Systematics , classification and phylogenetics


This palm group was first described by the celebrated  German physician and botanist Joseph Gaertner (1732-1791), in his publication De Fructibus et Seminibus Plantarum (1788) (Fig.1). At that time the species Hyphaene coriacea was described as the first taxon within the genus.

Fig. 1. Original description of the genus Hyphaene in De Fructibus et Seminibus Plantarum (1788).

Important contributions in the systematics of the group were carried out by two German botanists in palm science. The first one is the Bavarian palm expert Carl Friedrich Philipp von Martius (1794-1868, Fig. 2), who contributed with the description of several taxa in Hyphaene (i.e. H. petersiana) and made some generic transfers, in particular from the genus Corypha. The botanist and also gardener Hermann Wendland (1825-1923, Fig. 2) described also many important taxa in the group (i.e. H. compressa, H. macrosperma).

Fig. 2. Carl Friedrich Philipp von Martius (Photo courtesy of Botanische Staatssammlung München) and Hermann Wendland (Photo courtesy of Landeshauptstadt Hannover, Fachbereich Herrenhäuser Gärten)

The last taxonomic treatment of Hyphaene was published 90 years ago by Beccari (1924), who interestingly never saw Hyphaene in the wild (Fig. 3).  Taxonomical notes for some species were proposed by Furtado (1967, 1970a, 1970b, 1970c), who most probably only had seen H. dichotoma in India (Fig. 3). Working almost exclusively with herbarium material, these authors proposed more than 2/3 of the taxonomic names associated to the genus, most of these to the ranks of species and subspecies.
Some notes on the West African species were offered by Chevalier and Dubois (1938), and a partial treatment for the genus including the East African species was published by Dransfield (1986). A recent publication by Stauffer et al. (2014) briefly updates our knowledge of the tropical African taxa. The number of currently recognized synonyms for some taxa is impressive (H. compressa: 33; H. coriacea: 24; H. petersiana: 21; H. thebaica: 17), which reveals the tortuous taxonomic and nomenclatural history of the genus and highlights the urgent need to re-evaluate the status for most of the names therein included. A complete revision of H. guineensis was undertaken by van Valkenburg and Dransfield (2004), clarifying the taxonomic identity of this palm and providing data on its ecology and distribution.

Fig. 3. The Italian botanist Odoardo Beccari (1843-1920), left, and the Indian botanist Caetano Xavier Furtado (1897-1980), right, described most of the taxonomic diversity in the genus Hyphaene. Photo of OB courtesy of Riccardo Baldini (Florence Herbarium).

Phylogenetics of Hyphaene and Hyphaeninae

The placement of Hyphaene in the palm subfamily Coryphoideae was first proposed by Uhl and Dransfield (1987) and supported strongly in later studies relying on molecular phylogenetics (Dransfield et al., 2008). Modern molecular phylogenies confirm Hyphaene as a highly supported monophyletic group (Bayton, 2005; Asmussen et al., 2006; Baker et al., 2009, Fig. 4), with moderate to high support for a sister-relationship to the monotypic Medemia, endemic to the Egyptian Nubian Desert and north-eastern Sudan, (Bayton, 2005; Asmussen et al., 2006; Baker et al., 2009).

Fig. 4. Phylogenetic position of Hyphaene within Coryphoideae. Strict consensus of the 46 080 trees from the supermatrix analysis of all palm genera (tree length = 15 173, CI = 0.41, RI = 0.62). Values above branches are BPs (>50%). Adapted from Baker et al. (2009).

An NGS phylogenetic approach is now being undertaken at the laboratory of the Unité de Phylogénie et Génétique Moléculaires of the Conservatory and Botanic Gardens of Geneva. This work is leaded by Dr. Mathieu Perret, Dr. Yamama Naciri and Dr. Camille Christe, technically supported by Mrs. Regine Niba. The first 60 DNA samples include 6 species of Hyphaene represented by specimens from more than 12 African countries. The sister group Medemia and other taxa of Borasseae (i.e. Borassus, Bismarkia) and other Coryphoideae have been also included in these first analyses. A simplified working pipeline proposed by Dr. Camille Christe is presented in Fig. 5.

Fig. 5. NGS pipeline proposed for the Hyphaene project.

Available phylogenies support a sister relationship between Hyphaene and Medemia (Fig. 6) within the tribe Hyphaeninae (sensu Dransfield et al., 2008). Medemia (1-2 spp.), longtime believed to be extinct, was rediscovered but the populations remain heavily threatened. This palm is endemic to the Nubian Desert, in the border between Egypt and Sudan. This palm can be distinguish from Hyphaene by its hastula completely lacking and its petioles much less armed than in Hyphaene. More recently Morcote, Raz and Stauffer (Project on African Phytoliths) found morphological differences between the phytoliths of Medemia and those observed in Hyphaene (Fig. 7). Leaf surfaces (presence vs. absence of trichomes) in Borasseae seems to be informative to taxonomically separate some genera and this is a good character to observe in living and herbarium material. Whereas the presence of conspicuos trichomes is evident in Hyphaene and Medemia , they are almost completely absent in Borassus (Fig. 8).

 Fig. 6 Medemia argun. Growth habit in southern Egypt (Photo Bill Baker); detail of petiole (top right); detail of fruits (bottom-right, specimen at the K Herbarium).


Fig. 7. Phytoliths in Hypharninae. Medemia argun (left); Hyphaene coriacea (right). Project "Phytoliths in African Palms" (Gaspar Morcote, Fred Stauffer and Lauren Raz - Universidad Nacional de Colombia, Conservatory and Botanic Gardens of Geneva).

Fig. 8. Leaf surfaces in Borasseae. The presence of conspicuos trichomes is evident in Hyphaene (left) and Medemia (center), whereas they are almost completely absent in Borassus.

Go to top