Well, well, Wells: Jonathan Wells reacts

Jonathan Wells, the creationist who makes shoddy claims about developmental biology, has deigned to respond to my criticisms…but only indirectly, on another blog. It’s an interesting response, in that it once again reveals Wells’ misunderstandings of biology, and his sneaky way of inserting phony claims.

Here is Wells’ rebuttal.

The issue here is not all that complicated. Darwin thought that “community in embryonic structure reveals community of descent” and concluded that early vertebrate embryos “show us, more or less completely, the condition of the progenitor of the whole group in its adult state.” Darwin considered this “by far the strongest single class of facts in favor of” his theory. (Origin of Species, Chapter XIV; September 10, 1860 letter to Asa Gray)

But early vertebrate embryos do not look alike. They become somewhat similar (though not as similar as Haeckel made them out to be) midway through development, then they diverge again. This is illustrated by the “developmental hourglass” drawing on page 31 of my Politically Incorrect Guide to Darwinism and Intelligent Design. Developmental biologists (including P.Z. Myers, to judge from his Panda’s Thumb review of my chapter) are well aware of this pattern, which has been described repeatedly in the developmental biology literature.

But an hourglass pattern does not provide the evidence Darwin needed for his theory. If “community in embryonic structure reveals community of descent,” then a pattern of early differences followed by convergence followed by divergence makes no sense. Some modern Darwinists, instead of acknowledging the problem, simply attribute the early differences to evolution. In other words, they assume their theory is true and then use it to explain away anomalies in the very evidence that was supposed to provide the strongest support for it. Meanwhile, other Darwinists provide a smokescreen for this circular argument by calling their critics names…

Is that how science should be done?

So much slipshod nonsense! Let’s take this apart. First, I’ll show how Wells distorts Darwin. Then I’ll say a little about what the point of similarity in embryos is, as understood by Darwin and by modern developmental biology. Then I’ll explain a little more about the hourglass, although I think I covered it fairly well in that earlier effort. Finally, I’ll ask how intelligent design creationism should be done.

A little historical background

Wells is guilty of frequently imposing a discredited old philosophy on modern biologists, so first let me clarify exactly the two ideas he is confusing.

In the early part of the 19th century, particularly among a particular school of German philosophers, there was an idea that development represented the expression of a string of archetypes—that in human development, for example, we can talk about a “molluscan stage” and a “fish stage” and so forth, and that each species can be characterized by how high on the ladder of progress its development ascends. This idea got smacked down hard by the facts, but was revivified late in the century by Ernst Haeckel—who did not invent it, but was responsible for popularizing it with snazzy slogans and tying it to an evolutionary mechanism. Thanks to Haeckel, you’ll hear it called the Biogenetic Law, or Recapitulation Theory, or you’ll hear the phrase “Ontogeny Recapitulates Phylogeny”. It is the idea that during development, organisms re-enact their evolutionary history. They don’t. This theory required that organisms go through a long chain of similar developmental events that corresponded to the step by step changes in their evolution, an expectation that did not and does not fit the data, was already dismissed and replaced by an explanation from the premier embryologist of the day, and even when it was briefly popular, required much handwaving and special exceptions to accommodate widely known differences.

That premier embryologist who royally smacked down the idea of recapitulation was von Baer. He had a simple idea that explained reality better than a ladder of archetypes or development as a historical re-enactment, and he had good reason to do so: he disliked that school of German philosophy, never accepted evolution, and he was the fellow who first made the observation of similarities between vertebrate embryos. Von Baer’s principle was that development proceeded from the general to the specific. That is, embryos first sketched in the major features of their phylum—a notochord, a tail, pharyngeal structures, for instance—and that the reason for the resemblances is simply that early embryos represent pared down sketches of the animal, an outline or framework. Development then proceeds by filling in the outline, adding more and more specific details until the full set of derived characters are present. Von Baerian explanations of developmental similarities require only a single point of similarity, and that is based entirely on the presence of only key features of the phylum and the absence of derived features of the species. That, as I’ll show below, was the basis of Darwin’s use of embryology, and it is a better reflection of modern views. Recapitulation had its brief flurry of popularity, but it’s dead now, and has been for over a century.

Darwin’s views on embryology

So let’s consider Darwin. Biology has moved far beyond Darwin, but Wells wants to claim that this is an issue on which Darwin was wrong, so I think it’s worthwhile to show that the only person who has it wrong here is Wells. I warned you to be watchful whenever Wells starts chopping up quotes, and here again we have an example. Wells wants us to believe that Darwin was promoting Haeckelian recapitulation, that his “strongest single class of facts” in favor of common descent was that there was an early similarity (at the earliest stages!) between all vertebrates that progressively diverged. Unfortunately for Wells, not even Haeckel believed that. Darwin certainly didn’t. Here’s the appropriate section from the first edition of the Origin:

In two groups of animal, however much they may at present differ from each other in structure and habits, if they pass through the same or similar embryonic stages, we may feel assured that they have both descended from the same or nearly similar parents, and are therefore in that degree closely related. Thus, community in embryonic structure reveals community of descent. It will reveal this community of descent, however much the structure of the adult may have been modified and obscured; we have seen, for instance, that cirripedes can at once be recognised by their larvæ as belonging to the great class of crustaceans. As the embryonic state of each species and group of species partially shows us the structure of their less modified ancient progenitors, we can clearly see why ancient and extinct forms of life should resemble the embryos of their descendants,—our existing species.

There are some differences in what Wells quoted and what Darwin wrote here, but I’ll charitably assume it was because Wells is looking at a different edition (I didn’t feel like checking every edition to see where he got this). However, I think you can get the sense of what Darwin was talking about. Embryos and larvae are important sources of information in classifying species; taxonomists had noted for years before that access to larvae was a critical tool in resolving species relationships. What he was explaining is that even in highly derived adult forms, the more generalized earlier forms, before those obscuring later specializations have developed, expose deep similarities that are indicators of similarity in descent. Wells is a little dodgy here; does he want to deny that?

Note also the important modifiers, not preserved in Wells quoting, in that last sentence. The embryonic state partially shows the structure of less modified ancient progenitors. These modifiers are important tip-offs: Darwin wasn’t peddling Haeckelisms here, that’s von Baer! Von Baer was an opponent of any kind of recapitulation in development, and his proposal was that what we saw in development was a progression from the general to the specific and derived adult forms; not a series from one ancient form to another, but that development builds on a more generalized framework. What Darwin is saying is that development allows us to see that framework or body plan more clearly.

(One qualification, however. Von Baer would argue that embryos only resemble the embryos of progenitor species. Darwin is making the mistake of suggesting that embryos resemble the adults of progenitor species. This is not correct…but then, Darwin wasn’t perfect. Modern developmental biologists would not, I hope, make this error.)

There’s another clue that Darwin is discussing von Baerian developmental principles rather than Haeckelian recapitulation: the dates. The first edition of the Origin was published in 1859, and that letter to Asa Gray is from 1860. Recapitulation was first proposed by Muller in 1864, and Haeckel published his ideas on it in 1866. Isn’t it silly to blame Darwin for something someone else did years after Darwin published?

Darwin did revise the Origin multiple times. Perhaps he adopted Haeckel’s views later? Here’s the same section, from the 4th edition in 1866, where perhaps Darwin at least had an opportunity to be influenced by those Germans.

In two groups of animal, however much they may at present differ from each other in structure and habits, if they pass through the same or similar embryonic stages, we may feel assured that they have both descended from the same or nearly similar parents, and are therefore in that degree closely related. Thus, community in embryonic structure reveals community of descent; but dissimilarity in embryonic development does not prove discommunity of descent, for in one of two groups all the developmental stages may have been suppressed, or may have been so greatly modified as no longer to be recognised, through adaptations, during the earlier periods of growth, to new habits of life. Community of descent will, however, often be revealed, although the structure of the adult may have been greatly modified and thus obscured; we have seen, for instance, that cirripedes, though externally so like shellfish, can at once be recognised by their larvæ as belonging to the great class of crustaceans. As the embryonic state of each species and group of species shows us more or less completely the structure of their less modified ancient progenitors, we can see why ancient and extinct forms of life should resemble the embryos of our existing species, their descendants.

Note the italicized section. Rather than hardening in his views, Darwin expands them to mention that there can also be modification of developmental processes, even at earlier periods. This is true of modern biology as well, which sees no mechanism for barring genetic change in any stage of development, and recognizes variation where it occurs. As usual, Wells is warring with a straw man when he insists that “early vertebrate embryos do not look alike”; they certainly do, in a general sense, and that is exactly what Darwin is describing. Haeckel’s ‘laws’ were dead by the end of the 19th century, they are not accepted by any credible modern biologist, and most interestingly, Darwin doesn’t seem to have subscribed unquestioningly to them, either.

Similarities in embryos

So what was Darwin going on about with his talk of “community in embryonic structure”, and has it been shown to be wrong? It’s actually a very simple concept, and it is still valid.

Here’s what Darwin faced: life is very diverse, but he is arguing that it all can be traced back to a common root, that all the differences can naturally arise from something initially similar. Darwin is making the case that while you may find the differences overwhelming and the possibility of any unity of form improbable, all you have to do is look at embryos to see that they are based on similar plans.

For example, a bat is small, hairy, has prominent wings and big complicated ears, flies through the air and eats insects. A dolphin is large, smooth skinned and torpedo shaped, has greatly reduced limbs and no external ears, swims and eats fish. Folk classifications see no significant problem in grouping bats with birds and whales with fish, so obviously these are two extremely distinctive groups of animals. What reason do we have for assuming that they have any family relationship at all?

There is the evidence from anatomy and physiology—a good comparative anatomist can see all the mammal-specific details of their forms, and we can also find deep similarities that tie bats and dolphins more closely together than we can tie dolphins and fish, for instance: dolphins are endothermic and produce milk, as do bats but not fish. But what’s also very vivid and dramatic is to observe embryos of both species, and see that the commonalities in their body plans can leap out at you.


What these images show is that when you catch them before they’ve developed those species-specific attributes like wings and flippers, when what you’ve got is simply the most basic elements of the body plan, the similarities are apparent. This is what Darwin is talking about; you have to explain why a torpedo shaped dolphin and a fuzzy flappy bat have this core of morphological similarity to one another. Darwin’s explanation is historical. These similarities reflect a common ancestry, and what’s more, they illustrate that it is not absurd to imagine diverse organisms arising from a single common plan.

That darned developmental hourglass

I’ve already summarized the evo-devo explanation for the developmental hourglass, so rather than rehashing that, I want to address two points.

First, Wells is playing a dishonest game. In Icons of Evolution, he basically accuses biologists of fraud, perpetuating a myth of Haeckelian recapitulation in our textbooks. He makes it sound as if our only rationale for reconciling evolution and development is to reuse these old, discredited 19th century ideas. He complains when textbooks use drawings derived from Haeckel’s old work, but he also complains if textbooks use photographs of embryos at the phylotypic stage. Now, though, when his accuracy is called into question, he suddenly explains that he was right in his characterization of embryology because the developmental hourglass is “described repeatedly in the developmental biology literature”.

Is he aware that the hourglass model and the recapitulation model are not the same thing? Isn’t it a little dishonest to, on one hand, complain that we are using a false and antiquated data set, and on the other, that we’re describing embryology accurately as a convergence on a common form followed by divergence? Wells flip-flops between those accusations at his convenience. Which is it, Mr Wells? Are biologists slavishly repeating discredited Haeckelian models, or are they repeatedly describing the decidedly non-Haeckelian hourglass model? I suspect it doesn’t matter how he answers; at his next talk before a credulous creationist crowd, he’ll again tell them that we are followers of the biogenetic law.

Secondly, Wells may quote Darwin, but he doesn’t understand him. How else can he claim that “an hourglass pattern does not provide the evidence Darwin needed for his theory”? Darwin does not say that his theory requires that all embryos are identical at the onset of development and progressively diverge. He does not say that earlier stages are perfectly insulated from evolutionary change. Quite the contrary, he argues that in some lineages, earlier stages can be greatly modified, obscuring similarities. He was also basing his interpretation of embryology on the work of von Baer, who was quite aware of the appearance of early variations in embryos, in for instance, the extra-embryonic membranes. Since Darwin knew of differences in the earliest stages, and knew that the similarities were not apparent except in the post-gastrula embryo, I can’t imagine that he would be at all discomfited at an explicit statement about the pattern of development, such as the developmental hourglass—he already knew about it!

Since all Darwin was saying is that there are similarities in embryonic organization between vertebrates at a stage in development, and that was all he needed to argue for similarity in basic structure and for the visible demonstration of a transformation of a single form into different adult types, Wells’ assertion does not follow.

Your turn, Intelligent Design creationists

Wells ends his complaint by asking, “Is that how science should be done?” It’s an acutely ironic question, since Wells’ books, both Icons of Evolution and The Politically Incorrect Guide to Darwinism and Intelligent Design, are little more than extended slanders of modern biology and certainly fail to offer any alternative explanations of the phenomena he disputes. He has already admitted that developmental biologists document the pattern of morphological (and molecular, I would add) development, and have shown a convergence on a broadly similar form at the phylotypic stage. That’s a phenomenon that needs to be explained.

Evolutionary biologists have an explanation in terms of common ancestry and developmental constraints.

Where is the Intelligent Design creationists’ explanation?

And no, “the designer did it that way” is not an explanation. I would like to see a testable, coherent explanation for why dolphin and bat embryos should look so similar—one that can also lead to continuing investigations and that makes predictions about properties of embryos that are unique to the design hypothesis. Wells has never offered any. All he can do is wobble between outright denial that any similarities exist, untenable distortions of Darwin’s ideas and those of modern biologists, and rare admissions that yes, the similarities exist, followed by incorrect and nonsensical assertions that they don’t fit evolutionary theory.

No, that isn’t how science should be done. That’s why Intelligent Design creationism is not science.


  1. #1 Ichthyic
    November 7, 2006

    Of course, you do realize that this is the part that the Creationist then says “well, since you don’t agree with me, I now know that you’re going to burn in Hell with the rest of (the people who don’t think like me)”

    of course!

    I personally am counting on it. i think in an earlier thread I was given the job of organizing the endless “Mystery Science Theatre” horror fest when I get there.

    looking forward to it.

    ahh, hell… the only real way to escape from the fundiebots.

  2. #2 John Wilkins
    November 7, 2006

    Neo-Darwinism can mean:

    1. The views of Wallace and Weismann, c1895-1918, in which natural selection was the sole mechanism of evolution worth talking about, although Weismann allowed for something like drift. Moreover, these guys also rejected Darwin’s “pangenesis” inheritance theory in favour of a sequestered germline.

    2. The views of the original Modern Synthesis, c.1930-1950, in which Mendelian genetics and Natural Selection are the primary aspects of evolution (from which we get the “change in allele frequencies” definition of evolution). The contrast here is with people like Waddington and Sewall Wright.

    3. The c1964-1980 views of the post-Hamilton evolutionary theorists like Maynard Smith, Price, and others (like Dawkins, a decade later) who thought that genetic selection was the sole thing about evolution, and that all the action should be reduced to genes and game theory.

    This is very simplistic and unfair to the views of these thinkers, but there have been three distinct movements that get called by the name “Neo-Darwinism”. I’m on the road, so I can’t check my sources right now…

The site is undergoing maintenance presently. Commenting has been disabled. Please check back later!