606 
Telopea Vol. 6(4): 1996 
Plinlhanthesis has a complicated nomenclatural history. Bentham (1878) already 
separated it as Danthonia sect. Micmnlhera. Blake (1972) elevated this group to generic 
level, taking up Steudel's generic name Plinthanthesis, which he lectotypified by 
PIinthanthesis iirvillei Steud., while transferring P. temiius Steud. to Notodanthonia. 
Conert (1975) explicitly synonymised Plinlhanthesis under Danthonia, by transferring 
the type species, P. urvillei Steud. to Danthonia, and Veldkamp (1980) challenged the 
typification of Plinlhanthesis, an argument which was rejected by Jacobs (1982) and 
Connor and Edgar (1981). Our analysis supports the distinctness of Plinthanthesis, 
supported by four characters: villous palea margins, puberulous palea backs, and 
the absence of bulliform cells in the adaxial leaf epidermis. Blake used the 
characteristic brown colour of dried specimens, the remarkably long rhachilla 
internodes, the palea longer than the lemmas, and the glumes with prominent 
midnerves to separate Plinthanthesis from Danthonia s.s. and from Chionochloa, while 
Rytidosperma was separated from Plinthanthesis by a punctiform hilum. 
Joycea 
The two species of joycea consistently form a distinct group, robust to successive 
weighting, often placed as sister-group to the rest of the Rytidosperma clade. These 
two represent a group of three species (of which two were used in the analysis): 
Joycea pallida, j. clelandii and the recently described /. lepidopoda. These three species 
lack the distinct tufting in the upper row of lemma hairs typical of Rytidosperma s.l., 
and also have more linear caryopsis hila, and the group is defined by its longer, 
deep red anthers. 
Although this group has not been recognised before, J. pallida has always been 
considered to be taxonomically anomalous, and has consequently been placed in 
several genera: Veldkamp (1980) transferred the species to Notodanthonia, a position 
disputed by Connor and Edgar (1979), and Jacobs (1982) transferred the species to 
Chionochloa. Simon (1993), however, retained /. pallida in Danthonia. Vickery (1956) 
placed J. clelandii next to D. pallida, indicating the close relationship between them. 
Walsh (1991) also indicated the close relationship between his ]. lepidopoda and 
/. pallida. Ecologically, the three species are quite distinct from the rest of the 
Rytidosperma clade, and especially j. pallida is not commonly regarded in Australia as 
a member of the genus Rytidosperma. 
The morphological and ecological evidence indicates that this lineage may best be 
separated from typical Rytidosperma, so we separate the three species into a new 
genus, named Joycea after Joyce Vickery, who did such excellent work on the 
Australian grasses. 
Rytidosperma s.l. 
A clade including Rytidosperma sensii Zotov, Erythranthera, Pyrrhanthera, Monostachya 
and Karroochloa is retrieved by all analyses, and is robust to successive weighting. 
This clade is based largely on the lemma indumentum, which is more or less tufted, 
at least in the upper indumentum row. In addition, the hilum is more or less punctate. 
The hilum character is somewhat variable, as may be expected in such a large clade. 
Generally, the hilum is not truly punctate, but rather ovate to elliptical, and its size, 
relative to the caryopsis, is also variable. However, it is not linear, as is typical of so 
many other danthonioid genera. Consequently, simple measurements do not always 
capture the distinction between the different hilum types. It should be noted, though, 
that there are exceptions within both Rytidosperma s.l. and the rest of the tribe 
Danthonieae, so that this character is not absolute. 
