The Oklo Pyramid
New explanations in science are needed when an observation isn't explicable by current theory. The power of the scientific method lies in the extraordinary richness of understanding that can emerge from an attempt to devise a new explanation. It is like an inverted pyramid, with the first observation—often just a slight departure from the norm—as the point and then ever widening layers of inference, each dependent on a lower layer, until the whole pyramid supplies a satisfying and conclusive explanatory whole.
One of my favourite such explanations began with the observation of a small anomaly in a routine sample of uranium ore sent from Oklo, a region near the town of Franceville, in the Haut-Ogooué province of the Central African state of Gabon. Several natural nuclear fission reactorswere discovered in the uranium mines in the region in 1972.Africa and analysed in a French laboratory. Rock samples of naturally occurring uranium usually contain two types of uranium atoms, isotopes U238 and U235. Most of the atoms are U238 but about 0.7% are U235. In fact, to be accurate, the figure is .720%, but the sample that arrived in France had 'only' .717%, meaning that .003% of the expected U235 atoms were missing.
The only place such differences in proportion were known to occur was in the very artificial surroundings of a nuclear reactor, where U235 was bombarded with neutrons in a chain reaction that transformed the atoms and led to the change in the naturally occurring proportions. But the uranium ore had come from mines in the African state of Gabon and at the time there was no nuclear reactor on the whole continent of Africa, so that couldn't be the explanation. Or could it?
Unlike Olber's Paradox, where science had to wait nearly a hundred years for an explanation of an interesting observation, in the case of Oklo the explanation had already been published. Nearly twenty years before, a scientific paper by three scientists had suggested that somewhere on the Earth the conditions might have existed in the past for a uranium deposit to act like a natural nuclear fission reactor. They suggested three necessary conditions: 1. The size of the deposit should be greater than the average length that fission-inducing neutrons travel, which is about 70 centimeters; 2. Uranium235 atoms must be present in a greater abundance than they exist in natural rocks today, as much as 3% instead of .720%; 3. There must be what is called in a nuclear reactor today a moderator, a substance that 'blankets' the emitted neutrons and slows them down so that they are more apt to induce other uranium atoms to break apart.
These three conditions were exactly those that had applied to the Oklo deposits two billion years ago. The Oklo deposits were much larger than the minimum predicted size. Then, uranium235 has a half life of 704 million years and decays about six times faster that the U238 atoms, so several half lives ago, round about 2 billion years, there would have been much more U235 in natural deposits, just the sort of amount that would lead to a sustainable chain reaction. and so, extrapolating backwards, the relative proportions of the two isotopes would have been approximately 97 to 3 rather than 99.3 to .7 as it is today. And finally, the layers of rock had originally been in contact with natural water suggesting that what had happened was the following:
A chain reaction would start in rocks surrounded by water, and the atoms would split and generate heat. The heat would turn the water to steam and destroy its ability to moderate the reactions and the neutrons would escape, stopping the chain reactions. The steam would condense and turn back into water, beginning to blanket the neutrons that were still being emitted by the uranium, more neutrons would be retained, splitting the uranium atoms and restarting the chain reaction.
In explaining a tiny anomaly in the ratio of two types of atom in a piece of rock this size—I—the scientific method has led to a description of a series of events that happened in a specific location on earth billions of years ago. Over a period of 150 million years a natural nuclear reactor would produce heat for about half an hour and then shut down for two and a half hours before starting up again. It had done this over a period of 150 million years at an average power of 100 kilowatts, the kind of power produced in a typical car engine. Not only is this explanation deep, elegant and beautiful. It's also incontrovertible. It doesn't depend on someone's opinion or bias or desires, unlike many other 'explanations' of how the world works, and that's the power of the best science.