DISCOVERY 



111 



and consequently it must lose a considerable amount 

 before it is even collected in the field, especially when, 

 as in the case of thorianite, it occurs in gravels which 

 have been subject to the ' action of the weather for 

 hundreds or perhaps thousands of years. 



It is clear, therefore, that the helium now found 

 in a mineral can be only a fraction, and sometimes only 

 a small fraction, of the total amount which has been 

 generated within it, and which alone could give a true 

 estimate of its life-time. Returning to the thorianite 

 investigated by Strutt, it will now be seen that it would 

 be quite wrong to suppose that its age is 280 million 

 years. The actual age must be considerably greater 

 than that, and it should be carefully noticed that helium 

 determinations never can pro\'ide data for more than a 

 minimum estimate. All that the helium-ratio can tell 

 us is that the age of the mineral to which it refers is 

 greater than a certain minimum value. In the table 

 below some of the minimum values which were obtained 

 by Strutt are listed in order of geological age, and it 

 wiU be noticed that in general they are less than half 

 the corresponding values obtained from lead-ratios. 

 To these, and the more reliable figures deduced from 

 them, we may now turn our attention. 



The original suggestion that lead is the stable end- 

 product of the uranium family was made by Boltwood 

 in 1905, and more recent discoveries have uniformly 

 converged to demonstrate the correctness of his view. 

 It is found that in fresh, primary uranium-bearing 

 minerals of the same geological age, the amount of lead 

 is closely proportional to that of uranium. Moreover, 

 when series of minerals of different geological ages 

 ire compared, it is found that the ratio of lead to 

 uranium increases with the geological age. Before the 

 lead-ratio can be accepted as a rehable age-index, the 

 questions discussed in the case of helium must, however, 

 be answered satisfactorily. 



If lead should ha%'e been originally present in 

 uranium-mineral at the time of its formation, the age 

 deduced will clearly be too high. Although lead is 

 a negligible constituent in all the ordinary primary 

 minerals of igneous rocks, it is possible that it may be 

 present in certain radio-active minerals which are 

 themselves associated with galena, the common ore 

 •of lead. Fortunately, however, it is possible to detect 

 the presence of primary lead by means of an atomic 

 weight estimation. If lead is wholly primary, its atomic 

 weight is 2071 ; if generated from uranium, the 

 atomic weight is 206. Values between these figures 

 indicate a mixture, and pro\'ide a means of deter- 

 mining the proportion of the total lead which is of 

 radio-active origin. 



The rate of production of lead in uranium minerals 

 .is not likely to have varied during their long history 

 ■except for the gradual decrease due to the decay of the 



parent-element. As pointed out in connection with 

 helium, no known agency has proved itself capable of 

 affecting the rates of radio-active changes to the 

 slightest detectable degree. From the rate of produc- 

 tion of hehum, a rate that is now accurately known 

 from several independent methods of investigation, it 

 is easy to calculate that i gram of uranium produces 

 lead at the rate of i gram in 7,500 million years. 

 Knowing the percentage U of uranium in a mineral, 

 and that of lead, Pb, the age of a mineral to a first 



Pb 



approximation is evidently ,,- ' 7,500 million years. 



However, as the existing percentage of uranium is 

 necessarily less than that originally present, a more 



accurate formula is ^y— ; ^r, ' 7,500 million years, 



U +o-575Pb "-■ ^ ' 



where (U -f- o"575Pb) is the average value of the 



uranium present during the Hfe-history of the mineral. 



The figure taken from Pb may also have to be corrected 



by reference to an atomic weight estimation, but in 



practice none of these modifications make any serious 



difference to the order of the period of time finally 



arrived at. 



So far the influence of thorium as a generator of lead 

 has not been discussed. Although the evidence seems 

 conclusive that lead is also the end-product of the 

 thorium family, yet in thorium minerals of apparently 

 identical geological age the ratio Pb/Th is often 

 extremely variable. This inconsistency appears to be 

 related to the fact that thorium minerals are usually 

 very much altered, and in them neither the Pb/Th nor 

 Pb/U ratios give reliable results. It is of course obvious 

 that, if a mineral is altered, it has suffered chemical 

 changes whereby the normal lead ratio is upset , for either 

 introduction or eUmination of lead may have taken 

 place. The well-known radio-active minerals from 

 Llano Co., Texas, are a warning example of the effect 

 of iilteration. Some of them are fresh and give a ratio 

 of o'i7 for Pb/U, but others are secondary alteration 

 products, and give ratios ranging up to I"I5. 



In order, therefore, that age determinations should 

 be reliable, it is essential that the minerals tested should 

 be chemically stable, microscopically fresh, and of 

 definitely known geological age. Further, since 

 thorium minerals are frequently altered and give 

 widely varying results, the minerals chosen should be 

 those in which the parent radio-active element is pre- 

 dominantly uranium. If these conditions are fulfilled, 

 the reliability of the experimental results can then be 

 judged by two further criteria. The lead-ratios of a 

 series of minerals of the same age should be concordant 

 among themselves, and the atomic weight of lead 

 separated from the minerals should be of the correct 

 order. 



Satisfactory minerals are available for dating three 



