|
Author
|
Topic: ontogeny recapitulates phylogeny
|
Josh
Member
Member # 405
|
posted 06. November 2002 22:07
This started up on another post and I wanted to discuss it further, but the issue wasn't relevant to the other thread.
Question- Is the following section fair and unbiased without misrepresentation of facts?
So, I just got home and for review I decided to glance at my embryology text. It is “Patten’s Foundations of Embryology” Sixth Edition, Bruce M. Carlson, 1996. On page 38-39 it says
Recapitulation (title of section)
“The story of individual developmental sketches for us an approximate outline of the evolutionary changes passed through by our forebears. This concept is known as the biogenetic law of Muller and Haeckel. The general idea of recapitulation was first propounded by Muller (1864) on the basis of his studies of the development of invertebrates. Haeckel (1868) formulated his principles much more fully and called it the biogenetic law. In essence the law tells us that an animal in its individual development passes through a series of constructive stages like those in the evolutionary development of the race to which it belongs. More technically, and more succinctly, ontogeny is an abbreviated recapitulation of phylogeny.
In recent years the biogenetic law has been subjected to considerable criticism. Most of the objections to it have been directed against attempts to apply it too rigidly to details. It is readily apparent that recapitulation does not consist simply of adding more recent phylogenetic traits onto old ones. In particular, one would not expect the embryo to pass through stages in which many of the specialized structural features of present day lower chordates are emphasized, for many of these are specific adaptations that diverge from the mainstream of chordate evolution. Rather ontogenetic recapitulation is a conservative process which retains the basic ontogenetic stages of more primitive forms. Thus in a mammalian embryo only the most fundamental steps of early development and the establishment of major organ systems, such as the heart and large blood vessels, would resemble those of a fish embryo. There would be greater similarity between mammalian and reptilian or avian embryos (fig. 1-27), (Figure basically shows human, pig, reptile, bird embryos- all looking basically identical at “corresponding developmental stages.”) and this similarity would be apparent for a greater portion of embryonic life. In many cases ontogenetic processes leading to the formation of specialized structures in lower species are discarded or greatly reduced, and new ones are superimposed.
Often, however, a phylogenetically newer structure will make use of some component of the older one during the early phases of its development. This can be seen in the human placenta, the major organ of exchange between the embryo and the mother. The major blood vessels supplying and draining the placenta are homologous with those supplying the allantois, which subserves a similar exchange function in the embryos of birds and some lower mammals. In the human, the allantois itself remains vestigial but the allantoic blood vessels become incorporated into the phylogenetically newer circulatory system of the placenta.
IP: Logged
|
|
andyg
Member
Member # 415
|
posted 06. November 2002 23:05
I very much doubt that this is an appropriate topic for ISCID Brainstorms, but in my opinion as a developmental biologist, I feel the paragraphs you quote are fair and reasonable.
Since this is rather off topic, I would suggest you continue the discussion on the talk.origins newsgroup, which has a couple of developmental biologists who have discussed Haeckel in the past.
AndyG
IP: Logged
|
|
Paul A. Nelson
Member
Member # 26
|
posted 07. November 2002 10:42
A good formulation of the Biogenetic Law is "Ontogeny recapitulates phylogeny, except when it doesn't." ![[Big Grin]](biggrin.gif)
IP: Logged
|
|
andyg
Member
Member # 415
|
posted 08. November 2002 12:39
Paul Nelson Writ: quote:
A good formulation of the Biogenetic Law is "Ontogeny recapitulates phylogeny, except when it doesn't."
Actually, this is a good formulation of most natural laws, and underscores the provisional and tentative nature of science methodology.
AndyG
IP: Logged
|
|
Paul A. Nelson
Member
Member # 26
|
posted 08. November 2002 14:25
Will my HP laptop fall to the ground if I let it go in a gravitational field at a planet's surface? Yes, with the normal ceteris paribus clauses applying (no funny business with sudden very violent gusts of wind, et cetera).
Will any given metazoan ontogeny recapitulate its phylogenetic history? Dunno. Here's a coin; flip it and make your best guess.
"Von Baer's Law would be more useful if we knew where we should expect it to apply and where not." Richard Strathmann, "Larvae, phylogeny, and von Baer's Law," in Echinoderm phylogeny and evolutionary biology, eds. C.R.C. Paul and A.B. Smith (Oxford: Clarendon Press), 1988.
Yes, all scientific propositions are provisional and subject to ceteris paribus qualifiers. That doesn't mean, however, that Andy Grove's blood does not circulate (after all, he could be a cyborg), that I should ingest cyanide (I could be immortal and invulnerable to poisons), or -- to come at the matter from the other direction -- that any exception-ridden generalization nevertheless wins the honorific status of a "law." A law is just what ontogeny recapitulates phylogeny ain't.
[ 08. November 2002, 14:49: Message edited by: Paul A. Nelson ]
IP: Logged
|
|
charlie d.
Member
Member # 159
|
posted 08. November 2002 16:10
That I am aware of, pretty much nobody actually uses the term "biogenetic law" in biology (if not as a historical reference, often in quotes). There are 8 references to the term in Pubmed (13 for "ontogeny recapitulates phylogeny"). Certainly the term is never used the way people refer to Newton's Laws, or Mendel's. Incidentally, the latter are also subject to major caveats and restrictions Mendel could not have known about (such as genetic linkage, mutation, epigenetic effects, epistasis, etc etc) - although no one is arguing to take genetics out of the textbooks, so far.
Thus, to complain that Haeckel's "Law" does not behave like one seems very much like a straw man argument, or a matter of pure semantics.
As for the matter raised in the first post, I think the quote captures the essence of the concept of recapitulation and its limitations. Sounds reasonable to me, all in all.
IP: Logged
|
|
andyg
Member
Member # 415
|
posted 08. November 2002 17:04
I think Paul is gettng a bit worked up over nothing. Haeckel's biogenetic law is not taken seriously and has not been taken seriously for a very long time. Ontogeny does not recapitulate phylogeny in the sense that Haeckel meant it.
What *is* the case is that vertebrates have many characteristics in common, and this is especially true during their early development, when many common features can be observed shortly after the end of gastrulation. That is the only reason some textbooks used Haeckel's fraudulent drawings, not to illustrate Haeckel's law.
AndyG
IP: Logged
|
|
Moderator
Administrator
Member # 1
|
posted 08. November 2002 20:55
This thread is dangerously close to being closed down. It's bordering on a major re-hashing of old, worn-out issues.
If this thread is to stay alive, the participants need to start moving it in positive, new directions. What insights does Haeckel's biogenic law (correctly formulated) give us for research on biological systems? Why should some biosystems exhibit remarkable similarities at certain developmental stages? Why to they exhibit remarkable differences at other stages? Is this something that we would expect based on a Darwinian, Design, or Self-Organizational viewpoint? How would we test such an idea?
Other possible positive directions of discussion are left for participants to explore. But be warned--the discussion needs to move in positive, new (possibly even insightful!) directions instead of re-hashing old, worn-out arguments about Haeckel's embryos. I think having a discussion of embryological development here on Brainstorms would be very beneficial, and I'm waiting to see whether such a fruitful discussion can be generated.
[ 08. November 2002, 20:57: Message edited by: Moderator ]
IP: Logged
|
|
Josh
Member
Member # 405
|
posted 11. November 2002 17:15
quote:
This thread is dangerously close to being closed down. It's bordering on a major re-hashing of old, worn-out issues.
My main interest in this topic was investigating more closely the claims about embryology that I was taught and how that matches up with current science thinking. For myself, this teaching (and from the professor) conveyed the message that evolution was in fact confirmed positively by analysis of embryological development due to the biogenetic law as formulated in this text. No critical analysis was given outside to say that inapplicable criticisms are generated against the claim (as the text says to me.)
Since I am not an embryologist I would be very interested if those who claim expertise in this area were to address the moderator's questions. I will offer my answers to those questions relevant to my understanding.
quote:
Why should some biosystems exhibit remarkable similarities at certain developmental stages?
To me it seems obvious that if different organisms generate limbs and anatomical features that are similar, they will use a similar developmental paradigm to implement the construction of these features. Whether the pathway of development of different anatomical features suggests design or evolution is another question.
quote:
Why to they exhibit remarkable differences at other stages?
A parallel question for this is: Do we have enough information about the consequences of the differences in developmental stages to demonstrate a purpose for such differences between organisms. For example, is it known that spiral verses other-pattern cell divisions in the early embryo have specific consequences for later development? And are these differences linked to overall construction of the organism's anatomy, etc.?
I was not looking to be argumentative, but hoping to learn something from starting this, because I personally am not completely sure whether the text I posted is misrepresentative of the facts or exactly on target and fair.
IP: Logged
|
|
andyg
Member
Member # 415
|
posted 11. November 2002 19:26
quote:
Why should some biosystems exhibit remarkable similarities at certain developmental stages? Why to they exhibit remarkable differences at other stages?
Interesting questions. I'll restrict my answer to vertebrate embryos, since these were the subject of Haeckel's fraudulent drawings.
The interesting feature of embryonic development that the moderator alluded to above is what is known as the "developmental hourglass". Simply put, this means that vertebrate embryos start off by looking very different from one another (this is the first reason why Haeckel's idea about ontogeny recapitulating phylogeny was wrong). At a certain point in development, they tend to have very many features in common, before diverging in their final forms again.
The point at which embryos tend to converge in their appearance is shortly after the events of gastrulation. This is the process by which a ball or disc of primitive cells undergoes rearrangements to from the three "germ layers" of the embryo - the ectoderm (which will form the skin and nervous system), the mesoderm (which will form the blood, bones, muscle, dermis, kidneys, etc) and the endoderm (which will form the gut, lungs, liver, pancreas, etc). The important thing to understand is that the process of gastrulation is not simply characterized by cells rearranging. During this process, different populations of cells are signaling to one another by releasing protein signaling molecules that act on other cells to induce a particular cell fate. Thus, the events of neural induction, mesoderm induction and endoderm induction are all occurring during gastrulation, and all require that different populations of cells are brought into close apposition for this cell signaling to occur. The signaling molecules and pathways in this process seem to be remarkably well conserved across vertebrates.
The second point to bear in mind is that different vertebrate eggs have vastly different amounts of yolk at fertilization, and this places constraints on how the embryo grows to deal with this amount of yolk. For example, some amphibians internalize the yolk in their eggs during gastrulation, so that the yolk ends up inside the endoderm layer. Frogs don't need a huge amount of yolk since the tadpole begins to feed from a relatively early age. Other amphibians and fish develop on top of the yolk, but can swim around with the yolk attached like a little balloon. Birds and reptiles, on the other hand, have huge amounts of yolk, making it physically impossible for the embryo to envelop the yolk during development. Instead, the embryo develops on top of the yolk as a disc, and sends a network of blood vessels over the yolk to act as means of absorbing the yolk, taking up gas and depositing waste material. Mammals tend to develop in a similar way to birds and reptiles, (which isn't surprising since they are thought to share a common ancestor with them), but have now dispensed with the need for yolk at all, despite retaining the yolk sac, for example.
Put these two constraints together - the requirement for different cells to be placed in apposition to one another during gastrulation, and the vastly different amounts of yolk in different vertebrate eggs, and the reason why a "developmental hourglass" might occur becomes apparent. Different embryos start off by having very different forms to accomodate (in part) the very different amounts of yolk. However, they still need to undergo gastrulation, and this places constraints on the possible conformation the embryo could take on immediately after gastrulation.
Many embryologists doubt whether the concept of the "phylotypic stage" - a sort of idealized rendering of the "hourglass" where vertebrate embryos share the greatest number of features - is actually a very useful one. My own feeling is that it serves a pedagogical purpose in focusing attention on the common mechanisms in early embryogenesis. However, many people have pointed out that there are also many differences in early vertebrate embryonic development. Indeed, Michael Richardson and colleagues published a paper entitled
"There is no highly conserved embryonic stage in the vertebrates: implications for current theories of evolution and development." (Anat.Embryol. 196: 91 - 106. )
Scott Gilbert comments on this paper at:
http://zygote.swarthmore.edu/evo5.html
quote:
The Richardson et al. paper does not dispute that there is a highly conserved embryonic stage among the vertebrate classes. Indeed, at the late tailbud stage, vertebrate embryos of most all classes possess "somites, neural tube, optic anlagen, notochord, and pharyngeal pouches." However, these authors do criticise the notion that this stage is nearly identical in all species and that differences beteen the classes can be resolved only after subsequent development. Rather, they discover significant differences between groups. Size is one distinctive marker. The scorpion fish embryo is 700 microns long at the tailbud stage, while the mudpuppy salamander measures some 9 millimeters. Heterochrony is another problem. In some species of direct developing frogs, and in monotreme mammals, limb buds are already present at the tailbud stage, whereas in other species, these are not seen until significantly later. Birds are characterized by their prominent mesencephalon. Whereas most amniote embryos have a heart by this stage, the zebrafish does not. (Teleosts such as zebrafish are even the exception to the rule mentioned above. They eventually possess a notochord, somites, pharyngeal pouches, etc., but they do not have the pharyngeal pouches until after the tailbud stage)."
The embryos in the above image have many common features. In fact, row II of Haeckel's drawing above is actually not a bad representation of them. They do however, have differences - attached yolk, for example, and differences in size and the relative maturity of limbs, for example. My own view is that these differences stem from the fact that after gastrulation, different regions of the embryo develop in a much more autonomous fashion, with a corresponding decrease in the number of constraints on the speed at which a particular tissue develops.
quote:
Is this something that we would expect based on a Darwinian, Design, or Self-Organizational viewpoint?
Design, like divine intervention, can obviously predict anything it wants, based on the motives of the designer, so the above situation vis-à-vis different vertebrate embryos is entirely consistent with a design viewpoint. However, the evidence for a common evolutionary origin of vertebrates is also compelling, not just at the morphological level, but also at the genetic level.
AndyG
[ 11. November 2002, 21:07: Message edited by: andyg ]
IP: Logged
|
|
Moderator
Administrator
Member # 1
|
posted 15. November 2002 01:55
AndyG,
Excellent post! This thread is definitely "looking up". Since the moderator usually comes down on people, I thought I'd go ahead and "come up" on you, commend you on a well-written and scientifically refreshing post. We need more data, less hyperbole sometimes--and your post was excellent in that regard.
quote:
Design, like divine intervention, can obviously predict anything it wants, ...
Is this necessarily true? Remember that Design is not the same as devine intervention...I can design an automobile, but I'm definitely not devine. It seems that design can happen without miracles, and even without any sort of "creation" per se (again, imagine building an automobile--the matter it comes from was pre-existing and was molded and formed to human uses--no creation, no miracles). What does the science of engineering tell us about commonality between designed things? Do human engineers utilize certain patterns in multiple circumstances, and what might those circumstances be? What does the science of engineering tell us about biological development? When human engineers are designing semi-autonomous systems (self-adjusting, maybe even self-constructing), do they utilize certain common motifs? What motifs do they use that are different? Can any engineers out there cite any examples? Any TRIZ experts want to weigh in?
quote:
However, the evidence for a common evolutionary origin of vertebrates is also compelling, not just at the morphological level, but also at the genetic level.
It would be good to discuss more specifics here. What do we expect in terms of similarities or differences in the genetic networks regulating body plans in organisms that have evolved via a Darwinian pathway? Remember, there are both similarities and differences (at the whole-organism and genetic levels), so the question is: does the pattern of similarities and differences fit better with what we might expect from an intelligent designer (like ourselves, an engineer (or engineers)), or from what we'd expect from a trial-and-error process operating iteratively over time? The key question that I'd like to see discussed, is how do we make this distinction? How do we know which predictions really do follow from a Darwinian or engineering standpoint? Are we resigned to just say that they both predict the same thing, and hence the evidence doesn't help us any, or can we make statements about what sorts of evidence really would tilt the scales toward one explanation or the other (though not an absolute proof either way)? If we can get our minds just a bit closer to an understanding of these questions (and do so in a constructive, non-hostile, non-battlewarrior mentality), I'll consider this thread a success. It's looking good--keep it up!
IP: Logged
|
|
Argon
Member
Member # 276
|
posted 15. November 2002 11:24
The moderator considers the notion that one cannot make predictions about the creative output of unspecified designers. He asks: "Is this necessarily true?"
Yes. Unless the limitations and specifications of the designer can be determined it is difficult to predict what is possible. Unspecified designers do unspecifiable things. This is one of Sober's arguments about creation science in his book "Philosophy of Biology".
Although many of the leaders in the ID movement assume that the designer involved in the development of life was a god and in fact have a particular god in mind, almost everyone agrees that "design" capabilities are certainly not limited to gods. Nor is it necessary that the creation and subsequent modification of life required divine intervention: Just, really, really advanced technology and long attention span.
The appearance of commonality between designed things depends on the intentions, mechanisms of implementation, and technological capabilities of designers. One rule of thumb is that the more technologically advanced one is, the greater the number of options/routes available.
The human approach to biological engineering often involves a lot of horizontal transfer, at least for the addition of "new" functionality. With our current scientific knowledge, we are more likely to transfer slightly modified systems from other organisms than produce a new one or rearrange an existing one. However, greater knowledge could permit the ability create de novo designs for proteins and regulatory networks and to predict what various modifications will have on organisms and ecological relationships. This is especially relevant if we take into account the technological advances that a set of designers would likely achieve over the course of a few billion years. So in either case -- low capability (e.g. humans), or high technology entities (e.g. ???) -- I think it would be hard to make a case for patterns that generally mimic common descent.
As Arthur Clarke once wrote: "Any sufficiently advanced technology is indistinguishable from magic." With regard to beings with the expertise to travel across the galaxy and to design & guide the subsequent modifications of life over the course of several billion years, I suspect the question is not how "human-like" design would appear in biological systems but, "What does magic look like?" Design can, in principle, look like common descent. It can also look unlike common descent (For example, consider arguments made by Discovery Institute fellow, Walter ReMine). Without specifying a particular designer and design mode in advance, it's not easy to make robust predictions. It might also help to know the timing, number and extent of designer-driven "interventive" acts to make sure a designer's specifications are properly compared against the acts attributed to the designer.
Added 11/17/2002:
TRIZ examines human-generated processes in an attempt to derive general principles of human technological advancement. Thus TRIZ is used to characterize a type of designer based on what the designer produces. But the first problem is that TRIZ neither predicts what an unknown designer is capable of creating nor how it could create. Second, it cannot answer the question of whether the development of life on earth over the eons truely represents step-wise technological advancements on the part of "biotic engineers" over the course of a few billion years, or whether the biotic engineers had the technology to created a fully functional, modern ecology in say, six days with no need for "experimental" models (such as trilobites). Such answers require knowledge about the technological capabilities and intentions of a designer, which in the case of terrestrial life has not been determined. So I'm not sure how one would deploy TRIZ methodolgies in this instance.
[ 17. November 2002, 14:56: Message edited by: Argon ]
IP: Logged
|
|
andyg
Member
Member # 415
|
posted 15. November 2002 19:54
The moderator poses the question : can design, as I have suggested, predict anything it wants, in a similar manner to divine intervention? I would concur with Argon's post. A sufficiently intelligent designer could design anything we might be able to imagine. I think this discussion is best left to another thread, but I will make one point to lead into the main point of this post on vertebrate embryology. Intelligent design creationists are rather difficult to pin down about when they believe a design event occurred. Take three simple-minded hypotheses:
- Vertebrates arose by evolution from a common ancestor
- A designer created a common vertebrate ancestor and then left it to its own devices to evolve and diversify.
- A designer created modern vertebrates in the last few thousand years and cunningly altered each vertebrate species to make it look like they evolved from a common ancestor.
Depending on the nature and motives of the unspecified designer, the pattern of modern vertebrate life could be construed as having arisen by evolution, design , or both. In this regard, I have always been bothered by Behe's acceptance of common descent, but at the same time his implication that designers must have intervened at subsequent times to introduce more modern systems like the clotting cascade.
Anyway, this is getting off the topic of vertebrate embryology. To return to the issue at hand, with a quote from the moderator regarding the evidence for vertebrate evolution:
quote:
It would be good to discuss more specifics here. What do we expect in terms of similarities or differences in the genetic networks regulating body plans in organisms that have evolved via a Darwinian pathway? Remember, there are both similarities and differences (at the whole-organism and genetic levels), so the question is: does the pattern of similarities and differences fit better with what we might expect from an intelligent designer (like ourselves, an engineer (or engineers)), or from what we'd expect from a trial-and-error process operating iteratively over time?
Here are some genetic features of vertebrates that I believe are consistent with the idea of evolution from a common ancestor:
- Synteny. This refers to the fact that the arrangement of genes on particular regions of chromosomes are conserved between species, for example, between mice and humans, or humans and fish. This has proved very useful in identifying human disease gene candidates, for example.
- Gene duplication. We have examples of duplication both in "cis" - whereby a copy of a gene is placed adjacent to the original gene on a chromosome, and in "trans" - where a cluster of related genes is duplicated onto another chromosome. More primitive vertebrates, such as lampreys and hagfish tend to have fewer copies of genes in "cis" and in "trans". Even more primitive non-vertebrate chordates have still fewer copies of such genes. Examples of gene clusters are Hox genes, Dlx genes, Irx genes and so forth. References provided on request.
- Conservation of gene function. For example, the Math1 transcriptional regulatory gene from mice can functionally replace its fruit fly counterpart, and vice versa. The interesting and consistent wrinkle in this is that sometimes, a member of a gene family that has undergone duplication can only rescue some of the functions in other species, illustrating how duplication and divergence of genes can lead to a sharing of functions between genes.
- Conservation of regulatory regions of DNA. Gene activation is controlled by regions of DNA that lie on either side of a gene, but do not themselves code for protein. These so-called promoter or enhancer elements direct the correct spatial and temporal activation of genes during development. A nice example of this is the promoter region of the Dlx3 gene. Dlx3 is activated in the limbs, ear, etc in frogs, and in the limbs, ear, hair follicles and mammary gland in mice. If you take the frog promoter region and use it to drive expression of a reporter gene in mice, the frog promoter still activates gene expression in mouse hair follicles and mammary glands, even though frogs do not possess such structures! Sean Carroll is doing really nice work to show how small duplications or deletions in the promoter regions of closely related species of fruit flies can change the banding pattern on fly abdomens.
- The twin nested hierarchy - in other words, species that are grouped closely together on their morphological characteristics also tend to be grouped together on the similarity of their genome organization and gene sequences.
The point in my previous post which you called me on is that a designer could design all these features if they wanted to. Or they could have designed a prototype and let it evolve to give the present variety. At this point, some people appeal to the principle of parsimony and state that if the evidence can be explained by evolution or a designer, then evolution wins by appeal to parsimony. As a scientist, this is a point of view I have some sympathy for.
Different intelligent design creationists have different views on whether evolution of the sort that lead to vertebrate diversity occurs at all - for example, Michael Behe versus Jonathan Wells. Perhaps it's time to pause and ask someone else for their ideas. I would be particularly keen to hear from Paul Nelson, as I gather he is working on a paper on vertebrate embryology at the moment.
AndyG
IP: Logged
|
|
|
Printer-friendly view of this topic