Isochron dating is unreliable. The method assumes that the samples are
cogenetic, that is, that they form at the same time from a reasonably
homogeneous common pool. This assumption is invalid. In particular,
mixing two sources with different isotopic compositions gives meaningless
but apparently valid isochron plots.
Mixing can usually be detected by plotting the total daughter
isotopes against the ratio of daughter isotopes. These would not
likely fall on a straight line if mixing occurred.
Isochron plots from mixing can have any slope, even negative slopes.
If mixing were common, we would expect a high percentage of isochron
results to show negative slopes. They do not.
Other factors can produce false isochrons (Stassen 1998; Zheng 1989).
For example:
Protracted fractionation. This requires slow cooling (over millions
of years) and produces only a small error.
Inherited ages as from partial melting. The age given by this
method is the age of the source material. Furthermore, this factor
requires unusual conditions and usually produces scatter in the
isochron plot.
Metamorphosism. This produces apparent ages younger than the age
of the source material.
False isochrons can usually be avoided by choosing appropriate samples.
The samples must come from an (apparently) initially homogeneous source
and avoiding obvious signs of weathering and metamorphism.
Zheng, Y.-F., 1989. Influences of the nature of the initial Rb-Sr
system on isochron validity. In Chemical Geology (Isotope Geoscience
Section) 80, pp. 1-16.
Further Reading:
Dalrymple, G. Brent, 1991. The Age of the Earth. Stanford, CA:
Stanford University Press
Faure, Gunter, 1986. Principles of Isotope Geology, 2nd ed. New
York: John Wiley and Sons.