9 



Folia 4 poll longa ad 2^ poll. lata. Calyx vix 2 lin. longus. Petala 

 1^ lin. longa. Antherai l"lin. longye. 



3. P. cuneata, Stapf(sp.nov,); foliissubcoriaceispetiolatis oblongo-vel 

 elliptico-cuneatis acuminatis 3-plinerviis venis transversis inconspicuis, 

 floribus solitariis e ramis annotinis pedicellatis, petalis late oblongis 

 acutis, staminibus circiter 30. 



Hab. Borneo : Sarawak, Selabat Rock, on the coast, Haviland. 



Frutex 5 ped. altus. Folia ad 3 poll, longa, ad li poll, lata ; petiolns 

 \ poll, longus. FedicelH 4 lin. longi. Calyx 2-2| lin. longus. Fetak 

 3 lin. longa. /Stamina H Hn. lon^a. 



t r^^^ genus Plethiandra was placed in Astroniem hy the authors of the 

 •Genera Plantarum,' and subsequently by Baillon, Coguiaux, and 

 Krasser. I have, however, pointed out iii 'Trans. Linn. Soc,,' I.e., that 

 the affinity lies most distinctly with 2£edinil la, irom which Flethiandrn 

 ditlers mainly in the large number of stamens. Curiously enougli, 

 Cogniaux has described a new genus, MediniUojJsis, which he places 

 next to Medmilla, and which difiers, according to him, from Medinilla 

 m exactly the same points. Unfortunately I have not been able to see a 

 good specmien of MediniUopsis. But judging from a specimen of Beccari's 

 in the Kew Herbarium numbered 4004, and from Goguiaux's descrip- 

 tion there cannot be any doubt that it is identical with Plethiandra. 

 I here is only one point of difference so far as Cogniaux's description 

 IS concerned, viz. the rmniber of cells in the ovary. Cogniaux indicates 

 throe cells for Afedinillopsis, whilst I finci constantly six in P/e^/«««f/m ; 

 but the septa separating the cells are so extremely delicate that an 

 error as to their number is quite conceivable. In a rather advanced fruit 

 of Leccans No. 4004 I counted, in fact, four distinct septa, the otiu;r 

 two having been probably absorbed. The description of MediniHoif 



.1 -, ' . ,o „' -"-w^^.^^j ct^n.c'i rtiiiiu^L exactly wiuxi my -^ • *'"^' 



the oidy point of difTerence being the length of the petals. 



Another specimen from Beccari's collection (No. 4049) was de- 

 scribed by Cogniaux as Medinilla rohuHta. The one I have seen, 

 W.« +T . v^'-'l '° extremely like Cogniaux's Medlniliopsis lecca- 

 r^«mthatlt might be taken to be the same but for the very sl.ort 

 fw'" / /i^^.^^'^*^ «een flowers of it, but I feel convinced it is also 

 a species of l'lethuiudra.~0. Stapp. 



sortit of or'n-v"' fi V''"^- ^- '^'S^'' ^^^ 4. Smaller .sta..ens. 5. Lon Juuli.al 



ot o^a.j. 6. rrunsverse section of immature) fruit. All adanjed. 



% 



