The relative ages of rocks are often known from field relationships. If a dolerite intrudes a succession of rocks, the dolerite must be younger than the succession it intrudes. This relationship can be written as the inequality: Age (Dolerite) ≤ Age (Intruded succession). If the Radioisotope Age (Dolerite) > Radioisotope Age (intruded succession) then there is a violation of the relative age inequality. Many violations of a similar kind are described in this paper. These violations cannot be explained by constant half-lives, since constant half-lives would always produce radioisotope ages that are in sequence from old to young. This indicates that there is either a problem with the data or with the understanding of radioactivity. Since discrepancies occur in most of the data, it is thought that it is the latter. Therefore, a hypothesis is proposed to explain the discrepancies between the relative ages from geological field relationships and radioisotope ages: There exists fractionation in radioactive atoms of the same atom mass based on proton neutron configurations in their nuclei, causing faster and slower decaying atoms of the same atom mass. This hypothesis could explain why there is a general trend from old to young in radioisotope ages from the bottom of the stratigraphic profile to the top of the stratigraphic profile, and why discrepancies can occur.