THE COASTAL PLAIN OF ARKANSAS 



43 



This is evidently a very imperfect list, as shown by the large 

 number of interrogation points in it, but it was the best I could do 

 in a few hours without botanical literature or any facilities for 

 collecting specimens. Most of the species seen onl^' once are 

 omitted, as are those seen more than once which I was unable 

 to identify even generically, for lack of flowers or some other 

 reason. Notwithstanding the difficulty just mentioned, I recog- 

 nized more species of plants in this prairie in one day than I have 

 in the prairies of Long Island" in three summers. The flora is 

 evidently very rich, and it is reasonably certain that a careful 

 taxonomic study of it would reveal at least 150 species, a few 

 of them probably as yet undescribed. 



From what is already known it is evident that Compositae, 

 Leguminosae and Juncaceae are pretty well represented in pro- 

 portion to the total number of species in the prairie, or the size 

 of these families, or both. Gramineae and Cyperaceae are less 

 prominent, and Ericaceae appear to be entirely absent, as the}"" 

 are .from some other prairies. The very interesting geographical 

 affinities of this flora can be discussed much better after a more 

 complete and accurate list of plants is made. It might be re- 

 marked in passing, however, that some of the species occur also in 

 the prairies of Illinois, and some in the southeastern pine-barrens, 

 while others are pretty widely distributed in the eastern states. 



After crossing the White River near DeVall's Bluff the railroad 

 passes through about 12 miles of alluvial bottoms, partly culti- 

 vated, and then about 16 miles of prairie (from about Brinkley to 

 Palestine), more interrupted by oak groves and cultivated fields 

 than the Grand Prairie of Prairie County. In some places in 

 Monroe County east of Brinkley there was a good deal of Pinus 



^See Torreya 12: 277- 2S7, December, 1912. 



