304 On the Tree uMch produces the Gamboge of Commerce. [Oct. 



her of seeds are not subject to accidental loss in drying or examining. 

 The quinary proportion of stamens and uneven number of seeds afford 

 1 think almost irrefragable proof of the identity of Stalagmitis and 

 Xavtliochi/mns; the petals and sepals only being erroneously described. 



This view is confirmed by Mr. George Don, in his edition of Miller's 

 Dictionary, hawing reduced Roxburgh's 'X&nthoehymu-3 to Stalagmitis, I 

 presume on the authority of Murray's own specimens which he could 

 examine in the Bankean Herbarium ; an arrangement in which we, not 

 sufficiently adverting to his? opportunity of determining the identity of 

 these genera, did not think it safe to follow him. By thus uniting 

 GarctJiia to Cambogia, and Stalagmitis to Xantliochi/mus, the confused 

 assemblage is reduced to two very distinct genera. The only question 

 that remains to be considered is, whether or not it is advisable to leave 

 them as they now stand. 



In my opinion the genus Garcinia, as now constituted, ought not to be- 

 retained; a glance at the accompanying figures will explain my reasons 

 for thinking so by exhibiting in one view several of the incongruities 

 whichit presents. For example, the Garcinia mangostana, (PI. fig. 2) 

 speciosa, (PI. fig. 1) and cornea, have the filaments of the male flower 

 united, forming four large fleshy bodies covered with anthers, and two 

 of the three are known to have a glabose, not sulcated fruit. These 

 might form the type of a genus under Rumphius' original name Mango- 

 stana. G. Cambogia, (PI. fig. 3) has the stamens of the male flower 

 ranged in a single row, round a disk-like receptacle, with a sulcated 

 fruit. This might form the type of a second genus, for which, as 

 nearly corresponding with the character assigned by Linnaeus to his 

 Garcinia, that name might be retained. In G. Kydia, (PI. fig. 4) 

 Zeylanica, pedunculata, paniculata and ajjtinis (PI. fig. 5) the filaments 

 are united into a staminal column, terminating in a head covered with 

 anthers, fruit usually furrowed. These might form a third genus,, 

 retaining for them the now vacant name of Cambogia : and lastly, 

 G. putoria, Roxb. {Mangost. morella Gasrt ?) ; elliptica Wall. {fid. 

 Graham) and Dr. Graham's Ceylon plant, (PL fig. 6) would form a 

 fourth, distinguished by their united filaments, and cup-shaped, one- 

 celled, circumscissile, anthers; for which the now vacant name, Stalaga- 

 mitis, might be retained, in preference to disturbing Roxburgh's 

 Xanthochymus, now well established, by rigidly enforcing the rule of 

 priority, and restoring that of Murray, thereby causing considerable 

 confusion in the synonymy, that might thus be easily avoided. 



Palamcottah, 2d August, 1836, 



