RECOGNITION OF THE CHARACTER OF SOILS. 519 



age, at once reminds us that the Rhizobia do not flourish in 

 acid lands. The great prevalence of leguminous plants of all 

 kinds in the arid region clovers (not fewer than twenty-three 

 species in California alone), Lupins, Astragalus and related 

 genera, at once remind us of the universal prevalence of 

 calcareous soils in these regions, as shown above. Mutatis 

 mutandis, we find precisely the same general facts in the arid 

 regions of the other continents. 



Nevertheless, it must be kept in mind that not all plants of 

 the leguminous order are positively " calciphile." Within the 

 United States, it is especially the genera Desmodium (Mei- 

 bomia] and Lespcdcza, which are very numerously represented 

 in the long-leaf pine region of Mississippi, where the soils are 

 so poor in lime. Whether under these conditions these plants 

 develop the rhizobian nodules, has not, so far as the writer 

 is aware, been definitely observed. Certain it is that quite 

 a number of these plants occur on both calcareous and non- 

 calcareous soils, and on the latter assume a much more vigor- 

 ous development than in the pine woods. But it is evident that 

 they, with a few others (c. g. Galactia moll is, Cassia chamcc- 

 crista and nictitans) are more or less indifferent to the lime- 

 content of soils, and cannot therefore be relied upon in judg- 

 ing the quality of lands. In Mississippi and northern Ala- 

 bama, the Tcphrosia virginica (" devil's shoestring"), associ- 

 ated with chestnut and short-leaved pine, is characteristic of 

 the poorest non-calcareous lands, and bears seeds but very 

 scantily. It disappears so soon as calcareous lands are ap- 

 proached, together with the chestnut tree. 



EUROPEAN OBSERVATIONS AND VIEWS ON PLANT DISTRIBUTION 

 AND ITS CONTROLLING CAUSES. 



The writer has thus far presented and discussed mainly his 

 own observations made in the United States, without refer- 

 ence to the previous and contemporaneous work on the same 

 subject in Kurope. There arose certain discrepancies which 

 could not well be explained without a previous full consider- 

 ation of American conditions. 



As is well known, for nearly twenty years the accepted 

 theory in Kurope was that of Thurman. 1 which attributes the 



1 Essai de Fhytostatique appliquce a la chaine du Jura et aux contrccs voisines, 

 2 vols. Svo. Herne, 1849. 



