It is obvious, on looking at a map of this reign, and comparing it with the statement of Herrera, that the river spoken of might be either the Uruguay, which, on the right, takes a northern direction, or one of the various streams into which the Parana is broken by the islands at its mouth. Cabot would hardly follow the Uruguay, because it evidently struck into Brasil, and, at a much higher point of ascent, he is found avoiding, expressly for that reason, a great river on the right hand. In speaking of the position occupied by his ships he states it, according to Herrera, to be on the Brasil, meaning the northern side of the river, a mode of designation, which, supposing him, as we reasonably may, to have been aware of the general course of the great stream discovered by De Solis, would not distinguish any position up the Uruguay, both sides of which were equally within that region, according to the distribution with reference to which he spoke. But the position of St. Salvador is conclusively settled by information from another quarter. In Hakluyt, (vol. iii. p. 729,) is "a Ruttier for The River Plate." The pilot who prepared it gives the various methods of striking the mouths of the Parana in proceeding from the island of Martin Garcia. A caution is interposed—" and if you fall into the mouth of the river which is called the Uruay you must leave it on the right hand.” He adds that all the mouths of the Parana, which are five in number, have their eastern termination infested with shoals for an extent of more than two leagues. Describing one of the routes more particularly, he says, "From the isle of Martin Garcia unto St. Salvador is nine or ten leagues. This is an island which standeth two leagues within the first mouth, where Sebastian Caboto took possession." The pilot, it will be seen, gives the name of St. Salvador, not to the river, but to a port. Cabot himself does the same, for in describing the assault finally made on the upper fort by the natives, he speaks of a similar attack on the port of St. Salvador, where the ships lay.* .* It seems certain, then, that the first position fortified by

* "Lo mesmo hizieron de la poblacion que avian hecho en el puerto que llaman de S. Salvador adonde estaban los navios." (Herrera, Dec. iv. lib. viii. cap. xi.)

Cabot was in the most northern mouth of the Parana, on an island about two leagues from where it reaches the La Plata. On the map of Louis Stanislaus d'Arcy de la Rochette,* this most northern avenue is divided into two parts, the upper of which is designated as "Rio Paca," and the lower, that issues into the La Plata, as" Rio Naranjos." St. Salvador was, of course, situated on the latter, or perhaps on the stream next in order to the south which also communicates with the Rio Paca and thus forms with the Rio Naranjos a considerable delta. In a Memoir drawn up by Lopez Vaz, a Portuguese, and taken with the author by the fleet sent forth in 1586 by the Earl of Cumberland, the fort where Cabot left his ships is said to be then standing. Its distance from the sea is, however, misstated either by him or the translator. (Hakluyt, vol. iii. p. 788.)

It is desirable to fix this first point of occupation, not only as a matter curious in itself, but because Charlevoix, (Histoire du Paraguay, tom. i. p. 27,) with his usual wild inaccuracy, would throw the whole subject into confusion. He represents Cabot

to have finally left the ships at the island of St. Gabriel, and proceeded in boats up the Uruguay, by mistake, and he imagines two reasons why such a blunder was committed. He doos not even allow the Uruguay to have been the St. Salvador, but makes it one of the tributaries of that river a considerable distance up the stream.

In order to avoid the tedious interruption of the narrative, one other probable misconception was not adverted to at the moment.

* "Colombia prima or South America, in which it has been attempted to delineate the extent of our knowledge of that continent, extracted chiefly from the original manuscript Maps of His Excellency, the late Chevalier Pinto; likewise from those of Joao Joaquim da Rocha, Joao da Costa Ferreira, El Padre Francisco Manuel Sobreviela, &c. And from the most authentic edited accounts of those countries. Digested and constructed by the late eminent and learned Geographer, Louis Stanislas D'Arcy de la Rochette. London, published by William Faden, Geographer to His Majesty and to His Royal Highness the Prince of Wales, June 4th, 1807." This Map is in the Topographical Department of the King's Library, British Museum.

It has been assumed, with Herrera, that Cabot left his vessels at the island of St. Gabriel, and proceeded thence in boats. More probably, however, the island of Martin Garcia was the one intended. Eden says expressly, (fol. 316,) that De Solis was killed in attempting to take possession of the island of Martin Garcia, and that it was the same afterwards carried by Cabot. We must bear in mind that Herrera is giving, somewhat loftily and reluctantly, the details of an expedition to which he attaches little importance, and he might not care for minute accuracy. He saw the name of Gabriel conferred by Cabot, and did not choose, perhaps, to occupy the page of his History with describing the further progress of six leagues before the ships were quitted. The account of Eden, who approached the subject in a different temper, is confirmed by other considerations. The island is spoken of by Herrera as one standing by itself. Now the St. Gabriel is a group of small islets, correctly stated in the "Ruttier" to be five in number. But still more conclusively: Cabot's report, as given by Herrera, states that seven leagues from the island at which he left his ships, he came to the mouth of a river, which he called St. Salvador, and to which he afterwards brought up his ships. Now the "Ruttier" speaks of the position at St. Salvador, as nine leagues in all from the island of Martin Garcia, two of which being up the St. Salvador, there is, of course, an exact correspondence. The St. Gabriel group, on the contrary, is correctly stated in the "Ruttier" to lie six leagues lower down than the island of Martin Garcia. While the statement of Eden produces greater harmony in the accounts, the position of the fort is not contingent on success in this reconciliation, but seems conclusively settled by the language of the "Ruttier."

An incident is mentioned by Gomarra,* but without the attendant circumstances, as occurring at this point, from which it would

* Gomarra, cap. lxxxix. "En el puerto de San Salvador que es otro Rio quarenta leguas arriba, que entra en el de la Plata, le mataron los Indios dos Españoles i no los quisieron comer diciendo que eran Soldados que ià los havian probado en Solis i sus compañeros."

appear that the position was not gained without resistance. The natives killed and carried off two Spaniards, but declared, in a spirit of fierce derision, that they would not eat them, as they were soldiers, of whose flesh they had already had a specimen in De Solis and his followers!

CHAP. XXI.

CABOT PROCEEDS UP THE PARANA-ERECTS ANOTHER FORT CALLED SANTUS SPIRITUS, AND AFTERWards fort CABOT-ITS POSITION-CONTINUES TO ASCEND CURIOSITY OF THE NATIVES AS TO THE EXPEDITION-PASSES THE MOUTH OF THE PARANA-ENTERS THE PARAGUAY-SANGUINARY BATTLE THIRTY-FOUR LEAGUES UP THAT RIVER-THREE HUNDRED OF THE NATIVES KILLED, WITH A LOSS TO CABOT OF TWENTY-FIVE OF HIS PARTY-MAINTAINS HIS POSITION-GARCIA ENTERS THE RIVER-INTERVIEW WITH CABOT-MISTAKES OF CHARLEVOIX, &c.-CABOT RETURNS TO THE FORT SANTUS SPIRITUS.

HAVING Completed the Fort, and taken every precaution for the safety of the ships at St. Salvador, Cabot resolved to ascend the Parana. Leaving, therefore, a party under the command of Antonio de Grajeda, he proceeded in the boats and a caravel cut down for the purpose. The point at which he next paused, and built a second Fort, is not a matter of doubt. It was on the south bank of the Parana, near a river called by the natives Zarcaranna or Carcaranna. This name was subsequently changed by the Spaniards into Terceiro. On the map of De la Rochette, already referred to, and also on that of Juan de la Cruz Canoy Olmedilla,* it is designated at the early stages as Terceiro, but lower down, gathering strength, it reassumes the aboriginal

'Mapa Geografica de America Meridional dispuesto y gravado por de Juan de la Cruz Canoy Olmedilla, Geogfo. Pensdo. de S. M. Individuo de la Rl. Academia de S". Fernando, y de la Sociedad Bascongada de los Amigos del Pais; teniendo presentes varios mapas y noticias originales con arreglo á observaciones astronomicas Año de 1775. Este Mapa de los Dominios Españoles y Portugueses en America Meredional, es una copia literal y exacta de un Mapa Español mui raro; compuesto y gravado en Madrid, ano 1775, de orden del Rey España, por Dr. Juan de la Cruz Cano y Olmedilla, Geofo. Pedo. de S. M. C. Londres Publicardo por Guillermo Faden, Geografo del Rey, y del Principe de Gales, Enero 1. de 1799."

セコ