RELATIONSHIPS AND GENERIC CONCEPTS WITHIN STYPHELIEAE (EPACRIDACEAE)

Citation
Jm. Powell et al., RELATIONSHIPS AND GENERIC CONCEPTS WITHIN STYPHELIEAE (EPACRIDACEAE), Australian systematic botany, 10(1), 1997, pp. 15-29
Citations number
30
Categorie Soggetti
Plant Sciences
ISSN journal
10301887
Volume
10
Issue
1
Year of publication
1997
Pages
15 - 29
Database
ISI
SICI code
1030-1887(1997)10:1<15:RAGCWS>2.0.ZU;2-A
Abstract
A morphological data base of 30 characters has been assembled for 33 t erminal taxa in the tribe Styphelieae sensu Powell et al. (1996), as w ell as four outgroup taxa representing the other major affinity groups previously identified within the Epacridaceae on morphological and mo lecular data. Heuristic and bootstrap analyses provided strong support for the tribe, but indicated a need to modify several long-establishe d generic concepts. Four genera are shown here to be polyphyletic. Six new monotypic genera should be segregated from Astroloma. Brachyloma should be divided into three segregates. Two distinct affinity groups warranting generic status were identified within Cyathodes. The separa tion of the monotypic Western Australian genus, Croninia, from Leucopo gon s.l is supported, and four other separate entities are here recogn ised: four species from section Heteranthus Benth. should be united wi th Lissanthe, and the other three segregates warrant generic status. I n addition, Monotoca is paraphyletic, and should be redefined to incor porate Oligarrhena, while a new genus, 'Pseudactinia', is required to accommodate two new species from Western Australia There seems little justification for maintaining the distinction between Trochocarpa and Decatoca. The pattern of intergeneric relationships resolved suggests that Astroloma s.l., Coleanthera, Conostephium, Croninia, Melichrus, S typhelia and much of Leucopogon s.l. constitute an affinity group, and that Brachyloma s. str. and B. scortechinii are basal within the trib e, but there is only weak support for these aspects of the topology wi thin the data.