18 



THE VEGETATION OF THE UNITED STATES. 



number of growth-forms recognized is 55. The number of growth- 

 forms apportioned to each of the various classes of plants is shown 

 by the following table: 



II. Aquatic plants (6) : 



Amphibious 3 



Submerged 2 



Floating 1 



III. Non-vascular plants H 



I. Terrestrial plants (38): 



Trees 7 



Shrubs 9 



Climbers 4 



Parasites and saprophytes 2 



Grasses 3 



Succulents 3 



Small perennials 7 



Annuals 3 



The importance for us of a carefully elaborated and natural system 

 of growth-forms such as that of Drude lies not so much in its details 

 as in the criteria on which it is based. Some of the gross anatomical 

 or physiognomic criteria are of profound and obvious physiological 

 importance, such as the major distinction between terrestrial and 

 aquatic plants, the distinction between perennials and annuals, and 

 that between succulent and non-succulent forms. Other criteria are 

 of known physiological importance, such as the distinction between 

 saprophytic, parasitic, and autonomous plants, or between the decidu- 

 ous and perennial habits of leaves. When, however, we approach such 

 distinctions as those between broad and narrow leaves, between pov- 

 erty and richness of branching, and between the possession of rhi- 

 zomes and that of bulbs, we are on extremely controversial ground. 



There is much evidence to indicate that the form and size of leaves 

 has been overestimated as a criterion of importance in the ecological 

 classification of plants. Paleobotanical evidence shows that many 

 unusual forms of leaf, such as those of Liquidamhar, Platanus, and 

 Artocarpus, have persisted through long periods of time. The fact 

 that these trees have undergone extensive migrations and recessions, 

 undoubtedly encountering substantial changes of environment, affords 

 some basis for a belief that leaf-form is often as conservative as the 

 structure of the floral organs^ The importance of mere leaf-size in 

 relation to water-loss has also been overestimated, as it has been shown 

 that the transpiring power of a leaf bears no invariable relation to its 

 size. This explains the existence, side by side in the deserts of southern 

 Arizona, of such plants as Franseria ambrosioides, with leaves from 25 

 to 40 sq. cm. in area, and such plants as Hymenoclea monogyra and 

 Baccharis emoryi, with leaves from 1 to 2 sq. cm. in area; or the con- 

 comitant occurrence, in relatively dry habitats in the mountains of 

 Jamaica, of Bocconia frutescens, with leaves often 200 sq. cm. in area, 

 and Micromeria obovata, with leaves less than 0.25 cm. in area. 



'For a discussion of this topic from the paleobotanical standpoint, see: Berry, Edward W., The 

 Lower Eocene Floras of Southeastern North America, U. S. Geol. Surv.. Professional Paper 91, 

 351 p., 1916 (p. 73). 



