How the evo devo revolution is undermining Darwinism. II

Internal Constraints vs. External Pressures: The Revelations of Evo-Devo

Michael Denton March 14, 2016 3:29 AM

Editor's note: In his new book Evolution: Still a Theory in Crisis, Michael Denton not only updates the argument from his groundbreaking Evolution: A Theory in Crisis (1985) but also presents a powerful new critique of Darwinian evolution. This article is one in a series in which Dr. Denton summarizes some of the most important points of the new book. For the full story, get your copy of Evolution: Still a Theory in Crisis. For a limited time, you'll enjoy a 30 percent discount at CreateSpace by using the discount codeQBDHMYJH.

One of the major advances since the publication of Evolution: A Theory in Crisis has been the revolutionary increase in our understanding of development and especially the genetics of development, about which little was known in 1985. These advances have led to the emergence of a whole new field, nicknamed "evo-devo" (evolutionary developmental biology). At the heart of the new field has been the discovery that a limited set of highly conserved genes, gene circuits, and developmental mechanisms -- for example, Hox genes, signaling proteins such as "sonic hedgehog," chemical gradients, and gene regulatory networks -- are involved in the construction of the bodies of all bilaterally symmetric animals,1 all the morphological homologs, and indeed all higher organismic form. Sean Carroll has termed this set of conserved genes and developmental mechanisms the "toolkit."2

One now well-established finding of evo-devo was totally unexpected from Darwinian first principles. It is that what are referred to by evo-devo researchers as "constraints," or what might also be termed "internal causal factors," have played a far more important role in the origin of many of the homologs and Bauplans than was previously envisaged. Before evo-devo it was widely thought that internal factors played little or no role in shaping the course of evolution. But it is now increasingly obvious that internal causal factors have played a far more prominent role in the actualization of evolutionary innovations than cumulative selection.

One of the most fundamental of all evolutionary innovations occurred when some ancestral chordate switched its body plan from the design previously shared by all other animal groups (in which the nerve chord is in a ventral position and the heart and main blood vessel are placed dorsally) to a design that was the exact reverse (in which the nerve chord is placed dorsally and the heart placed ventrally), and that has remained invariant in all chordates ever since. The evidence for this dorsal-ventral inversion (D-V) was revealed when evo-devo studies showed that the signaling molecules that specify the back of an insect also specify the ventral side (the belly) of a vertebrate.3 Before evo-devo revealed that the same genes were involved in specifying the dorsal-ventral axis of chordates and non-chordates, no one took seriously Goeffroy's suggestion, made early in the 19th century, that all animals shared the same basic body plan.

Self-evidently, Darwinian scenarios must confront the obvious question as to whether this transition was gradual or sudden. On the one hand, it is very hard to imagine how gradual cumulative selection could carry out such a radical re-engineering of the basic body plan. What mystifying adaptive path can be proposed along which such a gradual transition might have occurred? On the other hand, if the change occurred suddenly in one massive macro-mutational saltation, then Darwinian causation is ruled out of court altogether.

In the case of this dramatic innovation, the revelations of evo-devo provide no support at all for the Darwinian notion that cumulative selection is the major causal engine of evolutionary transitions.

In taxa-defining homologs, it is self-evident that the emerging evo-devo picture provides no support whatsoever for the Darwinian claim that novelties came about during the course of evolution to serve a succession of functional ends. In this case, the structuralist inference that internal constraints and not cumulative selection played the key role in their actualization is difficult to refuse.

This example is not atypical. Evo-devo advances have revealed that in many, if not the great majority, of innovations in the history of life,internal causal factors have played a predominant role. Against all traditional expectations, Darwinian selection to serve adaptive ends could only have played, in most cases, a relatively peripheral causalrole. Before evo-devo, no one would have imagined that a vast amount of organic order arises from internal constraints and causal factors within organisms themselves and is not imposed by selection. What was heresy only three decades ago is now accepted doctrine.

References:

(1) Lewis I. Held, How the Snake Lost Its Legs: Curious Tales from the Frontier of Evo-Devo (New York: Cambridge University Press, 2014), Chapter 1.

(2) Sean Carroll, Endless Forms Most Beautiful, Chapter 3, see fig. 3.7.

(3) Held, How the Snake Lost Its Legs, Chapter 1; for alternate scenarios, see "Inversion (evolutionary biology)," Wikipedia, Wikipedia Foundation Inc., March 22, 2015 (accessed on August 19, 2015).

On the distinction between patients' rights and euthanasia.

Some Bioethicists Are Not as Smart as They Think

Wesley J. Smith March 9, 2016 4:51 PM

Witness this attack by University of Ottawa professor Stuart Chambers on the disability rights advocacy organization Not Dead Yet for its opposition to legalizing assisted suicide and euthanasia.

Chambers calls the organization's stance "unprincipled." Why? Because it distinguishes between the right to refuse life-sustaining medical treatment, which the group supports, and legalized euthanasia and assisted suicide, which it opposes implacably. From "Not Dead Yet: An Unprincipled Position Against Assisted Death":

Although the choices surrounding acts of commission and acts of omission could be similarly motivated by loss of autonomy, decrease in bodily function, or feelings of being a burden, Not Dead Yet only supports autonomous choice for passive treatment decisions (withholding or withdrawing life-sustaining treatment) that lead to premature death. To be consistent, Not Dead Yet should be lobbying for or against all end-of-life alternatives.

We've heard this same sophistry for decades. The truth is:

• Withdrawing medical treatment may lead to death, but that isn't the intent. Indeed, the point is to stop an unwanted bodily intrusion, not to kill. As Paul Ramsey put it, that is treating the "patient as a person."
• With the exception of a feeding tube, such deaths are uncertain. Sometimes -- if unexpectedly -- people live. For example, Karen Ann Quinlan lived about ten years after her respirator was removed.
• Death is certain in euthanasia and assisted suicide.
• When medical treatment is withdrawn or withheld on request, if it comes, the death is natural.
• In euthanasia and assisted suicide, death is unnatural, e.g., a result of homicide or suicide.
• In contrast to removing unwanted treatment, the intent of assisted suicide and euthanasia is to kill.

Not Dead Yet sees clearly that assisted suicide and euthanasia discriminate invidiously against people with disabilities because they treat them as a disposable caste whose lives are not worth saving if they become suicidal.

The vital distinction between "allowing to die" and "making dead" through homicidal or suicidal means was recognized 9-0 by the United States Supreme Court in the 1997 decision Vacco v. Quill :

[A] physician who withdraws, or honors a patient's refusal to begin, life sustaining medical treatment purposefully intends, or may so intend, only to respect his patient's wishes to cease doing useless and futile or degrading things to the patient when the patient no longer stands to benefit from them. ...

A doctor who assists a suicide, however, "must necessarily and indubitably, intend primarily that the patient be made dead." Similarly, a patient who commits suicide with a doctor's aid has the specific intent to end his or her own life, while a patient who refuses or discontinues treatment might not ...[and, indeed] may instead "fervently wish to live, but to do so free of unwanted medical technology, surgery, or drugs." [Citations omitted.]

In other words, the right to refuse unwanted medical treatment is not a "right to die" but a right to be free from unwanted bodily intrusions. So then were the nine Justices -- liberal and conservative -- who unanimously signed this decision "unprincipled?" Please.

We shouldn't be surprised by the attack on Not Dead Yet and the opposition of disability rights organizations to assisted suicide. They pierce the political and ethical sensibilities of those on the political Left. They appeal to secularist sensibilities and arguments. Most importantly, they are effective.

Advocates like Chambers would do well to learn the skill of three-dimensional thinking and making nuanced distinctions. It would help them understand cogent and sophisticated arguments against legalizing assisted suicide.

Yet more on geologists'rewrites of the cambrian narrative

"Settled Science" and the Cambrian Explosion -- Geologists Weigh In

Evolution News & Views March 12, 2016 3:46 AM

Yesterday we considered the "settled science" explaining the Cambrian explosion in light of one of two new papers in the Geological Society of America Bulletin. In the second paper, also by geologists from Harvard (Smith et al., with Francis A. MacDonald participating in both papers), we travel to Mongolia.

Until the 1990s, large regions of outcrops that cross the Ediacaran-Cambrian boundary were inaccessible to Western scientists. The Harvard team performed an extensive mapping of the area, correlating the outcrops with each other, measuring carbon and oxygen ratios for thousands of samples, and integrating their findings with outcrops in other parts of the world. They begin by introducing the problem:

Since Charles Darwin's observation of the apparently rapid appearance of fossils in what is now known as Cambrian strata (Darwin, 1859), the Precambrian-Cambrian boundary has been widely regarded as one of the evolutionary pivot points in the history of life. Despite the persisting interest on this topic, the causes, triggers, and tempo of change remain controversial. Following the extinction of Ediacaran biota and the calcifying metazoan Cloudina, the early Cambrian (or Terreneuvian Series) encompasses an evolutionary interval characterized by increases in diversity and disparity(Marshall, 2006) that coincide with multiple carbon isotope excursions with amplitudes of <8 2006="" 2007="" 2010a="" 2011="" al.="" e.g.="" erwin="" et="" maloof="" nbsp="" porter="" span="" style="border: 0px; margin: 0px; outline: 0px; padding: 0px;" zhu="">over a geologically brief interval in Earth history

 (Valentine, 2002).Most recently, Maloof et al. (2010a) suggested that the Cambrian fossil first appearances occurred in three discrete pulses associated with rapid reorganizations in the carbon cycle. To test this hypothesis and others about mechanistic links between environmental change and evolutionary milestones across the Ediacaran-Cambrian transition, it isnecessary to integrate records around the globe. However, previous global syntheses (e.g., Maloof et al., 2010a) have been limited to a small number of localities with large uncertainties in both local and global stratigraphic correlations and the lack of absolute ages directly linked to biostratigraphic and chemostratigraphic data, particularly with respect to critical data from Asia. Because the global data set is biased by just a few localities, refining local correlations and grounding them in regional geology arenecessary before an accurate global synthesis of fossil occurrence data can be constructed and interpreted. [Emphasis added.]

Similar problems were found in Mongolia as in the first paper: previous studies were flawed, and geochemical evidence did not correlate with fossil evidence. Once again, it's the local conditions that matter; "we suggest that this pattern is controlled largely by regional sedimentation and taphonomy [fossilization processes] rather than the rate of taxonomic origination," they say. ("Taxonomic origination" is a euphemism for "abrupt appearance of complex animals.")

One of their findings puts the squeeze on the evolutionary biologists. Their recalibration of the sequence puts the first appearance of small shelly fossils "hundreds of meters higher in the stratigraphy" than previously recorded, because earlier studies mistook an outcrop of phosphatic shale at one location with another fossil-bearing layer due to incorrect mapping. This compresses the time available for their evolution.

Reading these papers in detail, one gets the clear impression that geology is as much art as science. You can't just walk up to a wall of strata and read it like a book, much less use it like a Rosetta Stone to correlate with similar outcrops in other parts of the world. A great deal of interpretation is involved, even with empirical data like carbon and oxygen ratios. The authors use the word "interpret" frequently, even alleging that previous geologists misinterpreted things.

For example, one of the second paper's charts shows the small shelly fossils appearing in three pulses whose dates vary between Mongolia, Siberia, and China. Is this real, or an artifact of preservation?

We suggest that the apparent pulses of fossil first appearances are the result of intervals of nondeposition in the sections included in this compilation and do not represent global evolutionary patterns; FADs will not be found during periods in which sediment is not deposited. Charles Darwin(1859) suggested that the apparently rapid appearance of fossils found in Cambrian strata was a product of the incompleteness in the stratigraphic record -- at a smaller scale, this indeed may be the case.

If it "may" be the case on the smaller scale, it is clearly not the case on the large scale. Every Cambrian expert agrees that the fossil record is complete enough to consider the Cambrian explosion a major unsolved problem in biology.

Other complications appear in the paper:

1. Some of the strata are interpreted to be autochthonous (in their original position), and others are interpreted to be allochthonous (transported into place). There are flooding surfaces, intrusions, thrusts, bypass channels, subduction zones and unconformities. Much of the region is in a large basin that was infilled by sediments.
2. Some formations are "highly variable both in terms of thickness and lithology," with facies changes occurring over very short distances.
3. The authors infer periods of "depositional hiatus" in certain areas, one of them possibly up to 6-10 million years in length (this is to keep their correlations in sync).
4. They cannot account for the large "excursions" of carbon-isotope ratios (positive and negative) at certain levels. After considering various explanations, they say, "None of the hypotheses described above provides direct explanations for a mechanistic link between eustatic sea-level change and the isotopic variations." The measurements cannot, therefore, serve as unambiguous proxies for changes occurring in the global carbon cycle at different times. (This undercuts Maloof's 2010 hypothesis about three pulses of evolution tied to the carbon cycle and, instead, attributes the pulses to accidents of deposition.)
5. The dates of the carbon-isotope ratio excursions do not always match between different parts of the world, even though they are assumed to represent correlation "tie points." Mongolia has extra excursions, for instance, that do not appear in China or Siberia. "suggesting that the carbon cycle was oscillating even more rapidly than previously thought during the earliest Cambrian."
6. The first appearance of a trace fossil named Treptichnus pedum is considered diagnostic of the Ediacaran-Cambrian boundary around the world, but it appears at different dates in different locations. Because its appearance is strongly "lithofacies controlled" (dependent on the type of rock in the outcrop), "using T. pedum as a global chronostratigraphic marker has been problematic on most Cambrian paleocontinents, notably in Siberia, China, Mongolia, and Kazakhstan."
7. Some of the upper strata contain ultramafic minerals, representative of very high temperature volcanics that are atypical of the cooler mafic lavas observed today.
8. Some of the strata are stratified; others are not. Some contain conglomerates are even large boulders, representative of transport. Some of the conglomerates contain pebbles that are rounded and well sorted; others are angular and unsorted.
9. The strata contain chert, limestone, sandstone, siltstone, dolomite, ooids, and other minerals and structures that call for interpretation.
10. The Harvard team ties their carbon-isotope ratios to absolute ages from Morocco (a third of the way around the globe), but Morocco lacks the earliest Cambrian fossils. "Because there is no one section globally in which it is possible to integrate the ichnofossil record, body fossil record, carbon isotope chemostratigraphy, and absolute ages, the calibration of this evolutionarily important transition remains piecemeal, resulting in much uncertainty in determining rates of origination and geochemical change."

These complications make it likely that some future geologist will find flaws in this paper. If the science were settled, the Harvard team would not end with a call for "testable hypotheses that can be used to better constrain the relationships between biological and environmental change during this major transition in life." In other words, we geologists just see "first appearances" of complex creatures in a confusing bundle of rocks. Ask the biologists where they came from.

Information 'R' us?

To Protect Genetic Information, Cells Go to Extraordinary Lengths

Evolution News & Views March 10, 2016 3:43 AM |

If the real "stuff" of the universe is information, as William Dembski argues in his book Being as Communion, it makes sense that life would make it a high priority to preserve that information. To a Darwinian who deals in chance processes, though, information is just a haphazard by-product of variation and selection. Some recent papers provide an opportunity to compare the explanatory success of both positions.

That DNA bears coded information has been non-controversial since the 1950s, of course, but we should not forget how revolutionary a concept that was. Before that, heredity reduced to chemistry. Some envisioned a homunculus in the cell, a reduced form of the organism that merely grew to maturity. Mendel conceived of alleles in pairs that accounted for traits. But to find that the carrier of heredity was a digital code inscribed on molecules must surely count as one of the most revolutionary findings in the history of science. It also made sense: the vast diversity of living things could be explained by complex specified information encoded in a universal chemical language.

Now we are in the throes of another revolution. Epigenetics is revealing codes above codes; systems of data flow, quality control, and information processing that cannot be reduced to DNA itself. Let's look at a few of these with an "information" focus.

The Local Stiffs

To be transcribed, DNA has to be accessible. If the double helix is too rigid or too loose, its readability is compromised. The double helix of DNA wraps around chromatin to form nucleosomes. This would seem to lock it away from the transcription machinery, but a remarkable set of molecular tags is coming to light that modulates the stiffness of the DNA strand on chromatin, so that it can "loosen up" when needed for translation, but stay rigid enough to be durable at other times.

These tags, called methyl groups (5-mC), formyl groups (5-fC) and carboxyl groups (5-caC), attach to cytosine bases in the DNA at the fifth carbon. The type of tag affects the flexibility of the strand, probably because of each tag's unique electrostatic signature. How this works was reported recently in Nature Communications by American biochemists at Eastern and Midwestern universities.

Molecular dynamics simulations show that these modifications promote or dampen structural fluctuations, likely through competing effects of base polarity and steric hindrance, without changing the average structure. The increase in DNA flexibility increases the mechanical stability of the nucleosome and vice versa, suggesting a gene regulation mechanism where cytosine modifications change the accessibility of nucleosomal DNA through their effects on DNA flexibility. [Emphasis added.]

The authors list the following possible functions for these tags:

5-mC is found in most plants, animals and fungi, and has aprofound effect on genome stability, gene expression and development. 5-hmC, 5-fC and 5-caC may also function in gene regulation. For instance, 5-hmC is preferentially enriched at distal regulatory elements such as enhancers in mouse or human embryonic stem cells, whereas 5-fC or 5-caC tend to accumulate at poised enhancers and promoters.Possible mechanisms for the regulatory role of cytosine modifications include the following: (1) altering the physical properties of DNA; (2) modulation of chromatin accessibility; and (3) recruitment of proteins that recognize cytosine modifications as their substrates followed by transcriptional activators/repressors that translate the presence of modifications to downstream signalling of transcription machinery.

If the proteins are recognizing these cytosine modifications, they are in essence reading a code that is independent of the DNA sequence. It's like calling out, "Tag found here; call in the activator." The activator signals the transcription machinery accordingly. That's fascinating enough, but the tag actually modifies the physical properties of DNA as well:

Here we report a comprehensive survey of the effect of cytosine modifications on DNA flexibility using a single-molecule cyclization assay: 5-fC remarkably increases DNA flexibility while 5-mC reduces and 5-hmC enhances flexibility, and 5-caC does not have a measurable effect. Molecular dynamics simulations elucidate the microscopic mechanism of the DNA flexibility changes induced by cytosine modifications.

They go on to describe in detail how this works (interested readers can study this open-access paper). Depending on the tags, they stiffen the strand to lift it from the chromatin, making it accessible to transcription machinery, or loosen the strand to allow it to wrap more tightly around the nucleosome. For example, "Methylation-induced increases in DNA stiffness likely arise from the restriction of the conformational fluctuations caused by the bulky methyl groups," possibly inactivating a gene. The 5-fc tag, by contrast, loosens the strand so that repair enzymes can get to repair a spelling error. These tags are not necessarily independent. The authors think they work together in a combinatorial way, modulating each other's influence. In addition, they appear numerically significant: the more 5-fc, the more stable the strand:

The presence of two copies of 5-fC in 147 bp (that is, 0.68%) of DNA of the 601 sequence results in a remarkable enhancement of nucleosome mechanical stability. This observation together with the reduced mechanical stability of nucleosomes containing many copies of 5-mC gives further support to our previously reported correlation between DNA flexibility and nucleosome stability: the more flexible DNA is, the more stably it is bound to histone octamer core and vice versa. The increased nucleosome stability induced by 5-fC may explain enrichment of 5-fC at poised enhancers and thetranscription start sites (TSSs) of low-expression genes:enhancement of nucleosome stability limits exposure of nucleosomal DNA to transcription machinery at enhancers and transcription start sites.

It's a work in progress that warrants further attention. But we're not done yet.

Tag Positional Roles

A commentary in Nature about a recent paper explores the impact of a methyl tag's position. Transcriptional regulation appears to be influenced not just by the presence of a methyl tag (CH3), but also by which nitrogen atom in an RNA base it is attached to. One such form, called m1A, was thought to be a bug; now it appears to be a feature.

Dominissini and colleagues' findings represent an intriguing step for epitranscriptomics. More work is needed tounderstand the mechanism by which m1A influences translation initiation and regulation, as well as the changes in its levels in stress responses. It will be exciting to see how this modification affects, and is affected by, RNA structure. Moreover, we need to know about the hypothesized proteins that are 'writers', 'readers' and 'erasers' of m1A and thus take part in the dynamics of its regulation. Finally, as yet another modified base has been identified in mRNAs, one wondershow many more exist: is this the final one, or are we just seeing the tip of the epitranscriptome iceberg?

Definitely exciting! Searching for "hypothesized writers, readers, and erasers" of these tags makes sense in light of the design perspective. They must be doing something important. But then, what is telling themwhat to do?

Signal to Noise

A paper in Current Biology explains how cells keep their focus on the signal in a noisy environment:

Transcriptional regulation is noisy, yet despite this variability,embryonic development reproducibly generates form and function. Recent work demonstrates that patterns of transcriptional activity in embryos are stably inherited through mitosis. These observations have implications for how accuracy arises in development.

What do the words "signal" and "accuracy" imply? Information, of course. This recalls our exploration of targets as a measure of intelligent design. It also makes one think of the chicken embryo sequence in Flight: The Genius of Birds that leads to a working chick at the end of a "highly elaborate dance" of many protein parts. The author of this paper puzzles over how the embryo gets to the target. One thing is certain: it works. The chick grows up to lay eggs that will grow more chickens that will lay eggs, maintaining the accuracy of the embedded coded instructions over thousands of generations.

Running the Replication Gauntlet

Think about the problem of maintaining genomic integrity during cell division. Machines are reading and duplicating DNA extremely rapidlywhile the cell is transcribing genes. This presents many challenges, not the least of which is avoiding extra copies of gene products. The number of proteins generated has to be tightly controlled, but now the cell has two DNA strands exposed to the transcription machinery, increasing in a stepwise manner. How does the cell keep the "DNA dosage" constant during this dynamic whirlwind of activity? How does the cell maintain "expression homeostasis"?

Now there's a challenge many of us never thought about. Three Israeli scientists explore possible answers in Science Magazine:

Genome replication introduces a stepwise increase in the DNA template available for transcription. Genes replicated early in S phase experience this increase before late-replicating genes, raising the question of how expression levels are affected by DNA replication. We show that in budding yeast, messenger RNA (mRNA) synthesis rate isbuffered against changes in gene dosage during S phase. This expression homeostasis depends on acetylation of H3 on its internal K56 site by Rtt109/Asf1. Deleting these factors, mutating H3K56 or up-regulating its deacetylation, increases gene expression in S phase in proportion to gene replication timing. Therefore, H3K56 acetylation on newly deposited histones reduces transcription efficiency from replicated DNA, complementing its role in guarding genome stability.Our study provides molecular insight into the mechanism maintaining expression homeostasis during DNA replication.

Information, Information, Information

There's a lot more one could say. PhysOrg talks about how certain proofreading enzymes physically "yank" on RNA to keep the splicing machinery from starting at the wrong spot. News from Nencki Institutein Poland explores why it is that one base substitution in non-coding DNAcan lead to disease (that's like having one spectator leave a stadium and affect the outcome of the game). A research site in Belgium adds an additional layer of complexity to the human protein landscape, revealing that shortened versions of proteins are not only conserved, but are tightly regulated and have functions. And Science tries to come up with a unifying model of epigenetic regulation, seeking to get a handle on the "intricate 'web' of proteins, chemical modifications, and RNA that together organize genomic DNA into chromatin."

Everywhere in life, it's information, information, information. Machines are reading, writing, and erasing signals, regulating transcription and translation, and ensuring the correct information gets the organism to the target. Information is duplicated, proofread, and buffered against changes. Information at one level regulates information at another level. It's hard to keep up with what science is learning about biological information.

If early geneticists could barely conceive of digital codes as the basis of life, early 1950s biologists hardly knew what the next sixty years would reveal about additional levels of information that regulate the newly discovered genetic code. Intelligent design theory is well positioned to lead the charge into biology's information age.

How the evo-devo revolution undermines Darwinism.

The Evo-Devo Revolution: LEGOs or Transformers?

Michael Denton March 7, 2016 3:31 AM

Editor's note: In his new book Evolution: Still a Theory in Crisis, Michael Denton not only updates the argument from his groundbreaking Evolution: A Theory in Crisis (1985) but also presents a powerful new critique of Darwinian evolution. This article is one in a series in which Dr. Denton summarizes some of the most important points of the new book. For the full story, get your copy of Evolution: Still a Theory in Crisis. For a limited time, you'll enjoy a 30 percent discount at CreateSpace by using the discount codeQBDHMYJH.

he revolution associated with the field of evo-devo (evolutionary developmental biology) has brought with it the revelation that underlying the development of all organisms is a set of highly conserved gene circuits and integrated developmental modules that guide and constrain phylogeny without regard for the immediate adaptive needs of species. This new knowledge poses a widely acknowledged challenge (and indeed self-evident challenge) to Darwinian evolution.

And it has also undermined completely the previous neo-Darwinian mechanistic view of organisms as analogous to contingent assemblages of LEGO blocks. According to the Darwinian model, organisms are infinitely plastic "additive functional assemblages" which can be changed -- LEGO block by LEGO block -- from one shape to any other conceivable novel shape bit by bit without any significant constraints. On the contrary, all the evo-devo evidence suggests that there are deep and profound developmental constraints that work against such infinite pliability.

The new conception of organisms arising from the discoveries of evo-devo might be termed the "Transformer" model. Just as in a child's Transformer toy action figure, the number of forms that can be reached by combining the basic parts in different ways is severely limited by the shape and properties of the component building bricks. The finite set of forms that can be assembled is prefigured into the basic properties of the constructional units. One might view the constraints imposed by the building units in a Transformer set (unlike LEGO bricks) as being analogous to those imposed in living things by the highly conserved toolkit components and developmental modules.

Consequently, pigs will never fly -- not only because of functional constraints (far too heavy!), but because of deep internal structural constraints in the way a pig is put together, which greatly restrict the available paths that adaptive evolution can take. On this view, the major novelties actualized during evolution could only have occurred if they were compatible with the pre-existing inner "developmental logic" of the organism, analogous to prefiguring of the Transformer components for a specific set of forms.

In other words, development rules! Or, more precisely, what rules is the collective constraints of toolkit elements, developmental pathways, and modules that underlie the ontogeny of every class of metazoan organisms.

Although there are some dissenters,1 the existence of highly conserved developmental genetic mechanisms, gene circuits, and so forth restricting the paths of evolution has led to the widespread adoption of what might be termed a new constraints paradigm among many evo-devo researchers, who acknowledge that the deep logic of development is bound to restrict the direction of variation and hence evolutionary change along limited paths.2 But what is really radical about the "constraints paradigm" is that the constraints restrict the paths of evolution without any regard for immediate adaptive function."3

The notion of deeply shared non-adaptive internal causal factors channeling evolution in such a manner is certainly heretical. It is small wonder that, as Gerd Müller and Massimo Pigliucci concede, many leading biologists feel that evo-devo presents "challenges to the received theory... so substantial that no reconciliation with the classical framework [Darwinian, incremental functionalism] is at all possible."4

References:

(1) Hopi E. Hoekstra and Jerry A. Coyne, "The Locus of Evolution: Evo Devo and the Genetics of Adaptation," Evolution; International Journal of Organic Evolution 61, no. 5 (May, 2007): 995-1016; see also Lindsay R. Craig, "Defending Evo-Devo: A Response to Hoekstra and Coyne,"Philosophy of Science 76, no. 3 (2009): 335-344.

(2) Wallace Arthur, Evolution: A Developmental Approach; Jerry Fodor and Massimo Piattelli-Palmarini, What Darwin Got Wrong (London: Profile, 2010), Chapters 2, 3, and 4; Günter P. Wagner, Homology, Genes, and Evolutionary Innovation (Princeton: Princeton University Press, 2014).

(3) Ron Amundson, The Changing Role of the Embryo in Evolutionary Thought (Cambridge: Cambridge University Press, 2007), 8.

(4) Gerd B. Müller and Massimo Pigliucci, "Extended Synthesis: Theory Expansion or Alternative?" Biological Theory 5, no. 4 (2010): 275-276, 275.