This is a synopsis and review of

Homology, An Unsolved Problem
by Sir Gavin de Beer
Oxford University Press (1971) ISBN 0 19 914111 8

The post is longer than I would normally write. It is intended to be informative, to resolve what light de Beer sheds upon homology as a problem, and as an object lesson in how to make serious consideration of a cited reference.

I lack the depth of understanding required for this review to have any professional weight, but it may serve as a useful pointer. It is as fair assessment as I can honestly manage in a reasonable time.

The text is a short 16 page monograph that refers to prior work for “further reading”. Following some of this “further reading” was useful; but difficult. Citations are not included as part of the text, but must be inferred from the bibliography. All references are now at least 35 years old, and so I was generally forced to use more modern review articles to explore them.

My review is structured with the same twelve sections as the original.

1. The concept of homology. A historical review, going back to pre-Darwinian and pre-evolutionary explanations for homology, including the first detailed studies in the eighteenth century that sought to establish whether or not particular structures really are homologous. De Beer describes Darwin’s work as a bombshell that had a profound effect on the explanation of homology by introducing the notion of common ancestors; but did not affect how homology was identified.

De Beer concludes with a teaser for what follows:

So, provided with a cast-iron explanation in terms of affinity, of inheritance in evolution from a common ancestor, it looked as if the concept of homology was at last soundly based and presented no more problems of principle; however, as will be seen below, it unfortunately does.

2. Homology in plants: leaves and flowers. This is a straightforward description of the structures of leaves and flowers, showing structural evidence for homology that was first identified by Goethe in about 1790. No problems here; and subsequent work in genetics has abundantly confirmed this classic example of unambiguous homology.

3. Homology in animals: the ear ossicles. This is another classic example, without any conceptual difficulties; the homology of the bones of the inner ear in mammals with bones that make up part of the jaw in reptiles. He concludes:

What makes this study even more significant is that the results of comparative anatomy are confirmed by those of palaeontology, for there are fossil reptiles that show advances towards the mammalian condition, and the superseding of the quadrate-articular hinge of the lower jaw by the squamosal-dentary articulation. All this evolution took place without any functional discontinuity. It is a sobering thought that ever man carries in his ear ossicles the homologue of the lower jaw hinge of his reptilian ancestors. This is one of the most demonstrative examples of how comparative anatomy can determine homology of structures inherited from common ancestors in evolution.

4. Conservative effects of homology. This section considers, by an example, the consequence of a well known homology on certain other conserved structures, explaining an otherwise inexplicable anomaly.

Diverse vertebrate species have “branchial arches” in the embryo (sometimes called “gill slits”). In different species, these arches develop into different structures. The “recurrent laryngeal nerve” passes behind branchial arch number six in the vertebrate embryo. During embryonic development, several of the arches disappear, and others develop into various structures, in diverse manners for different species.

The result is that in mammals, the recurrent laryngeal nerve ends up with what would appear to be a very poorly designed route, “running backwards and looping round the ductus arteriosus, then runs forward again to innervate the muscles of the larynx.” In humans, the nerve is thus several inches longer than we should expect in a well designed organism; and in a giraffe it is several feet longer. De Beer shows how homology explains this curious anomaly.

The explanation is the homology between the mammalian ductus arteriosus and the 6th arterial arch of the fish, which is respected in descendent forms, resulting in apparently anomalous conditions.

5. The displacement of homologous structures. This section shows how homologous structures can, in the course of evolution, shift position so that they are expressed by different tissues in the embryo. This is a plain refutation of some confused criticisms raised by Michael Kent and Michael Denton. In this section, De Beer is not presenting a problem with the concept of homology (we’ll get to that in later sections) but clarifying a feature of homology that was surprising at the time.

De Beer uses to the vertebrate forelimb illustrate his point.

There is no doubt whatsoever that the forelimb in the newt and the lizard and the arm of man are strictly homologous, inherited with modification from the pectoral fins of fishes 500 million years ago. They have identical elbow and wrist joints and their hands end in five fingers. The bones and muscles that they contain also correspond. But a minute examination of their comparative anatomy reveals the astonishing fact that they do not occupy the same positions in the body. The limbs of vertebrates are always formed from material that is contributed from several adjacent sections of the trunk. So, in the newt the forelimb is from trunk segments 2, 3, 4 and 5; in the lizard from 6, 7, 8 and 9; in man from trunk sections 13 to 18 inclusive. [...]

[...] in the course of evolution, transposition has occurred; new adjacent segments further back in the trunk have been drawn into contribution to the formation of the limb, and segments further forward, which previously contributed, cease to do so.[...]

These examples illustrate the important principle of the pattern which is where the problem of homology lies, not in identity of position in the body.

That is, the problem of homology is not that different embryonic tissues give rise to homologous organs. This is a non-problem.

As a minor aside, a similar phenomenon has now pretty much resolved the best argument that used to be raised against the close relationship of birds and dinosaurs. See Digit numbering and limb development by Paul Myers, at the Pharyngula Blog. This is a caustic response to confusions published by Dr Sarfati at Answers in Genesis, but don’t be put off. Paul is very good at explaining the scientific details. The technical description is “A solution to the problem of the homology of the digits in the avian hand.”, by Wagner GP and Gauthier JA (1999), in PNAS 96:5111-5116.

6. Serial homology. “Serial homology is something really a misnomer.” This section is an aside to the main thread of argument. It concerns not the association of organs with their representatives in a common ancestor, but similarity between organs in an individual. For example, forelimbs and hind limbs are “serially homologous”, showing very closely related structure.

This is not real homology, as forelimb and hindlimb cannot be traced back to any ancestor with a single pair of limbs. At most it might be said that there had been reduplication of a pattern.

7. Latent homology. This covers the case in which related structures may be homologous, and yet not visible in the phenotype of the common ancestor. This is starting to get close to a “problem” in homology. Several examples are given; most especially:

[...] the problem of spiral cleavage. This is a very precise set of manoeuvres by which the fertilized egg is cleaved. [...] Spiral cleavage occurs in polyclad turbellarians, nemertines, marine annelids, and molluscs other than Cephalopoda. It surely indicates a general affinity between the different groups in which it is found, because it is difficult to see how such a complicated mechanism could have been evolved separately in each group, and this affinity is supported by other embryological and morphological considerations. But did the common ancestor of these groups itself develop by spiral cleavage? It is impossible to say and difficult to assert, because in many species of these groups it does not occur.

This example might be grist to the mill for an “intelligent design” hypothesis, in which related themes appear independently by design, without a related starting point in a common ancestor. I deliberately refrain from a critique of that hypothesis here; but would be inclined to engage discussion of this point should it arise.

A survey of the literature indicates that that conventional biology identifies spiral cleavage with protostomes, and radial cleavage with deuterostomes.

This is a very deep distinction within Animalia; and so unsurprisingly the details of phylogenetic relationships remain open to considerable debate. As an example of this debate, see The new animal phylogeny: Reliability and implications, by Adoutte et. al., in PNAS Vol. 97, Issue 9, 4453-4456, April 25, 2000.

The “problem” of spiral cleavage being incompletely characteristic of protostomes seems to be pretty much resolved. Some lineages have lost this characteristic, and mutations that have this effect are known. This is strong circumstantial evidence that the common ancestor had spiral cleavage, and that loss of this feature is a secondary characteristic in some lineages. For more details, see Cleavage patterns and the topology of the metazoan tree of life, by James W. Valentine, in PNAS Vol. 94, pp. 8001-8005, July 1997.

There is lots of scope in this section to explore real substantive disputes on evolutionary relationships, but not much basis for seeing a fundamental problem with the notion of homology.

8. Homology and functional change. Homologous organs can have different functions in different organisms. No “problems” are noted in this section, and several examples are given. Here is the first:

There are several proofs of this, of which one of the simplest is the case of muscles and electric organs in fishes. [...] In some fishes [...] the muscles of certain parts of the body are modified to produce electric organs [...] powerful enough to deter predators and to kill prey. As it was difficult to imagine how these specializations arose by natural selection, and what advantages could have been conferred by initial states of such specialization, Darwin warned that ‘it would be extremely bold to maintain that no serviceable transitions are possible [...]’ This prophecy has been fully verified [...] the weak electric discharges [...] of certain fishes function in a manner analogous to radar and provide the fish with information of the proximity of other objects, [...]

9. Non-homology This section shows how morphology can provide proof (in the scientific sense of the word proof, of course) that “certain organs and structures are homologous, it can also show that others are not.” This is not a claim that homology can always be reliably identified or refuted; but a claim that there are some cases in which homology can be reliably inferred or refuted. Examples are given.

10. Homology and embryology. This is the longest single section. It does not identify problems with the concept of homology, so much as difficulties with identifying homology. Various common assumptions about homology are refuted, and some of those refutations still stand today. The section makes three major points, and that I present with my own subheadings.

(10.1. Location of embryonic cells giving rise to homologous structures.) As an example, the “obviously homologous” structure, the vertebrate alimentary canal, may arise from cells in the roof, or the floor, of the embryonic gut cavity (sharks, or lampreys), from both roof and floor (frogs), or from the lower layer of the embryonic disk (reptiles). In conclusion:

correspondence between homologous structures cannot be pressed back to similarity of position of the cells of the embryo or the parts of the egg out of which these structures are ultimately differentiated.

This is closely related to the information in section 5, and is not really a problem.

(10.2. The master organizers for inducing development of homologous structures.) De Beer speaks of the diffusion of “substances from a master structure called an organizer” that will induce cells to undergo differentiations and form certain structures. He presents experimental evidence of cases in which distinct species use different “organizers” for triggering development of homologous structures. De Beer’s conclusion:

homologous structures can owe their origin and stimulus to differentiate to different organizer-induction processes without forfeiting their homology.

There are two major examples given; formation of the lens in the eye of different species of frog, and formation of the brain and spinal cord in vertebrates and tunicates. In both cases the experiments, though ground breaking, were crude by modern standards, and have since been investigated more thoroughly to elucidate the “organizers” involved.

Poor citation makes this unclear, but the frog experiments are most likely those performed by various researchers from 1901 to about 1904. Some details and better references are available in the review article, Sequential activation of transcription factors in lens induction, by Hajime Ogino and Kunio Yasuda, in Development Growth & Differentiation, Vol. 42 Iss. 5 p 437, Oct 2000. This review pointed to a systematic re-investigation of lens formation, at Reinvestigation of the role of the optic vesicle in embryonic lens induction, by RM Grainger et. al., in Development, Vol 102, Iss 3, pp 517-526 (March 1988).

It seems that de Beer was simply wrong. The evidence for diverse organizers is weak, and his conclusion poorly supported in the light of more careful experiments. Grainger et. al. have proposed a fairly complex multi-step process for lens formation, which is apparently well conserved.

The second example, of brain and spinal cord, has also been subject to considerable study since de Beer’s book. De Beer notes that “in true vertebrates, the spinal cord and brain develop as a result of induction by the underlying organizer; but in the ‘tadpole larva’ of the tunicates, which has a ‘spinal cord’ like the vertebrates, it differentiates without any underlying organizer at all.” De Beer is probably reporting conclusions of Spemann (uncited) who originated the organizer concept. But those conclusions are wrong. The organizers in question do not induce development of the central nervous system, but rather suppress development of epidermis. This is explained in Organizing the Embryo: The Central Nervous System, part of an on-line biology text by Dr. John Kimball. A more formal paper is Head induction by simultaneous repression of Bmp and Wnt signalling in Xenopus., by Andrei Glinka et. al., in Nature 389, pp 517-519 (2 October 1997). Xenopus is a vertebrate (frog). Many more references could be found by following a trail from this paper.

That is, brain development is the default, in both tunicates and vertebrates. But vertebrates have since evolved an additional feature, which must be suppressed in cells to allow the default to proceed. This is confirmed in Xenopus Embryos as a Model to Study the Genetic Mechanisms of Brain Development, by A. G. Zaraisky, in Molecular Biology, Vol. 38, No. 1, 2004, pp. 34–39 (trans from a Russian journal, 2004).

It would be rash to simply presume that “organizers”, or factors inducing formation of homologous structures, definitely cannot evolve or become co-opted from other parts of the genome or embryo. But I would expect this to be unusual; and the examples given by de Beer for this alleged phenomenon turn out to be invalid in the light of further research.

This research lends strong support to the common descent model, though in practice it is not necessary. Common descent is not seriously questioned by any rational biologist, and it is even accepted by many in the intelligent design movement who apparently propose that some designer has been stepping in to tweak lineages from time to time.

(10.3. Homology and germ layers.) Here de Beer refutes a claim that homologous organs must always arise from the same germ layer in the embryo. He cites his own work, The Differentiation of Neural Crest Cells into Visceral Cartilages and Odontoblasts in Amblystoma, and a Re-Examination of the Germ-Layer Theory, in Proc. of the Royal Soc. of London. Series B, Biol Sci, Vol 134, Iss 876, pp. 377-398 (1947).

I’m uncertain on this. I’d welcome comment from a biologist in the house. As I understand it, de Beer is recognized as having disproved a germ layer theory developed around the turn of the century; but the concept of gastrulation and germinal layers remains vital in embryology. This includes the notion that various structures are developed from various layers. So what did de Beer disprove here? I’m not sure.

De Beer speaks of identifying the germinal origin of cells by certain indicators; such as “melanin granules” to indicate ectodermal cells, and “small globules of yolk” to betray endodermal and mesodermal cells. On this basis, de Beer infers that cartilage in jaws and visceral arches are ectodermal. But that is fine; it appears to match modern germinal models. De Beer also claims that tooth enamel can arise from ectodermal or endodermal cells. In this, I suspect he is wrong. Enamel arises from ectodermal cells, not endodermal.

Anyone who can clarify, please speak up!

11. Homology and genetics. This, finally, identifies the major “problems”. It starts with dramatic flair...

Because homology implies community of descent from a representative structure in a common ancestor it might be thought that genetics would provide the key to the problem of homology. This is where the worst shock of all is encountered.

He has two points, the first of which is no problem at all, and the second of which is more interesting, but wrong. Again, the subheadings are my own.

(11.1. Identical genes for non-homologous structures.) There are two major cases considered. The first is a gene in fowls, which controls formation of a crest of feathers, and also controls a cerebral hernia (knob in the skull) to accommodate it. The two structures are not homologous, yet controlled by the same gene. The second is the gene ‘antenna’, which induces Drosophila to produce an extra antenna instead of an eye. The conclusion:

characters controlled by identical genes are not necessarily homologous.

There is no particular problem here. The genes in question are control genes, that are able to trigger expression of other genes, that are not identical. The genes actually coding for non-homologous structures are different, but triggered by the same factor. This is fascinating, but not remotely in conflict with common descent.

(11.2. Different genes for homologous structures.) This claim I find surprising, but I don’t think the case is made very well. The example is the “eyeless” gene in Drosophila, which de Beer speaks of as a gene that “deprives its possessor of eyes”; though it is recessive.

This is not quite true. Eyeless is actually a gene that enables or triggers the formation of eyes. It was named eyeless because a mutant allele prevented it from working. A fly homozygous for the defective allele does not form eyes.

By breeding pure homozygous populations for this allele, experimenters establish a population of flies without eyes. After a time, offspring with eyes appear in that population, yet the eyeless allele is shown to still be present by its ill effects when mated back into the wild stock.

De Beer concludes:

What has happened during inbreeding is that all the other pairs of alleles making up the gene complex have been reshuffled until a gene complex has been produced that prevents the phenotypic manifestation of the ‘eyeless’ allele. Other genes must therefore deputize for the absent normal gene that controls the formation of eyes. But why should they, and by what mechanism? Nobody can deny that the restored eyes that develop in genetically ‘eyeless’ stocks are homologous to the original normal eyes.

Before one can say whether this phenomenon truly indicates a disconnect between genotype and phenotype, one really needs to know more about the processes by which these ‘eyeless’ stock actually express eyes. Basically, this work actually indicates a phenomenon known as “variable expressivity”. It is a complication of conventional Mendelian inheritance, in which an allele may express to differing degrees in different individuals. The full causes of this are not known; though it is likely associated with other genetic information, and of course is subject to selection. I don’t have a good reference for variable expressivity in the eyeless allele; and I would very much like to find one.

De Beer’s major reference is

Morgan, T.H. (1929) The variability of eyeless. Publs Carnegie Instn. 399, 139.

I could not find this paper, though I found other articles referencing it. The word “The” in the title should probably be omitted.

This, then, is the really major issue raised in the paper. The conclusion of de Beer is

Therefore, homologous structures need not be controlled by identical genes, and homology of phenotypes does not imply similarity of genotypes.

This is a very strong and surprising conclusion; but I believe that work in understanding development indicates that de Beer is wrong.

12. Revision. De Beer begins his conclusion thus:

It is now clear that the pride with which it was assumed that the inheritance of homologous structures from a common ancestor explained homology was misplaced; for such inheritance cannot be ascribed to the identity of genes. The attempt to from ‘homologous’ genes, except in closely related species, has been given up as hopeless.

He was wrong. Many homologous genes have been found in very diverse species, and continue to be found. The findings fit the basic model of common descent with spectacular success, and are rightly identified as proof of common ancestry sufficient for any reasonable investigator. The eyeless gene is a case in point; it is homologous to Pax6 in humans. There is an enormous literature on this.

De Beer well illuminates complexity of the homology concept, but his arguments for a disconnect with genetic homology are wrong, and have been disproved by the enormous strides made in genetics over the last thirty years.

Cheers -- Sylas

Sylas, thanks for the thoughtful review, and for bringing us up to date on some of the research de Beer refers to. I am sure that it must have taken some time to write. That, unfortunately, is a commodity that is in short supply for me right now. My job is very demanding, and my wife has pointed out to me that I have neglected precious time I should have devoted to family participating in this thread.

I was surprised to find so many participating in this thread who are opposed to ID, and that not a single person besides me jumped in to defend it. You outnumber me too well, and I simply do not have the time to engage all of you. So if I did not respond to your post, I ask your forgiveness - this does not mean that your post was not well-written or unworthy of attention.

I am just a science layman and I am sure there are very many who are much better equipped to engage in a debate such as this than me. That, combined with the limited time that I have, has resulted in some errors in my posts, for which I apologize. I thank you for pointing them out to me, as the last thing I want to do is argue from error or misconceptions.

I would like to leave you with the following thoughts. Although our knowledge of it is advancing, life is mind-boggingly complex, and human knowledge of life is still very incomplete. Even in areas in which our knowledge seems complete, our understanding or comprehension of that knowledge often lags far behind.

The overall lay of the land can appear very different from the ground than from the air. The threads of a large tapestry can lead us to one conclusion about it when only one portion of it is within view, but when we back away, the "big picture" is not what we expected. Our limited knowledge should give us humility - a recognition that we could be wrong in some areas, and a tolerance for others who with integrity and rigor have honestly come to differing conclusions. I don't mind intolerance towards sloppiness or error - I appreciate it, and recognize that some ID'ers have been guilty of it. But I do ask that you not prejudicially judge all by some, or even all by most. That is just as wrong as prejudging all black men as lazy because all of those that you have met seemed to be.

Many of you have expressed a great deal of intolerance towards the theory of ID. Some of you no doubt believe that this is justified, but it has been around for a long time. As humans, we instinctively recognize whether or not something has been designed by another human or an animal. Children and uncivilized men even do this, and are usually correct in their perceptions. You may argue that life has only the appearance of design - but what if the appearance actually matches the reality?

The idea that we have been created by an intelligence has been held to by some of the wisest men who have ever lived, and some of the greatest moral teachings - teachings that have transformed murderous barbarians into civilized men who actually love their neighbors -have been associated with this idea. I don't think that we should be too quick to throw it away. Scientific ideas that seemed impregnible in their time have become obsolete with the passage of time. What if you are wrong, and life does owe it's origin to an Intelligence?

If I am wrong, if there is no Creator who will hold me accountable, then when I die, that will be it. But if you are wrong, if you really are accountable to a Great Intelligence who has created life - then there is much at stake. Pascal's Wager should at least cause you to tread cautiously.

Drawing an example from Theology, I think of the advice that Gamaliel gave to the Jewish Sanehedrin when they were considering killing the apostles and stamping out the fledgling Christian movement:

Originally posted by Acts 5:34

Then there stood up one in the sanhedrin, a Pharisee named Gamaliel, a doctor of the law honored among all the people. And he commanded the apostles to be put outside a little space. 35 And he said to them, Men of Israel, take heed to yourselves what you intend to do regarding these men. 36 For before these days Theudas rose up, boasting himself to be somebody; a number of men, about four hundred, joined themselves to him; who was slain. And all, as many as obeyed him, were scattered and brought to nothing. 37 After this one, Judas of Galilee rose up in the days of the Registration, and drew away considerable people after him. Yet that one perished; and all, as many as obeyed him, were scattered. 38 And now I say to you, withdraw from these men and let them alone. For if this counsel or this work is of men, it will come to nothing. 39 But if it is of God, you cannot overthrow it, lest perhaps you be found even to fight against God.

The Sanehedrin let them live. I likewise urge you to tolerate this view and those who hold to it, if for no other motive than a humble recognition that there is a chance - however small you think it may be - that you may be wrong. Hold us to high standards. Correct us when we are in error. But do, please, not too lightly regard this idea or those who hold to it.

Originally posted by Athanasius

Many of you have expressed a great deal of intolerance towards the theory of ID.

I hope you understand that some of this intolerance has built up over many years of hearing the same old tired arguments. Much of the intolerance you find is due to simple exasperation.

Some of you no doubt believe that this is justified, but it has been around for a long time. As humans, we instinctively recognize whether or not something has been designed by another human or an animal. Children and uncivilized men even do this, and are usually correct in their perceptions.

I readily distinguish between vegetable and mineral, also. I see human-designs every day and can pick them out in a second when placed against a natural background. But I do not discern designed from natural in the case of life. Why is everything that happens in life natural except its origins? Why is there so much evidence that supports a concept anathema to you and your designer? Not a very well designed universe, I'd say.

You may argue that life has only the appearance of design - but what if the appearance actually matches the reality?

I will confess that I have been wrong hundreds of times when interpreting natural phenomena. I am not afraid to be wrong. My real fear is that I may not have the nerve to attempt solving a problem someday. THat I will give up.

... What if you are wrong, and life does owe it's origin to an Intelligence?

I would have no problem with this. If that intelligence were unwilling to see that I used my mind to the best of my ability, then that intelligence would not have my respect. In the meantime, I await evidence to support ID.

Originally posted by Athanasius

Sylas, thanks for the thoughtful review, and for bringing us up to date on some of the research de Beer refers to. I am sure that it must have taken some time to write. That, unfortunately, is a commodity that is in short supply for me right now. My job is very demanding, and my wife has pointed out to me that I have neglected precious time I should have devoted to family participating in this thread.

No problem, and you are welcome. It took me all day Sunday to research and write. You are right to put your family first, and it is my normal expectation that a careful analysis at this level of detail cannot really be taken up easily in this informal context.

If you merely accepted instantly the information in my review, then the most important lesson has been lost! If you merely reject instantly with a gut reaction that fails to engage the substance; then nothing has been learned either. If you found it thought provoking, I am content; and I do not demand either response or explicit concessions.

I do want to respond to a couple of challenges you offer.

If I am wrong, if there is no Creator who will hold me accountable, then when I die, that will be it. But if you are wrong, if you really are accountable to a Great Intelligence who has created life - then there is much at stake. Pascal's Wager should at least cause you to tread cautiously.

Pascal's wager is arguably the low point of Western philosophy.

Have you considered that you have the wager backwards? Perhaps there is a Creator, who designed the whole universe to be fertile and good, and who (expressed in poetic language) called the Earth and the waters to bring forth life – which they have done.

Perhaps this Creator is the great Lord and Father of the Christian religion. And perhaps the refusal to recognize the created good capacity of the natural world to obey that command to bring forth life is culpable. Perhaps the persistent wilful blindness of intelligent design advocates is, by its distortions, bringing God into disrepute, and driving people away from the faith of salvation.

Perhaps intelligent design theorists will, at judgement, be called to account for gross failure to honestly consider the plain evidence of God’s power and Lordship over all natural processes, which work together to His purpose. Perhaps they will be held responsible to making Christian faith a laughingstock, and punished for the many souls lost as a result.

I don’t actually think so; but you might consider the possibility....

The solution, to my mind, is may be found in the words of Jesus, standing in stark contrast to the fear-mongering of Pascal’s wager -- “Do not be afraid.”

Do not be afraid to look for answers. Do not let your fears constrain the integrity of how you draw honest conclusions.

...I likewise urge you to tolerate this view and those who hold to it, if for no other motive than a humble recognition that there is a chance - however small you think it may be - that you may be wrong. Hold us to high standards. Correct us when we are in error. But do, please, not too lightly regard this idea or those who hold to it.

Shrug. I am well used to being wrong. In my world view, it is something to be expected, and not something to fear.

There are two particular aspects of my review which stand out as deficient. First, I don’t know enough about Morgan’s experiments on eyeless, or about how the cascades for eye formation become triggered in populations homozygous for the defective allele of the eyeless gene. I can speculate... there are homologues for this gene even in fruit flies, like “toy” (Twin-Of-eYless). Could that be involved somehow? I don’t know. I would love to get some better references on this experiment. Has it been repeated? This is important, as it seems to be the most important justification for de Beer’s most startling conclusion.

Second, I don’t know enough about the germinal model for embryonic development, or what exactly de Beer disproved.

Finally, I am quite positive that there are errors of fact in my review that I have not picked up. I know I make mistakes. I am very keen to fix them. Mere caution that “I might be wrong” tells me nothing. The question is... WHAT bits are wrong, and why? How can I fix them? How can I find out, in an attempt to be as accurate as possible?

I have decided to take this review further, and have passed on some of these questions to talk.origins, where there is a very active set of biologists who may help illuminate some of these questions. I am proposing that my review could become the basis of a new FAQ. We’ll see.

I always recognize that I don’t have any special access to revealed truth. But it is pretty clear who are best at actually picking up my errors for me. They are people who don’t claim to be infallible, but are sufficiently knowledgeable to point out what errors they see in concrete terms. I like that. I have sought their input.

Cheers -- Sylas

Originally posted by sylas

I have decided to take this review further, and have passed on some of these questions to talk.origins, where there is a very active set of biologists who may help illuminate some of these questions. I am proposing that my review could become the basis of a new FAQ. We’ll see.

Why not also submit it to trueorigins.org. and get some imput from the other side, too? Couldn't hurt.

I really need to get my hands on de Beer's article before I can critique your review. It has been over 20 years since I read portions of it, and my memory is not good enough for that. Unfortunately, the Cobb County library does not make it possible to order such publications online. When I get the chance I will pay them a visit or stop by the local university library.

Originally posted by Athanasius

Why not also submit it to trueorigins.org. and get some imput from the other side, too? Couldn't hurt.

Apples and origins...

Trueorigins is a website, not a discussion forum.

talk.origins is a Usenet discussion forum, in which all kinds of contributors are active. Anything sent to talk.origins is already open slather for anyone from any perspective. Some people miss this distinction. I have not sent anything to the talkorigins web site as yet, and if I do, it will be for on-line publishing, not for review.

For review, I have solicited input at an open Usenet discussion group, in which there are hundreds of active participants from many diverse perspectives. I'll welcome and consider any input from *any* perspective. (I gave a link to the TheologyWeb post, along with some specific questions. I hope linking in this way is okay.) The Usenet post is here.

It there are any useful contributions from anyone here at TheologyWeb, then they will also be very welcome indeed.

If you (or anyone) gets hold of de Beer's monograph, and can give specific feedback on the review as a review, that would be great. However, I am also soliticing information on the specific experiments referenced, and de Beer's paper is not required for those matters.

Two things I have singled out where I know my comments are weak on substance. Germ layer theory and variable expressivity of eyeless in Drosophila.

Thanks -- Sylas

Sylas, since I am a biochemist and have some experience with developmmental biology, let's see if I can add anything to your review.

Originally posted by sylas

This is a synopsis and review of

Homology, An Unsolved Problem
by Sir Gavin de Beer
Oxford University Press (1971) ISBN 0 19 914111 8

2. Homology in plants: leaves and flowers. This is a straightforward description of the structures of leaves and flowers, showing structural evidence for homology that was first identified by Goethe in about 1790. No problems here; and subsequent work in genetics has abundantly confirmed this classic example of unambiguous homology.

3. Homology in animals: the ear ossicles. This is another classic example, without any conceptual difficulties; the homology of the bones of the inner ear in mammals with bones that make up part of the jaw in reptiles. He concludes:

What makes this study even more significant is that the results of comparative anatomy are confirmed by those of palaeontology, for there are fossil reptiles that show advances towards the mammalian condition, and the superseding of the quadrate-articular hinge of the lower jaw by the squamosal-dentary articulation. All this evolution took place without any functional discontinuity. It is a sobering thought that ever man carries in his ear ossicles the homologue of the lower jaw hinge of his reptilian ancestors. This is one of the most demonstrative examples of how comparative anatomy can determine homology of structures inherited from common ancestors in evolution.

So, what you are saying is that DeBeer has no problems with SOME examples of homology. He can ascribe them to evolution. Right?

4. Conservative effects of homology.

Diverse vertebrate species have “branchial arches” in the embryo (sometimes called “gill slits”). In different species, these arches develop into different structures. The “recurrent laryngeal nerve” passes behind branchial arch number six in the vertebrate embryo. During embryonic development, several of the arches disappear, and others develop into various structures, in diverse manners for different species.

The result is that in mammals, the recurrent laryngeal nerve ends up with what would appear to be a very poorly designed route, “running backwards and looping round the ductus arteriosus, then runs forward again to innervate the muscles of the larynx.” In humans, the nerve is thus several inches longer than we should expect in a well designed organism; and in a giraffe it is several feet longer. De Beer shows how homology explains this curious anomaly.

The explanation is the homology between the mammalian ductus arteriosus and the 6th arterial arch of the fish, which is respected in descendent forms, resulting in apparently anomalous conditions.

Notice the part I bolded. You should emphasize this more. It was observations like this, Athanasius, that killed ID in the period 1830-1865. If we humans know a better design, then surely a much more intelligent ID can also figure it out. The theological problem for ID is that the ID isn't very smart much of the time. Evolution really got God off a very bad hook that ID put Him on.

8. Homology and functional change. Homologous organs can have different functions in different organisms.

Also called "exaptation".

(10.2. The master organizers for inducing development of homologous structures.) De Beer speaks of the diffusion of “substances from a master structure called an organizer” that will induce cells to undergo differentiations and form certain structures. ...
It seems that de Beer was simply wrong. The evidence for diverse organizers is weak, and his conclusion poorly supported in the light of more careful experiments.

DeBeer was thinking of an embryonic structure for his "organizer", not Homeobox genes. The key is that the same genes are turned on, but not necessarily in the the cells that actually differentiate to the tissue. See below:

The second example, of brain and spinal cord, has also been subject to considerable study since de Beer’s book. De Beer notes that “in true vertebrates, the spinal cord and brain develop as a result of induction by the underlying organizer; but in the ‘tadpole larva’ of the tunicates, which has a ‘spinal cord’ like the vertebrates, it differentiates without any underlying organizer at all.”

Sylas, by "organizer", De Beer is referring to embryonic tissue, not genes. In vertebrates, this tissue is the anterior ectodermis.

It would be rash to simply presume that “organizers”, or factors inducing formation of homologous structures, definitely cannot evolve or become co-opted from other parts of the genome or embryo.

Be careful here. As I said, when DeBeer is talking about "organizers", he may be referring to factors, but is more likely referring to embryonic tissues. It was after De Beer's paper that it was found that the embryonic tissues secreted SOLUBLE factors that acted on the cells in the adjacent embryonic tissue.

(10.3. Homology and germ layers.) Here de Beer refutes a claim that homologous organs must always arise from the same germ layer in the embryo. He cites his own work, The Differentiation of Neural Crest Cells into Visceral Cartilages and Odontoblasts in Amblystoma, and a Re-Examination of the Germ-Layer Theory, in Proc. of the Royal Soc. of London. Series B, Biol Sci, Vol 134, Iss 876, pp. 377-398 (1947).

I’m uncertain on this. I’d welcome comment from a biologist in the house. As I understand it, de Beer is recognized as having disproved a germ layer theory developed around the turn of the century; but the concept of gastrulation and germinal layers remains vital in embryology. This includes the notion that various structures are developed from various layers. So what did de Beer disprove here? I’m not sure.

In the original germ layer theory, you have three major germ layers: ectoderm, endoderm, and mesoderm. Ectoderm gives rise to the skin, oral mucosa, PNS (peripheral nervous system) and CNS (central nervous system). Endoderm gives rise to the functional cells of the liver, pancreas, intestines, etc. Mesoderm gives everything else, including cartilage and teeth. However, the head is derived from the neural crest and is, therefore, composed of "ectoderm" cells. It should give rise only to CNS cells. But these same neural crest cells also give rise to cartilage, the facial bones, and the odontoblasts (tooth forming cells). Thus, you have cells not of mesodermal origin giving rise to tissues that are supposedly mesodermal.

There is an elegant study of founder cells in C. elegans. I'll have to find the reference for you. It shows that the traditional germ layer idea is not completely accurate. IN GENERAL, ectodermal cells give rise to nerve cells in PNS and CNS. Yet one founder cell that gives skeletal muscle also gives 2 of the peripheral nerves. Apparently endoderm does give rise only to endodermal cells, but there is mixup between ectoderm and mesoderm.

De Beer also claims that tooth enamel can arise from ectodermal or endodermal cells. In this, I suspect he is wrong. Enamel arises from ectodermal cells, not endodermal.

He said "endodermal" and not "mesodermal". As you say, De Beer is quite wrong here. But identifying cells on the morphological criteria he used is dangerous. Cells change morphology on differentiation, nutrition, and attachment. He needed to follow those cells in time during development a bit better.

(11.1. Identical genes for non-homologous structures.) There are two major cases considered. The first is a gene in fowls, which controls formation of a crest of feathers, and also controls a cerebral hernia (knob in the skull) to accommodate it. The two structures are not homologous, yet controlled by the same gene. The second is the gene ‘antenna’, which induces Drosophila to produce an extra antenna instead of an eye. The conclusion:

characters controlled by identical genes are not necessarily homologous.

He's never heard of pleiotropy? Two functions of the same genes?

There is no particular problem here. The genes in question are control genes, that are able to trigger expression of other genes, that are not identical. The genes actually coding for non-homologous structures are different, but triggered by the same factor. This is fascinating, but not remotely in conflict with common descent.

Also remember that structures are also formed by the interaction of concentration gradients of two or more genes. So yes, knock out one of those genes and you don't get the structure, but that does not tell you that there are other genes also involved.

(11.2. Different genes for homologous structures.) This claim I find surprising, but I don’t think the case is made very well. The example is the “eyeless” gene in Drosophila, which de Beer speaks of as a gene that “deprives its possessor of eyes”; though it is recessive.

This is not quite true. Eyeless is actually a gene that enables or triggers the formation of eyes. It was named eyeless because a mutant allele prevented it from working. A fly homozygous for the defective allele does not form eyes.

Your explanation is correct. Can you please quote De Beer more from this section? I'd like to see his argument in more detail.

De Beer concludes:

What has happened during inbreeding is that all the other pairs of alleles making up the gene complex have been reshuffled until a gene complex has been produced that prevents the phenotypic manifestation of the ‘eyeless’ allele. Other genes must therefore deputize for the absent normal gene that controls the formation of eyes. But why should they, and by what mechanism? Nobody can deny that the restored eyes that develop in genetically ‘eyeless’ stocks are homologous to the original normal eyes.

Why? Because it turns out that there are several Hox genes involved in eye development. Eyeless (or Pax 6) is near the head of the list. However, if mutant transciption factors appear in the population that cause the expression of the next protein in the cascade, then of course eye development will proceed as almost normal. Also, look at the abstract of the first paper in the list. It turns out there is a twin Pax-6 gene in Drosophila. So, even if Pax -6 is mutated, this twin gene, if expression is increased, can compensate. That also explains the results De Beer is talking about.

Mol Cell 1999 Mar;3(3):297-307

twin of eyeless, a second Pax-6 gene of Drosophila, acts upstream of eyeless in the control of eye development.

Czerny T, Halder G, Kloter U, Souabni A, Gehring WJ, Busslinger M
Research Institute of Molecular Pathology, Vienna, Austria.

The Drosophila Pax-6 gene eyeless (ey) plays a key role in eye development. Here we show that Drosophila contains a second Pax-6 gene, twin of eyeless (toy), due to a duplication during insect evolution. [emphasis mine] Toy is more similar to vertebrate Pax-6 proteins than Ey with regard to overall sequence conservation, DNA-binding function, and early expression in the embryo, toy and ey share a similar expression pattern in the developing visual system, and targeted expression of Toy, like Ey, induces the formation of ectopic eyes. Genetic and biochemical evidence indicates, however, that Toy functions upstream of ey by directly regulating the eye-specific enhancer of ey. Toy is therefore required for initiation of ey expression in the embryo and acts through Ey to activate the eye developmental program.
PMID: 10198632, UI: 99214845

Dev Genes Evol 1998 Aug;208(6):352-6 Related Articles, Books, Nucleotide, LinkOut


Pax-6 origins--implications from the structure of two coral pax genes.

Catmull J, Hayward DC, McIntyre NE, Reece-Hoyes JS, Mastro R, Callaerts P, Ball EE, Miller DJ.

Department of Biochemistry and Molecular Biology, James Cook University, Townsville, Queensland 4811, Australia.

Vertebrate Pax-6 and its Drosophila homolog eyeless play central roles in eye specification, although it is not clear if this represents the ancestral role of this gene class. As the most "primitive" animals with true nervous systems, the Cnidaria may be informative in terms of the evolution of the Pax gene family. For this reason we surveyed the Pax gene complement of a representative of the basal cnidarian class (the Anthozoa), the coral Acropora millepora. cDNAs encoding two coral Pax proteins were isolated. Pax-Aam encoded a protein containing only a paired domain, whereas Pax-Cam also contained a homeodomain clearly related to those in the Pax-6 family. The paired domains in both proteins most resembled the vertebrate Pax-2/5/8 class, but shared several distinctive substitutions. As in most Pax-6 homologs and orthologs, an intron was present in the Pax-Cam locus at a position corresponding to residues 46/47 in the homeodomain. We propose a model for evolution of the Pax family, in which the ancestor of all of the vertebrate Pax genes most resembled Pax-6, and arose via fusion of a Pax-Aam-like gene (encoding only a paired domain) with an anteriorly-expressed homeobox gene resembling the paired-like class.


Genes Dev 1997 May 1;11(9):1122-31

Direct regulation of rhodopsin 1 by Pax-6/eyeless in Drosophila: evidence for a conserved function in photoreceptors.

Sheng G, Thouvenot E, Schmucker D, Wilson DS, Desplan C
Laboratory of Molecular Genetics, Howard Hughes Medical Institute, The Rockefeller University, New York, New York 10021, USA.

Pax-6 is a transcription factor containing both a homeodomain (HD) and a Paired domain (PD). It functions as an essential regulator of eye development in both Drosophila and vertebrates, suggesting an evolutionarily conserved origin for different types of metazoan eyes. Classical morphological and phylogenetic studies, however, have concluded that metazoan eyes have evolved many times independently. These apparently contradictory findings may be reconciled if the evolutionarily ancient role of Pax-6 was to regulate structural genes (e.g., rhodopsin) in primitive photoreceptors, and only later did it expand its function to regulate the morphogenesis of divergent and complex eye structures. In support of this, we present evidence that eyeless (ey), which encodes the Drosophila homolog of Pax-6, directly regulates rhodopsin 1 (rh1) expression in the photoreceptor cells. We detect ey expression in both larval and adult terminally differentiated photoreceptor cells. We show that the HD of Ey binds to a palindromic HD binding site P3/RCS1 in the rh1 promoter, which is essential for rh1 expression. We further demonstrate that, in vivo, P3/RCS1 can be replaced by binding sites specific for the PD of Ey. P3/RCS1 is conserved in the promoters of all Drosophila rhodopsin genes as well as in many opsin genes in vertebrates. Mutimerized P3 sites in front of a basal promoter are able to drive the expression of a reporter gene in all photoreceptors. These results suggest that Pax-6/Ey directly regulates rhodopsin 1 gene expression by binding to the conserved P3/RCS1 element in the promoter.
PMID: 9159393, UI: 97302962

Int J Dev Biol 1997 Dec;41(6):835-42

Pax-6, eyes absent, and Prox 1 in eye development.

Tomarev SI
Laboratory of Molecular and Developmental Biology, National Eye Institute, NIH, Bethesda, MD 20892-2730, USA. tomarev@fcrfv1.ncifcrf.gov

Eyes in different systematic groups including arthropods, molluscs and vertebrates probably have a common evolutionary origin. As a consequence of this, related genes are used for regulation of the early steps of eye development in different organisms. In this review, I briefly summarize data on three gene families which might be essential for eye development across species: Pax-6/eyeless, Eya/eyes absent and Prox/prospero with emphasis on our contribution here. Mechanisms of eye formation and the generation of different types of eyes in the course of evolution are discussed.

Mol Cell Biol 1999 Feb;19(2):1159-70

The trithorax group gene moira encodes a brahma-associated putative chromatin-remodeling factor in Drosophila melanogaster.

Crosby MA, Miller C, Alon T, Watson KL, Verrijzer CP, Goldman-Levi R, Zak NB
Department of Molecular and Cellular Biology, Harvard University, Cambridge, Massachusetts 02138, USA.

The genes of the trithorax group (trxG) in Drosophila melanogaster are required to maintain the pattern of homeotic gene expression that is established early in embryogenesis by the transient expression of the segmentation genes. The precise role of each of the diverse trxG members and the functional relationships among them are not well understood. Here, we report on the isolation of the trxG gene moira (mor) and its molecular characterization. mor encodes a fruit fly homolog of the human and yeast chromatin-remodeling factors BAF170, BAF155, and SWI3. mor is widely expressed throughout development, and its 170-kDa protein product is present in many embryonic tissues. In vitro, MOR can bind to itself and it interacts with Brahma (BRM), an SWI2-SNF2 homolog, with which it is associated in embryonic nuclear extracts. The leucine zipper motif of MOR is likely to participate in self-oligomerization; the equally conserved SANT domain, for which no function is known, may be required for optimal binding to BRM. MOR thus joins BRM and Snf5-related 1 (SNR1), two known Drosophila SWI-SNF subunits that act as positive regulators of the homeotic genes. These observations provide a molecular explanation for the phenotypic and genetic relationships among several of the trxG genes by suggesting that they encode evolutionarily conserved components of a chromatin-remodeling complex. [emphases mine]
PMID: 9891050, UI: 99108089

Catmull J, Hayward DC, McIntyre NE, Reece-Hoyes JS, Mastro R, Callaerts P, Ball EE, Miller DJ.

Department of Biochemistry and Molecular Biology, James Cook University, Townsville, Queensland 4811, Australia.

Vertebrate Pax-6 and its Drosophila homolog eyeless play central roles in eye specification, although it is not clear if this represents the ancestral role of this gene class. As the most "primitive" animals with true nervous systems, the Cnidaria may be informative in terms of the evolution of the Pax gene family. For this reason we surveyed the Pax gene complement of a representative of the basal cnidarian class (the Anthozoa), the coral Acropora millepora. cDNAs encoding two coral Pax proteins were isolated. Pax-Aam encoded a protein containing only a paired domain, whereas Pax-Cam also contained a homeodomain clearly related to those in the Pax-6 family. The paired domains in both proteins most resembled the vertebrate Pax-2/5/8 class, but shared several distinctive substitutions. As in most Pax-6 homologs and orthologs, an intron was present in the Pax-Cam locus at a position corresponding to residues 46/47 in the homeodomain. We propose a model for evolution of the Pax family, in which the ancestor of all of the vertebrate Pax genes most resembled Pax-6, and arose via fusion of a Pax-Aam-like gene (encoding only a paired domain) with an anteriorly-expressed

Originally posted by lucaspa

Sylas, since I am a biochemist and have some experience with developmmental biology, let's see if I can add anything to your review.

Thanks very much indeed! I am going to ask if we can move my review, and your followup here, to start a new thread. I will PM you and a moderator to see if this can be arranged somehow. I will refrain from followup until we get it fixed up. Others who like to join in make wish to take note...

Thanks again -- Sylas

Originally posted by lucaspa

So, what you are saying is that DeBeer has no problems with SOME examples of homology. He can ascribe them to evolution. Right?

Yes, of course. Sir Gavin de Beer was eminent zoologist, embryologist, evolutionist. The examples in the first two sections are pretty clear, and there is no credible doubt about shared ancestry in those cases.

The monograph seems intended to discuss difficulties with explaining homology, given a putative disconnect with genetics.

We should not be too hard on de Beer for failing to anticipate the findings of biology in the thirty five years since the book was written, particularly since at that time gene sequencing had not yet been developed. He made valuable contributions to development of modern evo-devo. At the time he was writing, genes could be identified and located to chromosomes. The genetic code had only just been cracked, and it was known that genes could express proteins. But genetic homology revealed by sequencing was not available.

Originally posted by lucaspa

Notice the part I bolded. You should emphasize this more. It was observations like this, Athanasius, that killed ID in the period 1830-1865. If we humans know a better design, then surely a much more intelligent ID can also figure it out. The theological problem for ID is that the ID isn't very smart much of the time. Evolution really got God off a very bad hook that ID put Him on.

Yes, quite definitely. Homologous structures often show constraints imposed on body forms that indicate the process was not comparable to the constructed artefacts made by intelligent designers. Section 4 is an excellent case in point.

Originally posted by lucaspa

DeBeer was thinking of an embryonic structure for his "organizer", not Homeobox genes. The key is that the same genes are turned on, but not necessarily in the cells that actually differentiate to the tissue. See below:

The concept of an “organizer” is due to Spemann, who won a well deserved Nobel prize in 1935 for this development. The organizer is tissue or embryonic cells that produce a substance which then controls the development of other cells. This concept remains of critical importance in embryology, and is not superseded by the discovery of control genes and homeobox genes. They are two different aspects in how development proceeds and is guided. (I know you know this; I’m speaking out loud stuff that may become part of a longer review article. So tell me if my account is accurate!)

Yes, of course de Beer did not know about homeobox genes. With the understanding gained in how these genes work, the most serious putative difficulties in de Beer’s monograph are resolved.

Originally posted by lucaspa

Originally posted by Sylas

The second example, of brain and spinal cord, has also been subject to considerable study since de Beer’s book. De Beer notes that “in true vertebrates, the spinal cord and brain develop as a result of induction by the underlying organizer; but in the ‘tadpole larva’ of the tunicates, which has a ‘spinal cord’ like the vertebrates, it differentiates without any underlying organizer at all.”

Sylas, by "organizer", De Beer is referring to embryonic tissue, not genes. In vertebrates, this tissue is the anterior ectodermis.

Sure. De Beer’s point is that the organizer is required in vertebrates, but not in tunicates, which are supposedly homologous. He sees this as an unexplained problem with the notion of homology of neural structures in tunicates and vertebrates. The bit in italics above is a direct quote from de Beer’s book.

The point is discussed in some detail on the link I provided in the original review, to Dr. John Kimball on-line biology text, and the section called Organizing the Embryo: The Central Nervous System. It explains that the operation of the Spemann organizer has only just been resolved quite recently. Kimball explains the operation of the organizer as follows.

• Cells on the ventral side of the blastula secrete a variety of proteins such as bone morphogenetic protein-4 (BMP-4)
• These induce the ectoderm above to become epidermis.
• If their action is blocked, the ectodermal cells are allowed to follow their default pathway, which is to become nerve tissue of the brain and spinal cord.
• The Spemann organizer blocks the action of BMP-4 by secreting molecules of the proteins
  • chordin and
  • noggin
• Both of these physically bind to BMP-4 molecules in the extracellular space and thus prevent BMP-4 from binding to receptors on the surface of the overlying ectoderm cells.
• This allows the ectodermal cells to follow their intrinsic path to forming neural folds and, eventually, the brain and spinal cord.

Hence that tunicates develop neural tissue without the organizer is no real surprise. De Beer thought the organizer was required to induce formation of neural tissue. In fact, the organizer blocks formation of epidermis.

On to the section I listed as 11.1, and the conclusion that “characters controlled by identical genes are not necessarily homologous”. Lucaspa asks:

Originally posted by Lucaspa

He's never heard of pleiotropy? Two functions of the same genes?

I think that is what he is describing. Pleiotropy was a term coined in 1910. I’m sure de Beer had heard of it; and here is pointed out an example, as part of an argument that there is an unsolved problem connecting homologous structure to related genes. His argument is in two parts. First, that identical genes may be involved in non-homologous structures; and second, that totally different genes may be involved in homologous structures. De Beer concluded that there was a problem understanding homology in the light of genetics.

It is the second half that of his argument that I claim is now out of date and inconsistent with better information not available in 1971.

On to the section I listed as 11.2, on “eyeless”…

Originally posted by lucaspa

Your explanation is correct. Can you please quote De Beer more from this section? I'd like to see his argument in more detail.

In Drosophila there is a gene, ‘eyeless’, which deprives is possessor of eyes. It is a recessive character, which is important because it means that when its effect is produced, the fly has inherited the ‘eyeless’ allele from both parents, and no normal eye-controlling allele is present. If a stock of individuals pure (homozygous) for the ‘eyeless’ gene is inbred for many generations, there is a high mortality as would be expected for the adverse effects of natural selection action on a gene with such lethal effects. But eventually, flies appear in the offspring possessing normal eyes. It can easily be shown that the ‘eyeless’ gene has not changed, because when one of these phenotypically eye-possessing byt genotypically homozygous ‘eyeless’ flies is mated with the original wild stock, i.e., the ‘eyeless’ gene is put back into the original gene complex, the virulent effects of the ‘eyeless’ gene reappear. What has happened during the inbreeding is that all the other pairs of alleles making up the gene complex have been reshuffled until a gene complex has been produced that prevents the phenotypic manifestation of the ‘eyeless’ allele. Other genes must therefore deputize for the absent normal gene that controls the formation of the eyes. But why should they, and by what mechanism? Nobody can deny that the restored eyes that develop in genetically ‘eyeless’ stocks are homologous to the original normal eyes. Therefore, homologous structures need not be controlled by identical genes, and homology of phenotypes does not imply similarity of genotypes.

—Gavin de Beer, “Homology, An Unsolved Problem”, page 15

It is an intriguing argument, a bit confused with the terms allele and gene being mixed up somewhat by comparison with modern usage. Sir Gavin was handicapped by not having access to genomes and gene sequences. We now know that flies have several homologues for the ‘eyeless’ gene; in particular ‘twin of eyeless’. Therefore the co-opted genes are strongly homologous to the eyeless gene, and so de Beer’s conclusion does not really follow.

Thanks for all your references on this bit.

One thing I did not include in the original review -- quite simply because I hit up hard against the post size limit! – was de Beer’s final example. This is T.M. Sonneborn’s work on Paramecium. The citation de Beer gives is

Sonneborn, T.M. (1970) Gene action in development. Proc. R. Soc. B 176, 347.

Very briefly, Sonneborn showed that there are three distinct types of inheritance in the singled celled protist Paramecium. They are Mendelian inheritance via nuclear genes, cytoplasmic inheritance via mitochondrial genes, and [b]cortical inheritance[/i]. Quoting a modern on-line text by Scott F. Gilbert (DevBio: A Companion to Devlopmental Biology), and the section Inheritance in Ciliates,

This cortical inheritance has not yet been observed in non-ciliate cells, and it remains one of the great exceptions to the rule that genes provide the only vehicles for inheritance from one generation to another.

De Beer concludes his book with the following comment.

To the question ‘Is the whole of development encoded in DNA (that is to say, in the genes)?’ the answer in Paramecium is ‘No’. Whether this is applicable to ‘patterns’ in higher organisms, and whether homologous structures are controlled by non-DNA mechanisms awaits further research.

Further research is long since in; and the answer is mostly that inheritance in higher animals is controlled by nuclear genomes. Basically, the “Problem of Homology” that de Beer focuses upon is a disconnect between genotype and phenotype. But de Beer did not have the advantage of insights from modern genetics. He was necessarily working with weak inferences from circumstantial indications. Homology continues to be a difficult problem, to identify and describe. But the particular arguments of de Beer were wrong.

Cheers -- Sylas

Originally posted by sylas

Yes, of course. Sir Gavin de Beer was eminent zoologist, embryologist, evolutionist. The examples in the first two sections are pretty clear, and there is no credible doubt about shared ancestry in those cases.

The monograph seems intended to discuss difficulties with explaining homology, given a putative disconnect with genetics.

So, once again, we have professional creationists taking things out of context to mean something the author did not mean, and then passing these off on unsuspecting creationists like Athanasius as tho they are true. And Athanasius gets conned by people he thinks he can trust.

We should not be too hard on de Beer for failing to anticipate the findings of biology in the thirty five years since the book was written, particularly since at that time gene sequencing had not yet been developed.

It's not just gene sequencing. Brian Hall didn't do his experiments to show soluble inductive factors in developmental biology until the early 1980s. Hall thought that tissues had to be in contact.

Yes, quite definitely. Homologous structures often show constraints imposed on body forms that indicate the process was not comparable to the constructed artefacts made by intelligent designers. Section 4 is an excellent case in point.

Another theological case against the IDer is the Panda's thumb. Everyone knows that it is a modified wrist bone, and IDers sometimes just look at the "thumb" and say how wonderfully complex it is. But they forget to tell people that the Panda has a real thumb! That's right, the Panda has 4 fingers and 1 thumb already. However, in bears the thumb and fingers are fused to form a single boney unit in the paw. Now, an IDer making Pandas from scratch is not constrained to have the thumb remain fused. Knowing the Panda is going to eat bamboo -- because the ID designed the Panda to eat bamboo -- the ID only has to free up the thumb from the other digit and Voila!, real thumb. What? God is suffering from advanced Alzheimer's so badly that He didn't remember that the Panda has a thumb?

With descent with modification, however, the problem goes away. The ancestor of the Panda -- a bear - already had the digits fused and thus there was no way to unfuse them. Add a new thumb instead. Since natural selection can only add information, not subtract it.

The concept of an “organizer” is due to Spemann, who won a well deserved Nobel prize in 1935 for this development. The organizer is tissue or embryonic cells that produce a substance which then controls the development of other cells. This concept remains of critical importance in embryology, and is not superseded by the discovery of control genes and homeobox genes. ... (I know you know this; I’m speaking out loud stuff that may become part of a longer review article. So tell me if my account is accurate!)

The point, Sylas, was that the "organizer" was thougth to be the tissue or cells themselves. That is, it was direct contact with the tissue or cells that did the "organizing". Remember the specifics of the theory. So, in tunicates DeBeer could not find cells or tissues that could serve as the organizer. As I said, it wasn't until BK Hall and others in the 1980s that data was found showing that "organizers" were, in fact, proteins secreted from cells that then acted either 1) on the same cells (autocrine) and/or 2) on adjacent cells (paracrine). So, while the "organizer" concept has survived, as you noted, it is now much modified from the original concept of Spemann and DeBeer.

De Beer’s point is that the organizer is required in vertebrates, but not in tunicates, which are supposedly homologous. He sees this as an unexplained problem with the notion of homology of neural structures in tunicates and vertebrates. The bit in italics above is a direct quote from de Beer’s book.

For tunicates, we have autocrine induction as far as I can tell. Another factor that is produced by the cells themselves is responsible for neural induction. We'll have to look at tunicate anatomy to see if they have normal epidermis in the region. But in vertebrates an additional level was added in the blastula to get epidermis. Thus, to get the nerve tissue to form, secretion of noggin or chordin is necessary to block the BMP-4.

BTW, notice that again we have an increase in information. The epidermal induction is not turned off, which would be the simplest thing to do. Instead, we get an additional level of control -- noggin and chordin -- to block epidermis and get neural tissue.

I think that is what he is describing. Pleiotropy was a term coined in 1910. I’m sure de Beer had heard of it; and here is pointed out an example, as part of an argument that there is an unsolved problem connecting homologous structure to related genes. His argument is in two parts. First, that identical genes may be involved in non-homologous structures; and second, that totally different genes may be involved in homologous structures. De Beer concluded that there was a problem understanding homology in the light of genetics.

Then I don't follow his logic. Since most structures are polygenic, and many genes are pleiotropic, I would expect the situation De Beer describes.

On to the section I listed as 11.2, on “eyeless”…

In Drosophila there is a gene, ‘eyeless’, which deprives is possessor of eyes. It is a recessive character, which is important because it means that when its effect is produced, the fly has inherited the ‘eyeless’ allele from both parents, and no normal eye-controlling allele is present. If a stock of individuals pure (homozygous) for the ‘eyeless’ gene is inbred for many generations, there is a high mortality as would be expected for the adverse effects of natural selection action on a gene with such lethal effects. But eventually, flies appear in the offspring possessing normal eyes. It can easily be shown that the ‘eyeless’ gene has not changed, because when one of these phenotypically eye-possessing byt genotypically homozygous ‘eyeless’ flies is mated with the original wild stock, i.e., the ‘eyeless’ gene is put back into the original gene complex, the virulent effects of the ‘eyeless’ gene reappear. What has happened during the inbreeding is that all the other pairs of alleles making up the gene complex have been reshuffled until a gene complex has been produced that prevents the phenotypic manifestation of the ‘eyeless’ allele. Other genes must therefore deputize for the absent normal gene that controls the formation of the eyes. But why should they, and by what mechanism? Nobody can deny that the restored eyes that develop in genetically ‘eyeless’ stocks are homologous to the original normal eyes. Therefore, homologous structures need not be controlled by identical genes, and homology of phenotypes does not imply similarity of genotypes.

—Gavin de Beer, “Homology, An Unsolved Problem”, page 15

It is an intriguing argument, a bit confused with the terms allele and gene being mixed up somewhat by comparison with modern usage. Sir Gavin was handicapped by not having access to genomes and gene sequences. We now know that flies have several homologues for the ‘eyeless’ gene; in particular ‘twin of eyeless’. Therefore the co-opted genes are strongly homologous to the eyeless gene, and so de Beer’s conclusion does not really follow.

Yeah, De Beer simply did not have all the data. And, even tho he talks about "gene cassette", he seems to think that the genes are completely independent (and expressed at the same time) rather than in a timed cascade. Thus, Pax-6 is really a transcription factor that turns on the next gene in the cascade. If that gene can be turned on by another means -- modification of the DNA sequence for transcription binding such that another protein can turn on that gene -- then the rest of the sequence happens. Or modification of the twin of Pax-6 such that it can now bind to the promotor region of the next gene in the cascade and start the cascade instead of Pax-6. "Why should they?" Random mutations.

As to breeding back, the modified twin Pax-6 allele is lost by sexual recombination.

One thing I did not include in the original review -- quite simply because I hit up hard against the post size limit! – was de Beer’s final example. This is T.M. Sonneborn’s work on Paramecium. The citation de Beer gives is

Sonneborn, T.M. (1970) Gene action in development. Proc. R. Soc. B 176, 347.

Very briefly, Sonneborn showed that there are three distinct types of inheritance in the singled celled protist Paramecium. They are Mendelian inheritance via nuclear genes, cytoplasmic inheritance via mitochondrial genes, and [b]cortical inheritance[/i]. Quoting a modern on-line text by Scott F. Gilbert (DevBio: A Companion to Devlopmental Biology), and the section Inheritance in Ciliates,

This cortical inheritance has not yet been observed in non-ciliate cells, and it remains one of the great exceptions to the rule that genes provide the only vehicles for inheritance from one generation to another.

De Beer concludes his book with the following comment.

To the question ‘Is the whole of development encoded in DNA (that is to say, in the genes)?’ the answer in Paramecium is ‘No’. Whether this is applicable to ‘patterns’ in higher organisms, and whether homologous structures are controlled by non-DNA mechanisms awaits further research.

You needed to quote further into the article on ciliates you cited. The section on Cortical Inheritance:

"As depicted in Figure 2.8 of the textbook, Paramecium can exchange genetic material by fusing together and conjugating. Here, a haploid nucleus from each of the two individuals migrates across the fused area of the oral apparatus to "fertilize" the stationary haploid nucleus. Thereafter, the conjugants separate. Occasionally, this separation is not made cleanly, and a piece of the mate’s cortex gets picked up and integrated into the cortex of the other partner."

Now, since other animals don't reproduce like the ciliates, of course they don't have this type of inheritance. It appears that De Beer was not aware of conjugation in Paramecium.

But the particular arguments of de Beer were wrong.

The arguments are not necessarily wrong, altho I obviously disagree with one of them. Instead, some of the conclusions were wrong because De Beer did not have enough data. Based on the data he did have, most of his conclusions and concerns were legitimate.

What is not legitimate is taking De Beer out of context, not looking at the new data, and telling people that De Beer's conclusions are still valid.

Originally posted by lucaspa

Since natural selection can only add information, not subtract it.

Are you sure about this?

Roy

Originally posted by rthearle

Are you sure about this?

Roy

Pretty much. The general equation for information is -log2(M/N), where "log2" is logarithm to the base 2. M is the number selected and N is the number present. To quote Dembski: "Let me stress that this formula is not an case of misplaced mathematical exactness. This formula holds universally and is non-mysterious. Take a simple non-biological example. If I am sitting at a radio transmitter, and can transmit only zeros and ones, then every time I transmit a zero or one, I choose between two possibilities, selecting precisely one of them. Here N equals 2 and M equals 1. The information -log2(M/N) thus equals -log2(1/2) = 1, i.e., 1 bit of information n is introduced every time I transmit a zero or one. "

So, let's apply it to natural selection. Observation says that, in a generation of any population, more individuals are born than survive and reproduce. Thus N is alway greater than M. This means that M/N is always a negative logarithm and -(-log) = a positive number = increase in information.

This is what we see in biology. Take the bird's beak. It is without teeth. But, during embryonic development teeth develop only to be resorbed later. It would be simpler just to remove the information to make teeth. But that isn't what happened. Instead, information is added to resorb teeth that are already formed. Another example: J Diamond, Evolving backward. Discover 19: 64-71, Sept. 1998. Discusses the loss of eyes in the blind mole rat. But it is not indiscriminateloss of eye. Rather it is addition of info to remove only certain parts of the eye.

This is what fooled Behe in irreducible complexity. Structures are complex in Byzantine ways because selection can't remove informatoin. So every time finer control was needed in a system, instead of removing something that put the system out of control, another layer of control (information) was added.