aware, for the first time, of the close and intimate connection between the subjects which have been treated of in the foregoing pages. We are here conducted into a field which has been as yet but little cultivated. I have ventured to follow the method first employed with such brilliant results in the Zoological works of Aristotle, and which is especially suited to lay the foundation of scientific confidence,—a method which, whilst it continually aims at generality of conception, seeks, at the same time, to penetrate the specialities of phenomena by the consideration of particular instances.
The enumeration of forms according to physiognomic diversity is, from the nature of the case, not susceptible of any strict classification. Here, as everywhere else, in the consideration of external conformation, there are certain leading forms which present the most striking contrasts: such are the groups of arborescent grasses, plants of the aloë form, the different species of cactus, palms, needle-trees, Mimosaceæ, and Musaceæ. Even a few scattered individuals of these groups are sufficient to determine the character of a district, and to produce on a non-scientific but sensitive beholder a permanent impression. Other forms, though perhaps much more numerous and preponderating in mass, may not be calculated either by the outline and arrangement of the foliage, or by the relation of the stem to the branches,—by luxuriant vigour of vegetation,—by cheerful grace,—or, on the other hand, by cheerless contraction of the appendicular organs, to produce any such characteristic impressions.
As, therefore, a "physiognomic classification," or a divi-
