Wednesday, February 16, 2005

The Cambrian Explosion Never Happened.

Blair and Hedges just published an article where they use molecular clocks to examine the validity of the Cambrian explosion. They conclude, "molecular clocks continue to support a long period of animal evolution before the Cambrian explosion of fossils." Their results indicate that the Cambrian explosion (a period ~520 million years ago [mya] when many animal phyla first appear in the fossil record) is an artifact of the fossilization process and not an adequate description of animal evolution.

Molecular clock estimates of divergence times work by determining the rate at which DNA sequences diverge (the rate of DNA evolution). Rates of sequence evolution are calibrated to real time (years) using the fossil record. Estimates of the divergence time of two taxa from the fossil record are used to determine the age of divergence, which can then be used to figure out how fast a genomic sequence should evolve. For instance, if two taxa diverged 50 mya and they differ at 10% of their nucleotide sites, we can say that the nucleotide sequence evolves at a rate of 2% divergence every 10 million years or 0.2% divergence every 1 million years. This rate of evolution is then used to calculate the divergence times of other taxa in which fossil evidence is scarce or non-existent.

Previous molecular studies have been inconsistent -- some support the Cambrian explosion and some refute it. Blair and Hedges argue that the results that support the Cambrian explosion are flawed because they either misapplied calibration points or used an improper model of nucleotide substitution. The allegation of calibration point misconduct is a bold one coming from the Hedges lab, considering a recent review in which the authors conclude that Hedges and collaborators' "divergence-time estimates were generated through improper methodology on the basis of a single calibration point that has been unjustly denuded of error." I'll stop at that, as I don't want any grief from the folks upstairs (and, no I don't mean god -- the Hedges lab is literally "upstairs" from me), but I will point out that Hedges and Kumar did refute the allegations here.

The concept of the Cambrian explosion is often used by anti-evolutionists as support for divine intervention during the origin of animals. If most animals appeared at the same time, they argue, evolution would fall apart since Darwin's theory depends on gradual change over time. It now appears, however, that the Cambrian explosion is a mere artifact of the fossilization process. Because fossilization is a chance process that requires multiple events of varying probabilities, the fossil record can be misleading as a true history of life on Earth. It provides a general guideline, but the first appearance of an organism or taxon in the fossil record cannot be taken as the first appearance of the taxon in history. Instead, the first appearance of a taxon in the fossil record is the first discovered fossilized account of that taxon.

If Blair and Hedges's interpretation of the molecular data is correct, many of the animal taxa that were thought to have arisen in a very short time period may have evolved over hundreds of millions of years. This is extremely consistent with Darwin's view of evolution as a gradual process, and it puts an axe through the anti-evolutionists claims that the Cambrian explosion is inconsistent with evolution.


At 4:33 PM, Blogger DarkSyde said...

Good post, Linked it on UTI.

At 9:37 PM, Blogger Marty said...

What is missing here, and is also missing from Blair and Hedges paper, is the point that a mutation does not have to be expressed at the time it occurs. This is even more fully explained in neutral mutation theory. A mutation that arises as a neutral and possibly unexpressed change in one environment may be conserved in the genome and later, when the environment changes, increase in frequency in the population and be expressed in a phenotype. Molecular clocks that don't reconcile with morphology are predictable artifacts of how population level genetics actually work.

At 10:25 AM, Blogger RPM said...


I'm not quite clear on what you mean by "expressed." Do you mean "selected?"

Molecular clock dating operates under the assumption that the majority of nucleotide substitutions (aka fixations) are neutral. This can be tested by determining if the observed phylogeny has equivalent branches of equal lengths. If there are unequal rates of evolution along branches, then some evolutionary force (selection, drift, etc) may have lead to an excess or deficiency of substitutions along one of the branches. These trees can either be linearized or discarded when dating divergence times.

A substantial amount of research has be done on evolutionary rates and whether they reject or fail to reject the neutral theory (this is reviewed in any good Molecular Evolution textbook). For the most part, molecular evolution is a fairly neutral process that occurs at a constant rate. This may be in contrast to morphological evolution.

Many researchers are interested in finding loci that reject the neutral hypothesis. Most of the popgen and molecular evolution tests for evidence of selection are based on rejecting the the null hypothesis of neutral evolution.

Molecular and morphological data may be inconsistent. This can be due to flaws in the molecular data, morphological data, or both. I can get into this a bit more if you'd like, but this comment is already long enough.

At 3:45 PM, Blogger DarkSyde said...

I think isee what he's saying. If a psuedogene which was unexpressed picked up a mutation, and then later that pseudogene got promoted to full gene status so to speak, and the biome in which it was expressed selected for it, that might mess up the molecular clock if it happened to several lineage's throgh various stages of speciation? (BGTW marty is a confirmed fossil/plasticity hound)Something like that I think. I'm not the evo pro you guys are.


Post a Comment

<< Home