ROSACEJS. 391 



The ^ower of Fup/troma^ is hermaphrodite, with a cup-shaped recep- 

 tacle, bearing on its edges four or five unequal sepals, and perhaps, 

 too, a perigynous corolla/ The stamens have the same insertion, and 

 are as numerous as the sepals and superposed to them. Their fila- 

 ments are broad at the base, where they cohere in a variable manner,' 

 but are free above and taper to a point." The gynseceum is very 

 much like that of Eiicrijjjkia, inserted in the bottom of the receptacle. 

 It consists of an ovary with three uniovulate cells, surmounted by a 

 filiform persistent style bearded below. The fruit is a capsule sur- 

 mounted by the remains of the style, and accompanied at the base 

 by the receptacle and the reflexed sepals. It has a pretty thick 

 mesocarp ; this separates from the woody endocarp which is septici- 

 dal, with its clefts prolonged into the style. Each stone then opens 

 along the ventral angle to free a descending seed prolonged below into 

 a long flattened wing, and containing within its coats a fleshy embryo 

 with its radicle superior, surrounded by a thin layer of albumen. 

 Only one species of this genus is known,^ a native of the north of 

 Brazil. Its woody branches bear alternate exstipulate petiolate 

 simple leaves. Its flowers are in terminal racemes." 



adopted our view, but they at first (Gen. 164, Walp., Rep., v. G59. — Hook. F., in Makt., Fl. 



195) shared that of Planchon {Ann. Sc. Nat., Bras., Rosac, GO. 



sen. 4, ii. 261), who places Eucri/phia among ® Next to F.uphronia has been placed the 



the SaxifragacecE. genus Canotia (ToBE., ap. W'ippl., 12), whose 



' Maet., & Zucc, Nov, Gen. et Spec, i. 12b flower is pentamerous, with a five-lobed per- 



t. 73.— Endl., Gen., n. 6400. — B. H., Gen., 317. sistent calyx and five hypogyuous stamens. Its 



616, n. 34. I'nit is a five-celled septicidal capsule, surmounted 



2 The fruits alone being known, the petals, if ''^ tlie subulate style. In each cell is a suspended 

 there ever were anv, must have fallen off in the ^'^ed, whose coats are prolonged be o.v into a mem- 

 specimens in the Munich herbarium. Vj""?' "J,"^' ""Z^^"'^ f" '''. '" ^ ^^ '''''' 



, „, , , , ., , ,. 1 , , "' neshy albumen. Ihe only species known is C. 

 They have been described as diade phous, ,,,i^,^,,tha, a shrub from New Mexico, with leaf- 

 four bemg united into one bundle. But we ,^^^ branches, whose alternate twigs end in long 

 have seen more thun one free filament around j^^^^ Hektham and Hookee sap of this plant 

 the one young fruit we have been able to ex- ,</-.,„„„„„„ i jx t t u i i- >y ,^ 



Jo •' brenus quoad ((ffinitatem value dubiitm {Gen., 



616, n. 35). As yet the corolla has not been 



The anthers are unknown. seen, so that th,j char.icters appear to have beeu 



* E. hirtelloides Mabt. & Zrcc, Lc. cit. — made out from fruiting specimens only. 



