22 
C. E. Moss. 
Grisebach’s formation was wider than Humboldt’s association, yet, 
from this date onwards, the term “ formation ” began to enter into 
direct competition with the term association. It is most interesting, 
however, to observe that the modern tendency is to retain the 
usage of the two great pioneers in so far as the term formation now 
denotes a vegetation unit more comprehensive than the association. 
Not only has the term association had its rivals, but various 
concepts, quite foreign to the prevailing one, have at various times 
been attached to it. Lecoq (1855, iv.: 90, et seq.) spoke of the 
seasonal appearances (the “aspects” of Clements, 1905: 315, etc.) 
as “ associations ”: thus, he referred to “ associations du mois 
d’avril,” “ associations vegetales du mois de juin,” and so on. Loew 
(1879: 592) regarded as an “association” those species which entered 
a country at the same time ; and Celakovsky (1869) is said to have 
used the term “ formation ” in this sense. Hock (1895: 227) also 
used “association” in the same way. Clements defined “asso¬ 
ciation” (1904: 9) as “the coming together and the staying 
together of individuals,” i.e., “ the arrangement of individuals in 
vegetation ” (1905 : 315). All these significations, however, have 
given way to the use of the term as a unit of vegetation of narrower 
limits than the formation, and characterized by minor differences 
in habitat and in floristic composition. 
Blasius (1840-1; cited in Grisebach, 1846: 133) described 
certain plant associations (sub nom. “ formations ”) in his account 
of Russian vegetation; for example, he spoke of a “ dwarf birch 
formation,” and a “ formation of reed grasses and Carices.” 
Many of the “formations” of Kernel’ (1863, etc.) are also 
either associations or groups of associations. 
Hult (1881, etc.) has described many associations [“ formations”] 
in the vegetation of north Finland, such as an Empetrum “formation,” 
a Nardus “formation,” a Betula nana “formation,” and a Scirpus 
cccspitosus “formation.” Drude (1889: 25) considers that Hult’s 
formations are too narrow, and terms them Bestande ; and 
Warming (1909: 140) also rightly urges that this narrow use 
of the term “ formation ” is not exemplary, and that “ it is better to 
speak of such smaller units as associations.” Some of Hult’s 
“ formations ” are perhaps only societies, e.g., his Poa annua 
“ formation ” (1881 : 53). 
Kernel' (1887-1891 ; English tr., 1895: 896) retained the term 
formation, although he regarded the selection of this name as not 
quite fortunate ; “ but, having been once introduced into the science, 
