This is it. This is the .50 caliber Desert Eagle that replaces the original 9mm Glock used to put down evolutionary biology. I got a little tired of dealing with everyone picking at various minor details that were difficult to ascertain precisely from the published papers, like the exact average rate of e. coli fixations over the 40,000 50,000 60,000 generations, and instead of trying to sort through the opaque summaries provided by the authors of the three related papers, I sent Athos to the repository and had him crunch the numbers directly. It turned out there were a number of problems with the published methodology, both in terms of how the first two studies overcounted “fixations” as well as the way none of them actually reported the precise averages due to various complications and unexpected developments across the 12 various populations. Moreover, the latter two studies contained data for 10k and 20k more generations than the original 40k in the 2009 paper.
The primary cause of the complications was the fact that half of the 12 populations being sequenced experienced a mutation that broke a genetic repair mechanism and introduced a “hypermutation” that wildly increased the number of bacterial mutations and increased the speed of the subsequent fixations to 78 generations per fixation. However, these hypermutations a) never happened in humans or the other Great Apes and b) would kill off a species in the wild by rapidly degrading the genome by flooding it with deleterious mutations. It is quite literally cancer of the genome. This is why those populations were not included in the original summaries or the final calculations, and is probably why the various scientists were a little coy with the numbers they reported in their papers.
Nevertheless, the final results were very much in line with what we’ve come to expect. And in order to deal with all the “but what about” nonsense, I’ve been careful to point out how the LTTE already incorporates every single evolutionary mechanism to which the evolutionists have been appealing save one trivial one that is impossible to incorporate due to the clonal nature of the bacteria utilized. And even the most inveterate Darwinian will freely acknowledge that ILS is not going to close the observed gap. As I mentioned in the previous post, the natural selection-only rate is now confirmed to average 4,615 generations per fixation, whereas the all-mechanisms rate averages 1,322 generations per fixation.
Abstract
The neo-Darwinian synthesis has never been subjected to a comprehensive empirical audit in which every proposed evolutionary mechanism operates simultaneously under observation. Defenders of the theory have constructed what we term the post-Darwinian polythesis: a shifting constellation of supplementary mechanisms — neutral drift, parallel fixation, supermutation, hitchhiking, epistasis, clonal interference, compensatory mutation — each invoked ad hoc to answer a specific mathematical objection, but never tested collectively. The Long-Term Evolution Experiment (LTEE) with Escherichia coli, now in its fourth decade, is the only experiment in existence where every one of these mechanisms operates simultaneously under controlled conditions. We analyze the total fixation throughput of the LTEE using two independent public datasets: Tenaillon et al. 2016 (clone-pair sequencing through 50,000 generations, 264 genomes) and Good et al. 2017 (metagenomic frequency trajectories through 60,000 generations). The resulting average for non-hypermutational populations is 1,322 generations per fixation at 60,000 generations. Applied to the human-chimpanzee divergence, the shortfall is 1,075,000-fold, and three independent lines of genomic evidence confirm that no hypermutational phase ever occurred on the human lineage. The post-Darwinian polythesis fails comprehensively by six orders of magnitude under the most favorable experimental conditions ever constructed.
1. The Post-Darwinian Polythesis
The original MITTENS framework posed a simple question: given the fastest fixation rate ever observed in any organism under any conditions, is there enough time for natural selection to produce the genetic divergence documented between humans and chimpanzees? The answer was no, by a factor exceeding one million.
The response from the evolutionary biology community was not a quantitative rebuttal. It was a qualitative one. Rather than demonstrating that the fixation rate used was wrong, which would have required producing a faster observed rate, which they cannot, defenders of the Modern Synthesis reached for a series of supplementary mechanisms. Neutral drift accounts for most substitutions, they argued. Parallel fixation allows many loci to change simultaneously. Hypermutation can accelerate the rate when needed. Hitchhiking carries neutral variants along with beneficial sweeps. Epistatic interactions create synergies. Clonal interference, frequency-dependent selection, and eco-evolutionary dynamics add further dimensions.
Each mechanism was invoked to answer a specific objection. None was invoked alongside the others. No calculations of their effects, individual or cumulative, were ever attempted or provided. No efforts were made to show how these various mechanisms were sufficient to supply the observed shortfall.
We call this collection of ad hoc supplementary mechanisms the post-Darwinian polythesis. It is “post-Darwinian” because each mechanism either supplements or replaces the core Darwinian claim that natural selection acting on random variation is sufficient to explain observed biological diversity. And it is a “polythesis” rather than a hypothesis because it is not a single testable proposition but a shifting collection of individually unfalsifiable claims, each deployed selectively depending on which mathematical objection is currently on the table.
The polythesis has a characteristic rhetorical structure. When MITTENS demonstrates that sequential fixation cannot fill the timeline, the defender invokes parallel fixation. When the Bernoulli Barrier demonstrates that parallel fixation is probabilistically self-defeating, the defender invokes neutral drift. When neutral theory’s own mathematics are shown to require an even longer timeline than the one it was invoked to rescue, the defender invokes supermutation. When supermutation is shown to produce diminishing returns, the defender returns to parallel fixation. The circle is never closed because the defender never has to run all the mechanisms simultaneously and measure their combined output.
This paper closes the circle.
The Long-Term Evolution Experiment with Escherichia coli was founded by Richard Lenski in 1988 and is now approaching 85,000 generations. It is the only experiment in existence where every mechanism in the polythesis operates simultaneously under direct observation. Beneficial sweeps, neutral hitchhiking, parallel fixation, clonal interference, compensatory mutation, epistatic interactions, hypermutation, frequency-dependent selection, and eco-evolutionary dynamics have all been documented in the LTEE populations. They are not theoretical possibilities invoked ad hoc. They are observed realities operating in concert.
The total fixation throughput of the LTEE is therefore not a measurement of any single mechanism. It is a measurement of all of the evolutionary mechanisms except one running at once, under conditions maximally favorable to rapid evolution: enormous effective population sizes, short generation times, strong selection pressure, no mate-finding costs, no recombination overhead, and an effectively unlimited mutation supply. Whatever number the LTEE produces, it is the empirical ceiling on what evolution can accomplish when every tool in its kit is deployed simultaneously under ideal conditions.
We measured that number. It is not enough. It is not even close to being enough.
You can read the rest or download the PDF at the repository if you are so inclined. And yes, there will be a third edition of PZ as well as a second edition of TFG eventually.