Color-coded diagram of a small bone bed containing at least twelve individuals of the Permian synapsid Suminia. From Frobisch and Reisz (2009) When I hear the phrase “early human relative” I cannot help but think of an ape-like creature. Something like Sahelanthropus fits the bill nicely; it may not be a hominin but it is still a close relative from around the time that the first hominins evolved. That is why I was a bit puzzled to see parroting a story written by the Discovery Channel which proclaimed “Early human relative predates even dinosaurs“! Was this another fossil that would change everything? While not quite as startling as a Precambrian rabbit, a 260-million-year-old-hominin (or even primate) would certainly be a shocker!

The truth of the matter, however, is that the fossil described in the MSNBC story is only a distant relative of humans. It is called Suminia getmanovi, and it was a synapsid that lived during the Permian in what is now Russia. It belonged to a diverse group of herbivorous, non-mammalian synapsids called anomodonts, a group that also contained creatures like the tubby Lystrosaurus and the small, tusked Robertia. As such Suminia did belong to the larger group that encompasses all living mammals (the synapsids), yet it was a distant relative of the first “true” mammals (which did not evolve until about 70 million years after Suminia lived). Despite its superficial appearance Suminia was more closely related to mammals than to living reptiles, but armadillos, cows, bats, whales, cats, elephants, and even our species are more closely related to each other than any is to Suminia.* Attempts to make Suminia relevant to human ancestry is a quick and dirty way of grabbing attention, but in this case I think it stirs more confusion than enlightenment.

*[In other words, all living mammals shared a common ancestor that lived more recently than Suminia, and living mammals shared a more recent common ancestor with Suminia than with reptiles. Common ancestry is the key is navigating through these relationships.]

My gripes about the msm-coverage of this synapsid aside, it really is an interesting creature. While this is the first many people have probably heard of Suminia it is not a “new” fossil. It was originally described in 1994, and the new Proceedings of the Royal Society B paper that has stirred all the media coverage is based upon a jackpot of new, better-preserved specimens studied by paleontologists Jorg Frobisch and Robert Reisz. To understand why Suminia is making headlines, though, we need a little more background information.

As any good paleontologist knows different environments influence fossil preservation in particular ways. An oxygen-depleted lakebed may allow for exquisitely detailed fossils to be formed while animals that live in the canopy of forests are much less likely to be preserved in the fossil record. In this latter example dead animals that make it down to the forest floor may be consumed or destroyed without ever being buried as there is no regular deposition of sediment to preserve them. Even if the body goes untouched by larger scavengers, detritivores like insects, fungi, and bacteria will still break the animal’s body down. If a dead animal falls out of a tree into a nearby lake or stream it might be covered up quickly enough, but in general arboreal animals are rare in the fossil record because they did not live in environments amenable to good fossil preservation.

Thus the fossils of Suminia came as something of a surprise to paleontologists. Frobisch and Reisz determined that Suminia was probably an arboreal animal, and it is (so far) the earliest arboreal vertebrate known. It was clambering through the trees over 30 million years before the chameleon-like drepanosaurids were occupying a similar niche, for example, and it is astounding that so many remains of a tree-living synapsid have been found from deposits of such great age.


Fleshed-out and skeletal restorations of Suminia. From Frobisch and Reisz (2009)

The new Suminia material consists of a single block containing the articulated remains of over a dozen individual animals. While such jumbles can be difficult to study they are also extremely informative, especially since variation between individuals from one location can be studied. Indeed, the individuals on the slab appear to range from sub-adults to adults, and the preservation of the bones suggests rapid burial caused by some sort of minor catastrophe.

This sort of excellent preservation allowed Frobisch and Reisz to study the anatomy of Suminia in detail. (All that was known of Suminia when it was described was a skull.) What the authors found most curious were the hands of the individuals, and it seemed that Suminia had hands and fingers similar to creatures that are arboreal. Suminia had very long fingers, large hands when compared to the rest of its arm, claw-like bones at the ends of its fingers, and a first digit set off at an angle to function like an opposable, grasping thumb. This is just the sort of arrangement that would be expected if it was grasping tree limbs to move through the canopy.


The hand of Suminia (right) compared to terrestrial anomodonts Galechirus (middle) and Robertia (left). Note the claw-like finger bones and divergent “thumb” of Suminia. From Frobisch and Reisz (2009)

The hands of Suminia widely differed from those of terrestrial anomodonts and more closely resembled those of other arboreal vertebrates from various groups and time periods. It appeared that lineages that adopted an arboreal lifestyle evolved similar traits over and over again, these similarities being due to convergent evolution. Comparisons between Suminia and living arboreal vertebrates supported this hypothesis, and Frobisch and Reisz make a very well-supported case that Suminia moved through the trees by clinging and grasping.

Why the ancestors of Suminia moved into the trees, however, is another question. The Permian deposits in which Suminia was found is relatively well-sampled. During the time the little synapsid was skittering through the trees the local environment supported a large number of herbivores (about 83% of all the vertebrates present) and a much smaller number of carnivores (about 13% of all the vertebrates present). While it is true that an arboreal lifestyle may have helped keep Suminia out of the jaws of predators, the authors hypothesize that it was competition with other herbivores that might have driven some synapsids into the trees. These plants would have been a resource unexploited by other vertebrates, and this may have allowed for the evolution of Suminia from more terrestrially-adapted ancestors.

Dinosaurs and hominins make headlines much more frequently, but Suminia justly deserves its 15 minutes of fame. The problem is that Permian synapsids are not often in the news, so any time one is mentioned a considerable amount of background information is necessary to help people understand how they fit into the “tree of life.” Indeed, “anomodont” is far from being a household word, but I hope this essay has provided a little more context for those who are curious about this strange creature from the distant past.

Jörg Fröbisch and Robert R. Reisz (2009). The Late Permian herbivore Suminia and the early evolution of arboreality in terrestrial vertebrate ecosystems Proceedings of the Royal Society B, Online First DOI: 10.1098/rspb.2009.0911


  1. #1 chris y
    July 29, 2009

    Thanks for this: fascinating material that I knew nothing about.

    And I agree that the media selling something like this as a “human relative” is doubly annoying because you just know who will pick up on that phrase and interpret it precisely as a “pre-cambrian rabbit”.

  2. #2 Zach Miller
    July 29, 2009

    Bah, you beat me to it sir.

    Very awesome little critter. Matt Celesky told me about the SVP talk, and I was sorry I missed it. I sent the paper to Will Baird–his head is probably exploding right now.

  3. #3 Anthony Parcero
    July 29, 2009

    I don’t know if MSNBC has updated the title of the article since you wrote this post but it now says: “Mammals’ family tree predates the dinosaurs”. I still think it overly-simplifies the relationship of synapsids to mammals and your arguments are just as valid.

  4. #4 Laelaps
    July 29, 2009

    Anthony; Thanks for the update. Yes, they have updated the headline since this morning. Obviously I can’t know for sure, but I would like to think that I have made a difference.

  5. #5 Zach Miller
    July 29, 2009

    Isn’t the fact that Suminia is arboreal cool enough? A headline like “Earliest arboreal vertebrate found” kind of sums things up nicely.

  6. #6 Will Baird
    July 29, 2009

    It’s FREAKIN Permian “monkey”, it is! Holy moly!

    You didn’t mention the tail. It’s significant in two ways. The first is that it is LONGER than the majority of anomodonts: they trended to shorter and shorter tails. This could have been for balance purposes. Alternately, which the authors actually bring up, it may have been prehensile.


    Monkey! I tell ya! Money! обезьяна! (it was found there after all!)

  7. #7 Laelaps
    July 29, 2009

    Will; Yes, that was an omission on my part. I might write it back in. I was just running short on time and decided to focus on the hands instead, but you are correct to point out the significance of the tail.

  8. #8 Mike Keesey
    July 29, 2009

    “Mammals’ family tree predates the dinosaurs” is accurate enough. Stem-mammals did originate much earlier than dinosaurs. But that entirely misses the point. We have hundreds (thousands?) of fossils of earlier stem-mammals. This is like writing an article on a new species of frog from the Amazon Basic and headlining it, “Amphibians Exist South of the Equator”.

  9. #9 Vasha
    July 29, 2009

    I find the drawing of Suminia remarkable because of how it jars with the way my expectations have been trained by living animals. If I start scanning at the hindquarters, my brain says “lizard”, and then I get to the head which is so entirely unreptilian that I’m startled.

  10. #10 Sean Craven
    July 29, 2009

    I’ve got a for-fun side project going and I’ve been wanting an intelligent species to take the blame for the Permian extinction.

    Here’s the ancestor. Thanks!

  11. #11 Allen Hazen
    July 30, 2009

    Tangential, but is “Robertia broomiana” an example of genus and species names both honoring the same person?

    Suminia, i.i.r.c., just SHOUTS “basal anomodont” when you look at its skull: more than a hint of Dicynodont in the overall shape, but a mouth full of TEETH!

  12. #12 PaoloV
    July 30, 2009

    Excellent article.

    It’s a shame that human links always get highlighted whenever possible – it’s a damning indictment on humankind’s self-importance. Kudos for flagging this – it’s a particular pet hate of mine. Suminia is fascinating in its own right.

  13. #13 David Marjanović
    July 30, 2009

    I’d like to mention the amazing reversal that’s shown in the 3rd figure you copy: the three bones marked “disc” are reappeared phalanges that had been lost. Galechirus and Robertia have a phalangeal formula of 2-3-3-3-3 just like us, but Suminia has reverted to 2-3-4-5-3 by making the lost phalanges (which had become disc-shaped before disappearing) reappear somehow.

    The German ö designates the French e/eu sound. It isn’t similar to o in pronunciation, so it shouldn’t be represented as such. You have plenty of options:

    - Write “oe”, which is the accepted ASCIIfication.
    - Copy & paste “ö” from the reference or from this comment.
    - Write “ö”.
    - Write the other code that I can’t find at the moment, the one with the number… for example, ? is “ć”.
    - Use the character table to insert it. In Windows, it’s rather hidden (Start – All Programs – Accessories – Character Table); on the Mac, just click on the flag near the top right corner of your screen, and “character table” appears in the menu.
    - Install a keyboard layout which has it (German, Swedish, Finnish, Estonian, Hungarian, Turkish, whatever) and then just type it. There’s a shortcut to switch between keyboard layouts in Windows, and additionally the abovementioned flag on the Mac.

  14. #14 David Marjanović
    July 30, 2009

    Start – All Programs – Accessories – Character Table

    Start – All Programs – Accessories – System Programs - Character Table.

  15. #15 cromercrox
    July 30, 2009

    Please belabour all media ignoramuses around the head with copies of Henry Gee’s ‘In Search of Deep Time’.

  16. #16 Edgar
    July 30, 2009

    “divergent thumb”=opposable thumb?

    Sort of disturbing to see a reptilelike being with primate hands and feets…….

  17. #17 JDP
    July 31, 2009

    It should be noted that various arboreal squamates have somewhat divergent thumbs (see: [i]Corytophanes[/i], for example). Numerous arboreal squamates have prehensile tails, as well (see: arboreal [i]Varanus[/i], the anguid genus [i]Diploglossus[/i], the scincid [i]Corucia[/i], chameleons, etc). It’s cool to see that basal therapsids were doing the same thing.

  18. #18 DDeden
    August 3, 2009

    @ 13 – DM, what indication is there that the ancestors of Suminia had had 2-3-4-5-3 configuration? (‘reverted’?) My interpretation is that 2-3-3-3-3 was the primitive condition since the separation from crustaceans (or so), since crabs share it, and the ‘missing’ thumb bones became the fused mandible in tetrapods.

    I would instead suggest the discs are derived duplicates (or bony growths) of adjacent bones rather than reversals to a former state.

    Just as acanthostega developed duplicate digits (8, for better hydrodynamics, like ray-fin fish) upon returning to full-time aquatic habit from a 5-digit amphibious tiktaalik-like predecessor, I’d think Suminia derived (genetic or environmental/epigenetic) duplicate boney disks in the digits, due to selection for improved grasping of round plant stems, thus indicating permanent dwelling in a new niche.

    Note that arboreal apes tend to have highly derived curved-bone digits, while man does not; if Suminia had no ancestors with curved digits, perhaps a simple mutation or even epigenetic “growths” forming bony disks (paralleling ear exostosis in the ear canal due to persistent immersion in cold water as in ‘surfers ear’) allowed closer grip and stronger muscle grasping of thick branches and faster locomotion in the arboreal canopy. Agree?

  19. #19 dinogorgon
    August 4, 2009

    “[i]My interpretation is that 2-3-3-3-3 was the primitive condition since the separation from crustaceans (or so), since crabs share it, and the ‘missing’ thumb bones became the fused [b]mandible[/b] in tetrapods.[/i]”

    i don’t get it. how do you link ‘crabfeet’ to the evolution of the tetrapod lower jaw? is this a malapropism?

  20. #20 DDeden
    August 4, 2009

    The missing basal thumb bones fused bilaterally to become the mandible, (later splitting off the hyoid & ear bones?), the associated muscle became the tongue, in some fashion, I’m unclear on precise details. Crab digits and joints match human hands, crabs lack mandibles, since they have jointed foreclaws, can’t have both. I see no evidence for ‘reverted’, thus the reason for the question to DM, though I’d like to hear from anyone interested.

  21. #21 dinogorgon
    August 5, 2009

    “[i]The missing basal thumb bones fused bilaterally to become the mandible, (later splitting off the hyoid & ear bones?)[/i]”
    -it has been shown quite conclusively by Shigetani [i]et al.[/i] (2005)with genetic evidence as to where the tetrapod lower jaw evolved from. see “” for an depth discussion and full reference.

    “[i]Crab digits and joints match human hands, crabs lack mandibles, since they have jointed foreclaws, can’t have both.[/i]”
    -this is purely coincidence. crabfeet have nothing to do with the tetrapod lower jaw, so DM may correct with his ‘reverted’ hypothesis. one way to check for evidnece would be to look at the phalangeal formula of the sphenacodont pelycosaurs. although it is uncertain if they gave rise to the Therapsida, they are certainly the best candidate.

  22. #22 DDeden
    August 6, 2009


  23. #23 altın çilek
    April 3, 2011

    Altın çilek sayesinde özellikle göbek basen gibi vücudumuzda yağın en çok depolandığı bölgelerdeki yağların yakılarak kurtulmanızda yardımcı olarak, spor yada diyet yapmanıza gerek kalmadan bünyenizdeki yağın yakımında önemli derecede rol alır.

  24. #24 tütüne son
    April 3, 2011

    It’s FREAKIN Permian “monkey”, it is! Holy moly!

    You didn’t mention the tail. It’s significant in two ways. The first is that it is LONGER than the majority of anomodonts: they trended to shorter and shorter tails. This could have been for balance purposes. Alternately, which the authors actually bring up, it may have been prehensile.