NATTJKAL AND ARTIFICIAL STSTEM8. 365 



Pansy ; V. rostrata, a remarkably long-spurred species ; V. rotundi- 

 folia, with rounded leaves ; V. lanceolata, with lanceolate leaves ; 

 V.pedata, with pedately parted leaves ; V. primulcBfolia, where the 

 leaves are compared to those of the Primrose ; and V. puhescens, 

 with pubescent or hairy herbage. Sometimes the specific name re- 

 fers to the country which the plant inhabits or was first found in, as 

 Viola Canadensis, the Canadian Violet ; or to the station where it 

 naturally grows, as V. palustris (Marsh Violet). Sometimes it com- 

 memorates the discoverer or describer, when it rightly takes the 

 genitive form, as Viola Muhleribergii, V. NuttaUii, &c. When com- 

 memorative names are given merely in compliment to a botanist un- 

 connected with the discovery or histoiy of the plant, the adjective form 

 is preferred ; as, Oarex Torreyana, G. Hookeriana, &c. : but this rule 

 is not universally followed. Specific names are sometimes substantive ; 

 •as. Magnolia Umbrella, Ranuncidus Flammula, Hypericum Sarothra, 

 Linaria Cymhalaria, &c. (most of these being old generic names 

 used as specific) ; when they do not necessarily accord with the 

 genus in gender. These, as well as all specific names taken from 

 persons or countries, are to be written with a capital initial letter. 



713. Varieties may be designated by names when they are re- 

 markable enough to require it. The name of the variety, when 

 used at all, follows that of the species, and is formed on the same 

 plan. Subgenera need to be designated by names, which are sub- 

 stantive, and on the same principle as generic names. These are 

 convenient to refer to, but are not a part of the proper name of 

 a plant, which is that of the genus and species only. 



714. The names of genera and species are the same in all botani-. 

 cal systems, and therefore are properly alluded to here. But those 

 of orders, and all other groups higher than genera, vary in plan 

 with the system adopted. Classifications are of two sorts, viz. 



715. Natural and Artificial Systems. A natural system carries out 

 in practice as perfectly as possible the principles sketched in this 

 chapter, arranging all known species in groups of various grades in 

 view of their whole plan of structure, so placing each genus, tribe, 

 order, &c. next to those it most resembles in all respects. An arti- 

 ficial system arranges the genera by some one character, or set of 

 characters, chosen for convenience, disregarding other considerations- 

 It aims only to provide an easy mode of ascertaining the names of 

 plants, and does not attempt to express their points of resemblance- 

 generally, but serves nearly the same purpose as a dictionary. 



31* 



