> Actually, Gordon's concerns about assigning dates to the appearance
> of a "mitochondrial eve" seem quite reasonable to me. There is no
> "molecular clock" in the sense that we can easily predict the rates
> at which sequence divergence will increase over long periods in a
> lineage.
I have no objection to Gordon's "concerns" regarding a mitochondrial clock.
What I object to is his claim that the reason why the clock is wrong is
because it is based on the assumption that the mutation rates of genes are
constant. Hence if the clock is wrong, so is the assumption. In point of
fact, however, Gordon is simply confusing the rate of mutation with the rate
of molecular evolution. For one thing, we know that the constancy of the
rate of gene mutation is not an assumption but an established fact. For
another, we also know that the mitochondrial clock is wrong not because of
any problem with the constancy of the rate of mutation, but because other
molecular mechanisms such as recombination affect the rate of molecular
evolution to a greater degree than the rate of mutation does. So my point is
that Gordon's main argument is erroneous because it is based on
misconceptions and a lack of critical knowledge.
Kevin L. O'Brien