INTERMEDIATE FORMS 2 
Aster species must usually be limited by arbitrary lines. Type 
specimens in As¢er are often remarkably distinct from each other ; 
one inspecting an herbarium with but few representatives of each 
species would gain no conception of the actual connections between 
the species still existing in nature, made by thousands of interme- 
diate plants, usually capable of such arrangement in gradual 
sequence as to show a remarkable transition between species, with 
no satisfactory dividing line. Taking asingle sequence from the 
long-established species of Gray’s Synoptical Flora, A. macrophyl- 
lus is thus gradually connected with A. Herveyi, A. Herveyi with 
A. spectabilis, A. spectabilis with A. surculosus, both these with A. 
gracilis, and all three with 4. radula. 
In short, of the long-known northeastern species, there is 
hardly one which does not afford intermediate forms to some other 
within its own region, unless we except A. Novae-Angliae, A. 
ptarmicoides, and A. subulatus ; it happens that these are species 
which by certain botanists are, on other grounds, separated from 
Aster. Doubtless when the southern and western species which 
now seem more sharply defined shall have been studied equally 
. thoroughly in the field, similar connecting forms will be found 
between many of those also. Continuous experience in the field 
over wide areas, and continued observation of the same localities 
through a series of years, confirm me in the conclusion that a 
sharply defined species of Aster is likely to remain an exception. 
Intermediate forms do not invalidate species in Aster, Should we 
admit that they do we must accept a series of reductions which 
would not end till the principal bulk of the whole vast genus had 
been reduced to one species. For instance, among narrow-leaved 
species, Gray's group the Æricoidei included his A. Porteri, A. 
polyphyllus and A. ericoides. Because of essential similarity and 
multitudes of closely connecting variants, the advocate of reduc- 
tion may unite all these into the one species, A. ericoides. If he 
seeks for connecting forms in Gray's next group, the Mu/tiffor:, he 
will find them, and he might similarly unite these into one species, 
A. multiflorus. But between this and A. ericoides numerous addi- 
tional connecting forms exist, so another coalescence would follow. 
The next group, Gray's Divergentes, including A. dumosus, vimi- 
meus, lateriflorus, etc., by the same plausible reasoning might be 
