the pathway, is a copious spring of water, into which we plunged the thermometer, which fell to 15.4 degrees. At a hundred toises distance from this spring is another equally limpid. If we admit that these waters indicate nearly the mean heat of the place whence they issue, we may fix the absolute elevation of the station at 520 toises, supposing the mean temperature of the coast to be 21 degrees, and allowing one degree for the decrement of caloric corresponding under this zone to 93 toises. We should not be surprised if this spring remained a little below the heat of the air, since it probably takes its source in some more elevated part of the peak, and possibly communicates with the small subterranean glaciers of which we shall speak hereafter. The accordance just observed between the barometrical and thermometrical measures is so much more striking, because in mountainous countries, with steep declivities, the springs generally indicate too great a decrement of caloric, for they unite small currents of water, which filtrate at different heights, and their temperature is consequently the mean between the temperature of these currents. The spring of Dornajito has considerable reputation in the country; and at the time I was there, it was the only one known on the road which leads to the summit of the volcano. The formation of springs demands a certain regularity in the direction and inclination of the strata. On a volcanic soil, porous and splintered rocks absorb the rain waters, and convey them to considerable depths. Hence arises that aridity observed in the greater part of the Canary Islands, notwithstanding the considerable height of their mountains, and the mass of clouds which navigators behold incessantly overhanging this archipelago.

From Pino del Dornajito to the crater of the volcano we continued to ascend without crossing a single valley; for the small ravines (barancos) do not merit this name. To the eye of the geologist the whole island of Teneriffe is but one mountain, the almost elliptical base of which is prolonged to the north-east, and in which may be distinguished several systems of volcanic rocks formed at different epochs. The Chahorra, or Montana Colorada, and the Urca, considered in the country as insulated volcanoes, are only little hills abutting on the peak, and masking its pyramidal form. The great volcano, the lateral eruptions of which have given birth to vast promontories, is not however precisely in the centre of the island, and this peculiarity of structure appears the less surprising, if we recollect that, as the learned mineralogist M. Cordier has observed, it is not perhaps the small crater of the Piton which has been the principal agent in the changes undergone by the island of Teneriffe.

Above the region of arborescent heaths, called Monte Verde, is the region of ferns. Nowhere, in the temperate zone, have I seen such an abundance of the pteris, blechnum, and asplenium; yet none of these plants have the stateliness of the arborescent ferns which, at the height of five or six hundred toises, form the principal ornament of equinoctial America. The root of the Pteris aquilina serves the inhabitants of Palma and Gomera for food; they grind it to powder, and mix with it a quantity of barley-meal. This composition, when boiled, is called gofio; the use of so homely an aliment is a proof of the extreme poverty of the lower order of people in the Canary Islands.

Monte Verde is intersected by several small and very arid ravines (canadas), and the region of ferns is succeeded by a wood of juniper trees and firs, which has suffered greatly from the violence of hurricanes. In this place, mentioned by some travellers under the name of Caravela,* (* "Philosophical Transactions" volume 29 page 317. Carabela is the name of a vessel with lateen sails. The pines of the peak formerly were used as masts of vessels.) Mr. Eden states that in the year 1705 he saw little flames, which, according to the doctrine of the naturalists of his time, he attributes to sulphurous exhalations igniting spontaneously. We continued to ascend, till we came to the rock of La Gayta and to Portillo: traversing this narrow pass between two basaltic hills, we entered the great plain of Spartium. At the time of the voyage of Laperouse, M. Manneron had taken the levels of the peak, from the port of Orotava to this elevated plain, near 1400 toises above the level of the sea; but the want of water, and the misconduct of the guides, prevented him from taking the levels to the top of the volcano. The results of the operation, (which was two-thirds completed,) unfortunately were not sent to Europe, and the work is still to be recommenced from the sea-coast.

We spent two hours and a half in crossing the Llano del Retama, which appears like an immense sea of sand. Notwithstanding the elevation of this site, the centigrade thermometer rose in the shade toward sunset, to 13.8 degrees, or 3.7 degrees higher than toward noon at Monte Verde. This augmentation of heat could be attributed only to the reverberation from the ground, and the extent of the plain. We suffered much from the suffocating dust of the pumice-stone, in which we were continually enveloped. In the midst of this plain are tufts of the retama, which is the Spartium nubigenum of Aiton. M. de Martiniere, one of the botanists who perished in the expedition of Laperouse, wished to introduce this beautiful shrub into Languedoc, where firewood is very scarce. It grows to the height of nine feet, and is loaded with odoriferous flowers, with which the goat hunters, that we met in our road, had decorated their hats. The goats of the peak, which are of a deep brown colour, are reckoned delicious food; they browse on the spartium, and have run wild in the deserts from time immemorial. They have been transported to Madeira, where they are preferred to the goats of Europe.

As far as the rock of Gayta, or the entrance of the extensive Llano del Retama, the peak of Teneriffe is covered with beautiful vegetation. There are no traces of recent devastation. We might have imagined ourselves scaling the side of some volcano, the fire of which had been extinguished as remotely as that of Monte Cavo, near Rome; but scarcely had we reached the plain covered with pumice-stone, when the landscape changed its aspect, and at every step we met with large blocks of obsidian thrown out by the volcano. Everything here speaks perfect solitude. A few goats and rabbits only bound across the plain. The barren region of the peak is nine square leagues; and as the lower regions viewed from this point retrograde in the distance, the island appears an immense heap of torrefied matter, hemmed round by a scanty border of vegetation.

From the region of the Spartium nubigenum we passed through narrow defiles, and small ravines hollowed at a very remote time by the torrents, first arriving at a more elevated plain (el Monton de Trigo), then at the place where we intended to pass the night. This station, which is more than 1530 toises above the coast, bears the name of the English Halt (Estancia de los Ingleses* (* This denomination was in use as early as the beginning of the last century. Mr. Eden, who corrupts all Spanish words, as do most travellers in our own times, calls it the Stancha: it is the Station des Rochers of M. Borda, as is proved by the barometrical heights there observed. These heights were in 1803, according to M. Cordier, 19 inches 9.5 lines; and in 1776, according to Messrs. Borda and Varela, 19 inches 9.8 lines; the barometer at Orotava keeping within nearly a line at the same height.)), no doubt because most of the travellers, who formerly visited the peak, were Englishmen. Two inclined rocks form a kind of cavern, which affords a shelter from the winds. This point, which is higher than the summit of the Canigou, can be reached on the backs of mules; and here has ended the expedition of numbers of travellers, who on leaving Orotava hoped to have ascended to the brink of the crater. Though in the midst of summer, and under an African sky, we suffered from cold during the night. The thermometer descended as low as to five degrees. Our guides made a large fire with the dry branches of retama. Having neither tents nor cloaks, we lay down on some masses of rock, and were singularly incommoded by the flame and smoke, which the wind drove towards us. We had attempted to form a kind of screen with cloths tied together, but our enclosure took fire, which we did not perceive till the greater part had been consumed by the flames. We had never passed a night on a point so elevated, and we then little imagined that we should, one day, on the ridge of the Cordilleras, inhabit towns higher than the summit of the volcano we were to scale on the morrow. As the temperature diminished, the peak became covered with thick clouds. The approach of night interrupts the play of the ascending current, which, during the day, rises from the plains towards the high regions of the atmosphere; and the air, in cooling, loses its capacity of suspending water. A strong northerly wind chased the clouds; the moon at intervals, shooting through the vapours, exposed its disk on a firmament of the darkest blue; and the view of the volcano threw a majestic character over the nocturnal scenery. Sometimes the peak was entirely hidden from our eyes by the fog, at other times it broke upon us in terrific proximity; and, like an enormous pyramid, threw its shadow over the clouds rolling beneath our feet.

About three in the morning, by the sombrous light of a few fir torches, we started on our journey to the summit of the Piton. We scaled the volcano on the north-east side, where the declivities are extremely steep; and after two hours' toil, we reached a small plain, which, on account of its elevated position, bears the name of Alta Vista. This is the station of the neveros, those natives, whose occupation it is to collect ice and snow, which they sell in the neighbouring towns. Their mules, better practised in climbing mountains than those hired by travellers, reach Alta Vista, and the neveros are obliged to transport the snow to that place on their backs. Above this point commences the Malpays, a term by which is designated here, as well as in Mexico, Peru, and every other country subject to volcanoes, a ground destitute of vegetable mould, and covered with fragments of lava.

We turned to the right to examine the cavern of ice, which is at the elevation of 1728 toises, consequently below the limit of the perpetual snows in this zone. Probably the cold which prevails in this cavern, is owing to the same causes which perpetuate the ice in the crevices of Mount Jura and the Apennines, and on which the opinions of naturalists are still much divided. This natural ice-house of the peak has, nevertheless, none of those perpendicular openings, which give emission to the warm air, while the cold air remains undisturbed at the bottom. It would seem that the ice is preserved in it on account of its mass, and because its melting is retarded by the cold, which is the consequence of quick evaporation. This small subterraneous glacier is situated in a region, the mean temperature of which is probably not under three degrees; and it is not, like the true glaciers of the Alps, fed by the snow waters that flow from the summits of the mountains. During winter the cavern is filled with ice and snow; and as the rays of the sun do not penetrate beyond the mouth, the heats of summer are not sufficient to empty the reservoir. The existence of a natural ice-house depends, consequently, rather on the quantity of snow which enters it in winter, and the small influence of the warm winds in summer, than on the absolute elevation of the cavity, and the mean temperature of the layer of air in which it is situated. The air contained in the interior of a mountain is not easily displaced, as is exemplified by Monte Testaccio at Rome, the temperature of which is so different from that of the surrounding atmosphere. On Chimborazo enormous heaps of ice are found covered with sand, and, in the same manner as at the peak, far below the