Dypsis thouarsiana Baill., Bull. Mens. Soc. Linn. Paris 2: 1163 (1894)

Primary tabs

no image available

Introduction

  • INSUFFICIENTL Y KNOWN SPECIES (J. Dransfield & H. Beentje, The Palms of Madagascar. 1995)A

Distribution

Map uses TDWG level 3 distributions (http://www.nhm.ac.uk/hosted_sites/tdwg/geogrphy.html)
Madagascarpresent (World Checklist of Arecaceae)B

Discussion

  • The interpretation of the name Dypsis thouarsiana has been one of the most intriguing problems in Madagascar palm nomenclature. The name was published by Baillon (1894a) in his paper entitled "Les palmiers malgaches à petites fleurs". Loosely translated, his protologue reads "The species that we name D. thouarsiana takes after the preceding one (D. lantzeana) in the deep division of the leaves and by the confluence of the terminal segments, despite the lower ones being separate; but it is also equally reminiscent of
    D. pinnatifrons in the narrow rigid elongate segments, nerved like the leaves of Gladiolus, linear lanceolate and nearly half a metre long. In the plant of Du Petit Thouars and in a variety which we attribute to the same species, called Vounouthre or Talanouc by the natives of northeast Madagascar (Boivin 1709, 17092, Sainte Marie, Tafondrou), the inflorescence has a short stocky compressed peduncle, and long divisions which are two to four decimetres, rather thick, rigid and carrying three-flowered glomerules. In the male flowers, the only ones that are well developed, there are only three stamens and obtuse concave strongly imbricate sepals, the base of which is prolonged on the outside in a sort of solid obtuse spur, equalling a fifth of the total height of the sepal and running into a short obtuse keel along the dorsal median line".
    Thus three collections are cited in this protologue. The collection made by Du Petit Thouars, without locality, consists of leaf fragments and what appears to be a first order branching system of a very young inflorescence; floral bracteoles have developed, and three imbricate sepals are partially developed, within which are three very small valvate structures which we interpret as petals. Stamens, however, have not yet developed. Boivin 1709 consists of two leaves while Boivin 17092 consists of a first order branching system of an inflorescence, still young, but with more floral details available. Within the petals are three antesepalous stamens; these have widely divergent anther thecae that seem to be pendulous from the tip of the filament, a most unusual stamen form, though not unique to this taxon (see also D. lokohoensis and D. fasciculata). The material is so incomplete that we cannot be certain that the leaves in Boivin 1709 are from the same species as the inflorescence in Boivin 17092. The Du Petit Thouars collection seems to be conspecific with the Boivin leaf collections. A fourth collection, probably made by Boivin, is annotated "talanouac" but consists of leaves only, matching those of the other collections.
    The leaves of the three collections described above are relatively small; they have distinctive broad multifold basal and apical segments and single fold mid-leaf segments. There is a long petiole and at its base a short entire ligule. It must be said that these leaves are reminiscent of the plant previously named Vonitra fibrosa that has long been called Vonitra thouarsiana, so this may go some way to explain why Beccari based his Vonitra thouarsiana on Baillon's name (while preparing his description from a collection made by the Rev. Baron). Boivin 17092 also carries the name "vounouthre" -i.e., vonitra, the consistently applied vernacular name for this important and common palm. Yet there are only three stamens in Boivin 17092, as described by Baillon and clearly evident in the specimen, while in Beccari's new genus Vonitra there are six biseriate stamens. It is most surprising that Beccari did not mention this anomaly. Could it be that the four early collections represent more than one taxon? The leaves could be interpreted as juvenile D. fibrosa leaves. Only in D. fasciculata do rachilla and stamen form approach those of Boivin 17092, but this species has inflorescences branched to two orders; the inflorescence fragment in Boivin 17092 could represent the whole branched portion of an inflorescence branched to two orders, or, more likely (because of "rachis" is flattened on one side) a whole first order branching system of an inflorescence branched to three orders. It is possible that the inflorescence in the Du Petit Thouars collection represents a fragment of a very young D. fibrosa inflorescence, but the stamens that would allow identification have not yet developed.
    Further collecting on Î;le Sainte Marie may sort out the problems, but at present we have a species, Dypsis thouarsiana, that appears to have been typified by Baillon on the Du Petit Thouars collection (the Boivin collections are mentioned as a variety) that are too young to show the diagnostic three stamens described by Baillon. The two numbered Boivin collections may not even belong to the same taxon, as one consists of leaves while the other of inflorescence fragments, and these are clearly not conspecific with Dypsis fibrosa. (J. Dransfield & H. Beentje, The Palms of Madagascar. 1995)A

Bibliography

A. J. Dransfield & H. Beentje, The Palms of Madagascar. 1995
B. World Checklist of Arecaceae