Phylogenetics of tribe Sabiceeae (Ixoroideae, Rubiaceae) revisited, with a new subgeneric classification for Sabicea LISE ZEMAGHO 1,2 , SIGRID LIEDE-SCHUMANN 1 *, BONAVENTURE SONK  E 2 , STEVEN JANSSENS 3 , OLIVIER LACHENAUD 3 , BRECHT VERSTRAETE 3 and STEVEN DESSEIN 3 1 Department of Plant Systematics, University of Bayreuth, Bayreuth, Germany 2 Plant Systematic and Ecology Laboratory, Department of Biology, Higher Teachers’ Training College, University of Yaound e I, Yaound e, Cameroon 3 Botanic Garden Meise, Meise, Belgium Received 24 April 2015; revised 4 March 2016; accepted for publication 18 July 2016 Tribe Sabiceeae (Ixoroideae, Rubiaceae) has undergone recent taxonomical changes with the incorporation of the related genera Ecpoma, Pseudosabicea and Stipularia into the type genus Sabicea. We use phylogenetic analysis and morphological data to verify the relationships among members of the tribe, including the most comprehensive taxon sampling of the tribe to date with 74 of 145 species. Sequence data from the nuclear internal transcribed spacer (ITS) and three plastid markers (petD, rps16, trnT–F) were used to infer relationships among the members of the tribe. Individual analyses using maximum likelihood, parsimony and Bayesian approaches reveal several supported clades: the former genus Stipularia is resolved as a monophyletic unit, but Ecpoma is monophyletic only if Sabicea urbaniana and Sabicea xanthotricha are included (corresponding to Sabicea subgenus Stipulariopsis sensu Wernham). Pseudosabicea is biphyletic, with one clade corresponding to section Anisophyllae of Hall e (1964) and the other one to the other sections (Floribundae and Sphaericae) of the genus. Eleven morphological characteristics were recorded for all species studied and seven have been mapped onto the phylogenetic tree to study their evolution in the group and assess their value for the classification of Sabicea s.l. Finally, our study shows that a combination of diagnostic characteristics should be used to differentiate each group and we propose to recognise four subgenera in Sabicea. © 2016 The Linnean Society of London, Botanical Journal of the Linnean Society, 2016, 182, 551–580 ADDITIONAL KEYWORDS: cpDNA – Ecpoma – Hekistocarpa – morphology – nrITS – Pseudosabicea – Stipularia – Tamridaea – Virectaria. INTRODUCTION Tribe Sabiceeae (Ixoroideae, Rubiaceae) was pro- posed for the first time by Bremekamp (1966) in sub- family Cinchonoideae, to accomodate the genus Sabicea Aubl. Bremekamp (1966) excluded Sabicea from tribe Mussaendeae because of its simple (not bifid) stipules, axillary (not terminal) inflorescences and the narrow (not large) testa cells. Several Rubi- aceae taxonomists did not accept Sabiceeae as a sep- arate tribe (Verdcourt, 1958; Hall e, 1961, 1963, 1966; Kirkbride, 1979, 1982; Robbrecht, 1988, 1993) and included Sabicea, with supposedly related genera, in tribe Isertieae or Mussaendeae. Andersson (1996), however, conducted a cladistic analysis using mor- phological characteristics of tribe Isertieae sensu Robbrecht (1988) and resurrected tribe Sabiceeae in which he included Acranthera Arn. ex Meisn. (Meis- ner, 1838), Amphidasya Standl. (Standley, 1936), Ecpoma K.Schum. (Schumann, 1896), Pentaloncha Hook.f. (Hooker, 1873), Pittierothamnus Steyerm. (Steyermark, 1962), Pseudosabicea N.Hall e (Hall e, 1963), Sabicea Aubl. (Aublet, 1775), Schizostigma Arn. ex. Meisn. (Meisner, 1838) and Temnopteryx Hook.f. (Hooker, 1873); a tenth genus included in his cladistic analysis, Stipularia P.Beauv. (Palisot-Beau- vois, 1810), was erroneously omitted from the list. He considered the tribe more closely related to *Corresponding author. E-mail: sigrid.liede@uni-bayreuth.de 551 © 2016 The Linnean Society of London, Botanical Journal of the Linnean Society, 2016, 182, 551–580 Botanical Journal of the Linnean Society, 2016, 182, 551–580. With 9 figures Downloaded from https://academic.oup.com/botlinnean/article-abstract/182/3/551/2707795 by guest on 14 June 2020