What happens when you clone a mouse, and then clone the clone, and then clone that clone—repeating the process over and over again for twenty years? A team of Japanese scientists led by Sayaka Wakayama has now answered that question in remarkable detail. Published in Nature Communications in March 2026, their study documents the longest serial cloning experiment ever conducted in mammals: 58 generations of cloned mice, produced from a single donor, spanning nearly two decades and more than 30,000 nuclear transfer attempts. The results are extraordinary, and they carry profound implications for how we understand the role of mutations in biology.
The cloned mice looked normal. They had superficially normal lifespans. They ate, they moved, they grew old and died of old age just like any other mouse. But hidden beneath that veneer of normalcy, their genomes were quietly falling apart. With each generation of cloning, new mutations accumulated. These were point mutations, insertions, deletions, and eventually large-scale structural rearrangements including the loss of an entire X chromosome and translocations between chromosomes.
By the 57th generation, the mice carried approximately 3,700 single-nucleotide variants, 80 insertions or deletions, and at least 16 large structural variants that had not been present in the original donor. The birth rate declined steadily after generation 27, and the 58th generation was the last. Every mouse born in that final generation died within a day.
This is a stunning piece of empirical science, and I suspect it will not be long before young-earth creationists begin citing it as evidence for what they call “genetic entropy”—the claim, popularized by retired Cornell geneticist John Sanford, that all genomes are irreversibly deteriorating and that this deterioration disproves evolutionary theory. I want to get ahead of that interpretation, because it fundamentally misunderstands what this paper shows. In fact, the Wakayama study is a beautiful empirical confirmation of a prediction that evolutionary biologists made more than sixty years ago. It demonstrates precisely why sexual reproduction exists, and in doing so, it undermines—rather than supports—the genetic entropy argument.
What Is “Genetic Entropy” and Why Do Young-Earth Creationists Love It?
Before we examine what the Wakayama paper actually demonstrates, let’s introduce the genetic entropy argument as creationists present it.
John Sanford’s 2005 book Genetic Entropy and the Mystery of the Genome has become one of the most widely cited resources in the young-earth creationist arsenal. The argument goes roughly as follows: every generation, organisms accumulate new mutations. The vast majority of these mutations are either neutral or very slightly deleterious—too slight for natural selection to effectively remove them. Over time, these nearly-neutral harmful mutations pile up in the genome like rust on a car, slowly degrading the organism’s fitness. Natural selection, Sanford argues, is simply too weak a force to stop this accumulation. The genome is therefore in a state of perpetual decline. Given enough time, this decline leads to what he calls “mutational meltdown” and eventual extinction.
The implications Sanford and his followers draw from this are sweeping. If genomes can only degrade, they argue, then the evolutionary model—which requires genomes to improve over time through the accumulation of beneficial mutations—is fundamentally bankrupt. Life could not have been evolving for millions of years because genomes would have collapsed long before now. Therefore, life must be young, consistent with a recent creation only thousands of years ago. Answers in Genesis, Creation Ministries International, and the Institute for Creation Research have all promoted versions of this argument, and it has become a staple of creationist presentations, often delivered with the confident assertion that mainstream genetics confirms what Genesis teaches about a creation groaning under the weight of the Fall.
There is a superficial appeal to this argument. It is true that most mutations are either neutral or slightly deleterious. It is true that natural selection cannot perfectly purge every harmful variant. And it is true that the concept of “genetic load”—the cumulative burden of deleterious mutations in a population—is a real topic in population genetics, one that serious researchers like Hermann Muller, Motoo Kimura, and Michael Lynch have grappled with extensively. What Sanford does, however, is take a real and nuanced problem in population genetics and strip it of every mechanism that actually addresses it. He dismisses the role of beneficial mutations as negligible. He ignores the power of recombination to shuffle genomes and expose deleterious variants to selection. He overlooks the well-documented ability of sexual reproduction to purge harmful mutations far more efficiently than asexual reproduction can. And he treats his simulation program, Mendel’s Accountant, as though it were a faithful model of real populations, when in fact it is built on assumptions that guarantee the very outcome it claims to demonstrate.
What the Wakayama Study Actually Found
Now let’s turn to the science. The Wakayama team began their serial cloning experiment in January 2005 using somatic cell nuclear transfer—the same basic technique used to create Dolly the sheep. They took cumulus cells (a type of cell surrounding the egg) from a single donor mouse and transferred the nucleus into an enucleated oocyte. The resulting clone was raised to three months of age, at which point its own cumulus cells were used to produce the next generation. This process was repeated for twenty years, producing approximately three to four generations per year.
Several findings are critical for our purposes. First, the cloned mice themselves appeared healthy throughout most of the experiment. Their lifespans were normal—roughly two years, indistinguishable from control cloned mice. Their body weights were normal. Even mice from the 50th generation of cloning looked and behaved like ordinary mice. This is important: the individual organisms were not visibly degraded, even as their genomes accumulated mutations.
Second, the mutations accumulated at a remarkably steady rate. Whole-genome sequencing revealed that each generation of cloning introduced approximately 70 new single-nucleotide variants and about 1.5 structural variants. Over the course of 57 generations, this produced a substantial mutational burden: roughly 3,700 SNVs, 80 small insertions or deletions, and at least 80–84 structural variants, including 16 large-scale structural variants of 40 kilobases or more. These large variants included loss of one X chromosome, a 147.5-megabase loss of heterozygosity on chromosome 4, and translocations between chromosomes 7 and 9. Crucially, these large structural variants are the type of mutation that is rarely seen in the germline of sexually reproducing organisms.
Third—and this is the key finding—the experiment identified a clear turning point around generation 25. Before this point, the success rate of cloning actually improved with each generation, suggesting that some form of selection was occurring among the somatic cells used for cloning. After generation 25, however, the accumulation of large-scale structural variants began in earnest, and the cloning success rate entered an irreversible decline. The proportion of mutations classified as “high impact” (loss-of-function) or “moderate impact” (missense) nearly doubled between the early and late phases of the experiment. By generation 57, the success rate had fallen to 0.6%. Generation 58 was the last—all mice born in that generation died within a day.
Fourth, and perhaps most striking: the researchers tested whether the damaged genomes could be rescued by sexual reproduction. They mated late-generation cloned mice (G50 and G55) with normal males. These cloned females could still become pregnant and produce offspring, though litter sizes were dramatically reduced—from about 10 pups in normal mice to only 2–3 pups from late-generation clones. Many embryos degenerated during development, presumably because they inherited lethal combinations of accumulated mutations. But here is the remarkable part: when the surviving offspring of the G55 clones were themselves mated (producing “grandchildren”), the litter size rebounded to 7 pups—and the grandchildren’s placentas returned to nearly normal size. The process of meiosis and fertilization had filtered out many of the most damaging mutations in a single generation.
Muller’s Ratchet: The Prediction That Came True
The Wakayama team explicitly frames their results as a confirmation of Muller’s ratchet—a concept first articulated by the Nobel Prize-winning geneticist Hermann Muller in the 1960s. The idea is elegantly simple. In an asexual population, genomes are inherited as indivisible blocks. There is no recombination, no shuffling of chromosomes, no way to combine the best parts of two different genomes into one offspring. This means that once the least-mutated genome in a population picks up a deleterious mutation, the pre-mutation state is gone forever (except by the vanishingly rare event of a back mutation). The “ratchet” can only turn in one direction: toward more mutations, never fewer. Over time, deleterious mutations inevitably accumulate, fitness declines, and eventually the population goes extinct in what population geneticists call a “mutational meltdown.”
This is exactly what the serial cloning experiment demonstrated. Cloning is the purest form of asexual reproduction imaginable—a single genome, copied over and over, with no partner, no recombination, no opportunity for selection to act across a genetically diverse population. Under these conditions, Muller’s ratchet predicts precisely what happened: steady accumulation of harmful mutations, progressive decline in viability, and eventual collapse.
But here is what makes the study so powerful as an evolutionary finding rather than a creationist one: the researchers also maintained control lines of mice that reproduced sexually through sibling mating for more than 60 generations over the same 20-year period. These sexually reproducing mice accumulated far fewer mutations—approximately 22 SNVs per generation compared to 70 in the cloned lines, and far fewer structural variants. More importantly, the sexually reproducing lines showed no decline in fertility or viability across 60 generations. The sexually reproducing mice were fine. The clonally reproducing mice went extinct. This is not a story about genomes inevitably falling apart. It is a story about what happens when you remove the very mechanism—sexual reproduction with its attendant recombination and selection—that genomes use to maintain genomic integrity.
How Creationists Will Misuse This Paper—and Why They’re Wrong
I anticipate that young-earth creationists will seize on the Wakayama study and present it something like this: “See? Even mainstream scientists confirm that mutations accumulate and destroy genomes. These mice went extinct in just 58 generations! This is exactly what Dr. John Sanford has been saying for twenty years. Genetic entropy is real, and it proves that life cannot be millions of years old.”
Let’s take a look at why this interpretation is wrong.
First, the experiment does not demonstrate that all genomes are inevitably deteriorating. It demonstrates that genomes deteriorate when they are propagated clonally—without recombination, without a genetically diverse population, and without the filtering power of meiosis. This is a crucial distinction. The entire point of the study is to identify why mammals reproduce sexually rather than asexually. The answer, confirmed by this experiment, is that sexual reproduction provides mechanisms for purging deleterious mutations that clonal reproduction lacks. To cite this paper as evidence for genetic entropy is like observing that a person who never eats will eventually starve, and then concluding that nutrition is a myth. The starvation proves the importance of eating, not its futility.
Second, the paper provides direct empirical evidence that sexual reproduction can rescue even heavily damaged genomes. When late-generation clones—carrying thousands of accumulated mutations including chromosomal translocations and loss of an entire X chromosome—were mated with normal males, the process of meiosis and fertilization filtered out many of the harmful variants. By the grandchild generation, litter sizes had substantially recovered and placental abnormalities had largely resolved. This is exactly the opposite of what the genetic entropy model predicts. If genomes can only decline and natural processes cannot restore them, this rescue should be impossible. But it happened, and it happened through ordinary biological mechanisms: meiosis, recombination, segregation, and natural selection acting on genetically variable offspring.
Third, the mutation rate in the cloned mice was approximately three times higher than in the sexually reproducing control lines. The Wakayama team found about 70 new SNVs per clonal generation versus about 22 per sexual generation. This difference reflects, in part, the fact that somatic cells (used in cloning) accumulate mutations at a higher rate than germline cells (used in sexual reproduction). In sexually reproducing organisms, the germline is specifically protected against mutation accumulation—it is maintained in a relatively quiescent state, and multiple DNA repair mechanisms operate to keep the germline mutation rate low. Cloning bypasses all of these protections. The genetic entropy argument treats mutation accumulation as an inescapable feature of all reproduction, but in reality, sexually reproducing organisms have evolved elaborate systems specifically designed—if I may use that word in its biological sense—to minimize and manage the mutational burden.
Fourth, and this gets to the heart of the matter, Sanford’s genetic entropy argument depends on the claim that natural selection cannot effectively purge slightly deleterious mutations. But the Wakayama study demonstrates something more fundamental: it is not just natural selection that matters, but the entire suite of mechanisms associated with sexual reproduction. Recombination breaks up linkage between mutations, allowing selection to act on individual variants rather than on entire chromosomes. Segregation during meiosis separates chromosomes bearing different mutations, generating offspring with varying mutational loads. Some offspring inherit fewer harmful variants and reproduce more successfully; others inherit more and are eliminated. This is not a minor detail—it is the central mechanism by which complex organisms maintain their genomes over deep time, and it is precisely the mechanism that the Wakayama study confirms by showing what happens when it is absent.
The Irony of Using Muller’s Ratchet to Defend Young-Earth Creationism
There is a deep irony in the creationist appropriation of this kind of research. Muller’s ratchet was proposed specifically to explain why sexual reproduction evolved and why it is maintained despite its significant costs (the so-called “twofold cost of sex”—a sexual population produces half as many offspring as an equivalent asexual population, because half the population consists of males who do not directly produce offspring). The very existence of sex has been one of the great puzzles of evolutionary biology. Why do organisms pay this enormous cost? The answer, as the Wakayama study beautifully illustrates, is that sexual reproduction provides an indispensable mechanism for managing mutations and maintaining genomic integrity over the long term.
This is a profoundly evolutionary insight. The reason sexual reproduction exists—the reason it was favored by natural selection despite its costs—is precisely because it solves the problem of mutation accumulation. To take the evidence that clonal reproduction fails (as predicted by evolutionary theory) and use it to argue that all reproduction must fail is to fundamentally misunderstand the science. It is like arguing that because a car with no steering wheel crashes, steering wheels must not work.
Moreover, the Wakayama study actually undermines a key prediction of the genetic entropy model in another way. Sanford argues that organisms alive today should show measurable signs of genomic decline, and he has pointed to declining human lifespans described in Genesis as evidence for this. But the serially cloned mice showed normal lifespans throughout the experiment—even at generation 50, with thousands of accumulated mutations. The individual organisms were not measurably degraded. What declined was the success rate of producing new clones—that is, the reproductive output of the lineage. This is consistent with a well-understood population-genetic phenomenon in which purifying selection eliminates the most severely affected embryos, leaving only those individuals whose mutation load is compatible with survival. The mice that were born were healthy precisely because the most damaged embryos never made it to birth. This is selection at work, filtering out the worst mutations even in the absence of recombination. It is exactly what evolutionary theory predicts.
What This Paper Tells Us About the Real World
Let me step back and consider what this experiment tells us about real populations—the ones that matter for the evolution/creation debate.
In real sexually reproducing populations, every generation involves meiosis, recombination, and the union of two genetically distinct gametes. Every offspring is genetically unique. Every generation, the genomes of the population are reshuffled, harmful mutations are exposed to selection, and the most damaging combinations are eliminated. This is not a theoretical abstraction; it is an observed, well-characterized, empirically confirmed process that we can watch in real time in laboratory populations, in natural populations, and—thanks to modern genomics—at the level of individual nucleotides.
The Wakayama study provides the most dramatic possible demonstration of what happens when you strip all of this away. Take a single genome. Copy it, over and over, with no recombination, no mating, no population. The result is extinction in 58 generations. Now compare that to the sexually reproducing control lines, which continued for more than 60 generations with no decline in fitness. The message is unmistakable: sexual reproduction is not merely a nice feature of biology. It is the mechanism that prevents the very genomic collapse that Sanford warns about—and it does so with stunning effectiveness.
This has direct implications for the creationist argument. If genetic entropy were real—if genomes were truly in a state of irreversible decline that no natural mechanism could halt—then we would expect to see signs of this decline in real populations. We would expect fertility to decrease over generations. We would expect lifespans to shorten. We would expect accumulating evidence of genomic meltdown in species that have large populations and have been reproducing sexually for millions of years. But we do not see this. Humans, mice, fruit flies, roundworms, and countless other organisms have been studied across hundreds and in some cases thousands of generations, and the pattern is clear: sexually reproducing populations maintain their fitness over time. The ratchet does not turn when recombination is present to stop it.
A Better Way to Read Both Nature and Scripture
As a conservative confessional Christian and a biologist, I find the Wakayama study fascinating and in many ways deeply satisfying—not because it scores points against anyone, but because it reveals something genuinely beautiful about how creation works. Sexual reproduction, with all its complexity and apparent inefficiency, turns out to be an elegant solution to a fundamental problem of information maintenance in a world of molecular imperfection. Scholarly evangelical Christian tradition has long affirmed that God works through secondary causes—that the ordinary operations of the natural world are the means by which providence sustains creation. The mechanisms of meiosis, recombination, and selection are precisely such secondary causes, and they are breathtaking in their effectiveness.
What troubles me about the genetic entropy argument is not just that it is bad science—though it is—but that it encourages believers to read the natural world through a lens of despair. Creation is not running down. Genomes are not collapsing. Life is not spiraling toward extinction because of some accumulated genetic curse. The living world is dynamic, adaptive, and—in the language of our tradition—sustained by the providence of a faithful Creator who has equipped his creatures with the means to endure. To insist that genomes can only deteriorate is not only to ignore the evidence; it is to tell a smaller story about creation than the one the evidence reveals.
I would encourage my fellow believers to look carefully at this paper before accepting the interpretive spin that young-earth creation apologists will inevitably put on it. Read what the researchers actually found. Note that the clonal mice went extinct precisely as evolutionary theory predicted. Note that the sexually reproducing mice did not. Note that even the most damaged genomes in the experiment could be partially rescued by a single round of sexual reproduction. And ask yourself whether the evidence is really pointing toward a world in genetic free-fall—or toward a world in which the mechanisms of reproduction are far more robust, far more elegant, and far more worthy of wonder than the genetic entropy narrative allows.
Blessings,
Joel Duff
References
Peer-reviewed literature:
Wakayama, S., Ito, D., Inoue, R. et al. “Limitations of serial cloning in mammals.” Nature Communications 17, 2495 (2026). https://doi.org/10.1038/s41467-026-69765-7
Muller, H. J. (1964). The relation of recombination to mutational advance. Mutation Research/Fundamental and Molecular Mechanisms of Mutagenesis, 1(1), 2-9.
Felsenstein, J. “The evolutionary advantage of recombination.” Genetics 78: 737–756 (1974).
Gabriel, W., Lynch, M., and Bürger, R. “Muller’s Ratchet and Mutational Meltdowns.” Evolution 47(6): 1744–1757 (1993).
Lynch, M. “Mutation and Human Exceptionalism: Our Future Genetic Load.” Genetics 202: 869–875 (2016).
Kimura, M. “The Neutral Theory of Molecular Evolution.” Cambridge University Press (1983).
Young-earth creationist sources:
Sanford, J.C. Genetic Entropy and the Mystery of the Genome. FMS Publications, 4th ed. (2014).
Sanford, J.C. and Carter, R. “A new look at an old virus: patterns of mutation accumulation in the human H1N1 influenza virus since 1918.” Theoretical Biology and Medical Modelling 9(42): 1–19 (2012).
Carter, R. “Genetic entropy and human lifespans.” Creation Ministries International (2012). creation.com.
Naturalis Historia 
Comments or Questions?