FLORA VITIENSIS. 85 
Var. 8; foliis latioribus nunc ovalibus apice rotundatis subtus haud flavescentibus. A. Gray, l.c. 
—Ovalau and north coast of Vanua Levu (U. S. Expl. Exped.); Kovovono, Viti Levu (Seemann! 
n. 172). 
The leaves of M. Vitiense are not strictly opposite, but rather one-sided, and somewhat swollen at the 
insertion of the leaves. These characters, as well as the shape of the leaves, distinguish it from another 
Polynesian species, viz. M. Harveyi (sp. nov.), Seem. ; glabra; ramulis junioribus acute angulatis ad inser- 
tionem foliorum haud incrassatis; foliis stricte oppositis ovatis v. ovalibus acuminatis basi acutis, supra 
nitidis atroviridibus venis prominulis, subtus subglaucis subaveniis; cymis axillaribus pedunculatis; pedun- 
culis petiolo longioribus ; bacca globosa 1-sperma.— Vavau and Lifuka, Tongan Islands (Harvey!). Leaves, 
including the short petioles, 2-23 inches long, 1-14 inch broad. Flowers fallen off. 
II. Astronia, Blume, Bijdr. p. 1080; Naud. Melast. p. 677. Calycis tubus hemisphzricus ; 
limbus sub anthesi clausus, post anthesin irregulariter lacerus. Stamina 10-12, qualia; antheris 
dolabriformibus apice obtusis antice rima duplici longitudinali dehiscentibus, connectivo postico basi 
mutico ant in tuberculum vix perspicuum desinente; filamentis complanatis brevibus. Ovarium cum 
calyce omnino adnatum, 2—5-loculare. Placente nunc in fundo loculorum depressæ planze, fere semi- 
lunares, nunc e basi column centralis products co-ovulate. Capsule globose limbi calycini vestigiis 
coronatee, in fragmenta undique disrupte, nervis coriaceis parenchyma solutum superantibus. Semina 
co, minuta, linearia aut acicularia, raphe laterali fusca hinc excurrenti lineata.—Arbores et arbuscule ; 
foliis petiolatis ovatis ovatove-oblongis acuminatis integerrimis 8-nerviis 3-plinerviisque; floribus in 
paniculas terminales digestis, albis aut purpurascentibus.— PAarmacum, Rumph. Amb. vol. iv. t. 69. 
Conostegie et Melastome sp. auct. ' 
l. A. fraterna, A. Gray, Bot. Wilkes, p. 576; glaberrima; foliis longe petiolatis oblongis acu- 
minatis basi acutis manifeste 3-plinerviis (pretermisso utrinque nervo tenui submarginali) ; cymis 
corymbosis; pedicellis gracilibus; alabastris ellipticis; calycis limbo clauso demum aperto ultra 
ovarium longiuscule producto, margine 4—5-fido; ovario 5-loculari; placentis ima columella ortis.— 
Melastoma glabra, Forst. Herb., Icon. (ined.) t. 137, 138, et in Guill. Zephyr. Tait. p. 61, nec Prodr. 
Astronia Forsteri, Naud. Melast. p. 258 ex parte.—In woods of the interior of Viti Levu (Milne!), 
Tahiti (Forster! in Mus. Brit.). 
Milne's specimen is merely a sterile branch, but as far as it goes agrees with the Tahiti specimens. 
It should here be mentioned that two generically different plants have been confounded under the name of 
Melastoma glabra (Conostegia glabra), the one having obscurely 3-nerved leaves and a calyx which has a per- 
sistent limb, splitting longitudinally in anthesis into 3—5 lobes, 10 stamens, and a 5-celled ovary, the other 
manifestly 3-ribbed leaves, a deciduous calyx-limb, which in anthesis separates by a transverse circumscission 
in the form of a calyptra, 8 stamens, and a 4-celled ovary. Asa Gray named the former Astronia fraterna, 
and says that he has seen Forster’s original specimens at the British Museum, which, he adds, belong to the 
latter species. But here he was mistaken. The six original specimens which the two Forsters collected 
at Tahiti, and which, with their own labels attached, are preserved at the British Museum, do all belong to 
Asa Gray's Astronia fraterna. The two plates, one of them coloured, which Forster in his MS. drawings 
devotes to the illustration of his Melastoma glabra, also represent A. Gray’s Astronia fraterna. Moreover, 
in the detailed MS. description of Forster, published by Guillemin (Zeph. Tait.), the calyx is described as 
“apice 4-fidum rarius 5-fidum, laciniis irregularibus laceris,” clearly referring to Asa Gray's A. fraterna. 
But it is evident, from the brief diagnosis in his * Prodromus,' that Forster also knew the second Tahitian 
species, and confounded it with that of which his specimens are preserved in the British Museum, of which he 
made two drawings, and of which he gave a detailed description, for he says, in the * Prodromus,’ of Mela- 
stoma glabra, “ foliis integerrimis 3-nerviis elliptico-lanceolatis scabris [evidently a misprint for glabris— B. S.], 
calycibus eireumscissis." Naudin, in his valuable Monograph of Melastomacee, re-names Forster's Mela- 
stoma glabra, Astroni@Forsteri, and seems to include both plants under that name, saying of the calyx, “sub 
 anthesi partim cireumscissum aut etiam in lobos 2-4 dentiformes obtusos laceratum fuisse ostendebant.” 
But it should be added that he also had no perfect specimens. : 
. On mentioning my ideas about the generic differences to M. Triana, who is now occupied with a 
revision of the Melastomacea, he examined these two plants, and agreed in my opinion; and he has since 
ascertained that M. Decaisne arrived at the same conclusion, and had given the name Naudinia to this 
