H. F. Wernham. 
37 & 
necessity, be the product of fusion in the course of descent. 1 In the 
writer’s opinion, however, this is scarcely probable in the case of 
any of the known Sympetalse. Apart from the abstract considerations 
which have just now been urged against such a probability, most of 
the component groups of this series are, as we have endeavoured 
to point out in these chapters, so clearly linked with the Archi- 
chlamydege and with each other in one evolutionary whole, that 
very few, if any, are left as matter for speculation from such a 
standpoint. Perhaps the only seriously isolated group is Plantaginales; 
but the floral organs of these plants, although combined in a peculiar 
way, reveal no fundamental differences from the corresponding 
organs in Archichlamydeae when considered separately. 
General Summary of the Sympetal/£. 
The accompanying diagram has been devised for the reader’s 
convenience ; it represents a portion of the evolutionary tree of the 
Vegetable Kingdom, namely, that portion, or most of it, which 
includes the branchings from the type of flower characterized by 
an elongated floral axis, and by indefiniteness and absence of 
cohesion in all the parts: this we have termed the Ranalian 
Common Ancestor. Proceeding from this, under the domination 
chiefly of the Economy Principle, is a line of descent which breaks 
ultimately into the three main lines leading respectively to the 
Geranial Plexus (Disciflorae of Bentham and Hooker), the Centro- 
spermal Group, and the Calycifloral Plexus. 2 It is from these three 
groups that the existing Sympetalse, according to our conclusions, 
have descended. 
The caryophylline branch of the Centrospermal Group, after 
the acquisition of a heterochlamydeous perianth and a unilocular 
ovary, gave rise to the Primulales, Plumbaginaceas corresponding 
to the uniovulate or chenopodial section of that branch. 
The general Calycifloral Plexus is exemplified in Rosales and 
Parietales (of Engler), and both display the transition from perigyny 
to epigyny. The Ebenales represent the sympetalous progeny of 
that section of Parietales in which the ovary is prevailingly superior 
(Guttiferales); and this progeny reflects the condition of indefinite¬ 
ness in numbers of the floral organs which is so marked a feature 
1 The Gnctaceaz possibly afford evidence of this. 
2 It is open to question whether Centrospermae arc derived from 
a ranalian floral type, but this does not concern us here ; 
it is with their descendants that we have to deal, not their 
ancestry. 
