ON THE NATURAL SYSTEM. 



Value of the Epigynous Character. 



In carrying out this mode of investigation especial attention is in the first place directed to the fact that there exists among those Cryptogams which approach the Phanerogams, two forms 

 of organisation differing from each other in that one of them, represented by the Bryacece, has its fructification contained in a seed vessel called a theca, which is surmounted by 1, 2, or 3 rows of 

 teeth resembling rudimentary leaves* ; and the other, represented by the Lycopodiacece, has a fructification of a very different kind. Of these forms of organization we find the former, which is 

 represented by the Bryacece, to be decidedly the lower of the two as it stands in immediate relation with the lowest forms of the vegetable system, and the latter, the vascular Cryptogams 

 represented by the Lycopodiacece, to be as immediately connected with the Gymnosperms, besides which the higher development especially of the fossil genera place it far in advance of the Bryacece. 



The first feature of the arrangement in the accompanying Tables is the division of the Class Exogens in an analogous way, placing the Epigynous Division first, as being connected with the 

 lower of these two groups of the Cryptogams, and all the other families, whether Perigynous or Hypogynous, in advance of them, as being connected with the more highly developed group of 

 which Lycopodiacece are the representative. We find however in systematic arrangements that the greater part of the Epigynous Division is generally placed among the higher developed sections 

 of the Class Exogens, but there are some peculiarities in the structure of a part of the families of this Division which, according to the generally received opinions of the value of characters, require 

 their being stationed at the commencement; besides which the fact of Hypogynous Exogens being allied exclusively to the Gymnosperms and Lycopodiacece claims for them a position in advance 

 of the Epigynous, provided it is admitted that there exists a connection between Bryacece and the latter. The peculiarities in structure to which I refer are as follows : — 



1. The ovary is often but feebly developed, as in the Composite, and, as I have endeavoured to show (Ann. Nat Hist. 2nd Ser. V.l. XL), in Calytrix virgata there is no perceptible trace of 

 a carpel, and supposing such an ovary to have the ovule of Balanophoracece, it would then be a much lower form of organization than any which occurs in Perigynous or Hypogynous families, in 

 which the ovary is always very distinct, even in the epigynous genera included among them. Camarina on the other hand has a seed-vessel remarkable for the excess of spiral vessels. 



2. The simplicity of the ovule of the Balanophoracece, and the absence of all integument in some genera of Loranthacece 9 are variations of structure which may be regarded as indicating a 

 lower degree of development than any which occurs among the Perigynous or Hypogynous Exogens. On the other hand, the ovule of Platanus has two coats so distinct as to be scarcely if at 

 all adherent, and the Acanthacea described by Mr. R. Kippist (Trans. Linn. Soc. Vol. XIX. p. 65), have a testa containing an unusual quantity of spiral vessels ; Dr. Lindley has also observed 

 spiral vessels on the outside of the testa very abundantly in Collomia (Bot. Reg. fol. 1 166) and other genera of Polemotiiacece, and Mr. Quekett in the testa of the almond. May not these two 

 extremes in the structure of the seed indicate the most widely separated parts of the Class Exogens? From the observations of Mr. Miers it appears that spiral vessels occur in the testa of Halesia 

 but in much less quantity. 



3. The number of the cotyledons also may be regarded as in some degree showing that the Perigynous and Hypogynous families are more highly developed than the Epigynous. If we take 

 a general survey of the Cryptogams, we find the lower families destitute of leaves, and it is only in those which approach the Dicotyledons that they are both numerous and highly developed ; on 

 this account it might be expected that families the seeds of which had the greatest number of cotyledons would occupy the highest station in an arrangement, and those which had but one, and that 

 comparatively less developed, the lowest. 



Agreeably to this suggestion we always place the Endogens at the commencement of a system of phanerogamous plants, and if this principle is adhered to, Conifers, which have many 

 cotyledons, and Proteacece, where 5 occur,f should be regarded as the highest developed forms of Eoxgens, 4 being the highest number known to be produced in the Epigynous families, and that 

 to a much less extent than in the Perigynous and Hypogynous. In Proteacece, however, the increase in the number of the cotyledons would perhaps scarcely deserve attention, but in the Comferce 

 the increase is much greater, and as their affinity is exclusively with the Hypogynous families, an argument is hence afforded for placing the Hypogynous families in advance of the Epigynous, so as 

 to be among phanerogamous plants the opposite extreme of the Endogens. 



4. The raphe of the pendulous anatropal ovule in all the apetalous families of the Epigynous Division is dorsal, (with one exception and there it is not next the placenta,) which perhaps 

 more decidedly than any other character shows that they are a lower form of organisation than the Perigynous and Hypogynous Divisions, because this is the general character of the Endogens, 

 or at least the pendulous anatropal ovule with the raphe next the placenta must be extremely rare among them. And I think also it deserves notice that in the epigynous families and families 



* For further remarks v. Table II., the accompanying explanation. 



f Persoonia has 5 cotyledons, but as in descriptions of this genus the number is not particularly stated further than that there are several, possibly there are sometimes more. 



