138 VERTEBRATE PALAEONTOLOGY 



isotopes of lead occur and these are formed during the breakdown 

 of uranium and thorium. Thus 



TJ238 ^ Pb 206 



TJ235 ^ Pb 207 



1^232 y p]^208 



The rate of decay of U 238 — >- Pb 206 is constant and can be deter- 

 mined by experimental observation. It is usually expressed as a 

 half-life period, i.e. the time taken for x g of uranium to decay 

 into x/2 g of uranium. Since the half-life period of the three 

 reactions above are known it is possible to get three checks on the 

 age of any piece of rock that contains U 235 , U 238 and Th 232 . In 

 less than one million years all three reactions come into equilibrium 

 and the ratio of the values U 238 /Pb 206 ; U 235 /Pb 207 and Th 232 /Pb 208 

 should be constant. The ratio of Pb 207 /Pb 206 should also be con- 

 stant since the ratio of U 235 /U 238 is constant. This means that on 

 old rocks all three methods should give results of approximately 

 the same value. 



This radiogenic dating has been of the greatest value in deter- 

 mining the age of the earth. These studies indicate that the 

 earth is much older than most people had thought and that it is 

 of the order of 4,500 million years old. But when the radiogenic 

 methods are applied to more recent rocks, especially those bearing 

 fossils, two serious handicaps arise. The first is that this method 

 can of course only be applied to rocks that contain radiogenic 

 lead; that is, lead derived from uranium or thorium. These rocks 

 are usually pegmatites, i.e. rocks formed from the residues after 

 granite has crystallised out from the liquid mass. This implies 

 that the material at some stage or another has been molten and 

 that therefore it is unlikely to contain fossils. Secondly there are 

 considerable differences in the age as determined from the differ- 

 ent ratio of the isotopes 206/207, 206/238, 207/235 or 208/232. 

 Thus Kulp (1955) has published a table giving data for forty-five 

 different samples of material, the lead ratios being determined by 

 mass spectrometer; and of these only seven are believed to be 

 accurate to within 5%. Some are very inaccurate, due, it is 

 believed, to the loss of radon by diffusion from the rocks in the 

 series U 238 /Pb 206 . Another difficulty is due to the amount of 



