Note 2 — STEPPES AND DESERTS.

p. 2.—“‘Banks’ is the name given by the natives to this phenomenon.”

The Llanos of Caraccas are occupied by a great and widely extended formation of conglomerate of an early period. In descending from the vallies of Aragua, and crossing over the most southern ridge of the coast chain of Guigue and Villa de Cura towards Parapara, one finds successively, gneiss and mica slate;—a probably silurian formation of clay slate and black limestone;—serpentine and greenstone in detached spheroidal masses;—and, lastly, close to the margin of the great plain, small hills of augitic amygdaloid and porphyritic slate. These hills between Parapara and Ortiz appear to me like volcanic eruptions on the ancient sea-shore of the Llanos. Farther to the north are the celebrated grotesque-shaped cavernous rocks of Morros de San Juan; they form a kind of rampart, have a crystalline grain like upheaved dolomite, and are rather to be regarded as parts of the shore of the ancient gulf than as islands. I term the Llanos a gulf, for when we consider their small elevation above the present sea level, their form open as it were to the equatorial current sweeping from east to west, and the lowness of the eastern coast between the mouth of the Orinoco and the Essequibo, we can scarcely doubt that the sea once overflowed the whole basin between the coast chain of Caraccas and the Sierra de la Parime, and beat against the mountains of Merida and Pamplona; (as it is supposed to have overflowed the plains of Lombardy, and beat against the Cottian and Pennine Alps). The strike or inclination of the American Llanos is also directed from west to east. Their height at Calabozo, 400 geographical miles from the sea, is barely 30 toises (192 English feet); being 15 toises (96 English feet) less than that of Pavia, and 45 toises (288 English feet) less than that of Milan, in the plains of Lombardy between the Alps and Apennines. The form of the surface of this part of the globe reminds one of Claudian’s expression, “curvata tumore parvo planities.” The horizontality of the Llanos is so perfect that in many portions of them no part of an area of more than 480 square miles appears to be a foot higher than the rest. If, in addition to this, we imagine to ourselves the absence of all bushes, and even in the Mesa de Pavones the absence of any isolated palm-trees, it will afford some idea of the singular aspect of this sea-like desert plain. As far as the eye can reach, it can hardly rest on a single object a few inches high. If it were not that the state of the lowest strata of the atmosphere, and the consequent changes of refraction, render the horizon continually indeterminate and undulating, altitudes of the sun might be taken with the sextant from the margin of the plain as well as from the horizon at sea. This great horizontality of the former sea bottom makes the “banks” more striking. They are broken strata which rise abruptly from two to three feet above the surrounding rock, and extend uniformly over a length of from 40 to 48 English geographical miles. The small streams of the Steppes take their rise on these banks.

In passing through the Llanos of Barcelona, on our return from the Rio Negro, we found frequent traces of earthquakes. Instead of the banks standing higher than the surrounding rock, we found here solitary strata of gypsum from 3 to 4 toises (19 to 25 English feet) lower. Farther to the west, near the junction of the Caura with the Orinoco, and to the east of the mission of S. Pedro de Alcantara, an extensive tract of dense forest sank down in an earthquake in 1790, and a lake was formed of more than 300 toises (1918 English feet) diameter. The tall trees (Desmanthus, Hymenæas, and Malpighias) long retained their foliage and verdure under the water.