88 MB. O. BENTHAM ON BIXACEiE AND SAMYBAOE^E. 



Tribus III. Homalie^. Calyx liber v. basi ovario adnatus, 4-15- 

 merus. Petala tot quot sepala v. rarius numero dupla. Sta- 

 mina petalis numero aequalia v. plura, singulatim v. fascicula- 

 tim iis opposita, cum glandulis pulvinatis sepalis oppositis 

 altemantia. 



* Ovarium supe-rum. 



Genera : — 9. Calantica, Tul. 10. Bivinia, Tul. 11. Dissomeria, 

 Benth. 



** Ovarium semi-infer -urn. 



Genera : — 12. Homalium, Jacq. (Blackwellia, Comrn.) 13. Byr- 

 Banthus, Ouillem. 



With the exception of Osmelia, Lunania, and Banara, I have, for 

 the present, but few remarks to offer on these genera. For Thio- 

 dia, Sadymia, and Samyda, I may refer to Grisebach's 'Erlaute- 

 rungen ' above mentioned. The large genus Gasearia is much in 

 need of a careful monographist, for some of the common species 

 have been repeated by various authors under several different 

 names. I see no reason for modifying the five sections into which 

 I proposed to divide it in the 4th vol. of ' Hooker's Journal of 

 Botany.' Grisebach has indeed well identified Valentinia, Sw., 

 with the section Hexanthera, Endl., and my Casearia brevipes with 

 Latia guidonia, Sw., both of which he proposes to establish as 

 distinct monotypic genera •, but I can scarcely concur in the 

 reasons which he gives. In Valentinia there appears to be 

 nothing but the prickly-toothed leaves to rely upon. In Gui- 

 donia the scales or staminodia are very short, and slightly united 

 outside the filaments, so that the stamens appear to be inserted 

 within the notches of a short crenulated tube ; yet it appears to 

 me that this cannot be considered as a structural difference, nor 

 as the character of a group of species, but merely as a difference 

 in degree, exemplified in a single one which has otherwise the 

 habit and characters of the Iroucanas. 



Osmelia, Thw., from Ceylon, has the perianth, stamens and ovary 

 of Casearia, but the styles are short and free from the base ; and 

 this character being accompanied by a very marked difference in 

 inflorescence, and confirmed in two other species from the Philip- 

 pine Islands, may be considered as amply sufficient for esta- 

 blishing a generic group. One of the latter species has been 

 published by Turczaninow under the name of Stachycrater, too 

 nearly after the appearance of the first part of Thwaites' ' Enu- 

 meration of Ceylon Plants' for him to have been aware of the 



