The 'tree of life' is the single evolutionary 'tree' that links all living, and extinct, plants, animals, and microbes. Constructing the tree is one of the major enterprises in evolutionary biology - a substantial, international research programme involving thousands of biologists.

The scientific aim is the same as that set out by Charles Darwin - to understand where life came from, the shape of evolution, and the place of humans in nature. It also hopes to determine the extent of modern biodiversity and where it is threatened.

A key problem in constructing the tree is to date the times when major 'branching' events happened. For example, when one group split into two such as when humans separated from apes. There are two lines of evidence to can be used to reconstruct the shape of the tree - fossils and molecules.

Molecules can be used to calculate the shape of the tree - what is related to what, and how closely. So, if one branching point in the tree is calibrated against a known date, the other dates can be estimated.

In their paper, Professor Mike Benton, a palaeontologist from the University of Bristol and Professor Francisco Ayala, a molecular biologist from the University of California at Irvine, show that while these dates often agree, molecular dates can sometimes be up to twice as old as fossil dates.

For example, molecular dates estimate that modern birds first appeared between 70 and 120 million years ago, however, the oldest fossil of a modern bird is dated at only 65 million years. On the other hand, both fossils and molecules show that humans split from apes about 6-7 million years ago.

Clearly fossils will always be under-estimates of actual dates since it is highly unlikely that the first member representing a new branch on the tree will ever be found fossilised. Similarly, if fast-evolving genes have been unwittingly used to estimate the molecular dates, then dates obtained from that will always be too old.

Clearly, both methods have their strengths and weaknesses. By recognising this and using both to assess each other, closer agreement can usually be found. So there is no need for fossil and molecular experts to fight - after all, only one date can be right!

'Dating the Tree of Life' by Michael J. Benton and Francosco J. Ayala is published in Science on 13 June 2003.