INTRODUCTION. 27 



These considerations force us to realize that the most commonly 

 used and most natural subdivisions of vegetation are based upon cri- 

 teria which have to do with the relations of the communities to water- 

 supply and water-loss. It is quite true that the water-relations of 

 plants have more to do with the control of the local and general dis- 

 tribution of vegetation than have any other conditions. This is not 

 true of the local and general distribution of the species themselves, for 

 we here find temperature relations plajdng a strong role. For the pur- 

 poses of our investigation into the correlations existing between vege- 

 tation and climate it is therefore significant that we are under the 

 necessity of using a classification of vegetation which rests so largely 

 upon a basis of the water-relations of plants. We might foresee from 

 this fact that strong correlations would be discovered between vegeta- 

 tion and water conditions and weak correlations between vegetation 

 and temperature conditions. 



In spite of the efforts of Schimper and of others to give the tempera- 

 ture-relations of plants a place in vegetational distinctions, by the 

 recognition of microtherms, mesotherms, and megatherms, we are as 

 yet unable to place a given species in its proper thermal category with- 

 out possessing facts which are still lacking for all but a very few plants. 

 We can not tell a megathermous plant from a microthermous plant 

 when we see them growing side by side, and it follows that we can not 

 go into the field in Georgia or Texas and pick out the plants in each 

 habitat which have great or small temperature requirements, as we 

 can rather satisfactorily distinguish those of great or small water 

 requirement. These categories are consequently of no present use in 

 delimiting vegetational areas. 



The system of life-zones w^hich was worked out by Merriam^ for the 

 United States, and has been elaborated by members of the United 

 States Biological Survey, constitutes an" effort to delimit biological 

 areas primarily with respect to the influence of temperature. With 

 such a classification of biological areas in hand it is not possible, how- 

 ever, to make an impartial effort to determine which of several climatic 

 factors is primarily concerned in conditioning the existence and limits 

 of the areas.- This attempt would indicate a very close correlation of 

 biological areas and certain temperature conditions, because the 

 climatic maps showing these temperature conditions were used as a 

 basis in the original form of the life-zone map. 



If we were to make a map of the mean rainfall of the growing-season 

 in the United States it would be found to possess certain isohyetal 

 lines which corresponded closely with the distribution of certain plants 

 or vegetations. If we were, then, to modify the rainfall map in slight 



^Merriam, C. Hart. Laws of temperature control of the geographic distribution of terrestrial 

 animals and plants, Nat. Geog. Mag., 6:229-238, 3 maps, 1894. — Life zones and crop zones of 

 the United States. U. S. Dept. Agric, Biol. Surv. Bull. 10, 1898. 



