164 , . 



measuring about one line ; corolla exceeding hardly ^in. in length, 

 unbearded ; its lobes imbricate in bud ; filaments 10, rarely 12, 

 fixed to the very base of the corolla, of equal length, not fringed ; 

 anthers lin. long, opening with two terminal pores ; styles about 

 2in, long ; berry |in. in diameter ; seeds pale brown, half a line 

 long, cuneate-ovate, angnlar, finely streaked. 



The anthers distinguish this species from nearly all its numerous 

 congeners, and bring it near the section Notopora (J. Hook. Icon, 

 plant. 1159), but they are fixed below not above the middle. 



There seem to be but few other species of Vaccinium known from 

 any part of Polynesia. Of these V. Vitiense (Paphia Vitiensis, 

 Seem, journ. of Bot. 1864, p. 77 ; Flor. Vit. p. 146, t. xxviii) differs 

 in its large flowers, reminding of certain Thibandias, of which genus 

 at least the species with free stamens should be included in Vac- 

 cinium, as indicated by the writer in the volume of the Acclima- 

 tisation Society of Victoria for 1872. 



V. cereum (G. Forst. florul. insul. Aiistr. prodr. p. 28 ; Andro- 

 meda cerea, Murr. syst. veg. 406) from Tahiti as well as F. mac- 

 gillivrayi (Seem, journ. of Bot. 1864, p. 67) from the New Heb- 

 rides, differ both by their acute and especially serrated leaves, their 

 cleft calcarate anthers and sphaerical fruits. In a similar manner, 

 V. reticulatum (Sm. in Rees's Cyclop. 1824),of the Sandwich Islands, 

 is removed from our plant ; besides it has lanceolate lobes of the 

 calyx, while the tube of the latter is as long as it is broad. 



V. plenduliflorum (Gandich. in Freycen. Voy. Bot. 454, t. 68) 

 also from Hawaia differs, irrespective of the serratures of the leaves, 

 in longer peduncles, much elongated lobes of the calyx, anthers 

 with dorsal appendages, slender style, almost globular fruit and 

 a shorter embryo. 



Dr. Asa Gray has offered on the Polynesian Vaccinia notes 

 referring to characteristics and synonymy in the Proceedings of the 

 American Academy for Arts and Science, 1862, p. 323-324. 



Among Indian species, V. RolUnsoni (Hook. bot. Magaz. 4612) 

 from Java, is- nearest, except perhaps the very imperfectly known 

 V. micropjiyllum (Reinw. in Blume's Bijdr. 851) from Celebes. The 

 former has almost precisely the same foliage, and also slightly 

 downy branchlets, while the flowers are not always terminal and 

 racemose, but also axillary and solitary ; the berries, however, are 

 globular.* 



I am not acquainted as yet with any true Vaccinium from 

 Australia and New Caledonia, but the genus will likely be found 

 well represented in the higher regions of New Guinea. 



The Samoan collection received from the Rev. S. T. 

 Whitmee, contains the following cotyledonar plants : — 



Stephania hernandifolia, Walp. 

 Cardamine sarmentosa, G. Forst. 

 Hibisciis abelmoschiis, Lin. 

 Trichospermum richei. Seem. 



* Since writing the above I have had access to the illustrated plate of 

 Vaccinium Kollisoni. It differs from V. Wliitmeci also in its hairy pedicels 

 And (especially) filaments. 



