commandant of the fort of Agana is one of the officers of the King of Spain, who can with the greatest impunity exercise an arbitrary power. He has no communication with Europe, and the Philippine islands except once a year; and if the nao is intercepted or if it is lost in a tempest, he remains for several years completely insulated. Although the distance from Madrid to Agana is 4,000 leagues east in a straight line, it is said that the governor of Guahan on seeing the galleon arrive two years in succession expressed a desire to reside in an island more remote from Spain, that he might be less exposed to the controul of minis

ters.

The galleon carries to the colony of the Mariana islands (islas de los Ladrones), besides the situado that is to destined to pay

the say,

money

the troops and the royal officers, woollens, linens, cottons, and hats for the dress of the small number of whites who inhabit this archipelago. The governor supplies the galleon with fresh provisions, particularly with pork and beef. Horned cattle have multiplied in a wonderful manner in this island, where there is a beautiful breed of white oxen with black ears. Commodore Byron* affirms, having seen at the island of Saypan situated to the north of Tinian, which

* Hawkesworth's Compilation. Vol, i. p. 121.

bas mountains of small elevation, huanaco's like those of Peru. This observation deserves

to be verified by naturalists. The Spaniards having introduced neither llamas nor huanacos nor alpacos into Mexico or the kingdom of New Granada, it appears very improbable, that they should ever have transported them into a group of islands in the vicinity of Asia*.

Besides the galleon of Acapulco, from time to time a vessel is also dispatched from Manilla to Lima. This navigation, one of the longest and most difficult, is ordinarily carried on by the same northern route with the passage from the Philippine islands to the coast of California. The galleon destined for Lima, after discovering the coast of Mexico, steers southwards to the 28 and 30% of south latitude where the south west wind prevails. When Peru, liberated from the yoke of the monopoly of the Philippine company, shall be allowed to trade without restriction to the East Indies, in returning from Canton to Lima the preference will most likely be given to a track which goes to the south of New Holland through seas where they are secure of favourable winds.

A few years before my stay at Lima, Don Josef Arosbide brought the galleon el Fillippino' in ninety days, by a' direct tract from west to

** Voyage de Marchand. T. i. p. 436.

east from Manilla to the Callao. Favoured by light variable winds which blow especially by night in the vicinity of the South Sea islands, he ascended between the parallels of 6° and 10° south against the current of rotation. The dread of falling into the hands of English cruisers led him to make choice of a track so extraordinary and opposite to the direction of the trade winds. Forgetting that chance had a great share in the success of a voyage during which the calms were interrupted by squalls from the south and south west*, M. Arosbide wished to try the route a second time, but after long struggling against the trade winds, he was obliged to ascend to high latitudes, and to follow the old method of navigation. He was obliged to put into the port of San Blas for want of provisions where he died worn out with fatigue and disappointment.

It has been asked how it was possible for Spanish vessels since the sixteenth century to cross the great ocean from the western coast of the New Continent to the Philippine islands, without discovering the isles with which that

M. de Fleurieu a learned navigator, has very truly observed that it is not uncommon, in the equinoctial region of the great ocean, and especially in the 15° and 18° of south latitude and the 114° and 118° of west longitude, for south south west and even north west winds to prevail for several days (Voyage de Marchand) T. II, p. 269.

This problem may consider that few Lima to Manilla, of which we owe labours of Wallis,

vast sea basin is strewed. be easily resolved, if we navigations take place from and that the archipelagoes, the discovery to the Bougainville, and Cook, are almost all contained between the equator and the tropic of Capricorn. For these three hundred years the pilots of the Acapulco galleon have been prudent enough constantly to run the same parallel in their course from the coast of Mexico to the Philippine islands; and it appeared to them so much the more indispensable to follow this track as they imagined they should fall in with shallows and shoals whenever they deviated to the north or south. At a period when the use of lunar distances and time-keepers was unknown to navigators, they endeavoured to correct the longitude deduced from the reckoning by the observation of the variation of the mag netic needle. It had been very early remarked that the variation was nearly O at the strait of San Bernardino; and in 1585 Juan Iayme embarked with Francisco Gali from Manilla to Acapulco, to prove an instrument of his invention for finding the variation of the needle*.

* Viage al estrecho de Fuca, p. 46. Voyage de La Perouse, T. II. p. 306. I found in the month of December 1803, the magnetic variation at Mexico (lat. 19° 25' 4" north, west

This method of correcting the reckoning might be useful at a period when few pilots knew their longitude within nearly eight or ten degrees. It has been proved by very accurate observations in our days that the variation of the magnetic needle is extremely slow in these parallels, even in approaching the straits of San Bernardino.

Moreover we are not to be surprised that galleons laden with cargoes of the value of six or seven millions of francs, have never been tempted to abandon this track prescribed to them. Real expeditions of discovery can only be carried on at the expence of a government; and it cannot be denied that under the reigns of Charles V., Philip II., and Philip III., the viceroys of Mexico and Peru gave encouragement to a great number of undertakings calculated to give celebrity to the Spanish name. Cabrillo visited in 1542 the coast of New California or New Albion to the 37° of latitude. Gali, in going out of his track to the north in his return from China to the coast of Mexico, discovered in 1582 the mountains of New Cornwall, covered with eternal snows, and situated in the 57° 30, north. The expedition of Sebastian Viscayno discovered the

long. 101° 25') 8° 8' to the east; and in the South Sea, at 13° 50' of north latitude 106° 26' of longitude, 6° 54'.