102 ENDOGENS. 



bumiiious monocotyledonous seeds of the latter; and the resemblances 

 bet\7een Hydropeltis and Hydrocharis in the other case are so very great, 

 that Schidtz and others actually refer them to the same class. 



Endogens probably contain more plants contributing to the food of man, 

 and fewer poisonous species in proportion to their whole number, than 

 Exoo-ens. Grasses, with their floury albumen, form a large portion of this 

 class, to which have to be added Palms yielding fruit, wine, sugar, sago ; 

 Amms, Arrow-roots, AmaryUids, &c., producing arrow-root; the nutritious 

 fruit of Plantains ; the aromatic secretions of Gingers ; and Orchisworts, 

 forming salep. Among the deleterious species we have no inconsiderable 

 number among Amaryllids, Arimis, Melauths, and even Lilies. 



In this, as m all other large groups, the extremes of development are so 

 far apart, that one would be tempted to doubt the possibihty of their being 

 mere forms of each other, were it not certain that numerous traces exist in 

 the vegetable kingdom of a frequent tendency to produce the typical struc- 

 ture of a natural association of whatever kind in both an exaggerated and 

 degraded state, if such figurative terms may be employed in science. For 

 instance, the genus Ficus contains some species creeping on the ground 

 like diminutive herbaceous plants, and others rising into the air to the 

 height of 150 feet, overspreading with the arms of their colossal trmiks a 

 sufficient space of ground to protect a multitude of men ; the type of 

 organisation in the willow is in like manner represented on the one hand by 

 the tiny Salix herbacea, which can hardly raise its head above the dwarf 

 moss and saxifrages that surround it; and on the other by Salix alba, 

 a tree sixty feet high. Then among natural orders we have the Rosal 

 structm-e exaggerated, on the one hand, into the arborescent Pomacese, 

 and degraded, on the other, into the apetalous imperfect Sanguisorbese ; 

 the M\Ttal type, highly developed in Myrtus, and almost obhterated in 

 Streamworts (Halorageae) ; the Urtical, in excess in Artocai-pus, and quite 

 imperfect in CeratophyUum ; Grasses, presenting the most striking differ- 

 ences of perfection between the moss-like Knappia, and Bamboos a hundred 

 feet high ; and the Lilial in equally different states of development, when 

 Asparagus is compared with the Dragon-tree, or an autumnal squill with 

 an arborescent Yucca. So, in like manner, we find at one extreme of the 

 organisation of the class of Endogens, Pahns, Plantains, and arborescent 

 Liliaceous species, and at the other, such submersed plants as Potamogeton, 

 Zannichellia, and Duckweed, the latter of which has not even the distinction 

 of leaf and stem, and bears its flowers, reduced to one carpel and two 

 stamens, without either calyx or corolla — and therefore at the minimum of 

 reduction, if to remain flowers at all — in little chinks in its edges. 



The classification of Endogens is not a subject upon Avhich there is any 

 very great diversity of opinion among botanists. If the natural orders are 

 sometimes not distinctly limited, they are, upon the whole, grouped much 

 better than those of Exogens ; and although it may be expected that some 

 changes have still to be introduced into this part of systematic botany, yet 

 there seems no probability of the limits of the natural orders themselves 

 being disturbed to any considerable extent. 



The principles of classification here adopted are the following : — 



In the first place, all those numerous species whose flowers are like 

 grasses are stationed by themselves, and constitute the Glumal alliance. 

 They are not perhaps so close upon flowerless plants as some hereafter to 

 be mentioned, but they form, as a whole, the lowest condition of structure 

 to which a great mass of Endogens is reduced. Their flowers may be 



