The present natural abundance ratio of uranium-235 to uranium-238 is about 0.7171%. The Oklo mine in Gabon, West Africa holds evidence of a natural chain reaction involving the fission of uranium which must have occurred about 1800 million years ago when the natural isotopic ratio of uranium-235 to uranium-238 was much higher at about 3%. The ratios of the two radioactive isotopes have changed during this time because the halflife of uranium-238 is much longer than the halflife of uranium-235.

Although both U-235 and U-238 isotopes decay exponentially with time, the rate of disintegration of uranium-235 is higher, so this isotope decreases in relation to the U-238 isotope. What is strange about the Oklo uranium deposit, is that the 0.296% U-235/U-238 isotopic ratio of the Oklo deposit is much less than the presently observed 0.7% ratio for uranium deposits elsewhere in the world. [Another source (The IAEA - aka International Atomic Energy Agency) claims that the U-235/U-238 ratio is 0.720% in naturally occurring Uranium deposits elsewhere in the World, but 0.717% at Oklo - which represents a major disagreement in percentages]. The Oklo deposit must have undergone natural depletion of the uranium-235 by a nuclear chain reaction involving nuclear fission using ground water as the moderator to slow down the neutrons so that they can cause further fission. The ore must have been sufficiently concentrated in one place for it to have been near the critical mass. This fission can only have happened when the U-235/U-238 ratio was 3%, about 1900 million years ago, and it must have continued for thousands of years (about 150,000 yrs) until the U-235/U-238 ratio fell to 1.5%, below which the reaction is no longer sustainable, hence the need to purify (increase the concentration of) uranium-235 for use in nuclear power stations today. During the reactor phase the average output power of the reaction was only about 100kW. But the induced fission of U-235 does not take place by the fast neutrons emitted by the spontaneous fission of U-235, because they are travelling too fast. If water was also present (highly likely) then this would have moderated (slowed) the fast neutrons to such a low speed as they were able to induce fission in U-235. In fact, it is now believed that the reaction oscillated like a geyser: as the water moderated the neutrons and sped up the induced fission of U-235, the water got hotter turning to steam, which because it is less dense than water, is less efficient a moderator, and the reaction abated, only to repeat when more groundwater trickled in. The evidence to support this oscillatory regime comes from the gaseous xenon (another fission product) found in nearby rocks, which could only become trapped if the reactor had just cooled. It is thus believed that it alternated between 30 minutes of activity separated by dormant 150 minute periods, and generated an average power of 100kW for several millenia. It should be noted that the current ratios of U-235/U-238, at 0.717%, are far too low for a sustainable fission reaction to occur.

The explanation that U-235 underwent induced fission is supported by the presence of insoluble fission products within the ore: yttrium and other lanthanides, and the siderophiles from ruthenium to niobium. (Normally, uranium would undergo only alpha decay, not fission and alpha). The more soluble fission products rubidium, caesium, strontium and barium were long since washed out of the ore. Now, 1800 million years later, this short period of accelerated decay (by induced fission) has resulted in the present U-235/U-238 ratio for this ore being much less than other naturally occurring uranium deposits.

Constancy of speed of Light, c Since 1984 the speed of light in a vacuum, c, has been fixed by convention at 299.792458 ×106 metres per second. It is explicit in Einsteins Theory of Relativity that the speed of light is fixed at a finite, but, it seems, arbitrary value, and cannot change over time. However, this view is always being tested to greater and greater precision. The Oklo deposit can be used to test whether the speed of light has remained constant over the duration of the deposit, 1800 million years. The rate of neutron capture events depends upon alpha, the fine structure constant, which is a measure of the strength of the electromagnetic force. The fine structure constant is (amongst two other constants: Plancks constant and the charge on the electron) inversely proportional to the speed of light. The Fine Structure Constant, alpha It is generally thought that if alpha changed over time, it would be because the speed of light changed over time (rather than either Plancks constant or the charge on the electron). Samarium-149, present in the Oklo deposit, captures a neutron to become samarium-150, and the rate at which it does so depends upon alpha, the fine structure constant. So, by measuring the ratios of several pairs of isotopes in the Oklo deposit, like Sm-149/Sm-150, any change in alpha can be calculated and thus any change in the speed of light measured. By this method, researchers have calculated that the speed of light has changed by 4.5 parts in 108 over the last 1800 million years. But the calculation depends somewhat critically upon knowing the temperature of the Oklo deposit throughout this period of time, and this can only be guessed at to within 200 Celsius, which throws into some doubt the validity of the measurement. It also depends critically on the assumption that there is no mechanism whereby one isotope of any isotope pair is not altered over this time, be leaching, etc, thereby altering the ratio of the isotopes. Also, the rate of neutron capture is dependant not only upon alpha, the fine structure constant, but also on alpha(s), which determines the strength of the strong nuclear force, and which also affects the attraction between neutrons and nucleus. By measuring the spectral lines emitted by both caesium and hydrogen between the years 1999 and 2003, it is known that any variation in the speed of light has been less than 1 part in 1015 (on Earth) over these four years.

The fine structure constant, alpha, is currently known to great precision, being equal to 1/137.03599976.