226 THE JOURNAL OF BOTANY 



P. Magclalena determined, but nowhere described it." Stephani's 

 description of P. dominicensis Tajl. is that of P. Magdalena 

 Gottsche, and I fail to see where it agrees with Taylor & Gottsche's 

 description. 



From P. dominicensis it differs in that the leaves are very de- 

 current ; in P. dominicensis they are not so. Insertion more erect, 

 patent-divergent to patent 70° to 50°, in P. dominicensis they are 

 horizontal to patent-divergent 90°-70° ; apex obtusate or subtruncate, 

 not rounded ; leaves more triangular with apex narrower ; postical 

 margin more ampliate, projecting beyond the stem, in P. dominicensis 

 to the middle of the stem only ; the cells are also rather larger with 

 thinner walls. 



Gottsche, who knew P. dominicensis, evidently considered our 

 plant distinct from it, or he would have referred I'Herminier's plant 

 to it. 



P. dominicensis Tayl., Dominica (Elliott), named by Spruce, 

 agrees exactly with the original, specimens of which are in the 

 Manchester Museum (ex herb. Greville). P. superha Nees, is a 

 much larger plant, with leaves of a different shape, margin ciliate. 



Dimensions. Stems 5 mm., with leaves 7*5 mm., leaves 4 mm. 

 long X 2-0 mm. broad at base, 1 mm. at apex, cells -06 mm. X 0*5 mm., 

 •OG mm. X '04 inni. 



Hah. Mountain Lake, Dominica, Jan. 1896. 



P. BAEBADEXSIS St. Sp. Hep. p. 563. 



The specimens of this species in Miss Armitage's Collection were 

 named by Stephani P. Guilleniiniana Mont. In his Sp. Hep. he 

 describes P. Guilleminiana with leaves oblongo-trigonous, apex 4 times 

 narrower than the base. In Miss Armitage's si^ecimens they are 

 semi-ovate, apex twice as narrow as at thg base ; he must have later 

 recognised its distinctness and published it as a new species, P. har- 

 hadensis, Barbados (Armitage). 



There are no specimens of P. Guilleminiana Mont, in the Man- 

 chester Museum ; a specimen so named, probably by Austin, from 

 Sandwich Islands (Baldwin 201), has nothing in common with the 

 description. The leaves are patent, oblong-trigonous to oblong- 

 subquadrate, apex acute or obtuse, antical margin very decurrent, 

 much recurved, entire, postical margin extending to the middle of 

 the stem, teeth few, irregular. 



The species is not recorded by Stephani nor Spruce from the 

 West Indies. 



Stephani (Sp. Hep. pp. 524-5) reduces two of Spruce's species 

 — P. oreocharis and P. rhizophila (Spruce Hep. Am. And. pp. 495, 

 49S) — to synonyms of P. Guilleminiana ; both differ from Linden- 

 berg's descriptions and figures. P. oreocharis is a very distinct 

 species ; the contorted, longly decurrent postical margin of leaf alone 

 distinguishes it from any other. P. rhizophila has semi-cordato-ovate 

 leaves, which are remarkably ampliate at their postical base. 



Stephani's list of 966 species of Flagiochilce may possibly require 

 reducing: but as Spruce took infinite trouble and time in studying 

 the species he named, I fear Stephani does not give him due credit . 



