the Rio Sabalos tributary when that portion of the valley was an arm of the lake, and before the old continental divide was cut down to its present elevation. The other rapids are due to the varying hardness of the underlying rock. Portions of the river bed in the lower part of this division are below sea level, showing that it was formed when the land was much higher than at present, and that the greater part of such sediment as now reaches it from its tributaries is transported by the waters of the main

stream.

The third division of the Rio San Juan, from the mouth of the San Carlos river to the sea, is quite different in character from either of the preceding ones. Its waters are swift, turbid, and shallow, and the channel shifts continually. This is due to the San Carlos and Sarapiqui tributaries, which, heading in the mountains of northern Costa Rica, transport and furnish to the main stream large quantities of sand, thus supplying material for the river's flood plain and helping to push the delta seaward year by year. The lower San Juan's sharp bend to the left is due to the northward sand-drift along the coast, caused by wave action, and the Rio Colorado is a comparatively new channel cut by the water in its effort to reach the sea without traversing the constantly increasing territory to the northward.

As regards rainfall, the Nicaraguan depression may be divided into two sections, the eastern, extending from the Atlantic coast to the divide between the Caribbean and the lake drainage, and the western, extending from this divide to the Pacific Ocean. In the eastern division the rainfall is heavy, particularly near the coast, where it approaches three hundred inches annually, and it is quite uniformly distributed throughout the year. In the western division it is much less, and there is a welldefined dry season of five or six months, during which little or no precipitation occurs. The eastern slope, bathed in constantly recurring showers, is clothed with a dense virgin forest, which forms an elastic and absorbent cushion, breaking the impact of the drops and distributing large volumes of water so slowly as to prevent sudden floods and consequent rapid erosion of the soil. The result is that the` surface of the country, composed principally of tenacious red clay, remains comparatively unchanged from year to year. But in the western division the conditions are very different: during the dry season the foliage is parched, extensive forest fires rage, and the surfaces of numerous clayey llanos shrink and crack in every direction. Consequently when the rains begin, the forest growth, largely denuded of its foliage, offers comparatively little protection to

the underlying soil, which falls a ready victim to the erosive action of the water, while the flat llanos, disintegrated as if by the action of frost, succumb in a measure to the same influence. As might be expected, this division is characterized by steep hills and rugged gullies, very different in appearance from the smooth slopes and shallow watercourses east of the divide.

It is unnecessary at this point to discuss the different classes of material through which the canal must be excavated, as this subject will receive due consideration in a subsequent chapter.

CHAPTER V

CANAL PROJECTS, PAST AND PRESENT

T

HE project of an interoceanic canal is but a natural development of the idea which impelled Columbus on his first adventurous voyage to the New World and led contemporaneous and subsequent explorers to trace the coast line of the two Americas from Labrador to Cape Horn. Failing in their attempts to find the apocryphal passage to the Indies, the project of excavating an artificial channel to repair the obvious neglect of Nature gradually took form within their minds. Philip II. sent a commission to the Isthmus to make surveys and decide upon the practicability of cutting a canal, but the report was so favorable and dwelt so strongly upon the advantages to international commerce of the proposed work that Philip, whose philanthropy did not extend to alien races, abandoned the project and even decreed death to him who should advocate it. For many years the struggle between England and Spain for supremacy in America kept the problem of trans-isthmian communication in the background, but in 1740 La Condamine,

But

whom a long residence in Central and South America had afforded ample opportunities for observation, presented a paper before the Academy of Sciences at Paris, calling attention to the importance of an interoceanic canal and advocating the construction of one through Nicaragua. Thirty-one years later the Viceroy of Mexico attempted to discover a suitable location across the Isthmus of Tehuantepec and from 1779 to 1781 Spanish engineers and explorers, unfamiliar with the use of locks, explored the Nicaraguan depression and, convinced of the impracticability of a sea-level cutting, reported unfavorably upon it. Hodgson and Lee, the English agents on the Mosquito Coast, were favorably impressed by the San Juan valley and expressed themselves so strongly as to interest the British Government and to materially encourage the disastrous invasion of 1780. During the Napoleonic wars Alexander von Humboldt travelled through Central America, reporting favorably upon the Nicaraguan route, which he declared well adapted to the construction of a canal of large dimensions. In 1826 a struggle between English and American projectors resulted in the acquisition by the latter of a liberal concession, but the funds necessary to carry on the work could not be obtained and no construction was attempted. Soon afterwards a