A new study refutes the notion that the dinosaurs' extinction led to an explosion of new mammal species.

The surprise finding overturns the widely held belief that the ancestors of modern mammals were able to quickly evolve and spread to fill many of the empty niches left behind following the mass extinctions of dinosaurs and many other large animals when a huge asteroid crashed into the Earth about 65 million years ago.

An international research team reached that conclusion after analysing the evolutionary links of some 4,500 mammals, creating for the first time a "supertree" of family relationships between almost all species of mammals alive today.

Armed with the information about those relationships, the researchers used DNA data and the fossil record to estimate diversification rates and work backwards to establish when specific groups of mammals first appeared on Earth.

The study, which included work by Robin Beck, a PhD student in the UNSW School of Biological, Earth and Environmental Science, suggests that while some early mammals may have benefited from the demise of the dinosaurs, many were on branches of the family tree not closely related to present-day mammals and died off early on.

They also found that modern mammal orders, such as primates, rodents and hoofed animals, did not diversify until much later, at least 10 to 15 millions years after the mass extinctions at the boundary between the Cretaceous and Tertiary periods, when dinosaurs became extinct.

"Modern mammals went though an early period of diversification about 93 million years ago, when the major groups - the superorders and orders - appeared, but after that the rate dropped and stayed fairly low for the next 40 million years. It was other groups of mammals, not those we see today, that took advantage of the extinction of the dinosaurs," says Mr Beck.

The research, which appears in the current issue of the journal Nature and was led by Olaf R.P. Bininda-Emonds of the Technical University of Munich and Andy Purvis of Imperial College in London, brought together a diverse group of scientists to produce the first comprehensive analysis of mammalian biodiversity from about 160 million years ago to the present day.

Supertrees are a kind of summary of evolutionary history for a large group of organisms constructed from many, smaller studies for separate groups based on genetic or physical analysis or both.

They let scientists trace the evolution of groups of species and map the relationships among organisms. With the growth of molecular analytic techniques, the construction and study of supertrees has required ever-greater computing power, as well as contributions from an increasingly diverse group of scientists.

"The big question now is what took the ancestors of modern mammals so long to diversify," said Ross MacPhee, a curator in the Division of Vertebrate Zoology at the American Museum of Natural History.

"It's as though they came to the party after the dinosaurs left, but just hung around while all their distant relatives were having a good time. Evidently we know very little about the macroecological mechanisms that play out after mass extinctions."

In collaboration with other researchers around the world, the team is now using the supertree to answer a whole range of questions about the evolution of mammals.

"The supertree itself is really just the first stage," says Mr Beck. "The information it provides allows us to look at the overall pattern of mammalian evolution in far greater detail than before. It has applications in ecology, conservation, physiology, palaeontology, amongst other fields, and it will also shed new light on the evolution of our own species - it's a big step forward."