242 MERRILL. 



apex. Corolla-tube very slender, 2.5 em long, about 1 nnu in diameter, 

 the aiitheriferous portion near the apex slightly inflated and about 2 mm 

 in diameter, this portion also densely bearded iiisidu, Uie outside with 

 scattered, short hairs, the lobes linear-lanceolate, !» to 10 nnu long, 2 mm 

 wide at the base, gradually narrowed upward to the slendci'ly acuminate 

 apex, more or less pubescent. Anthers 4 to .j mm long. Style 4 to 5 mm 

 long, cleft one-third to one-half into two arms. Tlie persistent leaf-like, 

 accrescent calyx-lobe is white, elliptic-ovate, ovate, or oblong-ovate, 

 membranaceous, 6 to 9 cm long, 3 to 5.5 cm wide, 5- to 7-nerved from 

 the base, the nerves more or less hirsute, the reticulations lax, the apex 

 shortly and sliarply acuminate, the base acute or rounded, the stipe about 

 2 cjii long. Fruit fleshy, narrowly obovoid. about 1..'^ mm long, black 

 when dry, with few, long, scattered hairs, the ealy.x-lobcs, otliov than the 

 accrescent one, early deciduous. 



Negros, Faraon, For. Bvr. 11/358 (type), 17350 Curran, September, 1909, and 

 near the same locality, For. Bur. 5539 Everett, September, 1906, For. Bur. 5218 

 Danao <C- Aspillera, June, 1906, in thickets, stream depressions, etc., at low 

 altitudes, locally known as aghoy. 



A species well cliaracterized by its long, slender, corolla-tubes which are white 

 instead of yellow, (lifl'ering in these characters from all Philippine species known 

 to me. 



Mussaenda philippica A. tJicli. in Mem. Soc. Hist. Nat. Paris 5 (1834) 245. 



Calycujjhj/llum grandifiorum Meyen Reise 2: 234; Walp. in Nov. Act. Acad. Nat. 

 Cur. 19 (1843) Suppl. 1: 356. 



Mussae7ida grandiflora Rolfe in Journ. Linn. Soc. Bot. 21 (1S84) 311: Vid. 

 Rev. PI. Vase. Filip. (1886) 152, non Benth. (1849). 



Mussaenda frondosa Auct. Philip., non Linn. 



There are at present about 60 sheets in the herbarium of the Bureau of Science 

 that are apparently referable to a single variable species, or perhaps to several 

 closely allied ones. The material has been identified at various times, some 

 specimens as Mus.-iaenda frondoaa Linn., some as M. grandiflora Rolfe, and some as 

 M. glabra Vahl. Rolfe in 1884 stated that he had seen no Mussaenda frondosa 

 Linn., from the Philippines, and I can only agree with liim in considering that 

 the typical form of Linneus' species does not e.xtend to the Archipelago. The 

 type of Mussaenda frondosa Linn., was from Ceylon, and the Director of the 

 Botanic Garden at Peradeniya has kindly supplied me with a full series of 

 specimens representing the Ceylon plant, presumably some of which are typical 

 Mussaenda frondosa Linn. None of tliis Ceylon material matches any of our 

 Philippine specimens. Mr. Rolfe transfen-ed CalycophyUum grandifiorum Meyen 

 to Mussaenda, to supply a specific name for the Philippine plant, but overlooked 

 the fact that the above specific name was invalidated in Mussaenda by the earlier 

 M. grandiflora Benth. 1 have accordingly adopted for the Philippine species the 

 name .\his.taenda philippica A. Rich., which was based on material secured in the 

 Philippines by Perrottet. Whether or not it is the oldest valid name for the 

 species, I am unable to determine at the present time; some of our Philippine 

 specimens apparently clo.sely match some Javan and Caroline 'Islands material 

 distributed as .1/. frondosa Linn., and M. glabra Vahl. Mussaenda philippica 

 A. Rich, is an erect .shrub or small tree, not at all scandent, and as at present 

 interpreted, extends from northern Luzon to southern Mindanao, and will probably 

 be fuuiiil fo extend to ofher parts of M;ilny;\. 



