Over the past couple years, a few pounds of rock from Australia have been the subject of a fierce scientific battle between geologists and paleontologists. Some paleontologists have claimed that microscopic marks in the 3.5 billion year old rocks are the oldest fossils of life yet found. Some geologists have recently argued that the marks are just odd mineral formations that could have been created without the help of life. Today in Science, the geologists have struck again. A team from Spain and Australia mixed up some silica, carbonate, barium, and other compounds that can be found in the Australian rocks. With a little lab cooking (which they argue is akin to how the rocks formed) they were able to create little lumpy chains. When UCLA's William Schopf discovered similar little lumpy chains in 1993, he declared that he had found fossils of cyanobacteria (also known as blue-green algae). Not only did the chains look like living chains of cyanobacteria, but Schopf also found organic carbon around them. The geologists who published the Science paper today point out that non-living processes can create "organic" carbon too, and when they added this carbon to their recipe, they found that they could readily coat their pseudofossils as well. Richard Kerr, the estimable senior writer for Science's news section, talked to some other geochemists and paleontologists, and many of them were impressed with the new work. And it's not the first challenge that Schopf has had to deal with. Last year, researchers argued that the rocks Schopf found were formed around hydrothermal vents--not exactly the place where photosynthetic cyanobacteria would be found. Likewise, other evidence for isotopic signatures of life in 3.8 billion year old rocks from Greenland have also been challenged. It's been bracing to watch these scientific battles, and it should serve as yet another refutation of the absurd notion embraced by certain board of education members that scientists who study evolution are "dogmatic." Even the most high-profile research on the most important aspects of the history of life are fair game for rigorous scientific challenges. Never willing to let self-consistency slow them down, antievolutionists have seized on these new reports, claiming that they call into question all evidence of ancient life (perhaps even radiometric dating). It would be nice if--just once--they would actually do the hard work involved to make such a claim: publish a paper in a peer-reviewed journal showing how they went into a lab and created mineral formations that mimic all fossils. Or even a few fossils. Even one. The fact is that other evidence for ancient life still stands. There are isotopic signatures dating back 3.7 billion years, for example, that have not been challenged. Fossils as old as 2.5 billion years are generally considered the real deal. While these dates are still inconceivably old, they raise some fascinating issues about how long it took for life to arise on Earth. It was starting to look as if life might have gotten started perhaps 4 billion years ago. Earth is 4.55 billion years old, but for several hundred million years it was colliding with assorted failed planets and other pieces of interstellar rubble, which would have literally boiled off the oceans and made it unlikely for any early life that might have gotten started to survive. Before these new challenges popped up, it looked as if life got started pretty quickly as soon as things calmed down. That might have suggested that life elsewhere in the solar system (or the universe) could be pretty common. Now it's not so clear whether life starts easily or needs hundreds of millions of years more to get going. Schopf and company haven't backed away from their original research, though, and they shouldn't be counted out. The research published today only shows that geological processes can create structures that look like bacterial fossils. There's plenty of evidence that bacteria can form these structures, too, particularly from younger bacterial fossils that haven't been so degraded by the ravages of time. And in Kerr's article, Schopf points out that he found internal walls between the lumps in his chains, which look a lot like the walls of bacteria. The pseudofossils are hollow tubes. Their creators told Kerr that if they altered their recipe a bit, they could probably make internal walls, but if that's true, they should have waited to actually get those results before they submitted their paper. This is a story that's far from over.