In article <D7DDBp.Jq1 at zoo.toronto.edu>, mes at zoo.toronto.edu (Mark Siddall) writes:
>> I often wish Zuckerlandl and Pauling never wrote that paper.
> There is no clock. Independent lineages (post cladogenesis) are just that
> "It's later than you think."
>> Mark :)
> Mark E. Siddall "I don't mind a parasite...
>mes at vims.edu I object to a cut-rate one"
> Virginia Inst. Marine Sci. - Rick
> Gloucester Point, VA, 23062
the molecular clock was "discovered" by Zuckerkandl and Pauling, not
"invented". The important thing about the clock is not whether or not
it "should" exist; rather the important thing is whether or not you
can "observe" it. There are many (many, many) reasons why molecular
clocks can vary in different lineages and they very often do. The surprising
thing is how often the clock model still holds quite well when you look
at the data.
Early work on the clock used fossil record dates to match sequence divergence
estimates. This can be criticised because the dates are pretty approximate
and just give the date of the earliest known fossil for a lineage; not the
date of cladogenesis. The fit is still pretty good in many cases though.
More recently, people have used the "relative rate test" of Alan Wilson.
In this case, all you need is a good outgroup. You can ask whether the
"amount" of evolution in one lineage is statistically different from the
"amount" in a second lineage, since the time of divergenec of the 2 lineages,
if you have a reference, outgroup. It is this test that Wen-Hsiung Li
and others use to try to say whether or not lineages evolve at the same rate.
Informally, you can see this if you have a rooted tree with branch lengths.
If the clock holds well, the tips of the tree will all be roughly the same
distance from the root. If the clock varies wildly, the tips will line up
Sadly, the debate about the existence of the molecular clock has been polluted
by debate about taxonomic methods over the past 30 years or so. Some groups
of taxonomists would much prefer there not to be a clock because they feel
that this might lead down the slippery slope of using statistical methods to
recover phylogenetic information from sequences. I, of course am completely
neutral on the subject :-).