Thus, it appears, from the intermatch distributions, that the Aboriginal Australian and southern Indian populations derive from the same ancestral population, whereas the highland PNG [Papua New Guinea] population derives from a completely different ancestral population.
Furthermore, the net separation between Aboriginal Australian populations and southern Indian populations appears to be much more recent than the separation between Aboriginal Australian populations and PNG highland populations. The precision of the estimated divergence times should be considered somewhat cautiously, since these estimates are associated with large uncertainties. However, the patterns are consistent with a separate origin (or ancient separation) for PNG highlanders and Australian Aboriginals and with recent genetic affinities between southern Indian populations and Aboriginal Australian populations.Interestingly enough, they take special note of Birdsell's supposedly discredited theory:
These findings are somewhat consistent with Birdsell’s trihybrid model for the peopling of Sahul . . . Birdsell hypothesized that Oceanic “Negritos” first populated Sahul, but that two later migrations replaced most of them in Australia but not in the Cairns area of northeast Queensland or in Tasmania and New Guinea. According to this model, the second migration of populations, with affinities to the Ainu of Japan, dispersed throughout Australia, whereas the third migration of populations, with affinities to tribal populations of India, entered northern Australia around the Gulf of Carpentaria. The gene tree in the present study shows that the PNG3 cluster shares sites with African sequences, a finding that may be consistent with Birdsell’s first-migration hypothesis. Our results also suggest that there may have been a migration(s) from an Indian source that reached Australia but not PNG. However, our results do not support two distinct source populations for the subsequent peopling of Australia, because Aboriginal Australian populations cluster together with southern Indian populations. . .Their summary is especially relevant for our purposes, as it highlights the connection they found between highland Papua New Guinea and Africa:
To summarize, our data indicate that the PNG highlanders contain distinct and divergent mtDNA, with evidence of ancient African ties, that were rare or absent in Aboriginal Australians and suggest a possible recent connection between Aboriginal Australian populations and populations from the Indian subcontinent (824 -- my emphases).10. In addition to the Tasmanians, another very interesting population can be found in the islands of the Torres Straits, separating the northernmost Cape York Peninsula region of Australia from New Guinea:
The islands' indigenous inhabitants are the Torres Strait Islanders, Melanesian peoples related to the Papuans of adjoining New Guinea. The various Torres Strait Islander communities have a distinct culture and long-standing history with the islands and nearby coastlines. Their maritime-based trade and interactions with the Papuans to the north and the Australian Aboriginal communities have maintained a steady cultural diffusion between the three societal groups, dating back thousands of years at least (my emphasis).It seems surprising to see a people with a fundamentally Melanesian morphology and culture living in such close proximity to Australian aborigines, yet maintaining, over thousands of years, an almost total independence of outlook. This is reflected in their musical culture as well, as their singing tends to be open-throated and polyphonic, in striking contrast to the tense voiced solos and unison choruses of their close neighbors to the south. Once again, as with Tasmania, we see the possibility that the islands of Torres Strait could have served as a refuge area.