A colleague suggested I post in several installments my past response to critics of Icons of Evolution. The series begins here, which includes the full reviewer citations.
Part 5: Four-winged fruit flies
According to neo-Darwinism (the modern form of Darwin’s theory), evolution results primarily from two factors: natural selection, which acts on variations already present in a population, and genetic mutations, which supposedly provide new variations which then become raw materials for evolution.
Since natural selection favors variations that benefit the organism, and tends to eliminate those that harm it, only beneficial mutations can provide raw materials for evolution. Some mutations benefit certain organisms by enhancing their ability to resist toxins (antibiotic resistance in bacteria is perhaps the best-known example of this). Such mutations typically act by deforming a molecule involved in the organism’s response to the toxin. Since organisms with the deformed molecule may survive in the presence of the toxin while others perish, such mutations are favored by natural selection. But Darwinian evolution needs a lot more than deformed molecules to explain the origin of new organs and body plans--it needs beneficial changes in anatomy.
To show how genetic mutations can provide raw materials for anatomical evolution, many biology textbooks feature pictures of a four-winged fruit fly. Fruit flies normally have two wings and two “balancers”--tiny appendages behind the wings that enable the insect to stabilize itself in flight. A skilled geneticist, however, can combine three separate DNA mutations to produce a fly in which the balancers are transformed into a normal-looking second pair of wings. Since some insects have four wings instead of two, the four-winged fruit fly seems at first glance to provide evidence for how one kind of insect evolved into another.
As I pointed out in Icons of Evolution, however, fruit flies with four normal-looking wings do not occur in nature; they must be engineered in a modern genetics laboratory. Furthermore, the extra wings have no muscles, so the mutant fly is a hopeless cripple that has great difficulty flying or mating. Outside the laboratory, natural selection would quickly eliminate it. Far from being raw material for evolution, the four-winged fruit fly is an evolutionary dead end. [16]
Reviewer Raff objects to my characterization of this icon: “Wells misuses the science he learned at Berkeley,” he writes, since “despite some pictures of suitably iconic four-winged Drosophila [the generic name for fruit flies], the discussion of genes and development in Icons of Evolution“ is “shabby and misleading.” (Raff, p. 374)
Raff doesn’t say exactly what he finds shabby and misleading. Presumably, though, it’s not my discussion of genes and development in four-winged fruit flies, since that part of my chapter was reviewed before publication by none other than Edward B. Lewis, the Nobel Prize-winning geneticist who made the first four-winged fruit fly. Although Lewis doesn’t endorse my criticisms of Darwinian evolution, he was kind enough to help me get my facts right.
Reviewers Padian and Gishlick (as we saw above in the discussion of the Cambrian explosion) object that “Wells entirely overlooks the explosive field of evolutionary developmental biology when he ignores the fact that… relatively early-acting, small, genetic changes in genes [can] affect features of body plans such as axis orientation, segmentation, and appendage formation.” (Padian & Gishlick, p. 35)
Since my entire chapter on four-winged fruit flies dealt with genetic changes that affect appendage formation, it’s difficult to see how Padian and Gishlick justify their claim that I ignore them. But if mutations affecting appendage formation are a problem for neo-Darwinism, mutations affecting axis formation and segmentation are even worse. As I pointed out in my book, developmental geneticists in the 1970s and 1980s used a technique known as “saturation mutagenesis” to screen for all possible mutations affecting embryo development in the fruit fly. Although this work shed considerable light on the role of genes in development (and led to some well-deserved Nobel Prizes), it also showed that mutations affecting axis formation and segmentation are invariably harmful, and indeed often fatal. [17] Such mutations cannot provide raw materials for evolution.
Reviewer Ussery feels that my ignorance of molecular biology extends further than developmental genetics. He writes: “Does Wells really believe that it is not true that DNA makes RNA makes protein?… I am seriously concerned that Wells claims himself to be a molecular biologist.” (Ussery, p. 74)
Of course, I have never implied that DNA doesn’t make RNA, or that RNA doesn’t make protein. In fact, a main point of my chapter on the four-winged fruit fly is precisely that DNA (through RNA) does make proteins--but that proteins alone are insufficient to specify the body plan of an organism, just as building materials are insufficient to specify the floor plan of a house. Defective 2x4’s can produce a deformed house, and mutant proteins can produce a deformed organism. Mutant proteins might even explain how some organisms might have lost previously existing features. But they do not account for changes in body plans. When it comes to the evolution of new morphologies or body plans, the question remains: Where is the evidence that DNA mutations can alter anatomy in beneficial ways and thereby provide raw materials for evolution?
It seems to me that this is a reasonable question. Despite my Berkeley Ph.D., I certainly don’t know everything about genes, development and evolution (after all, who does?). But until I actually see some good evidence for beneficial anatomical mutations, I’ll keep asking the question.
Come to think of it, shouldn’t all biologists be asking it?
NOTES:
[16] On what it takes to produce four-winged fruit flies, see E. B. Lewis, “A gene complex controlling segmentation in Drosophila,” Nature 276 (1978): 565-570; E. B. Lewis, “Control of Body Segment Differentation in Drosophila by the Bithorax Gene Complex,” pp. 269-288 in Max M. Burger & Rudolf Weber (editors), Embryonic Development, Part A: Genetic Aspects (New York, Alan R. Liss, 1982); E. B. Lewis, “Regulation of the Genes of the Bithorax Complex in Drosophila,” Cold Spring Harbor Symposia on Quantitative Biology 50 (1985): 155-164. On the absence of flight muscles in the extra pair of wings, see J. Fernandes, S. E. Celniker, E. B. Lewis & K. VijayRaghavan, “Muscle development in the four-winged Drosophila and the role of the Ultrabithorax gene,” Current Biology 4 (1994): 957-964; Sudipto Roy, L. S. Shashidhara & K. VijayRaghavan, “Muscles in the Drosophila second thoracic segment are patterned independently of autonomous homeotic gene function,” Current Biology 7 (1997): 222-227.
[17] On saturation mutagenesis in fruit flies, see Christiane Nüsslein-Volhard & Eric Wieschaus, “Mutations affecting segment number and polarity in Drosophila,” Nature 287 (1980): 795-801; Daniel St. Johnston & Christiane Nüsslein-Volhard, “The Origin of Pattern and Polarity in the Drosophila Embryo,” Cell 68 (1992): 201-219. Saturation mutagenesis has also been used in zebrafish; see Peter Aldhous, “ ‘Saturation screen’ lets zebrafish show their stripes,” Nature 404 (2000): 910; Gretchen Vogel, “Zebrafish Earns Its Stripes in Genetic Screens,” Science 288 (2000): 1160-1161.
