Joumal of Plant Biology, December 2007, 50(6) : 644-649 A Phylogeny of Euphorbieae Subtribe Euphorbiinae (Euphorbiaceae) Based on Molecular Data Ki-Ryong Park 1. and Robert K. Jansen 2 ~Department of Science Education, Kyung-Nam University, 449 Wolyoung-dong, Masan 631-701 Korea, 2 Section of Integrative Biology, Plant Resouro~ Center, and Institute of Cellular and Molecular Biology, University of Texas at Austin, Austin, TX 78712 USA Phylogenetic relationships within Euphorbiinae were inferred from our analysis of the 3' end of the chloroplast gene ndhF. A sampling of that subtribe covered 88 species; 3 closely related species from the subtribes Anthosteminae and Neoguillaumi- niinae and the tribe Hippomaneae were included as outgroups. A phylogenetic assessment was carried out using the parsi- mony approach. The relationships revealed via these ndhF data supported the monophyly of subg. Esula, subg. Chamaesyce, subg. Euphorbia, and subg. Lacanthis. However, the polyphyly of subg. Agaloma, subg. Lyciopsis, and subg. Eremophyton also was strongly suggested. The African succulent Euphorbiinae can be divided into primarily two independent groups: 1) spiny succulents, which form a strongly supported clade with three subclades (subg. Euphorbia, subg. Lacanthis, and &fonade- nium+Synadenium); and 2) non-spiny succulents, which consist of sect. &feleuphorbia, sect. Medusae, sect. Anthacantha, sect. Trichadenia, sect. Pseudeuphorbium, sect. Treisia, and sect. Pseudacalypha. In the ndhF tree, the subg. Esula clade is placed as a sister to the rest of the Euphorbiinae. Thus, the origin of the Euphorbia s.I. should be sought within the herba- ceous species of subg. s The core North American endemic Euphorbia groups -- Agaloma, Chamaesyce, and Poinsettia -- are monophyletic and independent of the South American subg. Agaloma. Instead, they are derived from the African Euphor- bia subg. Lyciopsis and Eremophyton. The Eurasian subg. Esula (lade forms two subclades, which are concordant to sect. Esula and sect. Tithymalus. Keywords: euphorbiaceae, ndhF data, phylogeny,subtribe euphorbiinae Approximately 2000 species of the subtribe Euphorbiinae are distributed worldwide; its herbaceous members, how- ever, are primarily found in Eurasia and America. The largest proportion of succulent Euphorbia (ca. 500 species) occurs in the drier regions of southern and eastern Africa (Carter, 1988; Govaerts et al., 2000). The subtribe includes the spe- cies with a true cyathium, i.e., having a completely fused involucre with glands on the rim. Because of this unique structure, most species of Euphorbiinae have been treated as members of the genus Euphorbia s.I. ]-he naturalness of the subtribe has been supported in recent phylogenetic analyses based on data both morpho- logical (Park and Elisens, 2000; Park and Backlund, 2002) and molecular (Steinmann and Porter, 2002). However, the classification within this subtribe remains problematic, and until now, only a few taxa, e.g., Euphorbia, Monadenium, 5ynadenium, Elaeophorbia, and Pedilanthus, have been seg- regated as independent genera (Carter, 1988). Although most of these are monophyletic, they have been nested within the Euphorbia s.s. in recent phylogenetic studies. Thus, the recognition of these groups as genera has resulted in the Euphorbia s.s. becoming a paraphyletic group (Park and Elisens, 2000). Generic systems within the subtribe can be traced back to treatments by 14aworth (1812), Rafinesque (1836), and Klotzsch (1859), 'who divided it into 15, 25, and 18 genera, respectively. Nevertheless, most recent classification systems in Euphorbia systematics do not accept these, but rather have adopted the subgeneric and sectional system as pro- *Corresponding author; fax + 82-055-249-2014, e-ma park@kyungnam ac.kr posed by Wheeler (1943), Boissier (1862), and Pax and Hoff- man (1931). Recently, however, Gilbert (1987), Park (1996), and Steinmann and Porter (2002) have criticized most of the sections and subgenera previously proposed. Thus, a strong phylogenetic signal is crucial to the establishment of a robust classification system within Euphorbiinae and to understand- ing their evolution. Molecular phylogenetic examinations of the Euphorbia s.I. have been newly undertaken. For example, Steinmann and Porter (2002) have used ITS and nclhF sequence data to evaluate Euphorbieae, while Ritz et al. (2003) have con- ducted phylogenetic analyses of the subsect Meleuphorbia based on ITS and four noncoding cpDNA regions. Likewise, Haevermans et al. (2004) have determined the ITS sequences for the Euphorbia subg. Lacanthis species. In the current study, we used ndhE data to conduct a phy- Iogenetic analysis of 88 Euphorbiinae species. Our objective w~ to elucidate the relationships and test the monophyly of traditional taxa within Euphorbiinae. The resulting phyloge- netic framework provides a strong guideline for establishing a new classification system for this problematic subtribe. MATERIALS AND METHODS 'We investigated 88 species representing most of the sub- genera and sections of the subtribe Euphorbiinae. Three closely related species from subtribes Anthosteminae and AFP endix'. Eu phorbiinae taxa, their, representative species,, voucher. information, and GenBank accession for phylogenehc study using ndhE 644