Polyalthia^] flora indica. 137 



In the absence of fruit, this species and the next cannot be better placed than in 

 the genus JJnona. They are probably congeners of Unona pannosa, Dalzell, and of 

 Uvaria virgata, Blume, 



12. U. cauliflora (H.f. et T.) ; foiiis liueari-oblongis vel lineari- 

 lanceolatis brevissime petiolatis basi rotunda tis obtusis, floribus secus 

 rainos fasciculatis, sepalis dense pilosis, petalis linearibus sericeis. 



Hab. In peninsula Malay ana ad Siugapur, Lobb ! — (v. s.) 



Arbor? Famuli dense fasco-tomentosi. Folia 5-6 poll, longa, 1^-lf lata, pe- 

 tiolo vix lineam longo, incrassato, tomentoso, tenuiter coriacea, supra lucida, subtus 

 (secus costam densius) fusco-pubescentia, nervis falcatis. F lores in fasciculo pau- 

 ciores quam in praecedente. Pedicelli pedunculo abbreviato liguoso ramoso iuserti, 

 i poll, lougi, tomentosi. Sepal a ovato-lanceolata, i-pollicaria. Fetala 2i poll, 

 longa, i poll, lata, (ex sicco) lsete rubra. Stamina indefinita, breviter cuneata, trun- 

 cata, antherarum loculis discretis. Ovaria ovalia, dense albo-strigosa. Ovula in 

 sutura ventrali 3-4. 



17. POLYALTHIA, Blume. 



Polyalthia, § 1, Blume, Fl. Jav. Anon, 70. 



Sepala 3. Fetala 6, biserialia, ovata vel elongata, coriacea, piano- 

 con vexa, oest. valvata. Stamina indefinita, connectivo truucato capi- 

 tato ; antlieris lineari-oblongis dorsalibus. Torus apice truncatus, pla- 

 nus. Ovaria indefinita, oblonga, in sutura ventrali prope basin biovulnta. 

 Stylus oblongus. Carpella Guatterice. — Arbores, foiiis coriaeeis, nervis 

 obliquis distantibus apice arcuatim connexis inconsjric/ds, floribus axillaris 

 bus vel extra-alaribus. 



The genus Pohjalthia, as originally instituted by Blume, included four very dis- 

 tinct groups, all of which were clearly distinguished by that author as sections. 

 Blume made the character of the genus to rest mainly on the two-ovuled ovary. 

 This indeed is the principal point of resemblance between the different groups which 

 he brought together under this genus, while they possess, it appears to as, too many 

 and important points of distinction to permit of their being associated together. 

 We have therefore considered each of Blume's sections as a distinct genus, except 

 Kentia, which is so closely allied to our genus Melodortim that it does not appear 

 necessary to retain it as a genus. As Blume has foreseen the probability of this being 

 done, and has given to each section a name, indicating at the same time to which he 

 desired the generic name to be attached, we have of course made no alteration in 

 that respect. The true Polyalthia, in the restricted sense, as characterized above, 

 is much more closely allied to Guatteria than to any other genus, the flowers being 

 in no respect different, except by the increased number and different position of the 

 ovules. * The species have, however, a peculiar habit, not like that of the majority 

 of GuafleTi^wevery short-pet ioled leaves giving them a peculiar facies. Guat- 

 teria suberosa, however, approaches the genus Pohjalthia in this l 

 and forms a direct transition from the one genus to the other. W< 

 that species of Polyalthia are numerous in the Malayan Archipelago, while within 

 our limits they are entirely confined to the Malayan peninsula. Besides Blume's 

 species and those described below, we have before us several species from the Philip- 

 pines, collected by Cuming. We have also seen an imperfect specimen from Ceylon, 

 in Dr. Wight's Herbarium, which resembles P. obliqua, but has oblong-lanceolate 

 leaves more membranous than those of that species, and long-pedicelled flowers op- 

 posite the leaves, not axillary as in P. obliqua: it is probably a very distinct species. 



T 



learn 



