Acute rheumatic fever (ARF) remains a major problem in tropical regions, resource-poor countries, and minority indigenous communities. It has long been thought that group A streptococcal (GAS) pharyngitis alone was responsible for acute rheumatic fever; this belief has been supported by laboratory and epidemiological evidence gathered over more than 60 years, mainly in temperate climates where GAS skin infection is uncommon. GAS strains have been characterised as either rheumatogenic or nephritogenic based on phenotypic and genotypic properties. Primary prevention strategies and vaccine development have long been based on these concepts. The epidemiology of ARF in Aboriginal communities of central and northern Australia challenges this view with reported rates of ARF and rheumatic heart disease (RHD) that are among the highest in the world. GAS throat colonisation is uncommon, however, and symptomatic GAS pharyngitis is rare; pyoderma is the major manifestation of GAS infection. Typical rheumatogenic strains do not occur. Moreover, group C and G streptococci have been shown to exchange key virulence determinants with GAS and are more commonly isolated from the throats of Aboriginal children. We suggest that GAS pyoderma and/or non-GAS infections are driving forces behind ARF in these communities and other high-incidence settings. The question needs to be resolved as a matter of urgency because current approaches to controlling ARF/RHD in Aboriginal communities have clearly been ineffective. New understanding of the pathogenesis of ARF would have an immediate effect on primary prevention strategies and vaccine development.