|
|
|
Ants in some pants (but no obscure dance all the way to France) Two points to get through first. First off, I give a cool nod to whoever got sentimental when they read the title of this article. Secondly, the meat of this paper was written as a student response to the ISCID 2002 Summer Workshop (I have since done some considerable revision and a great deal of addition). (Note 1) I responded to one of the papers posted on the suggested reading list: Abouheif, E and Wray, GA “Evolution of the Gene Network Underlying Wing Polyphenism in Ants.” Science 297: 249-52. (Ref. 1) This paper raises several points that deal with homology, ontogenetic variation, and generative entrenchment. While it is very interesting, the paper fails (at least in my limited understanding) to recognize the full implications of the particular study on which it reports. The field of evolutionary development (a.k.a., “evo-devo”), in a much broader sense, also fails in this regard. First, what is homology? I could pull a Bill Clinton and say, “It depends on what your definition of ‘is' is,” (Note 2) but I wouldn't want any of my readers (assuming I have any read-…well, nevermind) to vomit (again). This is a somewhat complicated issue in and of itself, however (Ref. 7), so I'll restrict my definition to a very general explanation. (Note 3) Homology is the “hypothesis that a particular similarity in two extant species predates their evolutionary divergence.” (Ref. 9, p. 675) Often homology has been used as evidence for evolution, which, as Jonathan Wells points out, is circular reasoning. (Ref. 8) Second, what do I mean when I speak of ontogenetic variation? At the earliest stages of development, noticeable differences are observed among a wide variety of organisms, but differences are also observed across the board at other stages of development. “Ontogeny” refers to development, so ontogenetic variation is simply varying developmental patterns exhibited by however many organisms are in question. Third, what is generative entrenchment? Although examples of early ontogenetic variation abound, development is known as a conserved process. In other words, it basically doesn't change. A decade ago Keith Stewart Thomson wrote that due to “the hierarchical nature of the development process, all structural features are basically linked together. Therefore it has been hard to see how major changes in any part of the phenotype could be produced with disruption of all – leading to hopeless monsters.” (Ref. 6, p. 111) A couple years ago Raff explained that “mutations in genes controlling early development would be expected to be deleterious, as they are bound to affect all of later development.” (Ref. 3, p. 76) Let's use an example, organism A. Organism A's development consists of ten different stages, stages 1-10. A change (mutation!) at stage 9 would affect stages 9-10; likewise, a change at stage 8 would affect stages 8-10. A change at stage 1, however, would affect all the subsequent stages (1-10). Thomson also noted, “it has long been a major tenet of evolution (and development) that any change to early developmental stages would be extremely disruptive or lethal.” (Ref. 6, p. 111) Because of their effects on later stages, earlier stages are more entrenched than those later stages; in other words, because mutations are known to be damaging, particularly the earliest stages of development will stay the same over time because mutations to them would be detrimental to organism A. (Note 4) The Abouheif and Wray paper was exceedingly technical in nature, so there were plenty of unfamiliar concepts. The authors themselves explain polyphenism, “the ability of a single genome to produce two or more alternative morphologies in response to an environment cue,” (Ref. 1, p. 249) but some other terms do not receive such attention. “Orthologous genes” are homologous genes. “Holometabolous” organisms, such as ants, are those that undergo complete metamorphosis, the process by which an organism changes from its larval stage to a completely distinct adult stage. “Imaginal” refers to the culminating fertile adult form, or imago, of an insect. “Imaginal discs” are structures that develop into wings and other appendages during metamorphosis. “Vestigial wing imaginal discs” are thus imaginal discs that don't develop into wings. “Embryogenesis” refers to the general developmental tract of (you guessed it) the embryo! An “instar” is “a stage in the life of an arthropod (as an insect) between two successive molts.” Molting is the “process by which the exoskeleton is shed” and replaced with another. “Epidermal” refers to the outer layer of skin, known as the epidermis. “Epidermal invaginations,” in the context of the paper, are small folded-in enclosures in the epidermis. “Eusociality” refers to a three-fold concept under which modern ants exist: 1) there is a caste system; 2) there is communal care for not-yet-fully-developed offspring; and 3) like in our human societies, many different generations of fit offspring exist at any given time. Phew! (Note 5) The authors explain the relevant characteristics of the different ant castes pretty well, so we can go over this quickly. There are two reproductive castes. Winged females (“queens”) make up one caste, while the other consists of winged males. The males develop from unfertilized eggs, making them haploid (like sperm cells), while the infertile females develop from fertilized eggs, making them diploid (like somatic cells). The soldier and worker castes are made up only of females, and none of them have wings. This caste structure is very important because a female queen cannot do very well by herself. The colony depends on the worker castes to take care of the queen's offspring, for example, and some colonies are made up of hundreds of thousands of ants. The problem at issue in this paper is how this polyphenism evolved. (Note 6) These authors “examined the expression of several ant genes orthologous to those from the wing-patterning network in Drosophila melanogaster (Fig. 1) to determine how the expression of this network changes during the development and evolution of winged and wingless ant castes.” (Ref. 1, p. 249) In other words, the paper assumes that the network in D. melanogaster basically works the same as the network in Pleidole morrisi, a species of ant. This assumption is not without evidence, of course (see below). I wonder if this may be a dangerous assumption, though, because homologous genes often do not have the same or even similar functions. Wray and Abouheif wrote as much in a paper written several years ago that grappled with the problem of homology. They noted that “homologous genes do not necessarily encode homologous structures,” (Ref. 9, p. 675) so perhaps concluding certain things about ant evolution may be jumping the gun until we have the wing-patterning network in P. morrisi understood more fully. But regardless, the gene expression patterns are conserved, so problems with homological analysis don't seem to apply here. It's worthwhile to point out that the relevant genes aren't orthologous unless we assume some sort of relatedness, however. Considering that the genes are involved in similar functions (i.e., wing development), it's not exactly surprising even from a design point of view that the genes would be similar in base sequence. The paper identifies several variations among wingless castes. The authors realize that this is “remarkable given that wing polyphenism evolved just once and that, within an ant species, all of the wing-patterning genes that are interrupted in workers must still retain the ability to specify and pattern the wings in queens and males, as well as the legs or central nervous system in all castes.” (Ref. 1, p. 251) From an evolutionary standpoint, no one knows how exactly this happened. Wray and Abouheif address the “simultaneous lability and conservation of the network underlying wing development.” (Ref. 1, p. 249) Basically, the genetic network regulating wing development in P. morrisi, is expressed similarly in flies and butterflies, both of which obviously have wings, so the network is conserved. The conservation of the wing development network is an example of generative entrenchment. At the same time, however, there are many detailed differences, alluded to above, among the four species of ants the authors examined (Pleidole morrisi, Neoformica nitidiventris, Crematogaster lineolata, and Myrmica Americana). The differences are in the expression of certain genes involved in the development of wings. It should be noted that this “simultaneous” developmental flexibility is assumed on the basis of common ancestry. If common ancestry is not assumed, then this lability morphs, linguistically speaking, into simple, unloaded differences (i.e., the differences in genetic expression among the four different ant species). If you bring evolution into the picture, then these differences need to be explained as arising through some mechanism. They can't just be explained away as unexplainably flexible developmental processes. I must add that I'm not faulting the paper for its assumption of an evolutionary paradigm, since I would hardly expect that mindset to be given up in the scientific literature any time soon, but am merely pointing out where this particular assumption comes into play. Developmental lability is not a new concept. In 2000, Raff noted that “early development evolves freely, allowing highly divergent ontogenies to evolve among closely related species.” (Ref. 3, p. 76) Due to this flexibility, “distinct developmental modes and larval features have evolved among sea urchins, starfish, ascidians, salamanders, frogs, nematodes, and even polyembryonic insects…These studies show that early development can evolve as radically as later development, and that it also can contribute marked evolutionary novelties.” (Ref. 3, p. 76-7) Keith Stewart Thomson wrote: “So the dilemma is easily resolved: because early stages have changed, they must be capable of change.” (Ref. 6, p. 112) Unfortunately for evolutionists, there is no evidence for evolvable early development beyond the assumption that it has to be possible! As explained earlier in this article, mutations in early developmental genes would harm the organism in question. Raff noted in 1996: “In developmental genetics, mutations are used to disrupt the functions of developmentally important genes. The resulting phenotypes are compared with the wild type. The defects are analyzed to define the role of the mutated gene.” [emphasis added] (Ref. 4, p. 217). In other words, mutations to developmental genes result in defects! This is an assumption, apparently, that development geneticists work under! Mutation is part of the two-fold neo-Darwinian mechanism, and yet how could the differences among the ant castes have evolved if mutations to developmental genes would result in defects, and would thus be acted against by natural selection? Paul Nelson and Jonathan Wells exposed this problem well in their paper on the subject. The base of the developmental hourglass records widespread early ontogenetic variation (Ref. 5), and such variation must be explained. Nelson and Wells concluded: “On the assumption that dramatic divergences in early development are due to evolution by natural selection from a common ancestor, we would expect to see substantial heritable variation in early development. The fact that we do not suggests that we may need to question basic assumptions.” (Ref. 2, p. 4) We may need to, indeed. Another problem for this study, a problem that seems to plague evolutionary explanations often (e.g., the Cambrian explosion), is that the changes must've been rapid. Wray and Abouheif write: “this lability occurs over relatively short time scales…despite the fact that the network has been largely conserved…fort the past 325 million years.” (Ref. 1, p. 251) So not only do evolutionists need to provide a mechanism, the explanation must fit in a 20-90-million-year time frame, as well. The authors speculate that the first interruption of the wing developmental network occurred “during the evolution of eusociality.” (Ref. 1, p. 251) This would make sense, of course, as this developmental interruption would have resulted in the caste system, one of three criteria for a eusocial society. But until a mechanism can be demonstrated, why should we assume that such an interruption in development even happened? Finally, the authors note that the “evolutionary mechanism driving developmental and genetic changes in the network since this event is not clear.” (Ref 1, p. 251) Yet this evolutionary mechanism is the exact same mechanism, I would think, that is required to explain developmental and genetic changes everywhere else, at least in a general sense. If the mechanism for worker and soldier castes spinning off from reproductive castes is not understood, then perhaps the evolutionary mechanism to account for developmental and genetic changes in the fuller story of life is also not understood. If this is true, then we are more than “justified in questioning basic assumptions.” (Ref. 2, p. 1) References: 1. Abouheif, E and Wray, GA “Evolution of the Gene Network Underlying Wing Polyphenism in Ants.” Science 297: 249-52. Notes: 1. I'm not sure if that's an exact quote or not. I have found two versions on the Internet, but I should at least get humor points if it's a misquotation. 2. I would highly recommend this workshop to my fellow students. We were able to discuss things live with Michael Behe, William Dembski, Paul Nelson, and many others. 3. I have read several papers on the subject of homology, and intend to write up one myself soon. It would, of course, be in the spirit of Paul Nelson and Jonathan Wells' 1997 paper (Ref. 7) and Wells' chapter on homology in Icons. (Ref. 8) 4. Paul Nelson explained this concept at the ISCID Workshop. http://www.iscid.org/workshops-2002-paulnelson.php If I have misunderstood it, it's completely my fault. The term apparently originated with William Wimsatt at the University of Chicago. 5. For the explanation of molting, I referred to “Introduction to Insects,” by Gary A. Dunn. It's available here: http://members.aol.com/YESedu/introbug.html . I depended on AOL's handy on-line dictionary for many quick definition look-ups: “embryogenesis,” “epidermal,” “epidermis,” “holometabolous,” “imaginal,” “imago,” “instar,” “invaginate,” and “invagination.” Merriam-Webster OnLine: Collegiate Dictionary. 2002. http://www.merriam-webster.com/dictionary.htm (11 Aug. 2002). The Dictionary of Molecular and Cell Biology – Online! was also helpful with “imaginal disc,” http://www.mblab.gla.ac.uk/~julian/dict2.cgi?3193 (11 Aug. 2002). I relied on http://www.anu.edu.au/BoZo/book/chap11.htm for the explanation of eusociality. 6. General information on ants was obtained from “ant.” The Columbia Encyclopedia, 6th ed. New York: Columbia University Press, 2001. www.bartleby.com/65/ . 11 Aug. 2002. 7. I should note that I have not read Raff's entire book. I try to make it a practice to read in their entirety all the books and papers I cite. I read about half of it, though, and feel that citing his other paper, which I have read in its entirety, for good measure should be enough for the purposes of this article. I have also read all the other papers cited, but not all of the web pages referred to, as they were purely for hard factual information. Comments? Contact Tristan Abbey at tabbey@idurc.org. Copyright 2002 idurc.org. All rights reserved. International copyright secured. Click here to return to the Papers Archive
|
