Uncommon Descent Serving The Intelligent Design Community

Have 99% of All Species Gone Extinct?


Dear readers,

It has been far too long since my last post, occasioned by the fact that I have entirely too many irons in the fire.

I hope you will forgive this brief “drive-by” post, with a request for some help and information.

One of the common refrains that comes up regarding the fossil record, or regarding claims about biodiversity and the evolution of species more generally, is that the vast majority of species that have ever lived on the Earth have gone extinct. This is often phrased as “99% of species that have ever lived have gone extinct” or similar wording. (Occasionally someone will temper the number to 98% or 95% or some other nearby figure, but 99% seems to be the most common claim.)

I am trying to track down a credible source for this incredible claim.

With the help of, yes, Wikipedia, I’ve managed to make a little bit of progress.

  1. Apparently, in 1991, University of Chicago paleontologist, David Raup, estimated that there might have been anywhere from 5 to 50 billion species that had existed during the history of the Earth.
  2. Given then-current estimates of existing species, Michael McKinney calculated that “well over 99% of earth’s species” had gone extinct. As near as I can tell, this was presented by McKinney at a symposium and published in 1997 along with other symposium papers and presentations in a book titled, “The Biology of Rarity: Causes and consequences of rare-common differences”.

No doubt this is not the only source for this claim, as many other biologists would have quite readily drawn similar conclusions based on Raup’s estimate.

I would be most grateful for any additional, or more solid, sources for the idea that “99% of all species that have ever lived have gone extinct.”

Incidentally, it is worth noting that the actual number of known (not estimated or projected or inferred) species is quite different. Compared with some 1-2M known species currently living, there are about 250K identified fossil species. Obviously these numbers are also subject to some margin of error, but at least they deal with known, identified organisms, rather than projections and estimates. Even allowing for lots of gray area due to the ever-elusive definition of “species” and the fact that observable fossils will obviously tend toward larger creatures (e.g., one suspects it will be tough to get a decent count of bacteria in the fossil record!), it will still not be lost on the reader that these numbers flip the common claim on its head, with the fossil record count making up but a minor percentage of currently-living species, rather than vastly dwarfing the latter.

This leads to a couple of important follow-up questions:

  1. Is the fossil record truly so poor that many billions of species have come and gone without leaving a discernable trace? Not thousands, not millions, but billions? Were Gould and Eldridge wrong to suggest that the fossil record, albeit imperfect, is generally reliable and tells a largely accurate story of the history of life on Earth, at least as it relates to the larger animals?
  2. Or is the “99% extinct” claim serving some other role — perhaps something of a modern incarnation of Darwin’s proposal that the number of long-dead intermediates must be “innumerable” and that the fossil record is not to be trusted, because it reflects but a miniscule part of the Earth’s actual biological history?

I would like to know one way or another.

Any help you can give in either (a) tracking down good sources for the 99% extinct claim, or (b) spelling out why the claim should be accepted, despite the actual physical data, would be most helpful.

Thanks, BA77, for the additional citations. Examples of predictable, repeatable, rapid adaptation are becoming more and more common in the literature. Although this adaptation is often trumpeted as evidence for "evolution" happening "right under our nose," as in Reznick's guppy population,* adjectives such as rapid, repeatable, and predictable have no rational place in naturalistic evolutionary scenarios. These are instead indicia of purposeful (whether deterministic or stochastic) outcomes, resulting from the engineered design of the organism and variable characteristics operating within a given set of operating parameters. Thoroughly non-Darwinian. Thoroughly non-naturalistic. There is a massive conceptual chasm separating evolutionary theory from the observed reality of adaptation. If we are willing to look at the evidence objectively, without the blinders of materialistic philosophy, the take home lesson from essentially all the classic examples of adaptation -- peppered moths, finch beaks, bacteria and antibiotics, cichlid fishes, insects and insecticides, Lenski's E. coli experiments -- the take home lesson from all of these is this: Organisms have the capacity to temporarily adapt to environmental challenges, while ultimately resisting fundamental change. Even in this area of so-called "micro-evolution," where even skeptics of evolutionary theory often give evolution a pass, Darwin had it fundamentally backwards. The importance of this can hardly be overstated, as it goes to the very heart of evolutionary theory. * BTW, thanks for the kind inclusion in your list. :) Eric Anderson
Yes EA, it is dated, but it does have very instructive moments. That is why I referenced it even though it is dated. In fact, I think I wrote out something he said on cichlids,,, Yeah, here it is,,,
"For all the diversity of species, I found the cichlids to be an unmistakably natural group, a created kind. The more I worked with these fish the clearer my recognition of “cichlidness” became and the more distinct they seemed from all the “similar” fishes I studied. Conversations at conferences and literature searches confirmed that this was the common experience of experts in every area of systematic biology. Distinct kinds really are there and the experts know it to be so. – On a wider canvas, fossils provided no comfort to evolutionists. All fish, living and fossil, belong to distinct kinds; “links” are decidedly missing." Dr. Arthur Jones - did his Ph.D. thesis in biology on cichlids - Fish, Fossils and Evolution - Cichlids at 29:00 minute mark (many examples of repeated morphology in cichlids) - video http://edinburghcreationgroup.org/video/14
All subsequent studies that I have run across conform to his overall thesis:
African cichlid fish: a model system in adaptive radiation research: "The African cichlid fish radiations are the most diverse extant animal radiations and provide a unique system to test predictions of speciation and adaptive radiation theory(of evolution).----(surprising implication of the study?)---- the propensity to radiate was significantly higher in lineages whose precursors emerged from more ancient adaptive radiations than in other lineages" http://www.pubmedcentral.nih.gov/articlerender.fcgi?tool=pubmed&pubmedid=16846905 Cichlid speciation attributed to “plasticity” - December 7, 2016 Excerpt: There is increasing evidence that phenotypic plasticity can promote population divergence by facilitating phenotypic diversification and, eventually, genetic divergence. When a ‘plastic’ population colonizes a new habitat, it has the possibility to occupy multiple niches by expressing several distinct phenotypes. These initially reflect the population’s plastic range but may later become genetically fixed by selection via the process of ‘genetic assimilation’ (GA). Through this process multiple specialized sister lineages can arise that share a common plastic ancestor – the ‘flexible stem’,,, https://uncommondesc.wpengine.com/speciation/cichlid-speciation-attributed-to-plasticity-now/ Multiple Genes Permit Closely Related Fish Species To Mix And Match Their Color Vision - Oct. 2005 Excerpt: In the new work, the researchers performed physiological and molecular genetic analyses of color vision in cichlid fish from Lake Malawi and demonstrated that differences in color vision between closely related species arise from individual species’ using different subsets of distinct visual pigments. http://www.sciencedaily.com/releases/2005/10/051011072648.htm Is the outcome of evolution predictable? - Oct 28, 2014 Excerpt: There are only very few circumstances in which one can investigate the repeatability of evolution, because spatially independent environments that are populated by the same species are extremely rare in nature. "The young and completely isolated crater lakes along the Central American Volcanic Arc in Nicaragua provide an ideal setting to study parallel evolution. Several crater lakes house populations of Midas cichlid fish that have developed independently from the ancestral population in the nearby great lakes of Nicaragua. This setting is like a natural experiment", explains Axel Meyer. In two of these crater lakes, Apoyo and Xiloá, new types of Midas cichlids evolved, independently from each other, in less than 10,000 years. These new species show identical morphological adaptations that are not found in the ancestral population: from the shallow, murky water to the new habitat of the deep, clear water of the crater lakes. "In each of the two crater lakes new species of the Midas cichlid evolved with an elongated body – a phenotype that does not exist in ancestral lakes from which the colonisers of crater lakes came from", explains Meyer,,, "Our study shows that complex parallel phenotypes in similar environments can evolve very rapidly, repeatedly and yet via different evolutionary routes. This is a microevolutionary example of rewinding Gould's tape and resulting in the evolution of two very similar species, albeit by non-parallel evolutionary routes", sums up Axel Meyer. http://phys.org/news/2014-10-outcome-evolution.html Real Time Evolution “Happening Under Our Nose” - October 12, 2015 Excerpt: Our research shows that these fish adapted to their new habitats in less than one year, or three to four generations, which is even faster than we previously thought.,,, Lee Spetner commenting on Reznick's earlier work: Reznick and his team took 200 guppies from the Aripo [river in Trinidad] and put them in a tributary of the river that is home to the killfish but has no cichlids and had no guppies. Changes soon appeared in the newly introduced guppies. The fish population soon changed to what would normally be found in the presence of the killfish, and Reznick found the changes to be heritable. The full change in the guppy population was observed as soon as the first samples were drawn, which was after only two years. One trait studied, the age of males at maturity, achieved its terminal value in only four years. The evolutionary rate calculated from this observation is some ten million times the rate of evolution induced from observations of the fossil record [Reznick et al. 1997]. Reznick interpreted these changes as the result of natural selection acting on variation already in the population. Could natural selection have acted so fast as to change the entire population in only two years? https://uncommondesc.wpengine.com/evolution/real-time-evolution-happening-under-our-nose/#comment-583041
BA77 @7: BTW, all. I watched the majority of the lecture BA77 linked to. It is a little old, but has some very instructive moments. One that jumped out at me was the lecturer's observation, as a scientist who spent a significant portion of his career studying cichlids, that a great deal of variety was obtained among cichlid species just by varying a couple of parameters (specifically, color and location of markings). These variations have often been viewed through the lens of Darwinian evolution as evidence for the "plasticity" of the organism and evidence for the evolutionary process of change over time. And yet, the cichlids remain cichlids, and are easily recognizable as such. With an engineering eye, this observation of cichlid variation is more consistent with pre-determined operating parameters and adjustable variables, rather than the essentially random, successive-slight-changes-turning-into-another-organism process Darwin proposed. Eric Anderson
ET @23: Yes. A strange result indeed. Even if the fossil record is woefully incomplete, it is not as though we don't have any samples. We have hundreds of thousands of data points, and they overwhelmingly seem to be not confirming the transitional evolutionary process. That is quite a sampling error! It is one thing to posit a made-up story about a rapid evolutionary process (a la Gould) for which there is essentially no empirical evidence and which makes the challenge with waiting times and so on even more problematic. But one might start to suspect that there is a hole in the theory when the data stubbornly refuse to confirm the theory. It starts to look pretty self-serving to keep blaming the data for 150 years, rather than looking in the mirror and asking the simple question, "I wonder if my theory is wrong?" Eric Anderson
There is something else that is interesting with respect to the existing fossil record- the vast majority of fossils, over 95%, is of marine invertebrates. And that is to be expected from what we know of the fossilization process. However, in that vast majority we do NOT see any evidence for universal common descent. And yet you will never be taught that in any class on evolution. ET
We seem to be dealing with two issues: 1. Is the fossil record woefully incomplete? 2. Even if woefully incomplete, why do the fossils that have been preserved not demonstrate the transitional process that should have happened under evolutionary theory? IOW, regarding #2, even if we only have a small % of total fossils, the thousands of fossils that are preserved should still confirm the overall story. Why then do they essentially invariably represent only the tips of the branches? Even if we only have a small % of total fossils, we should still see numerous clear transitionals, including many relatively-complete series. After all, the transitionals are supposed to be much more numerous than the final resulting forms. "Innumerable," Darwin said of the transitionals. Isn't it suspicious that the out of the allegedly-massive dataset, the thousands of data points that have been preserved just happen to be the ones that don't confirm the transitional evolutionary story? That is a very strange result indeed, and might suggest to the objective observer that something else is going on. Punctuate equilibrium doesn't help much, unless one makes the utterly questionable assumption that all the action just happens to occur when and where we can't observe it. A very interesting theory indeed -- one of the few theories in science that is based almost completely on the absence of evidence. "But we simply don't see evidence for the rapid evolutionary transitions that you claim must have occurred over and over in the history of life," the skeptic notes. "Exactly!" responds Gould, "That's just what my theory predicts -- that you won't find evidence for it." One might be forgiven for observing that this is just a little too convenient. Eric Anderson
Hi Eric, I don't know if he spells out how he derived that figure. It's a quote I've seen on several occasions, but had never tracked down the source, but it looks like it's from a 2001 book "What Evolution Is". Not a book I've read though. goodusername
Goodusername @19: Thanks for the detailed comment. Do you happen to recall what Mayr was basing his 99.99% figure on? Eric Anderson
 Is the fossil record truly so poor that many billions of species have come and gone without leaving a discernable trace? Not thousands, not millions, but billions? Were Gould and Eldridge wrong to suggest that the fossil record, albeit imperfect, is generally reliable and tells a largely accurate story of the history of life on Earth, at least as it relates to the larger animals?
  I have little doubt that Gould would agree that 99.9+% of all species have gone extinct.  With Punk Eek, they proposed that the rarity of intermediates isn’t only because of how rare it is for a creature to become of fossil, but also because of the “mode” of evolution – the intermediates were formed in small peripheral populations, i.e., peripatric speciation, and thus there were fewer chances of leaving fossils. And so I don’t think that Gould was really challenging the prevailing view of how imperfect the fossil record is, but instead was arguing that the many “gaps” in the fossil record were more than a mere “inconvenience” - that the pattern of spottiness of the fossil record was itself informative, and specifically was evidence for Mayr's theory of peripatric speciation. BTW, Mayr himself claims that well over 99.99% of all species have gone extinct. This might be a good illustration of how spotty the fossil record is… It’s estimated that the Permian - Triassic extinction event wiped out about 95% of all species (and looking up the numbers real quick, over 80% of genera, 57% of families, etc) - yet despite the vast majority of species being wiped out, it’s believed that not a single phylum went extinct. This makes sense because the larger, more diverse a taxonomic group, the harder it is to completely wipe out. For the same reason that it’s so incredibly difficult to wipe out an entire phylum, it would be equally difficult for a phylum to leave no trace in the fossil record unless the fossil record was really bad - and, yet, that’s precisely the case for most animal phyla. goodusername
One way to get a handle on extinction is to look at how long (typically or statistically) fossil "species" last in the fossil record. If the average duration of a bona-fide fossil species (say a certain shellfish in a continuous sedimentary deposit) lasted 5 million years, and complex life has been around about 500 million years, then the 5 represents 1% of the 500, lending some support to the 99% number. One could (in principle) construct a summary "tree of life" to specify how many "species" existed at any time during, say, those past 500 million years, taking into account mass extinctions, among other "known" facts. Then one could take the statistical distribution of the "known" fossil species durations, and apply it mathematically against the "tree", to get an estimate of how many species have existed during the 500 million years. One could further attempt to project the 500 back to 1 billion or so, to ensure one accounts for all metazoans. Of course, any such analysis would run afoul of the previously mentioned uncertainties, not to mention all the assumptions one would have to make, but it would at least provide a method or process to arrive at some defensible number. Both the process and the fossil data feeding it could then be subject to peer review and revision, to get a better idea about the "actual" percentage. Perhaps that is what the original authors of the 99% number did? If so, it could doubtless use an update. If not, some paleontologist with a modicum of mathematics could give it a go. Fasteddious
Thanks, Latemarch. It is instructive to note that they say a subphylum "had evolved into a large number of distinct orders". I'm willing to accept that the members of a subphylum had diversified into distinct orders, but this is going exactly backwards from the way Darwin proposed. The Virtual Fossil Museum seems to say that the fossil record exhibits a higher-to-lower transition in the case of insects -- exactly the opposite of what Darwin's tree of life was supposed to show. Am I misunderstanding? Eric Anderson
First of all we should note that the original statement that 99% of all species are extinct assumes evolution, a fact not in evidence. I don't believe that the fossil record supports the assertion. The excuse is that insects do not fossilize well and therefore all the supposed species that are extinct are missing from the record. This seems overly convenient. Evolution of the gaps? What we do know is that there were several orders of insects that went extinct. From the Virtual Fossil Museum
By the end of the Carboniferous, the subphylum insecta had evolved into a large number of distinct orders. During the Permian, new insects forms appeared. Blattoid and Orthopteroid orders attained their greatest diversity, and new groups like the Psocoptera, Homoptera, Hemiptera, Mecoptera and Flying termite in fossil amber Coleoptera became ubiquitous and diverse. The Permian extinction wiped out nine orders of insects, and more orders disappeared in the Triassic or the early Jurassic.
Unfortunately what percent of species that these orders represent is probably unknowable in an absolute sense. Just because you have several examples (species) of a particular order tells you nothing about how many others there may have been. Orders can have a large number or only a few representatives. Still this may allow estimation of a reasonable range. Latemarch
BA77 @6, Thanks for the link. I'll try to find time this week to watch the video. Eric Anderson
Latemarch @10. Thanks for the source. Very instructive. Eric Anderson
Thanks, Bob O'H. This is a great link. Did you happen to find a similar breakdown for fossil species? You mention that "big vertebrates and inverts with shells are over-represented, and the big vertebrates certainly only account for a minor proportion of total diversity of life." I'd be inclined to agree with you. Yet this does re-raise the question in a slightly different way. Specifically, even if large vertebrates and shell-carrying species make up the vast majority of the 250,000 fossil species, it still would not even approach 99% compared with today's extant species. Maybe 50%, 30%? Gould and Eldridge seemed to suggest that the fossil record is, if imperfect, a good representative sample for the large vertebrates and shell-carrying species you mention. If so, then, at least for those species, 99% isn't even in the ballpark. The quotes by Latemarch @10 are instructive. In your response @11 you mention that "only a small proportion of insects" have been found in fossils, but I can't tell if you are referring to individual organisms. No doubt it is true with respect to individual organisms. But is it true with respect to represented species? Eric Anderson
Latemarch - thank you for that link. The first paragraph is instructive for this thread:
Unlike the trilobite that has left a prodigious fossil record, the preservation of insects in sedimentary matrix is relatively rare, and essentially limited to the Lagerstätte sites. The reason for the scarcity of insect fossil is the poor preservation potential of the insect's exoskeleton. Like other Arthropods, insects have an external skeleton called an exoskeleton. Unlike the thick and calcified trilobite exoskeleton, the insect exoskeleton is made of a thin, plastic-like material called chitin, along with a tough protein. This thin, waterproof covering simply does not preserve well in most oxygenated environments, making insect fossils sparse despite the tremendous number that could have been preserved. The exception is in fossil resinite (amber, by street name), where it is possible for even the minutest details to be preserved right down to fragments of DNA. Despite their huge strength to weight ratio, insects were often too small to escape the sticky resin exuded by trees, and which later became a fossil itself, with physical properties akin to modern polymerized plastics. Countering low probability of fossilization is the prodigious fecundity of insects.
In other words, only a small proportion of insects have been found in fossils, and those from one particular type of fossilisation. Bob O'H
From the Virtual Fossil Museum
Taxonomic research on fossil insects has always been relegated to a subordinate role when compared to that of living species. There are large numbers of undetermined fossil insects in many collections throughout the world awaiting descriptions, but only a small fraction of systematic research has ever been devoted to these fossils.
They hint that the fossil record is actually reasonably complete but not well studied. No one is really interested. Possibly because....
So exquisite is insect design that most groups were well formed by the Cretaceous and remain largely unchanged in appearance during modern times.
(my bold) The Cretaceous is when we have abundant records of insects in amber. We find them almost identical to insects of today. No change for millions of years.....well....that's inconvenient. Latemarch
All of the Precambrian rabbits failed to fossilize! And the humans from 100 million years ago to just a couple thousand years ago all burned their dead, leaving nothing to fossilize. Yeah, that's the ticket... ET
It would be interesting to run the comparison the other way. The fossil species are about 20% of currently known species. How many of the current species have shells or skeletons that would be found in FUTURE fossils? I'll bet it's not far from 20%. polistra
Eric Anderson, welcome back and you come at a good time. No claims about numbers can make any sense at all without a science-based definition of "species." I've been harping on that point for years. Same with claims about "Anthropocene extinctions." Absent a scientifically robust measurement of what constitutes a species, people can make almost any point they want from a lectern or podium. Much of it sounds like doubtless well-meant rubbish. I keep hearing about the extirpation/extinction of animals later spotted in local urban areas over the summer. The real loser may be the wise use of public resources. What ARE we trying to preserve and why? We can't get anywhere with a key question like that if we are using a system of measurement based on centuries of "Well, he classified it this way." So what? News
Fish, Fossils and Evolution - The Myth Of +99.9% Extinct Species – 22:06 minute mark - Dr. Arthur Jones - video http://edinburghcreationgroup.org/video/14 "Stasis in the Fossil Record: 40-80% of living forms today are represented in the fossil record, despite being told in many text books that only about 0.1% are in this category. The rocks testify that no macro-evolutionary change has ever occurred. With the Cambrian Explosion complex fish, trilobites and other creatures appear suddenly without any precursors. Evidence of any transitional forms in the fossil record is highly contentious." Paul James-Griffiths via Dr. Arthur Jones Darwin on the rocks - Sept. 19, 2014 Q&A | DNA and Cambrian fossils, says Stephen Meyer, make macroevolutionary theory increasingly untenable Excerpt: What you found in the Cambrian was 23 distinct body plans, and fully 20 of those first appeared in the Cambrian. There are only about 27 body plans that have been preserved in the fossil records, total. So you can see this is a big event in the history of life. http://www.worldmag.com/2014/09/darwin_on_the_rocks Alleged Refutation of the Cambrian Explosion Confirms Abruptness, Vindicates Meyer - Günter Bechly - May 29, 2018 Excerpt: Even though Daley et al. (2018) do not bother to mention Darwin’s Doubt, they vindicate three main theses of Stephen Meyer’s book. First, the authors confirm Stephen Meyer’s refutation of the artifact hypothesis and my own argument from the absence of animals in recently discovered Burgess-Shale-type fossil localities from the Ediacaran period (Bechly 2018).,,, https://evolutionnews.org/2018/05/alleged-refutation-of-the-cambrian-explosion-confirms-abruptness-vindicates-meyer/ June 2019 - "It turns out that the fossil record itself reveals an ‘upside down’ pattern for the appearance of the various classifications (of fossils) than what Darwin’s theory predicted https://uncommondesc.wpengine.com/intelligent-design/why-some-biologists-are-beginning-to-question-the-biological-species-concept/#comment-679510
Is the fossil record truly so poor that many billions of species have come and gone without leaving a discernable trace?
Yes it is. Of the ~2m species described, about 1m are insects, and only about 65k are vertebrates (source). Insects don't fossilise well (no mineral shell or bone), so they're massively under-represented in the fossil record. Other inverts (e.g. worms) which have even softer bodies, so are even less likely to be fossilised. Big vertebrates and inverts with shells are over-represented, and the big vertebrates certainly only account for a minor proportion of total diversity of life. Bob O'H
Is there an estimate for the number of newly described fossil species being discovered each year? Another thing to look at could be: how often are organisms discovered that are very distantly related to all known organisms? Just some early thoughts on the topic. KJul3s
very good question, with good context kairosfocus
Let me describe the issue in another way: The 99% claim cannot be true (except by pure coincidence of a luck guess). It cannot be true because we know neither (a) how many species previously lived on the Earth, nor (b) how many currently live on the Earth. So we have excellent reason to take the claim with a huge grain of salt right out of the gate (and anyone repeating the claim, should pause to thoughtfully consider whether we need a little more evidence before just regurgitating the claim). What I'm trying to pin down is a less precise, but more substantive, number. Is the 99% even in the range of what our best information suggests is correct, or is it way off the mark - perhaps even to the point of having things upside down? I'm interested in the range, and it seems we should be able to get some handle on the range, if not the precise value. IOW, I'm less concerned about whether it is 99% or 98% or even 95%. But if it is 50% or 25%, that would be a fundamentally different situation indeed. Eric Anderson
Thanks, Latemarch. Yes, definitely one of the challenges in getting to any solid numbers. My guess is that a broader definition would tend to reduce the count of currently-existing species vs fossil species. I'm not sure it would appreciably change the numbers, and wouldn't impact the general issue under discussion, but it would be an interesting exercise for some enterprising post-doc to write a nice paper about. :) Eric Anderson
Would you care to define species?.......Oh wait.... Latemarch

Leave a Reply