Lowiaceae

Primary tabs

Lowiaceae

Description

Perennial, glabrous herbs without aerial stem, often tufted from a horizontal or vertical, sympodial rhizome bearing scale-like leaves. Leaves distichous, glabrous, eligulate, base sheathing, petiole well-defined, lamina entire, lanceolate, dorsiventral, enrolled in bud; venation pinnate-parallel with the secondary veins emerging from the prominent main vein at narrowly acute angles and regularly connected by fine transversal veins (conspicuous in dried material, but not in living plants). Inflorescence dense or lax, consisting of irregularly branching monochasial cymes. Flowers bracteate, bisexual, epigynous, with the ovary prolonged into a solid extension. Sepals (outer tepals) 3, sub-equal, narrow, fused basally into a tube. Petals (inner tepals) 3, very unequal, the two lateral small, narrow, the median large, forming a variously shaped labellum. Stamens 5, free from each other, the upper median reduced; filaments short, adnate to the base of the petals; anthers tetrasporangiate, introrse, opening by long slits, apically terminated by a short connective tip. Ovules numerous, anatropous, bitegmic, crassi-nucellate. Fruit a dry, loculicidal, elongate capsule terminated by a beak formed by the proximal part of the ovary extension. Seeds numerous, globose, hairy, endospermous, with a lacerate aril and an operculum opposite the radicle.

Distribution

from southern tropical China (Kwangtung and Hainan), through Vietnam, Laos, and Thailand, in Malesia to West Malaysia and Borneo: present
There is only one genus, Orchidantha. For the distribution, see there.from southern tropical China (Kwangtung and Hainan), through Vietnam, Laos, and Thailand, in Malesia to West Malaysia and Borneo

Pollination

Very little is known about the pollination. The flowers of Orchidantha fimbriatum have a foul smell (Corner writes on a label that they smell of "bugs and coconut-oil"). The labellum is white and placed close to the soil, and the succulent central part along the midvein contains cells with nutritious tissue, protein cells. Kress (pers. comm.) reports that the labellum is eaten during the night; he suggests that small primatess, tree shrews (Tupaiidae), feed on the labellum then. Orchidantha holttumii, on the other hand, has purple flowers smelling of horse-dung (Boyce) or fish (Dransfield), and the labellum is erect during anthesis. This species may be pollinated by flies.

Morphology

The distal partial inflorescences of Orchidantha siamensis consist of a bract subtending a branch with a basal prophyl and three bracts, the uppermost of which subtends a flower bud, the next from the apex subtends a branch of a similar construction; the axis continue>beyond the uppermost flower bud (pers. obs.). Holttum in his analysis of O. longiflon; described the flower as truly terminal. He also observed from old inflorescences a succession of scars where the flowers had fallen, with two bract scars between each; he concluded that the cyme is close to being a cincinnus The bracts are all more or less tubular at base and their orientation is not easy to establish The ovary has an apical prolongation which is sometimes called a calyx tube or hyp-anthium. It is, however, solid with a central vascular strand connecting the base of the style with the ovary; here the term ovary extension, introduced by Holttum, has beer. used. The sepals (or outer tepals) are similar; the petals (or inner tepals) are arranged as in an orchid flower with two lateral petals and a labellum. The flower is resupinate. The lateral petals are small and overlap the stamens; the labellum is variously shaped. in some species having a narrower proximal part and a broad distal part, in other species (e. g. O. holttumii) the whole labellum is rather broad but distally curved back and forming a claw-like structure. The labellum often has lamellae along the vascular strands. These parts may play a role in pollination.

Anatomy

The roots are relatively thick and air channels are present. Vessels are present in the roots only and have scalariform perforation plates with numerous bars. The rhizomes are rich in small, angular starch grains quite different from those found in Musaceae and Strelitziaceae. The chlorenchyma of the lamina is without a palisade layer and consists of an irregular mixture of small and large cells. The leaf axis includes two distinct systems of narrow air canals. Stomata are paracytic with deeply sunken subsidiary cells. Stegmata are present in the form of 'hat-shaped' silica-bodies. Raphide sacs are common, especially in the rhizome. Crystals of calcium oxalate are found in all parts. Tannins occur only in the rhizome.

All in all Lowiaceae are anatomically very different from the other families in Zingi-berales.

Taxonomy

The systematic position of the Lowiaceae is undisputed; since its establishment it has been recognized as belonging to the order Zingiberales.

The only genus Orchidantha was described by N.E. Brown, who published O. borneensis on 23 October 1886 in a note titled "A new genus oiScitamineaeT He stated: "for the present I place it in the Museae" The plant was introduced from Borneo without indication of locality and grown at Ghent, Belgium.

On the 25th of October of the same year Scortechini published Lowia as a new genus from the Malay Peninsula based on L. longiflora from Perak. Also he referred it to the tribe Museae, regarding it as closely related to Heliconia. On account of the developed labellum he suggested it being a link between Museae and Zingibereae.

In 1893 Ridley published Protamomum maxillarioides, also from the Malay Peninsula. He knew of Scortechini's description, but not of Brown's, and recognized his genus as belonging to the same group as Lowia; he proposed the family name Lowia-ceae, without any formal description.

In 1897 K. Schumann treated the group under Musaceae as Lowieae in Engler & Prantl, Nat. Pflanzenfamilien, and as Lowioideae in Das Pflanzenreich three years later. H. Winkler followed him in the second edition of Engler & Prantl's Pflanzenfamilien (1930). Ridley established the family in 1924, in 1941 T. Nakai added a Latin diagnosis. All subsequent authors have recognized the family with only one genus, Orchidantha. See also Cronquist (1980).

Cladistic analyses undertaken by Dahlgren et al. (1985), Kress (1990) and Smith et al. (1993), all using different outgroups, show its relationship to the group of families related to Musaceae, rather than to the Zingiberaceae-Marantaceae group, and are thus in full agreement with the observations and conclusions of Brown. The study by Smith et al. (1993) involved the analysis of chloroplast DNA of Orchidantha fimbriata and O. siamensis.

The most striking character of the family is the enlargement of the median petal to a labellum as in the orchids, hence the name Orchidantha. However, little is known about the family. To quote Kress (1990): "The Lowiaceae are among the most poorly known taxa in the order in terms of taxonomy, general morphology, embryology, chemistry, and ecology."

Cytology

2n = 18 has been found in Orchidantha maxillarioides and O. holttumii by the author and in O. siamensis by C. Artharamas, Thailand (pers. comm.).

Phytochemo

Formerly Lowioideae (Orchidantha), Strelitzioideae (including Heliconiaceae 3 or 4 genera) and Musoideae (1-3 genera) formed one family, Musaceae s.l. As the chemistry of all these taxa is rather poorly known at present, some general remarks about chemical characters of Musaceae s.l. seem to be appropriate here. For recent reviews see Tomlinson (1959, 1969), Hegnauer (1963, 1986) and Dahlgren et al. (1985). It should be stressed that some of the important anatomical and chemical characters of Musaceae s.l. are intimately connected: structures of starch grains, crystals of calcium oxalate (raphides and other crystal-types), silica bodies located in longitudinal rows of specific silica-cells, called stegmata, and 'tannins' in ordinary parenchymatic cells or deposited in idioblasts or articulated lactifers with their mucilagineous tannin-like and polyisoprenoid contents; rubber-like polyisoprenes seem to be restricted to Musaceae s.str. The tannins of Musaceae s.l. are of the condensed type; all seem to be proantho-cyanidin oligo- and polymers yielding pelargonidin, cyanidin and/or delphinidin on treatment with mineralic acids. Serious chemical investigations are restricted to Musa-ceae s.str. apart from a flavonoid survey of Zingiberales by Williams & Harborne (1977); leaves of most representatives of Musaceae s.l. contained glycosides of the flavonols kaempferol and quercetin, but lacked glycosides of myricetin, flavones and C-glycoflavones; flavonoid content of leaves of Orchidantha maxillarioides, the onh investigated taxon belonging to Lowiaceae, was so low that no positive identification-were possible. According to Tomlinson proanthocyanidins are restricted in Lowiaceae to subterranean parts. Remarkable observations were published recently for bananas (Musa acuminata, M. x paradisiaca). Fruits of vegetable varieties of banana are used in. India as anti-ulcerogenic medicine; the active principles were reported to be acylated (16:0; 18:1) 3-glycosidic derivatives of sitosterol, called sitoindosides I to IV (Ghosu. & Saini 1984; Ghosal 1985). Musa acuminata 'Giant Cavendish', has Colletotrichium. rawsae-resistant fruits; unripe fruits were shown to produce the phenalenone-type phyto-alexin 2-(4'-hydroxyphenyl-)naphthalic anhydride after infection with conidia of this pathogen (Hirai et al. 1994). A year before Luis et al. (1993) had isolated two phenale-nones, irenolone and emenolone, from banana leaves infected with Mycosphaerella fidjiensis. It is noteworthy that these phytoalexins strongly remind the yellow to red pigments of Haemodoraceae (see Flora Malesiana I, 11: 357 and Hegnauer 1986: 659 Tomlinson stressed the fact that from an anatomical point of view Lowioideae have an isolated position within Musaceae s.l.; one argument is based on the structure of their rather small and often composite starch grains which strongly deviate from the zingiber-alean type present in the rest ofMusaceae s.l. Other pleas for family status of Lowiaceae are based on epicuticular leaf waxes which are lacking in Orchidantha, but are represented in all other investigated Musaceae s.l. by the so-called Strelitzia-type (Fröhlich & Barth-lott 1988), and by a low diameter of their sieve-element plastides which belong to the P2cs-type (i.e. containing cuneate protein crystals and starch grains) in all Zingiberales (Behnke 1994). Chemically the Strelitzia-type wax rodlets of Musaceae are composed predominantly of wax esters which are accompanied by alkanes, alkanols, alkanals and free fatty acids; triterpenoids are lacking (Heliconia, Strelitzia) (Meusel et al. 1994).
1. R. Hegnauer