KECAPITULATION OF THE DOMINANT FEATURES OF THE FLOEA. 307 



which concerns us. In his own words : — " The fourth large province stretches from 

 the southern boundaries of the Montana [Rocky Mountains] and Californian floral 

 provinces, and from the very ill-defined western limits of the Atlantic States, in about 

 93° W. long., southward to the mouth of the Rio del Norte and away over the interior 

 highlands to the Sierra Madre in Mexico, where it terminates in a point a little within 

 the tropics, leaving the Atlantic and Pacific coasts up to the 27th parallel in the tropical 

 province " *. This he names the province of Texas and North Mexico. The remainder 

 of our territory comes within Drude's ' floral kingdom ' of tropical America generally, 

 though excluding the high Andes, and constitutes a province which he designates the 

 Mexican. This province is nearly conterminous with Grisebach's ; but it also includes 

 Yucatan and the Atlantic side of Central America. 



It is not our intention to discuss the merits of the divisions proposed by the authors 

 cited. Collectively they contain most of the ideas of our own, which were suggested 

 by a study of all the available material. 



It cannot be denied, however, that the divisions adopted in this work, as indicated 

 on the map at least, are also, to some extent, unsatisfactory and open to criticism, 

 particularly as the northward extensions of tropical types in the coast regions, and the 

 southward extensions of xerophilous types in the dry upland regions, are not shown. 

 But these are defects of detail, largely due to imperfect data, and after all very slightly 

 affecting any general conclusions, as evidenced by the synopsis of facts below. 



The country investigated in this work extends through 24° of latitude, the northern 

 limit being the thirty-third parallel on the western and the twenty-sixth on the eastern 

 side, with an altitudinal range of about 17,000 feet within the tropics. 



In the Enumeration habitats are recorded for the plants according to the political 

 divisions of the country f, and they are tabulated on the same basis, with an ultimate 

 reduction to three floral provinces — a northern, a central, and a southern. Unfortu- 

 nately a large number of the Mexican plants in herbaria are so vaguely labelled that it 

 is only from a knowledge of the itineraries of collectors that it has been possible to 

 assign them to their respective provinces ; but in some instances information on this 

 point is wanting. In all calculations the Mexican plants of uncertain origin are added 

 to the South-Mexican flora ; but it was perhaps a little precipitate to assume that they 

 all belonged to this province. Not that this course has unduly augmented the number 

 of species inhabiting South Mexico, though the North-Mexican total may have thereby 

 suffered a trifling reduction. Succulent plants, such as the Cactaceae and those of the 

 Agave and Yucca type, are largely known only from cultivated specimens, of Mexican 



* Ibid. p. 63. 



t We have very few plants from Yucatan and British Honduras, almost nothing from Spanish Honduras, 

 and although the general character of the vegetation of Nicaragua, Costa Rica, and Panama is known, it is 

 far from having been exhaustively investigated. 



biol. centr.-amek., Bot. Vol. IV., December 1887. 2 s 



