2t>G 



NATURAL HISTORY OF PLANTS. 



IFilhicas. Three species oi Kibara' have been described, but it may well 

 be tliat they are merely varieties of one and the same species, observed 

 in Java, i\rahicca, Sumatra, and Celebes, with leaves of variable thick- 

 ness, which may be entire or serrate, acute or obtuse at the apex. 



It is Avortliy of note that as the stamens become fewer, so they 

 tend to lose in length, and that in the oligandrous species of the 

 t'cnus Mollinedia the filaments disappear, and the connectives become 



Mullinedia {Kibara) coriacea. 



Fig. 328. 

 Female flower {^{'). 



Fig. 329. 

 Female flower, diagram. 



Fig. 330. 



Longitudinal section of female 



flow-er. 



broader than they are long (figs. 332-335). This is especially 

 marked in Epluppiandra'^ a plant from Madagascar, of which it has 

 been proposed to make a special genus. Its female flowers are 

 exactly those of other MoIUnedias, and the stamens of the male 

 flowers are usually arranged two superposed to the two outer divi- 

 sions of the perianth, and two to the inner, then two more to the 

 outer, and so on. Owing to the slight elevation of the connective 

 the two anther-cells tend to become horizontal ; but their essential 

 structure is the same as in all other species <di Mollinedia. The most 

 remarkable fact in this species, which allows us to make it the type 

 of a special section, is that, on anthesis, not only do the four divi- 



' Bl., Mus. Lugd. Bat., ii. 87.— Hook. & 

 TiioMS., Fl. Ind., i. 165. — TuL., Mon., 404. 

 — IIas.-<K., pi. Jar. Ear. (1818), 209, n. 134. 

 — Steud., ISo)iu;ncl., BIG. — Walp., Ann., iv. 

 HI.— A. DC, loc. cit. 



2 Decsne., Ann. Sc. Nat., ser. 4, ix. 278, t. 

 7 ; Traits Gener. de Bot., 517. — A. DC, 

 Prodr., xvi. s. post., 662. — H. Bn., Adansonia, 

 ix. 124. 



