A Revision of Borassus L. (Arecaceae: Coryphoideae)

A Revision of Borassus L. (Arecaceae: Coryphoideae)
of 26
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  A Revision of Borassus L. (Arecaceae)Author(s): R. P. BaytonSource: Kew Bulletin, Vol. 62, No. 4 (2007), pp. 561-585Published by: Springer  on behalf of Royal Botanic Gardens, Kew Stable URL: . Accessed: 13/04/2014 14:03 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at  .  . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact  .  Royal Botanic Gardens, Kew  and Springer   are collaborating with JSTOR to digitize, preserve and extendaccess to Kew Bulletin. This content downloaded from on Sun, 13 Apr 2014 14:03:57 PMAll use subject to JSTOR Terms and Conditions  KEW BULLETIN 2: 561-586 (2007) 561 A revision f Borassus L. (Arecaceae) R P. Baytoni Summary. A taxonomic revision of the genus Borassus L. (Arecaceae: Coryphoideae. Borasseae) is presented. Five species are recognised: B. aethiopum from Africa and Madagascar, B. akeassii from West and Central Africa, B. madagascariensis from Madagascar, B. flabellifer rom South and Southeast Asia and B. heineanus from New Guinea. Pollen morphology and leaf anatomy are examined and the taxonomic history, morphology, distribution, ecology and conservation stattus re discussed. Key words. Borasseae, Coryphoideae, natomy, morphology, palm, pollen, taxonomy. Introduction Borassus L. of subfamily Coryphoideae, tribe Borasseae, subtribe Lataniinae (Dransfield et al. 2005), is one of the most widespread genera in the Arecaceae with a distribution encompassing tropical Africa and Madagascar, southern and Southeast Asia and New Guinea (Map 1). A phylogenetic study using nuclear and chloroplast DNA sequences (Bayton 2005), suggests that Borassus is monophyletic in its current circumscription, but further commentary will be reserved for a future publication. Throughout its range, Borassus (and particularly the Asian palmyrah, B. fiabellifer .) provide a huge variety of products of great significance to local economies. Every part of the palm is utilised, though perhaps the most significant product is the sweet sap (toddy, amil) extracted by tapping the stem or inflorescences. This is fermented into palm wine (arrack, derived from Arabic) or, the crude sugar (jaggery, derived from Portuguese) is crystallised (Morton 1988). It is not the intention of this account to review the current and potential uses of Borassus and several such works already exist (e.g. Blatter 1912; Porteres 1964; Fox 1977; Kovoor 1983; Davis & Johnson 1987; Morton 1988; Burkill, 1997). However, it is noteworthy that despite its economic significance, the taxonomy of Borassus has remained problematic. Species delimitation in Borassus has always been challenging. Borassus palms are massive, with huge stems of up to 20 m tall and 1 m diameter, which makes gaining access into the crown challenging. The leaves and inflorescences are very large and consequently only small parts can be collected, while essential accompanying notes are often absent. The process of collecting, doctumenting and preserving herbarium specimens is extremely time-consuming. Therefore there is a dearth of herbarium material. Existing material is often incomplete with juvenile leaves substituted for the inaccessible mature leaves. In addition, all Borassus species are dioecious, decreasing further the comparability of existing specimens. Species delimitation in Borassus has relied upon morphological characters that reflect the narrow range of available herbarium material, or that can only be determined in the field. An additional problem with Borassus is missing type specimens. Of the seven species recognised in the most recent taxonomic revision of the genus (Beccari, 1924), only three have extant type specimens (a fourth is lectotypified with an illustration, Moore & Dransfield 1979). Taxonomic History Borassus and the type species B. flabellifer were described by Rheede tot Draakestein (1678) in the Hortus VWalabaficis nder the names Ampana (staminate palm) and Carimpana (pistillate palm). The description of B. flabellifer by Linnaeus (1753) cites Rheede's account (among others), and the accompanying illustrations have since been selected as the lectotype (Moore & Dransfield 1979). Borassus in Africa was considered by some authors to be only a variety of B. flabellifer (Kirk 1867; Warburg 1895; Drude 1896, Dammer 1901), but a second species, B. aethiopum Mart. was described by Martius (1838). The debate over whether or not to recognise Borassus aethiopum s largely Accepted for publication November 2006. 1 Centre for Plant Diversity & Systematics, School of Biological Sciences, The University of Reading, P0 Box 221, Reading, Berkshire, RG6 6AS, U.K. and Herbarium, Royal Botanic Gardens, Kew, Richmond, Surrey, W9 3AB, U.K. ? The Board of Trustees of the Royal Botanic Gardens, Kew, 2007 This content downloaded from on Sun, 13 Apr 2014 14:03:57 PMAll use subject to JSTOR Terms and Conditions  562 KEW BULLETIN OL. 62(s) resolved, with most recent authors accepting both species (Beccari 1914, 1924; Kovoor 1983; Dransfield 1986a). A limited molecular study by Kovoor & Hussein (1983) came to the same conclusion. The taxonomy of the two endemic Borassus species in Madagascar has received recent attention (Dransfield & Beentje 1995a; Bayton et al. 2003). Borassus madagascariensis (Jum. & H. Perrier) Bojer ex JLIm. & H. Perrier was first published in 1837i in a catalogue of the native and exotic plants of Mauritius (Bojer 1837). Bojer stated that the palm was. native to Madagascar and could be found in the vicinity of Majungay (now Mahajanga), and also in the Royal Botanic Gardens of Pamplemousses (now the Sir Seewoosagur Ramgoolam Botanical Gardens) in Mauritius. Bojer's account does not include a description of B. madagascariensis and his name is therefore a nomina nuda. Jumelle and Perrier de la Bathie (1907) accepted Bojer's account, but reduced the species to a variety of B. flabellifer. They later (1913) elevated it to specific rank and it is their accounts that inadvertently validate the epithet at both varietal and specific ranks. A second species of Barassus was also described byjumelle and Perrier de la Bathie (1913). Borassus sambiranensis Jum. & H. Perrier is restricted to the Sambirano Region in the northwest of the island. In the most recent taxonomic treatment of the Madagascar palms, Dransfield and Beentje (1995a) concluded that there were few morphological characters to distinguish the endemic Borassus species, either from each other or from B. aethiopum in mainland Africa. In 1914, Beccari published the first taxonomic revision of Borassus. In it, he described three neNw species, and two new varieties of B. aethiopum. He also transferred Borassus machadonis Ridl. into the new genus Borassodendron Becc. (Beccari 1914). This account was later expanded to include the other borassoid genera and was re-published after Beccari's death (Beccari 1924). Beccari's treatments recognised Borassus flabellifer as restricted to Asia and added Borassus sundaicus Becc. from Indonesia and Borassus heineanus Becc. from New Guinea. The latter species is morphologically divergent from other Borassus species, and in some ways resembles Borassodendron machadonis (Ridl.) Becc. (Beccari 1924). In Africa, Beccari described B. deleb Becc. from Sudan and recognised three varieties of Borassus aethiopum: var. senegalensis Becc. in WVest Africa, typified by material from modern-day Mali, var. bagamojensis Becc. in East Africa, based on material from mainland Tanzania and var. aethiopum from Central Africa (though the type was from Ghana). Chlevalier described a fourth variety, var. domesticus . Chev., from material collected in Mali (Chevalier & Dubois 1938). However, this name is not valid as there was no Latin diagnosis. n his treatment f the East African palmus, Dransfield (1986a) reducect all African taxa to a single species, with B. deleb and the varieties in synonymy. However in 1996, Ake Assi & Guinko (1996) examined a Borassus palm from West Africa that was morphologically distinctive. Further details were provided by Ouedraogo et al. (2002) and Arbonnier (2002) and the consensus of opinion was that these palms represented B. flabellifer, reviously unknown in Africa. However, this taxon was later recognised as a newv species, B. akeassii Bayton, Ouedraogo & Guinko (Bayton et al. 2006). Morphology Stem & Habit All Borassus species have a large single stem, branching only when damage has occurred (Ramassamv & Kannabiran 1991). The stem is clothed in dead leaf bases, though these later abscise cleanly leaving prominent scars. The stem of the African and Madagascar species is uisually ventricose with a prominent swelling below the crown. The anatomy and adaptive significance of this swelling has not been investigated, but Tuley (1995) and Henderson (2002) suggest that it is an adaptation to arid environments (though ventricose stems are not exclusive to palms from arid habitats); Barot & Gignoux (1999) proposed that it is connected with the onset of sexual maturity. Field observation of this character indicates that within a population, the swelling is usually at a similar height in each tree, and multiple swvellings have been observed in some individuals (Tuley 1995). Ventricose stems are characteristic of a number of unrelated palm genera, many of which occur in seasonally dry habitats. When the ventricose stem of Acrocomia crspa (Kunth) C. F. Baker ex Becc. (syn. Gastrococos crispa (Kunth) H. E. Moore) wvas examined, the cells within were found to have extremely high water content (Fisher et l. 1996). Leaves The leaves are costapalmate; the lamina is split to one third or half of its radius, with up to 130 induplicate leaflets. There is a prominent triangular cleft in the sheath below the petiole. Towards the stem, the leaf sheath disintegrates, forming a network of tough, woody fibres. The petiole is robust with a flat adaxial surface and a rounded abaxial surface; the margins can be armed with spines or unarmed. There is considerable variation in the density, shape, size and colour of the spines, which are influenced by light intensity and the age of the plant (Fig. 1). Adaxial and abaxial hastulae are usually present at the juncture of the petiole and lamina. The lamina is broad, undulating anld stiff towards the base. ? The Board of Trustees of the Royal Botanic Gardens, Kew, 2007 This content downloaded from on Sun, 13 Apr 2014 14:03:57 PMAll use subject to JSTOR Terms and Conditions  A REVISION OF BORASSUS L. (ARECACEAE) 563 InduLnentum may be present on the costa and leaf fold ridges of both surfaces, though it erodes in older leaves. The transverse commissural veins, which link the major parallel veins of the leaf, are distinct and their spacing varies significantly from species to species. The leaflet apices may be acute or cuspidate, bifid or entire. Leaf Anatomy The anatomy of palm leaves was examined by Tomlinson (1961), who noted that in the Borasseae, there was an unusually high incidence of isolateral leaf symmetry. Most palms have leaves with an asymmetrical (dorsiventral) anatomy i.e. a well defined adaxial palisade mesophyll, and stomata primarily on the abaxial surface. However, most Borassus species lack a clearly defined palisade, and have stomata on both surfaces, i.e. isolateral anatomy (Tomlinson 1961). Leaf symmetry is thouLght to be correlated with light exposure (Tomlinson 1961) and/or aridity (Barrow 1998). In the Borasseae, genera from exposed or arid habitats (Bismarckia Hildebr. & H. WVendl., Hyphaene Gaertn. and Medemia Wurttenb. ex H. Wendl.) exhibit isolateral leaf symmetry, while those from shady, humid habitats (Borassodendron, Lodoicea Comm. ex DC. and SatranalaJ. Dransf. & Beentje) show dorsiventral leaf symmetry. Latania Comm. ex Juss., a genus of coastal cliffs and savannas, has a somewhat intermediate leaf anatomy (Tomlinson 1961). Borassus reflects this overall pattern as most species of this primarily arid zone palm have isolateral leaf symmetry, but B. heineanus, from the rain forests of New Guinea, has dorsiventral leaf anatomy. I a I. 3 cm Fig. 1. Borassus petiole margin rmature. Borassus heineanus Papua New Guinea), Banka s.n. (K); B B. flabellifer Thailand), ilkin t al. 1 60 (K); C B. madagascariensis Madagascar), ayton & Ranaivojaona 51 (K); D B. aethiopum Madagascar), ayton Ranaivojaona 53 (K); E B. aethiopum juv., enya), Bayton & Obunyali 14 (K); F B. aethiopum Kenya), ransfield 811 (K); G B. akeassii juv., urkina Faso), Bayton t al. 72 (K); H B. akeassii Burkina aso), Bayton t al. 74 (K). DRAWN Y UCY . SMITH. O The Board of Trustees of the Royal Botanic Gardens, Kew, 2007 This content downloaded from on Sun, 13 Apr 2014 14:03:57 PMAll use subject to JSTOR Terms and Conditions  564 KEW BULLETIN OL. 62(4) Inflorescences Borassus is pleonanthic and dioecious though monoecy occurs rarely as a monstrosity (Louis & Mary 1975). The inflorescences are axillary, shorter than the leaves and the staminate and pistillate forms are superficially dissimilar. The prophyll is the first bract on the inflorescence and is large, bicarinate and coriaceous, however it is rarely collected duLe to its position deep within the leaf axils. NuLmerous large, primary bracts sheath the inflorescence, and these are largely indistinguishable from the prophyll. They often have hard, woody apices, fibrous margins and green longitudinal stripes. Staminate inflorescences are usually branched to one or two orders (Figs 3E, 5E), but pistillate inflorescences are typically spicate (Figs 4A, 5G). The first-order branches of the staminate inflorescence terminate in 1 3(- 5) rachillae that are covered in scale-like rachilla bracts (Figs 3E, 5F). These bracts are imbricate and partially connate and consequently, the rachillae resemble massive catkins. In the axil of each scale is a cincinnus (Fig. 3H) containing several staminate flowers that are obscured in a pit formed by the bracts (Figs 3F - G). Each flower is exserted from its pit sequentia'ly, by elongation of the floral receptacle. Pistillate inflorescences of most Borassus species are typically spicate, though branching is common in B. akeassii (Fig. 5H), perhaps due to human-inflicted damage (Bayton et al. 2006). The flower-bearing portion of the pistillate inflorescence is covered with large cupular bracts. The first few bracts are empty, but subsequent bracts subtend pistillate flowers. Unlike the staminate flowers, pistillate flowers are solitary and superficial on the rachis. Flowers The staminate and pistillate flowers are widely divergent. Staminate flowers (Fig. 3J K) are less than a centimetre long and are subtended by membranous bracteoles. The calyx is angular and membranous with three lobes. The receptacle and corolla are fused for most of the length of the flower, though there are three small, keeled corolla lobes. There are six stamens with very short triangular filaments and medifixed, latrorse anthers. A pistillode of varying size is usually present. The pistillate flowers (Fig. 4A - B) are solitary and much larger than the staminate flowers (approximately 2 x 3 cm). They are subtended by coriaceous bracteoles, and have three distinct imbricate sepals and three distinct imbricate petals. A staminodal ring is usually present. The ovary is tricarpellate with basal septal nectaries and the three stigmatic areas are superficial on the apex. Fruits nd Seeds The fruits (Fig. 4D) are very large (uLp to 35 cm long) with a coriaceoLls epicarp, wvhich varies in colour from green to orange to black. The sepals and petals remain attached to the rachis and persist around the base of the developing fruit up to and incltuding maturity. The mesocarp is pulpy and fragrant, containing many longitudinal fibres. There is a separate endocarp surrounding the locule of each carpel, and this encloses the seed forming a pyrene (Figs 4E - H). There are between one and three pyrenes in each fruit and the number of pyrenes in the fruit has an impact on the shape and size of both the fruit and the pyrenes (Fig. 4C). The pyrenes are woody and are often fibrous, and there is a great deal of variation in shape, size and ornamentation. Some pyrenes are bilobed and may have a deep longitudinal furrow that can penetrate the seed within (Fig. 4E). Other pyrenes have a prominent external crest, the vestigial remains of the link between the endocarps, wvhich divided duLring development. In some specimens of B. heineanus, there are internal flanges of the endocarp that penetrate the seed, though these are invisible externally (Fig. 4H). The seed almost entirely fills the endocarp cavity; the endosperm is homogeneouls and bony and there is a small central hollow. Germination is remote-tubular, and the cotyledonary petiole rapidly extends, penetrating the soil. The embryo is, in effect, buried underground where it is safe from herbivory and small fires (Tuley 1995; Barot & Gignoux 1999). Pollen Ferguson et al. (1986) examined the pollen of seven of the eight genera of tribe Borasseae and incltuded Borassus fiabellifer nd B. aethiopum. The eighth genus Satranala was discovered later (Dransfield & Beentje 1995b) and its pollen has since been examined (M. Harley, pers. comm.). Borassoid pollen is in general, monosulcate, tectate and either elliptical or subcircular in polar view. The aperture is almost as long as the longest axis in polar view (L), though in Borassodendron it is onlv a quarter of the length of L. Supratectal gemmae are present in Borassus (Fig. 2) and Ilyphaene (Ferguson et al. 1986). Slight differences between the pollen of Borassus flabellifer and B. aethiopum were noted; the gemmae of B. aethiopuum were variable in size and a dense band of small gemmae surrounded the aperture. In B. flabellifer the gemmae were of two distinct sizes and the aperture margins wvere smooth (FerguLson et al. 1986). However, the small sample size (one specimen per taxon) probably precludes meaningful conclusions. ? The Board of Trustees of the Royal Botanic Gardens, Kew, 2007 This content downloaded from on Sun, 13 Apr 2014 14:03:57 PMAll use subject to JSTOR Terms and Conditions
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks