580 
Telopea Vol. 6(4): 1996 
Everett (1990) analysed 30 morphological and anatomical characters from 37 taxa. The 
results are summarised in Fig. 1 being the strict consensus tree from PAUP 2.4 (Swofford 
1985). Tlie taxa are: Achmthenmi, 'Boreobtusae' and 'Obtusae' (Barkworth & Everett 1987), 
Nassella, Oryzopsis, Piptalherum sections Piptatlierum, Miliacea (included with Piptatherum 
in Fig. 1) and Virescentia, Piptochaetium, Ptilagrostis, Hespewstipa, Stipa s.s., Stipa sections 
Podopogon (included in Piptochaetium in Fig. 1) and Stephanostipa (mcluded in Nassella 
in Fig. 1), and 23 groups obtained from an analysis of Australian taxa (Everett 1990). 
Nardus and ]oinvillea were used as outgroups. The 23 Australian groups were obtained by 
grouping the species into minimum unequivocal monophyletic units. This confirmed that 
the Australian species represented a monophyletic group (Austrostipa in Fig. 1), a sister 
group to Achnathenim and Ptilagrostis (Fig. 1), supporting a similar more tentative conclusion 
reached by Barkworth and Everett (1987). The rDNA study so far is mdicating that the 
Australian species are more closely related to Nassella than to Achnathenim (Hsiao pers. 
comm.). If segregate genera are to be recognised then it is evident that the Australian 
species can no longer remain in Stipa s.s. Given that there is some conflict as to the closest 
relatives of the Australian species, and that all studies to date indicate that they are all 
more closely related to each other than any are to any non-Australian species, the best 
option is to place them in a new genus, here described as Austrostipa. 
- AUSTROSTIPA 
-PTILAGROSTIS 
- - ACHNATHERUM 
- - BOREOBTUSAE 
- sect. VIRESCENTIA 
- STIPA 
- - - HESPEROSTIPA 
PIPTOCHAETIUM 
NASSELLA 
ORYZOPSIS 
OBTUSAE 
PIPTATHERUM 
JOINVILLEA 
Fig. 1. A dadogram indicating relationships in the Stipeae, based on morphological and anatomical 
data, simplified (several groups removed) and redrawn from the consensus tree in Everett (1990). 
