"Molecular Clock" Can't Save Darwin from the Cambrian Explosion
Evolution News & Views October 28, 2015 3:59 AM
Current Biology has published yet another attempt to explain away the Cambrian explosion. On reading certain parts, you might think the authors, including Maximilian Telford, Philip Donoghue, and Ziheng Yang, have solved the problem. Indeed, their first Highlight in the paper summary claims, "Molecular clock analysis indicates an ancient origin of animals in the Cryogenian." (Cryogenian refers to the Precambrian "cold birth" era about 720 to 635 million years ago.) By itself that statement would be very misleading, because the title of the open-access paper is pessimistic: "Uncertainty in the Timing of Origin of Animals and the Limits of Precision in Molecular Timescales."
Yang appeared briefly in Stephen Meyer's book Darwin's Doubt with bad news. Meyer cited a paper Yang co-authored with Aris-Brosou in 2011 showing that molecular clock analyses are unreliable. They "found that depending on which genes and which estimation methods were employed, the last common ancestor of protostomes or deuterostomes (two broadly different types of Cambrian animals) might have lived anywhere between 452 million years and 2 billion years ago" (Meyer, p. 106).
Nothing has changed since then. The bottom line after a lot of wrangling with numbers, strategies, and analyses is that all current methods of dating the ancestors of the Cambrian animals from molecular clocks are imprecise and uncertain. They cannot be trusted to diffuse the explosion by rooting the animal ancestors earlier in the Precambrian.
Although a Cryogenian origin of crown Metazoa agrees with current geological interpretations, the divergence dates of the bilaterians remain controversial. Thus, attempts to build evolutionary narratives of early animal evolution based on molecular clock timescales appear to be premature. [Emphasis added.]
Check out the euphemisms. Translated into plain English, it means, "We can't tell our favorite evolutionary story because the clock is broken, but we're working on it."
In this new paper, they provide the latest and greatest analysis of molecular clock data so far. It's clear they believe that all the data place the root of the divergence in the Ediacaran or earlier, 100 million years or more before the Cambrian, but can they really defend their belief? They have to admit severe empirical limits:
Here we use an unprecedented amount of molecular data, combined with four fossil calibration strategies (reflecting disparate and controversial interpretations of the metazoan fossil record) to obtain Bayesian estimates of metazoan divergence times. Our results indicate that the uncertain nature of ancient fossils and violations of the molecular clock impose a limit on the precision that can be achieved in estimates of ancient molecular timescales.
Perhaps, a defender might interrupt, the precision, admittedly limited, is good enough. But then, there are those pesky fossils! The molecular clocks are fuzzily in agreement about ancestors in the Precambrian, but none of them have support from the very best observational evidence: the record of the rocks. Even the phyla claimed to exist before the explosion are contested:
Unequivocal fossil evidence of animals is limited to the Phanerozoic [i.e., the modern eon from Cambrian to recent, where animals are plentiful]. Older records of animals are controversial: organic biomarkers indicative of demosponges are apparently derived ultimately from now symbiotic bacteria; putative animal embryo fossils are alternately interpreted as protists; and contested reports of sponges, molluscs, and innumerable cnidarians, as well as putative traces of eumetazoan or bilaterian grade animals, all from the Ediacaran. Certainly, there are no unequivocal records of crown-group bilaterians prior to the Cambrian, and robust evidence for bilaterian phyla does not occur until some 20 million years into the Cambrian.
This severely limits their ability to "calibrate" the molecular clock. Meyer granted the possible existence of three Precambrian phyla (sponges, molluscs, and cnidarians). But there are twenty other phyla that make their first appearance in the Cambrian, many of them far more complex than sponges. What good are the molecular methods if you can't see any of the ancestors in the rocks?
The authors admit that the Precambrian strata were capable of preserving the ancestors if they existed.
No matter how imprecise, our timescale for metazoan diversification still indicates a mismatch between the fossil evidence used to calibrate the molecular clock analyses and the resulting divergence time estimates. This is not altogether surprising since, by definition, minimum constraints of clade ages anticipate their antiquity. Nevertheless, it is the extent of this prehistory that is surprising, particularly since the conditions required for exceptional fossil preservation, so key to evidencing the existence of animal phyla in the early Cambrian, obtained also in the Ediacaran.
The only way they can maintain their belief that the ancestors are way back earlier is to discount the fossil evidence as "negative evidence" and to put their trust in the molecular evidence. But how can they trust it, when the answers vary all over the place, depending on the methods used? One clever method is called "rate variation." Would you trust a clock that has a variable rate? How about one fast-ticking clock for one animal, and a slow-ticking clock for another?
When rate variation across a phylogeny is extreme (that is, when the molecular clock is seriously violated), the rates calculated on one part of the phylogeny will serve as a poor proxy for estimating divergence times in other parts of the tree. In such instances, divergence time estimation is challenging and the analysis becomes sensitive to the rate model used.
They try their trees with steady rates and with varying rates ("relaxed clock models" -- amusing term). They try data partitioning. They try Bayesian analysis. None of them agree. Meyer discussed molecular clock problems in detail in Chapter 5 of Darwin's Doubt. There's nothing new here. "Here we show that the precision of molecular clock estimates of times has been grossly over-estimated," they conclude. "....An evolutionary timescale for metazoan diversification that accommodates these uncertainties has precision that is insufficient to discriminate among causal hypotheses." In the end, these evolutionists have to admit that fossils would be much, much better:
Above all, establishing unequivocal evidence for the presence of metazoan clades in the late Neoproterozoic, as well as for the absence in more ancient strata, will probably have more impact than any methodological advance in improving the accuracy and precision of divergence time estimates for deep metazoan phylogeny. Realizing the aim of a timescale of early animal evolution that is not merely accurate, but sufficiently precise to effect tests of hypotheses on the causes and consequences of early animal evolution, will require improved models of trait evolution and improved algorithms to allow analysis of genome-scale sequence data in tandem with morphological characters.
Wait a minute; isn't that what Darwin provided? -- a model of trait evolution? Wasn't it natural selection of gradual variations? Let's parse this interesting quote that mentions Darwin:
The timing of the emergence of animals has troubled evolutionary biologists at least since Darwin, who was sufficiently incredulous that he considered the abrupt appearance of animal fossils in the Cambrian as a challenge to his theory of evolution by natural selection. There has been, as a result, a long history of attempts to rationalize a rapid radiation of animals through theories of non-uniform evolutionary processes, such as homeotic mutations, removal of environmental restrictions on larger body sizes, through to the assembly of gene regulation kernels --- proposed both as an explanation for rapid rates of innovation followed by subsequent constraint against fundamental innovation of new body plans after the Cambrian. Indeed, there have been explicit attempts to accommodate rapid rates of phenotypic evolution in the early Cambrian, compatible with these hypotheses and a semi-literal (albeit phylogenetically constrained) reading of the fossil record.
And yet our results, as have others before them, suggest that there is no justification for invoking non-uniform mechanisms to explain the emergence of animals and their phylum-level body plans.
That phrase "semi-literal (albeit phylogenetically constrained) reading of the fossil record" is curious. How else are you supposed to read it? They are saying that you have to read the fossil record with Darwin-colored glasses to see it correctly.
But they're trying to have it both ways. They want a slow-and-gradual fuse leading up to the Cambrian explosion (disliking "non-uniform evolutionary processes"), which requires a non-literal reading of the fossil record with Darwin glasses on, but they can't take the molecular data literally either, because it is so method-dependent. You can almost hear them crying out for fossils. As Meyer's book shows, the fossil record is more explosive now than it was in Darwin's time.
The Information Enigma Again
Notice how they mention "the emergence of animals and their phylum-level body plans." How do you get the information to build a phylum-level body plan? Once again, these authors ignore the information issue completely. They say, "Much of the molecular genetic toolkit required for animal development originated deep in eukaryote evolutionary history," skirting past that with a lateral reference to a paper about a microbe that had no animal body plan. Talk of "emergence" just doesn't cut it. What is the source of the information to build an animal body plan composed of multiple new cell types and tissues, with 3-D organization and integrated systems like sensory organisms, locomotion and digestive tracts? Is there an evolutionist who will please answer Meyer's primary challenge?
As we've seen over and over again, many Darwinian evolutionists think they have done their job if they can just push the ancestry back in time. The fossil record doesn't allow it, but even if it did, it wouldn't solve the information problem. Calling it "emergence" is unsatisfactory. Calling it "innovation" is unsatisfactory. Calling it latent potential waiting for environmental factors like heat or oxygen is unsatisfactory. Answer the question: what is the source of the information to build twenty new animal body plans that appeared suddenly in the Cambrian without ancestors? We have an answer: intelligence. What's yours?