460 : BOTANICAL GAZETTE | DECEMBER 
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 dis- 
tinguished and understood. Consequently I have erected (Bull. Torr. Bot. 
Club 20: 422) the four groups into coordinate species, as follows : 
A. Carex echinata. Old World. 
B. Carex sterilis. New World. 
a. var. excelsior. 
b. var. cephalantha. 
é¢. var. angustata. 
C. Carex Atlantica. 
D. Carex interior. 
a. var. capillacea.”’ 
II. THe Hysripiry Coro_Ltary.—Intermediate forms are 
the most confusing and disturbing elements in species-divisions. 
These forms may be (a) normal intergradients, (4) unusual or 
sportive aberrations, or (c) hybrids. The marks of hybrids are 
generally inconstant and evasive, and yet it is of the utmost 
importance to perspicuous taxonomy that one shall be able 0 
determine hybrids from normal variations. There are certain 
general evidences of hybridity which the student may apply 
with very satisfactory results to intermediates of which he sus- 
pects a hybrid origin. I have been in the habit of giving seven 
categories of tests to my students. These have been suggested 
mostly by a study of known hybrids in domestic plants. The 
student must be cautioned that the satisfying of any one of these 
tests is not a proof of hybrid origin, but if the suspected forms 
answer more or less closely to three or more of them hybridity 
generally may be inferred with some confidence. These tests 
are as follows: 
A. Evidences of variation. 
I. Intermediateness of characters between any two species 
is an evidence of hybrid origin, and this evidence is the greater 
the more unvariable the suspected parents normally are. 
2. Variation or gradation towards one or two related spe 
cies arouses a very strong suspicion of hybrid origin, and the evi- 
