Same thing(s)

Richard Owen

It’s grim how many basics have been borked at the outset in biology curricula and To their credit, many colleagues of mine have fought hard to improve matters, and much has changed for the better, but unfortunately, it’s slowest to improve at the opening, first-year stage.

  • No, metabolism does not “reverse entropy.” Not even sort of. (It accelerates it.)
  • No, biology is not “the study of life.” (It is the study of the nonliving phenomena of which life is composed and by which it is altered.)
  • No, creatures do not “adapt or die.” (They all die, at both the individual and species levels.)
  • And no, homology is not “similarity in structure and function.” (Which is what this post is about.)

Not all first-year bio textbooks commit these crimes but enough did for long enough that the discipline was affected for the worse, to say nothing of the resulting deficit for the larger culture. Creationists picked up on the entropy one instantly and you can still find that talking-point in their yipyap, thank you ever so much, textbook authors. However, of them all, I think the homology failure is the most pernicious.

The founding principle of biology is that observed life is subject to material investigation irrespective of superadded forces and intrinsic purpose. The founding observation of biology is that of the same organ in different creatures. That’s what homology is.

So what’s the “same?” The materials and arrangement. See what the things are made of, tissue type or chemical composition, and see how, in each, the piece is hooked up to everything else.

Note: inference, not evidence. In both the science and the teaching, there’s some difficulty with the classic Darwinian “evidence for evolution” discussion, which lies in the interesting historical detail that the fellow who coined the term homology in 1843, Richard Owen, was not a supporter of Darwin’s ideas about natural selection (published 1859). However, the unnamed concept had been utilized in evolutionary thinking for decades by that point. Owen provided the name but wasn’t the first to make the observation and didn’t invent the concept or control its use by others. So just what the named term means relative to ideas about selection can’t rest comfortably on a canonical founding quote but has been a matter of reinterpretation and a certain amount of piffle ever since.

[There’s a whole post in each topic of vestigial organs and Ernst Haeckel’s depiction of vertebrate embryonic features, but for this one, all I’m saying is that they demonstrate homology as the bedrock of their respective meaning. Same goes for the fundamentals of cell theory and the discovery of DNA structure.]

The hassle shows up right away in the teaching. Look at the famous illustration, or rather, this one of the thousands of its iterations. What’s to be learned from it, again? Specifically that in this case the same organs show different forms and activities. A fine point as such, but which loses its punch when you fail to include that this is a very tight taxonomic cluster of critters with a startling, arguably disproportionately high degree of homology from species to species throughout the body, and pertaining to developmental and genetic details as well as anatomy. Mammals are all pretty much the same thing.

Then when you put up the equally-famous analogy picture from a broader span of vertebrates types, it confuses the bejeesus out of students and no wonder. Wait, these things which fly or dig or whatever, they are made similarly and that “proves evolution,” but over here, these things which all fly are made dissimilarly, and that “proves evolution” too? You see the problem, I’m sure …

… That it shouldn’t be about “proving evolution.” It’s about showing the comparative degrees of homology as you expand the scope of the taxonomic comparison, independent of function, which itself permits the compelling inference of evolution. And as I’ve observed in 25 years of teaching – outstandingly, Zen-strike compelling.

Unless you frame this whole thing in comparing understandable categories of creature and pointing out the differences in the presence and degree of homology, the insight is lost. I do not hesitate, again, to call this an intellectual crime.

Taking the limb to the wider range of tetrapods still isn’t good enough. We must expand the taxonomic range and the range of organs such that less homology is observed as well as lots, to permit looking at what it is and what it isn’t at the same time. One of my best examples concerns the main mechanisms of breathing in vertebrates.

Check out the respiratory anatomies here. See how that works? Gills and lungs aren’t homologous at all. Lungs and swim bladder are homologous.

A wonderful man. I miss him.

I’ll take this moment to make a quick evolutionary point, that specific organs are not one-shot do-or-not “chances” to do a thing. Counter to “they lost their lungs and only use gills,” many teleost fish still make use of air, with a variety of air-gulping and sites of gas exchange. So fish used lungs to breathe air, and some still do, and also, many fish that converted the lungs to something else now use air with different mechanisms. Meaning? That fish didn’t wiggle up onto land and start suffocating and gasping around all over the place until they “solved” the problem or “saved” themselves by instantly switching to lungs. The whole category of “breathing air” is not synonymous with land vs. water habitat. I here acknowledge the late great Karel Liem’s work in this topic, as well as appreciation for his kindness to me and many, many other younger scientists.

Now, hold that thought and look at this, which if marked on the above diagram would go right at the vertical bar third from the left. It’s the origin of all vertebrate jaws from specific gill arches.

Now this one, which is a closer look at one bit of one branch in that same phylogeny, the one labeled “lungfish and tetrapods.”

Important: the red and the yellow bones at the back of the jaws correspond to the upper and lower green bones in the above figure – the rest of the jaws (blue and orange) are different bone at this point.

See what’s going on? A front-ish gill arch of jawless fishes is homologous to (i) the original vertebrate jaws, (ii) some or a small part of most tetrapods’ jaws, and (iii) two of the mammalian middle-ear bones. [special nerd point here] So across the species there are the same bones doing various jobs and thus a given job being done by different bones. None of these structural switches per job means very much regarding the job’s excellence or efficiency or any other value-laden or goal-oriented word. They all hear fine and bite fine.

See, now we can talk about analogy and make sense. Not the opposite of homology at all, it’s about examining homology in given parts when you expand the scope enough, and when you do add in similarity of function as a nice touchpoint. There are two very different sorts.

eyesConvergence means the parts aren’t homologous. Remember, on its own that isn’t special; randomly pick some part on this critter and also randomly a part of that critter, and the overwhelming majority of cases aren’t homologous. In each case, though, the parts are doing something, so all we’re doing with convergence is cherry-picking the ones which are doing something pretty similar. What’s cool is how the exigencies of physics produce similar structures from extremely different working bits. You do not find convergence which shows the same materials and arrangement.

Here’s the famous comparison between the vertebrate eye and the mollusc eye. Yes, they’re both eyes, specifically chambered eyes, specifically pinhole camera eyes – but in all matters of material parts, in the most literal possible meaning of the following phrase, they are not the same thing.

It’s neither magic nor mysterious because again, light and animal nervous systems interact in a given physical way with a limited range of how lenses can be involved. Also, remember, lots and lots of creatures have light-sensitive organs that aren’t eyes – in other words, the seeing comes first, you’re not “building” an eye “in order to see.” Each eye is a refinement of an existing function.

Parallelism means the parts are homologous but they do not share a common evolutionary origin. Remember what I said about mammals? Suffice to say that the same parts and kinds of variation are present in most of the subgroups thereof, so seeing the same kinds of changes occur separately among them isn’t a shock – it’d be a shock if it didn’t happen. Sames goes for the hundreds of examples found within any species-rich group; parallelism is really common, unlike convergence. The outcomes are never wholly exact, but in many cases extremely close.

Each of these represents a different origin of saber teeth in mammalian carnivorans. There are even more among wider-flung mammals.

M. woermanni, parallel little booger

My special curse from such a situation arrived in the form of long-tongued nectarivory in megabats, when I was working on and writing my dissertation. I wanted to map genital anatomy onto their phylogeny – only to discover that at that moment, many publications differed over whether several genera, the ones with long pink curling horrorshow tongues, comprise a closely-related subgroup or not. In the early 20th century, they’d all been placed together on that very basis of their tongue and associated skull parts, but now most people agreed that Megaloglossus woermanni is a parallel over there in southwest Africa, and the others might be a solid group. Except no, some others said, nearly every instance of this funky tongue in these bats is its own evolutionary event. (Cue me making Edvard Munch’s Scream face, all I wanted was a freaking phylogeny, is that so much to ask?)

So that’s the basics for homology and its importance. Now for the interesting parts.

To repeat: genes influence developmental events, developmental events result in organs … You can probably immediately see that this is the larger framework for the diversity among human genitals for my previous post Hermes & Aphrodite. The same organs differ specifically because the developmental processes are different, as signaled by minor differences in genetic influence. Seems easy to treat homology as a happy “linear unit” across the levels, right? But!

Full stop. This raises big issues with “sameness” – you’d think that an earlier step in that chain is a great indicator of identity at a later one, but it isn’t. Painfully, slowly, the biologists most deeply-engaged in these issues have realized that you don’t use one level to justify or claim homology for another, you have to stay with materials-and-arrangement within each level, period. Fortunately, in practice throughout the 20th century, people switched to the developmental level and thought in terms of materials-and-arrangement there, while nominally discussing the organs, so most historical conclusions about it stand intact.

The obvious next question of what to do with differing degrees of homology across the levels for the same comparison of parts is quite a thing. It brings up a serious intellectual failing in the history of this topic: the mistaken definition of homology as common evolutionary origin. That they are often associated, yes, absolutely, even obviously – but one is not the other. Saying they are diminishes the logical foundation of the discipline. This claim or interpretation is still out there in the textbooks – I bet when some readers saw the criticism of “similarity in structure and function” in the early part of this post, they thought I’d be offering it as the solution. I’m saying it’s a deeper problem and needs to be scrapped.

We have a fine word for the shared evolutionary origin of parts, mouthful though it be: synapomorphy. The trouble is defining homology as synapomorphy and then claiming you’ve spotted a synapomorphy because of homology – not only is it simply circular (il-)logic, it forces the common phenomenon of parallelism into a weird “exceptional” intellectual space that front-loads phylogeny construction and borks the teaching. I’ll save further talk on this point for systematics-focused posts, for which you may shiver with anticipation.

Arriving at a rigorous meaning and justified scope of applications for homology is hard work – it requires gaining historical perspective, integrating new and old information, ignoring traditions and expectations within subdisciplines, and questioning everything about the way and why we teach. But if that discussion is not happening, always, then we’re not doing biology at all.

[much of this post is indebted to the work of Robert J. Richards and Brian K. Hall; the included judgments and any errors here are mine]

Next: The Other


10 thoughts on “Same thing(s)

  1. Just a little history question:

    “The founding principle of biology is that observed life is subject to material investigation irrespective of superadded forces and intrinsic purpose.”

    This is as it was presented to me in Grade 11, 30 years ago. But when did it become the archstone of biological theory?

    19th century vitalism gets pushed out of the laboratories, lecture halls, and text books around when? And does it still hold on in a few tidepools and backwaters of the mentalspere?


  2. The origin is explicit: William Lawrence’s lectures to the Royal College of Surgeons in the 18-teens, collected as Lectures on Physiology, Zoology, and the Natural History of Man (1819). They received intensive pushback in the public and opinion-making sphere, not least from John Abernethy, Lawrence’s former mentor, now outraged – he was a major supporter of the superadded forces model. Lawrence even had his copyright stripped for “heresy” (the courts reached way back into the books for that one) – which predictably resulted in open-source publications throughout the land. [edited to fix names – RE]

    You are very optimistic and idealistic to think that vitalism was successfully pushed out of labs, lectures, and textbooks. Thomas Huxley tried to do so single-handedly for four or five decades with limited success, and I think there’s been considerable backsliding since, mostly through omission. As far as the mentalsphere is concerned, I think material thinking is the one in the “few tidepools and backwaters.”

    But this is not the topic for this post. I’ll discuss it in detail later, particularly as it relates to teaching. After that I’d encourage you to ask the right questions of your own students and colleagues, and you’ll see.

    Liked by 1 person

  3. Hey Ron
    I’m enjoying these posts a lot. There’s a point which I’m grappling with. How does one show that two organs in different animals are in fact homologous, without assuming evolution? I’ll buy that gill arches and mammalian ear-bones are homologous, because “evolution”. But how do I get there without it?


    • Hi! That’s exactly what I’m talking about: the definition of homology cannot include reference to evolution or, more specifically, to synapomorphy. It has to rely strictly on observing the materials and arrangement of the parts. For anatomy, that means looking at tissues and cell types, as well as the placement of the part in relation to others – i.e., how it’s hooked up. With any luck you’re seeing that this is a matter of going both down and up in scale a little bit – so cries of reductionism (a topic I’ll deal with soon) need not arise.

      This applies to the ear and jaw bones perfectly. There are three types of bone in the skull, and the relative position of each one’s individual bones is apparent to both casual inspection and, for the funkier bits, close inspection. You don’t have posit evolution at all merely to investigate each one, and both bone composition and relative placement tag the ear bones not only as jaw bones in the other amniotes, but as these exact ones in the back, and as the entire jaw bones in, say, sharks..

      In other words, the whole point is that one doesn’t say, in teaching or at any other time, “the mammalian ear bones are old jaw bones because evolution.” The only rigorous phrasing is to observe the homology and then state the inference correctly: that the best explanation we have for such phenomena, especially given its varying degrees depending on which creatures and which parts are examined, is that parts have undergone historical change.

      Next, discussion of synapomorphy vs. parallelism is an evolutionary topic, so the criteria for one vs. the other is a completely different question.


      • OK, thanks! I think what I was missing was the degree of detail in which one could study organs and thus reasonably posit homology just because of the sheer similarities.


        • It helps to consider how different part-to-part comparisons can be across creatures, in terms of materials & arrangement. One doesn’t have to be subtle or to guess.


  4. Pingback: The evolution of evolution | Man nor Beast

  5. Pingback: What are little species made of? | Man nor Beast

  6. Pingback: Count Bacula, part 2: Oh, the diversity! | Man nor Beast

  7. Pingback: All-natural, all-artificial selection | Man nor Beast

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s