php at rhi.hi.is (Petur Henry Petersen) writes:
>In <3rldgd$f10 at studium.student.umu.se> Agneta Guillemot <Agneta.Guillemot at historia.umu.se> writes:
>>But it is'nt koscher. Why? There is one vital factor that molecularr
>>systematicists never seem to take into account in their work:
>>Different species evolve at different rates at different times.
>>Evolution is'nt just a clockwork that goes on and on. What if,
> I would like to think it does _some_ of the time. Specially
>when we are talking of neutral changes in molecules. Of course evolution
>of quantitative characters is not a clockwork (most of the time).
>I think that this is the reason different molecules are chosen for different
>>Let's face it: Evolution in a species slows down and speeds up at
>>different times. Some species are left in the backwater, others
>>evolve fast in new evolutionary niches. There is no way of knowing
>>what happened when. The molecular clock does'nt exist!
While many biologists (particularly those with a non-molecular
orientation) have problems with the notion of the clock, I think they
fail to comprehend neutral evolution. Is the third position in most
codons selected for or against? This sort of random drift in DNA
sequences *does* appear to have clock-like behavior.
Hennig's insistence that evolutionary distances could not be measured
stemmed from the fact a) that he didn't have the power of molecular
methods available to him, and b) he was ignorant of statistics.
>>There is only one school of systematics whose theorethical basis
>>is untouchable. It is of course cladistics. If you apply cladistic
>>methodology to sequence data you come to the right, unquestionable
>>conclusions. Most molecular systematicists seem to ignore cladistics.
> I think molecular systematics are well aware of cladistics and
>dont ignore them though they might use other methods since they have
>different data. I think it is hard to use cladistics on DNA (or what?).
>What do you mean with cladistic methodology to sequence data?
Indeed, cladistics works only on shared dervived characteristics. With DNA
sequences this whole assumption breaks down, because there are only