Vestigial limb muscles in human embryos show common ancestry—for the gazillionth time

October 6, 2019 • 9:00 am

There are three kinds of vestiges that constitute evidence for evolution, or rather its sub-claim that modern species share common ancestors. I discuss all three in Why Evolution is True:

1.) Vestigial traits that persist in modern species but either have no adaptive function in a species or a function different from the one served in that species’ ancestors. The vestigial ear muscles of humans are one, the flippers of penguins (functional, but not for flying in the air) is another, and the coccyx in humans (sometimes with attached “tail muscles” that can’t move it) is a third.

2.) Vestigial genes that are functional in our relatives (and presumably in our ancestors) that have been inactivated in some modern species. There is no explanation for these “dead genes” save that they were useful in ancestors but aren’t useful any longer. Examples are “dead” genes that code for egg yolk proteins in humans (but don’t produce them); a dead gene for vitamin C synthesis in humans (we don’t make the vitamin because that gene is inactivated, but rather get it from our diet; and the many dead “olfactory receptor” genes in cetaceans (whales, dolphins, etc.)—genes that were active in their terrestrial ancestors but became inactivated because “smelling” underwater uses different genes and traits.

3.) Features in development that are transitory, and whose appearance makes sense only under the supposition that those features were present in common ancestors and persist in some descendants but not others. The lanugo (a transitory coat of hair in human embryos) is one.

Today’s paper, which just appeared in the journal Development, shows several other “transitory” evolution-attesting features. Diogo et al. show that human embryos develop muscles that disappear as development proceeds, but those muscles don’t disappear in some of our relatives, including closely related ones like other primates as well as distant relatives like reptiles.

Moreover, these muscles, which disappear in most human embryos, sometimes don’t disappear, persisting in adults as rare and nonfunctional variants. Or they appear in malformed individuals, with both phenomena often seen in “vestigial traits”. For example, some people are born without wisdom teeth, considered a vestigial holdover from our ancestors; and the functionality of human vestigial ear muscles that move the ears in our relatives, like cats and dogs, is variable: some people like me are able to move those muscles and wiggle their ears, while others can’t.

Click on the screenshot below to access the paper, and the pdf is here (reference at the bottom of this post).

The authors visualized the muscles in the embryonic arm and leg by doing immunostaining—using antibodies that would affix to proteins in the muscles and also carried ancillary molecules that would make those muscles more easily visualized under the microscope in a three-dimensional way. The authors used 70 antibodies, but the main ones bound to muscle-specific proteins like myosin and myogenin.

They stained the mounted limb sections of 13 embryos (presumably from abortions) ranging from nine to thirteen weeks after gestation (quantified as “gestational weeks”, or GWs), and with the standard measurement “crown-rump length” (CR) ranging from 2.5 cm to 8.0 cm (about 1 to 3 inches). These were thus very small embryos, but the sophistication of the technique, and the efficacy of the stain, combined with our knowledge of embryonic development and tetrapod muscle anatomy, enabled the authors to produce pictures like these: the muscles in the hands of a 10 and 11-GW fetus:


What they found is that human embryos show a number of muscles present in the adults of some other tetrapods (including our closest relatives, the chimps), but that disappear during human development, with a few of these “atavistic muscles” fusing with other muscles in human fetuses although remaining distinct in our tetrapod relatives.

Here’s how the authors describe the main results, listing some of the atavistic muscles in the embryos (I’ve put them in bold):

As summarized in Tables 2-5 and also noted above, various atavistic muscles that were present in the normal phenotype of our ancestors are present as the normal phenotype during early human ontogenetic stages and then disappear or become reduced and completely fused with other muscles, thus not being present/distinguishable in human adults. These include the upper limb muscles epitrochleoanconeus (Fig. 3), dorsoepitrochlearis, contrahentes 3-5 (Fig. 4) and dorsometacarpales 1-4 (Figs 3-5), and the lower limb muscles contrahentes 3-5, dorsometatarsales 1-4 (Fig. 6) and opponens digiti minimi (Fig. 6). These muscles are present in some other tetrapods, as shown in Tables 6 and 7, which summarize the comparisons with other limbed vertebrates. Of all these muscles, only the dorsometacarpales often remain in adults, fused with other muscles: all the others are normally completely absent in human adults. Fascinatingly, all these atavistic muscles are found both as rare variations of the normal adult population and as anomalies in individuals with congenital malformations such as those associated with trisomies 13, 18 and 21, reinforcing the idea that such variations and anomalies can be related to delayed or arrested development.

Here are two of the fetal atavistic muscles. First, the dorsometacarpales in the hand, which are present in modern adult amphibians and reptiles but absent in adult mammals. The transitory presence of these muscles in human embryos is an evolutionary remnant of the time we diverged from our common ancestor with the reptiles: about 300 million years ago. Clearly, the genetic information for making this muscle is still in the human genome, but since the muscle is not needed in adult humans (when it appears, as I note below, it seems to have no function), its development was suppressed:


Here’s a cool one, the jawbreaking “epitrochleoanconeus” muscle, which is present in chimpanzees but not in adult humans. It appears transitorily in our fetuses. Here’s a 2.5 cm (9 GW) embryo’s hand and forearm; the muscle is labeled “epi” in the diagram and I’ve circled it:

This muscle must have become nonfunctional, and reduced in development, over the last six million years or so, when the common ancestor of humans and chimps gave rise to our separate lineages.

An interesting sidelight of this study is that some of these vestigial muscles occur as rare variants in adult humans, either via developmental “accidents” or as part of congenital malformations. Presumably these screwups in development block the genetic changes that normally lead to the suppression and disappearance of muscles in embryos. Variable expression of vestigial traits is common in organisms where the traits haven’t evolved into something else that’s useful. (For more on human vestigial traits, see the Wikipedia article on “human vestigiality”). The authors note that when the muscles do appear in adults, they are “functionally neutral, not providing any type of major functional advantage or disadvantage.”

The presence of these vestigial muscles is pretty irrefutable evidence of evolution and common ancestry, for there’s no reason why either God or an Intelligent Designer (a pseudonym for “God” to ID advocates) would put a transitory muscle in a human fetus that’s of no use whatsoever, but just happens to resemble the fetal muscles that goes on to develop into adult muscles in our relatives.  I wonder how creationists, including IDers, will explain this as the work of a designer. Will they say the muscles are really functional in a fetus? If so, why do they disappear? And doesn’t the fact that they go on to develop into functional muscles in our relatives like chimps and reptiles say something about common ancestry?

Two more points:

1.) The order of appearance of these muscles in development doesn’t completely comport with their order of evolution. This shows that the “recapitulation theory”—that the order of development mimics the order of evolution—isn’t completely obeyed. But we’ve known that for a long time. The time of appearance of a trait in development can be changed by other factors, like its usefulness in “priming” the development of other features. But this doesn’t overturn the very strong conclusion that the presence of transitory muscles in the human fetus that remain in adults of our relatives is evidence for evolution.

2.) Finally, muscles in the arms and legs that appear “homologous” (i.e., have the same evolutionary origin) may have had independent evolutionary origins, and may involve different genes, so they’re not really “homologous” in the way evolutionists use that term. As the authors note,

These differences support the emerging idea that the topological similarities between the hand and foot of tetrapods, such as humans, are mainly secondary (see recent reviews by Diogo et al., 2013, 2018; Diogo and Molnar, 2014; Sears et al., 2015; Miyashita and Diogo, 2016). This idea is further supported by the fact that the order of developmental appearance of the hand muscles is markedly different from that of the corresponding foot muscles (Tables 6, 7). As an illustrative example, whereas the lumbricales are the first muscles to differentiate in the hand, together with the contrahentes (Table 6), in the foot the lumbricales differentiate only after most other foot muscles are already differentiated (Table 7). Thus, these developmental data and evidence from comparative anatomy and from the evolutionary history of human limb muscles (see Tables 6, 7) indicate that several of the muscles that seem to be topologically similar in the human upper and lower limbs actually appeared at different evolutionary times; appear in a markedly different ontogenetic order; derive from different primordia; and/or are formed by the fusion of different developmental units in each limb.

Now the authors didn’t do this study to demonstrate evolution; like most rational people, they accepted it long ago. Rather, their stated aim was to “build an atlas of human development comprising 3D images. . . that can be used by developmental biologists and comparative anatomists, as well as by professors, students, physicians/pathologists and the broader public.” But one of the bonuses, especially for the broader public, is the very clear demonstration of the common-ancestry tenet of modern evolutionary theory.

h/t: Liz


Diogo, R., N. Siomava, and Y. Gitton. 2019. Development of human limb muscles based on whole-mount immunostaining and the links between ontogeny and evolution. Development 146:

51 thoughts on “Vestigial limb muscles in human embryos show common ancestry—for the gazillionth time

      1. Thanks for the link. I had no idea they’d published this but I consider it a new low for Quillette, to publish a Discovery Institute propaganda piece, even as ‘right of reply’.

        I don’t know if Jerry will reply or whether he’s been offered the chance to, but surely at some point with these DI people life is just too fucking short.

        1. I found it very amusing that an ad, smack in the middle of this dubious article, is for something called “The Messianic Prophecy Bible Project”. I wonder how *that* happened. Intelligent design? 😉

      2. Wow. I waded through the equivocation on “lifetime”, and got as far as the incredulity about a tiny part of the Cambrian: “an enormous transition within 13 million years”. As if 13000000 years isn’t enough time for hundreds of thousands of generations, extensive environmental changes, etc.
        So, just out of curiosity, I thought “what were human evolutionary ancestors like 13 million years ago?”. That would be somewhere in the middle of the Miocene epoch, which *spanned* 18 million years, from 23 myo to 5 myo. No hominids until the end of the Miocene or later (maybe – at a stretch – Ardipithecus just around the end:; that gets at most only half-way to 13 million years ago.
        It seems that the biggest problem they have – and I’m not talking about the YEC crowd – is a failure to grasp geologic time scales. The second biggest problem seems to be a misconception (probably due to a misinterpretation of Gouldian punk eek) that a species persists completely unchanged for a “lifetime” of millions of years, then suddenly poofs (or is poofed) into something completely different (confirmed misconception, two paragraphs later).

        1. There was a big row a handful of years ago about a commercially traded fossil from the Messel Shale of modern Germany, which was touted as “our earliest ancestor. I’m trying to remember it’s nick name, but the specific name was, I’m pretty sure, darwin-something-sis.
          Ah, found it, Darwinius masillae . But Messel is older than I’d remembered being well down in the Eocene (36-56 Myr BP), substantially before your 13 Myr window.
          How about Proconsul – a family of mid-Miocene apes from East Africa? They’re pretty unspectacularly similar to various apes of today – you’d need to get thoroughly into their anatomy to differentiate them from a modern ape.
          There have been a number of recent arguments – based on incomplete fossils in the approximate time range you’re looking at – that people use to argue for origins of the hominid line being in either SE Europe or SE Asia, such as Lufengpithecus, Gigantopithecus and Dryopithecus. They’re sufficiently homo-like that it is hard to be sure if the line originated in Africa, or in SE Asia, or even in SE Europe, and then went through various migrations and speciations before the rise of Australopithecine apes in E Africa at around 4 to 5 Myr.
          Personally, I think it significant that the first serious archaeological exploration of the “Sahel” region turned up Sahelanthropus tchadensis at about 7 Myr. It’s a difficult and dangerous (politically) area to work in, but they popped out a significant result on their first exploration – that’s strongly encouraging. Pity that there are secessionist and religious-fundamentalist movements operating in the area.
          The short version is that while the development of the Australopith -to- human line in East Africa is well established, we’re really not sure about the previous 5-plus million years. Which would probably attract more attention from the God-Squad, if it didn’t require them to address the existence of the rest of the fossil record. The jaw-breaker names probably put some of them off too.

    1. I know I’m commenting here kind of late, but I hope Coyne will eventually respond to the Quillette article. If you search for that article’s title at Google, it’s evident that a lot of people have been talking about it, and a large portion of them find Bechly, Miller and Berlinsky’s arguments persuasive. The absence of any rebuttal to arguments like these is one of the ways that ideas such as Intelligent Design grow in popularity.

  1. The bonus for most of us here as readers is the clear demonstration as you state of common-ancestry within evolution. Certainly you must be reasonable because there are some who do not give it a thought, these vestiges or if they do it is only to find a way to deny them. Kind of reminds us of another behavior these days.

  2. “muscles in the arms and legs that appear “homologous” (i.e., have the same evolutionary origin) may have had independent evolutionary origins, and may involve different genes, so they’re not really “homologous” ” that really smashes me, I find that so deeply counterintuitive, and the anatomy appears so clear, but yet…. I’m kinda shellshocked now.

  3. And yes (recovering a bit from being shellshocked), vestiges, all three types of them, are about the strongest arguments against ‘intelligent’ design, and in favour of evolution.
    [For 3 I think the hindleg buds in cetacean embryos is the most classical example (at least it is the one I use in lectures)]
    And what wonderfully beautiful immuno-staining photographs!

    1. “vestiges, all three types of them, are about the strongest arguments against ‘intelligent’ design”

      Four thousand people a year get injured by their own clothing. Eight hundred people injure themselves while cleaning their teeth.
      I have a very close relative who managed to accidentally _post her mobile phone to Australia._

      Once you start looking for arguments against ‘intelligent’ design, you see them everywhere.

        1. 🙂 I’ve never been able to get the precise details out of her. She really doesn’t like being reminded of it. Which is a shame to me because it’s something I like bringing up at every opportunity.

          A couple of weeks ago I was talking to her on the phone – she works in international aid and she’s in charge of a very large sum of money, so I jokingly said she should be careful she doesn’t accidentally post it to Australia. I thought it was funny – she did not.

  4. Proponents of the Great Designer will be aware that He is not only intelligent, but mischievous and possessed of a quirky sense of humor. This explains why so much is designed to look as if it evolved, and also why so much (like benign prostatic hyperplasia) doesn’t seem so intelligent.

  5. Thanks for posting about these utterly fascinating findings.

    BTW, my wife was born without wisdom teeth. Lucky woman! Incidentally, at 50, mine are still impacted, and I’ve never had a dentist that wanted to dig them out. One said I had a nerve that wasn’t in the right place and he didn’t want to risk damaging it.

    My mother (who doesn’t accept evolution) can also wiggle her ears. Oh the irony.

  6. That was a beautifully written summery of a scientific paper which anyone interested in nature can enjoy.

  7. As someone interested in genetics and biology, I really enjoyed this article. Sadly, intelligent design is one of those examples of ‘common sense’ that isn’t. Like flat earth theorists, ID is like looking at the world from the bottom of a very deep, very narrow well.

  8. Very interesting. I had considered penguin flippers to be re-purposed wings; flying wings that are now adapted for swimming. Their contour and flight feathers do seem vestigial.

  9. “indicate that several of the muscles that seem to be topologically similar in the human upper and lower limbs actually appeared at different evolutionary times; appear in a markedly different ontogenetic order; derive from different primordia; and/or are formed by the fusion of different developmental units in each limb.”

    As noted in comments above, this seems surprising, but that is a pretty solid list of reasons to think it is true.

    I am also amazed at how intricate things have gotten in the hand by just 10 weeks into development, and agree with the kudos to the investigators for some impressive imaging work.

    1. I’m rather surprised as well. And this fact seems to contradict the main point in Jerry’s post. If those apparently homologous organs don’t have a common evolutionary origin, they are not a proof of evolution. Am I missing something?

  10. “Clearly, the genetic information for making this muscle is still in the human genome”

    Is this really clear? In the case of brain development, we know there isn’t enough information in the genome to precisely specify all the neurons and their interconnections. Rather, neural pathways proliferate early in development and are then winnowed by a functional selection process that determines which connections persist and which wither away.

    Could something similar be true of muscular development? Maybe it’s routine for vertebrate embryos to begin building proto-muscles at a wide variety of skeletal sites that happen to trigger that development program. The ones that turn out to be load-bearing or otherwise functional persist, and the ones that don’t get resorbed or fused with neighboring muscles.

    On this view the loss of certain muscles in adult humans as compared to chimps would be less a matter of genes for muscle building than of different skeletal anatomy producing a different selective environment during muscle development.

    1. Pruning is clearly going on with these muscle tissues. This does not negate that muscle genes would remain in the human genome. The genes that spawn excess neurons are not lost either.

      1. I didn’t mean to imply that there aren’t any genes for muscle-building; clearly there are. What I’m questioning is the (apparent) suggestion that there are genes for specific muscles, and other genes that suppress specific muscles. I think it plausible that the specific muscles we have are the result of an interplay between a generic muscle-building program and the specifics of our skeletal growth.

        1. Yes, of course there are not genes for specific muscles; what I said is that information required to make these muscles remains in the genome. Nor did I say that there are genes to suppress specific muscles, but there are genes that had a role in suppressing the appearance of these muscles. That information in all probability is still in the DNA, but gets un-suppressed so that the muscles appear in the same way they appear in the embryos of our relatives. I stand by that, as I see no alternative explanation “an interplay between a generic muscle-building program and the specifics of our skeletal growth” is pretty much another way of saying what I said above, except that I doubt that there is a “generic muscle-building program” that simply says “build muscles”. There must be genes turning muscle formation on and off at specific times and places. Remember, we still have egg yolk protein genes in our DNA, despite the fact that our lineage hasn’t made egg yolks since we diverged from the reptiles (except for monotreme mammals).

          1. I agree that there must be genes for turning muscle formation on and off. But I think “at specific places” would be better expressed as “under specific conditions”.

            When I was 13 I broke my left arm and wore a cast on it for a couple of months. When the cast came off, I had pubic hair on the back of my left hand. Not because I had genes specifying pubic hair in that location, but because at the time the pubic hair program became active, my left hand happened to satisfy the conditions for its growth.

            I’m speculating that muscle formation may be like that. Muscles form wherever conditions are right for their formation, and not because the genome specifies a set of places where muscles ought to be.

    2. Maybe it’s routine for vertebrate embryos to begin building proto-muscles at a wide variety of skeletal sites that happen to trigger that development program. The ones that turn out to be load-bearing or otherwise functional persist,

      The problem with that is that the muscles disappear before birth (otherwise they’d have been recorded at autopsy, long since) leaving a period for the selection to occur when the muscles are not bearing loads and there is no call for manual dexterity. So … how is this culling (of muscles) managed?

      1. That’s a fair question, to which I don’t have a good answer. But we do know that fetuses aren’t entirely passive; they do move in utero. Maybe part of the reason for that is to test muscle function during development.

        1. And the mechanism for feeding back the results of that test into development is …? The energy expended in performing tests like this is wasted if there isn’t a way of acting on the results.
          We know that half-to-a-majority of conceptuses spontaneously abort, but that is long before the development of skeletal muscles.

          1. Muscles that do work get stronger, and muscles that are just along for the ride tend to atrophy. That’s how it works in adults; why shouldn’t it work the same way during fetal development?

  11. I saw this article – it is very interesting & I was about to comment when you answered my question in your next paragraph!

    I am curious though – presumably no one has done the same experiment with the other great apes – have they ‘lost’ traits we have retained, that are found in foetuses?

    The same research should also be done in all other vertebrates…

    1. As for the god botherers, I suppose the CoE type would say they accept evolution happens & that it was god’s way of setting up life – he started the ball rolling.

      Of course we all know that leaves lots of additional philosophical issues…

    2. Foetuses of great apes other than humans are sufficiently rare and valuable that the amount of experimental material is going to remain very limited.

  12. Fascinatingly, all these atavistic muscles are found both as rare variations of the normal adult population

    I’m thinking, for the second time in a week, of the alleged Roman senator and horse, Incitatus, who … is being the subject of misleading memory. I had thought it (he? or gelded?) had vestigial toes to either side of the main hoof. But no mention of that.
    On the other hand,

    The 2000 BBC Radio 4 comedy Me and Little Boots, by Shaun McKenna, told the story of Caligula (Latin for “little boots”) from the point of view of Incitatus, who was played by Leslie Phillips.

    “Ohh, I say! Left hand down a bit!”
    Sorry, it was Julius Caesar’s horse that had funny feet. Suetonius, page 61, “He rode a remarkable horse, too, with feet that were almost human; for its hoofs were cloven in such a way as to look like toes. This horse was foaled on his own place, and since the soothsayers had declared that it foretold the rule of the world for its master, he reared it with the greatest care, and was the first to mount it, for it would endure no other rider. Afterwards, too, he dedicated a statue of it before the temple of Venus Genetrix. “

  13. I’ve been reading my son’s freshman college biology book (for fun), and I’ve just finished the chapter on gene activation/suppression, including in embryos. So I read this with a little more understanding than I would have earlier. Pretty cool stuff, and amazing images. Thanks for posting.

  14. Has anyone read the book “Evolution Driven by Organismal Behavior” by the same author of this paper (Rui Diogo)?, any comments on the book?

Leave a Reply