IV.] SPECIES -MAKING. 135 



to one who attempts to describe species as so many enti - 

 ties which have distinct and personal attributes. So the 

 garden has always been the bugbear of the botanist. 

 Even our lamented Asa Gray declared that the modern 

 garden roses are ' ' too much mixed by crossing and 

 changed by variation to be subjects of botanical study." 

 He meant to say that the roses are too much modified to 

 allow of species -making. The despair of systematic 

 botanists is the proof of evolution ! 



I repeat that mere species - making, in the old or con- 

 states. It has been the fashion to throw these all together into a composite 

 si)ecies, calling it Carex echinata. In this arrangement, the sub-groups or swb- 

 forms do not stand out clearly, and it is impossible to contrast them forcibly. 

 Moreover, the characters which separate the most marked sub-forms are of as 

 great or even greater classifactory importance than characters which are used to 

 separate Carex echinata itself from its fellow species. The old arrangement might 

 be graphically presented as follows: 

 Carex echinata. 

 Group B. 



Sub-group a. 



Sub-group b. 



Sub-group c. 

 Group C. 

 Group D. 



Sub-group a. 

 This classification, from a taxonomic standpoint, is untrue, for, as Carex 

 species go, groups B, 0, D are coordinate witli C. echinata, and not subordinate to 

 it. The mere fact that there are now and then intermediate forms between 

 these various groups should not deprive us of the privilege of expressing the 

 taxonomic facts. In nearly every instance, specimens can be clearly referred to 

 one or the other of the groups by one who is familiar with them ; but so long as 

 the various groups are represented to be of minor and variable importance — as 

 the above arrangement does represent them to be, to a botanist's mind — so long 

 will they remain to be comparatively little distinguished and understood. Conse- 

 quently, I have erected (Bull. Torr. Bot. Club. xx. 422) the four groups into 

 coordinate species, as follows : 



A. Carex echinata. Old World. 



B. Carex tterilis. New World. 



a. variety excelsior. 



b. var. cephalantha. 



c. var. angustata. 



C. Carex Atlantica. 



X 



D. Carex interior. 



a. var. capillacea. 



