582 CONCLUSION. 



activities of individual plants. The attack upon this problem must, 

 however, be made by methods quite different from those employed in 

 purely physiological investigations. The conditions must be measured 

 rather than controlled, and the plant material must be examined 

 throughout its range of occurrence, much as a large series of experi- 

 mental cultures is scrutinized for the discovery of the effect produced 

 by controlled conditions. The methods that must be employed hinge 

 very largely upon the interpretation of a vast series of uncontrolled 

 experiments under the varying conditions of natural environment. 

 It is to the geographical aspects of the problem that we must ascribe 

 many of its complexities and much of its difficult nature. 



Although the results secured in this investigation are only general in 

 their applicability, we have endeavored to develop and make use of 

 methods which are specific and definite enough to warrant more 

 extended use. The basic data, both as to climate and vegetation, are 

 scanty in many cases, and the methods used could well be employed to 

 greater advantage with fuller data, or for the investigation of similar 

 problems in smaller areas. , 



The presentation of vegetational data that has been given takes no 

 account of the minor plant communities that occupy relatively small 

 areas in all plant formations, and owe their existence to the modifica- 

 tion of the fundamental conditions of climate through differences in 

 what might be designated as the response of soils to the climatic condi- 

 tions. No account has been taken of the developmental changes of 

 vegetation in regions with rapidly shifting topography, since these 

 changes depend mainly on differences in the character of the soil, or 

 on changes of environment due to the plant covering. All develop- 

 mental changes in vegetation are due to changes of environmental con- 

 ditions, and it is only rarely, or over very long periods of time, that 

 these changes are in the nature of definite alterations of the general 

 climate of the region. 



Our presentation of cHmatic conditions has been limited chiefly to 

 those elements of the climate that are commonly measured, as it is 

 impossible to secure well-distributed series of data for other conditions, 

 although many of these are well known to be of great importance to 

 plants. A departure from the customary climatological procedure 

 has been made in securing many of the data on temperature and mois- 

 ture conditions for the period of the average frostless season as well as 

 for the calendar year. The length of the average frostless season has 

 been determined for each of the stations from which other climatic data 

 have been used. The data for the summation of temperature have 

 been worked out by three methods, in addition to the one used by 

 Merriam. It seemed advisable to give this promising means of secur- 

 ing additive temperature data a thorough test, and to attempt to arrive 

 at a method with fewer objections from a physiological standpoint than 

 could be urged against the older methods. 



