The taxonomic treatment for the grasses of southern Africa was one of the firstto be based on computerised data and the DELTA system. These data, based on over 70,000 herbarium records, are amenable for analysis of species parameters includingabundance, frequency and distribution. This information is suitable for the allocation of species into the seven categories of rarity proposed by Rabinowitz using a combination of habitat specificity ("Narrow" or "Broad"), population structure ("Sparse" or "Abundant")and distribution ("Restricted" or "Widespread"). We compare the species lists obtainedfor each combination of these three aspects to published Red Data Lists (RDLs) forsouthern and South Africa. Ninety-three species are placed in the most sensitive orpotentially threatened category (Narrow habitat, Sparse populations and Restricted dis-tributions; RSN). This is substantially more than the number of species listed in currentRDLs for the region. Chi-square tests indicate a statistically significant bias in taxa fromthe Fynbos Biome for three of the categories (RSN, RAN and WSN), from the SavannaBiome for the WAN category and from the arid Succulent Karoo and Desert Biomes forthe RAB category. Analyses of habitat requirements indicate that many grasses listed(especially those associated with a "Narrow" habitat) are found in some form of wetlands (ephemeral or permanent), especially those at higher altitudes (montane). Despite concernsabout the subjective nature in determining the boundaries between the categories, thismethod is shown to provide a meaningful and valuable list of taxa that require prioritisationfor more detailed assessment according to the IUCN criteria. Keywords: Rabinowitz categories, Grasses, Hygrophytes, Poaceae, Rarity, Red Data List, Southern Africa, Wetlands.