seated some of the black Congo monkeys (Mycetes palliatus) which at times, especially before rain and at nightfall, make a fearful howling, though not so loud as the Brazilian species. Screaming macaws, in their gorgeous livery of blue, yellow, and scarlet, occasionally flew overhead, and tanagers and toucans were not uncommon.

Twelve miles above Castillo we reached the mouth of the Savallo, and stayed at a house there to breakfast, the owner, a German, giving us roast wari, fowls, and eggs. He told me that there was a hot spring up the Savallo, but I had not time to go and see it. Above Savallo the San Juan is deep and sluggish, the banks low and swampy. The large palm, so common in the delta of the river, here reappeared with its great coarse leaves twenty feet in length, springing from near the ground.

Our boatmen continued to paddle all day, and as night approached redoubled their exertions, singing to the stroke of their paddles. I was astonished at their endurance. They kept on until eleven o'clock at night, when we reached San Carlos, having accomplished about thirty-five miles during the day against the current. San Carlos is at the head of the river, where it issues from the great lake of Nicaragua, about one hundred and twenty miles from Greytown. The mean level of the waters of the lake, according to the survey of Colonel O. W. Childs, in 1851, is 107 feet, so that the river falls on an average a little less than one foot per mile. The height of the lowest pass between the lake and the Pacific is said to be twenty-six feet above the lake, therefore at that point the highest elevation between the two oceans is only about 133 feet; but even allowing that an error of a few feet may be discovered when a thorough

survey is made across from sea to sea, there can be no doubt that at this point occurs the lowest pass between the Atlantic and the Pacific in Central America. This fact, and the immense natural reservoir of water near the head of the navigation, point out the route as a practicable one for a ship canal between the two oceans. * Instead of cutting a canal from the head of the delta of the San Juan to the sea, as has been proposed, the Colorado branch might be straightened, and dredged to the required depth. Higher up, the Torre, Castillo and Machuca Rapids form natural dams across the river. These might be raised, locks formed round them, and the water deepened by dredging between them. In this way the great expense of cutting a canal, and the fearful mortality that always arises amongst the labourers when excavations are made in the virgin soil of the tropics, especially in marshy lands, would be greatly lessened between the lake and the Atlantic. Another great advantage would be that the deepening of the river could be effected by steam power, so that it would not be necessary to bring such a multitude of labourers to the isthmus as would be required if a canal were cut from the river; the whole track, moreover, passes through virgin forests rich in inexhaustible supplies of fuel.*

San Carlos is a small town at the foot of the great lake, where it empties its waters into the San Juan river, its only outlet to the ocean. On a hill behind the town, and commanding the entrance to the river, are the

* Since the publication of the first edition of this work the commissioners, appointed by the Government of the United States to examine into the practicability of making a canal across the isthmus, have reported in favour of the Nicaraguan route. The total cost is estimated at £12,250,000 sterling.

ruins of a once strong fort built by the Spaniards, the crumbling walls now green with the delicate fronds of a maidenhair fern (Adiantum). The little town consists of a single rugged street leading up from the lake. The houses are mostly palm-thatched huts, with the bare earth floors seldom or never swept. The people are of mixed origin, Indian, Spanish, and Negro, the Indian element predominating. Two or three better built stores, and the quarters of the military governor, redeem the place from an appearance of utter squalor. Behind the town there are a few small clearings in the forest, where maize is grown. Some orange, banana, and plantain trees exhaust the list of the productions of San Carlos, which is supported by being a calling place for all vessels proceeding up and down the river, and by the Ulleros or rubber-men who start from it for expeditions up the Rio Frio and other rivers. We found there two men who had just been brought down the Rio Frio by their companions, greatly injured, by the lianas up which they had made their ladder to ascend one of the rubber trees, having broken and precipitated them to the ground. I learnt that this was a very unusual accident, the lianas generally being very tough and strong, like great cables.

Most fabulous stories have been told about the Rio Frio and its inhabitants; stories of great cities, golden ornaments, and light-haired people, and it may be useful to relate what is known about it.

The Rio Frio comes down from the interior of Costa Rica, and joins the San Juan, near where the latter issues from the lake. The banks of its upper waters are inhabited by a race of Indians who have never been

subjugated by the Spaniards, and about whom very little is known. They are called Guatuses, and have been said to have red or light-coloured hair and European features, to account for which various ingenious theories have been advanced; but, unfortunately for these speculations, some children, and even adults, have been captured and brought down the river by the Ulleros, and all these have the usual features and coarse black hair of the Indians. One little child that Dr. Seemann and I saw at San Carlos, in 1870, had a few brownish hairs amongst the great mass of black ones; but this character may be found amongst many of the indigenes, and may result from a very slight admixture of foreign blood. I have seen altogether five children from the Rio Frio, and a boy about sixteen years of age, and they had all the common Indian features and hair; though it struck me that they appeared rather more intelligent than the generality of Indians. Besides these, an adult woman was captured by the rubber-men and brought down to Castillo, and I was told by several who had seen her that she did not differ in any way from the usual Indian type.

The Guatuse (pronounced Watúsa) is an animal about the size of a hare, very common in Central America, and good eating. It has reddish-brown fur, and the usual explanation of the Nicaraguans is that the Indians of the Rio Frio were called "Guatuses" because they had red hair. It is very common to find the Indian tribes of America called after wild animals, and my own opinion is that the origin of the fable about the red hair was a theory to explain why they were called Guatuses; for the natives of Nicaragua, and of parts much nearer

home, are fond of giving fanciful explanations of the names of places and things: thus, I have been assured. by an intelligent and educated Nicaraguan, that Guatemala was so called by the Spaniards because they found the guaté (a kind of grass) in that country bad, hence "guaté malo," "bad guaté,"-whereas every student of Mexican history knows that the name was the Spanish attempt to pronounce the old Aztec one of Quauhtemallan, which meant the Land of the Eagle. I shall have other occasions, in the course of my narrative, to show how careful a traveller in Central America must be not to accept the explanations of the natives of the names of places and things.

The first people who ascended the Rio Frio were attacked by the Indians, who killed several with their arrows. Exaggerated opinions of their ferocity and courage were in consequence for a long time prevalent, and the river remained unknown and unexplored, and probably would have done so to the present day, if it had not been for the rubber-men. When the trade in india-rubber became fully developed, the trees in the more accessible parts of the forest were soon exhausted, and the collectors were obliged to penetrate farther and farther back into the untrodden wilds of the Atlantic slope. Some more adventurous than others ascended the Rio Frio, and being well provided with firearms, which they mercilessly used, they were able to defy the poor Indians, armed only with spears and bows and arrows, and to drive them back into the woods. The first Ulleros who ascended the river were so successful in finding rubber, that various other parties were organised, and now an ascent of the Rio Frio from San