EDIT

<citation>Kew Bull. 46: 62 (1991)</citation>

<type>Madagascar; Dransfield et al.; JD 6453</type>

<type_loc>Holotypus K; isotypi BH, MO, NY, P, TAN</type_loc>

</nomenclature>

<div type="introduction"><p></p></div>

<div type="etymology"><p>"Hovitra vari mena"</p></div>

<div type="uses"><p></p></div>

<div type="discussion"><p>The most difficult problem still requiring solution is the affinities of this majestic palm. The pinnate leaf and the arrangement of flowers in triads of a central pistillate and two lateral staminate clearly indicate that it belongs to subfamily Arecoideae (Uhl and Dransfield 1987). The reduplicate vernation, the presence of only two large bracts in the inflorescence (the prophyll and a peduncular bract) and the pseudomonomerous gynoecium (i.e., a gynoe- cium with a single fertile locule but with vestiges of two further locules) further narrow the affinity to tribe Areceae, but to which subtribe of this, the most variable of tribes in Arecoideae, the palm is most closely related is not obvious. The corky-warted fruit is present in Manicariinae and two genera of Iguanurinae (Sommieria and Pelagodoxa), but the former has trilocular, triovulate gynoecia while the fruits of genera in the latter subtribe (which consists of genera with pseudomonomerous gynoecia) have a conspicious operculum in the endocarp that is lacking in the "red-lemur palm" Besides, such corky-warted fruits are found in several quite unrelated palm genera (see Uhl and Dransfield 1987) suggesting that this epicarp condition may have arisen several times in the evolutionary history of the family, Multistaminate flowers occur in many arecoid subtribes and the ontogeny of multistaminy is various (Uhl and Moore 1980), again suggesting that the presence of more than six stamens may have arisen several times. One of the most curious features is the internal fruit structure. The endocarp bears a basal heart-shaped button which seems to represent the sclerified vascular supply next to the chalaza. Such a structure is also present in Orania and Halmoorea in the Oraniinae (and in members of subfamily Phytelephantoideae); however, members of Oraniinae are tricarpellate and triovulate and are vegetatively and in inflorescence structure both rather uniform and quite different from the red-lemur palm. Apart from the presence of more than six stamens, the staminate flowers do bear a resemblance to those of larger members of the quintessentially Madagascar subtribe Dypsidinae, but, without drastically altering the circumscription of the Dypsidinae, an otherwise apparently very natural group, the "red-lemur palm" cannot be included. It may be necessary to create a new monotypic subtribe for this very distinctive new genus, but even so this does not solve the problem of which subtribe of Areceae is the sister group. A greater understanding of the relationships will require developmental study and may be helped by cladistic studies in progress (Dransfield and Uhl, in prep.) and until then I defer the subtribal placement.</p></div>

<div type="materials_examined"><p>MADAGASCAR. Maroantsetra, Andranofotsy River, Sahavary, hills east of village, Andilampananina, primary forest, deep narrow valley at 350-450 m, 23 Oct. 1986, Dransfield, D. N. Cooke, A. Rakotozafy,J. H. Beach, P. P. Lowry and G. Jean JD 6402 (BH, K, MO, NY, P, TAN); 10 Feb. 1988, DransJield, A. J. Henderson and M. Staniforth JD 6453 (holotype K; isotypes BH, MO, NY, P, TAN). </p></div>

</treatment>

</taxonxBody>

</taxonx>