Ep 127: Devo-lutionary theory (with Scott Gilbert and Tobias Uller)

What role does development play in evolution? How can biology incorporate a broader, more complex definition of what it means to be an organism?

In this episode, we talk with Scott Gilbert and Tobias Uller, two of the authors who wrote Evolution Evolving: The Developmental Origins of Adaptation and Biodiversity. Scott is a professor emeritus in Biology at Swarthmore College, and Tobias is a professor of Biodiversity and Evolution at Lund University. In Evolution Evolving, Scott, Tobias and their co-authors (Kevin Lala, Natalie Feiner, and Marcus Feldman), argue that development plays an underappreciated, but central role in evolution. One concept they emphasize is of “developmental bias”, which is the capacity of organisms to generate certain kinds of traits and trait combinations more than others. Scott and Tobias argue that such developmental mechanisms can provide a more robust explanation for patterns of evolutionary change than traditional theory. However, they also provide a very broad definition of development in the book, and challenge readers to consider the complex interactions and feedbacks between an organism and the environment that shape an organism’s evolution. Could this be the start of a new devo-lutionary theory? Listen and decide for yourself.

Cover art: Keating Shahmehri.

Marty Martin 0:05
So, Cam read anything interesting lately?

Cameron Ghalambor 0:08
Actually, I just started reading Ed Yong's book: I Contain Multitudes. I bought it a couple years ago, and I recently found it in a box.

Marty Martin 0:17
Oh yeah, that's his book about the microbiome. How is it so far?

Cameron Ghalambor 0:20
Fascinating, even though I already knew a lot about the importance of microbiomes, the book is actually making me rethink Dobzhansky's famous quote that nothing makes sense in biology except in light of evolution. Now I'm starting to think nothing makes sense in biology except in light of the microbiome.

Marty Martin 0:39
That's surprising to hear you say- you are such a traditionalist when it comes to evolution. Does this mean you're going to start focusing on microbiomes in your own work?

Cameron Ghalambor 0:47
Not exactly, although we are doing a little bit of microbiome work. What this book is really making me think a lot about is a lot of my commonly held assumptions and definitions about really basic concepts, like, what is a genotype, what is a phenotype? And what do we mean when we say whole organism?

Marty Martin 1:05
Yes, score one for heterodox biology. I love it when exposure to new ideas challenges traditional ways of thinking along these lines. Some of our long term listeners might recall our conversation in season one with Fred Tauber. We talked with him about how the whole field of immunology is having to reassess some of its fundamental metaphors, such as self and non self, and even the difference between benign and beneficial and nasty microbes.

Cameron Ghalambor 1:30
You and art also spoke with Corey Moreau in season four about the cool relationships ants have with their microbes, from helping leaf cutter ants to farm fungus to the ubiquitous effects of Wolbachia on all aspects of ant biology. We really can't understand ants without understanding their microbes.

Marty Martin 1:48
So the bigger question, how do new ways of thinking enter biology and change the way we ask questions and interpret our results?

Cameron Ghalambor 1:56
Philosophers of science think about this a lot. For example, Thomas Kuhn challenged the idea that scientific progress occurred as a steady accumulation of knowledge. Instead, he argued for episodic revolutions, or paradigm shifts that break up periods of what he called "normal science."

Marty Martin 2:14
We've covered some of these ideas already on Big Biology, and I know I'm beating a dead horse, but I think Karl Friston's free energy principle might constitute one of those paradigm shifts. More people need to learn about it.

Cameron Ghalambor 2:28
They say Thomas Huxley was Darwin's Bulldog because he publicly championed Darwin's ideas. And I think you might be Karl Friston's bulldog.

Marty Martin 2:36
Yes, probably guilty as charged, but Karl and I might be wrong, which raises another complicated issue, what's a paradigm shift for one set of researchers might not be viewed the same way by others.

Cameron Ghalambor 2:48
True and perhaps the highest profile current example of this debate is whether we need to replace traditional evolutionary theory with a new, extended evolutionary synthesis.

Marty Martin 3:00
Some of our previous guests, such as Massimo Pigliucci, Eva Jablanca, Dennis Walsh, Sonia Sultan, and many others have argued that phenomena like epigenetic inheritance or phenotypic plasticity or organismal agency in niche construction challenge the basic assumptions and metaphors associated with traditional theory. In other words, a new paradigm shift is needed.

Cameron Ghalambor 3:20
But some of our other previous guests, like Arvid Agrin, Eric Svensson, Nick Barton and others, feel that traditional evolutionary biology, rooted in theoretical population and quantitative genetics, is flexible and general enough to incorporate these topics, even if they were left out of the models that motivate a lot of the modern evolutionary research that occurs.

Marty Martin 3:42
Our guest today, Scott Gilbert, Emeritus Professor at Swarthmore College in the US, and Tobias Uller, professor at Lund University in Sweden, are co authors along with Kevin Lala, Natalie Feiner, and Marcus Feldman on a new book entitled evolution, evolving the developmental origins of adaptation and biodiversity.

Cameron Ghalambor 4:00
Scott, Tobias, and colleagues are advocates of the extended evolutionary synthesis, but the focus of their book is on the importance of incorporating a broadly developmental perspective on evolutionary change.

Marty Martin 4:15
In Evolution Evolving. They argue that we need to think more deeply about how variation is generated and why living systems generate certain kinds of traits and trait combinations more than others. This is what they call developmental bias.

Cameron Ghalambor 4:28
Tobias and Scott argue that while we often think about development as a constraint, developmental bias can work alongside natural selection to provide a more robust explanation for why certain evolutionary outcomes occur.

Marty Martin 4:45
We also talk with Scott and Tobias about some of the general principles of development, like how it's modular, epigenetic and, perhaps most importantly, completely intertwined with the external and internal environment of the organism. They emphasize that development, in the very broad way they define it, has the unique disposition to generate and filter variation.

Cameron Ghalambor 5:05
They say that the interdependence between developmental processes and the environment means that organisms don't simply follow a linear path that can be predicted simply by looking at their genetic makeup. Instead, they argue that reciprocal causation and feedbacks bias the generation of certain phenotypes, which in turn bias the direction of evolution.

Marty Martin 5:27
It should be clear by now that Scott and Tobias's definition of development is much more than the nuts and bolts of what happens as an embryo transitions into an adult. Their form of development captures the broad range of physiological and behavioral interactions that occurs between the organism and its environment over its lifetime.

Cameron Ghalambor 5:45
For example, in the same way that developmental processes interact with the environment to bias the production of certain traits or characters within an organism, processes like niche construction affect how organisms experience their environments. When beavers modify their habitat by building dams, they bias both the forms of selection they experience and the phenotypes that are favored.

Marty Martin 6:09
Ultimately, much of what Scott Tobias and their co-authors advocate boils down to how we should think about biological causation. They advocate for embracing and appreciating the complicated interactions between diverse mechanisms, rather than viewing evolution as simply a product of selection acting on genetic variation.

Cameron Ghalambor 6:28
While we agree with many of the points raised by Tobias and Scott-

Marty Martin 6:36
We didn't agree with all their ideas, and some evolutionary biologists are likely to be even more uncomfortable with some of their arguments, but we think that diverse views that could be tested and vetted can improve our explanations. And of course, that's a good thing.

Cameron Ghalambor 6:49
So whether Thomas Kuhn would consider these ideas normal science or paradigm shifting science, time will tell.

Marty Martin 6:55
Indeed. And now that we've switched to Substack, you can give feedback and discuss episodes under the discussion tab. We'd love to hear what you think. So make sure you become a subscriber and share your ideas.

Cameron Ghalambor 7:06
I'm Cameron Ghalambor.

Marty Martin 7:08
And I'm Marty Martin,

Cameron Ghalambor 7:09
And you're listening to Big Biology.

Cameron Ghalambor 7:23
Scott Gilbert and Tobias Uller, thanks so much for joining us on Big Biology today.

Scott Gilbert 7:27
Great to be here.

Tobias Uller 7:28
Thanks for having us.

Cameron Ghalambor 7:29
So we're here to talk about your new book Evolution Evolving. So I'm curious, you know, a big endeavor to write a book about evolution. Can you tell us a little bit about the origins of the book and how it came about, and who your target audience is for this book?

Cameron Ghalambor 7:45
Do you want me to go first on this one Scott?

Tobias Uller 7:53
Sure, sure.

Tobias Uller 7:55
Because it kind of goes back to a big grant that Kevin Lala and I had on various sort of aspects of evolutionary biology. And then we had this idea that we should try to synthesize what we learned through that project by, you know, in book length format, it was too much to say to just be a paper. So that's sort of the origin. Goes back really a long time. So it has been brooding for a long time. And then eventually we sat down with Scott, and with Mark Feldman and Natalie Feiner, and tried to think about what to say and how to say it. Then now we're finally, finally, it's out.

Marty Martin 8:30
And so who were you writing it to? Was this a book mostly meant for your peers? Was this inspirational for undergraduates? Was this that next cohort of graduate students that will try to integrate all of these threads in evolutionary biology?

Scott Gilbert 8:44
All the above, all the above.

Marty Martin 8:46
Ok perfect.

Scott Gilbert 8:47
The book written for several audiences, and that was one of the difficulties. I think the main difficulty in writing this book was deciding what audience we are aiming at. And I think that that's why we have 100 pages of footnotes. It's so that you could read this book in a reasonable amount of time. But if you want the details, if you want the information about how the data was obtained, it's there. And so I think that I got involved, you know, I wrote a paper withJohn Odling-Smee and Kevin Lala in 2008 and this was a result of a meeting we went to when we I think we were both frustrated with the way things were going. And we said a, you know, the notion of niche selection and the notion of developmental plasticity are talking to each other. You coming from the evolution side, me coming from the development side, we have to, we have to at least put out a paper saying, "look, we should try to make bridges between evolutionary and developmental biology, because that's where a lot of the action is right now and and this is where things are going." Kevin and I were both very interested in process philosophy and developmental systems theory. We had been listening to these people, and we just thought, you know, they're merging our disciplines. And so we wanted to, kind of get the scientists involved, to merge the disciplines as well and and so this has been kind of a long standing interest of I think all of us here, and I think that a lot of it comes out of, you know- why do you write a book in the first place? You don't write a book unless you're frustrated or dissatisfied with what already exists.

Scott Gilbert 9:02
Yeah, yeah. So can you, can you explain maybe a little bit about the title- Evolution Evolving and what the sort of rationale was for coming up with this title?

Scott Gilbert 10:54
It's almost a play on words, because we're talking about evolution evolving, the actual processes by which organisms come into existence, and that those processes have evolved and are evolving. And then the discipline of evolutionary biology evolving, and that it is not a static entity. It is not something that was fixed in the modern synthesis in the, you know, 50s or 60s and DNA and whatnot. It's evolving. And so I think it plays on both the natural and the disciplinary notions of the word evolution.

Marty Martin 11:32
Yeah, yeah. It's a wonderful title. I mean, very witty. It was good that you spelled it out in the book why you named it that way, but it's a wonderful title in the past. And Scott, I don't know whether you would sort of claim allegiance to this particular word, but Tobias, I think you were part of some of the earlier writings where the words extended evolutionary synthesis were used. And maybe there's something specific about evolution evolving that you would want to distinguish? But I was struck, and Cam mentioned this too, that that extended evolutionary synthesis, I think, was only mentioned once or twice in the entire book. So what's the reason for the change? Was this deliberate or just we moved on?

Speaker 1 12:12
No, it was deliberate when we talked about this, how to sort of frame this, right? And I guess the authors have perhaps slightly sort of different takes on it, but the way that I see the extended evolutionary synthesis sort of concept is that it's like an umbrella term for a lot of different kind of ways of making the organism central to evolution one way or another, right? But of course, being something that people have different sort of ideas on what it actually is, it's a little bit risky to kind of frame it in that way, because it means that people come in with particular sort of views on it and think that this is what it is, right? And I don't think it is a thing. So that's one reason. Another reason is that some people, for example, Gerd Muller, would say that the things that we cover in the book, so the developmental basis of adaptation, biodiversity, is just a subset of the things that go into the extended evolutionary synthesis. So from that perspective, of course, it, it is sort of perhaps literature in the vein of that kind of tradition, but it isn't sort of about the synthesis itself.

Cameron Ghalambor 13:17
Yeah, I mean, so, you know, kind of getting into the book. Now, like if somebody were to ask me to summarize the book in one sentence, because you cover a lot of material. I guess I would say something along the lines of basic argument is that all of the things that happen during the lifetime of an organism from, you know, an embryo to an adult have implications in the evolutionary process. What do you think of that summary?

Scott Gilbert 13:38
I think that really the one sentence I say in the book is that developmental mechanisms are central to an organism's capacity to evolve, and therefore take on much greater significance in evolutionary biology than we had assumed. And you know, philosopher in Vienna, Anne Sophie Meinke has said that adding development to evolutionary theory is not an add on, it's a whole restructuring. When development becomes central to evolutionary theory, it transforms evolution from being a biology of things, organisms, genes, populations, to a biology of process, development, symbiosis, metabolism, so that it really does change what evolution is? And I think that, you know, we have like five different parts of development in here. We talk about developmental plasticity, we talk about bacteria-induced development, we talk about epi-allelic inheritance. We talk about niche construction. We talk about evolvability and evolutionary bias. All of these things are development coming into evolutionary theory, and totally, I think, restructuring it

Scott Gilbert 13:47
Right, right? And you know, people that listen to the show will know very much that Cam and I are huge fans of these ideas, and especially the sort of process thinking, as opposed to "thing" thinking. You said something though, Scott, that, I guess I want to push you on a little bit, because it's a very particular word set that I preach to my undergraduates, that we have to be careful. You said that when organisms evolve, or something to that effect, those aren't exactly your words, but

Scott Gilbert 15:37
Populations evolve, organisms develop.

Marty Martin 15:40
Ok so I just wanted to make sure you meant something. This was mentioned a couple of times in the book, so that the same phrasing was used, but it was not used in the sort of, you know, intentional way. I just noticed it there. So I just want to make sure. But I wanted to bring that up because I would like to hear, and I'm sorry to sort of jump to the end of your story, and then we'll walk back through what is development and all of that, but, but let's just sort of maybe try to do the biggest possible thing. When you say that your motivation in the book is to infuse evolutionary theory with development, it feels like you almost stopped short of what I thought the extended evolutionary synthesis was supposed to be. And Tobias, you spoke to the point that, you know, Gerd Muller and others may say that what you're trying to do is a facet of the extension. But do you sort of think that there is the possibility of synthesis? So focusing on the organism and the components of your book really sort of lends towards the possibility that we could have a cohesive theory of biology is that the kind of thing that you're aspiring for, what's what's the difference what's missing from what you're trying to do, and what that kind of all inclusive theory would be, if such a thing is even realistic. I mean, maybe it would just say that, let's give that up, that's not possible, and that's why you're settling for this more modest current position, at least.

Speaker 1 17:01
Well so I don't think there is going to be a theory of evolution, right? I mean, it's not going to be like a particular, concrete thing that you just sort of take off the shelf and you apply it and then you're fine. So here's my take on, on, on what, what people are trying to do. So they it's an incredibly complicated reality out there. Developmental process are enormously complicated. If you just look in a cell, you know, there's so much stuff going on that it's mind boggling. If you zooming out and looking at the ecosystem, it's crazy as well. You know, a handful of soil has, you know, several thousands worm individuals and other kinds of creatures, and thousands and millions of bacteria and kilometers of high fungal hyphae. It's incredibly complicated. The entire ecosystem just in a handful of soil, right? So that reality that is so complicated needs to be handled by making representations of the world that are much, much simpler and abstract and idealized in various kinds of ways. That's how we do science right now when it comes to evolutionary change, and now those representations, they basically give rise to different theoretical frameworks, like, for example, population genetics or quantization, genetics, adaptive dynamics, whatever it could be, right?

Tobias Uller 18:11
So one thing that's really important, I think, to realize here is that evolutionary change, you said that before you know populations evolve, and of course, they evolve and organisms don't evolve. But that's not entirely true, right? Because, you know, organisms do evolve, depending on what you mean with the organism, individuals don't evolve, but organisms certainly evolve. So, you know, you can you zooming in. So first you're zooming in to try to figure out, what are the mechanisms, what are the process they give rise to something new? Well, those are the things we call developmental processes. We have actually define them as the process that give rise to phenotypes, right? So those happen within the life cycle of an individual. That's the right domain for trying to understand where new stuff comes from. That's the dynamical process that gives rise to that new stuff. Then you have to zoom out, because organisms come in populations. So if you want to, you know, it doesn't matter if something is new and functional, it has to actually spread and become the new sort of standard and a stepping stone for something new again, right? So it has to become prevalent in the population. So then you're zooming out and trying to figure out, what are those things then that make this prevalent, right? One thing, of course, is natural selection, the course of relationship between a feature and its survival and reproductive success. You can zoom out even further, and populations and individuals disappear out of sight, and the only thing you see is lineages that split, then you got transformation of one thing to another. So all of these different aspects are important, right? And if you say, is there a theory that can handle all of these different aspects, all that causal complexity for understanding evolution, my answer would be no, that is not really realistic. So we need multiple different representations to pick out particular kinds of courses, particular kinds of process that are contributing to the things that we want to explain, adaptation and biodiversity and so on.

Marty Martin 19:54
Okay, well, we're going to circle back to this at the end, when we tie up. But Scott, I imagine you have a perspective on this you'd like to share now?

Scott Gilbert 20:03
Yeah, yeah, no, I think that what Tobias said is absolutely the case. And I think that one of the things that we were trying to do, and it was explicitly stated, like probably in the preface of the book, is we need to go back to one of the assumptions of the modern synthesis was, which is that you don't need proximate causation to understand evolution, that you don't need to have the mechanisms of development by which you get the phenotypes as a mode of explanation in evolution. And we're just saying that this is inaccurate, it's outdated. It's from, you know, the 50s and the 60s. You know, Dobzhansky could say in the 50s that evolution is a change in the genetic composition of populations. Well, it's more than that. He says that, you know, the study of the mechanisms, and he's talking about the mechanisms, the study of the mechanisms of evolution falls within the province of population genetics. And we're saying, well, not totally. And so, you know, one of the philosophical things that Kevin has, he goes back to this notion, the paper of Ellokosky and Korokhovsky in Finland, who are giving five criteria by which alternative explanations can be evaluated, and one of them is precision. Another one is non-sensitivity. A third is factual accuracy, a fourth is degree of integration, and the fifth is salience. And I think that what the book is saying is, you know, on at least four of these five things. If you add development to the mix, you get better accuracy, you get better integration, you get better precision, you get better non-sensitivity. And so the only thing that this salience, it's it's harder to understand, yeah, yeah. But it's a difficult but the other four, I think he's saying, yes, you need development to get a better explanatory theory of evolution.

Cameron Ghalambor 22:11
Yeah so I think this is a good place maybe, to define kind of what it is that we mean by development. Because I think you know, one sort of representation of development, maybe is outdated, is just thinking about the processes and mechanisms you know, of an embryo developing into an adult and and I think in your book, you define development as the process of progressive and continuous change that generates complex phenotypes. Is that, is that a fair sort of definition that you would, you would ascribe to, and because that, that seems a little bit more broad, and I don't know, encompassing of a lot of different processes than just, you know what genes get turned on and off as cells divide and tissues form, and, you know, different layers, you know the way that I kind of remember taking developmental biology. So, yeah, can you explain a little bit more about that, that definition and, and, and why you maybe chose this, maybe broader perspective?

Speaker 1 23:26
I just want to say something before we let the development of biology team in here properly, right? It's a good question. It's a really good question to the author of the best textbook on development of all the currently in its 13th edition. Don't you get your development definition wrong. This is, this is a really, really tricky one, because it doesn't matter how many times you say in talks and so on. I use development to mean the processes that bring phenotypes into being, no matter if they're physiological or morphological or behavioral. Then people, some people, think development is egg to adult, and then they'll, they sort of won't get what you mean. I wish there was a different word for it. You know that we could just use a word that means the process that brings phenotypes into being, because it becomes really tedious to say that all the time. And there are people, for example, some people in Evo Devo that uses development in this broad, broad way. So it's not entirely, sort of taken out of thin air, but I do recognize, and I am well aware, that this does cause problem, because development means different things to different people. And over to you, Scott, when it comes to this.

Scott Gilbert 24:31
Yeah, yeah, development certainly means different things to different people. We tried to get a book cover from a radical environmentalist, and we told him it was for a book called ecological development. And he said, I don't give books about anything to developers. To him, development meant bulldozers,

Marty Martin 24:54
Yeah

Scott Gilbert 24:55
Okay, you know, we have an office in our college, you know, called the development office, and it's all about getting money has nothing to do with development. You know, okay, so development is a metaphor that's unfortunately too widespread. Now I agree with the definition here, which I didn't write is, interestingly enough, Kevin put that in. But the whole idea of development is changing, and what is development is changing. And that was reflected in the book. And I like to think of there being three movements that we talk about. And, again, what Tobias was talking about is, you know, there's the movement of metabolism, the physiology of the organism. Now that's metabolic movement. There's evolutionary movement, you know, from, you know, lineages splitting and so forth. And then in between them is development. And I've done a kind of calculus metaphor, which is really strange, because I married my calculus tutor. I'm not kidding, but you know when you have metabolism, you know anatomy and physiology, and you take the first derivative with respect to time, d metabolism over time, and that's development. And then you take the second derivative, development over time, and that's evolution. That evolution is the selectable changes of development over time, and so you get a very broad notion of what's development now, one of the problems is there's no such thing as metabolism, development or evolution alone, because what we found is that metabolism is cometabolism, because the body is this whole of ions, composite organism. So we have cometabolism, not metabolism. We have codevelopment, because the symbionts are actually involved in their life cycle, becoming part of our life cycle. So we have codevelopment. And of course, we have coevolution as well. And so all these things that were thought of unitary streams have all these tributaries feeding into them. It's so much richer than we thought, but so much more complicated. Because the metabolic stream, the evolutionary stream, they all fit into the developmental stream. It's very difficult to separate them. And I think that's one of the interesting things the book does, is that it recognizes that this is not something that's easy to separate. And you say, you know, it's phenotype production. And I agree with that definition and phenotype production is more than the readout of the genes.

Marty Martin 27:41
Well, I think all of us are on board with that broad conviction, but it is the breadth of the definition. And sorry, Tobias, because I know you've heard this many times before, but it's the breadth of the definition that's difficult for me. I mean, maybe it's our own biases, and we are going to talk about developmental biases, so this is a bit strange, not that kind of bias, but my bias as a physiologist. Scott, you just talked about derivatives of physiology as development. I'm fine with that, but you started from physiology. So is physiology and development the same thing? I mean, it is in your definition, but it's definitely not in my mind that I don't think it's defensibly as such, but so I guess what I'm where I'm coming from, is that you're almost saying that development is life, and that's refreshing, but it's so expansive and inclusive that I don't know what we do with that.

Speaker 1 28:30
Yeah, no. I mean, you're right. So, so one to go back to, I think, what, what cam asked before about, sort of, what's the what's the book, you know, in a sentence? It is roughly that it matters to evolution, that organisms are organisms, how they work, is actually really important. Now we have chosen to focus on the processes that bring phenotypes into being, and we call that development. You could have written very much the same and focus on the physiology, and say, "this is what I mean by physiology." Okay, maybe there are some things that would have fallen outside, because you wouldn't want to push your physiology definition too far, perhaps, but you could, in a way, then replace development by saying developmental, physiological and cognitive processes, if you so wish. This is a bit problematic, of course, because it is easy to get lost in the different definitions and what people actually mean. But it is, I think you're right, that the problem is that we're trying to grasp and trying to cover that it's important that life is actually life, and that is sort of the key.

Marty Martin 29:28
And it's, it's crazy that we have to emphasize that, as biologists, but it's always blown me away that we can talk for hours and days about evolution and never mentioned oh yeah, these are living systems, by the way.

Scott Gilbert 29:38
Yeah, yeah. Now, one of the interesting things I find is, you know, when you talk about physiology and development, it depends which Kingdom you're studying in. You know, in animals, development and physiology became separate disciplines. And development was originally developmental physiology. In 1894, Wilhelm Roux developmental physiology, as opposed to adult physiology. And that these were two separate disciplines studied by different sets of people who had different hiring tracks, different journals. That didn't happen in plants. Plant physiology is studied by plant developmental biologists and verse and vice versa, that plants with their plasticity, plants with their symbiosis, plants with their coming into being each season. There's no difference between plant phys and plant development. You want to study auxin. It doesn't matter what you call yourself. And so I think that that's a disciplinary thing, that this is, you know, the structure of the German Academy, it's not something about nature and how nature divides things. It's how we've separated things into, yeah.

Marty Martin 30:56
Yeah what your brain has done.

Cameron Ghalambor 30:58
Well, I would even add plant ecology and plant physiology are also much more integrated than animal ecology and animal physiology are

Marty Martin 31:08
Intertwined totally, yeah.

Scott Gilbert 31:10
Right.

Cameron Ghalambor 31:15
You talk in the book about five general principles of development that it's it's modular, it's epigenetic, it's constructive, it's interchangeable, and it's biased, and the bias, the developmental bias, plays a very prominent role in in the book, as as something that has a very strong influence on on evolutionary processes. So can you expand a little bit on what it is that you mean by developmental bias and, and, and how that maybe that idea differs from historic perspectives on development and its role?

Speaker 1 31:57
In a way, it just means that the if you have a developmental system, whatever organism you're dealing with, and you perturb it by changing something in its genome, for example, or changing something about its environment, it's going to react to that potentially, right? And what's going to happen is it might, might change, but that change is going to be structured. It's going to somehow be structured according to the rules of development, if you so wish. And that means that the distribution of phenotypes that you can get actually depends very much on that developmental system, it's internal organization, it's it's material instantiation as well how things are actually connected and so on. So there will always be some kind of distribution of phenotypes that are easy to produce, that will be frequent, and some that will be rare, and some that will be impossible. It's kind of means a trivial point, because, of course, that is the case. I mean, no one would, would, would ever claim that you can mutate a bluebell and you could get anything. It's going to get the bluebell that looks a little bit different, of course. But it is that is as simple as it is, it is just development. In a sense. It's just the fact that you actually get structured responses somehow that are reflecting something about the properties of development, the way that the system is organized and so on.

Scott Gilbert 33:18
Yeah, yeah. Certainly going back to the notion of phenotype production, it's which phenotypes can be produced and how readily can they be produced? One never sees eyes in the back of the head. This would be selectable. I mean, if only for teachers, it would be great to have eyes and parents, eyes in the back of the head would be wonderful. But the developmental systems for making eyes only exist in the front of the anterior part of the body, and so we can't get fingertips with eyes. We can't get, you know, eyes in the back of our heads, even though they might be very selectable. And this has to do with, you know, the mechanisms of development. Simply, we never see mammals with six or eight limbs. That might be really good to have, but we don't see them because the developmental mechanisms cannot make limbs in any part of the body. They make limbs only where certain genes are being expressed. And Pere Alberch, wonderful article many years ago, had a paper called The Logic of Monsters. And that's basically saying you can't get every type of monster, the monsters that you see, the mutations that you see, allow development to occur only in certain ways. And so he was mainly focusing on constraints. But we know that, for instance, it's relatively easy to get webbed feet because you know that we already have a webbing and we have to, if we can stop it from being degraded, we can get webbed feet. It's easy to get beaks. It's easy to get structural changes in our faces, because the system for making more divisions in one direction or another are already there. So I think that the bias is certain phenotypes are much more easy to make than other phenotypes.

Cameron Ghalambor 35:18
Yeah, so I mean, I think that that seems to me fairly uncontroversial, although in the in the book, and you mentioned this, Scott, the idea of constraint and you sort of pose developmental bias versus developmental constraint as kind of a dichotomy where in some cases you have bias towards certain types of phenotypes, and in other cases, you know, that constrains responses to selection. And you know, I remember many years ago reading John Maynard Smith et al's quarterly review paper on developmental constraint. And what I took away from that was the, you know, development was a more of a continuum. It was, in some cases, it was a constraint, like certain combinations just couldn't happen, like eyes on the back of our heads, but in other cases, because of that underlying developmental system, certain phenotypes were actually facilitated more and could more readily respond to evolution. So, so is it really a dichotomy, or is it? Is it more of a continuum?

Scott Gilbert 36:29
Well I see constraints as a portion of bias. I see it as a subset of bias, and that and constraints could be constructive. It doesn't mean that, you know, you're just limiting things. You can say, okay, you know, gravity is a constraint, but that allows us to make limbs of all sorts of things, and you know, ways of dealing with it. So I think that, you know, seeing a constraint as just a restriction isn't correct either. But I think that you have the biases, and within the realm of biases are can the notions of constraints. And I, you know, again, you see the constraint on wing development in Drosophila, and then you say, Yeah, but if you modify that, you get all sorts of interesting shapes, and you get all sorts of interesting adaptations, you know, if you make this part bigger and this part smaller, you can play with it, and the environment can play with it. And so you can get ants with different casts due to this pattern of wing development, which is very set. So, you know, again, it's, it's not an either or thing. And making all sorts of new possibilities possible thing.

Cameron Ghalambor 37:44
So, so you you would say that developmental bias is a, is a, is a broader, more encompassing term, and that constraint would be subsumed within that, as opposed to more of like you can have very strong bias or very strong constraint as part of the the same kind of the continuum, right?

Scott Gilbert 38:08
That, that's the way I would see it. If I would, you know, looking, if I were writing a chapter, I'd have developmental constraints, and then a developmental bias, and then a subset be developmental constraints, okay, okay? Other people, other people might see it very differently, though.

Speaker 1 38:24
Yeah. So I think that there is always bias, as you said, I mean, it's entirely uncontroversial, of course. So whether the developmental biases constrain or facilitate evolution depend on what you compare it to. So often we tend to think, we tend to compare, implicitly and sometimes explicitly, any kind of structured variation from, say, an isotropic variation. So if you think about, this is very abstract, but if you think about an adaptive landscape, okay, that you used to sort of thinking about an evolution of all you got your various kinds of fitness peaks and so on, and then you have your morphospace, and you put that on top of the adaptive landscape. Now, a particular kind of bias would be one that aligns pretty well with the landscape, that would then facilitate evolutionary change. Another kind of bias could be that you actually prevent it from reaching the fitness peaks, and that would be a constraint. So, you know, constraint and facilitation are relative to what you're actually comparing to, and we're implicitly, somehow, often very, very much sort of having as the baseline that anything is possible or that you can always get it. But you know, there isn't really a case where everything is possible. There's always some kind of bias.

Marty Martin 39:38
I mean, I might be asking you a big question, but that's my tendency, so apologies in advance. You know, the aspiration, I think, would be to use historically successful things with, you know, better, modern ideas. So let's just take as a given that the developmental bias is something that we want to try to account for as we understand evolution. How do we put it into existing modeling frameworks? I mean, so you write in the book some things about the G matrix and its limitations with respect to, you know, we're talking about with bias. So maybe it's not possible to use those frameworks. But have you thought about a way to sort of take into account that eyes will only show up on the front of the head when we're trying to mathematically model the evolution of populations? I mean, this sort of fundamental thing is the difficulty that I think a lot of people got frustrated about when it came to extending evolutionary syntheses. So what do you think about that part, Tobias?

Speaker 1 40:35
So one way to do that, of course, is to couple, say the population level perspective, maybe quantitative genetics is a good framework, in this case, with a mechanistic representation of development. So that you allow the mechanistic representation of development to actually produce the phenotypes that can exist, or that do exist, they come into being, and how frequently they come into being. That sort of when you're assuming in on that level, where you say, here is the domain development is the domain of generation of variation, and then you're zooming out and saying, "Okay, now I'm going to generate these phenotype distributions. They come from the developmental process. And those phenotype distributions can then be shaped by selection, because some things don't work and other things do work." And those kinds of models do exist. There's some very nice work by Lisandro Milocco and Isaac Salazar Ciudad, for example, that used the famous tooth model in Evo Devo fromJukka Jernvall's lab to demonstrate how you can do this sort of coupling between a developmental mechanistic model and a quantitative genetic model. And it is possible we are, in fact, Lisandro and I are working on some ways of formalizing this beyond the mechanistic, specific mechanistic models where you can, you can perhaps say something a little bit more general about these relationships.

Cameron Ghalambor 41:45
Yeah, you know, so I struggled with this issue years ago. I was studying how pregnancy influenced the capacity for fish to escape from their predators. And I was interested in how this behavior evolved, you know, how could you become faster. And, you know, from a sort of quantitative genetics perspective, I was thinking, "Well, you know, there's genetic variation and covariance between different traits, but that could tell me something about how the traits would evolve," but I you couldn't really capture this fundamental trade off, that if you produce more babies as a female, you had to carry those babies, and that would slow you down. And there was this like sort of physical constraint. And so, you know, the solution, I thought, was you simply just add another matrix of of how traits are biased and or, you know, facilitated or constrained, and then that's, it's the combination of the underlying genetics with the sort of functional trade offs that then determine the response to selection. Would you see that as like, one way to incorporate this into these ideas together?

Speaker 1 43:01
Well, if you want to, if you want to, somehow include the generative processes, the developmental processes that give rise to variation, you have to represent them somehow. And in a way, it's not that quantity genetic doesn't represent them because, you know, they have something called the M matrix, you know, which is the variation you fuel by getting from mutations. And they make particular assumptions about development, which means that they can generate that matrix, you know, without having a specific model system in place. But those assumptions are, of course, not really the kind of assumptions that you would probably want to live with if you're actually interested in the generative processes themselves, then you would probably want to represent development in a different way.

Tobias Uller 43:41
So one way, as I said, is to do that, is to do a mechanistic representation of teeth or Drosophila wings or whatever it is, and see what is the variation that this mechanism can actually generate, and how does that translate into phenotype distributions that selection can act upon? And how does that sort of then generate evolutionary change? There are maybe other ways of doing things, for sure. It would be good if not everything was a special case with a special mechanistic model. And for sure, there are general properties of development, right? People, sometimes this might be a good way to bring up something that is kind of interesting, because I think Scott has experienced this as well, where people say, "Well, isn't just development just making things awfully complicated? Because it is crazily complicated, and everything's a special case. You know, you have to just know so much to be able to parameterize any model and build any model. So it's just like, everything's just a special case. What's generalities?" And that is true, right? Because that is one of the problems, is the development is incredibly complicated. But ecology is also awfully complicated. Ecology is crazily complicated as well all these organisms running around, doing things to each other, but we have found ways of encompassing that where, you know, it could just be one minus s, right? Selection. So surely, there are ways of representing the developmental process in ways that are useful for understanding the role of development, and there are such models. Can even do the population genetics now, Arlin Stoltzfus the sort of arrival bias, mutation bias, which is, you know, the ways in things appear. Art Lewis has really nice work on this as well, both empirical and theoretical, showing that the arrival of things is really important for the evolution of things. So you don't necessarily, just because development is complicated doesn't mean that you have to throw up your hands and give up.

Scott Gilbert 45:32
I also think that, you know, one of the assumptions we're making is that mathematical modeling is the be all and end all of science and that, and that and that, a science, a discipline, is scientific to the extent that it is mathematical. And that's a supposition that pervades science. The undergraduates come in thinking this, that you know math is the highest and physics, then chemistry, then biology. And then, if you want the social sciences, you know, which are all murky the and I think that one of the reasons that the modern synthesis took off the way it did was it made biology mathematical. It took biology out of the Ag schools, you know, and made it into a science. And so I think that, you know, we have this bias that to be scientific is to be mathematical and to have mathematical models and and I think that the mathematical models should give some legitimacy to other models as well other more, because when you model something, you have to abstract. I mean, that's the definition of a model that you're abstracting things, and you have to make assumptions as to what is possible to abstract without losing what's important. And I think that one of the problems with the modern synthesis is it abstracted developmental biology, that's not important. It abstracted paleontology that's no longer important. And it abstracted really matter. It abstracted the mechanisms. And so I think that's what we're trying to bring back into this very abstract notion. And I think the cover up of our book is really interesting. I wrote an email to say, at first I didn't like the cover, but now I can appreciate it more, because it looks like the squares are kind of like Mondrian squares. And Mondrian was saying, you know, mathematical representation, representation is superior to the physical object. And then what do we have in modrian squares? We have these naturalistic representations of real organisms. And so it's kind of combining the formal with the natural, which I think is something that ecological developmental biology, but I think in general, bringing in development into the synthesis is trying to do. It's to it's to say, hey, you abstracted us away. No, that was a wrong move. Let's put it back.

Cameron Ghalambor 48:07
Yeah. I mean, I'm sympathetic to that view, but at the same time, I mean, just again, to follow up with what Tobias was saying earlier about, you know, ecology being complex, development being complex. I mean, the power of some of these models is that they are abstractions. And, you know the saying, I think maybe this came up earlier:All models are wrong. Some models are helpful. You know, are you capturing the most important parts? And so, for example, you know, if we think about the response to natural selection, there are a lot of mechanisms that are left out in sort of quantitative genetics models of, you know, selection acting on genetic variation, but they also do a pretty good job. I mean, animal and plant breeders have had a lot of success, even before we knew anything about DNA, in selecting for amazing diversity and form and function without any detail of the of the underlying mechanisms, genetics included. So so, I mean, I guess, yeah, it seems like, you know, there has to be some somewhere in between, a balance.

Scott Gilbert 49:25
Well, you're, you're bringing up the breeders. I mean, you know, I love to talk about cows, hey, because hey, I think because the cow is so obviously a holobiont, nutritionally, but also developmentally, because the bacteria help build the rumen that will house it, house the bacteria. So I like the cows. But if you want to go back to cow biology, you go to Sewall Wright. And Sewall Wright got his information on drift and small. Population bias, founder effect from looking at animal herds, cattle herds. And he was the chief animal husbandman of the US agricultural service, okay? And I think it's just fascinating that if you want to talk about cow evolution, you could talk about it in the genetic terms that Sewall Wright did, but you could also now look at it in terms of it being a holobiont, and the cow doesn't exist without its bacterial component. And so that abstraction of, you know, if you and if you want to talk about evolution, you have to talk about the evolution of herbivory. How did an organism become an herbivore? How did the world become its food court? I mean, this is a really fascinating problem, and so you have a whole redefinition of what your problems are in evolutionary biology by going yeah, the genetics thing worked well for one part of it, but you have to bring in the development to understand a whole other big part, which is, how did you get a rumen in the first place? How did you become a ruminant? And how did ruminants become success, a successful mammalian clade?

Marty Martin 51:14
So I think that, I mean, I'm on board with you, Scott, but your position on the utility of you know, there's no reason to platform mathematical models as a sort of be all and end all that. That is definitely a little bit overblown. But there is this space, I'm sort of with Cam, in the sense that there's the space between the models as we use them in evolutionary theory and points that you guys make in the book. Let's talk about evolvability, because I think this is a really this is really nice potential connector between the history and the future. As regards development, you say specifically that researchers should study the propensity for developmental systems to vary, their variability, in addition to the observed state of variation. I know that's very hard to get one's head around as I just spout it out, but, but that sort of helps us to point to what we may want to describe and then lump in as parameters to models, right? I mean, these are the little engines of a sort for evolvability. Or did I? Did I misunderstand that?

Speaker 1 52:15
No, I mean, so, yeah, okay, so let's, let's just take one step, step back. So evolvability means different things to different people, right?

Marty Martin 52:25
Yeah, please define that. I'm sorry. I just brought on a big term without defining it.

Speaker 1 52:29
Yeah, that's right. So, so this has to do with this what I what I often refer to as zooming in and zooming out. So you're zooming in on the developmental processes to understand the generation of variation, you're zooming out at the population level to see and understand why some things persist and other things disappear. And if you're at that sort of population level, where I guess that most of evolutionary theory and most of evolution of all these feel most comfortable, then evolvability looks like roughly something of how much standing at the variation is there in the direction of selection, if you're thinking about some multivariate trait. But if you're zooming in on developmental processes, then it is the evolvability is more something like along the lines of, to what extent can this system generate novel and potentially functional forms through, for example, genetic mutation? You know, genes, of course, being fundamental for making sure things are heritable, so the major source of evolutionary change.

Tobias Uller 53:25
So in a way, all that variation that you see at the population level, of course, in a comes from that developmental process. So somehow these concepts of evolvability must be linked, but they are not really exactly the same concept. So that, I think is sort of important to remember, yeah. And then, and then, what you said then, was that, okay? Well, isn't that sort of a way of seeing what we need to model? Well, that's right. because if you're interested in, say that you're interested in the repeated evolution of similar forms, convergent evolution or parallel evolution. One potential explanation for that is that ecological conditions are such that the same things get selected in all these different organisms, different lineages. Another potential reason is that certain things will get thrown up much more commonly than other things. So the developmental process is generating structured phenotypic variation in such a way where it facilitates the evolution of particular adaptations rather than other possible things. So there you clearly see that if you would be interested in the conversion of parallel evolution, you might want to zoom in on developmental mechanisms and say, I can understand this by studying the developmental mechanisms, because this is sort of part of the evolvability. Are some things more evolvable than others? But to answer that question, I need to actually zoom in on the developmental side of things. I can't just be at the population level things, because that is not the proper place to be if I'm interested in development.

Marty Martin 54:49
Well, I guess my question is, and maybe you are saying it, I'm just not hearing it. If there are, sorry, I'm going to use an example that's close to home, because it's the one that I can articulate the clearest. Is, if you have regulatory regions in the genome, sort of say, let's talk about gene promoters, and heterogeneity in those promoters, in their disposition to be methylated, right? And so methylation is going to influence gene expression in some form, that's phenotypic plasticity. That is one of these sort of repositories of a sort for evolvability, in the sense that environmentally dependent methylation can induce plasticity and that can evolve, right? So are these the kinds of things that you're talking about? I think that this distinction between population and developmental thinking, There's something more about it that I'm not I'm sort of not latching on to. I mean, that sort of seems to be a cohesive point of interface between the evolutionary perspective and the develop developmental perspective.

Speaker 1 55:45
Well, it seems what you referred to were was a particular aspect of development, or the or the developmental genetics, perhaps that is potentially important for generating the capacity to well, is potentially important for the capacity to generate new phenotypes, and that would then be a part of evolvability in this developmental sense. And the development is sort of absolutely full, developmental process are just full of these different kinds of mechanisms that of course, influence the extent to which environmental or genetic perturbation produces change, but change that is also potentially functional. So it can be sort of, and

Marty Martin 56:20
I think this is the physiologist in me, because when we started talking about the staggering diversity of developmental mechanism, there's no argument about that. I could say the same thing for hormones or cytokines or, you know, all of us have this staggering diversity. But these kind these are categories of factors, right? So we can pile lots of different things into these categories, and potentially, I mean factors that drive homeostatic loops, or they're the homeostatic loops themselves. So it just seems to me that despite this giant universe of variation, we may be able to come up with categories that are doing somewhat distinct kinds of things and putting some attention to those. And that's the potential interface point between this large scale evolution and the development that you guys are emphasizing, I guess I'm just not daunted by it as much as maybe I should be.

Speaker 1 57:07
No, I think you're right. I mean, this isn't, this isn't what you're saying. Exactly. What makes it possible to actually get a handle on this is that despite all the complexity in terms of the nitty gritty details, there are certain coherent principles. You know, there are motifs, there are feedback loops, there are things there that are really, truly quite general to how organisms work. And those generalities provide, I think, opportunities for us to develop an understanding of the role of development in evolution that isn't going to be everything is just a single you know, everything is a unique case study, just like there are general principles about ecological interactions that allow us to extract from that and say they're actually general things here going on. So I don't need to represent all that ecological interaction in nitty gritty detail. I can take it and represent in a simple way. And I think you're absolutely right that both the generative processes and ecological interactions that give rise to selection, have these, have these properties that make it possible to be a scientist, right?

Scott Gilbert 58:09
Yeah. I mean, I in working on the turtle, one gets really into questions of evolvability, because, you know, they said, you know, if you're in the late Triassic, you know, you wouldn't have put your money on the turtle surviving, you know. I mean, no other mammal for that matter. But anyway, the turtle has, you know, scutes, and no other organism really has scutes on their back. And we looked into these scutes, and we found all sorts of things involving paracrine factors and whatnot. And the reviewers of our paper says this cries out for a mathematical model. And given that we were working in Jukka Jernvall's lab, and Isaac was down the hall from us and so forth, yeah, we could, and Isaac said we could do this in a month. Several months later, they got a mathematical model for scute formation, which was amazing. It was a reaction diffusion mechanism coupled with a second reaction diffusion mechanism. And once, and I remember Isaac talking to us and with Ralph and saying I'm going to show you the most boring chart ever. We have put the parameters into, you know, the matrix, and we get the same result every time we can't change this scute pattern. And I said, you're kidding. You know, the scute pattern is identical in tortoises and marine turtles. I mean, there's only one pattern that we know of, and so they were able, by using mathematics, to figure out, using a Turing reaction, how the scutes could form, which was, I thought, a brilliant way of using mathematical modeling.

Scott Gilbert 59:54
Now the other question that we have, of course, with turtles is, how does the turtle get its shell? And what we found were the turtle got its upper shell, the carapace, by redirecting the growth of its ribs into the dermis. And the ribs form the shell and then produce factors which ossify the dermis. Now, okay, because ribs undergo, you know, endocrine, endochondral ossification and stuff. This was really weird, because it was basically saying, you know, the rib can be used, not only as, you know, a structure for him, making a rib cage to allow you to breathe. It can make a shell, you know, kind of a Kaufman screwdriver type of thing, you know, you can't tell me what a rib can be used for. So, so we were looking at this, which was an evolvability, which we would not have expected, unlike the Turing thing, which we said, let's look for a reaction diffusion mechanism. No, this we did not expect. And what was interesting, and what was in the book Kevin wrote this, is that this was not what paleontologists were looking for. When they were looking for turtle shell evolution, they were looking for, you know, bony regions within the dermis that fused, whereas the evolution embryology showed something totally different. And then when the paleontologist went back to their fossils, they found things that mimicked what was going on in the embryo. And so for me to look at evolvability, you really have to bring in the paleontological element of this What did happen. And again, the models, the mathematics, can only tell what could happen. But paleo and development put together can actually tell you, you know, a good approach is, good approximation of what actually did happen, and then you have to try to find the mechanisms for it.

Marty Martin 1:01:52
Yeah, that's the very frustrating thing with biology, right? There's always what happened yesterday. That's a big part of the solution, and there's no way around it, yeah.

Speaker 1 1:01:59
Yeah and see now. Now what Scott did was that he zoomed out even further, right. He zoomed out to that place where we no longer see the populations and individuals anymore with lineages and transforming. And he was thinking about evolvability from that really zoomed out level again, showing another sort of concept of evolvability.

Scott Gilbert 1:02:17
Yeah, this is what, what Rudy Raff said way back when evo devo was being founded, he said that those population geneticists, they're interested in species, we're interested in bodies. And yeah, that's big difference. That is a difference in the approaches here.

Cameron Ghalambor 1:02:35
Yes, so I think this idea of, like, zooming in and zooming out is a really good way of, like, kind of focusing, kind of where the interesting debates are. And one question that I have when I was reading the book, I'm interested in a lot of these questions. I'm really interested in how organisms interact with their environments, how that generates plasticity, and how that generates variation that selection can act on, and so like the processes that happen within the lifetime, what the implications are for evolvability and responding to selection. But then, when I zoomed in here, I there were some things that I struggled with, because I agree that there are the idea of reciprocal causation and feedbacks. But what I have trouble conceptualizing is that how processes like developmental bias, you depict them as as causes of selection in the book, and and I, and that, to me, is a is a hard idea to get around that plasticity or developmental bias or niche construction, that these interactions with the environment actually are causes of selection. I see them more as changing the distribution of phenotypes, which then cause a change in the strength of selection, which then feeds back on the population. But I don't see them directly as the causes of selection. And so I'm wondering if you could expand, sort of, how you, how you see those ideas, because I find that really confusing.

Tobias Uller 1:04:28
Yeah, let me see if I if I can clarify some, clear up some of that confusion. So with causes of selection. So well, put it this way, selection being some kind of perhaps summed across individuals' relationship here between particular features of organisms and survival and reproductive success or their fitness. You know, it's, of causal link between a particular variant of a character and their reproductive success. And that link, whether or not a particular variant, a particular feature of the organism actually is going to cause variation in survival and reproductive success depends on what organisms do. I mean, this is very well recognized in the huge literature on the oral behavior, here. For example, lizards thermo regulating, influencing whether they expose themselves to selection on physiological mechanisms that need to perform at particular temperatures and so on, for example, right? So in that sense, organisms, what they do and how they work, are causes of selection, in the sense that they are causes of the things, of the particular features of their own biology that's going to be important for survival and reproductive success. So that is sort of in the way that why there is the, why there are courses of selection, does that help?

Cameron Ghalambor 1:05:47
But well, not well, sort of, because it seems like those are two separate questions, like I was thinking, like, let's say, you know, you have a daphnia that is exposed to the pheromones of a fish, and it develops a spike. And so, you know, if we could look at the distribution of phenotypes in the presence and absence of the spike, and clearly in the, you know, you see a shift in this distribution, and then that shift is what determines how strong selection is. So if you don't develop the spike, you're probably at higher risk of being consumed by the fish, and if you do, you have a higher probability of surviving and having you know, you have higher fitness. So why the spike evolved, as opposed to, say, another form of anti-predator behavior, you know, that reflects maybe these developmental bias or the past selection but, but to me, that's a different question than the role of selection as a sorting process. So, you know, I recognize that the environment has is doing two things here. It has a dual role. It interacts with the underlying genetics to create the phenotypic variation, and then it also acts as a source of selection to sort that variation. I have trouble thinking about the generation of the variation as a cause of selection like that. If because then that to me, then we have to completely that changes my definition of what phenotypic plasticity is, as the capacity of the genotype to produce different phenotypes like that goes against, I guess, my worldview that way. So I don't know if, if there's a resolution to that way of thinking, or if that's, you know, something we agree to disagree on. But, yeah, that's, that's sort of just where I struggle with that. And I think it gets back to the idea that you do touch on in the book about how we define causation, and I think it also depends on these, this, this reciprocal causation as well, so yeah.

Speaker 1 1:08:00
I was just going to say, I'll leave it over to you, Scott, soon. But I just wanted to say that one view of selection is that it's sort of the outcome of all those interactions that go on, all that complex ecology. And all that complex ecology goes on where there are causal relations between features of organisms, features of individuals, I should say, and their fitness, that those interactions are really the course that you know, that's where the course of action for for evolution is, and the selection is sort of the outcome of that. So on that view, you know, you wouldn't necessarily even see selection as a cause at all, right? So some people will claim that it's not really you had Dennis Wallace on the show at some point, I think. And he would say, you know, that natural selection is a is not a course. It's a sort of a second order effect or high level effect. So that's one potential view of this, which perhaps makes it easier to see that there are courses of selection when you think about it like that. Doesn't mean you have to think about it like that. But of course, the thing that the relationships between organisms and their environment that results in some organisms doing well and others doing not so well. Some individuals doing well and others not so well. That has causes and we can call those causes of selection. I think that wouldn't really be super controversial,

Cameron Ghalambor 1:09:23
As opposed to causes of variation?

Speaker 1 1:09:26
They would be the causes of, yes, so not as opposed to, I think so. So when, when, when a when a lizard thermo regulates, then there are causes of that response, that behavioral response, it puts itself in a particular situation where they might generate selection of life history traits, but not so much on physiology or vice versa, depending on, you know, the environment. So that behavioral response is, there's the cause of that behavioral response, and there's also. Selection on both the behavioral response, but also on all these other traits and as a result of how the organism engages with its environment. Selection, in a way, selection isn't sort of an external force that just pushes organisms. It emerges out of those interactions that organism have with themselves, with each other and with their environment.

Scott Gilbert 1:10:19
I would definitely agree that, you know, it's kind of this coevolution of organism within its environment. And that was, of course, the big thing about niche construction is, you know, that the organism helps make its environment. It doesn't have a pre, a pre-existing environment that it's trying to fit into. But, you know, again, I think the same meeting that caused Kevin and I and John Odling-Smee to write a paper together, got got me into a big argument with Michael Ruse, who just died a few days a few weeks ago, I'm told, and Michael was easy to get into arguments with. Okay, that's a separate issue, but Michael and I ended up having a debate in biological theory about the creativity of natural selection, and that he felt that natural selection was a creative force. And I said no, development is a creative force, and that the metaphor that I eventually came up with was that development has the creativity of the artist and natural selection as the creativity of the curator. Now they will they will enter, they will interact with each other. An artist can't make anything and expect it to be getting into a museum or anything there's there are these interactions between the artistic creativity and the curatorial creativity, but I see this as one metaphor of looking at natural selection and development.

Cameron Ghalambor 1:11:48
Yeah, so, you know, that's a really interesting point, because I'm actually surprised by Ruse's position on that. Because, I mean, I always think of selection as it just it can only do what variation it has that's available for it to act on. Exactly following your metaphor with the artist and the curator, but, but I do see that argument, especially coming from you know, people who are critical of, you know the standard evolutionary theory where they argue for selection as a creative force. And I'm like, I, I've never thought of selection as a creative force, so I it's, it's hard for me to wrap my head around that as well. And maybe that's again, because I'm, I'd like to compartmentalize some of these things more than than you like to think of them as being more integrated with one another, but that that that's been a that I see that come up over and over again, and I I always find that very confusing.

Speaker 1 1:12:52
Maybe I can try to be on the side of natural selection here as a creative force. So I don't think it's a force, but it's creative in a certain sense, because it is because natural selection results in the persistence of particular forms, and those particular forms are then the stepping stones for new things. And one interesting thing that's come out of a lot of modeling work in the evolvability realm is that you can't really select for the evolvability in the sense of producing functional variation in response to mutation, because the individual fitness gains of being able to produce something good in the future is very low. But even without selection for evolvability, you can get the evolution of evolvability so you can get evolution of systems that are actually better able at producing novel functional phenotypes in the future. And, in that sense, and the reason for that is, well, one of the reasons in certain models is actually because there has been selection for, for example, for phenotypic plasticity. So in that sense, selection, natural selection, results in creative organisms, organisms with evolvability, but it doesn't mean the same, typically, as what people usually mean when they say that selection is creative force. But there is a way in which natural selection results causes creativity of the evolutionary process in organisms, yeah, but it's still.

Cameron Ghalambor 1:14:15
Yeah, but it's still, ultimately, this sorting process. It's the curator that, you know, decides what passes through the filter and what doesn't, and so it doesn't create. The artist creates the developmental system, the genetic variation creates, but it,

Scott Gilbert 1:14:35
But it can, but it, but it can make the conditions for development to create. And I think that that's really important, that the face has this, you know, BMP responsive neural crest cell population, which is just crazy, because it can respond to different levels of BMP4, and can make a beak. You can make a narrow, you know, it could change the shape of the beak. It can change where the eyes are. I mean, it can, it can really be the ability to change with BMP concentrations and calmodulin concentrations is really an aspect of the cranial neural crest cell which is allowing it to, you know, make different faces really remarkable. And the fact that each one of us has a recognizable face, and that our faces are more like our parents than the general population, I mean, that there's this genetic component to the ability to respond is, I think, just amazing.

Marty Martin 1:15:40
Yeah oh Scott, that's a perfect setup for where I would like to go. But your emphasis on these potentials, the potential that the neuro crest cells in the face to be responsive, that is just so elegant and exciting to me. I mean, I sort of see that as one of these unifying things, where you get all the biologists into the room that we could, we could go for that kind of stuff. And I guess some of that excitement also derives from my excitement about systems and networks, and you mentioned GRN's, gene regulatory networks, a few times in the book, but it was, surprisingly not that much emphasized, given how rapidly our ability to make measurements of everything all the time is becoming was that deliberate? Or there's a lot that systems biologists have to say about what you're writing about, and whereas some things were there, there wasn't a lot there. So I'm just wondering how you think about systems and why the book took the form that it did.

Speaker 1 1:16:33
Interesting that you say that because I thought that there were quite a lot of GRN.

Marty Martin 1:16:37
Well I'm biased here. I wanted to see a lot more, so I might not be representative.

Speaker 1 1:16:42
No, I'm completely with you, you know, we were using this all the time, both theoretically and empirically, of course. But so I think one, one way to answer your question, why isn't there more is we I think, agree, I think, as authors, that the right, a good way of representing developmental processes in terms of regulatory systems, regulatory interactions and regulatory networks. It's a representation of development, of course, it's actually not the real thing, right? So it's important to remember. But that representation, I think, that we all find very powerful, and we've certainly made a lot out of it, I think in the book. Then, whether you want those regulatory networks to be purely sort of gene regulatory networks is a slightly different issue, right? Because you don't necessarily need to think of those regulatory interactions as just being GRNs, because you could have higher level phenotypes, like cells interacting and of course, the environment comes in and so on. So maybe that it wasn't conscious, if you think that we're using it less than expected, but I can see that maybe one reason is because we actually want to emphasize all these other parts of the regulatory system, rather than just, so to say, the genes, which the GRN concept, of course, invites you to just think about wnt signaling or something.

Marty Martin 1:18:01
Right, and that's, and that's actually Tobias. What my surprise was that there was an emphasis on GRNs and then allusions occasionally to other regulatory networks. So much about what you're writing, it lends one especially that thinks about networks as I do, that you would expect these kinds of things, integration across networks as a conspicuous piece of it, which is almost practically intractable, so that maybe that's why you left it out, but, but it was that sort of next level, the other regulatory networks, probably those comprise a lot about what we need to think about in the developmental context. And it was that absence that I was surprised by.

Speaker 1 1:18:34
Absolutely, so you're absolutely right that when you think about development and you represent development, whether it's a model or just in your head. Thinking about regulatory interactions is, I think, a really powerful way of doing it. If you want to formalize that and doing things with hierarchical networks and so on, it becomes very, very messy. And of course, that is one of the reasons why connecting what you refer to as sort of systems biology and evolution is still something that seems to be not really had taken off yet in that way, because, I think partly because of just the sheer complexity of those kinds of models.

Scott Gilbert 1:19:10
Yeah, , again, it's what, who is our audience and, and I, I mean, we do mention, for instance, the work of Hanna and Abouhei, the wonderful thing on, on, you take a wing. GRN, okay, the standard wing GRN, worked out in Drosophila, but it's also in butterflies, ants, beetles, whatever. Then you take out a gene, or you make a gene environmentally responsive to growth hormones, and all of a sudden you get a wingless phenotype. Okay, now you have a wingless phenotype and a winged phenotype, and this, they claim, might be the origins of eusociality. Okay, so by changing a gene expression and making it environmentally dependent in a wing GRN might give you something as radical as eusociality. I mean, we can mention this, but we're not gonna. But, you know, their papers are complicated, because you have to know a huge amount of developmental genetics and a huge amount of behavioral genetics in order to make sense of this. So we are, we mention it, but we don't go into the details.

Cameron Ghalambor 1:20:20
Yeah, so embracing this complexity. So, you know, I'm thinking about these GRNs in the context of reciprocal causation. So you have underlying genetic variation that feeds into these networks and pathways which are influenced also by the environment. This gives rise to either stable phenotypes, you know, during development, or plastic phenotypes. And I think that's, I don't I think that's, you know, not very controversial in and of itself. But then when we start to think about the way these systems evolve. And maybe this was in the book, and I missed it, but you know the idea of context dependency and how that plays into how a population, for example, will respond to selection. So like, if we think about developmental bias, like if I think about the analogy of just like standing genetic variation, we know that a concept like heritability is very context dependent. You know, heritability changes across different environments because the environment changes the phenotypic variation and what's available, that additive component that's available for selection to act on, you know, changes. Sometimes that changes in the same population across years. But then, if we, if we bring in these more complex mechanisms, aren't they also sort of like that where, under certain conditions, you expose variation that becomes available to selection, and in other cases, they mask the variation and and, and sort of act more as a a buffer or a constraint. I mean, this is maybe analogous, again, to the more simple version of a behavioral thermoregulation as well. But do do you see what I mean I'm trying to get at basically, is,

Cameron Ghalambor 1:20:52
I think you want to talk about plasticity.

Cameron Ghalambor 1:21:45
Not cam, no.

Cameron Ghalambor 1:21:53
I mean the networks themselves are plastic, right?

Speaker 1 1:22:33
Yeah, no exactly. So, I mean to first thing, perhaps, is to sort of start with the organism. So the organism is sort of central and one thing that is important for the organism is all those regulatory interactions, which include genes and regulatory elements and all that kind of stuff. Plasticity, what is that? Well, it's basically just that the developmental system, developmental process, developmental dynamics, are responsive to environmental input and produce some kind of change that's somehow observable as a different phenotype of some sort. Again, if you go back to imagine your metaphorical adaptive landscape, and then you have your developmental system that produces potential phenotypes, which is the morphospace, so those two concepts are, are people often quite used to thinking about. And you overlay the morphospace on the adaptive landscape. Now you ask yourself, what is the evolutionary role of plasticity? Well, one thing that plasticity can do, the environmental responsiveness of that system, is that a population that is in one part of the morphospace might actually move to a different part of the morphospace when the environment changes, right? And that is kind of the one way to think about this idea that the plastic response are evolutionary consequential, because the kinds of phenotypes that are brought into being at a particular time and point in a particular environment depends not Just on the genetic variation, of course, but it depends on the environmental conditions. So the environmental conditions here are really crucial. Like you say, aren't they, or aren't they also very responsive, you know, isn't it just like, Isn't it like heritability is environmentally dependent? Isn't the phenotype distribution in general, environmentally dependent? Yes, of course it is. Because the environment is always part of the development system, you can't escape that. So that is crucial, and that is potentially a good way of trying to make sense of the different arguments for how plasticity comes into the evolutionary study, even without sort of any kind of inheritance of acquired characters.

Scott Gilbert 1:24:37
And I would want to change landscapes. I would want to go from the adaptive landscape to the epigenetic landscape. Okay, we'll go to Waddington's model. You know, again, kind of, instead of going to standing genetic variation, I want to go to standing developmental competence. Okay, that the competence is a technical term, meaning the ability to respond to an inducing signal, and it's the ability to respond in different ways. And here we have the standing developmental differences in competence. So you have a salamander or a frog, and let's say a tadpole, tadpole of a frog. Now it's under going to undergo metamorphosis. Thyroid hormone is being produced. Well thyroid hormone tells the cells in the tail to degenerate. It tells the cells in the limb bud to grow. It tells the cells in the brain to change their connections and their neural connectivity. So it's doing different things to different cells. The cells have the competence to respond to triiodothyronine, but they respond in different ways. And so, and this is an organism with the same genome. All the cells have the same genome, but they are responding in different ways because they have different histories. And so then to get plasticity, you change the way the organism is going to respond to, you know, thyroxine in this case. Or make it environmental to cold weather, you know, when you know, or warm weather, when metamorphosis is primed so that you change the responsiveness of the cells, and so you can get plasticity just through a simple GRN by changing the epigenetic landscape of this, again, dealing with bodies rather than species.

Marty Martin 1:26:37
We want to sort of zoom out to 30,000 feet and ask some really big questions, to give you a lot of space, hit on things that we haven't hit, but the first one, we'll just go ahead and assume that evolution needs development. You've convinced us. We've totally had the Kool Aid. We're completely on board. Now, what do we do? How do we train our students? How do we focus our grant proposals? How do we adjust our mathematical models, and especially, what do we do about these networks and complexity that we've been spending the last ten minutes on?

Speaker 1 1:27:09
Yeah, okay, thanks for that simple question. So I there was so much in it, so I'm gonna stick with the first part. How does this affect how we teach our students? Traditionally, developmental biology, and I think that even organisms isn't really a big part of evolutionary biology. In traditional education, we tend to have screened off the organism already, perhaps when the students come in, certainly, in my experience, the students come in to the master courses in evolutionary biology, they haven't spent a lot of time in their bachelor's thinking about what organisms are and how they work, and they typically haven't gotten almost any developmental biology background at all. So I think that one of the things that we need well. So here's a suggestion. I've tried this, and many people think that it's stupid, but one way to approach education here is to actually show the student the complexity and how to manage it, rather than like we do now, which is, if you want to be a bit mean, you start with Mendelian inheritance and Hardy Weinberg, and then you sort of do a little bit more complicated things after that. So you actually don't show the students that the real world is incredibly complicated. Well, we can still handle it. So if you do that, the ambition is and this is certainly our ambition with our master program in evolution biology at my university, we want those students, no matter if they end up being behavioral ecologists or Evo Devo people, that they should be able to communicate with each other, which is something that certainly the senior academics aren't very good at if they are in these kinds of different fields. And one way to do that is to show them the complexity and how to handle it, explain to them not just how population genetics work and doing things, but why is there such a thing as population genetics? It's really useful, but why is it useful, and what can it do, and what can't it do? And if you try to develop education along those lines, the hope would be that the students will be a little bit more prepared, and then they will perhaps see that I actually need developmental biology to make sense of some of these things, because for some of the evolutionary questions, I need to represent developmental systems in terms of regulatory systems, and I need to actually understand something about development, just like I need ecology to be able to study selection.

Marty Martin 1:29:24
Scott, any thoughts there?

Scott Gilbert 1:29:26
Oh, yeah. Again, I'm gonna go. I'm gonna go to 50,000 feet because I teach undergrads and so and most of the people I teach are not going to be biologists. That's just a fact of life.

Marty Martin 1:29:42
Sad as it may be, but

Scott Gilbert 1:29:43
Sad as it may be, so, what I want to teach them about is kind of okay using this. What is an individual? What does being a holobiont mean, for instance, in terms of your interactions with nature? It's certainly the end of genetic determinism, you know, and plasticity, plasticity and holobionts really spell the end of genetic determinism. But genetic determinism is written into our laws in America. It's one of the bases for the abortion crisis. You know that you are, you get, when you get your genome, you become a person, and the genome becomes your soul, and that and the sociologists have shown that. So I have to tell them that, no, the genome is not your soul, that your phenotype is determined by the environment, by microbes, by many things. Then what is the cause of disease? You know, and they're learning about microbes, and they're learning about genes, and that all the genetic causes of disease and whatnot. I have to say, actually, it's a lot more complicated than that. Then, you know, going into the whole notion of evolutionary biology, and you know, we still teach it in terms of economic metaphors. Darwin's metaphors were mercantile capitalism. We've now gone into investment capitalism, but now what happens when you bring in symbiosis and you bring in a notion of competition mixed with cooperation. That, you know, it's not just competition, but it's competition to form a cooperative entity which then competes with other cooperative entities. It's more like teams. And so I want them, at least, to think about how they're thinking about life, and what is, you know, evolution, what are we talking about? Because evolution, at least in America, has become a synonym for life and death struggle. So I think that Evo Devo can actually, you know, get around that.

Cameron Ghalambor 1:31:53
Yeah. I mean, that's interesting. I I do teach some undergrads, but I mostly deal with with graduate students who, you know, are more interested in biology and and it's interesting because, I mean, especially in terms of how I teach, and I think also some of the some of the historic debates you know that that you guys have been involved with, with like the extended evolutionary synthesis, and the the more of the shortcomings of what you see is part of standard evolutionary theory. I guess, like, you know, I start with, like, Well, yeah, there's a standard evolutionary theory. But evolution as a discipline is very diverse, and people are asking all kinds of different questions and that it is complex. And, you know, and I'll, I may focus on things like, you know, interactions between the organism and and its environment, and how these play out, and but, in the end, I'm still kind of confined to the like, the standard theory, like I'm comfortable, comfortable with that. Like it doesn't make me uncomfortable, but I get the sense that, like, you know, when I finished reading the book, I mean, you ultimately, I feel like part of the motivation here is that, you know, you would like to see the theory changed or updated. Maybe is a better word is that, is that a fair, like, kind of assessment that like, what is the end the end goal? Like, is it to to change the way we think and teach, or to to build, like tearing things down and rebuilding? Or is it, is it just more like expanding and showing the richness and diversity, I guess. You know, the models are great, yeah, but they're a good starting place. But we need to, we need to incorporate a lot more to have a better understanding. I don't know is that, does that make sense?

Scott Gilbert 1:33:53
I'm such a developmental bio nerd, okay, I'm gonna say, I'm not gonna go with your metaphors. I'm gonna say the metaphor that I'd use for the change of evolution is metamorphosis. You know, some things degenerate, some things grow, and some things are repurposed. That how I see evolution evolving. It's more evolution, I don't want to say developing, but evolution undergoing a metamorphic molt. And the metamorphic molt metaphor was actually used in the 1950s by Haldane, saying that, you know, Waddington is bringing these new ideas in, and he is predicting that we will have a molt. We're not there yet, but we should get ready for another instar.

Cameron Ghalambor 1:34:39
Yeah, I like that metaphor. I think that's a good one.

Speaker 1 1:34:41
It's also important here to to, perhaps to connect to this idea of pluralism here, in the sense of the multiple representations designed to do different kinds of explanatory jobs. If you want to explain adaptive change in terms of fitness differences, you use a particular representation when you model it, and probably when you think about the problem and go out and design your experimental study. If you're interested in the extent to which developmental processes are enabling or facilitating adaptive evolutionary change, or responsible for why some organisms diversify so much, you represent the world in a different way, because you're zooming in on those, those developmental systems, and then you design your study accordingly, and there isn't really a conflict between those two, because they actually tackle different things. Now there is a conflict, ocasionally there will be true conflict. Occasionally things will actually not sort of add up, or they will, they will, because it's not the case that they always address the different things, one way to one response has often been that, yeah, right. You know, it's fine. Development can do all these kinds of things, but it just doesn't really explain adaptation. Well, it doesn't necessarily explain adaptation on its own, but it certainly could be a cause of adaptive evolutionary change, and that's typically what we're interested in, you know, we're not so interested in just showing that organism adapted over and over again, because we know that if they wouldn't have a certain relation to the environment, they would would be dead, because organisms need to be in a certain relationship environment

Scott Gilbert 1:36:10
And I think that one of the things that's happening is, you know, where you used to be able to say, I'm going to talk about the evolution of the eye. Okay, well now you say I'm going to talk about the evolution of the mechanisms that make the eye, and that's a big difference. And I think that's what you know to me, bringing in development into evolutionary theory to restructure it, in some ways, is about. You're talking about not the adult organism you're talking about the mechanisms that made it

Marty Martin 1:36:44
Yeah. Okay, so, and for sake of time, I want to maybe do one more thing about pluralism, as you, as you mentioned it Tobias, because I think and Scott you alluded to, there's so much genetic determinism that still pervades how we're teaching undergraduates and teaching evolutionary medicine, I try to make that go away as much as I can at every opportunity. And the people writing those books are trying really hard too, but so just sort of want to push back a little bit on what you're saying, that this sort of multiple ways of looking at biology is fine, and invoke a completely non controversial concept, like agency cam, Cam's favorite topic, but Tobias, you'd mentioned that you'd heard our episodes on agency, and I guess that is a wonderful thing for me, and it does have its historical difficulties, and there's plenty of things we talked about, if people want to go listen to that. But to me, it is sort of this potential opportunity for thinking about unification in biology. So we it's fine that there's a lot of biology that can be done at any level, but maybe there is a thread to be running through. And so what do you guys think about? I don't think you use the word agency in your book, but your book is fairly resonant with a lot of the things people talking about agency would say. So what's your perspective on the concept?

Speaker 1 1:37:54
Yeah, that's a good point. The agency concept that I think has also become much more used in the last sort of five years than it was before. And it sort of seems to be almost like an additional wave of this organism centered evolutionary body, or attempts to do something different, bring the organism back to the center stage, or whatever you want to call it. So I think that the one point that it makes with the agency is to really emphasize that it matters to evolution, that organisms are organisms. So it matters to evolution that organism, or maybe put it as a question, does it matter to evolution that organisms are goal oriented, purposful beings? Well, if the answer to that is yes, we better take it seriously and do something about it. And I'm very sympathetic to that. My current problem with the agency concept is when it comes to operationalization of agency. So what am I supposed to do with it? As a practicing scientist, I'm very sympathetic to the organizational account of agency and sort of systems that are goal oriented and so on. I think that is the right way to go, but I haven't really found that I can do very much with that concept in my normal sort of scientific work. But that might just be because I haven't found a way for it yet.

Marty Martin 1:37:59
You sound like Cam.

Cameron Ghalambor 1:38:47
I'm sort of, I'm in the same category, you know. I I think it's, you know, there are these terms like fitness, which we throw around and use very commonly, but are really difficult to measure and have different definitions. And, you know, are challenging as practicing biologists to measure.

Speaker 1 1:39:40
There are aspects here to agency. If you think about one aspect of agency is arguably that living systems have the goal orientation and the goal directedness of organisms, and that ought to be pretty important in evolution, because that goal directedness, goal orientation is actually one of the reasons for why development is always structured, why it produces structured phenotypic variation. It is one of the reasons for why fitness differences arise because organisms try to do certain things and expose themselves in particular situations and so on. So you may certainly want to do something with a goal orientation and really think about that long and hard and see how you can get that into your evolutionary thinking. So there, agency, the agency concept isn't used directly, not necessarily in that work, but it structures a research agenda or a problem agenda for things that you want to address. And I think this is a very important function of agential concepts in evolution of all the in general is that they structure problem agendas, research agendas, explanatory agendas, they focus on things that we want to try to understand, but not necessarily that the concept itself is sort of used as an explanans or something like that.

Marty Martin 1:40:51
Yeah. Scott thoughts on agency before we wrap. Well,

Scott Gilbert 1:40:55
Well I begin my developmental bio course with Aristotle. And Aristotle has is, you know, four causes. And the four causes are, you know, still viable. You know, you know, the material cause, the efficient cause, the formal cause, the final cause. Okay, let's look at these. And anything that changes, any of these causes, is gonna change development. And so I kind of looked at agency, kind of as a distributed function. Okay, how do you get the sex of a turtle? Okay, again, get me talking always brings back to turtles. How do you get the sex of a turtle? Well, it depends on the temperature of egg incubation, and that depends on calcium transport genes, and the calcium transport genes are activated by the temperature and they affect the genes that are involved in nucleosome modification, and the genes that are involved in nucleosome modification will allow or not allow the transcription of certain genes. Okay, so you have environmental factors working, you have tissue-level factors working. You have genetic factors working, you know, and they have to be there together to work. And if you change any of those, you'll get a different sex. So I like to think of agency as a very distributed function. And to use a whitehead term, the organism is kind of a compressence, you know, it's a coming together confluence of these different causes. And you know, I'm looking primarily at environment as a cause. You know how environment is an agent, rather environment like temperature, diet cause changes in gene expression, you know? So, yeah, the genes are there. They're important. But how does the environment cause them to become expressed differently? So yeah, the environment is one cause, the genes are another cause, and you can take this all the way back, as far as I'm concerned, to Aristotle, now, who was the first, the first embryologist of notes, right?

Cameron Ghalambor 1:43:07
That might be a good place for us to wrap up. Thank you so much for taking time and speaking with us today. I really enjoyed the book and enjoyed talking to you. So thanks. Thanks again.

Marty Martin 1:44:07
Yeah, same here. Thank you very much.

Tobias Uller 1:44:09
Thank you so much for taking your time to talk to us.

Scott Gilbert 1:44:12
Yeah, and thanks for this opportunity to talk about what our book's about.

Cameron Ghalambor 1:44:26
Thanks for listening to this episode. If you like what you hear, let us know via X, Facebook, Bluesky, Instagram, or leave a review wherever you get your podcast, and if you don't, well, we'd love to know that too. All feedback is good feedback.

Marty Martin 1:44:40
Thanks to Steve Lane, who manages the website, and Molly Magid for producing the episode.

Cameron Ghalambor 1:44:45
Thanks also to interns, Dayna de la Cruz, Caroline Merriman and Brady Quinn for helping with this episode. Keating Shahmehri produces our awesome cover art.

Marty Martin 1:44:55
Thanks to the College of Public Health at the University of South Florida and the National Science Foundation for support.

Cameron Ghalambor 1:45:00
Music on the episode is from Podington Bear and Tieren Costello

Big BiologyJanuary 2, 2025

Ep 127: Devo-lutionary theory (with Scott Gilbert and Tobias Uller)

Episode Transcript