t f p g+ YouTube icon

Endless Forms Most Beautiful, Part 2

Bookmark and Share

June 18, 2013 Tags: History of Life
Endless Forms Most Beautiful, Part 2

Today's entry was written by Sean Carroll. Please note the views expressed here are those of the author, not necessarily of BioLogos. You can read more about what we believe here.

Note: Yesterday and today on the BioLogos Forum, we feature excerpts from biologist Sean Carroll’s book, Endless Forms Most Beautiful: The New Science of Evo Devo, a well-wrought account of how the field of evolutionary development biology, or Evo Devo, has shaped our understanding of how animals evolve, by studying their development. The Psalmist wrote of how God knits us together in our mother’s womb, and thanks to the methods and tools now available to biologists, these processes that were once hidden are being revealed through this exciting field. While we cannot attest to the author’s faith persuasion, Sean Carroll does an outstanding (and witty) job of making a complex set of ideas more manageable to understand, and we encourage you to read this book in its entirety.

In today’s excerpt, Carroll summarizes a few significant facets of evolution that Evo Devo has helped illuminate: the “master tool kit,” (an ancient set of genes and their protein products shared across the animal kingdom that controls development, which is powerful evidence for descent with modification), why understanding the relationship between animals based on appearances alone can be misleading, and how new animal forms do not require new genes. Today’s excerpts comes from the final chapter of the book and refers to earlier chapters.

The editorial policy used in these excerpts can be found at the bottom of this post.

The Tools for Making the Kingdom are Ancient

The first and still perhaps the most stunning discovery of Evo Devo is the ancient origin of the genes for building all sorts of animals (chapters 3 and 6). The fact that such different forms of animals are shaped by very similar sets of tool kit proteins was entirely unanticipated. The ramifications of these revolutionary findings are powerful and manifold.

First of all, this is entirely new and profound evidence for one of Darwin’s most important ideas—the descent of all forms from one (or a few) common ancestor. The shared genetic tool kit for development reveals deep connections between animal groups that were not at all appreciated from their dramatically different morphologies.

Second, the discovery that organs and structures that were long viewed as independent analogous inventions of different animals, such as eyes, hearts, and limbs, have common genetic ingredients controlling their formation has forced a complete change in our picture of how complex structures arise. Rather than being invented repeatedly from scratch, each eye, limb, or heart has evolved by modification of some ancient regulatory networks under the command of the same master gene or genes (chapter 3). Parts of these networks trace back to the last common ancestor of bilaterians (Urbilateria), and earlier forms (chapter 6).

Third, the deep history of the tool kit reveals that the invention of these genes was not the trigger of evolution. The bilaterian tool kit predated the Cambrian (chapter 6), the mammalian tool kit predated the rapid diversification of mammals in the Tertiary period, and the human tool kit long predated apes and other primates (chapter 10). It is clear that genes per se were not “drivers” of evolution. The genetic tool kit represents possibility—realization of its potential is ecologically driven.

Existing Genes and Structures Provide the Means for Innovation

We have seen that insects, pterosaurs, birds, or bats did not invent “wing” genes (chapter 7), butterflies a “spot” gene (chapter 8), or humans a “bipedalism” or “speech” gene (chapter 10). Rather, innovation in all of these groups has been a matter of modifying existing structures and of teaching old genes new tricks.

The key to innovation at the genetic level is the multifunctionality of tool kit genes. The multifunctionality of tool kit genes stems from their deployment at different times and places through batteries of genetic switches. In this manner, a protein such as Distal-less can act at one time to promote limb formation, and at another to promote eyespot development. The protein made each time is identical, so the difference in function is due to its action on different switches in these different contexts.

At an anatomical level, multifunctionality and redundancy are keys to understanding the evolutionary transitions in structures. We saw this especially in arthropods, where the shifting of a function such as feeding to one of a battery of appendages freed other appendages to become specialized for walking, swimming, or other activities. In a similar fashion, the gill branches in aquatic arthropod ancestors became modified into book gills, book lungs, tubular tracheae, spinnerets, and wings.

Evo Devo has revealed the continuity among forms that was masked or about which there were uncertainties based on appearance alone. By revealing the developmental similarities among structures, Evo Devo presents a wholly new kind of evidence that is far more objective than morphology alone. These insights into the evolution of novelty strengthen aspects of Darwin’s original ideas that some have found most difficult to grasp.

The history of these structures also illustrates how “endless forms” evolve through cycles of invention and expansion. New structures open up new ways of living. The insect wing led to the evolution of dragonflies and mayflies, butterflies and beetles, fleas and flies, and more. The expansion of these groups was catalyzed in turn by a cycle of innovation and expansion by making modifications to the wings or body plan—scale coloration systems in moths and butterflies, a hard covering in beetles, a sophisticated balancing hindwing in flies.

Why are existing body parts and genes the more frequent pathway to innovation? This is a matter of probability. Variation in existing structures and genes is more likely to arise than are new structures or genes, and this variation is therefore more abundant for selection to act upon. As François Jacob explained so eloquently, Nature works as a tinkerer with available materials, not as an engineer does by design. The invention of wings never occurred from scratch, but by modifying a gill branch (insects) or forelimbs (three times). Trends in evolution reflect the paths that are most available and therefore those taken most frequently.

Evo Devo has revealed that evolution can and does repeat itself at the levels of structures and patterns, as well as of individual genes. If evolution takes the most probable path, via existing structures and genes, then when confronted with similar selection pressures, different species may follow the same path to adaptation. We saw this in the evolution of feeding appendages in crustacea (chapter 6), pelvic spine reduction in sticklebacks (chapter 7), and other cases of limb reduction in vertebrates. We also saw that melanic fur or plumage patterns can arise through mutations in the very same gene in different species, and even the very same position in this gene (chapter 9).

These instances of evolution repeating itself directly address difficulties some have had in grasping the role of random mutation in the evolutionary process. Some people have found it hard to imagine how novelty and complexity arise from “a random process.” The key distinction is that while the generation of genetic variation by mutation is a completely random process, the sorting of these variations as to which will persist and which will be discarded is determined by a powerful, selective nonrandom process. Of the hundreds of millions or billions of individual base pairs in an animal genome, all are equally susceptible to random copying errors or physical damage that cause mutations. But only a tiny fraction of all possible mutations can alter a mammal’s coat in a viable manner, or reduce a stickleback’s spines without causing catastrophic collateral damage. In large populations of animals, over eons of time, such mutations will arise simply as a matter of probability. When they do occur, positive selection upon the trait they affect will cause them to spread in populations over time.

Jacques Monod captured this interplay of randomness and selection in evolution most eloquently in the title of his landmark book, Chance and Necessity (a reference to the Greek philosopher Democritus who said, “Everything existing in the Universe is the fruit of chance and necessity”). Evolution is indeed a matter of chance, but in the random lottery of mutations, some numbers and combinations better meet the imperatives of ecological necessity, and they arise and are selected for repeatedly.

We also saw in rock pocket mice that the same species can use different paths to a similar solution. And, while pterosaurs, birds, and bats evolved wings out of their forelimbs, they did so in fundamentally different ways. Similar ecological demands and opportunities have selected for similar adaptations, but the developmental solutions will sometimes differ in detail.

By revealing the genetic and developmental mechanisms underlying change, Evo Devo allows us to compare and contrast the evolutionary paths of different groups. Long-standing mysteries such as Batesian mimicry in butterflies, melanism in moths, and even the evolution of finch beak size and shape now lie within our grasp. We shall soon have detailed pictures of many of the classic examples of natural selection and understand in depth how variation arises and is selected for.

Today’s blog was an excerpt taken from the Introduction of Endless Forms Most Beautiful, (c. 2006), which was a finalist for both the Los Angeles Times Book Prize and the National Academy of Sciences Communication Award, as well as being a Discover magazine and USA Today “Top Science Books of the Year.” Learn more here.

Editorial Policy: The editing for these excerpts involves removing the odd sentence or two—indicated by putting [SNIP] at the appropriate point(s)—and sometimes inserting annotations where warranted [also enclosed in square brackets] to provide background information.


Sean Carroll is a Professor of Molecular Biology, Genetics, and Medical Genetics at the University of Wisconsin–Madison. He studies the evolution of cis-regulation in the context of biological development, using Drosophila as a model system. He is a Howard Hughes Medical Institute investigator. Since 2010, he has been vice-president for science education of the Howard Hughes Medical Institute.

< Previous post in series

View the archived discussion of this post

This article is now closed for new comments. The archived comments are shown below.

Page 1 of 1   1
Roger A. Sawtelle - #81170

June 18th 2013

Evo Devo unmasks another misunderstanding of NeoDarwinism, the primacy of random mutation of genes.

The big question, the elephant in the room, is natural selection, which it seems no one is addressing, but is the key to the whole question.  Evo-devo might, but I do not see where it has. 

Lou Jost - #81195

June 19th 2013

Come on, Roger, natural selection is not mysterious. Any elementary text in pop gen or evolution explains it. I showed you how it is a mathematical consequence of differential reproductive success, a few threads ago. You even admitted, sarcasticly, that this was just common sense. So now this is just crazy of you, to say (just a few days after that discussion) that no one is addressing natural selection.  Read a genetics book for g-d’s sake!

beaglelady - #81191

June 19th 2013

Sean Carroll is one of the lecturers for the 2005 Holiday Lectures presented by the Howard Hughes Medical Institute.  (As noted, Carroll is an investigator for the HHMI)  Since the target audience is bright high school students, it is very accessible. 

beaglelady - #81193

June 19th 2013

And Sean Carroll also has a relevant video called The Making of the Fittest: Evolving Switches, Evolving Bodies about the transformation of the bodies of Stickleback fish.

Roger A. Sawtelle - #81196

June 19th 2013


Genetics has nothing to do with natural selection.  That is the problem.

Since you do not believe in God (or do you?) please do not take God’s Name in vain.

Lou Jost - #81197

June 19th 2013

Genetics has nothing to do with natural selection?? Please explain how genetics has nothing to do with the process whereby favorable genes spread through the gene pool.

Lou Jost - #81198

June 19th 2013

You’re not even trying to think carefully about this. PLEASE read the relevant part of a genetics text.

beaglelady - #81199

June 19th 2013

Looking forward to another one of these conversations, Lou?

Lou Jost - #81201

June 19th 2013

I’m a glutton for punishment. I’ll give up, though, if he doesn’t look up the basics of natural selection and tell us exactly what is wrong with it, or with the example I gave in the earlier thread.

beaglelady - #81205

June 19th 2013

You’re adventurous,  I’ll say that for you.

Lou Jost - #81300

June 24th 2013

I’ve given up now. I tried my best.

glsi - #81202

June 19th 2013

There seems to be widely divergent views about natural selection within the evolutionist community.   Here’s one thing Lynn Margulis said about it:  “Natural selection eliminates and maybe maintains, but it doesn’t create”.  That’s from an interview in the pro-evolution Discover magazine and it’s not somehow out of context.  It’s one of many blunt criticisms she levels against much of neoDarwinism.

Lou Jost - #81203

June 19th 2013

There’s no divergence in opinion at all about the efficacy of natural selection in changing the allele frequencies of populations. Margulis is correct in the sense that natural selection does not create a new mutation, but no one ever claimed that it did. Natural selection changes the frequencies of alleles, so that favorable ones become more common in the population, and less favorable ones become rare or disappear. By doing this across the genome, it does create novel combinations that would be very unlikely to arise by chance alone. Natural selection transforms the population over time. In this sense it creates.

I don’t think there is any controversy here except perhaps if people use words loosely. We can always come back to the math if we need an unambiguous reference.

glsi - #81206

June 19th 2013

I’m pretty sure she’s saying natural selection doesn’t create new body forms or create more and more complex information.  Very similar to what many ID people are saying.  Sorry, I don’t have the date of the Discover article.  I xeroxed it w/o writing the date on it.  It’s from 2 or 3 years ago I believe.

“But neo-Darwinists say that new species emerge when mutations occur and modify an organism.  I was taught over and over again tht the accumulation of random mutations led to evolutionary change—led to new species.  I believed it until I looked for evidence.”—L. Margulis

Lou Jost - #81207

June 19th 2013

It’s time for bed so will continue tomorrow. Speciation (in the sense of cladogenesis) and evolutionary change are not synonymous, and the mechanisms of each are different.  Still, there is lots of evidence in both processes for the role of natural selection acting on random genetic variation (not only mutations).

glsi - #81208

June 19th 2013

Okay, I’m leaving for Mexico in the morning and probably won’t be thinking about any of this for several weeks.

Lou Jost - #81219

June 20th 2013

Enjoy your trip! Meanwhile I’ll try to find the Discover article.

Lou Jost - #81222

June 20th 2013

Found it, April 2011. Lots of discussion of it in creationist and scientific blogs.

Lou Jost - #81226

June 20th 2013

The article contains many very odd statements. Will take a while to unpack all of it. Hope I can find the time….maybe by the time you come back from Mexico…

Roger A. Sawtelle - #81211

June 20th 2013


The role of natural selection is to select.  The question is not whether natural selection selects, but how does it select? 

Or what is the mechanism that enables NS to select so it is not Darwinian magic?  This is the question that genetics does not answer. 

Maybe you can answer it since genetics can not.

Variation proposes, Natural Selection disposes.  They are separate processes working together to create evolution.  This is why evolution is not monistic. 

beaglelady - #81216

June 20th 2013

Would you care to take a free course on genetics and evolution? 

Lou Jost - #81220

June 20th 2013

“Or what is the mechanism that enables NS to select so it is not Darwinian magic?”

What precisely is mysterious about the example I gave in the other thread? I showed that a gene which happened to improve nitrogen uptake would spread through the population because it increased the number of spores that could be made. Any gene that arises by chance and improves reproductive success will be expected to increase in the population. We can predict the rate of increase and the final proportions based on the fitnesses and the population demographic parameters. That is all natural selection is. What’s the big deal? Where’s the “Darwinian magic”?

Roger A. Sawtelle - #81236

June 20th 2013

Any gene that arises by chance and improves reproductive success will be expected to increase in the population.

You made a statement of expectation, but not an statement of process. 

How do genes improve reproductive success? 

Is is not by enabling life forms to better adapt to their environment?  All the examples that you have given point in that direction. 

If so, let us just cut out the evolutionist jargon and say it straight out.  Life forms which are better adapted to their environments have an ecological edge and will thrive.    

Ergo, Natural Selection is ecological selection.

Lou Jost - #81237

June 20th 2013

Maybe we are getting somewhere. “Any gene that arises by chance and improves reproductive success will be expected to increase in the population…..How do genes improve reproductive success? Is is not by enabling life forms to better adapt to their environment?” Yes that is exactly right. So maybe natural selection is not Darwinian magic or myth, but simply “evolutionary jargon” for what you are calling “ecological selection”?

Roger A. Sawtelle - #81239

June 21st 2013

Ah Hah! 

Yes, but ecological selection is  very different in form and structure from Darwinian selection.  Darwinian selection even as defined by eco-devo is based on Malthus, while ecological selection is based on Lovelock.

As Noble has pointed out, the NeoDarwinian Selfish Gene evolutionary process is linear and monistic.  The Ecological evolutionary process is a holistic, dynamic, interdependent process worthy of the name of Evolution.  

Also of course the working of the NeoDarwinian natural selection process of competitive struggle has not been verified.    

Lou Jost - #81244

June 21st 2013

Don’t get distracted by your philosophical concepts. You were on track in your #81236….you seemed to agree that natural selection is exactly what you are calling “ecological selection”. It actually doesn’t involve Malthus (favorable genes would still increase in relative frequency even if population growth was unlimited). Evolutionary theory is also not linear but dynamic. If you would stop trying so hard to demonize the theory and make a straw man out of it, you’d probably find that you agree with it.

Roger A. Sawtelle - #81245

June 21st 2013


On what basis can you say that favorable genes would increase faster than others if there was nothing to limit population growth?  In other words if there were no competition for limited resources, what good is relative advantage? 

To say that if one factor would change it would not make a change in the other factors indicates a non-scientific point of view. 

Ecology is not philosophy, it is science, even if Dawkins refuses to recognize it. 

It is Dawkins who is wrapped up in philosophy as he freely admits.  Noble says that the Selfish Gene concept of NeoDarwinism is dead wrong.  Do you agree with him or not?  

I agree with Noble’s view and have for some time.

Which side are you on?      

Lou Jost - #81249

June 21st 2013

To see why favorable genes would still increase faster than the others even if population were unlimited, just work an example. Suppose we have two genes, one that allows a moss to make 100 spores, and another that allows a moss to make 1 spore. Suppose the initial population size is 2, one individual with the favorable gene and the other individual with the unfavorable one. If every spores sprouts (no competition, no premature death, infinite resources), in the next generation there will be 100 individuals with the favorable gene, and 1 individual with the unfavorable gene. In the third generation, there will be 10000 individuals with the favorable gene, and 1 individual with the unfavorable one. In the k-th generation, there will be 100^(k-1) individuals with the favorable allele, and one individual with the unfavorable allele.

Math is an important deductive tool for deriving consequences from premises. Verbal arguments are much less trustworthy.

I have still not formed an opinion on Noble’s ideas, because I am still not sure what they really are. I think he exaggerates the novelty and importance of his points, but I have still not read enough of him to make a more definite statement than that.

Lou Jost - #81253

June 21st 2013

Please note that this comment refers to my earlier moss example on another thread. The moss is haploid (one chromosome), and has non-overlapping generations (all parents die after reproducing one time only, and they die before the spores of the next generation sprout. You need that information to make sense of my explanation above.

Roger A. Sawtelle - #81270

June 22nd 2013


You example is bogus. 

Genes do not limit the number of offspring.  Generally speaking life forms produce more than enough to reproduce themselves, at least thousands of times if not millions. 

The exception would be advanced mammals like humans who reproduce in a limited manner because they must provide nurture for their offspring. 

The reproductive advantage is therefore not in the ability a life form to reproduce, but the relative ability of its offspring to survive and reproduce.  If all offspring of all rival life forms survive and reproduce, there would be no relative advantage and the earth would be swamped with life. 


Lou Jost - #81283

June 22nd 2013

Youasked in your earlier comment “On what basis can you say that favorable genes would increase faster than others if there was nothing to limit population growth?” I just proved to you that what you said was not true. Even if there are no limits to growth, the fitter genes will still dominate the population.

Of course the earth would eventually be swamped with life, which is why this can’t really happen for long in nature. But you asked the question involving unlimited growth, so I answered.

But you are right about this: “The reproductive advantage is therefore not in the ability a life form to reproduce, but the relative ability of its offspring to survive and reproduce.” In my moss example I explained that I would assume the plants didn’t die once they sprouted, so I could simplify the math. This simplification doesn’t matter for the question at hand. But it’s easy to incorporate that, along with lots of other things. Look at Templeton’s population genetics text which I cited earlier for you, and you can get the full treatment. Like I said before, if you would just bother to learn the theory and stop trying to demonize it, you would discover (perhaps to your horror?) that you agree with it.

Roger A. Sawtelle - #81288

June 23rd 2013


All the concrete examples that Beagle Lady have presented confirm with my point of view. Now you are agreeing with an important statement that natural selection determines relative advantage (not genes.)

Changes in life forms are caused by changes in the ecology in which they live. That is true of the pocket mice and true of the stickleback fish. Noble is saying that life forms use genes to adapt to their environment, not the other way round. That is what the evidence demonstrates.

It is not genes that initiate change in life forms, but the impact of environmental change, which causes these changes to take place.

Evo-devo also says that most if not all changes are not initiated by random mutation, but are found in the elements found in the DNA toolkit. Evo-devo takes most of the chance out of evolution.

I disagree with the Darwinian of natural selection as presented by Dawkins and others. I asked you to present some real scientific examples to prove your case. The two examples found by Beagle Lady confirm my position. Thank you, Ms. Case closed.

Lou Jost - #81290

June 23rd 2013

At times you seem close to understanding the theory, but then you pull back. I wish you the best and hope that some day you look closely at a text on population genetics and evolution.

Roger A. Sawtelle - #81371

June 27th 2013


Best wishes to you also and I hope that you will someday see the light.

Page 1 of 1   1