to which Art Chadwick responded:
>Can you falsify a molecular clock?
Art, this response seems a little overboard from you, even if you are just
trying to respond in kind. Certainly the molecular clock can be falsified
and has been in many situations. There are a multitude of rate tests that
can be applied to the molecular data to test for rate heterogeneity (i.e.
non-clock like substitution rates). Genes from holoparasitic plants
frequently violate molecular clock assumptions. See Nickrent and Starr
(1994) for an example relative rate tests to test the molecular clock in
rDNA genes of holoparasitic plants.
Nickrent and Starr, 1994. High rates of nucleotide substitution in nuclear
small-subunit (18S) rDNA from holoparasitic flowering plants. Journal of
Molecular Evolution 39: 62-70
Does this mean that there are never situations in which a molecular clock
is a viable assumption? I don't think so, but it isn't an assumption that
should go untested or questioned in any data set. It has been my
observation that many molecular systematists are very aware of difficulties
in the assumption of a molecular clock for their genes and are careful in
their use of it to estimate timing of events (divergence or migration times
for example).
Well, you have worked it out by
>discrediting Popper. I guess that is one way to deal with those who oppose
>you: define them out of existence! Check this out: If differences in
>organisms are related only to differences in their physiological
>requirements rather than some fanciful relationship to evolutionary
>history, what differences would you predict to see? It is interesting to
>see that universally, the more comparisons that are used in a molecular
>cladistic analysis, the less certain any particular line of relationship
>becomes.
What studies are you specifically referring to here? Who has given you the
impression that there are significant disagreements among multiple
cladistic analyses? At what level (species, genera, family etc...) are
you talking about when you say "the less certain any particular line of
relationship becomes?" It is unfair to overgeneralize in this area because
there are so many issues that can come into play when addressing molecular
data sets many of which much be addressed on an individual bases.
Regards,
Joel