582 



THE COMMUNITY 



may coincide with them. We shall attempt 

 to distinguish as ecological, rather than 

 historic in the geological sense, the post- 

 glacial period. This in fact coincides some- 

 what with the ordinary distinction of his- 

 toric time (in the familiar sense) from 

 geological time. Postglacial time has been 

 found, furthermore, on the evidence of the 

 development of endemic species, to agree 

 fairly well with the time required for spe- 

 cies difiFerentiation in vertebrates (Griscom, 

 1932). 



It is scarcely necessary to reject the terms 

 "animal formation" and "plant formation" 

 in favor of the biotic concepts supported in 

 the present work. It will nevertheless be 

 more than ever evident in the characteriza- 

 tion and discussion of the biomes that the 

 animal component is in many respects 

 secondary to the plant matrix and de- 

 pendent upon it. A more extended treat- 

 ment of the biomes, though under quite 

 difiFerent terminologv, is to be found in 

 Hesse, Allee, and Schmidt (1937), and 

 useful source books for North America are 

 Shelford (1926) and "The Biotic Provinces 

 of North America," by L. R. Dice (1943). 

 Otherwise, regional geographic works often 

 give better accounts of the biomes than do 

 ecological or zoological studies. The study 

 of the plant matrices of the biome, analyzed 

 under the term "vegetation," has long been 

 treated as a distinct department of eco- 

 logical botany and has a voluminous litera- 

 ture; e.g., Grisebach (1872), Schimper, 

 "Pflanzengeographie auf okologischer 

 Grundlag;e" (1898), and Engler and Drude 

 (1896-1928). 



We have attempted to answer the es- 

 sential question as to the geographic inclu- 

 siveness of the term "biome" by the concept 

 of biome-type. Much of the literature em- 

 ploying the term "biome," as is true equally 

 of the ecological "life zone," has avoided 

 this question by limiting itself to the con- 

 sideration of biomes in North America 

 alone. There is perhaps a tendency to re- 

 gard the biome as a Idnd of supercommu- 

 nity (Phillips, 1934-1935). The American 

 treatments of the subject have the validity of 

 recognizing the climatic veeetational 

 climaxes as the key to a geographic classi- 

 fication of the complex of serai stages, to- 

 gether with the correlated climax, into the 

 unified biome, as suggested in the principle 



of convergence (p. 575). When the biome 

 concept is still further broadened to include 

 the major biotic formations of the world, it 

 is necessary to combine with it the con- 

 ceptual geographic grid of historically de- 

 veloped biota characterized by regional 

 endemism. For example, to regard the 

 tropical forest of the Americas (with its 

 associated animal life) as a component of a 

 circumtropical or pantropical "tropical 

 forest biome" involves so extreme a tenuity 

 of connection in time that the biome con- 

 cept at this level becomes metaphysical. 

 The partition of the terrestrial world into 

 satisfactorily characterized biomes must 

 lean, therefore, upon historical biogeog- 

 raphy whenever major physiographic 

 separations are involved. 



The ecological distribution of terrestrial 

 communities on a world scale being essen- 

 tially climatic, and aflFected by physio- 

 graphic and certain rainfall boundaries that 

 dominate the east-west partition of conti- 

 nents and climates, as the temperature and 

 wind zones dominate their north-south 

 zonation, the terrestrial biomes are frag- 

 mented into geographic regions in low lati- 

 tudes and in the southern hemisphere, and 

 appear to be continuous only in the circum- 

 polar tundra and ice waste at the north. 



The major marine community, in spite 

 of its great regional biotic variation, is so 

 lacking in effective barriers to dispersal, is 

 so much subject to slow continuous circula- 

 tion of its medium, and exhibits so much 

 interdependence of its components from 

 region to region and area to area, that it 

 may be regarded as a single biome-type 

 (cf. pp. 595-597). 



The smaller fresh-water communities are 

 for the most part transient and are so 

 intimately related to their associated terres- 

 trial communities through succession (p. 

 572), and through the edaphon (p. 510) 

 that sharp segregation of fresh-water biomes 

 is scarcely practical. Even large lakes, such 

 as the North American Great Lakes, repre- 

 sent an edaphic climax on a large scale 

 rather than an independent biome. The ex- 

 istence of a few ancient fresh-water lakes 

 with regionally distinct faunas must be re- 

 garded as individual communities, and their 

 geographic relations may be relegated to 

 the province of historical biogeography 

 (Hesse. Allee, and Schmidt, 1937, p. 345). 



