TINY tubes in lava 3.5 billion years old may be the world鈥檚 oldest trace fossils, pushing back hard evidence for life by around 800 million years. Hints of life from even further back have been seen in isotope traces, but till now the oldest undisputed fossil evidence has dated from only 2.7 billion years ago.
Harald Furnes of the University of Bergen in Norway is basing his claim on the similarity between tubes in ancient rock and holes drilled by microbes in modern seafloor lavas. But the evidence does not convince Martin Brasier, the University of Oxford palaeobiologist who two years ago cast doubt on claims of fossil bacteria dating back 3.5 billion years (快猫短视频, 9 March 2002, p 5).
Furnes and his team studied pillow lavas 鈥� which are formed when they erupt on the seafloor 鈥� in the 3.5-billion-year-old Barberton greenstone belt in South Africa. In last week鈥檚 Science (vol 304, p 578), they describe a multitude of tubes 1 to 9 micrometres across and up to 200 micrometres long, stretching away from cracks where water once flowed.
Advertisement
The tubes look very much like those dug by bacteria in modern pillow lavas. When Furnes and his group looked at samples of these, they found that bacteria and other microbes seep into water-filled cracks, and then dissolve holes in glassy parts of the rock. They extract chemical energy from iron in the rock by oxidising it or making it react with sulphur.
Calcite in the ancient tubes contains anomalously low levels of the isotope carbon-13. Living cells also have low levels of this isotope, so this backs up the idea that a microbe was involved. 鈥淲ho did it we cannot tell, but someone did it,鈥� says Furnes.
Furnes is convinced that the tubes are almost as old as the rock because some of them are broken into segments. This probably happened when the rocks were altered by heat and pressure shortly after solidifying.
Martin Fisk of Oregon State University in Corvallis says the South African find is the best evidence yet for very early life on Earth. But Brasier doubts Furnes鈥檚 conclusions. He says he has 鈥減hotos which look absolutely identical鈥� of tubes in Australian rocks from the same period, which he says are not directly produced by organisms. 鈥淭hey are a product of chemical reactions in volcanic glass,鈥� he says.
Brasier thinks the tubes were formed when mineral grains were pushed through the molten glass. He isn鈥檛 sure what pushed them, but adds, 鈥渨e should assume it鈥檚 not biological until we have rejected all possible mechanisms that are not biological鈥�. He has yet to examine Furnes鈥檚 South African samples.
If Furnes鈥檚 claims hold up, researchers could go on to look for similar patterns in rocks from Mars. Fisk has already seen a few short tube-like structures in Martian meteorite samples, but he recognises that calling them evidence for life would be premature.