The Earth’s rock record begins around 3.8 billion years ago with a period that is known as the Archaean. The Archean can be split into different eras, where the early Archaean extends from 3.8 to 3.6 billion years ago, the Paleoarchaean extends from 3.6 to 3.2 billion years ago, the Mesoarchaean extends from 3.2 to 2.8 billion years ago, and the Neoarchaean extends from 2.8 to 2.5 billion years ago. At this point, a new period known as the Proterzoic begins and it will extend all the way until about 550 million years ago, the time of the Precambrian.
I have long assumed that life appeared on this planet approximately 3.5 billion years ago. But lately, the evidence for such ancient life seems to be evaporating. The crown jewel among the ancient microbial fossils has been the filamentous cyanobacteria from 3.5-billion-old Australian chert that were first described by William Schopf from UCLA. Yet in 2002, Martin Brasier and colleagues made a strong case that those fossils are not remnants of living things, but represent the activity of ancient and exotic geochemical processes . And a recent study has just confirmed these are not fossils (HT to Joe):
Twenty years ago the palaeontological community gasped as geoscientists revealed evidence for the oldest bacterial fossils on the planet. Now, a report in Nature Geoscience shows that the filament structures that were so important in the fossil descriptions are not remnants of ancient life, but instead composed of inorganic material.
This finding is an important reminder that the evidence for such ancient life is not as strong as I had assumed:
Their finding is not stirring feelings of jubilation. “After nearly 30 years of effort at pushing evidence for life to or beyond 3.5 billion years ago, we are reminded that the ancient record is more fraught with complications than we ever thought,” says geologist Stephen Mojzsis at the University of Colorado, Boulder.
Olcott Marshall suggests that this carbonaceous material may have been accidentally sampled by other research teams and played a part in them identifying the filaments as biological. Brasier disagrees, saying instead that people wanted to find life so badly that they ignored the obvious. “There is a willful blindness about these structures that sometimes has more to do with local politics than global truth,” he says.
Whatever their origin, the biomarkers must have entered the rock after peak metamorphism 2.2 Gyr ago and thus do not provide evidence for the existence of eukaryotes and cyanobacteria in the Archaean eon. The oldest fossil evidence for eukaryotes and cyanobacteria therefore reverts to 1.78–1.68 Gyr ago and 2.15 Gyr ago, respectively. Our results eliminate the evidence for oxygenic photosynthesis 2.7 Gyr ago and exclude previous biomarker evidence for a long delay (300 million years) between the appearance of oxygen-producing cyanobacteria and the rise in atmospheric oxygen 2.45–2.32 Gyr ago.
Perhaps Stephen Moorbath, from the Department of Earth Sciences at Oxford University, says it best:
For the time being, the many claims for life in the first 2.0-2.5 billion years of Earth’s history are once again being vigorously debated: true consensus for life’s existence seems to be reached only with the bacterial fossils of the 1.9-billion-year-old Gunflint Formation of Ontario. 
Or, from another review of the fossil evidence:
In general, the most ancient microfossils are dated as follows: (a) the indisputable remains of cyanobacteria found in the Belcher Supergroup are about 2.0 billion years old; (b) the possible microfossils of early eukaryotes found in the Negaunee Iron Formation are slightly older than 1.9 billion years; (c) the very probable remains of cyanobacteria found in the Transvaal Supergroup are 2.52–2.55 billion years old; (d) possible remains of cyanobacteria and heterotrophic bacteria are 2.69–2.76 billion years old (the Fortescue Series) or even 3.4–3.5 billion years old (the Onverwacht, Fig Tree, and Warrawoon Series). 
Anyone who has read my book might recognize the distinctions between the possible, the plausible, and the probable. That is, the safe date for life’s first appearance is about 2 billion years ago, things become probable around 2.5 billion years ago, but once we get beyond that, we enter the realm of the possible.
So while I am not expert on this subject, nor can I claim to following this area of research closely, I must say that I have begun to take a more tentative stance on life’s first appearance. Is there any solid evidence that life is really older than 2.5 billion years? And if so, how do we know that evidence will not evaporate in the next decade or so?
1. Brasier MD, Green OR, Jephcoat AP, Kleppe AK, Van Kranendonk MJ, Lindsay JF, Steele A, Grassineau NV. 2002. Questioning the evidence for Earth’s oldest fossils. Nature 416:76-81.
2. Moorbath, S. 2005. Dating earliest life. Nature 434: 155-156.
3. Sergeev VN, Gerasimenko LM, Zavarzin GA. 2002. Proterozoic history and present state of cyanobacteria. Microbiology 71: 623-637.