16 THE VEGETATION OF THE UNITED STATES. 



categories in which we can place the rich variety of types that the 

 plant organism has assumed. They constitute the beginnings of an 

 ecological classification of plants from a physiological standpoint. 

 Everyone must be aware that such a classification should have its 

 beginnings in physiological work and not in the descriptive work of 

 plant geography. It is not strange, however, that need for it should 

 arise in geographical work and should be felt more by plant geographers 

 and ecologists than by- most physiologists. 



In a brief review of the attempts that have thus far been made to 

 establish systems of growth-forms that will be of service in plant 

 geography, it will suffice to mention only a few. The first was proposed 

 by Humboldt^ in 1805, in connection with his effort to determine the 

 features that give distinctive character to the vegetation of different 

 altitudes in tropical America. Humboldt saw, in the types which he 

 recognized, the distinctive vegetational units that serve to bring about 

 the physiognomic diversity of the different regions of the earth, 

 rather than groups of possible physiological affinity. His list of 19 

 types included the coniferous tree, the palm, the cactus, the tamarind- 

 like tree, grasses, aroids, and the like. Grisebach" described 60 vege- 

 tative forms, and his classification, like that of Humboldt, had to do 

 largely with the conspicuous types of plants which determine the 

 physiognomy of vegetation and aid in differentiating the great floral 

 regions of the earth. Following upon these early classifications have 

 come the systems of Drude,^ Krause,^ Pound and Clements,^ Raun- 

 kiar,^ and Warming.'^ The systems proposed by these men are far 

 more elaborate than the earlier ones; they embrace all cryptogamic as 

 well as phanerogamic plants; they include aquatics as well as land 

 plants; they take into account seasonal behavior as well as form and 

 differentiation, and, what is best of all in an attempt to devise a natural 

 system, they introduce subordinate categories. 



The system of growth-forms most widely used at the present time, 

 and the one that seems to have attracted the most attention to this 

 subject, is that proposed by Raunkiar. His system is based enth-ely 

 on the character of the perennating organs of plants and their position 

 with respect to the substratum. His five groups are as follows: 



'Humboldt, Alexander von., Essai sur la Geographie des Plantes, Paris, 1805. 



^Grisebach, A. R. H., Die Vegetation der Erde, Leipzig, 1872. 



^Drude, O., Deutschlands Pflanzengeographie, 1896, and earlier papers. 



^Krause, E. H. L., Die Eintheilung der Pfianzen nach ihrer Dauer., Ber. d. deut. Bot. Ges. 9: 

 233-237, 1891. 



*Pound, R., and F. E. Clements, The phytogeography of Nebraska, Lincoln, 1898. 



«Raunkiar, C. Types biologiques pour la geographie botanique. Bull. Acad. Roy. Sc. Dane- 

 mark, Copenhague, 1905. — Livsformernes Statistik som Grundlag for biologisk Plantegeografi. 

 Botan. Tidssk., 29, Kjobenhavn, 1908 (translation in Beih. Bot. Centralbl. 87, 1910). — Formations- 

 undcrsogelse og Formationsstatistik. Botan. Tidssk., 30, Kjobenhavn, 1909 (English abstract 

 in Bot. Centralbl. 113:662, 1910). 



'Warming, E., Om Planterigets Liv.sformer, Festekr. ugd. af Univeraitet, Kjobenhavn, 1908. 



