4 



ON THE NATURAL SYSTEM. 



containing epigynous genera, included among the Perigynous and Hypogynous Divisions, the raphe is never next the placenta in pendulous ovules, at least not in the epigynous genera themselves, 

 unless part of Dipterocarpece are an exception. Such genera are numerous and widely distributed among these Divisions, and therefore if it is admitted that the raphe next the placenta in 

 pendulous ovules is a character indicating a high degree of development, this fact would materially assist in showing that the epigynous character is one of comparatively low development. 



5. And finally, when excess of development and monstrosity occurs, a superior ovary I believe never becomes inferior, but the opposite may take place, an instance of which occurred in 

 Cannabis sativa, from the mode in which it was cultivated, the embryo becoming tricotyledonous, the leaves 3 in a whorl, and the female plants producing 3 or 4 male flowers (Ann. Nat. Hist. Ser. 

 3, Vol. 1, p. 102). The fruit has the calyx so completely adherent that no trace of it can be seen except in the variety C. hidica, hwt in these plants it became in some few flowers, a nearly detached 

 semi-transparent membranous cup, and in a somewhat monstrous flower with 3 stigmas, it was free and of a greenish colour, having three faintly marked but distinct green ribs. Three well 

 characterized instances are described by Dr. M. T. Masters: one of Daucus carota, in which the ovary had become nearly superior and the calyx partially free (Trans. Linn. Soc. Vol. XXIII. 

 PI. 54, f. 4); one of a Campanula, in which the ovary was quite superior (Trans. Linn. Soc. Vol. XXIII. PI. 34, f. 6); and the third of a Fuchsia, with the ovary superior (ib. f. 7). Two 

 instances are also mentioned by Dr. Lindley, one of them a Cucurbitaceous plant, becoming hermaphrodite with a half-superior ovary, and the other an Umbelliferous plant, (Veg. Kingd, 

 p. 774,) in which the ovary had become superior, the calyx surrounding it loosely. Prof. Oliver has also observed the calyx detached in a species of Rudbeckia (Trans. Linn. Soc. V. XXIII. p. 365). 

 And agreeably to this view it will be seen that the epigynous families, which are in the Tables placed among the perigynous and hypogynous, and also the families containing epigynous genera find 

 their natural station almost exclusively among the retrograde subdivisions, i.e., those which show an inferior degree of development, as regards the position of the raphe and single carpel. Thus 

 in Table III. the families containing epigynous genera find their place for the most part in the Scleranthal Subdivision. Table IV. may possibly be a retrograde branch of Table VI., but whether 

 so or not, the single carpel is rarely always anterior, and the raphe very rarely in pendulous ovules next the placenta, and in this Table the inferior ovary frequently occurs. And in Table VI. it 

 will be seen I have arranged the families of the Proteal Subdivision into Heterocarpous and Procarpous Sections, and the inferior ovary and highly perigynous stamens occur exclusively in the 

 Heterocarpous Section. 



From these peculiarities of structure it appears to me as certain that in the ovary, ovule, seed, and embryo of the apetalous families of the Class Exogens, its two extremes or most widely 

 separated parts are indicated ; and such an arrangement of phanerogamous plants would to some extent correspond in principle to that of the two highest Classes of the Vertebrata, i.e., as far as the 

 embryo is concerned,— Endogens having but one cotyledon, imperfectly foliaceous,* might be compared to Birds, they having less perfectly developed limbs comparatively with the higher Orders 

 of the Mammalia, and, on the other hand, Conifers and Proteacece having the highest number of cotyledons might be regarded as showing some analogy with the Quadrumana in their more 

 perfectly developed limbs. It might be said that the Endogens show an imperfect tendency to become epigynous, and that they should, as being of a lower form of organization than Epigynous 

 Exogens, so far correspond with them ; but still it will be admitted that they show, although in a minor degree, a tendency to* be epigynous or perigynous. But supposing it to remain a question 

 as to whether the inferior ovary is a lower form of organization than the superior, yet the arguments in support of the Epigynous Exogens being a strictly natural assemblage of families are, I 

 believe, such as cannot be well superseded ; and that they are not so highly developed in the particulars of structure referred to as the Perigynous and Hypogynous Divisions, remains as an 

 argument for the arrangement in the accompanying Tables. And it may also deserve notice, that as far as there exists a tendency to enclosed fruit among the Cryptogams, there is no doubt 

 that it is confined exclusively to the comparatively lower forms, and among the Phanerogams the same law might be expected to obtain, however distant the affinities between them and the 

 Cryptogams might be. 



Value of the Characters Procarpous and Heterocarpous. 



The value of these characters will be fully adverted to in the remaining part of the work - On Relative Position," and the meaning intended to be attached to these terms will be sufficiently 

 understood by reference to the explanatory notes at the foot of each Table. In part of the Subdivisions these characters will appear of doubtful value, but I am able to add that they have appeared 

 of more value withm the last twelvemonth than they did three or four years since, and I do not see how the Epigynous Division could be arranged in Subdivisions by any other character. The 

 single dagger (f) I regard as a neutral character and such families may be placed in either Procarpous or Heterocarpous Sections; on the other hand I estimate the single asterisk (•) to be 



* Besides having with few exceptions imperfectly developed leaves, the number of the floral envelopes in the Endogens never rises so high as in the Exogens, the usual number of 5 sepals and 5 petals in the latter not being known among the former. 



