NOTES ON ARISTEA AITON (IRIDACEAE, NIVENIOIDEAE) - TAXONOMY, CHROMOSOME CYTOLOGY, AND PHYLOGENY

Capsules with long pedicels, a thicker stem, wider leaves, and inflore scences borne on long leafless scapes confirm the current taxonomy tha t maintains Aristea alata separate from A. abyssinica and A. cognata. The two latter, however, are difficult to distinguish. Although some s outhern African plants referred to A. cognata are smaller than tropica l African A. abyssinica, variation within and between populations sugg ests that smaller individuals are depauperate, due to local edaphic or seasonal conditions. I conclude that A. cognata and A. abyssinica are best treated as conspecific, and I do not recommend recognition of in fraspecific taxa. Section Ancipites, to which these species belong, is probably heterogeneous, and the removal of A. nitida to section Clado carpae is recommended. Three species of section Pseudaristea should al so be added to section Cladocarpae. Lectotypes, lacking for sections A ncipites and Pseudaristea, are designated here. New collections of sou thern African A. singularis, only species of section Singulares, chang e our understanding of the species. It is probably not primitive in th e genus, but appears specialized in growth form, inflorescence structu re, seed morphology, and number of seeds per locule. The species is mo st likely related to, and perhaps nested within, section Aristea (= se ction Cyaneae Weim.). Original chromosome counts for seven species, fi ve not before counted, confirm a base number of x = 16 for Aristea.