JOHNSTON: FLORA OF MARGARITA ISLAND. 295 



III order then to make an intelligent and thorough comparison of 

 the flora of ^Margarita and Coche with that of other regions about the 

 Caribbean Sea, it would be necessary for one to have access to speci- 

 mens of plants and lists of plants collected in all lands bordering the 

 Caribbean and also to have traveled in those regions. So far as the 

 lists of plants and the collections are concerned it must be said that 

 despite the many visits of American botanists to the West Indies and 

 tropical .America, and despite their intense activity in the United 

 States, there is yet very much to be known about the plants in those 

 regions. Parts of iSIexico and of Central America are being well 

 worked over. Colombia and Venezuela have had so little work done 

 on them that comparison of lists of their plants is almost valueless. 

 The West Indies as a whole have the useful works of Grisebach and 

 of Professor Urban but these contain no lists of plants of the individual 

 islands, so that they are scarcely to be used in comparative work. 

 A few of the American botanists frequent ^Mexico and parts of Central 

 America, a very few have visited Colombia, still fewer Venezuela, and 

 some frequent Cuba, Ha\1:i, Porto Rico, and Jamaica. Seldom is the 

 traveling extensive or in more than one region. Both plant and ani- 

 mal surveys of the United States are fairly thorough through many 

 parts, but in tropical America and the west Indies, biologically closely 

 related to our southern States, little such systematic work has been 

 done. 



In view of the situation as above discussed it is perhaps particularly 

 desirable that I make such a complete comparison of the flora of 

 Margarita and of Coche as may be possible with that of other regions, 

 especially considering that I have what I believe to be a complete 

 list of all plants ever collected ui or recorded from Venezuela and 

 also that I have been enabled to visit personally many parts of Vene- 

 zuela, British Guiana, and Panama, and many of the W'est Indian 

 islands. 



In a preceding chapter on the composition of the flora of Margarita, 

 it would seem that the vegetation of Margarita partook equally of the 

 nature of the W^est Indian and of the South American elements. 

 This appearance I believe to be due to the large proportion of cosmo- 

 politan plants present. In reality Margarita is, as would be expected, 

 distinctly South American in its flora as will appear in the following 

 pages. 



The two islands under discussion, Margarita and Coche, are the 



