2 ON THE NATURAL SYSTEM. 



departs so much from the character of Bignoniacece that it could not, without setting aside one of the most fundamental principles of classification, be referred to its true station. I believe the 

 same remark is applicable to entire families as to Combretacece and Humiriacece. 



The cause of this difference I would suggest is, that* the Vegetable Kingdom does not become so highly developed comparatively with the Animal, since cryptogamous plants generally 

 cannot be regarded as analogous to any higher Orders in the Animal Kingdom than the primordial forms of the Radiata, and the highest forms of phanerogamous plants cannot well be compared 

 to any higher developed portion of it than the Actiniadae. Hence it would appear that with the exception of the ovule and seed of the higher departments of the Vegetable Kingdom, which may 

 be compared to the mode of reproduction of some of the higher and perhaps the highest departments of the Animal Kingdom, all further comparisons between them must be impracticable. This 

 being admitted, it might be expected as a consequence, that the primary differences in structure which would take place in the Vegetable Kingdom would be fewer and less strongly marked, and 

 thus the difficulty in forming an arrangement would be accounted for. 



That there is much analogy between the Actiniadse and phanerogamous plants might I believe be admitted, if it can be proved that a leaf is a true branch, and therefore analogous to the 

 limb of an animal, and of this I have no doubt. I have seen monstrous cabbage leaves which distinctly show that there is no difference between a leaf and a branch,* and it appears to follow as 

 an unavoidable inference that leaves are analogous to limbs. Three of the leaves of a cabbage produced a great number of small leaves along their midribs on the upper surface, but none from 

 their footstalks, and one of them produced a comparatively much larger leaf from about the centre, which had its dorsal glaucous surface turned towards the axil of the leaf on which it grew ; and 

 from its base on the upper side, which no doubt was its axil, grew a small branch having several rudimentary leaves, thus becoming precisely like the terminal ramifications of a Ruscas. The 

 greatest difference between the Actiniadse and phanerogamous plants probably exists in the ovule and seed which may be compared to those of the Mammalia ;— if, therefore, a phanerogamous plant 

 had the spores of a Fungus it would show a decided analogy with the Actiniadae, each of the parts of the latter being more or less modified to suit the altered circumstances. A phanerogamous 

 plant in this view would be analogous to an Actinia, producing perfectly formed individuals from its ovaries instead of rudimentary bodies scarcely differing from the simplest kind of spores. 



Principles of the Arrangement. 



In attempting therefore to form divisions in phanerogamous plants new principles must be sought for, and in the arrangement now brought forward the foremost is that of tracing each 

 family to its nearest affinities, so as to demonstrate as far as possible of what family it is as it were a branch, and thus finally to form races of plants. In endeavouring to arrive at this desirable 

 result, all characters, and especially those of relative position, have been taken into consideration, and the inferences as regards the Class Exogens are : — 



1 . That Epigynous Exogens form a Natural Division, and that this character is the only one by which the Class Exogens can be divided into two parts, but that several 

 families which contain genera not epigynous must be admitted into the Epigynous Division. 



2. That the Divisions Hypogynous and Perigynous do not form natural assemblages, the latter requiring to be subdivided and arranged among the former in widely different 

 stations. 



3. That the characters Apetalous, Polypetalous, and Monopetalous are also of minor value, but that the Monopetalous families are without exception monopetalous 

 forms of the Polypetalous, and the latter are either directly or indirectly polypetalous forms of some one or other of the Apetalous; and that the foremost point in endeavouring 

 to form a Natural System of Exogens should be to determine these relations. It is already admitted by some botanists that the Apetalous families are not a distinct Class, but only 

 apetalous forms of the Polypetalous, or in other words, the Polypetalous are Apetalous families with the acquisition of petals; — this being conceded, the conclusion that the 

 Monopetalous are in their turn Polypetalous families with their petals united appears unavoidable, besides which, it is well borne out by practical observation and analysis. In 

 a limited sense the Apetalous families are dimorphous forms of the Polypetalous, and the latter dimorphous forms of the Monopetalous ; and, if this is admitted, the separation 

 of either of these three Classes from the other two would, it appears to me, render the formation of a Natural System connecting the whole impossible. 



4. That the Epigynous Division is less highly developed than the great mass of the Class Exogens, and should therefore constitute the first of a series of Divisions. 



* Some petioles have a perfect circle of woody fibre and a central pith, while in some foliaceous branches these are imperfectly marked, the broken circle of woody fibre exactly resembling the broken circle so common in petioles. Between 

 the venation of leaves and foliaceous branches there is no difference which can be depended on, as the venation of some leaves agrees with that of foliaceous branches, as that of Mercurialis annua with the species of Xylophylla, although it is usually 

 different. , 



