Generic limits in the Rytidosperma 
(Danthonieae, Poaceae) complex 
H. Peter Linder and G. Anthony Verboom 
597 
Abstract 
Linder, H. Peter and Verboom, G. Anthony (Bolus Herbarium, University of Cape Town, Rondebosch 
7700, South Africa) 1996. Generic limits in the Rytidosperma (Danthonieae, Poaceae) complex. Telopea 
6(4): 597-627. The generic limits of the Australasian danthonioid grasses are re-assessed using 
morphological and anatomical data from virtually all danthonioid species. These data, encoded 
in the DELTA system, were analysed using parsimony. The position of these genera within the 
Poaceae, and the broad structure of the relationships between them, has been assessed previously 
using both embryological information and molecular analyses. The results of the present study 
support the recognition of Plinthanthesis, Notochloe and Schismus, as well as corroborating the 
distinction between Danthonia and Rytidosperma. Rytidosperma sensu Zotov is segregated into 
three genera: Rytidosperma s.s. with 35 species, which includes Monostachya, Erythranthera and 
Pyrrhanthera; Notodanthonia with 28 species; and two new genera, Thonandia (5 species) and 
Joycea (3 species). These genera are characterised morphologically and ecologically, and the 
species included in each genus listed. The relevant new combinations are made. 
Introduction 
The last comprehensive, critical taxonomic study of the Australian danthonioids was 
by Vickery (1956), whose exemplary account of the species boundaries and 
nomenclature laid a solid foundation for further work. Vickery placed all species in a 
very broadly defined Danthonia, a concept which was current at that time. Within a 
few years, however, the dismantling of this large, unwieldy genus had begun. Zotov 
(1963) initiated this process by segregating the New Zealand species into four genera: 
Chionochloa, Notodanthonia, Erythranthera, and Pyrrhanthera. During the next decade, 
Conert, following the results of De Wet (1956, I960), removed most of the African 
species into separate genera: Dregeochloa (Conert 1966), Karroochloa (Conert & Tiirpe 
1969), MerxmueUera (Conert 1970, 1971), and Pseudopentameris (Conert 1971). The 
Australasian segregation was finally completed when Blake (1972) recognised 
Notodanthonia, Plinthmthesis and Monachather, thus leaving no Danthonia species native 
to Australasia and southern Africa. Blake did not make the relevant combinations for 
Notodanthonia: that was eventually done by Connor and Edgar (1979). Nicora (1973) 
completed the redelimitation of IDanthonia by removing six South American species 
from Danthonia to Rytidosperma. This reduced Danthonia from a ubiquitous genus of 
some 150 species, to a small genus of ca. 23 species, ranging from South America (9 
species), through North America (8 species) to Europe (3 species) and North-East 
Africa (2 species) and Asia (2 species). Subsequently, Connor and Edgar (1979) 
transferred all the Australasian species of Notodanthonia to Rytidosperma. 
This fragmentation of Danthonia has not gone unchallenged. The separation of 
Chionochloa has been generally accepted (Blake 1972, Conert 1975, Connor 1991, 
Jacobs 1994): Chionochloa is sharply discontinuous from the rest of Danthonia s.l. by 
its bizarre leaf anatomy, shaggy three-rowed lemma indumentum, and the shape of 
the paleas (Linder in prep.). Although the genus is clearly danthonioid, it is not 
closely related to the Rytidosperma-Danthonia clade (Linder in prep.). There is one 
