In the interval leading up to the coldest and driest phase of the last glacial cycle environment pressures on large mammal presumably increased as Australia became more arid. Many people have suggested that this changed at least contributed to the late Pleistocene extinctions (Arch 1984; Field & Dodson 1999; Horton 1984; Kohen 1995). The effects on megafauna of the deterioration ice age climate have been conceived in two major ways. First, the major impact could have come from an increase in climate variability, as might well have accompanied the transition from one climate regime to another. Main (1978) suggested that an unstable climate would have been to the disadvantage of large-bodied species because of their generally low rates of population growth. A population of a small-bodied species knocked down by an extreme climate event, like a severe drought, might be able to recover while the next one hit; populations of large-bodied species, unable to rebound quickly, could be driven down to very small numbers and ultimately to extinction by a series of extreme events.

Second, a general reduction in the rainfall would have reduced the avail-ability of drinking water and the productivity and nutritional quality of vegetations. Large mammals are the most vulnerable to those changes because of their large-requirements for food and water.

Merrilees(1968) and Jones (1968) argued the landscape burning by Aboriginal people brought about major changes to Australian vegetation, which if not the sole cause were a very significant part of the complex of factors that drove the megafuana extinct. Surely, the argument went, this level of use of fire must have changed the vegetation dramatically over practically the whole of the continent, and it may have destroyed the habitat of many megafauna species. Many of the extinct megafauna seem to have been animals of open grasslands, woodlands and shrub lands. When Europeans first came to Australia they found vast areas of such habitats under Aboriginal burning. If there is no reason to think that species like Diprotodon optatum and procoptodon goliah needed radically different habitats to those that existed over most of the continent two hundred years ago, it is hard to imagine how a changed in habitat cause by Aboriginal burning could have wiped them out. In many cases the most sensible view might be that burning favoured them, as it does some of the largest herbivores in Australia today. As Horton(2000) said ‘If creating grasslands in Australia was good for cows and sheep [it] would have been just as good for the Diprotodons.”

A more subtle version of the hypothesis is as follows. Before the arrival of people many habitats of inland Australia had a complex layer of shrubs and small trees, which was important to the large number of megafauna species that were browsers. An increase in burning removes or simplified this vegetation layer over large areas. Nonetheless, landscape burning continues to be invoked in general terms as a factor that might have contributed to at least some of the extinction.