To determine accurately the rates of late Pleistocene megafaunal loss, it is fundamentally important to have accurate taxonomic information for every species. In Australia, accurate taxonomic information is lacking for several Pleistocene groups, including the largest marsupial ever to live, Diprotodon Owen, 1838. Diprotodon taxonomy has been complicated by early nomenclatural problems and by the occurrence of two distinct size classes of individuals that do not reflect an ontogenetic series. Traditionally, the two size classes have been regarded as separate species. However, a taxonomic investigation of large samples (> 1000 teeth) of Diprotodon material from several different fossil localities in Queensland, New South Wales, South Australia and Victoria suggests that there is little evidence for the discrimination of more than one morphospecies. Thus, Diprotodon is here considered a monotypic genus and the single morphospecies, D. optatum Owen, 1838 is considered to have been highly sexually dimorphic. By drawing analogy with extant sexually dimorphic megaherbivores and marsupials, the large form was probably male, and the small form was probably female. Diprotodon optatum probably moved in small, gendersegregated herds, and exhibited a polygynous breeding strategy. As a single morphospecies, D. optatum had a near-continental geographical distribution, similar to that of extant megaherbivores, possibly indicating its niche as a habitat generalist.