A history of birthday surprises for geologists

Age does matter when we're talking about earth, finds Cherry Lewis.

How old is the earth? In 1899 the age favoured by most geologists was 100 million years (Ma), although estimates ranged from 3 million to 2 billion. Deduced by estimating the maximum thickness of sediments in the world and their rate of deposition, or the amount of salt in the sea and its rate of increase, this vast range was inevitable.

But physicists, led by Lord Kelvin, felt that "essential principles of thermo-dynamics" had been "overlooked". Kelvin calculated the time taken for the earth to cool from when it had been a molten globe. By 1899, Kelvin's "age" was just 20 Ma.

The discovery of radioactivity three years earlier led Ernest Rutherford and Frederick Soddy to reveal that one element decayed to another. Radium, a metal, explosively emitted an electron and turned to radon, a gas, releasing helium in the process.

When Pierre Curie showed that radium also emitted heat as it decayed, it was evident that even if the earth had cooled from a molten globe, there were enough heat-producing radioactive elements within it to slow that cooling. Rutherford measured the production rate of helium and how much accumulated in a rock over time. In 1904 the very first rock was dated at 500 Ma.

In 1907 Arthur Holmes won a scholarship to study physics at Imperial College, London. Holmes recognised the need for a reliable geological timescale.

When uranium (U) was established as the parent to radium, radon and lead (Pb), the last element in the decay chain, Holmes dated his rocks using the U/Pb method. He obtained ages as great as 1,500 Ma. But when in 1913 Soddy discovered isotopes, the laborious chemical techniques available to Holmes became more complex with the need to distinguish which of the three known isotopes of Pb was being analysed.

It was the late 1930s before estimates increased. Development of the mass spectrometer, a machine that separates isotopes, enabled Alfred Nier, an American physicist, to reliably date rocks. Holmes recognised that within Nier's data lay the clue to the age of the earth - 238U decays to 206Pb, and 235U to 207Pb. Since the rate of increase of the 206Pb/207Pb ratio was known, if it could be determined for when the earth first formed, then the time taken to get from the original to the present day ratio, would be the age of the earth. In 1946 Holmes obtained an age of 3,350 Ma. Subsequently it was found that no rocks on earth contain that original ratio, so meteorite values were used. In 1956 Clair Patterson dated both meteorites and the earth at 4,550 Ma.

Today, a consensus is developing that the lead isotope clock of the earth may have been set by formation of the earth's core as late as 4,500 Ma, which is significantly younger than most meteorites, which are 4,560 Ma old. Time is the framework onto which we hang all geological events - it has become indispensable.

Cherry Lewis is the author of The Dating Game: One Man's Search for the Age of the earth. The William Smith millennium meeting "Celebrating the Age of the Earth" will be held from June 28-29 at the Geological Society, Burlington House, London.