Oldest evidence for animals found? New sponge-like fossil is 890 million years old, several hundred million years older than next oldest animal

July 29, 2021 • 9:15 am

First, we have to know what biologists mean by “animals”. In brief, they are multicellular organisms comprising eukaryotic cells (“true cells” with a nucleus and nuclear membrane, as well as organelles like mitochondria). Or, to be more specific, I’ll give the Wikipedia definition:

Animals (also called Metazoa) are multicellular, eukaryotic organisms in the biological kingdom Animalia. With few exceptions, animals consume organic material, breathe oxygen, are able to move, can reproduce sexually, and go through an ontogenetic stage in which their body consists of a hollow sphere of cells, the blastula, during embryonic development.

Long before animals existed, living organisms existed, but these were cyanobacteria (“blue green algae”) and other microbes, not regarded as animals. The first cyanobacteria date back about 3.5 billion years, only a billion years after the Earth formed. The cyanobacteria are identified in fossil stromatoliteslayered reef-like structures formed by the accretion of bacteria. Stromatolites are still forming in some places on Earth, like Shark Bay, Australia.

But when did the first metazoan, or “animal” appear? For that you can use either fossil or molecular evidence.

The earliest fossil scientists regard as an animal is Dickinsonia from the Ediacara fauna, dated about 540 million years ago.  Scientists think it’s an animal because its lipid biomarkers, which you can extract from fossils and the sediments above and below them, include cholesteroids, compounds found exclusively in animals. Dickinsonia is known only from imprints, like the one below, and its affinities are a mystery.


Molecular data, from which you can construct a phylogenetic tree of living animal groups and then extrapolate backwards, have shown that animals probably originated between 650 and 850 million years ago, but we have no animal fossils from that period. Those trees also show that perhaps the earliest animal was similar to sponges, for sponges seem to be the most “basal” animals—those that branched off the animal tree before other groups. This makes sponges the “sister group” of all other animals.

Now a new paper in Nature by Elizabeth C. Turner of Laurentian University in Canada has pushed the oldest animal fossil back a long way: several hundred million years—to 890 million years ago! And, in fact, the fossil shows features of early sponges, verifying the molecular conclusions.

Now not all paleobiologists agree that what Turner found is an animal—some say the structures observed may have a microbial origin—but Turner herself is pretty confident, as are some other paleontologists. So let’s take this conclusion as “likely, but not certain”. Surely further work will either strengthen or weaken Turner’s evidence.

You can access Turner’s paper by clicking on the screenshot below, or downloading the pdf here. The reference is at the bottom of this post.

Investigating the Little Dal Reef Formation in Northwestern Canada, itself a kind of stromatolite, Turner collected rocks between 1992 and 2018, and, in thin sections of those rocks, observed “vermiform” (worm-shaped) microstructures filled with calcite “spar”, or calcium carbonate crystals. These tube-like structures join and divide in a branching network, just like the tubules of modern sponges, some of which have a calcite skeleton. (The tubules of modern sponges allow them to circulate water through their bodies, getting food and oxygen.) These wormlike structures are surrounded in the fossils by a calcite “groundmass”, which may be the external body of the sponge.

Here’s what Turner says about these interconnecting tubules and why she regards them as early sponges:

The shape, size, branching style and polygonal meshworks of the Little Dal vermiform tubules closely resemble both spongin fibre networks of modern keratosan sponges (Fig. 2a–c) and vermiform microstructure either demonstrated or interpreted to be sponge-derived in diverse Phanerozoic microbial, reefal and non-reefal carbonate rocks. The compositional and textural homogeneity of the microspar groundmass supports an origin through permineralization of a pre-existing biological substance, rather than incremental accumulation of detrital sediment or microbial carbonate that passively incorporated complexly anastomosing tubular microfossils. Variable preservation and association with geopetal peloid accumulations are familiar aspects of Phanerozoic sponge taphonomy In previous work, detailed comparison of the three-dimensional characteristics of vermiform microstructure with branching cylindrical organism types yielded no convincing alternative to the sponge interpretation

Here are subfigures (a)-(b) of her Figure 2 showing the fossil network compared to that of a modern sponge (c), with the captions below (click photo to enlarge).

(From Fig. 2 of the paper): a, Well-preserved vermiform microstructure exhibits a polygonal meshwork of anastomosing, slightly curved, approximately 30-μm-diameter tubules embedded in calcite microspar (KEC25). Scale bar, 500 μm. b, Enlarged rectangle from a, showing branching tubules forming three-dimensional polygons intersected at various angles by the thin section; clear calcite crystals, about 10–20 μm in width, fill tubules in groundmass of more finely crystalline calcite (dark grey). Scale bar, 50 μm. c, Three-dimensional fragment of spongin skeleton from a modern keratosan sponge, illustrating its branching and anastomosing network of fibres (incident light). Scale bars, 100 μm (main panel), 20 μm (inset).

There are other pictures as well, but the first two are the heart of the matter. You may not think they look like much, but they do show the interconnecting, ramifying tubules with the light-colored calcite crystals typical of some groups of sponges. The area where these putative fossils are found is 890 million years old.  And these fossils are older than the next oldest and indisputable sponge fossils by 350 million years!

Turner hypothesizes that these early organisms couldn’t compete with the reef-building cyanobacteria, but were able to find “oxygen oases” to use the oxygen produced by the cyanobacteria. The association of these putative sponges with oxygen-producing bacteria may be one piece of evidence that these are indeed metazoans, which of course require oxygen.

As I said, some paleobiologists disagree about whether these are animals. You can hear a ten-minute Nature-sponsored discussion with Turner, some supporters, and some doubters here. I highly recommend that you listen to this short but lucid discussion.

One other point: these organisms must have survived at least one of the periods of extensive glaciation and freezing known as “Snowball Earth“, when the entire planet was either completely frozen or almost covered with ice except for some open water. (The most extensive was between 700 and 600 million years ago.)  In the linked article, author Laura Poppick says this about that period:

What did life on Earth look like at the time, and how did it change as a consequence of these events?

There were certainly bacteria and there were also algae and unicellular primitive animals, or protists.

There is also evidence that the first multicellular animals originated at this time, probably something like sponges.

Well, according to Turner, the first multicellular animals, probably something like sponges, originated nearly 200 million years earlier than this.

Stay tuned to see how the dispute about the nature of these fossils progresses. Are they animals or simply remnants of bacterial activity? As Turner says in the interview, “We are quite confident” that these are spongelike animals. “It’s almost,” she adds, “a no-brainer.”

And here’s Turner in the field:

(From source): Elizabeth C. Turner, geology professor at Laurentian University, conducting geological fieldwork on northern Baffin Island in 2012. (Supplied photo/Laurentian University)


Turner, E.C. 2021. Possible poriferan body fossils in early Neoproterozoic microbial reefs. Nature (2021). https://doi.org/10.1038/s41586-021-03773-z

30 thoughts on “Oldest evidence for animals found? New sponge-like fossil is 890 million years old, several hundred million years older than next oldest animal

  1. Appreciate this well-written analysis of the claim. Turner probably should not have tagged on the “almost a no-brainer” thing.

    In my experience discussing evolution with people having thin knowledge, they first have to get acclimated to the “earth is 4.5 billion years old.” Unless they are indoctrinated to something “significantly shorter” by a belief project, they gradually accept this fact.

    However, is very surprising to them that “animals” have only been around for 1/9 of that time — it seems scary-short to them. Only 532 or so million years back. It hits home eventually that the Cambrian Explosion was indeed incredible.

    The news of this claim that animals (sponge) existed ‘a few’ additional millions might help cement understanding of the deep-time scales involved. It might not make an impression on the ‘shorters.’

    1. I interpreted “no-brainer” as just subtle irony by Turner. All sponges are no-brainers because they lack neurons.

      The most satisfying aspect of this discovery is that, if true, it would put a stake in the heart of the “ctenophores first” hypothesis.

    2. Welll …. disputed evidence for considerably earlier metazoans does exist. While the existence of the fossils is rarely questioned, the interpretation (of these fossils as metazoans, as animal fossils, or even as “animals with a worm-like life habit” (“vermiformes”, if Linneaus hadn’t bagged the term) is rather more a ground for dispute. They look nothing like the current “simplest” animals.
      This discovery takes the fossil record of “sponges” – our evidence for the previous existence of “sponges” – back considerably further than previously. But Horodyskia (and some other less established morphological taxa) still exists, even if there is no consensus about what type of life it represents. My Latin isn’t canine enough to misquote Galileo with an “Eppur si existe“, but it still exists, and possibly existed as far back as 1.5 Gyr (one third of the age of the Earth, to comfort your recovering creationist friends)
      Wake me up when they find metazoan fossils in the lower part of the Banded Iron Formations. Say,2.8 Gyr ago.

    3. The Cambrian explosion seems less and less “explosive” all the time as more fossils accumulate. And coalescent times for the total genetic diversity of living animals (including sponges) have long been known to go back much further than 532 million years ago. We always need to remember that, for any lineage, fossils tell us only the minimum time a lineage has been around.
      Well-calibrated genetic coalescent times are usually much more informative. But all of this well-established information may have little effect on those who might think the Earth is much less than 4.5 billion years old.

      1. @Frank

        “And coalescent times for the total genetic diversity of living animals (including sponges) have long been known to go back much further than 532 million years ago.”

        Please clarify. Are you saying there is evidence of animals larger and more complex than microbes, single-cell, sponges, etc., …. how far back? Can you give example.

  2. “Well, according to Turner, the first multicellular animals, probably something like sponges, originated nearly 200 years earlier than this.”

    Should this be “200 million years?”

  3. Good and interesting science, regardless of the outcome. Have not yet had a chance to watch the discussion. Hopefully tonight. Am a bit surprised there’s no geological or inorganic processes that produce similar-looking meshworks of microtubules in calcite or other similar minerals. Not that I know of any in specific, I don’t. It just that that’s not the sort of thing I would rule out as improbable for geological processes to do.

      1. Maybe this helps:

        “Turner was able to rule out other interpretations of the microstructures owing to the configuration of the fossil material.

        “It consists of little tubes that branch divergently and then rejoin to form a complex, three-dimensional meshwork,” she said in an email. “Of the branching organisms that could be considered as alternative interpretations, none of them have that kind of three-dimensional meshwork—not true algae, not bacteria, and not fungi.””

        [ https://gizmodo.com/animals-emerged-350-million-years-earlier-than-previous-1847377402 ]

  4. We should of course learn an attitude of love and respect for our ancient sponge-like ancestors. Anything less would cause harm and offense in the contemporary spongex community.

    1. Off actual topic, but:

      So, the first Canadian (if we grant retroactive citizenship to all resident animals) was maybe here before the first anywhere else, a billion years ago. We can start a sports competition with Australia, etc. on that. Looks like this is somewhere near the Gayna River, NWT, draining to the MacKenzie, a bit more than halfway up the border of NWT with Yukon, and near it.

      I’m happy to see stuff from ULaurentian from which I know people, visited for talks, and have had many relatives near there in Sudbury, where I did my final two high school years. It is very difficult for northern (relative to where some reasonable number of people live, not by where this discovery is) universities to recruit good students, even from the north.

      Also happy to see good stuff happening from my native northern mining areas—my dad was a mining engineer, degree from QueensU, where she did her (not unconnected) ‘geology/geobiology’ Ph.D. about 70 years later.

        1. Yes, I’ve been very aware of this bad situation. As you may know, I’m an academic mathematician, 11 years retired from teaching, but still minimally active in research.

          My closer friends in math who were there are also retired now. They included an undergraduate friend way back, a prof who moved there from my place (Waterloo), and a fellow who did research closer to mine and also competed in age group aerobic sports, not against me, being 10 years younger. In fact he provided accommodation for me when there were ski races up there, lives elsewhere now. He won the 50+ age group in the Detroit Marathon (running) a decade or two back, way more natural talent than I’ve ever had!

          But I’m too old for it to affect me directly.

          It’s been very bad for that present day math department among many others. I don’t know whether the idiot administrators who mostly caused this were themselves academics—I doubt it.

          At this point ULaurentian may be lucky to keep her from moving to greener pastures, though I have no idea about specifics. If this research holds up, she will become very prominent in the field, maybe already is. She already won some big award from them, only the second to do so, a few years ago.

  5. “Turner hypothesizes that these early organisms couldn’t compete with the reef-building cyanobacteria, but were able to find “oxygen oases” to use the oxygen produced by the cyanobacteria.”

    This is a neat feature of the fossil data. Many living sponge species are stuffed with cyanobacteria (and eubacteria) living in the sponge tissues. The sponges often benefit from photosynthesis by the cyanobacteria, and from metabolites that the microbes produce (which happen to deter grazers like fish). Some microbiologists think of sponges as a sort of playground for bacterial chemical warfare.

    These very old reefs might have been the places where that symbiosis originally evolved. It’s cool that in the original association, it was the sponge trying to eke out a living among the microbes.

  6. This looks to be an amazing find. For the untrained eye it sure looks like similar much younger fossils, in the paper and in the video here [@2:30]: https://www.youtube.com/watch?v=BXuyOPSDAYU&ab_channel=naturevideo . The absence of spicules used for defense and the classification in the most diverse sponge class of demosponges speak for keratose sponges as an early stem group. And it would reconcile genetic molecular clocks, trace molecule finds and the macroscopic fossil record.

  7. This article touch on the Cryogenian survival and a CNN article puts a broader context on that [ https://edition.cnn.com/2021/07/28/world/early-animal-sponge-fossil-scn/index.html ]. The early evolution before the Neoproterozoic oxygenation event implies tolerance for low oxygen levels that would have been present in local “oxygen oasis” environment “in “shadowy nooks and crannies” on and below large reefs made from water-dwelling bacteria that were photosynthesising”. It also links to the recent find of below shelf ice sponges [ https://edition.cnn.com/2021/02/15/world/antarctic-ice-shelf-sponge-discovery-scn/index.html ]. “”It’s about 160 kilometers further under the ice shelf than we had ever seen a sponge before.” … “Somehow, some really specialized members of the filter-feeding community can survive,” he said.”

  8. Sponges should be considered a proxy for other animals that ushered in the Cambrian explosion -apparently with “a long fuse” – since it looks like a gene loss clade. Especially it may have lost many of the body plan patterning genes (Hox, ParaHox, NK) that most other animals rely on:

    Figure 4“Consequently, our discovery of ghost Hox and ParaHox loci in a sponge, and a ParaHox locus containing Trox-2 alongside a ghost Hox locus in a placozoan, implies that the last common ancestor of animals possessed distinct Hox and ParaHox loci (). This, in turn, implies loss of these homeobox genes during the evolution of some basal animal lineages, which, in terms of these developmental control genes, have been simplified relative to the last common ancestor of animals.”

    [ https://www.sciencedirect.com/science/article/pii/S0960982212009888 ; see especially figure 4.]

  9. Good to see that the faculty at Laurentian in Sudbury, Ontario is on the ball, unlike the administration who have so mismanaged the university that it has had to file for bankruptcy. So far no heads have rolled.

    I wonder if Elizabeth Turner still has a job.

      1. I imagine the best will leave which makes this community poorer. All because of power hungry administrators

    1. Universities across the board do seem to be determined to root out good departments. Sheffield has done that to its archaeology unit. Terrible for younger academics though anyway…

      1. Laurentian U cancelled its Midwifery program. The medical school was saved by the provincial government.

Leave a Reply