392 CORRELATION OF DISTRIBUTIONAL FEATURES. 



an abundant and well-distributed set of climatic data. The State of 

 Texas is particularly well suited to such an investigation, both by reason 

 of its sharp gradients of vegetational and climatic conditions and by 

 reason of the absence of mountains and their complicating effects. 



While many distributional limits are sharp, the great majority of 

 plants pass gradually from abundance to rarity, sometimes over a 

 wide belt of territory. It is often possible by purely observational 

 methods to discover that the outermost localities for plants, on the 

 edges of their ranges, present conditions which are rare in that region 

 but are common or universal in the region in which this plant is abun- 

 dant. All of the deciduous trees of the Mississippi Valley find their 

 western limits along the watercourses of the Great Plains, and all of 

 the cacti of the Arizona Succulent Desert find their northern limits 

 on arid slopes of southern exposure along the southern edge of the 

 Colorado Plateau. These cases make it difficult to correlate vegeta- 

 tion and climate on a coarse scale, and they invite the use of more 

 refined methods with reference to small areas. 



The study of distributional limits and the climatic extremes by 

 which they appear to be controlled is only one phase of the correla- 

 tion of climate and vegetation. An equally important field is that in 

 which attention is given to the ecological centers of the distribution of 

 plants or the development of vegetations, and to the optimal condi- 

 tions which appear to determine the location of these centers. Faunal 

 naturalists have long been interested in the subject of ecological 

 centers, and numerous writers have proposed criteria by which they 

 may be recognized. Adams^ has brought these together, the criteria 

 which are apphcable to plants being as follows: (a) location of greatest 

 differentiation, (6) location of greatest individual abundance, (c) 

 location of closely allied forms, (d) location of maximum size of indi- 

 viduals, (e) ocation of greatest reproductive activity, (/) location of 

 convergence of lines of dispersal, {g) location of greatest catholicity of 

 habitat, (h) location of convergence of Hnes of individual variation. To 

 these might be added, for plants, the location of most rapid rate of 

 growth and the location in which a form is accompanied by the largest 

 number of individuals which are specifically distinct but of the same 

 growth-form. 



It is obvious that the locating of the ecological center for a plant 

 or a vegetation is a matter which requires the assembling of a considera- 

 ble body of data. There are cases in which one or two of these cri- 

 teria have been very carefully determined, but we know of no case in 

 which all of them have been determined for any plant. If it had been 

 possible to secure such data in the pubhshed literature, it would have 

 been an easy and fruitful task to have appHed our climatological 

 figures to such centers for the sake of lear ning the optimum climatic 



lAdams, C. C, Southeastern United States as a center of geographical distribution of flora 

 and fauna, Biol. Bull. 3: 115-131, 1902. 



