NATURE . VOL 384, PAGE 21-22. 7 NOVEMBER 1996
The Itsaq genesis complex in southern West Greenland
ITSAQ is Greenlandic for 'ancient thing'. The Itsaq Gneiss Complex, a geological region in southern West Greenland, is indeed ancient. And, importantly, it contains the metamorphosed remains of the Earth's oldest-known sediments, providing evidence that there was something like an ocean 3.85 billion years (Gyr) ago. Until now it has seemed that these sediments might be relics of a prebiotic world, but the article on page 55 of this issue may well remove that option. Specifically, new analyses of the traces of organic carbon present in the sediments change our estimate of its 13C/12C ratio. The new value is within the range characteristic of biological debris from sediments 2.7 to 1.5 Gyr old.
In modern oceans, both carbonates and organic molecules are derived from
CO2 through precipitation and photosynthesis. Isotopic effects associated
with these reaction pathways lead to an isotopic difference, or fractionation,
of about 25 parts per thousand, with the organic materials being depleted
in 13C relative to the carbonates. A similar isotopic signal persists, with
some variations, back through time to rocks that are 3.5 Gyr old.
The new finding seems to extend that record to the very bottom of our planet's sedimentary pile, crucially altering earlier views of these oldest sediments and leaving almost no time between the end of the 'late heavy bombardment' of bodies within the inner Solar System by giant meteorites and the first appearance of life. Two earlier papers published in this journal conclude that these asteroid impacts might have sterilized the planet as recently as 3.8 Gyr ago. It could follow that the 3.85-Gyr-old ltsaq organic material derives from biochemical processes that developed with breath taking rapidity after the last large impact, or even that it is a product of a biota that a was wiped out, then supplanted by our ancestors. Can this all be true?
Isotopic fractionations result from physical phenomena associated with chemical reactions. The task now is to determine whether the observed fractionations are due to biochemical processes, other low-temperature chemical reactions, or some post-depositional effect. The magnitude of the observed fractionations favours the biochemical alternative, but does not fit perfectly. About half of the new analyses indicate fractionations larger than any in the next-oldest organic material, found in Australian rocks about 5 Gyr old, which also contain bacterium-like bodies and laminated sedimentary structures that resemble algal mats. These features are not considered a sure sign of life, but the carbon-isotope differences are roughly equal to those in still younger rocks that have compelling evidence for biology, so many investigators have proposed that life was present 3.5 Gyr ago and that it was controlling redox chemistry of carbon at Earth' surface.
The new Itsaq data were gathered using an ion-microprobe mass spectrometer that can selectively analyse features as small as 20 microm. That selectivity has been used to examine carbon occluded in Apatite grains (see Fig. 1 on page 56). Apatite, a phosphate mineral, is known for its biological associations, but there have been no previous measurements of the isotopic composition of organic carbon closely associated with it. Interpretations of the new data may evolve as points of comparison become available. Although the age of the samples and the tantalizing results command interest and prominent publication, the novelty should also inspire caution.
All previous analyses of such ancient materials have dealt only with bulk organic carbon, which shows all too well the hazards of an early birth. Any sediment that sits around for long is likely to be buried by still further sedimentation and then dragged into the crust's tectonic traffic. The Itsaq rocks were churned downwards within a billion years and heated to 500š C at pressures of 5,000 atmospheres. From the point of view of anyone seeking to reconstruct details of their initial, low-temperature environment, they've been badly overcooked. Although some previously analysed samples contain as much as 2.5 per cent carbon, it looked more like graphite than biological debris.
During graphitization, mobile chemical products wheeze or ooze from the geological pressure cooker wherever a vent can be found. If carbonate minerals are present, their carbon is thrown into the chemical mêlée. In the process, isotopic evidence is first altered (as some carbon depleted in 13C is lost) then destroyed (as the residue exchanges carbon with metamorphic fluids). For this reason, results of earlier analyses of the Itsaq materials were set aside as they were received, because they showed the damage done by graphitization. When the authors examined bulk rather than apatite-associated carbon, their results agreed with those of previous analyses.
What is missing is an understanding of why the apatite should act as a safe-deposit vault. Until that is known, other questions will claim attention. First, why are the 13C/12C ratios so variable and, in half of the cases, so low? Biological fractionation is usually very consistent, and ratios like those in the lower half of the present distribution do not appear for more than a billion years, when they are a thought to be produced by methane-consuming bacteria. These organisms require oxygen, which fits well with other indicators of increasing oxygenation 2.7 Gyr ago, but not 3.85 Gyr ago. Second, is there a plausible non-biological cause for the fractionation, either an isotope effect associated with one of the infinite number of uninvestigated prebiotic chemical reactions, or some obscure post-depositional effect peculiar to microscopic inclusions in apatite? These are worries and long shots. It's most likely that the new results .are indeed evidence for life 3.85 billion years ago.
The use of an ion microprobe to measure carbon isotopes is new. The probe works by firing caesium-ion bullets into a polished chip to break up crystals and molecules at its surface. Negative ions (C¯ in this case) are extracted and accelerated, and the isotopes are electromagnetically separated by a mass spectrometer. It is assumed that any instrumental effects biasing the ratio observed from the thin films of carbon within the mineral matrices can be removed by comparison with a reference sample of pure graphite. Tests indicating consistency with results of conventional analyses are impressive, but more documentation of the probe's suitability for carbon-isotope analyses will be welcomed. As soon as that is available, curators of collections of martian meteorites and of ancient micro-fossils should start getting in line.
John M. Hayes is at the Woods Hole Oceanographic Institution, MS8, 360 Woods Hole Road, Woods Hole, Massachusetts 02543-1539, USA.