4 VARIATION IN ASTER 
Variation within the species. — Each species of Biotian Aster, 
the most variable group (and to a large degree this is true of the 
other groups also), is to be expected to develop quite a new aspect 
in any or all of the following classes of modifications : 
1. Branch-forms, plants which seek to replace a lost upper- 
stem by unusual branch-growth; involving not only a changed 
aspect by reason of the branching habit, but also many minor 
details of growth, particularly in leaf-form. 
2. Sprout-forms, which spring up when the whole original 
stem is lost, and are usually unbranched and just the opposite 
from the preceding in aspect. 
3. Arrest-forms, when sudden or gradual arrest of leaf-develop- 
ment or petiole-development, or of certain internodes or of certain 
leaf-types, may effectually disguise the relationship of the plant 
for a time. 
4. Dwarf and luxuriant, condensed and expanded, smooth and 
pubescent forms, and their intermediates ; dependent on nutrition, 
shade, etc. 
For all the above variations, which may be called regular 
irregularities, allowance must be made when in the field and when 
assigning value to specific characters. But I do not make allow- 
ance for these variations in keys or in descriptions of species, my 
plan being to describe the typical development of the species and 
then to add references to its variants. 
The large number of coincident leaf-forms simultaneously ex- 
isting on the normally developed plant must also be taken con- 
stantly into consideration. But a still greater source of misinter- 
pretation is the remarkable series of consecutive leaf-forms (soon to 
be described in detail) which are assumed in different years or in 
different relations to light, moisture and crowding. 
Variations between the species. — When the norm of a given 
species has been satisfactorily determined — and in order to do 
this I have felt it necessary to watch a given species in several 
localities during several years of practically unaltered natural con- 
ditions — there still remains beyond this norm, and sometimes in 
every direction, a number of outstanding forms which often estab- 
lish a perfectly graded connection to other neighboring species. 
But we will not deem a species nullified because nature has not 
yet extinguished all the light that could illuminate its descent. 
ERIS rie e EER de oi nini a o Dc dci aid 
