The tree of life may be in need of some radical surgery. New ideas suggest the last common ancestor - the ancient organism from which all known life has descended - may have been a more complex entity than most scientists had assumed.
Furthermore, the study suggested that the birthplace of life was more likely to have been a warm little pond than the boiling chemical soup around ocean-bottom hydrothermal vents that many believe it was.
The commonly held view is that the first lifeforms on Earth, 3.5 billion years ago, were single-celled organisms with a free-floating loop of DNA inside, possessing the same genetic mechanism that all life possesses today.
This branch of the tree of life is made up of the prokaryotes that include bacteria.
The tree branched twice to form the archaea - recently discovered single-celled organisms that have many cellular and genetic differences from bacteria - and, later still, the eukaryotes - cells with nuclei, such as those that make up plants and animals.
However, this progression towards more complex life is not the only way to explain the meagre evidence scientists have gleaned about the roots of the tree of life.
Nicolas Glansdorff, professor of microbiology at the Flemish University of Brussels, Belgium, has proposed that many of the clues about the identity of the last common ancestor, extracted from statistical analyses of genetic material, may point in a very different direction.
His hypothesis, developing ideas proposed by Patrick Forterre, suggests that the ancestor may have had a far greater array of genes and metabolic tricks than previously assumed, with more in common with the eukaryotes than the other two, more primitive, branches of the tree of life.
Glansdorff believes the archaea and, later, bacteria evolved to become simpler versions of their common ancestor to compete more efficiently in different environments.
His argument hinges on the interpretation of genetic similarities between different members of the three branches, found in recent comparative studies. Glansdorff rejects the suggestion that this shows very different forms of life "swapped" genes.
Instead, he proposes that the last common ancestor may have had a large number of duplicated genes. As the descendants diverged, some lost one set of such copies while others lost different ones.
The work is published in Molecular Microbiology .