About the SA Blog Network



Exploring evolution through genes, computers and history
Thoughtomics Home

Coelacanths are not living fossils. Like the rest of us, they evolve

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

This stuffed coelacanth, described by Smith in 1939, achieved worldwide fame. Source.

It was supposed to be extinct. Yet here it lay, with fins round and fleshy, scales as hard as bone and a tail unlike any living fish. “Lass, this discovery will be on the lips of every scientist in the world”, James Smith said to Marjorie Courtenay-Latimer, curator of the East London Museum. Smith had good reasons to make such a grand claim. This was a coelacanth.

Naturalists had known about coelacanths for a long time – but only as fossils. It was Louis Agassiz who first described the group in 1839. Paleontologists had found dozens of different coelacanth species since then, but always in rocks older than 70 million years. The lack of coelacanth fossils in younger strata led them to conclude that coelacanths had gone extinct a long time ago. But the fish that now lay before Smith was no fossil. This creature had been caught only weeks ago, as bycatch by a fishing trawler off the coast of South Africa.

It was 1939, and the discovery of the first living coelacanth was on the lips of scientists around the world. The press heralded the fish as a ‘missing link‘, ‘prehistoric fish’ and ‘living fossil’. In doing so, they branded the coelacanths as a backwards fish for years to come.

The same stereotypes have haunted the coelacanth to this very day. In most popular accounts, the coelacanth is portrayed as a a forgotten survivor that has been left at the evolutionary wayside. In this modern fable, coelacanths had been trapped in a private bubble of time for millions of years until they re-emerged in the 20th century.

Axelrodichthys araripensis, an extinct coelacanth from South America. Photo by Ghedoghedo.

But the truth is that evolution leaves no fish behind. Coelacanths are as much affected by evolution as finches, ferns and flying lemurs. They have their own evolutionary history – we only need to look for it. This is what Japanese and African coelacanth researchers did not long ago when they took stock of the genetic diversity amongst coelacanths in the Indian Ocean. Through their research, they uncovered a small part of the coelacanth’s history of change.

Coelacanths have popped up in many places along the East African coast. Figure from first reference.

Coelacanths have popped up in several places in the Indian Ocean over the years, but the majority of them has been found in the Comoros archipelago. In the late eighties, Hans Fricke filmed how coelacanths inhabit rocky crevices and caves around the Comoros. At night he saw them drifting along the up- and downwelling currents, using their fins as stabilizers, to sneak up on unsuspecting fish. A short stroke with its fan-like tail, a sudden and forceful bite and its prey is gone.

Other coelacanths have been captured off the coast of Madagascar, South Africa, Mozambique and Kenya, but these fish have been dismissed as strays. Biologists reasoned that coelacanths would not be able to survive on the flat and sandy sea floors near Mozambique and South Africa. They presumed that strong ocean currents had swept the creatures away from the Comoros. These dead-end drifters were destined for death.

But there’s evidence that these stragglers represent distinct coelacanth populations. Geologists have identified several marine canyons near South Africa and Mozambique in which coelacanths could live. A dozen coelacanths have been caught near Tanzania every year since 2003. It’s unlikely that these are all strays. Indeed – when marine biologists let a remotely operated submersible descend in Tanzanian waters, they were able to capture footage of nine living coelacanths. Could this be the second home of coelacanths in the Indian Ocean?

In the paper that was published a few months ago, researchers have compare the DNA of Tanzanian and Comoran coelacanths. They found that some Tanzanian fish carry unique genetic variants. These variants were not found in any Comoran fish or anywhere else. This was especially true for coelacanths captured off northern Tanzania. The team believe their results indicate that coelacanths from northern Tanzania form a separate breeding population from the coelacanths from the South and the Comoros. These last two populations are much closer to each other genetically.

The researchers think the last common ancestor of the Tanzanian and Comaran coelacanths lived at least 200,000 years ago. For your sense of time: this was around the same time when the first modern human walked the earth. The researchers arrived at this estimate with a simple technique, known as the molecular clock: the more genetic differences exist between two lineages, the longer ago they diverged. But calibrating the clock can be tricky. Using a different calibration point, the researchers dated the split between the two populations to a few millions years ago.

Whatever the exact figure is, fact is that the Indian Ocean harbours distinct populations of coelacanths. If the Comoros Archipelago is the ancestral home of coelacanths, some fish have packed their things and settled somewhere else. Given enough time these populations might evolve into distinct species. We know this has happened in the past, for there are two species of coelacanth alive today. Aside from the West Indian Coelacanth, there exists a second species of coelacanth that was discovered at a local fish market two decades ago, near Indonesia.

Scientists have just started to collect and sequence coelacanth DNA. The amount of DNA analyzed in genetic studies (including this one) has been tiny so far. As more sequences will become available, more evidence of the continued evolution of the coelacanth will come to light.

Let’s leave the silly concept of ‘living fossils’ behind. Watch the movie above, and see the coelacanth sail the currents with subtle movements of its fins. Marvel at the mysterious headstand these creatures perform. Peer into its eyes, and see how the light is reflected back at you. These creatures are no fossils. They are very much alive.

As Smith wrote in the paper that announced the discovery of a second specimen:

“Numbers of successful modern fishes appear less well equipped for survival than the coelacanth. [..] Coelacanths can scarcely be regarded as degenerate fish. They are apparently full of vigour.”

Nikaido, M., Sasaki, T., Emerson, J., Aibara, M., Mzighani, S., Budeba, Y., Ngatunga, B., Iwata, M., Abe, Y., Li, W., & Okada, N. (2011). Genetically distinct coelacanth population off the northern Tanzanian coast Proceedings of the National Academy of Sciences, 108 (44), 18009-18013 DOI: 10.1073/pnas.1115675108
Schartl, M., Hornung, U., Hissmann, K., Schauer, J., & Fricke, H. (2005). Genetics: Relatedness among east African coelacanths Nature, 435 (7044), 901-901 DOI: 10.1038/435901a
Fricke, H., Hissmann, K., Schauer, J., Reinicke, O., Kasang, L., & Plante, R. (1991). Habitat and population size of the coelacanth Latimeria chalumnae at Grand Comoro Environmental Biology of Fishes, 32 (1-4), 287-300 DOI: 10.1007/BF00007462
Fricke, H., Reinicke, O., Hofer, H., & Nachtigall, W. (1987). Locomotion of the coelacanth Latimeria chalumnae in its natural environment Nature, 329 (6137), 331-333 DOI: 10.1038/329331a0
SMITH, J. (1939). A Living Fish of Mesozoic Type Nature, 143 (3620), 455-456 DOI: 10.1038/143455a0
Fricke, H., Hissmann, K., Froese, R., Schauer, J., Plante, R., & Fricke, S. (2011). The population biology of the living coelacanth studied over 21 years Marine Biology, 158 (7), 1511-1522 DOI: 10.1007/s00227-011-1667-x

Lucas Brouwers About the Author: Lucas Brouwers is fascinated by evolution. He writes about science on his blog and for a Dutch daily newspaper. Follow on Twitter @lucasbrouwers.

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 17 Comments

Add Comment
  1. 1. JamesDavis 7:30 am 02/7/2012

    Maybe we should consider coming up with a new term for extinct. Maybe what we think is extinct really is not and all we did is find bones of what it looked like million of years ago and to rediscover it, we need to look for similarities in the things around us. There is a spider that looks the same as it did millions of years ago, so maybe the big extinction didn’t really happen and all these animals just evolved to accommodate a new environment that had less food or water or shelter or protection.

    Link to this
  2. 2. Bill Crofut 11:00 am 02/7/2012

    Re: “Paleontologists had found dozens of different coelacanth species since then (1839), but always in rocks older than 70 million years.”

    It simply doesn’t occur to me how anyone can describe an alleged 70-million-year-old fossil and an extant biological organism, each as a coelacanth, and still claim an evolutionary history. How will additional DNA sequences will make any difference on what has been observed, and published?

    Link to this
  3. 3. Lucas 11:37 am 02/7/2012

    @JamesDavis: I don’t think we need to redefine extinction. Many ancient species of coelacanth have gone extinct, we can be quite sure of that. It’s in assuming that the whole group has gone extinct that we erred.

    @Bill Crofut: You say you don’t understand how a fossil and an extant creature can both be called coelacanths. The trick is to don’t think of ‘coelacanth’ as a species, but as a name for a group, like ‘primate’.

    Contrary to common belief, the extant coelacanth (Latimeria) has never been found as a fossil. The fossils that scientists did find are extinct coelacanth species. While all of these fish belong to the same order of fish as the living coelacanths, not a single coelacanth fossil exists which is anatomically identical to Latimera. Just like there are plenty of primate fossils which are unlike primates of today.

    Both the fossil record and DNA analyses indicate that coelacanths evolve.

    Link to this
  4. 4. Bill Crofut 1:01 pm 02/10/2012


    Re: “The trick is to don’t think of ‘coelacanth’ as a species, but as a name for a group, like ‘primate’.”

    Yet, in your own text (para. 11) you’ve used the term, “species” three times:

    “Given enough time these populations might evolve into distinct species. We know this has happened in the past, for there are two species of coelacanth alive today.” Aside from the West Indian Coelacanth, there exists a second species of coelacanth that was discovered at a local fish market two decades ago, near Indonesia.”

    According to one source, “primate” would seem to constitute more than simply a group:

    “The primates include the most familiar of the placental mammals, because they include us, Homo sapiens. Primates also include familiar animals, such as the chimpanzees, gorillas, and monkeys, as well as the somewhat less familiar lemurs, lorises, galagos, pottos, sifakas, indris, aye-ayes, and tarsiers.”

    Is there any claim of intermediates (in either direction) in the peer-reviewed literature?

    Link to this
  5. 5. Lucas Brouwers in reply to Lucas Brouwers 7:22 am 02/14/2012

    @Bill Crofut: When I write coelacanth, I mean coelacanth. When I write species, I mean species. Coelacanths are a diverse order of fish that unites many different species, which lived in both past and present.

    You’re right that ‘primate’ means more than ‘simply a group’. It is a group of animals with a single common ancestor.

    I cannot make myself clearer than this. I’m all open for discussion, but if this turns into a quibble about semantics, I am afraid it ends here.

    Link to this
  6. 6. Bill Crofut 10:26 am 02/18/2012


    Re: “I’m all open for discussion, but if this turns into a quibble about semantics, I am afraid it ends here.”

    Fair enough. Allow me to re-ask the question that closed comment 4:

    Is there any claim of intermediates (in either direction) in the peer-reviewed literature?

    Link to this
  7. 7. Torbjörn Larsson, OM 12:04 pm 02/19/2012

    Interesting, and the stream figure hints at how the split happened.

    @ Bill Crofut:

    Are you asking whether evolution happens?

    Yes, there is the overall claim, more than a century old and supported in countless peer-reviewed literature. When nested biological hierarchies are observed they are related by intermediates (whether observed or not), because that is the only way to preserve the nesting property.*

    Or in other words, your cousins and you are related by way of your parents, whether you were early adopted or not. You can not not have parents, even if you believe yourself to be creationist.
    * I am not really sure you need to know about heredity to get that result.

    Link to this
  8. 8. Bill Crofut 10:38 am 02/20/2012

    Torbjörn Larsson, OM,

    Re: “Are you asking whether evolution happens?”

    No. As a militant young-Earth Biblical creationist, the only meaningful “evolution” is transformism. If the transformation of one “species” into another can be demonstrated, that will be cause for me to reconsider my position.

    Re: “…intermediates (whether observed or not)”

    Please provide an example of an unobserved intermediate.

    Re: “You can not not have parents…”

    You’ll get no argument from me. However, that leaves unexplained the origin of the first life.

    Link to this
  9. 9. Lucas Brouwers in reply to Lucas Brouwers 10:45 am 02/26/2012

    Your understanding of evolution is flowed. Evolution does not equal transformation of one living species into the next. No one claims that chimpanzees transformed into humans, or African coelacanths into Indian coelacanths.

    Evolution is simply change in frequency of heritable properties (genes). When biological populations become isolated, genetic difference accumulate until the two populations are no longer genetically compatible . At this point biologists will regard the two populations as two species. You want ‘intermediate species’? Look no further than the Tanzanian coelacanths, that have become isolated from the Comoran population. Genetic isolation is the prelude to speciation.

    Link to this
  10. 10. Bill Crofut 7:33 pm 02/26/2012


    If my understanding of evolution as transformism is incorrect, how do you account for the claim that reptiles (dinosaurs) evolved into birds?

    Link to this
  11. 11. @stegorawrrr 9:59 am 03/23/2012

    Adding to the discussion, dinosaurs evolved to birds as gradually as everything else are, it’s just mutations and perfecting from natural selection creating differences in a dinosaur, it’s merely that there are less evidence preserved for dinosaur-bird than older coelacanths- Latimeria, so most laymen assume that it just species morphing instantly, rather than a gradual transition.

    Link to this
  12. 12. Bill Crofut 5:48 pm 03/23/2012


    Please provide an estimate of the number of morphological changes required for the transition from a dinosaur (reptile) into a bird.

    Please provide an example of a mutation that has made a contribution to the alleged transition.

    It’s my understanding that a mutation is neutral at best, lethal at worst.

    Are you aware of any intermediate status for a mutation?

    Link to this
  13. 13. Gerdien 4:24 am 03/26/2012

    The present day species Latimeria chalumnae and Latimeria medanoensis are not identical in their skeleton to any fossil member of the order Coelacanthiformes. They are not ‘living fossils’ in any sense – we might as well call sharks ‘living fossils’! Sharks are as old as coelacanths, only there are many species of them, and they were not unexpectedly found.

    Professional literature on coelacanth evolution, far from an exhaustive search:

    G. Clement, 2005. A new coelacanth (Actinistia, Sarcopteryggi) from the Jurassic of France, and the question of the closest relative fossil to Latimeria. Journal of Vertebrate Paleontology 25(3):481–491, September 2005

    P.L. Forey, 1991. Latimeria chalumnae and its pedigree. Environmental Biology of Fishes 32: 75-91, 1991.
    Latimeria is the product of a long coelacanth lineage, usually viewed as having changed very little. In this paper a classification of better known coelacanth genera is proposed based on a cladistic computer analysis of 56 morphological characters. Biometrical data are then matched with the classification to explore the possibility of identifying subtle change. It is concluded that throughout coelacanth history there have been changes in the structure of the vertebral column involving an overall increase in the number of vertebral elements, and a consequent crowding of these elements within the abdominal region. These changes may be associated with increasing lobation of the second dorsal and anal fins. In the skull, parameters involving the intracranial joint have also changed in such a way that the anterior part of the skull has lengthened in relation to the posterior part and this may be associated with an increase in length of the basicranial muscle

    Link to this
  14. 14. Bill Crofut 10:47 am 03/26/2012


    Re: “The present day species Latimeria chalumnae and Latimeria medanoensis are not identical in their skeleton to any fossil member of the order Coelacanthiformes.”

    How non-identical are they? The abstract from the P. L. Forey paper would seem to indicate close proximity. In my research experience (spanning 30+ years) it’s a rare occurrence to find the term “living fossil” in the scientific literature where it’s not enclosed in quotation marks; that seems appropriate to me.

    Re: “we might as well call sharks ‘living fossils’!”

    “Sharks and their kin are sometimes described as “living fossils,” and they are indeed part of an ancient clade of vertebrates.”

    Link to this
  15. 15. Lucas Brouwers in reply to Lucas Brouwers 3:14 pm 03/26/2012

    @Bill: You come here and ask for evidence. People deliver. You ask for more.

    This is pointless. No matter how much evidence lies before you, you will not be swayed. I therefore will delete future comments from your account; this should be a place of open discussion and debate, not a stage for dogma and denial.

    Nick Lane voiced my objections better than I ever could:
    “To doubt that life evolved, is to doubt the convergence of evidence from molecules to men, from bacteria to planetary systems. [..] It is to doubt the veracity of experiment and observation, to doubt the testing in reality. It is, in the end, to doubt reality.”

    Link to this
  16. 16. bgrnathan 1:02 pm 04/19/2013


    The recent news is that scientists have discovered that some snippets of DNA from a fish, they once thought was extinct, can cause mice to grow limbs. This is wrongly being hailed as evidence that fish had developed legs. Isn’t it interesting that the fish they got the DNA from didn’t have legs!

    These snippets of DNA from the fish seem to be “triggering” mechanisms. They can only trigger (“turn on”) formation of limbs if the genes for limbs first exist, and since genes for limbs exist in mice then these triggering mechanisms, even if from a fish, will work. None of this means fish evolved legs.

    Imagine an evolving fish having part fins and part feet, with the fins evolving into feet. Where’s the survival advantage? It can’t use either fins or feet efficiently. There are no fossils of such fish. These fish exist only on automobile bumper stickers!

    A partially evolved species, with partially evolved tissues and organs, waiting (supposedly) millions of years to evolve would be unfit for survival.

    Genetic and biological similarities between species are no proof of common ancestry. Such similarities are better and more logically explained due to a common Genetic Engineer or Designer (yes, God) who designed similar functions for similar purposes in various species. Genetic information, like other forms of information, cannot arise by chance, so it’s more rational to believe that DNA or genetic similarities between species are due to intelligent design.

    The genes already exist for micro-evolution (variations within a biological kind such as varieties of dogs, cats, horses, cows, etc.), but not for macro-evolution (variations across biological kinds such as from sea sponge to human). Only variations for already existing genes and traits are possible.


    Babu G. Ranganathan
    B.A. Bible/Biology


    Link to this
  17. 17. esguerra25 11:57 pm 10/4/2014

    sorry to interrupt, Mr. Bill Crofut, but the thing you said-” However, that leaves unexplained the origin of the first life.” bothers me a lot, because it is clearly stated in Genesis 1:1 that “In the beginning God created the heavens and the earth”. therefore we already know the origin of life, which is God.

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American

Email this Article