“There has been enough time for the mutations that evolution requires to arise and spread.”
§1 The claim, as taught
Life on Earth has been accumulating mutations and passing them through the filter of natural selection for 3.8 billion years — more than enough time for evolutionary processes to produce its grand history.University of California Museum of Paleontology. *Understanding Evolution*, “Evolution 101.”
§2 The strongest evidence offered
The arithmetic is not controversial: it is standard population genetics. What is debated is which numbers to put in.
Single steps are fast in microbes
In large bacterial and viral populations, any particular single-letter mutation arises many times a day. Single beneficial mutations are routinely seen to arise and spread in laboratory populations.Two mutations, in humans
Mathematicians Rick Durrett and Deena Schmidt calculated that, in a human-sized population, waiting for a pair of specific mutations, the first neutral, to create a new regulatory binding site would take more than 100 million years, against a few million in fruit flies. They concluded that it is a real constraint, and that the answer is that many different sites would do.Strings of letters
Sanford and colleagues simulated a population of 10,000 hominins and reported that even a specific two-letter change took tens of millions of years to establish, and longer strings far more. The Lab's own model reproduces the paper's figures for single mutations. For strings of two letters or more, the paper scaled its results from runs at a raised mutation rate, and the Lab finds its figures too short, not too long.Tunnels through valleys
Lynch and Abegg argued that in large populations, combinations can arise without each step being fixed first (“stochastic tunnelling”), shortening waiting times greatly. In small populations the advantage largely disappears.The cost of substitution
Haldane calculated in 1957 that replacing one gene by another throughout a population costs deaths, limiting how fast substitutions can occur, roughly one per 300 generations in his example. Walter ReMine applied this to the human line and argued it allows only around 1,700 beneficial substitutions in ten million years.§3 The grade: two claims hiding in one
The answer depends on the size of the population and the number of changes needed at once. Graded separately: Hover or tap a rung to see why it got its score. How the grading works →
Single beneficial mutations arise and spread quickly in large populations
BStrongly supportedThere has been time for the coordinated changes needed in slow-breeding animals, including humans
DExtrapolation§4 What it would take
To show there was enough time, the following would be needed:
- Count the stepsFor a specific adaptation, such as a human-specific trait, state how many coordinated mutations it required and which ones.
- Run the numbers with real parametersUse measured mutation rates, effective population sizes and generation times for the lineage concerned, and show the waiting time fits the time available.
- Show the targets are manyWhere the answer depends on many alternative sequences working equally well, measure how many actually do.
§5 Another reading of the same data
The arithmetic here is unusual in the evolution debate, because both sides use the same equations. Durrett and Schmidt are mainstream mathematicians, and their paper was written partly to answer Michael Behe. Their result still showed that a pair of specific mutations, in a human-sized population, would take more than a hundred million years to appear. Their reply was not that the number was wrong, but that evolution rarely needs one specific pair: many different changes might do the same job.
That reply may be right. It is also the point where the argument becomes empirical. How many different sequences would have produced, say, a new regulatory switch for a human brain gene? Pillar 4 looked at how rare working protein sequences are, and found the estimates range widely. Here the same question arises for every adaptation that needs more than one change at once.
Read through a design lens, the waiting-time results suggest that the complex, coordinated features of animals and humans were not assembled by waiting for random mutations, but were there from the start, with the capacity to vary built in. That reading predicts that the adaptations we see arising in real time will mostly be single steps: tuning, switching and breaking. The evidence in Pillars 4 and 5 fits that prediction.
§6 Reasoning check
The grade weighs the evidence. This check looks at the reasoning: the step from the evidence to the conclusion. The same tests are applied to both readings. How the reasoning check works →
- Clear case
Waiting times that are short for single mutations in populations of billions are assumed to be short for multiple mutations in populations of thousands.
- Possible
“Many different targets would work” is often assumed because it is needed, not because the number of working targets has been measured.
- PossibleIn the evidence: The cost of substitution
Haldane's figure was an estimate for his example, not a law. Many geneticists argue that soft selection and overlapping substitutions lift the limit. Whether they lift it far enough is the open question.
- Possible
Critical calculations sometimes assume that one specific sequence was required. If many would have served, the waiting time shortens. Specifying the target as the one that exists can assume the conclusion.
§7 Questions to ponder
- If many different mutations would have done the job, how would we know?
- Why might the same equations give reassuring answers for bacteria and troubling ones for mammals?
- Try the Lab's Probability Simulator. At what population size does the waiting time for two coordinated changes exceed the time available?
§8 Sources
- Haldane JBS. “The cost of natural selection.” Journal of Genetics 55:511–524 (1957).B
- Durrett R, Schmidt D. “Waiting for two mutations: with applications to regulatory sequence evolution and the limits of Darwinian evolution.” Genetics 180:1501–1509 (2008).B
- Iwasa Y, Michor F, Nowak MA. “Stochastic tunnels in evolutionary dynamics.” Genetics 166:1571–1579 (2004).B
- Lynch M, Abegg A. “The rate of establishment of complex adaptations.” Molecular Biology and Evolution 27:1404–1414 (2010).B
- Perfeito L, Fernandes L, Mota C, Gordo I. “Adaptive mutations in bacteria: high rate and small effects.” Science 317:813–815 (2007).B
- Behe MJ, Snoke DW. “Simulating evolution by gene duplication of protein features that require multiple amino acid residues.” Protein Science 13:2651–2664 (2004).C
- Behe MJ. “Waiting longer for two mutations.” Genetics 181:819–820 (2009), with the authors' reply.C
- Sanford J, Brewer W, Smith F, Baumgardner J. “The waiting time problem in a model hominin population.” Theoretical Biology and Medical Modelling 12:18 (2015).C
- ReMine WJ. The Biotic Message: Evolution Versus Message Theory. Saint Paul Science (1993).D
Source grades are provisional. In production each links to a full published evaluation that readers can challenge.