III. FLORA OF BRITAIX. 403 



The first dozen orders in the above series are mostly 

 well-marked groups of plants, not merely conventional 

 combinations of genera. That they are such real groups 

 is in some degree evidenced by their having received 

 expressive names, not derived from particular genera, but 

 suggested by some obvious structural or physiognomical 

 character. Amentiferce and Conifera; are the only other 

 groups in the series which accord with them in this 

 respect. Rosacea and Caryophyllacece are the most de- 

 cided exceptions ; and the former group is perhaps mis- 

 placed among the highest, and also too forced or arbitrary 

 as an ordinal combination of plants. Indeed, by Lindley 

 and various other authorities this Rosaceous group is 

 divided into four distinct orders. And Caryophyllacem 

 are so imperfectly distinguished from allied orders, that 

 certain genera are tossed to and fro between them, or 

 even split into more genera (e. g. Sj^ergularia, cut from 

 Arena7-ia) for the very purpose of being placed under dif- 

 ferent orders ; — a rich illustration of the real conven- 

 tionality of these pretendedly "natural" orders. Though 

 latterly re -named after certain genera, ScrophulariacecB 

 and Lamiacece were long more familiar under the desig- 

 nations of Personate and Labiate. Whilst Cyperoides 

 and OrchiclacecB might equally well admit of descriptive 

 class-names, without resorting to those of genera which 

 imperfectly represent the groups. 



The dozen orders reckon up to 839 species. Throwing 

 out Rosacea', — as divisible into several orders of lower 

 grade, and also placed unduly high in the series, through 

 excessive subdivision of Riibus into segregate species, — 

 the other eleven orders will still count up to 757 species; 

 being upwards of one-half of the total flora, after making 

 some abatement also for spurious species of Hieracmm. 

 The remaining 77 orders have consequently less than half 



