• Darwin’s Doubt – Chapter 1 – Part 8

    We continue the last section that we began (location 567 Kindle edition* and pg 20 print), An Old Fossil Recovered.

    I wanted to spend some time on this because it shows a very common mistake made by many people, not just creationists.  Creationists, however, capitalize on this misconception for their own ends… even when they know better.  There’s two possibilities here.  Either Meyer really doesn’t know any better, in which case, he should never have written this book in the first place.  Or, he does know better and is still saying it incorrectly, in which case… well…

    This misconception has to do with what is expected from the fossil record.

    Rather, the problem, according to Agassiz, was the selective incompleteness of the fossil record.

    Why, he asked, does the fossil record always happen to be incomplete at the nodes connecting major branches of Darwin’s tree of life, but rarely—in the parlance of modern paleontology—at the “terminal branches” representing the major already known groups of organisms? These terminal branches were well represented (see Fig. 1.8), often stretching over many generations and millions of years, while the “internal branches” at the connecting nodes on Darwin’s tree of life were nearly always—and selectively—absent. As Agassiz explained, Darwin’s theory “rests partly upon the assumption that, in the succession of ages, just those transition types have dropped out from the geological record which would have proved the Darwinian conclusions had these types been preserved.”53 To Agassiz, it sounded like a just-so story, one that explains away the absence of evidence rather than genuinely explaining the evidence we have.

    And it accompanied by this graphic

    Darwin's Doubt 1.8


    You’ll notice that the left half is a common graphical representation of evolution.  There are two common ways of showing evolutionary relationships, cladograms and phylogenetic trees.  This isn’t a cladogram because nodes aren’t organisms in a cladogram.  So this must be a phylogenetic tree.

    So, why is this a problem for me in this case?  It’s actually very simple.

    Unlike what many people think, evolution doesn’t occur at an individual level.  It occurs at the population level.  Remember a while back we talked about how there is often much more variation within populations than between parent/offspring?  This is why we don’t expect to find that single individual that is the exact last common ancestor between two groups.

    An illustration is in order here. This is from Scitable by Nature Education.  I was thinking about doing this by hand, but this one is so much better.  I’ve made some minor modifications to emphasize what we’re talking about.

    This the phylogenetic tree that we all think of.  Species A and B are closely related.  Species C and D are closely related and all four share a common ancestor.  Now the problem as stated by Agassiz (via Meyer) is that we don’t have examples of the fossils at the nodes (I’ve circled the nodes in red).



    But those nice clean lines are really meaningless.  Species aren’t a single organism and evolution doesn’t happen to individuals.  Evolution occurs in populations.  So in reality, the tree looks more like this.


    Those nice clean lines are really aggregates of millions, possibly billions of individuals over a period of millions of years (generally).  Look at part A.  Each line represents an individual.  Two individuals briefly come together and a whole bunch of new lines form, the offspring.

    Let’s look at a blow up of one of those nodes.



    I’ve added some of the tracks of populations.  Notice how some groups follow one branch early on.  Some begin to follow one track, but move over to another track.  Some then move back.

    Now, the question for Meyer and Agassiz is, which one of those population tracks contains the individual that is the dividing point between species C and species D?

    Or let me ask this in another way.



    Which of the above breeds of dogs is the nodal individual for the varying species of dog that will appear in 3 million years (give or take).

    Yes, that is a valid question.  Because it’s why Meyer and Aggasiz’s point here is wrong.  All of those dogs are the same species… now.  But after 3 million years of evolution (whether it be human induced or not), do you really think that corgis, pugs, and spaniels will be the same species as the sheepdogs and deerhounds.  Yes, over the next 3 million years, there will be some mutts that are cross breeds between pugs and beagles and between sheepdogs and collies.  But there will be less and less between the small and toy dogs and the large and giant dogs.  It will take time, more time than humans have been alive, but these will become separate species.  It’s already happened with the foxes.

    Now, before you say “That’s a just-so story”, it’s not.  One group of organisms have had several species genomes completely analyzed and this pattern appeared.  There was some cross-breeding, so much that it’s visible at the genomic level, for several million years.  I refer to this image that details this, also from Nature.

    a, Schematic description of ILS states and percentage of bases assigned to each state. b, Effective population sizes and split times inferred from ILS and based on a molecular clock with a mutation rate of 10−9 yr−1. Myr, million years. We note that other estimates of mutation rates will correspondingly affect the estimates of the split times. c, Overlap between predicted ILS transposons and the closest HMM ILS assignments within 100 bp of a transposon insertion. d, Proportion of ILS in exons, introns and across the whole genome, counted within ~1-Mb segments of alignment (Supplementary Information, section 8). e, Proportion of ILS dependent on recombination rates. Errors, 95% confidence interval.

    Of course, the species I’m referring to here are the human, chimpanzee, and bonobo.

    Now, looking at the data above (I’m sure the authors would be willing to share the original data if you have a good experimental plan), which individual in that 3.5 million year period, is the one that is the nodal species?

    One last point.  The fossil record of these types of splits is very limited.  In general, they occupy a brief period of time, geologically speaking.  One spring, many, many moons ago, I spent some time doing field work in south Texas.  We were looking at a marine/estuary fossil bed that was between 48 million and 42 million years old.  It was compressed into a fossil layer about 6 inches thick.  Nearly six million years of living things compressed into 6 inches of rock.

    Only specific regions have good chances of fossilization happening.  Rain forests, mountains, and similar environments offer very, very poor chances of fossilization.  Any place that there are significant scavenger species, ranging from vultures and lions to beetles and ants, are poor areas for fossilization.

    Further, as Gould and Eldridge pointed at, species tend to remain stable over long periods of time, then rapidly diversify due to various pressures.  This is called punctuated equilibria and is a well known process to modern scientists.

    Finally, as we’ve pointed out before and like the dogs show, being able to decide whether an organism is a particular species or another is not a trivial task.  Careers and reputations have been made and ruined by claims of this sort.  We don’t know which dogs will become separate species in the future.  We can’t tell which snail became a new species, because at the time, they are all the same.  It gets much worse when all you have to look at is a few bones or a shell of an animal.

    All of these factors contribute to what Agassiz calls the lack of fossils at the internal branches.  But… and this is a big but… there are a few.


    Tiktaalik.  Which further proves the point that there is no fixed time when species diverged.  But more on that later.

    Please ask questions,  I hope I was clear, but I don’t want any misunderstandings.


    * Please note that I’m doing this on a Kindle.  I like e-books.  I encourage their use.

    Category: Book ReviewCreationismEvolutionScience


    Article by: Smilodon's Retreat

    • Buho

      Love your graphics. The tangled rope is a fantastic visualization! It clarifies a fuzzy mental picture I’ve had of evolution for years.

      If specimen A and specimen B are on opposite sides of the rope, your complaint would be valid. Meyer is pointing out that A and B are MILES apart. Always. Sure, the fossil record is incomplete, but as Meyer gets into in chapter 3, a pattern has emerged: they cluster in groups separated by miles, not inches (the fine gradation), suggesting there the lack of connections is indicative of reality. In other words, it’s not incomplete enough anymore. The problem Agassiz pointed out remains today, but now much more sharply.

      In chapter 7 Meyer goes into much more detail with Gould, whom you quote in support. Punctuated equilibrium is the opposite of what evolution needs, the opposite of your visuals. Meyer shows how Gould backed away from Punk Eek over the years (and his reasons), returning to the more mainstream slow-and-steady theory evolution.

      I’m looking forward to the next post!

      • SmilodonsRetreat

        A and B do not have to be Miles apart. Indeed, that’s the entire point of the process of speciation. One species, over time, becomes two. Just like breeds of dogs, they are different. Indeed, if only great Danes and teacup chihuahuas were left we would consider them different species because they could not interbreed. Yet, because the intermediate breeds exist, we know that both are just breeds of dog.

        That dividing line between A and B is, often, so subtle that we can’t see it… until well after the fact.

        Which is the entire point. We can’t see new phyla forming… until hundreds of millions of years down the line. We can’t see new orders, new classes forming until well after the fact.

        At the time of the cambrian and precambrian… what we now understand to be two vastly different phyla may have been as close together as Great Danes and chihuahuas. And that’s the very, very important point that Meyer ignores here.

        Oh, and I can’t wait to get to Meyer quotemining Gould. He was a good scientist, and a great writer, but he was not very careful how he spoke and is a huge target for quotemining.

        BTW: I can show you evidence that punctuated equilibria actually happens. Can Meyer show evidence that ID happens? I guess we’ll see.

        • Buho

          I understand your analogy, Smilodon. But great Danes and chihuahuas are “inches” apart in comparison to, say, trilobites and brachiopods (figures 1.3 and 1.4), the stuff of the subject of the book. Two completely different body plans.

          • SmilodonsRetreat

            Exactly. Thank you for supporting my point perfectly.

            Now, when one says that trilobites and brachs are so far apart, we have two ideas about how that my have happened. The first is that they were created, fully formed and complete… according to creationists. I don’t know what Meyer says, I’m just getting into Chapter 2.

            The other idea is that of evolution. If they exist now, then other forms must have existed previously. Hmmm….

            Here’s some work that you might be interested in reading about. I wonder if Meyer mentioned any of it. I guess we’ll see,

            • Buho

              The diagram that began this post illustrates the pattern in the fossil record: fossils tend to cluster in narrow “inch-wide” bands separated by “miles.” Do you disagree this is the case in the fossil record?

            • SmilodonsRetreat

              Yes. There are no distance markers in the fossil record or in that diagram. ANy discussion of “miles” and “inches” is analogy without meaning.

    • Christine Janis

      Another point is that the “terminal branches” are *not* complete (as shown in Meyer’s figure 1.8). They consist of point occurrences joined together by lines. So, they show the issue of sampling. The likelihood (or not) of a sample being discovered precisely at the point of splitting from another group is the same as the likelihood for any other predetermined point on the line —- pretty unlikely.

      I’ve heard creationists say things like “4000 bat fossils and yet no intermediate forms” — and nobody seems to see how absurd this is as any bat found after the earliest Eocene has no potential for being an “intermediate”. So, it’s irrelevant how many fossils are known in the past 54 million years. Also, any transition event is not only going to be rapid, but local. One of the best arguments for the transitional nature of the early whale fossils, that I hardly ever see anyone mention, is that they’re all known from a small area in Pakistan.

      • SmilodonsRetreat

        Excellent points. Thanks!

    • Jeremy Mohn

      Thanks for your post. This was a very common misconception that my students had about phylogenetic trees, so I made this Prezi:


      I just thought your readers might find this useful as an illustration of the point you were making.

      Being able to visualize the population evolving from one generation to the next on such a large scale has really helped my students grasp the concepts relating to Hardy-Weinberg equilibrium as well.

      • SmilodonsRetreat

        This is awesome!!

        Everyone… click on that link and watch!

    • Joshua Hedlund

      Have you spent several beautiful paragraphs on a straw man? The problem as described in the original left/right graphic is not merely that we don’t have fossils at the red circle nodes, but that we don’t have fossils at the nodes OR anything prior to them. You spent several paragraphs explaining why we shouldn’t expect fossils at the nodes, but unless I misunderstand you didn’t explain why shouldn’t expect to find fossils BEFORE the nodes ether (i.e. the entire difference between the left and right graphs at the top, not just the nodes)

      • SmilodonsRetreat

        OK, I can see that. In this specific case, we do have fossils before the middle Cambrian that Meyer spends a lot of time talking about. We have many fossils from the Early Cambrian (I’ve given to understand that Meyer ignores the small, shelly fauna fossils of late Ediacaran to the early Cambrian) and the mid-Ediacaran fossils. In fact, the very earliest megafossil we have is from nearly 70 million years prior to the start of the Cambrian. These include fossils with radial and bilateral symmetry.

        Now, is the only acceptable evidence an unbroken record of organisms showing minimal changes all the way from say Aspidella through trilobites? In that case, you will be disappointed. In all probability, that will never happen. It’s the nature of the fossil record of half a billion years ago and where these rock layers are (not subducted or metamorphised; not buried under ten thousand feet of limestone; not eroded away to nothingness; etc).

        Of course, that’s not required and never has been. I occasionally ask creationists if they can create a verifiable unbroken string of names from themselves all the way back to Noah (or Adam). If they can’t, then how do we know that they really are humans? Of course it’s a silly question. But it’s silly in both cases.

        I’m hopeful that as we go through Meyer’s book, then we will discover many of these fossils for ourselves (whether Meyer mentions them or I have to bring them up). There’s any number of good texts on the subject as well. I suggest the following:

        The Cambrian Explosion: The Construction of Animal Biodiversity


        The Rise of Animals: Evolution and Diversification of the Kingdom Animalia

        • Joshua Hedlund

          I don’t read Meyer as asking for an unbroken chain of organisms; I read him as asking for a chain without wildly statistically improbable gaps (i.e. lots of dots along the “terminal branches” of all the current phyla” with little to no dots along the pre-phyla “inner branches”)

          However, are you essentially saying, regardless of whether the diagrams represent individuals or populations, that the “diagram on the right” in the first graphic is demonstrably false, that we do have fossils that represent common ancestors of multiple phyla in the animal kingdom? Meyer does talk about the Ediacaran a little later (I’m only a couple chapters ahead of you at this point), but he argues that they may not be common ancestors of the later fossils or that even in a best case there are still something like 17 phyla that appear in the record within ‘only’ a few million years and with no prior ancestry. I would love to get your take on it, because either I’ve missed it in the reviews I’ve read or I’m too much of a layperson to infer it from them but I don’t think I’ve seen anyone explicitly rebut that point.

          • SmilodonsRetreat

            I’ll see what I can do. I’m trying to find some specific sources right now. I’ve got a few others that specifically refute this point.

            I’d like you to read this page: http://www.skepticink.com/smilodonsretreat/2013/09/16/where-higher-orders-of-life-come-from/

            And tell me what you think.

          • Joshua Hedlund

            I read it and I think I understand at least on some level… I’m trying to walk through how that changes expectations for the fossil record compared to Meyer’s expectation with his alleged 17 phyla… OK, so we’ll call them species (or distinct groups of similar species) because they’re not phyla yet, how does that change things?

            Well, maybe they’re not that distinct. I’m assuming they are because otherwise someone could say, No, we actually don’t have the first known representatives of 17 modern phyla in a few million years, some of them could be ancestors of more than one modern phyla, so we only have these 8 or 10 or whatever, and we have these 7 other species that could actually be transitional between some of the modern phyla. I don’t see anyone saying that, so I assume these early representatives are already pretty distinct species.

            Now, if they *are* all distinct fuzzy tangled ropes descended from the same fuzzy tangled rope, is it statistically likely that we have lots of fossils from *all* the tangled ropes around time X but little to no fossils from the fewer tangled ropes above them around time X – Y, even though we have other fossils from around time X – Y? Are there likely other tangled ropes in the picture that should affect the expectations?

            I guess I’m trying to clear up whether the rebutters are saying that Meyer is wrong about the number of distinct species groups appearing in a short amount of time, or if he’s right about them appearing but wrong about them not having ancestors in the record, or if he’s right about them appearing and not having ancestors in the record but wrong about expecting to have ancestors in the record. It seems like you’re arguing for the last point, but in my ignorance I’m not seeing how calling them species instead of phyla changes things. Maybe it will come to me when you get to chapters 3/4.

    • Buho

      With Chapter 1 finished, I’m curious what your takeaway is with Meyer’s
      main point for Chapter 1. Meyer’s main point was that Darwin had a doubt.
      Do you agree or disagree?

      • SmilodonsRetreat

        Any scientist has doubts. That’s the nature of science. The only people without doubts are people like Meyer and other creationists.

        Historically speaking, Darwin listed several things that would disprove his ideas. Those things, with our modern knowledge of genetics and mathematics, have all been tested and shown to support evolution… modern evolution, not Darwin’s evolution.

        Chapter 1 was a poorly written history. I say poorly written because Meyer interspersed his own ideas without telling the readers that he was doing so. Using modern references (quotemined) to support the questions that Darwin presented. And without telling the readers, that those questions have been answered.

        • Buho

          Thanks for answering! The details you’re providing are great, but I worry you may loose sight of the main points being made. Meyer hits up the latest research later in the book. I’m looking forward to your thoughts on them.