem since m. boisgiraud, professor of experimental philosophy at poitiers, has proved that in europe rain is generally sufficiently cold, relatively to the air, to cause precipitation of vapour at the surface of every drop. from this fact he traces the cause of the unequal quantity of rain collected at different heights. when we recollect that one degree only of cooling precipitates more water in the hot climate of the tropics, than by a temperature of 10 to 13°, we may cease to be surprised at the enormous size of the drops of rain that fall at cumana, carthagena and guayaquil.]

our passage from the island of cuba to the coast of south america terminated at the mouth of the rio sinu, and it occupied sixteen days. the roadstead near the punta del zapote afforded very bad anchorage; and in a rough sea, and with a violent wind, we found some difficulty in reaching the coast in our canoe. everything denoted that we had entered a wild region rarely visited by strangers. a few scattered houses form the village of zapote: we found a great number of mariners assembled under a sort of shed, all men of colour, who had descended the rio sinu in their barks, to carry maize, bananas, poultry and other provisions to the port of carthagena. these barks, which are from fifty to eighty feet long, belong for the most part to the planters (haciendados) of lorica. the value of their largest freight amounts to about 2000 piastres. these boats are flat-bottomed, and cannot keep at sea when it is very rough. the breezes from the north-east had, during ten days, blown with violence on the coast, while, in the open sea, as far as 10° latitude, we had only had slight gales, and a constantly calm sea. in the aerial, as in the pelagic currents, some layers of fluids move with extreme swiftness, while others near them remain almost motionless. the zambos of the rio sinu wearied us with idle questions respecting the purpose of our voyage, our books, and the use of our instruments: they regarded us with mistrust; and to escape from their importunate curiosity we went to herborize in the forest, although it rained. they had endeavoured, as usual, to alarm us by stories of boas (traga-venado), vipers and the attacks of jaguars; but during a long residence among the chayma indians of the orinoco we were habituated to these exaggerations, which arise less from the credulity of the natives, than from the pleasure they take in tormenting the whites. quitting the coast of zapote, covered with mangroves,* we entered a forest remarkable for a great variety of palm-trees. we saw the trunks of the corozo del sinu* pressed against each other, which formed heretofore our species alfonsia, yielding oil in abundance; the cocos butyracea, called here palma dolce or palma real, and very different from the palma real of the island of cuba; the palma amarga, with fan-leaves that serve to cover the roofs of houses, and the latta,* resembling the small piritu palm-tree of the orinoco. this variety of palm-trees was remarked by the first conquistadores.* the alfonsia, or rather the species of elais, which we had nowhere else seen, is only six feet high, with a very large trunk; and the fecundity of its spathes is such that they contain more than 200,000 flowers. although a great number of those flowers (one tree bearing 600,000 at the same time) never come to maturity,* the soil remains covered with a thick layer of fruits. we often made a similar observation under the shade of the mauritia palm-tree, the cocos butyracea, the seje and the pihiguao of the atabapo. no other family of arborescent plants is so prolific in the development of the organs of flowering. the almond of the corozo del sinu is peeled in the water. the thick layer of oil that swims in the water is purified by boiling, and yields the butter of corozo (manteca de corozo) which is thicker than the oil of the cocoa-tree, and serves to light churches and houses. the palm-trees of the section of cocoinies of mr. brown are the olive-trees of the tropical regions. as we advanced in the forest, we began to find little pathways, looking as though they had been recently cleared out by the hatchet. their windings displayed a great number of new plants: mougeotia mollis, nelsonia albicans, melampodium paludosum, jonidium anomalum, teucrium palustre, gomphia lucens, and a new kind of composees, the spiracantha cornifolia. a fine pancratium embalmed the air in the humid spots, and almost made us forget that those gloomy and marshy forests are highly dangerous to health.

[* rhizophora mangle.]

[* in spanish america palm-trees with leaves the most different in kind and species are called corozo: the corozo del sinu, with a short, thick, glossy trunk, is the elaeis melanococca of martius, palm. page 64 tab. 33, 55. i cannot believe it to be identical with the elaeis guineensis (herbal of congo river page 37) since it vegetates spontaneously in the forests of the rio sinu. the corozo of caripe is slender, small and covered with thorns; it approaches the cocos aculeata of jacquin. the corozo de los marinos of the valley of cauca, one of the tallest palm-trees, is the cocus butyracea of linnaeus.]

[* perhaps of the species of aiphanes.]

[* pedro de cieca de leon, a native of seville, who travelled in 1531, at the age of thirteen years, in the countries i have described, observes that las tierras comarcanas del rio cenu y del golfo de uraba estan llena de unos palmares muy grandes y espessos, que son unos arboles gruessos, y llevan unas ramas como palma de datiles. [the lands adjacent to the rio cenu and the gulf of uraba are full of very tall, spreading palm-trees. they are of vast size and are branched like the date-palm.] see la cronica del peru nuevamenta escrita, antwerp 1554 pages 21 and 204.]

[* i have carefully counted how many flowers are contained in a square inch on each amentum, from 100 to 120 of which are found united in one spathe.]

after an hour’s walk we found, in a cleared spot, several inhabitants employed in collecting palm-tree wine. the dark tint of the zambos formed a strong contrast with the appearance of a little man with light hair and a pale complexion who seemed to take no share in the labour. i thought at first that he was a sailor who had escaped from some north american vessel; but i was soon undeceived. this fair-complexioned man was my countryman, born on the coast of the baltic; he had served in the danish navy and had lived for several years in the upper part of the rio sinu, near santa cruz de lorica. he had come, to use the words of the loungers of the country para ver tierras, y pasear, no mas (to see other lands, and to roam about, nothing else.) the sight of a man who could speak to him of his country seemed to have no attraction for him; and, as he had almost forgotten german without being able to express himself clearly in spanish, our conversation was not very animated. during the five years of my travels in spanish america i found only two opportunities of speaking my native language. the first prussian i met with was a sailor from memel who served on board a ship from halifax, and who refused to make himself known till after he had fired some musket-shot at our boat. the second, the man we met at the rio sinu, was very amicably disposed. without answering my questions he continued repeating, with a smile, that the country was hot and humid; that the houses in the town of pomerania were finer than those of santa cruz de lorica; and that, if we remained in the forest, we should have the tertian fever (calentura) from which he had long suffered. we had some difficulty in testifying our gratitude to this good man for his kind advice; for according to his somewhat aristocratic principles, a white man, were he bare-footed, should never accept money “in the presence of those vile coloured people!” (gente parda). less disdainful than our european countryman, we saluted politely the group of men of colour who were employed in drawing off into large calabashes, or fruits of the crescentia cujete, the palm-tree wine from the trunks of felled trees. we asked them to explain to us this operation, which we had already seen practised in the missions of the cataracts. the vine of the country is the palma dolce, the cocos butyracea, which, near malgar, in the valley of the magdalena, is called the wine palm-tree, and here, on account of its majestic height, the royal palm-tree. after having thrown down the trunk, which diminishes but little towards the top, they make just below the point whence the leaves (fronds) and spathes issue, an excavation in the ligneous part, eighteen inches long, eight broad, and six in depth. they work in the hollow of the tree, as though they were making a canoe; and three days afterwards this cavity is found filled with a yellowish-white juice, very limpid, with a sweet and vinous flavour. the fermentation appears to commence as soon as the trunk falls, but the vessels preserve their vitality; for we saw that the sap flowed even when the summit of the palm-tree (that part whence the leaves sprout out) is a foot higher than the lower end, near the roots. the sap continues to mount as in the arborescent euphorbia recently cut. during eighteen to twenty days, the palm-tree wine is daily collected; the last is less sweet but more alcoholic and more highly esteemed. one tree yields as much as eighteen bottles of sap, each bottle containing forty-two cubic inches. the natives affirm that the flowing is more abundant when the petioles of the leaves, which remain fixed to the trunk, are burnt.

the great humidity and thickness of the forest forced us to retrace our steps and to gain the shore before sunset. in several places the compact limestone rock, probably of tertiary formation, is visible. a thick layer of clay and mould rendered observation difficult; but a shelf of carburetted and shining slate seemed to me to indicate the presence of more ancient formations. it has been affirmed that coal is to be found on the banks of the sinu. we met with zambos carrying on their shoulders the cylinders of palmetto, improperly called the cabbage palm, three feet long and five to six feet thick. the stem of the palm-tree has been for ages an esteemed article of food in those countries. i believe it to be wholesome although historians relate that, when alonso lopez de ayala was governor of uraba, several spaniards died after having eaten immoderately of the palmetto, and at the same time drinking a great quantity of water. in comparing the herbaceous and nourishing fibres of the young undeveloped leaves of the palm-trees with the sago of the mauritia, of which the indians make bread similar to that of the root of the jatropha manihot, we involuntarily recollect the striking analogy which modern chemistry has proved to exist between ligneous matter and the amylaceous fecula. we stopped on the shore to collect lichens, opegraphas and a great number of mosses (boletus, hydnum, helvela, thelephora) that were attached to the mangroves, and there, to my great surprise, vegetating, although moistened by the sea-water.

before i quit this coast, so seldom visited by travellers and described by no modern voyager, i may here offer some information which i acquired during my stay at carthagena. the rio sinu in its upper course approaches the tributary streams of the atrato which, to the auriferous and platiniferous province of choco, is of the same importance as the magdalena to cundinamarca, or the rio cauca to the provinces of antioquia and popayan. the three great rivers here mentioned have heretofore been the only commercial routes, i might almost add, the only channels of communication for the inhabitants. the rio atrato receives, at twelve leagues distance from its mouth, the rio sucio on the east; the indian village of san antonio is situated on its banks. proceeding upward beyond the rio pabarando, you arrive in the valley of sinu. after several fruitless attempts on the part of the archbishop gongora to establish colonies in darien del norte and on the eastern coast of the gulf of uraba, the viceroy espeleta recommended the spanish government to fix its whole attention on the rio sinu; to destroy the colony of cayman; to fix the planters in the spanish village of san bernardo del viento in the jurisdiction of lorica; and from that post, which is the most westerly, to push forward the peaceful conquests of agriculture and civilization towards the banks of the pabarando, the rio sucio and the atrato.* the number of independent indians who inhabit the lands between uraba, rio atrato, rio sucio and rio sinu was, according to a census made in 1760, at least 1800. they were distributed in three small villages, suraba, toanequi and jaraguia. this population was computed, at the period when i travelled there, to be 3000. the natives, comprehended in the general name of caymans, live at peace with the inhabitants of san bernardo del viento (pueblo de espanoles), situated on the western bank of the rio sinu, lower than san nicolas de zispata, and near the mouth of the river. these people have not the ferocity of the darien and cunas indians, on the left bank of the atrato; who often attack the boats trading with the town of quidbo in the choco; they also make incursions on the territory of uraba, in the months of june and november, to collect the fruit of the cacao-trees. the cacao of uraba is of excellent quality; and the darien indians sometimes come to sell it, with other productions, to the inhabitants of rio sinu, entering the valley of that river by one of its tributary streams, the jaraguai.

[* i will here state some facts which i obtained from official documents during my stay at carthagena, and which have not yet been published. in the sixteenth and seventeenth centuries the name of darien was given vaguely to the whole coast extending from the rio damaquiel to the punta de san blas, on 2 1/4° of longitude. the cruelties exercised by pedrarias davila rendered almost inaccessible to the spaniards a country which was one of the first they had colonized. the indians (dariens and cunas–cunas) remained masters of the coast, as they still are at poyais, in the land of the mosquitos. some scotchmen formed in 1698 the settlements of new caledonia, new edinburgh and scotch port, in the most eastern part of the isthmus, a little west of punta carreto. they were soon driven away by the spaniards but, as the latter occupied no part of the coast, the indians continued their attacks against choco’s boats, which from time to time descended the rio atrato, the sanguinary expedition of don manuel de aldarete in 1729 served only to augment the resentment of the natives. a settlement for the cultivation of the cocoa-tree, attempted in the territory of urabia in 1740 by some french planters under the protection of the spanish government, had no durable success; and the court, excited by the reports of the archbishop-viceroy, gongora, ordered, by the cedule of the 15th august, 1783, either the conversion and conquest, or the destruction (reduccion o extincion) of the indians of darien. this order, worthy of another age, was executed by don antonio de arebalo: he experienced little resistance and formed, in 1785, the four settlements and forts of cayman on the eastern coast of the gulf of urabia, concepcion, carolina and mandinga. the lele, or high-priest of mandinga, took an oath of fidelity to the king of spain; but in 1786 the war with the darien indians recommenced and was terminated by a treaty concluded july 27th, 1787, between the archbishop-viceroy and the cacique bernardo. the forts and new colonies, which figured only on the maps sent to madrid, augmented the debt of the treasury of santa fe de bogota, in 1789, to the sum of 1,200,000 piastres. the viceroy, gil lemos, wiser than his predecessor, obtained permission from the court to abandon carolina, concepcion and mandinga. the settlement of cayman only was preserved, on account of the navigation of the atrato, and it was declared free, under the government of the archbishop-viceroy: it was proposed to transfer this settlement to a more healthy spot, that of uraba; but lieutenant-general don antonio arebalo, having proved that the expense of this removal would amount to the sum of 40,000 piastres, the fort of cayman was also destroyed, by order of the viceroy espeleta in 1791, and the planters were compelled to join those of the village of san bernardo.]

it cannot be doubted that the gulf of darien was considered, at the beginning of the sixteenth century, as a nook in the country of the caribs. the word caribana is still preserved in the name of the eastern cape of that gulf. we know nothing of the languages of the darien, cunas and cayman indians: and we know not whether carib or arowak words are found in their idioms; but it is certain, notwithstanding the testimony of anghiera on the identity of the race of the caribs of the lesser antilles and the indians of uraba, that pedro de cieca, who lived so long among the latter, never calls them caribs nor cannibals. he describes the race of that tribe as being naked with long hair, and going to the neighbouring countries to trade; and says the women are cleanly, well dressed and extremely engaging (amorosas y galanas). “i have not seen,” adds the conquistador, “any women more beautiful* in all the indian lands i have visited: they have one fault, however, that of having too frequent intercourse with the devil.”

[* cronica del peru pages 21 and 22. the indians of darien, uraba, zenu (sinu), tatabe, the valleys of nore and of guaca, the mountains of abibe and antioquia, are accused, by the same author, of the most ferocious cannibalism; and perhaps that circumstance alone gives rise to the idea that they were of the same race as the caribs of the west indies. in the celebrated provision real of the 30th of october, 1503, by which the spaniards are permitted to make slaves of the anthropophagic indians of the archipelago of san bernardo, opposite the mouth of the rio sinu, the isla fuerte, isla bura (baru) and carthagena, there is more of a question of morals than of race, and the denomination of caribs is altogether avoided. cieca asserts that the natives of the valley of nore seized the women of neighbouring tribes, in order first to devour the children who were born of the union with foreign wives, and then the women themselves. foreseeing that this horrible depravity would not be believed, although it had been observed by columbus in the west indies, he cites the testimony of juan de vadillo, who had observed the same facts and who was still living in 1554 when the cronica del peru appeared in dutch. with respect to the etymology of the word cannibal, it seems to me entirely cleared up by the discovery of the journal kept by columbus during his first voyage of discovery, and of which bartholomew de las casas has left us an abridged copy. dice mas el almirante que en las islas passadas estaban con gran temor de carib: y en algunas los llamaban caniba; pero en la espanola carib y son gente arriscada, pues andan por todas estas islas y comen la gente que pueden haber. [and the admiral moreover says that in the islands they passed, great apprehension was entertained on account of the caribs. some call them canibas; but in spanish they are called caribs. they are a very bold people, and they travel about these islands, and devour all the persons whom they capture.] navarete tome 1 page 135. in this primitive form of words it is easy to perceive that the permutation of the letters r and n, resulting from the imperfection of the organs in some nations, might change carib into canib, or caniba. geraldini who, according to the tendency of that age, sought, like cardinal bembo, to latinize all barbarous denominations, recognizes in the cannibals the manners of dogs (canes) just as st. louis desired to send the tartars ad suas tartareas sedes unde exierint.]

the rio sinu, owing to its position and its fertility, is of the highest importance for provisioning carthagena. in time of war the enemy usually stationed their ships between the morro de tigua and the boca de matunilla, to intercept barques laden with provisions. in that station they were, however, sometimes exposed to the attack of the gun-boats of carthagena: these gun-boats can pass through the channel of pasacaballos which, near saint anne, separates the isle of baru from the continent. lorica has, since the sixteenth century, been the principal town of rio sinu; but its population which, in 1778, under the government of don juan diaz pimienta, amounted to 4000 souls, has considerably diminished, because nothing has been done to secure the town from inundations and the deleterious miasmata they produce.

the gold-washings of the rio sinu, heretofore so important above all, between its source and the village of san geronimo, have almost entirely ceased, as well as those of cienega de tolu, uraba and all the rivers descending from the mountains of abibe. “the darien and the zenu,” says the bachelor enciso in his geographical work published at the beginning of the sixteenth century, “is a country so rich in gold pepites that, in the running waters, that metal can be fished with nets.” excited by these narratives, the governor pedrarias sent his lieutenant, francisco becerra, in 1515, to the rio sinu. this expedition was most unfortunate for becerra and his troop were massacred by the natives, of whom the spaniards, according to the custom of the time, had carried away great numbers to be sold as slaves in the west indies. the province of antioquia now furnishes, in its auriferous veins, a vast field for mining speculations; but it might be well worth while to relinquish gold-washings for the cultivation of colonial productions in the fertile lands of sinu, the rio damaquiel, the uraba and the darien del norte; above all, that of cacao, which is of a superior quality. the proximity of the port of carthagena would also render the neglected cultivation of cinchona an object of great importance to european trade. that precious tree vegetates at the source of the rio sinu, as in the mountains of abibe and maria. the real febrifuge cinchona, with a hairy corolla, is nowhere else found so near the coast, if we except the sierra nevada of santa marta.

the rio sinu and the gulf of darien were not visited by columbus. the most eastern point at which that great man touched land, on the 26th november, 1503, is the puerto do retreto, now called punta de escribanos, near the punta of san blas, in the isthmus of panama. two years previously, rodrigo de bastidas and alanso do ojeda, accompanied by amerigo vespucci, had discovered the whole coast of the main land, from the gulf of maracaybo as far as the puerto de retreto. having often had occasion in the preceding volumes to speak of new andalusia, i may here mention that i found that denomination, for the first time, in the convention made by alonso de ojeda with the conquistador diego de sicuessa, a powerful man, say the historians of his time, because he was a flattering courtier and a wit. in 1508 all the country from the cabo de la vela to the gulf of uraba, where the castillo del oro begins, was called new andalusia, a name since restricted to the province of cumana.

a fortunate chance led me to see, during the course of my travels, the two extremities of the main land, the mountainous and verdant coast of paria, which columbus supposes to have been the cradle of the human race, and the low and humid coast extending from the mouth of the sinu towards the gulf of darien. the comparison of these scenes, which have again relapsed into a savage state, confirms what i have elsewhere advanced relative to the strange and sometimes retrograde nature of civilization in america. on one side, the coast of paria, the islands of cubagua and marguerita; on the other, the gulf of uraba and darien, received the first spanish colonists. gold and pearls, which were there found in abundance, because from time immemorial they had been accumulated in the hands of the natives, gave those countries a popular celebrity from the beginning of the sixteenth century. at seville, toledo, pisa, genoa and antwerp those countries were viewed like the realms of ormuz and of ind. the pontiffs of rome mentioned them in their bulls; and bembo has celebrated them in those historical pages which add lustre to the glory of venice.

at the close of the fifteenth, and the beginning of the sixteenth century, europe saw, in those parts of the new world discovered by columbus, ojeda, vespucci and rodrigo de bastidas, only the advanced capes of the vast territories of india and eastern asia. the immense wealth of those territories in gold, diamonds, pearls and spices had been vaunted in the narratives of benjamin de tudela, rubruquis, marco polo and mandeville. columbus, whose imagination was excited by these narrations, caused a deposition to be made before a notary, on the 12th of june, 1494, in which sixty of his companions, pilots, sailors and passengers certified upon oath that the southern coast of cuba was a part of the continent of india. the description of the treasures of cathay and cipango, of the celestial town of quinsay and the province of mango, which had fired the admiral’s ambition in early life, pursued him like phantoms in his declining days. in his fourth and last voyage, on approaching the coast of cariay (poyais or mosquito coast), veragua and the isthmus, he believed himself to be near the mouth of the ganges.* these geographical illusions, this mysterious veil, which enveloped the first discoveries, contributed to magnify every object, and to fix the attention of europe on regions, the very names of which are, to us, scarcely known. new cadiz, the principal seat of the pearl-fishery, was on an island which has again become uninhabited. the extremity of the rocky coast of paria is also a desert. several towns were founded at the mouth of the rio atrato, by the names of antigua del darien, uraba or san sebastian de buenavista. in these spots, so celebrated at the beginning of the sixteenth century, the historians of the conquest tell us that the flower of the castilian heroes were found assembled: thence balboa set out to discover the south sea; pizarro marched from thence to conquer and ravage peru; and pedro de cieca constantly followed the chain of the andes, by autioquia, popayan and cuzco, as far as la plata, after having gone 900 leagues by land. these towns of darien are destroyed; some ruins scattered on the hills of uraba, the fruit-trees of europe mixed with native trees, are all that mark to the traveller the spots on which those towns once stood. in almost all spanish america the first lands peopled by the conquistadores, have retrograted into barbarism.* other countries, discovered later, attract the attention of the colonists: such is the natural progress of things in peopling a vast continent. it may be hoped that on several points the people will return to the places that were first chosen. it is difficult to conceive why the mouth of a great river, descending from a country rich in gold and platina, should have remained uninhabited. the atrato, heretofore called rio del darien, de san juan or dabayba, has had the same fate as the orinoco. the indians who wander around the delta of those rivers continue in a savage state.

[* tambien dicen que la mar baxa a ciguare, y de alli a diez jornadas es el rio de guangues: para que estas tierras estan con veragua como tortosa con fuenterabia o pisa con venecia.” [also it is said that the sea lowers at ciguara, and from thence it is a ten days’ journey to the river ganges; for these lands are, with reference to veragua, like tortosa with respect to fuenterabia, or pisa, with respect to venice.] these words are taken from the lettera rarissima of columbus, of which the original spanish was lately found, and published by the learned m. navarrete, in his coleccion de viages volume 1 page 299.]

[* in carefully collating the testimonies of the historians of the conquest, some contradictions are observed in the periods assigned to the foundation of the towns of darien. pedro de cieca, who had been on the spot, affirms that, under the government of alonzo de ojeda and nicuessa, the town of nuestra senora santa maria el antigua del darien was founded on the western coast of the gulf or culata de uraba, in 1509; and that later (despues desto passado) ojeda passed to the eastern coast of the culata to construct the town of san sebastian de uraba. the former, called by abbreviation ciudad del antigua, had soon a population of 2000 spaniards; while the latter, the ciudad del uraba, remained uninhabited, because francisco pizarro, since known as the conqueror of peru, was forced to abandon it, having vainly demanded succour from st. domingo. the historian herrera, after having said that the foundation of antigua had preceded by one year that of uraba or san sebastian, affirms the contrary in the following chapter and in the chronicle itself. it was, according to the chronicle, in 1501 that ojeda, accompanied by vespucci, and penetrating for the first time the gulf of uraba or darien, resolved to construct, with wood and unbaked bricks, a fort at the entrance of culata. it appears, however, that this enterprise was not executed; for, in 1508, in the convention made by ojeda and nicuessa, they each promised to build two fortresses on the limits of new andalusia and of castillo del oro. herrera, in the 7th and 8th books of the first decade, fixes the foundation of san sebastian de uraba at the beginning of 1510, and mentions it as the most ancient town of the continent of america, after that of ceragua, founded by columbus in 1503, on the rio belen. he relates how francisco pizarro abandoned that town, and how the foundation of the ciudad del antigua by entiso, towards the end of the year 1510, was the consequence of that event. leo x made antigua a bishopric in 1514; and this was the first episcopal church of the continent. in 1519 pedrarius davila persuaded the court of madrid, by false reports, that the site of the new town of panama was more healthful than that of antigua, the inhabitants were compelled to abandon the latter town, and the bishopric was transferred to panama. the gulf of uraba was deserted during thirteen years, till the founder of the town of carthagena, pedro de heredia, after having dug up the graves, or huacas, of the rio sinu, to collect gold, sent his brother alonzo, in 1532, to repeople uraba, and reconstruct on that spot a town under the name of san sebastian de buenavista.]

we weighed anchor in the road of zapote, on the 27th march, at sunrise. the sea was less stormy, and the weather rather warmer, although the fury of the wind was undiminished. we saw on the north a succession of small cones of extraordinary form, as far as the morro de tigua; they are known by the name of the paps (tetas) of santero, tolu, rincon and chichimar. the two latter are nearest the coast. the tetas de tolu rise in the middle of the savannahs. there, from the trunks of the toluifera balsamum, is collected the precious balsam of tolu, heretofore so celebrated in the pharmacopoeias of europe, and in which is a profitable article of trade at corozal, caimito and the town of tocasuan. in the savannahs (altas del tolu) oxen and mules wander half wild. several of those hills between cienega de pesquero and the punta del comissario are linked two-and-two together, like basaltic columns; it is, however, very probable that they are calcareous, like the tetas de managua, south of the havannah. in the archipelago of san bernardo we passed between the island of salamanquilla and cape boqueron. we had scarcely quitted the gulf of morosquillo when the sea became so rough that the waves frequently washed over the deck of our little vessel. it was a fine moonlight night. our captain sought in vain a sheltering-place on the coast to the north of the village of rincon. we cast anchor at four fathoms but, having discovered that we were lying over a reef of coral, we preferred the open sea.

the coast has a singular configuration beyond the morro de tigua, the terminatory point of the group of little mountains which rise like islands from the plain. we found at first a marshy soil extending over a square of eight leagues between the bocas de matuna and matunilla. these marshes are connected by the cienega de la cruz, with the dique of mahates and the rio magdalena. the island of baru which, with the island of tierra bomba, forms the vast port of carthagena, is, properly speaking, but a peninsula fourteen miles long, separated from the continent by the narrow channel of pasacaballos. the archipelago of san bernardo is situated opposite cape boqueron. another archipelago, called rosario, lies off the southern point of the peninsula of baru. these rents in the coast are repeated at the 10 3/4 and 11° of latitude. the peninsulas near the ensenada of galera de zamba and near the port of savanilla have the same aspect as the peninsula baru. similar causes have produced similar effects; and the geologist must not neglect those analogies, in the configuration of a coast which, from punta caribana in the mouth of the atrato, beyond the cape of la vela, along an extent of 120 leagues, has a general direction from south-west to north-east.

the wind having dropped during the night we could only advance to the island of arenas where we anchored. i found it was 78° 2′ 10′ of longitude. the weather became stormy during the night. we again set sail on the morning of the 29th of march, hoping to be able to reach boca chica that day. the gale blew with extreme violence, and we were unable to proceed with our frail bark against the wind and the current, when, by a false manoeuvre in setting the sails (we had but four sailors), we were during some minutes in imminent danger. the captain, who was not a very bold mariner, declined to proceed further up the coast and we took refuge, sheltered from the wind, in a nook of the island of baru south of punta gigantes. it was palm sunday and the zambo, who had accompanied us to the orinoco and did not leave us till we returned to france, reminded us that on the same sunday in the preceding year, we had nearly been lost on the north of the mission of uruana.

there was to be an eclipse of the moon during the night, and the next day an occultation of alpha virginis. the observation of the latter phenomenon might have been very important in determining the longitude of carthagena. in vain i urged the captain to allow one of his sailors to accompany me by land to the foot of boca chica, a distance of five miles. he objected on account of the wild state of the country in which there is neither habitation nor path. a little incident which might have rendered palm–sunday more fatal justified the prudence of the captain. we went by moonlight to collect plants on the shore; as we approached the land, we saw a young negro issue from the thicket. he was quite naked, loaded with chains, and armed with a machete. he invited us to land on a part of the beach covered with large mangroves, as being a spot where the surf did not break, and offered to conduct us to the interior of the island of baru if we would promise to give him some clothes. his cunning and wild appearance, the often-repeated question whether we were spaniards, and certain unintelligible words which he addressed to some of his companions who were concealed amidst the trees, inspired us with some mistrust. these blacks were no doubt maroon negroes: slaves escaped from prison. this unfortunate class are much to be feared: they have the courage of despair, and a desire of vengeance excited by the severity of the whites. we were without arms; the negroes appeared to be more numerous than we were and, thinking that possibly they invited us to land with the desire of taking possession of our canoe, we thought it most prudent to return on board. the aspect of a naked man wandering on an uninhabited beach, unable to free himself from the chains fastened round his neck and the upper part of his arm, was an object calculated to excite the most painful impressions. our sailors wished to return to the shore for the purpose of seizing the fugitives, to sell them secretly at carthagena. in countries where slavery exists the mind is familiarized with suffering and that instinct of pity which characterizes and enobles our nature is blunted.

whilst we lay at anchor near the island of baru in the meridian of punta gigantes i observed the eclipse of the moon of the 29th of march, 1801. the total immersion took place at 11 hours 30 minutes 12.6 seconds mean time. some groups of vapours, scattered over the azure vault of the sky, rendered the observation of the immersion uncertain.

during the total eclipse the lunar disc displayed, as almost always happens, a reddish tint, without disappearing; the edges, examined with a sextant, were strongly undulating, notwithstanding the considerable altitude of the orb. it appeared to me that the moon was more luminous than i had ever seen it in the temperate zone. the vividness of the light, it may be conceived, does not depend solely on the state of the atmosphere, which reflects, more or less feebly, the solar rays, by inflecting them in the cone of the shade. the light is also modified by the variable transparency of that part of the atmosphere across which we perceived the moon eclipsed. within the tropics great serenity of the sky and a perfect dissolution of the vapours diminish the extinction of the light sent back to us by the lunar disc. i was singularly struck during the eclipse by the want of uniformity in the distribution of the refracted light by the terrestrial atmosphere. in the central region of the disc there was a shadow like a round cloud, the movement of which was from east to west. the part where the immersion was to take place was consequently a few minutes prior to the immersion much more brightly illumined than the western edges. is this phenomenon to be attributed to an inequality of our atmosphere; to a partial accumulation of vapour which, by absorbing a considerable part of the solar light, inflects less on one side the cone of the shadow of the earth? if a similar cause, in the perigee of central eclipses, sometimes renders the disc invisible, may it not happen also that only a small portion of the moon is seen; a disc, irregularly formed, and of which different parts were successively enlightened?

on the morning of the 30th of march we doubled punta gigantes, and made for the boca chica, the present entrance of the port of carthagena. from thence the distance is seven or eight miles to the anchorage near the town; and although we took a practico to pilot us, we repeatedly touched on the sandbanks. on landing i learned, with great satisfaction, that the expedition appointed to take the survey of the coast under the direction of m. fidalgo, had not yet put to sea. this circumstance not only enabled me to ascertain the astronomical position of several towns on the shore which had served me as points of departure in fixing chronometrically the longitude of the llanos and the orinoco, but also served to guide me with respect to the future direction of my journey to peru. the passage from carthagena to porto bello and that of the isthmus by the rio chagres and cruces, are alike short and easy; but it was to be feared that we might stay long at panama before we found an opportunity of proceeding to guayaquil, and in that case the voyage on the pacific would be extremely lingering, as we should have to sail against contrary winds and currents. i relinquished with regret the hope of levelling by the barometer the mountains of the isthmus, though it would then have been difficult to foresee that at the present time (1827), while measurements have been effected on so many other points of mexico and columbia, we should remain in ignorance of the height of the ridge which divides the waters in the isthmus. the persons we consulted all agreed that the journey by land along the cordilleras by santa fe de bogota, popayan, quito and caxamarca would be preferable to the sea-voyage, and would furnish an immense field for exploration. the predilection of europeans for the tierras frias, that is to say, the cold and temperate climate that prevails on the back of the andes, gave further weight to these counsels. the distances were known, but we were deceived with respect to the time it would take to traverse them on mules’ backs. we did not imagine that it would require more than eighteen months to go from carthagena to lima. notwithstanding this delay, or rather owing to the slowness with which we passed through cundinamarca, the provinces of popayan and quito, i did not regret having sacrificed the passage of the isthmus to the route of bogota, for every step of the journey was full of interest both geographically and botanically. this change of direction gave me occasion to trace the map of the rio magdalena, to determine astronomically the position of eighty points situated in the inland country between carthagena, popayan, and the upper course of the river amazon and lima, to discover the error in the longitude of quito, to collect several thousand new plants, and to observe on a vast scale the relations between the rocks of syenitic porphyry and trachyte with the fire of volcanoes.

the result of those labours of which it is not for me to appreciate the importance have long since been published. my map of the rio magdalena, multiplied by the copies of the year 1802 in america and spain, and comprehending the country between almaguer and santa marta, from 1 degree 54 minutes to 11° 15′ latitude, appeared in 1816. till that period no traveller had undertaken to describe new grenada; and the public, except in spain, knew the navigation of the magdalena only by some lines traced by bouguer. that learned traveller had descended the river from honda; but, being in want of astronomical instruments, he had ascertained but four or five latitudes, by means of small dials hastily constructed. the narratives of travels in america are now singularly multiplied. political events have led numbers of persons to those countries: and travellers have perhaps too hastily published their journals on returning to europe. they have described the towns where they resided, and landscape scenery remarkable for beauty; they have furnished information respecting the inhabitants and the different modes of travelling in barks, on mules or on men’s backs. these works, several of which are agreeable and instructive, have familiarized the nations of the old world with those of spanish america, from buenos ayres and chili as far as zacatecas and new mexico. but unfortunately, in many instances, the want of a thorough knowledge of the spanish language and the little care taken to acquire the names of places, rivers and tribes, have occasioned extraordinary mistakes.

during the six days of our stay at carthagena our most interesting excursions were to the boca grande and the hill of popa; the latter commands the town and a very extensive view. the port, or rather the bahia, is nearly nine miles and a half long, if we compute the length from the town (near the suburb of jehemani or xezemani) to the cienega of cacao. the cienega is one of the nooks of the isle of baru, south-west of the estero de pasacaballos, by which we reach the opening of the dique de mahates. two extremities of the small island of tierra bomba form, on the north, with a neck of land of the continent, and on the south, with a cape of the island of baru, the only entrances to the bay of carthagena; the former is called boca grande, the second boca chica. this extraordinary conformation of the land has given birth, for the space of a century, to theories entirely contradictory respecting the defence of a place which, next to the havannah and porto cabello, is the most important of the main land and the west indies. engineers differed respecting the choice of the opening which should be closed; and it was not, as some writers have stated, after the landing of admiral vernon, in 1741, that the idea was first conceived* of filling up the boca grande. the english forced the small entrance when they made themselves masters of the bay; but being unable to take the town of carthagena, which made a gallant resistance, they destroyed the castillo grande (called also santa cruz) and the two forts of san luis and san jose which defended the boca chica.

[* don jorge juan in his secret notices addressed to the marques de la ensenada says: la entrada antigua era por un angosto canal que llaman boca chica; de resultas de esta invasion se acordo deja cioga y impassable la boca grande, y volver a abrir la antigua fortificandola. [the old entrance was by a narrow channel called the boca chica; but after this invasion it was determined to close up the boca grande and to open the old passage, fortifying it.] secr. not. volume 1 page 4.]

the apprehension excited by the proximity of the boca grande to the town determined the court of madrid, after the english expedition, to shut up the entrance along a distance of 2640 varas. from two and a half to three fathoms of water were found; and a wall, or rather a dyke, in stone, from fifteen to twenty feet high, was raised on piles. the slope on the side of the water is unequal, and seldom 45°. this immense work was completed under the viceroy espeleta in 1795. but art could not vanquish nature; the sea is unceasingly though gradually silting up the boca chica, while it labours unceasingly to open and enlarge the boca grande. the currents which, during a great part of the year, especially when the bendavales blow with violence, ascend from south-west to north-east, throw sand into the boca chica, and even into the bay itself. the passage, which is from seventeen to eighteen fathoms deep, becomes more and more narrow,* and if a regular cleansing be not established by dredging machines, vessels will not be able to enter without risk. it is this small entrance which should have been closed; its opening is only 250 toises, and the passage or navigable channel is 110 toises. if it should one day be determined to abandon the boca chica, and re-establish the boca grande in the state which nature seems to prescribe, new fortifications must be constructed on the south-south-west of the town. this fortress has always required great pecuniary outlays to keep it up.

[* at the foot of the two forts san jose and san fernando, constructed for the defence of the boca chica, it may be seen how much the land has gained upon the sea. necks of land are formed on both sides, and also before the castillo del angel which, northward, commands the fort of san fernando.]

the insalubrity of carthagena varies with the state of the great marshes that surround the town on the east and north. the cienega de tesca is more than fifteen miles long; it communicates with the ocean where it approaches the village of guayeper. when, in years of drought, the heaped-up earth prevents the salt water from covering the whole plain, the emanations that rise during the heat of the day when the thermometer stands between 28 and 32° are very pernicious to the health of the inhabitants. a small portion of hilly land separates the town of carthagena and the islet of manga from the cienega de tesca. those hills, some of which are more than 500 feet high, command the town. the castillo de san lazaro is seen from afar rising like a great rocky pyramid; when examined nearer its fortifications are not very formidable. layers of clay and sand, belonging to the tertiary formation of nagelfluhe, are covered with bricks and furnish a kind of construction which has little stability. the cerro de santa maria de la popa, crowned by a convent and some batteries, rises above the fort of san lazaro and is worthy of more solid and extensive works. the image of the virgin, preserved in the church of the convent, has been long revered by mariners. the hill itself forms a prolonged ridge from west to east. the calcareous rock, with cardites, meandrites and petrified corals, somewhat resembles the tertiary limestone of the peninsula of araya near cumana. it is split and decomposed in the steep parts of the rock, and the preservation of the convent on so unsolid a foundation is considered by the people as one of the miracles of the patron of the place. near the cerro de la popa there appears, on several points, breccia with a limestone cement containing angular fragments of lydian stone. whether this formation of nagelfluhe is superposed on tertiary limestone of coral, and whether the fragments of the lydian stone come from secondary limestone analogous to that of zacatecas and the moro de nueva barcelona, are questions which i have not had leisure to investigate. the view from the popa is extensive and varied, and the windings and rents of the coast give it a peculiar character. i was assured that sometimes from the windows of the convent and even in the open sea, before the fort of boca chica, the snowy tops of the sierra nevada de santa marta are discernible. the distance of the horqueta to the popa is seventy-eight nautical miles. this group of colossal mountains is most frequently wrapped in thick clouds: and it is most veiled at the season when the gales blow with violence. although only forty-five miles distant from the coast, it is of little service as a signal to mariners who seek the port of saint marta. hidalgo during the whole time of his operations near the shore could take only one observation of the nevados.

a gloomy vegetation of cactus, jatropha gossypifolia, croton and mimosa covers the barren declivity of cerro de la popa. in herbalizing in those wild spots, our guides showed us a thick bush of acacia cornigera, which had become celebrated by a deplorable event. of all the species of mimosa the acacia is that which is armed with the sharpest thorns; they are sometimes two inches long; and being hollow, serve for the habitation of ants of an extraordinary size. a woman, annoyed by the jealousy and well founded reproaches of her husband, conceived a project of the most barbarous vengeance. with the assistance of her lover she bound her husband with cords, and threw him, at night, into a bush of mimosa cornigera. the more violently he struggled, the more the sharp woody thorns of the tree tore his skin. his cries were heard by persons who were passing, and he was found after several hours of suffering, covered with blood, and dreadfully stung by the ants. this crime is perhaps without example in the history of human turpitude: it indicates a violence of passion less assignable to the climate than to the barbarism of manners prevailing among the lower class of the people.

my most important occupation at carthagena was the comparison of my observations with the astronomical positions fixed by the officers of the expedition of fidalgo. in the year 1783 (under the ministry of m. valdes) don josef espinosa, don dionisio galiano and don josef de lanz proposed to the spanish government a plan for taking a survey of the coast of america, in order to extend the atlas of tofino to the western colonies. the plan was approved; but it was not till 1792 that an expedition was fitted out at cadiz, and they were enabled to commence their scientific operations at the island of trinidad.