16 FLORA OF ST. CROIX AND THE VIRGIN ISLANDS. 



after tlie separation of St. Croix from the latter, and immigration would 

 finally have ceased by the separation between them, as it exists at the 

 present period. Tims, the plants found in the Yirj'in Islands, but not 

 in St. Croix, would seem to have been more recently created in the 

 probable centre of vegetation, Porto Itico, or some other of the larger An- 

 tilles; the enilt'inic ones, as in the other islands also, being the youngest 

 of all, not having been formed till after the complete separation between 

 the islands had been efiected. This latter suggestion, which perhaps 

 seems contradictory to the general accepted theory of considering the 

 endemic forms on oceanic isles as the remnants of the oldest original 

 vegetation,* appears to be confirmed by the fact that even on such 

 recent formations as the Bahamas, which have as yet been but imper- 

 fectly explored, already no less than eighteen endemic species have been 

 discovered.t 



The supposition that the islands may have been separated from the 

 beginning, and have received their floras through immigration over the 

 sea, is sullieiently confuted, partly by the great number of species com- 

 mon to them all, which clearly indicates the connection in fomier times 

 with a larger country, partly by the circumstance that most of the spe- 

 cies common to the islands are in no way better adapted for migration 

 over the water than those jjcculiar to the Virgin Islands only ; in fact, 

 but few of them apparently possess the faculty of crossing salt-w^ater 

 even for a limited distance. 



Sui>posing the theory of a prolonged or oftener repeate<I connection 

 between Porto Pico and the Virgin Islands to be correct, it remains 

 still to explain how St. Croix can have obtained a number of species 

 which do not occur in the latter grou]). A few of these species, viz, 

 Castela erecia, Maytemis ehrodenclroides^ Zizuphus reticulatus, Anthacan- 

 thus jamai('C7hsu, and Buxns Vahlii^ occur in St. Croix on the tertiary 

 limestone only, and seem thus to have avoided the Virgin Islands as 

 not tinding tliere tin* substratum suited to their organisaticm. The 

 greater i)art, however, might, for all apparent reasons, as well occur in 

 the Virgin group as in St. Croix, and their absence in the former cannot 

 be exi>lained in this way. It must, however, be un<lerstood that wliilst 

 my iiivestigalitin of St. Croix has been tliorough, and carried on for 

 several years, my e\i»Ioiatiou «»f the N'iigiu Islands has been so for only 

 a jiart of them, especially the Danish ones, my collections from the 



• Hooker : On Iiisiilar Floras. 



t Griseb.': Geogr. Verbr. dor Pll. Wcstindieus, p. 55. 



