Molecular clocks can be evaluated by comparing absolute rates of evolu
tion and by performing relative-rate tests. Typically, calculations of
absolute rates are based on earliest observed occurrences in the foss
il record. Relative-rate tests, on the other hand, merely require an u
nambiguous outgroup. A major disadvantage of relative-rate tests is th
eir insensitivity to concomitant and equal rate changes in all lineage
s. Apparent differences in absolute rates, in turn, may be artifacts t
hat are attributable to an incomplete fossil record. Recently develope
d methods in quantitative biostratigraphy recognize the incompleteness
of the fossil record and allow us to place confidence intervals on th
e endpoints of taxon ranges. These methods are applicable to taxa whos
e fossil records are of markedly different quality. When we extend the
se methods and integrate molecular and paleontologic data, we can test
the null hypothesis that seemingly disparate rates of molecular evolu
tion are in fact equal under the simplifying assumption that fossils a
re randomly and independently distributed over their temporal ranges a
nd that fossils can be accurately placed in a phylogenetic context. We
can also estimate the range of ticking rates, if any, that are compat
ible with known fossil data. Ultimately, more accurate rate estimates
for widely divergent taxa should allow for more meaningful comparisons
of evolutionary rates. DNA hybridization data for monotremes and mars
upials suggest a 17-fold difference for 14 different rate calculations
with a mean value of approximately 1% divergence per million years. V
ariation among marsupials is sevenfold. However, when we apply appropr
iate statistical tests and make additional allowances for fossils of u
ncertain taxonomic assignment, etc., all 14 rates are compatible with
a molecular clock ticking at approximately 0.4% divergence per million
years. In addition, this analysis brings relative- and absolute-rate
tests into accord.