To these phenomena belongs especially the boundless forest district which, in the torrid zone of South America, connects the river basins of the Orinoco and the Amazon. This region deserves, in the strictest sense of the word, to be called a _primeval_ forest—a term that has, in recent times, been so frequently misapplied. Primeval (or primitive), as applied to a forest, a nation, or a period of time, is a word of rather indefinite signification, and generally but of relative import. If every wild forest, densely covered with trees, on which man has never laid his destroying hand, is to be regarded as a primitive forest, then the phenomenon is common to many parts both of the temperate and the frigid zones; if, however, this character consists in impenetrability, through which it is impossible to clear with the axe, between trees measuring from 8 to 12 feet in diameter, a path of any length, primitive forests belong exclusively to tropical regions. This impenetrability is by no means, as is often erroneously supposed in Europe, always occasioned by the interlaced climbing “lianes,” or creeping plants, for these often constitute but a very small portion of the underwood. The chief obstacles are the shrub-like plants which fill up every space between the trees, in a zone where all vegetable forms have a tendency to become arborescent. If travellers, the moment they set foot in a tropical region, and even while on islands, in the vicinity of the sea-coast, imagine that they are within the precincts of a primeval forest, the misconception must be ascribed to their ardent desire of realizing a long-cherished wish. Every tropical forest is not _primeval_ forest. I have scarcely ever used the latter term in the narrative of my travels; although, I believe, that of all investigators of nature now living, Bonpland, Martius, Pöppig, Robert and Richard Schomburgk, and myself, have spent the longest period of time in primeval forests in the interior of a great continent. Notwithstanding the striking richness of the Spanish language in designations, (descriptive of natural objects, of which I have already spoken), yet one and the same word _monte_ is employed for a mountain and a forest, for _cerro_ (_montaña_), and for _selva_. In a work on the true breadth and the greatest extension of the chain of the Andes towards the east, I have shown how this twofold signification of the word _monte_ has led to the error, in a fine and extensively circulated English map of South America, of marking ranges of high mountains in districts occupied only by plains. Where the Spanish map of La Cruz Olmedilla, which formed the basis of so many others, indicated Cacao Woods, _Montes de Cacao_[67], Cordilleras were supposed to exist, although the Cacao-tree affects only the hottest of the low lands.