Edwards is ornithology curator of the world's fifth largest collection at the Harvard Museum of Comparative Zoology in Massachusetts.

Founded in 1859 by British botanist Thomas Nuttall, it's an impressive, encyclopaedic collection of more than 7000 species, and 400,000 individual specimens, representing about 85 per cent of the world's bird species.

Among its treasures are birds that have gone extinct in recent decades - Mexico's imperial woodpecker and Florida's dusky seaside sparrow - as well as specimens collected and prepared by the great American frontier bird artist John James Audubon.

But, as Edwards points out, museum collections are far from lifeless. They can reveal a wealth of fine-scale detail about genetic variation and link geographic regions with genetic diversification.

''Look at these - you can see the geographic variations in some of the subtle differences,'' he says, carefully picking through a tray of grey-crowned babbler specimens at the Australian National Wildlife Collection in Canberra.

''You get not-so-subtle differences in some species, like different plumage patterns and colours.''

Earlier this month, Edwards visited the University of Canberra for a series of workshops and lectures on the molecular ecology of birds. Currently professor of evolutionary biology at Harvard University, he also heads the Edwards Laboratory at Harvard which specialises in the population genetics, geographic variation (known as phylo-geography) and evolutionary biology of birds. Busy? Well, he's also on the editorial boards of several academic journals, a member of the Council of the American Genetic Association, and is on the National Geographic Committee for Research and Exploration.

And as one of only a handful of high-achieving African-Americans involved in higher-level environmental science research, he's also initiated a program to increase student diversity in environmental and evolutionary sciences.

''There's an under-representation generally of African-Americans in science,'' he says. ''We've made good advances toward more diversity in law and medicine but the current lack of diversity in the environmental sciences can be disheartening for African-American students. A lot of students need to have good role models - people who will encourage their research interests, be a mentor and support them through the inevitable difficulties that arise with any research career …

''Even if we could increase that diversity by as little as 4 or 5 per cent, that would still make a difference.''

Edwards has been a regular visitor to Australia since 1986, working chiefly in remote areas of Cape York and the Kimberley. As a first-year doctoral student at the University of California, he spent 10 months as a volunteer ecologist in Papua New Guinea researching birds of paradise.

''Of course, it's just a short distance from Australia, and when I came here for the first time, I just loved it,'' he says. ''So I thought, 'Hmm, how can I drum up a project here? What's a group of birds that would have a strong interest?' So I looked around, and Julian Ford [a West Australian ornithologist] suggested grey-crowned babblers, so I'm greatly indebted to him.

''They were a great choice, because they're a relatively common species which meant I didn't have to spend years finding them. And that meant I could probably get to understand their biology within the timespan of a dissertation.''

His doctoral research looked at the genetics and population structure of grey-crowned babbler populations. A later research project also detected differing genetic patterns and wide variations in plumage and calls across the northern geographic range of the birds. Edwards explains that when isolated populations of birds form in a landscape, one of the first traits to change is song - ''just like geographical differences in human language''.

Over the past decade, he's been studying genetic divergence in bird populations across the Gulf country's savanna in north-west Queensland. There's a region known as the Carpentarian Barrier - ''a line running south from the Gulf through Normantown'' - which appears to mark an evolutionary population divergence for a number of species, including grass finches, wrens and treecreepers.

''It's intriguing, because if you drive across it, there aren't immediately visible differences,'' he says. ''It's more open, there are fewer trees and the vegetation on either side looks quite similar. It's very subtle, nothing that would shout out at you, 'Things cannot pass through this barrier!' ''

''There are some possible culprits, such as desertification, that may have caused genetic paths to diverge. It's possible sea levels may have risen and fallen and that may have served to isolate populations.''

Using DNA samples, Edwards and research colleagues can measure genetic differences on both sides of the barrier. They can also establish if populations diverged at around the same time, or if successive waves of genetic divergence occurred. The value of this research lies in mapping the genetic geography of Australia's biodiversity, he says.

''It's the geography of genetic diversity. There's a lot of interest in maintaining biodiversity, particularly given the impacts of climate change, and one of the goals of conservation is to preserve areas of habitat that have a high level of genetic diversity…

''There's also a growing interest in - and it's a heady concept - of preserving evolutionary processes, and setting aside areas where evolution is proceeding rapidly. So that's where phylo-geography can be beneficial, in mapping areas of genetic divergence.''

The perspective on conservation is changing, he says. Last century, writers like Aldo Leopold were ''ruminating on the importance of species'' and the need to preserve ecosystems as nature reserves.

''Now, the question has shifted to how do we keep these ecosystems functioning? How do we mange them against the onslaught of invasive species and climate change, and maintain their diversity, fertility and ecosystem function?

''We've lost the preservationist battle. It's no longer a viable option to try to restore these places to what they once were, because every habitat has experienced some sort of insult. We have already gone too far.''

■ Rosslyn Beeby is science and environment reporter.