Ancient DNA Discovery Reveals Woolly Mammoths, Wild Horses Survived Thousands of Years Longer Than Believed

Pleistocene Scenery

Researchers used DNA capture-enrichment technology developed at McMaster to isolate and rebuild the fluctuating animal and plant communities during the Pleistocene-Holocene transition. Credit: Julius Csotonyi

Mere spoonsful of soil pulled from Canada’s permafrost are opening vast windows into ancient life in the Yukon, revealing rich new information and rewriting previous beliefs about the extinction dynamics, dates, and survival of megafauna like mammoths, horses, and other long-lost life forms.

In a new paper, published in the journal Nature Communications, researchers from McMaster University, the University of Alberta, the American Museum of Natural History, and the Yukon government present a 30,000-year DNA record of past environments, drawn from cored permafrost sediments extracted from the Klondike region of central Yukon.

Researchers used DNA capture-enrichment technology developed at McMaster to isolate and rebuild, in remarkable detail, the fluctuating animal and plant communities at different time points during the Pleistocene-Holocene transition, an unstable climatic period 11,000-14,000 years ago when a number of large species such as mammoths, mastodons, and saber-toothed cats disappeared.

Yukon Soil Collection

Cored permafrost sediments are extracted from the Klondike region of central Yukon. Credit: Tyler Murchie

They reconstructed the ancient ecosystems using tiny soil samples which contain billions of microscopic genomic sequences from animal and plant species.

The analysis reveals that mammoths and horses were already in steep decline prior to the climatic instability, but they did not immediately disappear due to human overhunting as previously thought. In fact, the DNA evidence shows that both the woolly mammoth and North American horse persisted until as recently as 5,000 years ago, bringing them into the mid-Holocene, the interval beginning roughly 11,000 years ago that we live in today.

Through the early Holocene, the Yukon environment continued to experience massive change. Formerly rich grasslands—the “Mammoth Steppe”—were overrun with shrubs and mosses, species no longer held in check by large grazing herds of mammoths, horses, and bison. Today, grasslands do not prosper in northern North America, in part because there are no megafaunal “ecological engineers” to manage them.

Tyler Murchie

Tyler Murchie is a postdoctoral researcher in McMaster’s Department of Anthropology and a lead author of the study. Credit: Georgia Kirkos

“The rich data provides a unique window into the population dynamics of megafauna and nuances the discussion around their extinction through more subtle reconstructions of past ecosystems,” says evolutionary geneticist Hendrik Poinar, a lead author on the paper and director of the McMaster Ancient DNA Centre.

This work builds on previous research by McMaster scientists who had determined woolly mammoths and the North American horse were likely present in the Yukon approximately 9,700 years ago. Better techniques and further investigation have since refined the earlier analysis and pushed forward the date even closer to contemporary times.

Hendrik Poinar

Hendrik Poinar is an evolutionary geneticist, a lead author on the paper and director of the McMaster Ancient DNA Center. Credit: Georgia Kirkos

“Now that we have these technologies, we realize how much life-history information is stored in permafrost,” explains Tyler Murchie, a postdoctoral researcher in McMaster’s Department of Anthropology and a lead author of the study.

“The amount of genetic data in permafrost is quite enormous and really allows for a scale of ecosystem and evolutionary reconstruction that is unparalleled with other methods to date,” he says.

“Although mammoths are gone forever, horses are not,” says Ross MacPhee of the American Museum of Natural History, another co-author. “The horse that lived in the Yukon 5,000 years ago is directly related to the horse species we have today, Equus caballus. Biologically, this makes the horse a native North American mammal, and it should be treated as such.”

Scientists also stress the need to gather and archive more permafrost samples, which are at risk of being lost forever as the Arctic warms.

Reference: “Collapse of the mammoth-steppe in central Yukon as revealed by ancient environmental DNA” by Tyler J. Murchie, Alistair J. Monteath, Matthew E. Mahony, George S. Long, Scott Cocker, Tara Sadoway, Emil Karpinski, Grant Zazula, Ross D. E. MacPhee, Duane Froese and Hendrik N. Poinar, 8 December 2021, Nature Communications.
DOI: 10.1038/s41467-021-27439-6

26 Comments on "Ancient DNA Discovery Reveals Woolly Mammoths, Wild Horses Survived Thousands of Years Longer Than Believed"

  1. If grasslands would be created in the far north if there were large herds of big herbivores then that supports the theory being espoused by those scientists who want to bring the mammoth back from extinction. And the role they could then play in the environment there. Will humans undertake such a vast effort? Would they succeed if they did? Well all the money is being flushed down the outer space funnel. Total waste. We live here. Elephants are about as alien to us as anything we might find in space maybe in a million years.

    • False dichotomy. Doing one does not mean we can’t do the other. Also, the technology that will allow us to live and thrive in space is exactly the technology we need here. Living independently of Earth in space means very good recycling, compact high productivity food production, clean water and clean air. It is an almost zero footprint scenario.
      Used on Earth, our cities would require far less water, and far less acreage to grow food, and fiber for clothing bedding and toilet paper. Alternatives to wood would likely have to be developed to build homes and other structures.

    • Torbjörn Larsson | December 10, 2021 at 4:30 pm | Reply

      You can’t “deextinct” a species.

      There are a few groups that want to modify subpopulations of extant species to be more like some extinct species. The ethics and rationale is dubious – the ecology is gone – but not quite as bad as you claim.

      • Torbjörn Larsson | December 10, 2021 at 4:35 pm | Reply

        I should add that research is never wasted, since it is an integral whole. However, we do want to set rational priorities on our resources. Pandemic research used to be underfunded, now we risk that after the current pandemic it will be overfunded (though pandemic frequency will grow in tandem with growth in urbanization, so maybe not).

  2. Babu G. Ranganathan | December 10, 2021 at 7:16 am | Reply

    Babu G. Ranganathan*
    (B.A. Bible/Biology)


    When you divide a cake, the cake never gets bigger. However, when we were just a single cell and that cell kept dividing we got bigger. New material had to come from somewhere. That new material came from food.

    Just as the sequence of various letters and words in human language communicate a message and direct workers to build and assemble something so, too, the sequence of various molecules in our DNA (our genes or genetic code) directed molecules (i.e. various amino acids, the building blocks of proteins) from our mother’s food, that we received in the womb, to become new cells, eventually forming all the tissues and organs of our body. Food isn’t just for energy. It’s also material used by the body to make new cells. After all, when we eat meat or vegetables we’re actually eating cells of animals and plants.

    When you feed a cat your food the cat’s DNA will direct the food molecules to become the cells, tissues, and organs of a cat, but your DNA will turn the same food into human cells, tissues, and organs.

    What we call “genes” are actually segments of the DNA molecule. When you understand how your DNA works, you’ll also understand how egg yolks can turn into chickens. Read my popular Internet article: HOW DID MY DNA MAKE ME? Just google the title to access the article.

    This article will give you a good understanding of how DNA, as well as cloning and genetic engineering. You also learn that so-called “Junk DNA” isn’t junk at all. You will learn why it is not rational to believe that DNA code could have arisen by chance. Science points (not proves, but points) to an intelligent cause for DNA code.

    What about genetic and biological similarities between species? Genetic information, like other forms of information, cannot happen by chance, so it is more logical to believe that genetic and biological similarities between all forms of life are due to a common Designer who designed similar functions for similar purposes. It doesn’t mean all forms of life are biologically related! Only genetic similarities within a natural species proves relationship because it’s only within a natural species that members can interbreed and reproduce.

    Nature cannot build DNA code from scratch. It requires already existing DNA code to direct and bring about more DNA code or a genetic engineer in the laboratory using intelligent design and highly sophisticated technology to bring DNA code into existence from scratch. Furthermore, RNA/DNA and proteins are mutually dependent (one cannot come into existence without the other two) and cannot “survive” or function outside of a complete and living cell. DNA code owes its existence to the first Genetic Engineer – God!

    Protein molecules require that various amino acids come together in a precise sequence, just like the letters in a sentence. If they’re not in the right sequence the protein won’t function. DNA and RNA require for various their various nucleic acids to be in the right sequence.

    Furthermore, there are left-handed and right-handed amino acids and there are left-handed and right-handed nucleic acids. Protein molecules require for all their amino acids to be left-handed only and in the right sequence. DNA and RNA require for all their nucleic acids to be right-handed and in the right sequence. It would take a miracle for DNA, RNA, and proteins to arise by chance!

    Mathematicians have said any event in the universe with odds of 10 to 50th power or greater is impossible! The probability of just an average size protein molecule (with its amino acids in the right sequence) arising by chance is 10 to the 65th power. Even the simplest cell is made up of many millions of various protein molecules along with and DNA/RNA..

    The late great British scientist Sir Frederick Hoyle calculated that the odds of even the simplest cell coming into existence by chance is 10 to the 40,000th power! How large is this? Consider that the total number of atoms in our universe is 10 to the 82nd power.

    Also, so-called “Junk DNA” isn’t junk. Although these “non-coding” segments of DNA don’t code for proteins, they have recently been found to be vital in regulating gene expression (i.e. when, where, and how genes are expressed, so they’re not “junk”). Also, there is evidence that, in certain situations, they can code for proteins through the cell’s use of a complex “read-through” mechanism.

    Visit my latest Internet site: THE SCIENCE SUPPORTING CREATION (This site answers many arguments, both old and new, that have been used by evolutionists to support their theory)

    Author of the popular Internet article, TRADITIONAL DOCTRINE OF HELL EVOLVED FROM GREEK ROOTS

    *I have given successful lectures (with question and answer period afterwards) defending creation before evolutionist science faculty and students at various colleges and universities. I’ve been privileged to be recognized in the 24th edition of Marquis “Who’s Who in The East.”

    • Your logic needs work. Assuming God created life, it is presumptuous and inappropriate to dictated how God had to have done it or what this life must be like. You are putting yourself above God when you do.
      The presence of junk DNA with no role, says absolutely nothing about whether or not God created life. In the Bible, it says what God created was “good”. It does not say it was perfect, or would remain perfect if it was.
      You make anyone who accepts that God created life look like a nincompoop.
      Similar arguments were made hundreds of years ago about orbits. It was heretical to claim that orbits were anything other than circular, because circular was assumed to be perfect. No room in their thinking for eclipses. But where did they get this? Entirely fabricated. Nonsense logic.

    • “The probability of just an average size protein molecule (with its amino acids in the right sequence) arising by chance is 10 to the 65th power.”

      There are many unstated and unexamined assumptions that go into such a calculation. You are also assuming that Hoyle was in possession of perfect knowledge and that therefore his calculation is correct. Only considering the ‘facts’ that support your position is known as “cherry picking.”

      You might find the following short article to be interesting:

      • Your comment has nothing whatsoever to do with mine. You need to reread.

        • Mindbreaker,
          You are right. The comment was addressed to Ranganathan. If you re-read and pay attention to the indentations imposed by the website, you will see that it was not addressed to you. However, to make things easier on you, I will try to remember to open with the person my comment is addressed to.

      • Torbjörn Larsson | December 10, 2021 at 4:27 pm | Reply

        We know the nodel is irrelevant for the same reason why evolution happens all the time – it isn’t modifying proteins wholesale but massively parallel mutating them from what works to what may work better (or worse). It was rapid and fecund and evolution all the way from the split between geology and biology [ ].

        Anyway, creationism is even more improbable since it want to poof multiple ancestors with magic. A Nature paper estimated that the natural universal common ancestry had a fantastic likelihood of 10^2000+ more likely compared to multiple (say, creationist) ancestry. So when creationist say something like 10^-100 likelihood or so, it would still be far larger than their model (which they refuse to quantify, but we can see why – it is ridiculous based on probabilities alone, even before we get to the nonexistence of magic agents).

    • Torbjörn Larsson | December 10, 2021 at 4:17 pm | Reply

      Irrelevant, and erroneous anti-science superstition anyway – so pseudoscience links at best.

      Please stick to science when commenting on science.

  3. While this is a great technique for collecting genomes, I am dubious on the dating. Any animals roaming around are going to be moving the soil around, especially when herds move across the ground. This churning of the soil, would likely limit using depth, especially the first 6 or so inches, to indicate a gradation of time.
    And one would expect, if something were to go extinct 10k-30k years ago, for this churning to produce less and less DNA higher, but still be present as high as you want to look. Take enough samples, and you will get mammoths that, this technique says died last week.

    • Not just physical churning of the soil, but diffusion of very small molecules, particularly when suspended in water, is going to make the dating of the DNA problematic.

      It is interesting work, but I’m not confident that the dates associated with the DNA can be trusted.

    • Torbjörn Larsson | December 10, 2021 at 3:54 pm | Reply

      I haven’t looked into the paper, but datings are usually done by several methods and in the case of ancient DNA by rough dating it as well. Ancient DNA spontaneously breaks into fragments as well as converts the cytosine nucleobase to uracil by deamination [ ] and that can be used for rough dating.

    • Torbjörn Larsson | December 10, 2021 at 3:59 pm | Reply

      I should add that they wouldn’t have published a dating unless they had established uncertainty of the method. (And come to think about it, since this is a regional climate change of environment, they have more proxies for dating than usual – e.g. change in temperatures, soil pH, spores, pupas, et cetera.)

      By your own argument we know it isn’t a fuzzy date precisely because they couldn’t find traces of several species younger than 5 kyrs.

      • Torbjorn,
        Precision and accuracy are two different concepts. One can have high precision with low accuracy. And, that is my point. Whatever method that is used, even if the matrix is known with high precision and accuracy, if the DNA is moving around, it might be known with high precision but low accuracy. Estimating age by DNA mutations can be species specific and generally has both low precision and low accuracy.

      • Torbjorn
        The actual article says:
        “However, interpreting cryptic populations with sedaDNA necessitates caution. As noted previously, Arnold et al.91 found that although permafrost contains a wealth of well-preserved eDNA, the favourable characteristics of perennially frozen ground increases the likelihood for allochthonous organics to survive transport and be redeposited within younger strata. They argue that while reworking is of lesser concern when assessing first appearance dates and “abundant” sedaDNA signals, reworking of older sediments can be an inherent problem when assessing last appearance dates in high-energy fluvial contexts or in areas of thermokarst where older sediments thaw and mobilize followed by potential re-aggradation of permafrost.”

        They dismiss migration through ice, but don’t acknowledge that the tundra may have previously melted.

  4. I am also suspicious of the motivations of these researchers. These are Anthropologists, not Paleontologists. One can be forgiven for thinking that they might be motivated to exonerate the early humans of the charge of wiping out the Mammoths, and other megafauna.
    I think, to support their claims, they need to show changes in the prevailing genetic make up of the Mammoths over the time period investigated. Obviously, if climate was changing, there should be some genetic shift, even if insufficient to save them.
    If, on the other hand, it is just churned soil, there should be no pattern of change, beyond what sample size can explain.

    • Torbjörn Larsson | December 10, 2021 at 4:05 pm | Reply

      Author list is mixed:

      McMaster Ancient DNA Centre, McMaster University, Hamilton, Canada

      Tyler J. Murchie, George S. Long, Tara Sadoway, Emil Karpinski & Hendrik N. Poinar

      Department of Anthropology, McMaster University, Hamilton, Canada

      Tyler J. Murchie & Hendrik N. Poinar

      Department of Earth and Atmospheric Sciences, University of Alberta, Edmonton, Canada

      Alistair J. Monteath, Matthew E. Mahony, Scott Cocker & Duane Froese

      School of Geography and Environmental Science, University of Southampton, Southampton, United Kingdom

      Alistair J. Monteath

      Department of Biology, McMaster University, Hamilton, Canada

      George S. Long & Emil Karpinski

      The Hospital for Sick Children, Toronto, Canada

      Tara Sadoway

      Yukon Government, Palaeontology Program, Department of Tourism and Culture, Whitehorse, Canada

      Grant Zazula

      Collections and Research, Canadian Museum of Nature, Ottawa, Canada

      Grant Zazula

      Division of Vertebrate Zoology/Mammalogy, American Museum of Natural History, New York, United States

      Ross D. E. MacPhee

      Department of Biochemistry, McMaster University, Hamilton, Canada

      Hendrik N. Poinar

      Michael G. DeGroote Institute for Infectious Disease Research, McMaster University, Hamilton, Canada

      Hendrik N. Poinar

      CIFAR Humans and the Microbiome Program, Toronto, Canada

      Hendrik N. Poinar.

      Conclusions too:

      ““I think a combination of climate, ecological, and anthropogenic pressures best explains the losses, but more research is needed to really solve that problem that Quaternary scientists have been grappling with for some 270 years,” Murchie said.”

      [ ]

  5. “Scientists also stress the need to gather and archive more permafrost samples, which are at risk of being lost forever as the Arctic warms.”

    They seem to not be aware that the Arctic has warmed several times over the time period they are investigating.

    • Torbjörn Larsson | December 10, 2021 at 4:09 pm | Reply

      They likely mean recent samples, which is the ones they study.

      But if man made global warming takes us out of the glacial cycles – which it may if it goes towards 2-4 degC increase, most or all will eventually disappear. So that is another concern, but not the one I think they referred to.

  6. To treat horses as natives of North America because they lived here a long time ago would not have helped the natives who lived here without them before the Spaniards came 🙂

    • Torbjörn Larsson | December 10, 2021 at 4:14 pm | Reply

      They didn’t seem to domesticate horses, if that is what you mean. The latest genetic evidence seems to say that it happened in Asia two times.

      The first time didn’t lead to modern horses but some eloped and were ancestors to Przewalski’s horse. The second time of domestication selected for genes that made for stronger horse backs, and became ancestors to the modern horse. But similar to the results here, I’m sure that research is still an open area, so the story will evolve too.

  7. I think ot could ne interesting to see large herbivores on grasslands I. Canada and Siberia. BUT damn, if I think they would appreciate a wolly coat and climate change.

  8. ericwynnetaylor | December 11, 2021 at 9:37 pm | Reply

    I am waiting for the Camel to be brought back to America, perhaps a Camel 🐫 National Park as well as a Horse 🐎 National Park; these nature beings must have tasted pretty 😍 good to have been ghosted from their birth lands?

Leave a comment

Email address is optional. If provided, your email will not be published or shared.