576 A. E. Verrill — The Bermuda Islands. 



The remainder are as follows : — 



#S i irrtm ■icocetenuior'Liim. Button Peperomia magnolicefolia,* (=-P. 

 Weed. Annual. obtusifolia in Lef.) A succu- 



A ren a r ia a Is h i oides . Willd. 

 Callicarpaferruginea. " Turkey- 

 berry." 



lent-leaved plant. Fig. 35. 

 Statice Lefroyi Heinsl. =(S. Caro- 

 liniana Lef.) Sea Lavender. 

 (Endemic.) 



The last named species and two of the ferns appear to be endemic. 

 It is not improbable that the fyomoea and the Sicyos were both intro- 

 duced by man, though there is no evidence of this. Lefroy consid- 

 ered them native. The White Jasmine (Jasminum gracile), which 

 thickly covers the rocks and drapes the cedars to their very tops, in 

 parts of this tract, is known to have been introduced there about a 

 century ago, but it has not spread elsewhere to any marked extent. 

 Hemsley also classed the Passifiora as an introduced species, but 

 without any particular evidence. 



A few species were found only on Boaz Island, about twenty-five 

 years ago, in a place that had not then been much disturbed. 

 Whether they still exist there is not known. Several species, mostly 

 ferns, are not known to occur except in some particular spots in some 

 of the marshes, especially in Pembroke Marsh. Many of these very 

 localized species will probably disappear before many years. One 

 endemic species of sedge (Carex Bermudiana Hemsl.) is known only 

 from specimens collected, about 1099, by John Dickinson (Sloane 

 Herb.). It may now be extinct. 



Among the plants entirely restricted to certain marshes, or nearly 

 so, or to other limited localities, are the following : — 



Waltheria Americana L. Pembroke Marsh. A shrubby, downy 

 plant 2-3 feet high, with clusters of small yellow flowers. In all 

 tropical countries. 



* Hemsley states that the correct name of the Bermuda plant is somewhat 

 uncertain. My photograph shows that it does not agree well with the descrip- 

 tions of P. magnolicefolia and P. amplexicaulis (considered varieties of one 

 species by him and others), for both these West Indian forms are described as 

 having sessile or subsessile leaves, while they are petiolate in the Bermuda plant. 

 A specimen labelled as P. magnolicefolia, from southern Florida, in the Eaton 

 herbarium, agrees in this respect with the Bermuda plant. Some of the several 

 Cuban forms of P. obtusifolia, in the same herbarium, which 1 have studied, 

 have petiolate leaves of the same form as those of the Bermuda variety. There- 

 fore the latter may rather belong to a variety of P. obtusifolia, if this be really 

 a distinct species. But in that case the Florida form probably belongs to the 

 same variety. 



