Page:Aspects of nature in different lands and different climates; with scientific elucidations (IA b29329668 0002).pdf/20

From Wikisource
Jump to navigation Jump to search
This page needs to be proofread.

The deserts to the south of the Atlas, and the immense plains or steppes of South America, must be regarded as only local phenomena. The latter, the South American steppes, are clothed, in the rainy season at least, with grass, and with low-growing almost herbaceous mimosas. The African deserts are, indeed, at all seasons devoid of vegetation; seas of sand, surrounded by forest shores clothed with perpetual verdure. A few scattered fan-palms alone recall to the wanderer's recollection that these awful solitudes belong to the domain of the same animated terrestrial creation which is elsewhere so rich and so varied. The fantastic play of the mirage, occasioned by the effects of radiant heat, sometimes causes these palm trees to appear divided from the ground and hovering above its surface, and sometimes shews their inverted image reflected in strata of air undulating like the waves of the sea. On the west of the great Peruvian chain of the Andes, on the coasts of the Pacific, I have passed entire weeks in traversing similar deserts destitute of water.

The origin of extensive arid tracts destitute of plants, in the midst of countries rich in luxuriant vegetation, is a geognostical problem which has hitherto been but little considered, but which has doubtless depended on ancient revolutions of nature, such as inundations or great volcanic changes. When once a region has lost the covering of plants with which it was invested, if the sands are loose and mobile and are destitute of springs, and if the heated atmosphere, forming constantly ascending currents, prevents precipitation taking place from clouds[9], thou-