Personal Narrative of Travels to the Equinoctial Regions of America Volume Iii Part 6

The island of Cuba being surrounded with shoals and breakers along more than two-thirds of its length, and as ships keep out beyond those dangers, the real shape of the island was for a long time unknown. Its breadth, especially between the Havannah and the port of Batabano, has been exaggerated; and it is only since the Deposito hidrografico of Madrid published the observations of captain Don Jose del Rio, and lieutenant Don Ventura de Barcaiztegui, that the area of the island of Cuba could be calculated with any accuracy. Wishing to furnish in this work the most accurate result that can be obtained in the present state of our astronomical knowledge, I engaged M. Bauza to calculate the area. He found, in June, 1835, the surface of the island of Cuba, without the Isla dos Pinos, to be 3520 square sea leagues, and with that island 3615. From this calculation, which has been twice repeated, it results that the island of Cuba is one-seventh less than has. .h.i.therto been believed; that it is 32/100 larger than Hayti, or San Domingo; that its surface equals that of Portugal, and within one-eighth that of England without Wales; and that if the whole archipelago of the Antilles presents as great an area as the half of Spain, the island of Cuba alone almost equals in surface the other Great and Small Antilles. Its greatest length, from Cape San Antonio to Point Maysi (in a direction from west-south-west to east-north-east and from west-north-west to east-south-east) is 227 leagues; and its greatest breadth (in the direction north and south), from Point Maternillo to the mouth of the Magdalena, near Peak Tarquino, is 37 leagues. The mean breadth of the island, on four-fifths of its length, between the Havannah and Puerto Principe, is 15 leagues. In the best cultivated part, between the Havannah and Batabano, the isthmus is only eight sea leagues. Among the great islands of the globe, that of Java most resembles the island of Cuba in its form and area (4170 square leagues). Cuba has a circ.u.mference of coast of 520 leagues, of which 280 belong to the south sh.o.r.e, between Cape San Antonio and Punta Maysi.

The island of Cuba, over more than four-fifths of its surface, is composed of low lands. The soil is covered with secondary and tertiary formations, formed by some rocks of gneiss-granite, syenite and euphotide. The knowledge obtained hitherto of the geologic configuration of the country, is as unsatisfactory as what is known respecting the relative age and nature of the soil. It is only ascertained that the highest group of mountains lies at the south-eastern extremity of the island, between Cape Cruz, Punta Maysi, and Holguin. This mountainous part, called the Sierra or Las Montanas del Cobre (the Copper Mountains), situated north-west of the town of Santiago de Cuba, appears to be about 1200 toises in height. If this calculation be correct, the summits of the Sierra would command those of the Blue Mountains of Jamaica, and the peaks of La Selle and La Hotte in the island of San Domingo. The Sierra of Tarquino, fifty miles west of the town of Cuba, belongs to the same group as the Copper Mountains. The island is crossed from east-south-east to west-north-west by a chain of hills, which approach the southern coast between the meridians of La Ciudad de Puerto Principe and the Villa Clara; while, further to the westward towards Alvarez and Matanzas, they stretch in the direction of the northern coast. Proceeding from the mouth of the Rio Guaurabo to the Villa de la Trinidad, I saw on the north-west, the Lomas de San Juan, which form needles or horns more than 300 toises high, with their declivities sloping regularly to the south. This calcareous group presents a majestic aspect, as seen from the anchorage near the Cayo de Piedras. Xagua and Batabano are low coasts; and I believe that, in general, west of the meridian of Matanzas, there is no hill more than 200 toises high, with the exception of the Pan de Guaixabon. The land in the interior of the island is gently undulated, as in England; and it rises only from 45 to 50 toises above the level of the sea. The objects most visible at a distance, and most celebrated by navigators, are the Pan de Matanzas, a truncated cone which has the form of a small monument; the Arcos de Canasi, which appear between Puerto Escondido and Jaruco, like small segments of a circle; the Mesa de Mariel, the Tetas de Managua, and the Pan de Guaixabon. This gradual slope of the limestone formations of the island of Cuba towards the north and west indicates the submarine connection of those rocks with the equally low lands of the Bahama Islands, Florida and Yucatan.

Intellectual cultivation and improvement were so long restricted to the Havannah and the neighbouring districts, that we cannot be surprised at the ignorance prevailing among the inhabitants respecting the geologic formation of the Copper Mountains. Don Francisco Ramirez, a traveller versed in chemical and mineralogical science, informed me that the western part of the island is granitic, and that he there observed gneiss and primitive slate. Probably the alluvial deposits of auriferous sand which were explored with much ardour* at the beginning of the conquest, to the great misfortune of the natives came from those granitic formations (* At Cubanacan, that is, in the interior of the island, near Jagua and Trinidad, where the auriferous sands have been washed by the waters as far as the limestone soil. Martyr d'Anghiera, the most intelligent writer on the Conquest, says: "Cuba is richer in gold than Hispaniola (San Domingo); and at the moment I am writing, 180,000 castillanos of ore have been collected at Cuba."

Herrera estimates the tax called King's-fifth (quinto del Rey), in the island of Cuba, at 6000 pesos, which indicates an annual product of 2000 marks of gold, at 22 carats; and consequently purer than the gold of Sibao in San Domingo. In 1804 the mines of Mexico altogether produced 7000 marks of gold; and those of Peru 3400. It is difficult, in these calculations, to distinguish between the gold sent to Spain by the first Conquistadores, that obtained by washings, and that which had been acc.u.mulated for ages in the hands of the natives, who were pillaged at will. Supposing that in the two islands of Cuba and San Domingo (in Cubanacan and Cibao) the product of the washings was 3000 marks of gold, we find a quant.i.ty three times less than the gold furnished annually (1790 to 1805) by the small province of Choco. In this supposition of ancient wealth there is nothing improbable; and if we are surprised at the scanty produce of the gold-washings attempted in our days at Cuba and San Domingo, which were heretofore so prolific, it must be recollected that at Brazil also the product of the gold-washings has fallen, from 1760 to 1820, from 6600 gold kilogrammes to less than 595. Lumps of gold weighing several pounds, found in our days in Florida and North and South Carolina, prove the primitive wealth of the whole basin of the Antilles from the island of Cuba to the Appalachian chain. It is also natural that the product of the gold-washings should diminish with greater rapidity than that of the subterraneous working of the veins. The metals not being renewed in the clefts of the veins (by sublimation) now acc.u.mulate in alluvial soil by the course of the rivers where the table-lands are higher than the level of the surrounding running waters. But in rocks with metalliferous veins the miner does not at once know all he has to work. He may chance to lengthen the labours, to go deep, and to cross other accompanying veins. Alluvial soils are generally of small depth where they are auriferous; they most frequently rest upon sterile rocks. Their superficial position and uniformity of composition help to the knowledge of their limits, and wherever workmen can be collected, and where the waters for the washings abound, accelerate the total working of the auriferous clay. These considerations, suggested by the history of the Conquest, and by the science of mining, may throw some light on the problem of the metallic wealth of Hayti. In that island, as well as at Brazil, it would be more profitable to attempt subterraneous workings (on veins) in primitive and intermediary soils than to renew the gold-washings which were abandoned in the ages of barbarism, rapine and carnage.); traces of that sand are still found in the rivers Holguin and Escambray, known in general in the vicinity of Villa-Clara, Santo Espiritu, Puerto del Principe de Bayamo and the Bahia de Nipe. The abundance of copper mentioned by the Conquistadores of the sixteenth century, at a period when the Spaniards were more attentive than they have been in latter times to the natural productions of America, may possibly be attributed to the formations of amphibolic slate, transition clay-slate mixed with diorite, and to euphotides a.n.a.logous to those I found in the mountains of Guanabacoa.

The central and western parts of the island contain two formations of compact limestone; one of clayey sandstone and another of gypsum. The former has, in its aspect and composition, some resemblance to the Jura formation. It is white, or of a clear ochre-yellow, with a dull fracture, sometimes conchoidal, sometimes smooth; divided into thin layers, furnishing some b.a.l.l.s of pyromac silex, often hollow (at Rio Canimar two leagues east of Matanzas), and petrifications of pecten, cardites, terebratules and madrepores.* (* I saw neither gryphites nor ammonites of Jura limestone nor the nummulites and cerites of coa.r.s.e limestone.) I found no oolitic beds, but porous beds almost bulbous, between the Potrero del Conde de Mopox, and the port of Batabano, resembling the spongy beds of Jura limestone in Franconia, near Dondorf, Pegnitz, and Tumbach. Yellowish cavernous strata, with cavities from three to four inches in diameter, alternate with strata altogether compact,* and poorer in petrifications. (* The western part of the island has no deep ravines; and we recognize this alternation in travelling from the Havannah to Batabano, the deepest beds (inclined from 30 to 40 degrees north-east) appear as we advance.) The chain of hills that borders the plain of Guines on the north and is linked with the Lomas de Camua, and the Tetas de Managua, belongs to the latter variety, which is reddish white, and almost of lithographic nature, like the Jura limestone of Pappenheim. The compact and cavernous beds contain nests of brown ochreous iron; possibly the red earth (tierra colorada) so much sought for by the coffee planters (haciendados) owes its origin to the decomposition of some superficial beds of oxidated iron, mixed with silex and clay, or to a reddish sandstone* (* Sandstone and ferruginous sand; iron-sand?) superposed on limestone. The whole of this formation, which I shall designate by the name of the limestone of Guines, to distinguish it from another much more recent, forms, near Trinidad, in the Lomas of St. Juan, steep declivities, resembling the mountains of limestone of Caripe, in the vicinity of c.u.mana. They also contain great caverns, near Matanzas and Jaruco, where I have not heard that any fossil bones have been found. The frequency of caverns in which the pluvial waters acc.u.mulate, and where small rivers disappear, sometimes causes a sinking of the earth. I am of opinion that the gypsum of the island of Cuba belongs not to tertiary but to secondary soil; it is worked in several places on the east of Matanzas, at San Antonio de los Banos, where it contains sulphur, and at the Cayos, opposite San Juan de los Remedios. We must not confound with this limestone of Guines, sometimes porous, sometimes compact, another formation so recent that it seems to augment in our days. I allude to the calcareous agglomerates, which I saw in the islands of Cayos that border the coast between the Batabano and the bay of Xagua, princ.i.p.ally south of the Cienega de Zapata, Cayo Buenito, Cayo Flamenco and Cayo de Piedras. The soundings prove that they are rocks rising abruptly from a bottom of between twenty and thirty fathoms. Some are at the water's edge, others one-fourth or one-fifth of a toise above the surface of the sea. Angular fragments of madrepores, and cellularia from two to three cubic inches, are found cemented by grains of quartzose sand.

The inequalities of the rocks are covered by mould, in which, by help of a microscope, we only distinguish the detritus of sh.e.l.ls and corals. This tertiary formation no doubt belongs to that of the coast of c.u.mana, Carthagena, and the Great Land of Guadaloupe, noticed in my geognostic table of South America.* (* M. Moreau de Jonnes has well distinguished, in his Histoire physique des Antilles Francoises, between the Roche a ravets of Martinique and Hayti, which is porous, filled with terebratulites, and other vestiges of sea-sh.e.l.ls, somewhat a.n.a.logous to the limestone of Guines and the calcareous pelagic sediment called at Guadaloupe Platine, or Maconne bon Dieu. In the cayos of the island of Cuba, or Jardinillos del Rey y del Reyna, the whole coral rock lying above the surface of the water appeared to me to be fragmentary, that is, composed of broken blocks. It is, however, probable, that in the depth it reposes on ma.s.ses of polypi still living.) MM. Chamiso and Guiamard have recently thrown great light on the formation of the coral islands in the Pacific. At the foot of the Castillo de in Punta, near the Havannah, on shelves of cavernous rocks,* covered with verdant sea-weeds and living polypi, we find enormous ma.s.ses of madrepores and other lithophyte corals set in the texture of those shelves. (* The surface of these shelves, blackened and excavated by the waters, presents ramifications like the cauliflower, as they are observed on the currents of lava. Is the change of colour produced by the waters owing to the manganese which we recognize by some dendrites? The sea, entering into the clefts of the rocks, and in a cavern at the foot of the Castillo del Morro, compresses the air and makes it issue with a tremendous noise. This noise explains the phenomena of the baxos roncadores (snoring bocabeoos), so well known to navigators who cross from Jamaica to the mouth of Rio San Juan of Nicaragua, or to the island of San Andres.) We are at first tempted to admit that the whole of this limestone rock, which const.i.tutes the princ.i.p.al portion of the island of Cuba, may be traced to an uninterrupted operation of nature--to the action of productive organic forces--an action which continues in our days in the bosom of the ocean; but this apparent novelty of limestone formations soon vanishes when we quit the sh.o.r.e, and recollect the series of coral rocks which contain the formations of different ages, the muschelkalk, the Jura limestone and coa.r.s.e limestone. The same coral rocks as those of the Castillo and La Punta are found in the lofty inland mountains, accompanied with petrifications of bivalve sh.e.l.ls, very different from those now seen on the coasts of the Antilles. Without positively a.s.signing a determinate place in the table of formations to the limestone of Guines, which is that of the Castillo and La Punta, I have no doubt of the relative antiquity of that rock with respect to the calcareous agglomerate of the Cayos, situated south of Batabano, and east of the island of Pinos. The globe has undergone great revolutions between the periods when these two soils were formed; the one containing the great caverns of Matanzas, the other daily augmenting by the agglutination of fragments of coral and quartzose sand. On the south of the island of Cuba, the latter soil seems to repose sometimes on the Jura limestone of Guines, as in the Jardinillos, and sometimes (towards Cape Cruz) immediately over primitive rocks. In the lesser Antilles the corals are covered with volcanic productions. Several of the Cayos of the island of Cuba contain fresh water; and I found this water very good in the middle of the Cayo de Piedras. When we reflect on the extreme smallness of these islands we can scarcely believe that the fresh-water wells are filled with rain-water not evaporated. Do they prove a submarine communication between the limestone of the coast with the limestone serving as the basis of lithophyte polypi, and is the fresh water of Cuba raised up by hydrostatic pressure across the coral rocks of Cayos, as it is in the bay of Xagua, where, in the middle of the sea, it forms springs frequented by the lamantins?

The secondary formations on the east of the Havannah are pierced in a singular manner by syenitic and euphotide rocks united in groups. The southern bottom of the bay as well as the northern part (the hills of the Morro and the Cabana) are of Jura limestone; but on the eastern bank of the two Ensenadas de Regla and Guanabacoa, the whole is transition soil. Going from north to south, and first near Marimelena, we find syenite consisting of a great quant.i.ty of hornblende, partly decomposed, a little quartz, and a reddish-white feldspar seldom crystallized. This fine syenite, the strata of which incline to the north-west, alternates twice with serpentine. The layers of intercalated serpentine are three toises thick. Farther south, towards Regla and Guanabacoa, the syenite disappears, and the whole soil is covered with serpentine, rising in hills from thirty to forty toises high, and running from east to west. This rock is much fendillated, externally of a bluish-grey, covered with dendrites of manganese, and internally of leek and asparagus-green, crossed by small veins of asbestos. It contains no garnet or amphibole, but metalloid diallage disseminated in the ma.s.s. The serpentine is sometimes of an esquillous, sometimes of a conchoidal fracture: this was the first time I had found metalloid diallage within the tropics. Several blocks of serpentine have magnetic poles; others are of such a h.o.m.ogeneous texture, and have such a glossiness, that at a distance they may be taken for pechstein (resinite). It were to be wished that these fine ma.s.ses were employed in the arts as they are in several parts of Germany. In approaching Guanabacoa we find serpentine crossed by veins between twelve and fourteen inches thick, and filled with fibrous quartz, amethyst, and fine mammelonnes, and stalactiforme chalcedonies; it is possible that chrysoprase may also one day be found. Some copper pyrites appear among these veins accompanied, it is said, by silvery-grey copper. I found no traces of this grey copper: it is probably the metalloid diallage that has given the Cerro de Guanabacoa the reputation of riches in gold and silver which it has enjoyed for ages. In some places petroleum flows* from rents in the serpentine. (* Does there exist in the Bay of the Havannah any other source of petroleum than that of Guanabacoa, or must it be admitted that the betun liquido, which in 1508 was employed by Sebastian de Ocampo for the caulking of ships, is dried up? That spring, however, fixed the attention of Ocampo on the port of the Havannah, where he gave it the name of Puerto de Carenas. It is said that abundant springs of petroleum are also found in the eastern part of the island (Manantialis de betun y chapapote) between Holguin and Mayari, and on the coast of Santiago de Cuba.) Springs of water are frequent; they contain a little sulphuretted hydrogen, and deposit oxide of iron. The Baths of Bareto are agreeable, but of nearly the same temperature as the atmosphere. The geologic const.i.tution of this group of serpentine rocks, from its insulated position, its veins, its connection with syenite and the fact of its rising up across sh.e.l.l-formations, merits particular attention. Feldspar with a basis of souda (compact feldspar) forms, with diallage, the euphotide and serpentine; with pyroxene, dolerite and basalt; and with garnet, eclogyte. These five rocks, dispersed over the whole globe, charged with oxidulated and t.i.tanious iron, are probably of similar origin. It is easy to distinguish two formations in the euphotide; one is dest.i.tute of amphibole, even when it alternates with amphibolic rocks (Joria in Piedmont, Regla in the island of Cuba) rich in pure serpentine, in metalloid diallage and sometimes in jasper (Tuscany, Saxony); the other, strongly charged with amphibole, often pa.s.sing to diorite,* has no jasper in layers, and sometimes contains rich veins of copper; (Silesia, Mussinet in Piedmont, the Pyrenees, Parapara in Venezuela, Copper Mountains of North America). (* On a serpentine that flows like a penombre, veins of greenstone (diorite) near Lake Clunie in Perthshire. See MacCulloch in Edinburgh Journal of Science 1824 July pages 3 to 16. On a vein of serpentine, and the alterations it produces on the banks of Carity, near West-Balloch in Forfarshire see Charles Lyell l.c. volume 3 page 43.) It is the latter formation of euphotide which, by its mixture with diorite, is itself linked with hyperthenite, in which real beds of serpentine are sometimes developed in Scotland and in Norway. No volcanic rocks of a more recent period have hitherto been discovered in the island of Cuba; for instance, neither trachytes, dolerites, nor basalts. I know not whether they are found in the rest of the Great Antilles, of which the geologic const.i.tution differs essentially from that of the series of calcareous and volcanic islands which stretch from Trinidad to the Virgin Islands. Earthquakes, which are in general less fatal at Cuba than at Porto Rico and Hayti, are most felt in the eastern part, between Cape Maysi, Santiago de Cuba and La Ciudad de Puerto Principe. Perhaps towards those regions the action of the crevice extends laterally, which is believed to cross the neck of granitic land between Port-au-Prince and Cape Tiburon and on which whole mountains were overthrown in 1770.

The cavernous texture of the limestone formations (soboruco) just described, the great inclination of the shelvings, the smallness of the island, the nakedness of the plains and the proximity of the mountains that form a lofty chain on the southern coast, may be considered as among the princ.i.p.al causes of the want of rivers and the drought which is felt, especially in the western part of Cuba. In this respect, Hayti, Jamaica, and several of the Lesser Antilles, which contain volcanic heights covered with forests, are more favoured by nature. The lands most celebrated for their fertility are the districts of Xagua, Trinidad, Matanzas and Mariel. The valley of Guines owes its reputation to artificial irrigation (sanjas de riego).

Notwithstanding the want of great rivers and the unequal fertility of the soil, the island of Cuba, by its undulated surface, its continually renewed verdure, and the distribution of its vegetable forms, presents at every step the most varied and beautiful landscape.

Two trees with large, tough, and glossy leaves, the Mammea and the Calophyllum calaba, five species of palm-trees (the palma real, or Oreodoxa regia, the common cocoa-tree, the Cocos crispa, the Corypha miraguama and the C. maritima), and small shrubs constantly loaded with flowers, decorate the hills and the savannahs. The Cecropia peltata marks the humid spots. It would seem as if the whole island had been originally a forest of palm, lemon, and wild orange trees.

The latter, which bear a small fruit, are probably anterior to the arrival of Europeans,* who transported thither the agrumi of the gardens; they rarely exceed the height of from ten to fifteen feet. (*

The best informed inhabitants of the island a.s.sert that the cultivated orange-trees brought from Asia preserve the size and all the properties of their fruits when they become wild. The Brazilians affirm that the small bitter orange which bears the name of loranja do terra and is found wild, far from the habitations of man, is of American origin. Caldcleugh, Travels in South America.) The lemon and orange trees are most frequently separate; and the new planters, in clearing the ground by fire, distinguish the quality of the soil according as it is covered with one or other of those groups of social plants; they prefer the soil of the naranjal to that which produces the small lemon. In a country where the making of sugar is not sufficiently improved to admit of the employment of any other fuel than the baga.s.se (dried sugar-cane) the progressive destruction of the small woods is a positive calamity. The aridity of the soil augments in proportion as it is stripped of the trees that sheltered it from the heat of the sun; for the leaves, emitting heat under a sky always serene, occasion, as the air cools, a precipitation of aqueous vapours.

Among the few rivers worthy of attention, the Rio Guines may be noticed, the Rio Armendaris or Chorrera, of which the waters are led to the Havannah by the Sanja de Antoneli; the Rio Canto on the north of the town of Bayamo; the Rio Maximo which rises on the east of Puerto Principe; the Rio Sagua Grande near Villa Clara; the Rio de las Palmas which issues opposite Cayo Galiado; the small rivers of Jaruco and Santa Cruz between Guanabo and Matanzas, navigable at the distance of some miles from their mouths and favourable for the shipment of sugar-casks; the Rio San Antonio which, like many others, is engulfed in the caverns of limestone rocks; the Rio Guaurabo west of the port of Trinidad; and the Rio Galafre in the fertile district of Filipinas, which throws itself into the Laguna de Cortez. The most abundant springs rise on the southern coast where, from Xagua to Punta de Sabina, over a length of forty-six leagues, the soil is extremely marshy. So great is the abundance of the waters which filter by the clefts of the stratified rock that, from the effect of an hydrostatic pressure, fresh water springs far from the coast, and amidst salt water. The jurisdiction of the Havannah is not the most fertile part of the island; and the few sugar-plantations that existed in the vicinity of the capital are now converted into farms for cattle (potreros) and fields of maize and forage, of which the profits are considerable. The agriculturists of the island of Cuba distinguish two kinds of earth, often mixed together like the squares of a draught-board, black earth (negra o prieta), clayey and full of moisture, and red earth (bermeja), more silicious and containing oxide of iron. The tierra negra is generally preferred (on account of its best preserving humidity) for the cultivation of the sugarcane, and the tierra bermeja for coffee; but many sugar plantations are established on the red soil.

The climate of the Havannah is in accordance with the extreme limits of the torrid zone: it is a tropical climate, in which a more unequal distribution of heat at different parts of the year denotes the pa.s.sage to the climates of the temperate zone. Calcutta (lat.i.tude 22 degrees 34 minutes north), Canton (lat.i.tude 23 degrees 8 minutes north), Macao (lat.i.tude 22 degrees 12 minutes north), the Havannah (lat.i.tude 23 degrees 9 minutes north) and Rio Janeiro (lat.i.tude 22 degrees 54 minutes south) are places which, from their position at the level of the ocean near the tropics of Cancer and Capricorn, consequently at an equal distance from the equator, afford great facilities for the study of meteorology. This study can only advance by the determination of certain numerical elements which are the indispensable basis of the laws we seek to discover. The aspect of vegetation being identical near the limits of the torrid zone and at the equator, we are accustomed to confound vaguely the climates of two zones comprised between 0 and 10 degrees, and between 15 and 23 degrees of lat.i.tude. The region of palm-trees, bananas and arborescent gramina extends far beyond the two tropics: but it would be dangerous to apply what has been observed at the extremity of the tropical zone to what may take place in the plains near the equator. In order to rectify those errors it is important that the mean temperature of the year and months be well known, as also the thermometric oscillations in different seasons at the parallel of the Havannah; and to prove by an exact comparison with other points alike distant from the equator, for instance, with Rio Janeiro and Macao, that the lowering of temperature observed in the island of Cuba is owing to the irruption and the stream of layers of cold air, borne from the temperate zones towards the tropics of Cancer and Capricorn. The mean temperature of the Havannah, according to four years of good observations, is 25.7 degrees (20.6 degrees R.), only 2 degrees centigrade above that of the regions of America nearest the equator. The proximity of the sea raises the mean temperature of the year on the coast; but in the interior of the island, when the north winds penetrate with the same force, and where the soil rises to the height of forty toises, the mean temperature attains only 23 degrees (18.4 degrees R.) and does not exceed that of Cairo and Lower Egypt. The difference between the mean temperature of the hottest and coldest months rises to 12 degrees in the interior of the island; at the Havannah and on the coast, to 8 degrees; at c.u.mana, to scarcely 3 degrees. The hottest months, July and August, attain 28.8 degrees, at the island of Cuba, perhaps 29.5 degrees of mean temperature, as at the equator. The coldest months are December and January; their mean temperature in the interior of the island, is 17 degrees; at the Havannah, 21 degrees, that is, 5 to 8 degrees below the same months at the equator, yet still 3 degrees above the hottest month at Paris.

It will be interesting to compare the climate of the Havannah with that of Macao and Rio Janeiro; two places, one of which is near the limit of the northern torrid zone, on the eastern coast of Asia; and the other on the eastern coast of America, towards the extremity of the southern torrid zone.

The climate of the Havannah, notwithstanding the frequency of the north and north-west winds, is hotter than that of Macao and Rio Janeiro. The former partakes of the cold which, owing to the frequency of the west winds, is felt in winter along all the eastern coast of a great continent. The proximity of s.p.a.ces of land covered with mountains and table-lands renders the distribution of heat in different months of the year more unequal at Macao and Canton than in an island bounded on the west and north by the hot waters of the Gulf-stream. The winters are therefore much colder at Canton and Macao than at the Havannah: yet the lat.i.tude of Macao is 1 degree more southerly than that of the Havannah; and the latter town and Canton are, within nearly a minute, on the same parallel. The thermometer at Canton has sometimes almost reached the point zero; and by the effect of reflection, ice has been found on the terraces of houses. Although this great cold never lasts more than one day, the English merchants residing at Canton like to make chimney-fires in their apartments from November to January; while at the Havannah, the artificial warmth even of a brazero is not required. Hail is frequent and the hail-stones are extremely large in the Asiatic climate of Canton and Macao, while it is scarcely seen once in fifteen years at the Havannah. In these three places the thermometer sometimes keeps up for several hours between 0 and 4 degrees (centigrade); and yet (a circ.u.mstance which appears to be very remarkable) snow has never been seen to fall; and notwithstanding the great lowering of the temperature, the bananas and the palm-trees are as beautiful around Canton, Macao and the Havannah as in the plains nearest the equator.

In the island of Cuba the lowering of the temperature lasts only during intervals of such short duration that in general neither the banana, the sugar-cane nor other productions of the torrid zone suffer much. We know how well plants of vigorous organization resist temporary cold, and that the orange trees of Genoa survive the fall of snow and endure cold which does not more than exceed 6 or 7 degrees below freezing-point. As the vegetation of the island of Cuba bears the character of the vegetation of the regions near the equator, we are surprised to find even in the plains a vegetable form of the temperate climates and mountains of the equatorial part of Mexico. I have often directed the attention of botanists to this extraordinary phenomenon in the geography of plants. The pine (Pinus occidentalis) is not found in the Lesser Antilles; not even in Jamaica (between 17 3/4 and 18 1/2 degrees of lat.i.tude). It is only seen further north, in the mountains of San Domingo, and in all that part of the island of Cuba situated between 20 and 23 degrees of lat.i.tude. It attains a height of from sixty to seventy feet; and it is remarkable that the cahoba* (mahogany (* Swieteinia Mahogani, Linn.)) and the pine vegetate at the island of Pinos in the same plains. We also find pines in the south-eastern part of the island of Cuba, on the declivity of the Copper Mountains where the soil is barren and sandy. The interior table-land of Mexico is covered with the same species of coniferous plants; at least the specimens brought by M. Bonpland and myself from Acaguisotla, Nevado de Toluca and Cofre de Perote do not appear to differ specifically from the Pinus occidentalis of the West India Islands described by Schwartz. Now those pines which we see at sea level in the island of Cuba, in 20 and 22 degrees of lat.i.tude, and which belong only to the southern part of that island, do not descend on the Mexican continent between the parallels of 17 1/2 and 19 1/2 degrees, below the elevation of 500 toises. I even observed that, on the road from Perote to Xalapa in the eastern mountains opposite to the island of Cuba, the limit of the pines is 935 toises; while in the western mountains, between Chilpanzingo and Acapulco, near Quasiniquilapa, two degrees further south, it is 580 toises and perhaps on some points 450. These anomalies of stations are very rare in the torrid zone and are probably less connected with the temperature than with the nature of the soil. In the system of the migration of plants we must suppose that the Pinus occidentalis of Cuba came from Yucatan before the opening of the channel between Cape Catoche and Cape San Antonio, and not from the United States, so rich in coniferous plants; for in Florida the species of which we have here traced the botanical geography has not been discovered.

About the end of April, M. Bonpland and myself, having completed the observations we proposed to make at the northern extremity of the torrid zone, were on the point of proceeding to Vera Cruz with the squadron of Admiral Ariztizabal; but being misled by false intelligence respecting the expedition of Captain Baudin, we were induced to relinquish the project of pa.s.sing through Mexico on our way to the Philippine Islands. The public journals announced that two French sloops, the Geographe and Naturaliste, had sailed for Cape Horn; that they were to proceed along the coasts of Chili and Peru, and thence to New Holland. This intelligence revived in my mind all the projects I had formed during my stay in Paris, when I solicited the Directory to hasten the departure of Captain Baudin. On leaving Spain, I had promised to rejoin the expedition wherever I could reach it. M. Bonpland and I resolved instantly to divide our herbals into three portions, to avoid exposing to the risks of a long voyage the objects we had obtained with so much difficulty on the banks of the Orinoco, the Atabapo and the Rio Negro. We sent one collection by way of England to Germany, another by way of Cadiz to France, and a third remained at the Havannah. We had reason to congratulate ourselves on this foresight: each collection contained nearly the same species, and no precautions were neglected to have the cases, if taken by English or French vessels, remitted to Sir Joseph Banks or to the professors of natural history at the Museum at Paris. It happened fortunately that the ma.n.u.scripts which I at first intended to send with the collection to Cadiz were not intrusted to our much esteemed friend and fellow traveller, Fray Juan Gonzales, of the order of the Observance of St. Francis, who had followed us to the Havannah with the view of returning to Spain. He left the island of Cuba soon after us, but the vessel in which he sailed foundered on the coast of Africa, and the cargo and crew were all lost. By this event we lost some of the duplicates of our herbals, and what was more important, all the insects which M. Bonpland had with great difficulty collected during our voyage to the Orinoco and the Rio Negro. By a singular fatality, we remained two years in the Spanish colonies without receiving a single letter from Europe; and those which arrived in the three following years made no mention of what we had transmitted. The reader may imagine my uneasiness for the fate of a journal which contained astronomical observations and barometrical measurements, of which I had not made any copy. After having visited New Grenada, Peru and Mexico, and just when I was preparing to leave the New Continent, I happened, at a public library of Philadelphia, to cast my eyes on a scientific Publication, in which I found these words: "Arrival of M.

de Humboldt's ma.n.u.scripts at his brother's house in Paris, by way of Spain!" I could scarcely suppress an exclamation of joy.

While M. Bonpland laboured day and night to divide and put our collections in order, a thousand obstacles arose to impede our departure. There was no vessel in the port of the Havannah that would convey us to Porto Bello or Carthagena. The persons I consulted seemed to take pleasure in exaggerating the difficulties of the pa.s.sage of the isthmus, and the dangerous voyage from Panama to Guyaquil, and from Guyaquil to Lima and Valparaiso. Not being able to find a pa.s.sage in any neutral vessel, I freighted a Catalonian sloop, lying at Batabano, which was to be at my disposal to take me either to Porto Bello or Carthagena, according as the gales of Saint Martha might permit.* (* The gales of Saint Martha blow with great violence at that season below lat.i.tude 12 degrees.) The prosperous state of commerce at the Havannah and the multiplied connections of that city with the ports of the Pacific would facilitate for me the means of procuring funds for several years. General Don Gonzalo O'Farrill resided at that time in my native country as minister of the court of Spain. I could exchange my revenues in Prussia for a part of his at the island of Cuba; and the family of Don Ygnacio O'Farrill y Herera, brother of the general, concurred kindly in all that could favour my new projects. On the 6th of March the vessel I had freighted was ready to receive us.

The road to Batabano led us once more by Guines to the plantation of Rio Blanco, the property of Count Jaruco y Mopox.

The road from Rio Blanco to Batabano runs across an uncultivated country, half covered with forests; in the open spots the indigo plant and the cotton-tree grow wild. As the capsule of the Gossypium opens at the season when the northern storms are most frequent, the down that envelops the seed is swept from one side to the other; and the gathering of the cotton, which is of a very fine quality, suffers greatly. Several of our friends, among whom was Senor de Mendoza, captain of the port of Valparaiso, and brother to the celebrated astronomer who resided so long in London, accompanied us to Potrero de Mopox. In herborizing further southward, we found a new palm-tree with fan-leaves (Corypha maritima), having a free thread between the interstices of the folioles. This Corypha covers a part of the southern coast and takes the place of the majestic palma real and the Cocos crispa of the northern coast. Porous limestone (of the Jura formation) appeared from time to time in the plain.

Batabano was then a poor village and its church had been completed only a few years previously. The Sienega begins at the distance of half a league from the village; it is a tract of marshy soil, extending from the Laguna de Cortez as far as the mouth of the Rio Xagua, on a length of sixty leagues from west to east. At Batabano it is believed that in those regions the sea continues to gain upon the land, and that the oceanic irruption was particularly remarkable at the period of the great upheaving which took place at the end of the eighteenth century, when the tobacco mills disappeared, and the Rio Chorrera changed its course. Nothing can be more gloomy than the aspect of these marshes around Batabano. Not a shrub breaks the monotony of the prospect: a few stunted trunks of palm-trees rise like broken masts, amidst great tufts of Junceae and Irides. As we stayed only one night at Batabano, I regretted much that I was unable to obtain precise information relative to the two species of crocodiles which infest the Sienega. The inhabitants give to one of these animals the name of cayman, to the other that of crocodile; or, as they say commonly in Spain, of cocodrilo. They a.s.sured us that the latter has most agility, and measures most in height: his snout is more pointed than that of the cayman, and they are never found together. The crocodile is very courageous and is said to climb into boats when he can find a support for his tail. He frequently wanders to the distance of a league from the Rio Cauto and the marshy coast of Xagua to devour the pigs on the islands. This animal is sometimes fifteen feet long, and will, it is said, pursue a man on horseback, like the wolves in Europe; while the animals exclusively called caymans at Batabano are so timid that people bathe without apprehension in places where they live in bands. These peculiarities, and the name of cocodrilo, given at the island of Cuba, to the most dangerous of the carnivorous reptiles, appear to me to indicate a different species from the great animals of the Orinoco, Rio Magdalena and Saint Domingo. In other parts of the Spanish American continent the settlers, deceived by the exaggerated accounts of the ferocity of crocodiles in Egypt, allege that the real crocodile is only found in the Nile. Zoologists have, however, ascertained that there are in America caymans or alligators with obtuse snouts, and legs not indented, and crocodiles with pointed snouts and indented legs; and in the old continent, both crocodiles and gaviales. The Crocodilus acutus of San Domingo, in which I cannot hitherto specifically distinguish the crocodiles of the great rivers of the Orinoco and the Magdalena, has, according to Cuvier, so great a resemblance to the crocodile of the Nile,* that it required a minute examination to prove that the rule laid down by Buffon relative to the distribution of species between the tropical regions of the two continents was correct. (* This striking a.n.a.logy was ascertained by M.

Geoffroy de Saint Hilaire in 1803 when General Rochambeau sent a crocodile from San Domingo to the Museum of Natural History at Paris.

M. Bonpland and myself had made drawings and detailed descriptions in 1801 and 1802 of the same species which inhabit the great rivers of South America, during our pa.s.sage on the Apure, the Orinoco and the Magdalena. We committed the mistake so common to travellers, of not sending them at once to Europe, together with some young specimens.)

On my second visit to the Havannah, in 1804, I could not return to the Sienega of Batabano; and therefore I had the two species, called caymans and crocodiles by the inhabitants, brought to me, at a great expense. Two crocodiles arrived alive; the oldest was four feet three inches long; they had been caught with great difficulty and were conveyed, muzzled and bound, on a mule, for they were exceedingly vigorous and fierce. In order to observe their habits and movements,*

we placed them in a great hall, where, by climbing on a very high piece of furniture, we could see them attack great dogs. (* M.

Descourtils, who knows the habits of the crocodile better than any other author who has written on that reptile, saw, like Dampier and myself, the Crocodilus acutus often touch his tail with his mouth.) Having seen much of crocodiles during six months, on the Orinoco, the Rio Apure and the Magdalena, we were glad to have another opportunity of observing their habits before our return to Europe. The animals sent to us from Batabano had the snout nearly as sharp as the crocodiles of the Orinoco and the Magdalena (Crocodilus acutus, Cuv.); their colour was dark-green on the back, and white below the belly, with yellow spots on the flanks. I counted, as in all the real crocodiles, thirty-eight teeth in the upper jaw, and thirty in the lower; in the former, the tenth and ninth; and in the latter, the first and fourth, were the largest. In the description made by M.

Bonpland and myself on the spot, we have expressly marked that the lower fourth tooth rises over the upper jaw. The posterior extremities were palmated. These crocodiles of Batabano appeared to us to be specifically identical with the Crocodilus acutus. It is true that the accounts we heard of their habits did not quite agree with what we had ourselves observed on the Orinoco; but carnivorous reptiles of the same species are milder and more timid, or fiercer and more courageous, in the same river, according to the nature of the localities. The animal called the cayman, at Batabano, died on the way, and was not brought to us, so that we could make no comparison of the two species.* (* The four bags filled with musk (bolzas del almizcle) are, in the crocodile of Batabano, exactly in the same position as in that of the Rio Magdalena, beneath the lower jaw and near the a.n.u.s. I was much surprised at not perceiving the smell of musk at the Havannah, three days after the death of the animal, in a temperature of 30 degrees, while at Mompox, on the banks of the Magdalena, living crocodiles infected our apartment. I have since found that Dampier also remarked an absence of smell in the crocodile of Cuba where the caymans spread a very strong smell of musk.) I have no doubt that the crocodile with a sharp snout, and the alligator or cayman with a snout like a pike,* (* Crocodilus acutus of San Domingo.

Alligator lucius of Florida and the Mississippi.) inhabit together, but in distinct bands, the marshy coast between Xagua, the Surgidero of Batabano, and the island of Pinos. In that island Dampier was struck with the great difference between the caymans and the American crocodiles. After having described, though not always with perfect correctness, several of the characteristics which distinguish crocodiles from caymans, he traces the geographical distribution of those enormous saurians. "In the bay of Campeachy," he says, "I saw only caymans or alligators; at the island of Great Cayman, there are crocodiles and no alligators; at the island of Pinos, and in the innumerable creeks of the coast of Cuba, there are both crocodiles and caymans."* (* Dampier's Voyages and Descriptions, 1599.) To these valuable observations of Dampier I may add that the real crocodile (Crocodilus acutus) is found in the West India Islands nearest the mainland, for instance, at the island of Trinidad; at Marguerita; and also, probably, at Curacao, notwithstanding the want of fresh water.

It is observed, further south, in the Neveri, the Rio Magdalena, the Apure and the Orinoco, as far as the confluence of the Ca.s.siquiare with the Rio Negro (lat.i.tude 2 degrees 2 minutes), consequently more than four hundred leagues from Batabano. It would be interesting to verify on the eastern coast of Mexico and Guatimala, between the Mississippi and the Rio Chagres (in the isthmus of Panama), the limit of the different species of carnivorous reptiles.

We set sail on the 9th of March, somewhat incommoded by the extreme smallness of our vessel, which afforded us no sleeping-place but upon deck. The cabin (camera de pozo) received no air or light but from above; it was merely a hold for provisions, and it was with difficulty that we could place our instruments in it. The thermometer kept up constantly at 32 and 33 degrees (centesimal.) Luckily these inconveniences lasted only twenty days. Our several voyages in the canoes of the Orinoco, and a pa.s.sage in an American vessel laden with several thousand arrobas of salt meat dried in the sun had rendered us not very fastidious.

The gulf of Batabano, bounded by a low and marshy coast, looks like a vast desert. The fishing birds, which are generally at their post whilst the small land birds, and the indolent vultures (Vultur aura.) are at roost, are seen only in small numbers. The sea is of a greenish-brown hue, as in some of the lakes of Switzerland; while the air, owing to its extreme purity, had, at the moment the sun appeared above the horizon, a cold tint of pale blue, similar to that which landscape painters observe at the same hour in the south of Italy, and which makes distant objects stand out in strong relief. Our sloop was the only vessel in the gulf; for the roadstead of Batabano is scarcely visited except by smugglers, or, as they are here politely called, the traders (los tratantes). The projected ca.n.a.l of Guines will render Batabano an important point of communication between the island of Cuba and the coast of Venezuela. The port is within a bay bounded by Punta Gorda on the east, and by Punta de Salinas on the west: but this bay is itself only the upper or concave end of a great gulf measuring nearly fourteen leagues from south to north, and along an extent of fifty leagues (between the Laguna de Cortez and the Cayo de Piedras) inclosed by an incalculable number of flats and chains of rocks. One great island only, of which the superficies is more than four times the dimensions of that of Martinique, with mountains crowned with majestic pines, rises amidst this labyrinth. This is the island of Pinos, called by Columbus El Evangelista, and by some mariners of the sixteenth century, the Isla de Santa Maria. It is celebrated for its mahogany (Swietenia mahagoni) which is an important article of commerce. We sailed east-south-east, taking the pa.s.sage of Don Cristoval, to reach the rocky island of Cayo de Piedras, and to clear the archipelago, which the Spanish pilots, in the early times of the conquest, designated by the names of Gardens and Bowers (Jardines y Jardinillos). The Queen's Gardens, properly so called, are nearer Cape Cruz, and are separated from the archipelago by an open sea thirty-five leagues broad. Columbus gave them the name they bear, in 1494, when, on his second voyage, he struggled during fifty-eight days with the winds and currents between the island of Pinos and the eastern cape of Cuba. He describes the islands of this archipelago as verdant, full of trees and pleasant* (verdes, llenos de arboledas, y graciosos). (* There exists great geographical confusion, even at the Havannah, in reference to the ancient denominations of the Jardines del Rey and Jardines de la Reyna. In the description of the island of Cuba, given in the Mercurio Americano, and in the Historia Natural de la Isla de Cuba, published at the Havannah by Don Antonio Lopez Gomez, the two groups are placed on the southern coast of the island. Lopez says that the Jardines del Rey extend from the Laguna de Cortez to Bahia de Xagua; but it is historically certain that the governor Diego Velasquez gave his name to the western part of the chain of rocks of the Old Channel, between Cayo Frances and Le Monillo, on the northern coast of the island of Cuba. The Jardines de la Reyna, situated between Cabo Cruz and the port of the Trinity, are in no manner connected with the Jardines and Jardinillos of the Isla de Pinos.

Between the two groups of the chain of rocks are the flats (placeres) of La Paz and Xagua.)

A part of these so-styled gardens is indeed beautiful; the voyager sees the scene change every moment, and the verdure of some of the islands appears the more lovely from its contrast with chains of rocks, displaying only white and barren sands. The surface of these sands, heated by the rays of the sun, seems to be undulating like the surface of a liquid. The contact of layers of air of unequal temperature produces the most varied phenomena of suspension and mirage from ten in the morning till four in the afternoon. Even in those desert places the sun animates the landscape, and gives mobility to the sandy plain, to the trunks of trees, and to the rocks that project into the sea like promontories. When the sun appears these inert ma.s.ses seem suspended in air; and on the neighbouring beach the sands present the appearance of a sheet of water gently agitated by the winds. A train of clouds suffices to seat the trunks of trees and the suspended rocks again on the soil; to render the undulating surface of the plains motionless; and to dissipate the charm which the Arabian, Persian, and Hindoo poets have celebrated as "the sweet illusions of the solitary desert."

We doubled Cape Matahambre very slowly. The chronometer of Louis Berthoud having kept time accurately at the Havannah, I availed myself of this occasion to determine, on this and the following days, the positions of Cayo de Don Cristoval, Cayo Flamenco, Cayo de Diego Perez and Cayo de Piedras. I also employed myself in examining the influence which the changes at the bottom of the sea produce on its temperature at the surface. Sheltered by so many islands, the surface is calm as a lake of fresh water, and the layers of different depths being distinct and separate, the smallest change indicated by the lead acts on the thermometer. I was surprised to see that on the east of the little Cayo de Don Cristoval the high banks are only distinguished by the milky colour of the water, like the bank of Vibora, south of Jamaica, and many other banks, the existence of which I ascertained by means of the thermometer. The bottom of the rock of Batabano is a sand composed of coral detritus; it nourishes sea-weeds which scarcely ever appear on the surface: the water, as I have already observed, is greenish; and the absence of the milky tint is, no doubt, owing to the perfect calm which pervades those regions. Whenever the agitation is propagated to a certain depth, a very fine sand, or a ma.s.s of calcareous particles suspended in the water, renders it troubled and milky. There are shallows, however, which are distinguished neither by the colour nor by the low temperature of the waters; and I believe that phenomenon depends on the nature of a hard and rocky bottom, dest.i.tute of sand and corals; on the form and declivity of the shelvings; the swiftness of the currents; and the absence of the propagation of motion towards the lower layers of the water. The cold frequently indicated by the thermometer, at the surface of the high banks, must be traced to the molecules of water which, owing to the rays of heat and the nocturnal cooling, fall from the surface to the bottom, and are stopped in their fall by the high banks; and also to the mingling of the layers of very deep water that rise on the shelvings of the banks as on an inclined plane, to mix with the layers of the surface.

Notwithstanding the small size of our bark and the boasted skill of our pilot, we often ran aground. The bottom being soft, there was no danger; but, nevertheless, at sunset, near the pa.s.s of Don Cristoval, we preferred to lie at anchor. The first part of the night was beautifully serene: we saw an incalculable number of falling-stars, all following one direction, opposite to that from whence the wind blew in the low regions of the atmosphere. The most absolute solitude prevails in this spot, which, in the time of Columbus, was inhabited and frequented by great numbers of fishermen. The inhabitants of Cuba then employed a small fish to take the great sea turtles; they fastened a long cord to the tail of the reves (the name given by the Spaniards to that species of Echeneis*). (* To the sucet or guaican of the natives of Cuba the Spaniards have given the characteristic name of reves, that is, placed on its back, or reversed. In fact, at first sight, the position of the back and the abdomen is confounded.

Anghiera says: Nostrates reversum appellant, quia versus venatur. I examined a remora of the South Sea during the pa.s.sage from Lima to Acapulco. As he lived a long time out of the water, I tried experiments on the weight he could carry before the blades of the disk loosened from the plank to which the animal was fixed; but I lost that part of my journal. It is doubtless the fear of danger that causes the remora not to loose his hold when he feels that he is pulled by a cord or by the hand of man. The sucet spoken of by Columbus and Martin d'Anghiera was probably the Echeneis naucrates and not the Echeneis remora.) The fisher-fish, formerly employed by the Cubans by means of the flattened disc on his head, furnished with suckers, fixed himself on the sh.e.l.l of the sea-turtle, which is so common in the narrow and winding channels of the Jardinillos. "The reves," says Christopher Columbus, "will sooner suffer himself to be cut in pieces than let go the body to which he adheres." The Indians drew to the sh.o.r.e by the same cord the fisher-fish and the turtle. When Gomara and the learned secretary of the emperor Charles V, Peter Martyr d'Anghiera, promulgated in Europe this fact which they had learnt from the companions of Columbus, it was received as a traveller's tale. There is indeed an air of the marvellous in the recital of d'Anghiera, which begins in these words: Non aliter ac nos canibus gallicis per aequora campi lepores insectamur, incolae [Cubae insulae] venatorio pisce pisces alios capiebant. (Exactly as we follow hares with greyhounds in the fields, so do the natives [of Cuba] take fishes with other fish trained for that purpose). We now know, from the united testimony of Rogers, Dampier and Commerson, that the artifice resorted to in the Jardinillos to catch turtles is employed by the inhabitants of the eastern coast of Africa, near Cape Natal, at Mozambique and at Madagascar. In Egypt, at San Domingo and in the lakes of the valley of Mexico, the method practised for catching ducks was as follows: men, whose heads were covered with great calabashes pierced with holes, hid themselves in the water, and seized the birds by the feet. The Chinese, from the remotest antiquity, have employed the cormorant, a bird of the pelican family, for fishing on the coast: rings are fixed round the bird's neck to prevent him from swallowing his prey and fishing for himself. In the lowest degree of civilization, the sagacity of man is displayed in the stratagems of hunting and fishing: nations who probably never had any communication with each other furnish the most striking a.n.a.logies in the means they employ in exercising their empire over animals.

Three days elapsed before we could emerge from the labyrinth of Jardines and Jardinillos. At night we lay at anchor; and in the day we visited those islands or chains of rocks which were most easily accessible. As we advanced eastward the sea became less calm and the position of the shoals was marked by water of a milky colour. On the boundary of a sort of gulf between Cayo Flamenco and Cayo de Piedras we found that the temperature of the sea, at its surface, augmented suddenly from 23.5 to 25.8 degrees centigrade. The geologic const.i.tution of the rocky islets that rise around the island of Pinos fixed my attention the more earnestly as I had always rather doubted of the existence of those huge ma.s.ses of coral which are said to rise from the abyss of the Pacific to the surface of the water. It appeared to me more probable that these enormous ma.s.ses had some primitive or volcanic rock for a basis, to which they adhered at small depths. The formation, partly compact and lithographic, partly bulbous, of the limestone of Guines, had followed us as far as Batabano. It is somewhat a.n.a.logous to Jura limestone; and, judging from their external aspect, the Cayman Islands are composed of the same rock. If the mountains of the island of Pinos, which present at the same time (as it is said by the first historians of the conquest) the pineta and palmeta, be visible at the distance of twenty sea leagues, they must attain a height of more than five hundred toises: I have been a.s.sured that they also are formed of a limestone altogether similar to that of Guines. From these facts I expected to find the same rock (Jura limestone) in the Jardinillos: but I saw, in the chain of rocks that rises generally five to six inches above the surface of the water, only a fragmentary rock, in which angular pieces of madrepores are cemented by quartzose sand. Sometimes the fragments form a ma.s.s of from one to two cubic feet and the grains of quartz so disappear that in several layers one might imagine that the polypi have remained on the spot. The total ma.s.s of this chain of rocks appears to me a limestone agglomerate, somewhat a.n.a.logous to the earthy limestone of the peninsula of Araya, near c.u.mana, but of much more recent formation. The inequalities of this coral rock are covered by a detritus of sh.e.l.ls and madrepores. Whatever rises above the surface of the water is composed of broken pieces, cemented by carbonate of lime, in which grains of quartzose sand are set. Whether rocks formed by polypi still living are found at great depth below this fragmentary rock of coral or whether these polypi are raised on the Jura formation are questions which I am unable to answer. Pilots believe that the sea diminishes in these lat.i.tudes, because they see the chain of rocks augment and rise, either by the earth which the waves heave up, or by successive agglutinations. It is not impossible that the enlarging of the channel of Bahama, by which the waters of the Gulf-stream issue, may cause, in the lapse of ages, a slight lowering of the waters south of Cuba, and especially in the gulf of Mexico, the centre of the great current which runs along the sh.o.r.es of the United States, and casts the fruits of tropical plants on the coast of Norway.* (* "The Gulf-stream, between the Bahamas and Florida, is very little wider than Behring's Strait; and yet the water rushing through this pa.s.sage is of sufficient force and quant.i.ty to put the whole Northern Atlantic in motion, and to make its influence be felt in the distant strait of Gibraltar and on the more distant coast of Africa." Quarterly Review February 1818.) The configuration of the coast, the direction, the force and the duration of certain winds and currents, the changes which the barometric heights undergo through the variable predominance of those winds, are causes, the concurrence of which may alter, in a long s.p.a.ce of time, and in circ.u.mscribed limits of extent and height, the equilibrium of the seas.* (* I do not pretend to explain, by the same causes, the great phenomena of the coast of Sweden, where the sea has, on some points, the appearance of a very unequal lowering of from three to five feet in one hundred years. The great geologist, Leopold von Buch, has imparted new interest to these observations by examining whether it be not rather some parts of the continent of Scandinavia which insensibly heaves up. An a.n.a.logous supposition was entertained by the inhabitants of Dutch Guiana.) When the coast is so low that the level of the soil, at a league within the island, does not change to extent of a few inches, these swellings and diminution of the waters strike the imagination of the inhabitants.

The Cayo bonito (Pretty Rock), which we first visited, fully merits its name from the richness of its vegetation. Everything denotes that it has been long above the surface of the ocean; and the central part of the Cayo is not more depressed than the banks. On a layer of sand and land sh.e.l.ls, five to six inches thick, covered by a fragmentary madreporic rock, rises a forest of mangroves (Rhizophora). From their form and foliage they might at a distance be mistaken for laurel trees. The Avicennia, the Batis, some small Euphorbia and gra.s.ses, by the intertwining of their roots, fix the moving sands. But the characteristic distinction of the Flora of these coral islands is the magnificent Tournefortia gnaphalioides of Jacquin, with silvered leaves, which we found here for the first time. This is a social plant and is a shrub from four feet and a half to five feet high. Its flowers emit an agreeable perfume; and it is the ornament of Cayo Flamenco, Cayo Piedras and perhaps of the greater part of the low lands of the Jardinillos. While we were employed in herborizing,* our sailors were searching among the rocks for lobsters. (* We gathered Cenchrus myosuroides, Euphorbia buxifolia, Batis maritima, Iresine obtusifolia, Tournefortia gnaphalioides, Diomedea glabrata, Cakile cubensis, Dolichos miniatus, Parthenium hysterophorus, etc. The last-named plant, which we had previously found in the valley of Caracas and on the temperate table-lands of Mexico, between 470 and 900 toises high, covers the fields of the island