Evolution Icon Evolution

Mammoth Support for Devolution

Michael Behe
Image credit: Thomas Quine, CC BY 2.0 , via Wikimedia Commons.

The more science progresses, the more hapless Darwin seems.

In my 2019 book Darwin Devolves I showed that random mutation and natural selection are powerful de-volutionary forces. That is, they quickly lead to the loss of genetic information. The reason is that, in many environmental circumstances, a species’s lot can be improved most quickly by breaking or blunting pre-existing genes. To get the point across, I used an analogy to a quick way to improve a car’s gas mileage — remove the hood, throw out the doors, get rid of any excess weight. That will help the car go further, but it also reduces the number of features of the car. And it sure doesn’t explain how any of those now-jettisoned parts got there in the first place.

The Bottom Line

The same goes for biology. Helpful mutations that arrive most quickly are very much more likely to degrade genetic features than to construct new ones. The featured illustration in Darwin Devolves was the polar bear, which has accumulated a number of beneficial mutations since it branched off from the brown bear a few hundred thousand years ago. Yet the large majority of those beneficial mutations were degradative — they broke or damaged pre-existing genes. For example, a gene involved in fur pigmentation was damaged, rendering the beast white — that helped; another gene involved in fat metabolism was degraded, allowing the animal to consume lots of seal blubber, its main food in the Arctic — that helped, too. Those mutations were good for the species in the moment — they did improve its chances of survival. But degradative mutations don’t explain how the functioning genes got there in the first place. Even worse, the relentless burning of genetic information to adapt to a changing environment will make a species evolutionarily brittle and more prone to extinction. The bottom line: Although random mutation and natural selection help a species adapt, Darwinian processes can’t account for the origins of sophisticated biological systems.

In Darwin Devolves, I also mentioned work on DNA extracted from frozen woolly mammoth carcasses that showcased devolution: “26 genes were shown to be seriously degraded, many of which (as with polar bear) were involved in fat metabolism, critical in the extremely cold environments that the mammoth roamed.” It turns out that was an underestimate. A new paper1 that has sequenced DNA from several more woolly mammoth remains says the true number is more than triple that — 87 genes broken compared to their elephant relatives. The authors write of the advantages provided by destroyed genes (references omitted for readability):

Gene losses as a consequence of indels and deletions can be adaptive and multiple case studies investigating the fate of such variants have uncovered associations between gene loss and mammalian phenotypes under positive selection. In laboratory selection experiments, gene loss is a frequent cause of adaptations to various environmental conditions. Given that we focused on those indels and large deletions that are fixed among woolly mammoths, the majority of these protein-altering variants likely conveyed adaptive effects and may have been under positive selection at some point during mammoth evolution. We did not find specific biological functions overrepresented among these genes (see methods), but many of the affected genes are related to known mammoth-specific phenotypes, such as total body-fat and fat distribution (EPM2A, RDH16, and SEC31B), fur growth and hair follicle shape and size (CD34, DROSHA, and TP63), skeletal morphology (CD44, ANO5, and HSPG2), ear morphology (ILDR1 and CHRD), and body temperature (CES2). In addition, we find several genes associated with body size (ZBTB20, CIZ1, and TTN), which might have been involved in the decreasing size of woolly mammoths during the late Pleistocene.

There’s Lots More

The point is that these gene losses aren’t side shows — they are the events that transformed an elephant into a mammoth, that adapted the animal to its changing environment. A job well done, yes, but now those genes are gone forever, unavailable to help with the next change of environment. Perhaps that contributed to eventual mammoth extinction.

As quoted above, the mammoth authors note that gene losses can be adaptive, and they cited a paper that I hadn’t seen before. I checked it out and it’s a wonderful laboratory evolution study of yeast.2 Helsen et al. (2020) used a collection of yeast strains in which one of each different gene in the genome had been knocked out. They grew the knockout yeast in a stressful environment and watched to see how the microbes evolved to handle it. Many of the yeast strains, with different genes initially knocked out, recovered, and some even surpassed the fitness of wild-type yeast under the circumstances. The authors emphasized the fact of the evolutionary recovery. However, they also clearly stated (but don’t seem to have noticed the importance of the fact) that all of the strains rebounded by breaking other genes, ones that had been intact at the beginning of the experiment. None built anything new, all of them devolved.

Well, Duh

That’s hardly a surprise. At least in retrospect, it’s easy to see that devolution must happen — for the simple reason that helpful degradative mutations are more plentiful than helpful constructive ones and thus arrive more quickly for natural selection to multiply. The more recent results recounted here just pile more evidence onto that gathered in Darwin Devolves showing Darwin’s mechanism is powerfully devolutionary. That simple realization neatly explains results ranging from the evolutionary behavior of yeast in a comfy modern laboratory, to the speciation of megafauna in raw nature millions of years ago, and almost certainly to everything in between.


  1. Van der Valk, Tom, et al. 2022. Evolutionary consequences of genomic deletions and insertions in the woolly mammoth genome. iScience 25, 104826.
  2. Helsen, J. et al. 2020. Gene loss predictably drives evolutionary adaptation. Molecular Biology and Evolution 37, 2989–3002.

Michael J. Behe

Senior Fellow, Center for Science and Culture
Michael J. Behe is Professor of Biological Sciences at Lehigh University in Pennsylvania and a Senior Fellow at Discovery Institute’s Center for Science and Culture. He received his Ph.D. in Biochemistry from the University of Pennsylvania in 1978. Behe's current research involves delineation of design and natural selection in protein structures. In his career he has authored over 40 technical papers and three books, Darwin Devolves: The New Science About DNA that Challenges Evolution, Darwin’s Black Box: The Biochemical Challenge to Evolution, and The Edge of Evolution: The Search for the Limits of Darwinism, which argue that living system at the molecular level are best explained as being the result of deliberate intelligent design.



Arcticbrown bearDarwin DevolvesDarwinian processesdeletionsdevolutionDNAevolutiongenesgenetic informationindelsmetabolismmutationsnatural selectionPleistocenerandom mutationResearchseal blubberwoolly mammothyeast