Digitale Ausgabe

---

The kingdom of New Spain, the most north-erly point of Spanish America, extends from thesixteenth to the thirty-eighth degree of latitude.The length of this vast region from south-south-east to north-north-west is about 270 myria-meters (610 common leagues); its greatestbreadth is in the latitude of 30°. From the Redriver in the province of Texas (Rio Colorado) tothe island of Tiburon, on the coast of the pro-vince of La Sonora, is 364 leagues. That part of Mexico in which the twooceans, the Atlantic and the Pacific arenearest to each other, is unfortunately notthat in which the two ports of Acapulco andVera Cruz, and the capital of Mexico, aresituated. According to my astronomical obser- |380|vations, there is from Acapulco to Mexico anoblique distance of 2° 40′ 19″ of a great circle(or 155,885 toises); from Mexico to Vera Cruz2° 57′ 9″ (or 158,572 toises); and from the portof Acapulco to the port of Vera Cruz, in a directline, 4° 10′ 7″. It is in these distances that theold maps are the most faulty. According to theobservations published by Mr. Cassini, in theJournal of Chappe’s Voyage, the distance fromMexico to Vera Cruz is stated to be 5° 11′ oflongitude, instead of 2° 57′, which it has beenfound to be by more accurate observation.Assuming for Vera Cruz the longitude given by Chappe, and for Acapulco that of the map ofthe French Depôt of the Marine, compiled in1784, the breadth of the isthmus of Mexico,between the two ports, would be 175 leagues,which is 71 leagues too much. The isthmus of Tehuantepec, to the south-east of the port of Vera Cruz, is the part wherethe continent of New Spain is the narrowest;the distance being 45 leagues from the Atlantic to the Pacific ocean. The proximity of thesources of the rivers Huascualco and Chima-lapa seems to favor the project of a canal forinternal navigation, which was long meditatedby Count Revillagigedo, one of the most intel-ligent and active of the viceroys. When we |381|describe the province of Oaxaca, we shall re-turn to this subject so important to all civilizedEurope. We will here content ourselves withconsidering the problem of the communicationbetween the two rivers under the most generalpoint of view. We shall present nine pointswhich are not known in Europe, and all ofwhich afford more or less easy means of facili-tating navigation, either by canals, or by inter-nal communications between the rivers. At amoment when the new continent, profiting bythe misfortunes of Europe, and its perpetualdissentions, makes rapid progress in civiliza-tion; at a time when the commerce with Chinaand the north-west coast of America becomesfrom year to year more advantageous, the sub-ject, which we here treat in a summary manner,is highly interesting for the balance of com-merce and the political preponderance of na-tions. The nine points which I have marked inPlate IV. of my geographical and physicalatlas, have at different periods attracted theattention of enlightened merchants and states-men who have made a long stay in the colo-nies: they present very different advantages.We shall arrange them according to their geo-graphical position, commencing with the most |382|northern part of the New Continent, and fol-low the coasts to the south of the isle of Chiloe.It is not till after we have examined all theprojects here formed for the communication ofthe two oceans, that we shall be able to decidewhich deserves the preference. Previous to thisexamination, for which the correct materialsare not yet collected, it would be imprudentto dig canals in the isthmuses of Huasacualco,of Nicaragua, of Panama, or of Cupica. I. Under 54° 37′ north latitude, in the pa-rallel of Queen Charlotte’s Island, the sourcesof the river de la Paix or Ounigigah (Unjigah)approach within seven leagues the sources ofthe Tacoutché Tesse, which was supposed to bethe same as the river Columbia. The first ofthese rivers empties itself into the Polar Sea,after having mingled its waters with those ofSlave lake and Mackenzie river. The secondriver, the Columbia, falls into the Pacific oceannear Cape Disappointment, to the south ofNootka Sound, and, according to the celebrated Vancouver, in latitude 46° 19′. The Cordilleraof the Rocky Mountains, abounding in coals, wasfound by Mr. Fiedler to be 3520 English feetabove the level of the neighbouring plains.*

---

* If it is true that this chain of mountains entersthe limits of perpetual snow (Mackenzie, Vol. III.|383|It separates the sources of the rivers Peace andColumbia. According to the account of Mac-kenzie, who crossed this chain in August 1793,the portage is tolerably practicable, and themountains do not appear to be very elevated.To avoid the great detour which the Columbiamakes, a shorter road for commerce might beopened from the sources of the TacoutchéTesse as far as Salmon river, the mouth ofwhich is to the east of Princess Royal Islands,in latitude 52° 26′. Mr. Mackenzie justly ob-serves that a government which should openthis communication between the two oceans,forming regular establishments in the interiorof the country, and at the extremities of therivers, would thereby become master of thewhole fur trade of North-America, from the48th degree of latitude to the pole, except thatpart of the coast which has long since beenunder the dominion of Russia. Canada, by thenumber and course of its rivers, affords facilities

---

p. 331) their absolute height must be at least from 1000to 1100 toises; from whence it would result that theneighbouring plains, on which Mr. Fiedler was placedto take his measure, are from 450 to 550 toises abovethe level of the sea, or that the summits of which thistraveller has marked the height are not the most ele-vated of the chain crossed by Mackenzie.|384|for inland trade similar to those which exist inEastern Siberia. The mouth of the river Co-lumbia seems to invite Europeans to establisha fine colony. The banks of this river are fer-tile, and covered with fine timber. It must beconfessed, however, notwithstanding the exami-nation made by Mr. Broughton, only a verysmall part of the Columbia is yet known, which,like the Severn and the Thames, appears todecrease extremely in breadth in proportion tothe distance from the coast. Any geographer whowill carefully compare the maps of Mackenziewith those of Vancouver will be surprised that theColumbia, descending from the Rocky Mountainswhich we are inclined to consider as a prolon-gation of the Andes of Mexico, can traversethe chain of mountains which approaches thegreat ocean, and the principal summits ofwhich are mount St. Helen and mount Rainier.Mr. Malte Brun had already alleged importantdoubts against the identity of the TacoutchéTessé and the Rio Columbia, before it was dis-covered, as it now is, that the Columbia or Ore-gon is entirely different from the TacoutchéTessé or Fraser’s river. In latitude 50°, Nelson’s river, the Saskasha-wan and the Missouri, which may be regarded asone of the principal branches of the Missisippi, |385|furnish equal facilities of communication withthe Pacific ocean. All these rivers rise at the footof the Rocky Mountains. We have not yet anysufficiently accurate data respecting the natureof the ground where the portage must be fixed,to decide on the utility of this communication.The expedition which Captain Lewis per-formed at the expence of the Anglo-Americangovernment on the Mississippi and Missouri,may throw great light on this interestingproblem. II. In latitude 40°, the sources of the Rio delNorte or Rio Bravo, which falls into the Gulphof Mexico, are separated from the sources ofthe Rio Colorado by a mountainous district oftwelve or thirteen leagues in breadth. Thistract is the continuation of the Cordillère desGrues, which extends towards Sierra Verde andthe Lake of Timpanogos, celebrated in the his-tory of Mexico. The Rio St. Raphael and theRio St. Xavier are the principal sources of theriver Zaguananas, which, with the Rio Nabajoa,forms the Rio Colorada, and mingles its waterswith those of the Gulph of California. Thecountries watered by these rivers abound inrock salt: they were examined in 1777 byFathers Escalante and Antonia Velez, twomonks of the order of Saint Francis. How- |386|ever interesting the Rio Zaguananas and theRio del Norte may one day become to the in-land commerce of this northern part of NewSpain, and however easy the portage acrossthe mountains may be, no communication willever result from it affording advantages equalto those of a canal between the two oceans. III. The isthmus of Tehuantepec, under thesixteenth degree of latitude, comprehends thesources of the Rio Huasacualco or Goazacoal-cos, which falls into the Gulph of Mexico, andthe sources of the Rio Chimalapa. The watersof this latter river mingle with those of thePacific ocean near the barra de San Francisco.I consider here the Rio del Pasco as the prin-cipal source of the river Huasacualco, thoughthe latter does not take its name till it reachesthe Passo de la Fabrica, after one of its arms,which comes from the mountains de losMixes, has joined the Rio del Passo. Thisisthmus of Tehuantepec is the point which Fer-dinand Cortes, in his letters to the Emperor Charles V. calls the secret of the strait, an ap-pellation which sufficiently proves the impor-tance attached to it at the commencement ofthe sixteenth century. It has again attractedthe attention of navigators since the hostilitiescarried on by the castle of San Juan d’Uloa |387|have caused the commerce of Vera Cruz toturn to the Barra d’Alvarado and to the coastof Tabasco, near the mouth of the Rio Huasa-cualco. The ridge, which forms the divisionof the water between the two oceans, is inter-rupted by a valley; but I much doubt whetherin the time of the great inundations this valleyis filled (as has been lately stated) with a quan-tity of water sufficient to allow a natural pas-sage for the boats of the natives. Similar tem-porary communications exist between the ba-sins of the Missisippi and the river Saint Law-rence, that is to say between Lake Erié and theWabash, between Lake Michigan and the riverof the Illinois. We shall return in the sequelto the possibility of digging a canal, six or sevenleagues long, in the forests of Tarifa.* Since aroad was opened in 1798 from the port of Tehu-antepec to the Embarcadero de la Cruz (whichroad was improved in 1800), the river Huasa-cualco forms a commercial communication be-tween the two oceans. During the war withthe English, the indigo of Guatimala, the most

---

* The Spanish Cortes decreed the opening of thiscanal in 1814. The execution of the canal was con-fided to the Consulado de Guadalaxara, who proposedto issue an invitation to the capitalists of Europe.|388|valuable known, came by this isthmus to the portof Vera Cruz, and thence to Europe. IV. The great Lake of Nicaragua communi-cates not only with the Lake of Leon, butalso on the east, by the river San Juan, withthe sea of the Antilles. The communicationwith the Pacific ocean would be effected bydigging a canal across the isthmus which sepa-rates the lake from the gulph of Papagayo.On this narrow isthmus are the volcanic andinsulated summits of Bombacho (in latitude 11°7′) of Grenada and Papagayo (in latitude 10°50′). Old maps even indicate a communicationby water across the isthmus. Other maps,rather more recent, represent a river, under thename of the Rio Partido, one of the branchesof which flows into the Pacific and the otherinto the lake Nicaragua. But this bifurcationseems very uncertain. It is not noticed in thelatest maps published by the Spaniards andthe English. In the archives of Madrid are several Frenchand English Memoirs on the possibility ofuniting lake Nicaragua with the Pacific ocean.The commerce, which the English carry on, onthe Mosquito shore, has greatly contributed togive celebrity to this project of making a com-munication between the two seas. None of |389|these memoirs which have come to my know-ledge clear up the principal point, which isthe elevation of the ground of the isthmus. From the kingdom of New Grenada to theenvirons of the capital of Mexico, there is nota single mountain, plateau, or town, the ele-vation of which above the surface of the seais known to us. Is there an uninterruptedchain of mountains in the provinces of Vera-gua and Nicaragua? Has this range, which issupposed to connect the Andes of Peru withthe mountains of Mexico, its central chain tothe east or the west of the Lake Nicaragua?Does the isthmus of Papagayo offer a moun-tainous soil, or only a simple barrier? Theseare problems, the solution of which is as inte-resting to the statesman as to the geographer.The various works which have been publishedsince the commencement of the wars for theindependence of Spanish America confine them-selves to the same ideas developed in the firstedition of this work; I except some useful in-formation which Mr. Davis Robinson has givenon the bar of the river of San Juan de Nicara-gua. He assures us that “this bar has twelvefeet of water, and that on one point only it has anarrow pass, twenty-five feet in depth.” Inthe Rio de San Juan itself there are from four to |390|six fathoms; in the lake Nicaragua from three toeight fathoms. According to Mr. Robinson theSan Juan is navigable for brigs and schooners. There is not a spot upon the globe so full ofvolcanoes as this part of America, from latitude11° to 13°; but it seems that the trachytic moun-tains, through which the subterraneous firemakes its way, forms only insulated groups,and that, separated from each other by vallies,they rise from the plain itself. It must notexcite surprise that we were ignorant of facts ofthis importance, for we shall soon see that eventhe height of the chain which traverses theisthmus of Panama is as little known now as itwas before the invention of the barometer, andthe application of that instrument in the mea-surement of mountains. Perhaps too a com-munication between lake Nicaragua and thePacific ocean might be made by lake Leon,by means of the river Tosta, which descendsfrom the volcano of Telica, on the road fromLeon to Realexo. In fact the ground does notseem very high, and Dampier’s account of hisvoyage may infer that there is not a real chainof mountains between Lake Nicaragua and theSouth Sea. “The coast of Nicoya,” says thisgreat navigator, “is low, and overflowed at thetime of high water. Between Realexo and |391|Leon you traverse a flat country, covered withMango trees.” The city of Leon itself is situ-ated in a savannah. There is a small river,which falling into the sea near Realexo,might facilitate the communication betweenthat port and Leon. From the western bank ofthe lake Nicaragua, it is only four leagues tothe bottom of the gulph of Papagayo, and sevento that of Nicoya, which navigators call Caldera. Dampier expressly says that the ground betweenLa Caldera and the lake is not very hilly, andfor the most part a level plain. The isthmus of Nicaragua, by the position ofits inland lake, and the communication betweenthis lake and the sea of the Antilles by means ofthe Rio San Juan, has many features of resem-blance with the defile in the Highlands of Scot-land, where the river Ness forms a natural com-munication between the mountain lakes andthe gulph of Murray. At Nicaragua, as in theHighlands, there is to the west merely a barrierto pass; to the east it would perhaps be suffi-cient to canalize the Rio San Juan, withoutdeviating from the bed of the river, which hasno bars except in the dry season. If it is truethat the isthmus to be crossed has a few hillswhere it is the narrowest, between the westernbank of the Nicaragua, and the gulph of Papa- |392|gayo, it is on the other hand formed of uninter-rupted savannahs and plains, affording an ex-cellent road for carriages, between the city ofLeon and the coast of Realexo. It is the highroad by which merchandise is sent from Guati-mala to Leon, landing in the gulph of Fonsecaor Amalapa, to the port of Conchagua. Theelevation of Lake Nicaragua above the SouthSea is equal to the fall of the Rio San Juan inthe course of thirty leagues: accordingly theelevation of this basin is so well known in thecountry that it was formerly regarded as an in-surmountable obstacle to the execution of a canal.It was apprehended that it might cause eitheran impetuous overflow towards the west, or adiminution of the water of the San Juan, which,during the dry season, has, above the antientcastillo de San Carlos, several rapids, and thebanks of which, during their present uncultivatedstate, are extremely unwholesome. The art ofthe civil engineer is, however, sufficiently ad-vanced in our days not to fear such dangers.The lake of Nicaragua may serve as the upperbasin, as lake Oich on the Caledonian canal.Regulating sluices will admit into the canalonly sufficient water to feed it. The small differ-ence of level supposed to exist between the seaof the Antilles and the Pacific ocean, probably |393|arises only from the unequal height of the tides.A similar difference is observed between the twoseas which are united by the great Caledoniancanal, and were it even six toises and perma-nent, as that between the Mediterranean andthe Red Sea, it would be no less favourable toa junction between the oceans. The windsblow strong enough on lake Nicaragua to savethe necessity of towing by steam-boats the ves-sels which are to pass from one sea to the other;but the employment of steam would be veryuseful in voyages from Realexo or from Panamato Guayaquil. During the months of August,September, and October, calms are alternate inthese seas, with a wind blowing in a contrarydirection to this voyage. The coasts of Nicaragua are rather dangerousin the months of August, September, and Octo-ber, on account of the winds and terrible rain;in January and February, on account of theviolent north-east and east north-east winds,which are called by the name of Papagayos.This circumstance renders navigation very in-convenient. The port of Tehuantepec, in theisthmus of Huasacualco, is not more favouredby nature; it gives its name to the hurricaneswhich blow from the north-west, and whichmake all vessels fly from the little ports of |394|Sabinas and Ventosa. It results from the aboveconsiderations that the possibility of the canalof Nicaragua is threefold, namely, from thatlake to the gulph Papagayo, from the same tothe gulph of Nicoya, or from the lake of Leonor Managua to the mouth of the Rio de Tosta.The distance from the south-eastern extremityof the lake of Nicaragua to the gulph of Nicoyais very differently laid down (from 25 to 48miles) in Arrowsmith’s map of south America,and the fine map in the Deposito Hidrografico atMadrid, which is called Mar de las Antillas, 1809. V. The isthmus of Panama was crossed forthe first time by Vasco Numez de Balbo, in theyear 1513. Since this memorable epoch in thehistory of geographical discoveries, the projectof a canal has been a very general subject ofconsideration; yet even now, after the lapse ofthree centuries, there is no survey of the ground,nor any very correct determination of the exactposition of Panama and Portobello. The longi-tude of the first of these two ports has beenreferred from Carthagena; that of the secondhas been determined from Guayaquil. Theoperations of Fidalgo and Malaspina are doubt-less entitled to great confidence: but errors mul-tiply insensibly when by chronometrical opera-tions which embrace the whole coast of the |395|Terra Firma, from the isle of Trinidad to Por-tobello, and from Lima to Panama, one po-sition is made dependent upon another. Toform an idea of the uncertainty which stillprevails respecting the shape and breadth ofthe isthmus (for example at Nata) we needonly compare the maps of Lopez with thoseof Arrowsmith and the most recent ones of theDeposito Hidrografico at Madrid. The riverChagre, which falls into the sea of the An-tilles to the west of Portobello, notwithstandingits sinuosities and rapids, presents many advan-tages to commerce; it is 120 toises broad at itsmouth, and 20 near Cruces, where it begins tobe navigable. At present, ships ascend the RioChagre from its mouth to Cruces, in four or fivedays. If the water is very high, they must con-tend against the current for ten or twelve days.From Cruces to Panama the goods are conveyedon mules for five short leagues. The barometri-cal measurements given in Ulloa’s travels in-duce me to suppose that there is in the RioChagre a difference of level of thirty-five to fortytoises between the sea of the Antilles and theEmbarcadero or Venta de Cruces. This differ-ence must appear very small to those who haveascended the Rio Chagre; they forget that thestrength of the current depends both on the great |396|accumulation of water near the sources, and onthe general fall of the river, that is to say aboveCruces. On comparing the barometrical levelof Ulloa with that which I made in the riverMagdalena, we perceive that the elevation ofCruces above the ocean, far from being small, ison the contrary very great. The fall of the RioMagdalena from Honda to the dike of Mahatesnear Barancas is 160 toises, and yet this dis-tance is not, as might be expected, four times,but eight times greater than from Cruces toFort Chagre. The engineers who have proposed at the courtof Madrid to establish a communication betweenthe two oceans by the Rio Chagre have pro-jected the digging of a canal from the Venta deCruces to Panama. This canal would have topass through a mountainous tract, with the ele-vation of which we are entirely ignorant. Weonly know that from Cruces there is first a rapidascent and then a descent for several hours to-wards the coasts of the South Sea. It is verysurprising that neither La Condamine andBouguer, nor Don George Juan and Ulloa, hadthe curiosity, when they crossed the isthmus,to look at their barometers, to inform us what isthe elevation of the highest point on the roadfrom Fort Chagre to Panama. These gentle- |397|men remained three months in this region whichis so interesting to the commercial world; buttheir long stay has hardly added any thing tothe observations of Dampier and Wafer. Itseems indubitable that the principal chain, orrather range of hills, which may be consideredas a prolongation of the Andes of New Grenada,is, between Cruces and Panama, nearer to theSouth Sea than to the sea of the Antilles. It isfrom the summit of this range that persons havepretended to see the two oceans at once, anobservation which would not imply an absoluteelevation of more than 290 metres. LionelWafer complains that he could not enjoy thisspectacle; and he assures us that the hills, whichform the central chain, are separated from eachother by vallies, which leave a free course for therivers. If the last assertion be correct, we maybelieve in the possibility of a canal from Crucesto Panama, the navigation of which would beinterrupted by only very few sluices. From some slight indications of the tem-perature of these places, and the geographyof the indigenous plants, I should be inclinedto believe that the barrier in the road fromPanama to Cruces does not attain an elevationof 500 feet. Mr. Robinson supposes it to be400 feet at the most; besides, we find in almost |398|all mountainous countries, when carefully ex-amined, instances of natural openings acrossthe barriers. The hills between the basins ofthe Saone and the Loire, which the canal of theCentre would have had to pass, were 800 to 900feet high, but a defile or interruption at thepond of Long Pendu presented a ledge which is350 feet lower. There exist other points in which, accordingto memoirs drawn up in 1528, it has been pro-posed to cut through the isthmus, namely, byjoining the sources of the rivers called Caimitoand Rio Grande, with the Rio Trinidad. Theeastern part of the isthmus is narrower, but theground seems to be much more elevated. Atleast this is what we observe in the horribleroad taken by the mail from Portobello to Pa-nama. This road is two days’ journey, passesby the village of Pequeni, and presents verygreat difficulties. In all ages and in all climates, people havebelieved that of two neighbouring seas, one washigher than the other. Traces of this vulgaropinion are found in the antient writers. Strabo says that it was supposed that the level of thegulph of Corinth near Lechœa was above thatof the waters of the gulph of Cenchræa. Heimagined that it would be very dangerous to |399|cut through the isthmus, where the Corinthians,by the aid of particular machines, had esta-blished a portage. In the isthmus of Panamait is commonly supposed that the South Sea ismore elevated than the Sea of the Antilles. Thisopinion is founded on a bare appearance. Afterhaving contended several-days against the cur-rent of the Rio Chagre, we think we have as-cended, much more than we afterwards descend,the hills from Cruces to Panama. In fact, no-thing is more deceitful than the opinion whichwe form, of a difference of level upon a longslope, which consequently is very gentle. InPeru I could hardly believe my eyes when Ifound by my barometer that the city of Lima is91 toises above the port of Callao. The rock ofthe Isle of San Lorenzo must be entirely coveredwith water, in consequence of an earthquake,before the ocean could reach the capital of Peru.Don George Juan has already combatted theopinion of a difference of level between the twoseas; he found the height of the mercury to bethe same at the mouth of the Chagre and at Pa-nama. The imperfection of the meteorological instru-ments used at that time, may leave some doubtswhich even appear to have acquired more weightsince the French engineers attached to the ex- |400|pedition in Egypt found the level of the RedSea to be six toises above the Mediterranean.Till a geometrical survey shall have been madeof the level of the isthmus of Panama, we musthave recourse to barometrical admeasurements.Those which I made at the mouth of the RioSinu in the Sea of the Antilles and on the Pe-ruvian coasts of the South Sea, prove, afterevery correction has been made for the tem-perature, that if there exists a difference oflevel between the two oceans, it cannot be abovesix or seven metres. The navigation of the Rio Chagre is difficult,not so much on account of the number of itssinuosities as of the celerity of its current, whichis often one or two metres per second. Thesinuosities however afford the advantage of acounter-current, which is formed by eddies to-wards the banks, and by means of which, smallvessels, called Bongos and Chatas, ascend,either by means of oars and poles, or by tow-ing. If these sinuosities were cut through, theadvantage would cease, and it would be verydifficult to go from the Sea of the Antilles toCruces. The minimum of the breadth of the isthmusof Panama is not fifteen miles, as marked bythe first maps of the Deposito Hidrografico at |401|Madrid, but 25¼ miles, that is to say, 8½ sealeagues, or 24,580 toises; for the dimensions ofthe gulph of San Blas, called also Ensenada deMandinga, on account of the small river whichfalls into it, have given rise to serious errors.This gulph enters seventeen miles less into thecontinent than was supposed in 1805, in thesurvey of the Mulatto Islands. Whatever confi-dence the last astronomical observations seem tomerit, on which the map of the isthmus, publishedby the Deposito Hidrografico in 1817 is founded,it must not be forgotten that these operationscomprehend only the Northern Coast, and thatthey have not yet been connected either by a seriesof triangles or chronometrically with the south-ern coasts. Now the problem of the breadth ofthe isthmus does not depend on the determina-tion of the latitudes alone. From the whole of the information which Iwas able to procure during my stay at Cartha-gena and Guayaquil; it seems that we mustgive up all hopes of a canal seven metres indepth, and from 22 to 28 in breadth, whichshould traverse the isthmus of Panama fromsea to sea, and receive the same vessels thattrade between Europe and the East-Indies.The elevation of the ground would oblige theengineer to have recourse to subterraneous gal- |402|leries, or to sluices; consequently, the mer-chandize intended to pass the isthmus of Pa-nama, must be conveyed in flat-bottomed boats,incapable of keeping the sea. Depôts would benecessary at Panama and Portobello. All na-tions which should desire to carry on trade bythis way, would become dependent on the nationwho should be mistress of the isthmus and thecanal. This inconvenience would be very great,particularly for vessels dispatched from Europe.Even in case the canal should be dug, it is pro-bable that the greater number of ships, fearingthe delays caused by the numerous sluices,would still prefer the route by the Cape of GoodHope. We see that the passage of the Soundis much frequented, notwithstanding the exis-tence of the canal of the Eyder, which unites theocean with the Baltic. It would not be the same with the productionsof Western America, or the merchandize whichEurope sends to the coasts of Quito and Peru orthe Pacific Ocean: these would cross the isth-mus with less expense; and in time of war es-pecially, with less danger, than by doubling thesouthern extremity of the new continent. Inthe present state of the roads, the conveyanceof three hundred weight on mules from Panamato Portobello, costs three or four piastres. But |403|the rude state in which the government has leftthe isthmus is such, that the number of beasts ofburden from Panama to Cruces is much toosmall for the copper of Chili, the bark of Peru,and above all, the 70,000 fanegas of cocoa,which are annually exported from Guayaquil,to cross this slip of land; consequently, theslow, dangerous, and expensive navigation roundCape Horn is preferred. In 1802 and 1803, when the English privateerseverywhere hindered the commerce of Spain, agreat part of the cocoa of Guayaquil was sentacross the kingdom of New Spain, and shippedat Vera Cruz for Cadiz. The voyage fromGuayaquil to Acapulco, and a land journey of ahundred and thirty-five leagues from Acapulco to Vera Cruz, was preferred to the danger of along passage by Cape Horn, and the difficulty ofstruggling against the currents along the coastsof Peru and Chili. This example proves, that ifthe construction of a canal, either across theisthmus of Panama, or that of Huasacualco,should be accompanied with too many difficul-ties, on account of the multiplicity of the sluices,the commerce of Western America would gainprodigiously by good roads from Tehuantepecto the Embarcadero de la Cruz, and from Pa-nama to Portobello. It is true that in the |404|Isthmus the pasturages are at present not favour-able to the support and increase of cattle, butin so fertile a country, it would be easy to makesavannahs by cutting down the forests, or tocultivate the Paspalum purpureum, the Miliumnigricans, and particularly the Luzerne (Me-dicago sativa), which grows abundantly in Peruin the hottest parts. The introduction of camelswould be a means still more calculated to di-minish the expenses of conveyance. These landships, as the Orientals call them, are at presentto be found only in the province of Caracas,where the Marquis of Toro introduced themfrom the Canaries. No political consideration ought to opposethe progress of population, agriculture, com-merce, and civilization in the isthmus of Pa-nama. The more cultivated this tongue ofland becomes, the more resistance it will opposeto a foreign enemy. If some enterprising nationwished to make itself master of the isthmus, itcould more easily do so in its present state.There are numerous fine fortifications destituteof arms to defend them. The insalubrity ofthe climate, though already diminished at Porto-bello, renders a military enterprize in the Isth-mus difficult. It is from Saint Charles de Chiloe,and not from Panama, that Peru can be attacked. |405|It would take three or four months to advanceagainst the currents from Panama to Lima,whereas the navigation from Chili to Peru iseasy and always rapid. Notwithstanding thedisadvantages presented by the Isthmus, thepossession of it is still of great importance to anenterprising nation. The whale fishery, whichso far back as 1803, employed sixty Englishvessels in the South Sea, the facility of the tradewith China, and the furs of Nootka Sound, arevery seducing temptations; they suffice, sooneror later, to attract the masters of the ocean to-wards a point of the globe, which nature seemsto have destined to effect a change in the com-mercial system of nations. VI. To the south-east of Panama, followingthe coasts of the Pacific, from Cape St. Michaelto Cape Corrientes, we come to the small portof Cupica. The name of this bay has becomecelebrated in the kingdom of New Grenada, onaccount of a new plan for a communication be-tween the two seas. From Cupica, we cross, fora distance of five or six sea leagues, a leveltract* very well adapted for a canal, which would

---

* This information was communicated to me in1803, by an inhabitant of Carthagena: but the geo-graphical position of Cupica is as uncertain as that of|406|end at the Embarcadero of the Rio Naipi, orNaipipi. This latter river is navigable, and fallsbelow the village of Zitara, into the Rio Atrato,which empties itself into the sea of the Antilles.Mr. Gogueneche, a very intelligent Biscayanpilot, has the merit of having first drawn the at-tention of the government to this bay of Cupica;he attempted to prove that it might become tothe new world what Suez had antiently been to Asia. Mr. Gogueneche proposed to send all thecocoa of Guayaquil by the Naipi to Carthagena.The same channel offers the advantage of a veryspeedy communication between Cadiz andLima. Instead of sending the mails by Cartha-gena, Santa-Fé, or Quito, or by Buenos Ayres,and Mendoça, the dispatches should be sent bythe mouths of the Atrato to Cupica, and for-warded by little swift sailing packet-boats fromCupica to Peru. If this route had been opened,the viceroy of Lima would not sometimes haveremained six months without receiving orders *

---

* the junction of the Naipi with the Atrato. I cannot findon any Spanish map the port of Cupica; but PuertoQuemado, or Tupica, in 7° 15′ latitude. It would bevery important to know whether schooners can ascendfrom the mouth of the Atrato to the junction of theNaipi. It is to be hoped that all these points will besoon cleared up by observations made on the spot.|407|from his court. Besides, the environs of the bayof Cupica furnish fine timber, which might besent to Lima. The tract lying between Cupicaand the mouth of the Atrato, is perhaps theonly place in the whole of America in whichthe chain of the Andes is wholly interrupted.To form an idea of this extraordinary depres-sion of the western Cordillera of New Grenada,we must recollect that in the second degree oflatitude, in the assemblage of mountains whichcontains the sources of the Rio Magdalena, the Andes divide into three chains. The most east-erly extends, deviating towards the north-east,by Timana, Bogota, and Pemplona, to the snowymountains of Merida: between lake Maracayboand the city of Valencia, it joins the Cordillera,of the coast of Venezuela. The intermediatechain, that of Panama, Guanacas, and Quindiù,separates the longitudinal valley of the RioCauca from that of the Rio Magdalena. In theprovince of Antioquia, it joins the most westerlychain of New Grenada, which gradually disap-pears in the district of Choco, in latitude 7°, alittle to the west of Zitara, between the left bankof the Atrato, and the shores of the Pacific. Itwould be interesting to know the configurationof the ground between Cape Garachine, or theGulph of Saint Michael, and Cape Tiburon, es-pecially towards the sources of the Rio Tuyra |408|and Chucunaque, or Chuchunque, that wemight determine with precision where the moun-tains of the isthmus of Panama commence, theline of whose summits appears to be not abovea hundred toises in height. The interior of Dar-four is not more unknown to geographers thanthe damp, unwholesome, and woody tract whichextends to the north-west of Betoi, and ofthe junction of the Bevara with the Atrato, to-wards the isthmus of Panama. All that wepositively know at present is that betweenCupica and the left bank of the Atrato, there iseither a land strait (detroite terrestre), or a totalabsence of any chain. The mountains of theisthmus of Panama, on account of their direc-tion and geographical position, may be consi-dered as a continuation of the mountains of the Antioquia and Choco; but in the plains to thewest of the lower Atrato there is scarcely aledge, or a slight barrier. Between the isthmusand the Cordillera of Antioquia, there is not agroup of the mountains, interposed like that whichindubitably connects (between Barquesimeto,Nirgua, and Valencia), the eastern chain ofNew Grenada, (La Sierra de la Suma Paz,and the Sierra Nevada de Merida), to the Cor-dillera of the coast of Venezuela. VII. In the interior of the province of Choco,the little ravine (Quebrada) of the Raspadura |409|unites the Rio de Noanama, vulgarly called theRio San Juan, to the small river of Quibdo.This latter, augmented by the waters of theAndagueda and the Rio Zitara, forms the RioAtrato, which falls into the sea of the Antilles,while the Rio San Juan empties itself into thePacific. A very active monk, priest of the vil-lage of Novita, caused his parishioners to dig alittle canal in the ravine of the Raspadura. Bymeans of this canal, which is navigable duringthe rainy season, boats loaded with cocoa havepassed from one sea to the other. Here then isan inland communication which has existedsince 1788, and is unknown in Europe. Thesmall canal of the Raspadura connects thecoasts of the two oceans in two points aboveninety-five leagues distant from each other. Itwill never be any thing more than a canal forboats; but it might be easily enlarged by joiningto it the streams known by the name of Caño delas Animas, del Caliche, and Aguaclaras. Re-servoirs and tributary channels are easily madein a country like Choco, where it rains all theyear, and thunders every day. According tothe information I obtained at Honda and Vilela,near Cali, from persons employed in the tradeof the gold dust from Choco, the Rio Quibdo,which communicates with the canal of Mina de |410|Raspadura, unites, near the village of Quibdo,(vulgarly called Zitara) with the Rio Zitara andthe Rio Andagueda; but, according to a manus-cript map which I have just received fromChoco, on which the canal of the Raspaduralikewise joins (in latitude 5° 20′) the Rio SanJuan and the Rio Quibdo, a little above themine of Animas, the village of Quibdo is placedat the confluence of the small river of this namewith the Rio Atrato which has received the RioAndagueda, three leagues higher up near Lloro.From its mouth (lat. 4° 6′) to the south of thepoint of the Charambira, the great Rio SanJuan receives successively, in ascending towardsthe N. N. E. the Rio Calima, the Rio del No,(above the village of Noanama) the Rio Ta-mana, which flows near Novita, the Rio Iro, theQuebrada de San Pablo, and lastly, near thevillage of Tado, the Rio de la Platina. Theprovince of Choco is inhabited only on thebanks of these rivers; it has commercial com-munications towards the north with Carthagena,by the Atrato, the banks of which are entirelydeserted from latitude 60° 45′; towards the southwith Guayaquil, and (before 1786) with Val-parairo, by the Rio San Juan; to the east withthe province of Popayan, by the Tambo deCalima and by Cali. The ravine of the Raspa- |411|dura, which serves as a canal, and which I believeI was the first to make known in Europe, isoften confounded on maps with the portage ofCalima and of San Pablo. The Arastradero deSan Pablo also leads to the Rio Quibdo, butseveral leagues above the mouth of the Raspa-dura. It is by the road from this Arastradero ofSan Pablo that goods are generally sent fromPopayan by way of Cali, Tambo de Calima,and Novita to Choco del Norte, that is to say toQuibdo. It cannot be doubted that on anypoint of equinoctial America, whether in theisthmus of Choco, or those of Panama, Nicara-gua, and Huasacualco, the union of two neigh-bouring ports by a canal, from four to six feetdeep, (Canal en petite section) or by a riverconverted into a canal, would give rise to a veryactive commerce. This canal would act as arailway, and, however small, would animate andabridge the communications between the west-ern American coasts and the United States, andEurope: but however advisable enterprizes ofthis kind may be, they never can have thatpowerful influence on the commerce of the twohemispheres that a real oceanic canal wouldhave. VIII. In ten degrees south latitude, two or |412|three days’ journey from Lima, you come to thebanks of the river Guallaga (or Huallaga) bywhich, without doubling Cape Horn, you may goto the coasts of the Grand Para in Brazil. Thesources of the Rio Huanuco, which falls intothe Guallaga, are distant, near Chinche, four orfive leagues from the sources of the Huaura,which empties itself into the Pacific. Even theRio Xauxa, which falls into the Apurimac orUcayale, rises near Jauli, a short distance fromthe sources of the Rio Rimac, which traversesthe city of Lima. The height of the Cordilleraof Peru, and the nature of the ground, render theexecution of a canal impossible; but the con-struction of a convenient road from the capitalof Peru to the Rio Huanuco, would facilitate theconveyance of merchandize to Europe. The greatrivers Ucayale and Guallaga, in five or sixweeks, would bring the productions of Peru to themouth of the Amazons, and to the coasts near-est to Europe, whereas a voyage of four monthsis required to carry the same goods to the samepoint, if they double Cape Horn. The cultiva-tion of the beautiful regions on the eastern slopeof the Andes, and the prosperity and richesof their inhabitants depend on a free naviga-tion of the river of the Amazons. This liberty,which the court of Portugal denied to the |413|Spaniards, might have been acquired in conse-quence of the events preceding the peace of 1801. IX. Before the coast of Patagonia was suffi-ciently explored, it was supposed that the gulphof Saint George, situated between 45° and 47° ofsouth latitude, penetrated so far into the land asto communicate with the arms of the sea, whichinterrupt the continuity of the western coasts,that is to say the coast opposite to the Archipe-lago of Chayamapu. If this supposition werefounded on a solid basis, vessels bound forthe South Sea might cross South America, 175leagues to the north of the straits of Magellan,and shorten their route above 700 leagues. Na-vigators would by this means avoid the dangerswhich, notwithstanding the improvements in thescience of navigation, still attend the voyageround the Cape of Good Hope, and along thewestern coasts of Patagonia, from Cape Pilaresto the parallel of the Chonos islands. In 1790these ideas had attracted the attention of thecourt of Madrid. Gil Lemos, viceroy of Peru,an upright and zealous governor, sent a smallexpedition under Don Jose de Moraleda, to exa-mine the southern coast of Chili. I have seen,in the instructions which he received at Lima,that he was enjoined to preserve the strictestsecrecy, if he should be so fortunate as to discover |414|a communication between the two seas. DonMoraleda found in 1793 that the Estero ofAysen, which had been visited in 1763 by theJesuits, Fathers Jose Garcia and Juan Vicuña,is of all the arms of the sea, that, by which thePacific Ocean stretches the farthest to the east.This Estero however is not more than eightleagues long, and terminates abruptly near theisle of la Cruz, where it receives, near a hotspring, a river of small depth. This Estero deAysen, situated in 45° 28′ of Latitude, is there-fore eighty-eight leagues distant from the gulphof Saint George. This last gulph was accuratelysurveyed by the expedition of Malaspina. In1746 another communication had been suspectedin Europe between the bay of St. Julien (latitude50° 53′) and the Pacific Ocean. I have drawn upon one plate the nine pointswhich seem to offer means of communicationbetween the two seas, by uniting neighbouringrivers either by canals or by roads which couldfacilitate the conveyance of goods to places wherethe rivers become navigable. It is for the go-vernment, which possesses the most beautifuland fertile part of the globe, to perfect what Ihave only been able to hint at in this essay.Two Spanish engineers, Messrs. Le Maur havelaid down with much care the plan of the canal |415|of Los Guines, which was intended to crossthe whole isle of Cuba, from the Batabano tothe Havannah. A similar survey made in theisthmus of Guasacualco, at lake Nicaragua, be-tween Cruces and Panama, and between Cupicaand the Rio Naipi* would direct the statesmanin his choice: he would learn whether it is inMexico, in Nicaragua, or in Darien, that thisgrand enterprise should be executed, whichwould immortalize a government which shouldturn its attention to the true interests of hu-manity. The long voyage round South America would then become less frequent; a road wouldbe opened, if not for ships, at least for merchan-dize, which might go from the Atlantic to thePacific Ocean. When a canal shall unite the two oceans, theproductions of Nootka Sound and China, will bebrought above two thousand leagues nearer to

---

* The information which Major Alvarez lately com-municated to Captain Cochrane, is not favourable tothe utility of a canal between the Rio Naixo or Naipi,(which flows into the Atrato) and the bay of Cupica orTupica. This traveller assures us that the Naipipi isfull of bars, and that the isthmus between the river andthe coasts of the Pacific, is traversed by three ranges ofhills. (Vide, Captain Cochrane’s travels in Columbia.)|416|Europe and the United States. Then, and notbefore, great changes will be effected in the poli-tical state of Eastern Asia; for this tongue ofland, against which the waves of the Atlantic break, has been for ages the bulwark of theindependence of China and Japan.