Posted: Jun 20, 2010 2:32 pm
Darwinsbulldog wrote:@ susu.exp
Sorry, we seem to be discussing several things at once here. As I read it, Gould and Eldridge cam up with PE theory based on the fact that there are gaps in the fossil record. The traditional view was that the non-randomness of fossil preservation made it seem like there were gaps. Or that local extinctions were re-populated from refugia migrating into a region that had experienced some extinction event. Thus the transition seems "jerky"
Well, a short note on gaps in the fossil record. A gap is the time between the first appearance of a clade and the first appearance of its sister clade. These are explained through the spotty record and the fact that character evolution takes some time after the split - so one clade can usually be identified in older fossils than the sister clade.
What G&E noted were two things, one is the general absence of species to species transitional forms (there are exceptions to this, but they are rare). The other - and more crucial one really - is that there is stasis. We find some species that do not change a lot through large chunks of geological time. A lot of index fossils fit that bill, biozones defined through the occurance of a specific species lasting for more than 1Ma. So you´ve got a lot of species that remain morphologically stable for extended periods, but then change rapidly (you find them by the 1000s, that´s why they are index fossils - if they weren´t common, they´d be of little use - so not finding the transition means it must take a short time compared to the time they remain almost unchanged). Eldredge had worked on Trilobites where this is rather striking. The main support for PE is stasis. A splotchy record can explain a lack of transitional forms, but if there is an excellent record of stasis there´s no good explanation for why it should turn bad just when the interesting bit happens. Then you find things like the Steinheimer Becken, where the interesting thing does happen (neat transitional forms for the snails of Steinheim), or the snails of Kos, or... And in each of these you have an extraordinary temporal resolution, laminated lake sediments where you can pretty much count through annual layers. It all happens in a few 1000 years. So the combination of these things - big changes that show up as gradual on a scale measured in 100s of years and next to no changes on a scale measured in millions - is what propted PE. Both were well known, but the novelty of the PE paper was that it connected the two for the first time into a view on the distribution of rates of morphological change.
Darwinsbulldog wrote:Surely the rates of morphological change are based on how when and where clades are able to adapt to environmental by playing with their [non-conserved] developmental gene complexes [or their upstream or downstream expression]. I think that this is highly germane to the discussion. An organism may look wildly different, but the underlying Bauplan is highly conservative. In mammals for example, the cervical vertebra number only seven, but humans and giraffes look very different. The genetic/molecular evidence thus has confirmed many homologies that anatomists have always known about, but by no means all. Thus you can have huge changes in morphology that is based on small [but important] changes in the Hox complexes. Thus fruit flies and humans look very different, yet at the genetic level, they share homologies.
Indeed. But there we may enter into another can of worms, that for some reasons tends to enter discussions on PE (mainly because Gould wrote some things about this as well - constraints of baupläne and ontogenetic effects one of his key areas). It´s an interesting area to look into - where cutting edge science meets with von Baers law and Fishers geometrical argument (changes in HOX genes affect early development, leading to large changes in adult form, the larger a change is the less likely it is to be adaptive [Fisher], thus HOX genes tend to be highly conserved and so is early development [von Baer]). But such changes are compatible with both PE and PG. In the graphs they´d end up on the far right (where the PE and PG curves don´t differ much). The second peak in the PE graph does not correspond to fundamental changes in the baupläne, it does correspond to fast changes of features within them. If you look at the giraffe neck vertebra, Giraffa first appears about 11.6Ma ago and the containd species have short necks. The first long necks appear about 1.8Ma ago and Giraffa camelopardalis does not further increase the neck lenght. Now, the time span during which this happened is probably some 100ka long, but compared to 10Ma with no significant changes that´s rather short (1% of the time span). So that´s where PE comes in.
So at the genetic level, it looks more like gradualism rather than PE. Further, these genetic changes seem to pre-date the appearance of morphological change as seen in the fossil record.
I note a typo in the literature list. The last one should be:
Gould et al., 1977 "The Shape of Evolution: A Comparison of Real and Random Clades", Paleobiology, 3, 23-40.