24 
C. E. Moss. 
Betula, and of Calluita vulgaris. These were added to in his later 
memoirs (1900 a and b, 1905). 
Schimper (1898; English tr., 1903-4) regarded “formations” 
either as “ climatic formations ” or “ edaphic formations.” Many 
of the latter, e,g., the Pes-caprae “ formation,” are associations (cf. 
Warming, 1909: 227, etc.). 
Cowles (1899) followed Warming (1895) in rejecting the term 
“formation,” and used the term “society.” Cowles (1899: 111) 
defined his “ society ” as “ a group of plants living together in a 
common habitat and subjected to similar life-conditions. The term 
is taken to be the English equivalent of Warming’s Plantesamfund , 
translated into the German Pflanzenverein. The term formation, 
as used by Drude and others, is more comprehensive, in so far as 
it is not synonymous.” It will be seen that this definition does not 
clearly distinguish between the association and the formation; but, 
on the whole, the “societies” described by Cowles (1899, 1901) are 
associations. The latter term has, however, recently been adopted 
by Cowles (1909, 1910). 
It may be remarked that in Britain it is usual to employ the 
term community for a vegetation unit whose rank is not specified, 
like Warming’s Plantesamfund, as in the sub-title—“an introduction 
to . . . plant-communities ”—of Warming’s new book (1909). 
In the use of the term “society,” Cowles has had a considerable 
following in America, among whom may be mentioned Livingston 
(1901, 1903), Whitford (1901), Brown (1905), and Transeau (1905-6). 
This is natural, as Cowles shares with Pound and Clements the 
great honour of being one of the pioneers of American ecological 
plant-geography. 
Smith and Moss (1903), Smith and Rankin (1903), Lewis (1904), 
and Smith and Smith (1904-5) used the term association very 
cautiously, and the term formation scarcely at all. 
Ganong(1903: 349, et seq.) divided the salt marsh formation 
of the Bay of Fundy into three associations, namely, a Spartinetum, 
a Salicornietum, and a Staticetum. Ganong’s concepts were clear, 
and his terms unequivocal. 
Cajander (1903: 24) regarded as comprised within the same 
association all those Bestdnde in which the same species is dominant; 
thus, all pine Bestdnde would belong to the same association. It is 
perhaps doubtful if this view is quite correct; for certain communities 
which are characterized by the same dominant species sometimes 
occur on different habitats, and are accompanied by very different 
