1905] OLSSON-SEFFER—PH YTOGEOGRAPHIC NOMENCLATURE 189 
and are necessary to distinguish. It has been suggested’ that the 
division of the vegetation into formations must be founded upon the 
concept of habitats. This principle is a good one, the only difficulty 
seeming to lie in the practical working of the rule. Any one conversant 
with the great variety of forms of habitat is aware that such a classi- 
fication is no easy undertaking. In all attempts made the authors 
have decided for one or another environmental factor that has influ- 
enced the development of the formation, and consequently the classi- 
fication has been more or less artificial. The task of identifying, 
classifying, and naming plant aggregations, or features of the vege- 
tation, is extremely difficult because of the comprehensive data 
necessary to illustrate the complex factors influencing distributional 
phenomena. The use of one class of names that refer to habitat, 
however, are inevitable and absolutely necessary. For such terms 
we can turn to the vernacular language, which often possesses very 
expressive names that combine in one word the main features of 
environment. Together with the term formation as representing 
the large topographical units of vegetation such vernacular terms are 
very adequate. Let me give a few examples to illustrate this. The 
chaparral formation of California and southwestern United States 
generally is one of the most: peculiar anywhere. There can be no 
doubt about the meaning and scope of the term, when the formation 
has once been clearly defined, because it has no counterpart in any 
other region of the world, although it certainly is paralleled in many 
places by related formations. The expression chaparral formation 
gives not only a general idea of the component plants, but it also 
includes a conception of the topographic aspect of the country where 
the formation occurs, and whose physiognomy it assists to mold. 
Still the term is strictly devoid of any reference to the dominant 
species concerned in the aggregation of plants in any part of the 
formation. Formational names should always be so. A formation 
can be subclassified into associations, and these into communities. 
The latter can be designated adequately by adding the suffix etwm to 
the scientific name of the dominant plant after the method first sug- 
gested by Hutr.® The limitations of this article do not allow me 
5 CLEMENTs, be G. 
® Férs6k till en analytisk behandling af vixtformationerna. Medd. Soc. F. Fl. F. 
8:1-155. 188; : 
. 
