Signs of life? The isotopic signature of these oolite--carbonate rock from an ancient glacial ocean--suggests the world did not ice over.

A Not-so-Comatose Snowball

If ice really did cover Earth from pole to pole during ice ages more than half a billion years ago, as some Earth scientists have proposed, primitive life would have slowed to a near standstill. But new analyses of the ocean sediments laid down during two of those "Snowball Earth" periods seem to show that ocean life remained productive. The findings are renewing the sometimes rancorous debate over how extreme our planet's climate has been.

In the December issue of Geology, geologist Martin Kennedy of the University of California, Riverside, and his colleagues describe the pulse of the planet 600 million and 750 million years ago, as measured by isotopic ratios. Plants and green algae prefer the lighter isotope of carbon, C-12, when they build new organic matter from carbon dioxide, leaving more of the heavier C-13 in the environment. Volcanoes push the isotopic ratio the other way by spewing gas rich in C-12. When ice nearly shuts down photosynthesis, volcanoes should win out in the isotopic shoving match, Kennedy reasoned, pushing the ocean to match the isotopic composition of their emissions.

The carbon isotopic composition of carbonate rocks precipitated from glacial ocean waters, however, revealed no sign that volcanoes had prevailed over life in the isotope game. Carbonates from Australia, Namibia, and eastern California had a composition on the heavy side, as modern oceans do, rather than the distinctly light composition that volcanoes were supposed to impose on them. That means the oceans weren't almost dead for 10 millions years, as the Snowball Earth theory implies, says Kennedy. Snowball proponents "can't wiggle much any more," he says.

But geochemist Daniel Schrag of Harvard University, who helped formulate a new version of the theory in 1998, could not disagree more; he thinks Kennedy's data support the snowball hypothesis rather than undermine it. "Their chemical arguments are oversimplified and mostly just plain wrong," says Schrag. He figures an ice-covered ocean should have an isotopic composition much like what Kennedy finds, because carbonate sediments with heavier isotopes must dissolve into the ocean before the carbonates examined by Kennedy can precipitate.

Bystanders are anxious to see Schrag put his criticisms down on paper--which he promises to do soon. "I don't know whether the whole [snowball] story is right or not," says geochemist Louis Derry of Cornell University in Ithaca, New York, but "I think Kennedy has a pretty interesting data set."

Related sites

The Snowball Earth theory, by Harvard scientists Paul Hoffman and Daniel Schrag
Martin Kennedy's home page