| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
a Department of Life Sciences, Arizona State University West, P. O. Box 37100, Phoenix, Arizona 85069
The 21st century will be the century of biology, and evolutionary biology will be the linchpin. Yes, this is a bold statement, but it is certainly in keeping with Dobzhansky's famous dictum about evolution. Dobzhansky was one of the major architects of the Modern Synthesis. This book by Schlichting and Pigliucci (S&P) is written with the books of the Modern Synthesis firmly in mind, especially that of Schmalhausen (1949). S&P are making a bold attempt at what I have termed the Final Synthesis (Scheiner, 1992).
Modern biology is built around two central paradigms. One paradigm is that of evolution and is the centerpiece of a range of disciplines including systematics, ecology, paleontology, natural history, and genetics. The Modern Synthesis was the gathering together of these disciplines under the evolution paradigm. The other paradigm is that of gene expression and is the focus of biochemistry, molecular biology, cell biology, physiology, developmental biology, and genetics. Over the past 45 years, ever since the work of Crick and Watson, these disciplines have been gathering under the umbrella of the Molecular Synthesis.
The Final Synthesis is the joining of these two major paradigms through the disciplines of genetics and developmental biology. The Modern Synthesis took place from ~ 1930 to 1955 and the Molecular Synthesis from ~ 1955 to 1990; the Final Synthesis is just gathering steam. Note that these dates are just rough guides and not meant to imply that significant work did not occur outside those periods. What is critical is that within these periods the relevant disciplines all become focussed on the same research program. We are now in the midst of this final joining of biological disciplines. This joining, if history is a guide, will take the next couple of decades.
This joining will be driven from the evolutionary side. It is the evolutionary side within which the disciplines are linked by a synthetic theory. The molecular side is primarily linked by a shared set of techniques and a viewpoint that centers all questions ultimately on DNA sequences. Because the evolution side is much more theory driven, that is where synthetic concepts tend to arise. This book by S&P is an excellent example of such theory-driven synthesis. Also, if truth be told, as a whole evolutionary biologists tend to have a larger perspective than those in many other biological disciplines.
Looking back on the Modern Synthesis is useful for the task ahead (Mayr and Provine, 1980). In particular, if we learn anything from that episode in our history, we should learn the perils of trying to make a theory too all encompassing. By the end of the Modern Synthesis much of ecology and related disciplines were in the grip of what Gould termed pan-adaptationism. Most traits were seen as adaptations, and organisms were viewed as being nearly perfectly adapted to their environment. This viewpoint came about by taking the theory of natural selection and attempting to fit the entire world under its banner. I must mention two caveats about pan-adaptationism. First, it is clear from reading the works of the central figures of the Modern Synthesis that they were never pan-adaptationists. It tended to be ecologists, rather than evolutionary biologists, who carried this banner. Second, at the time strong adaptationism was a rational position (Provine, 1986). Natural selection was the focus of many studies so the investigated traits, not surprisingly, were shown to be adaptations. And, to a first approximation, most traits are shaped primarily by natural selection. Since then we have spent much more effort on the other processes that contribute to evolution such as drift, mutation, and developmental constraints. This more complex view of evolution has now emanated from the originators of the Modern Synthesis.
Just as many advocates of the Modern Synthesis were guilty of pan-adaptationism, S&P are guilty of a similar sin. They try to draw too much into their developmental reaction norm (DNR) theory. The DNR is an attempt to draw together space — environmental heterogeneity — and time — ontogeny — into a single compound descriptor for an organism. While I am quite sympathetic to the outlook of S&P, I take exception to a number of their claims and attempts to apply their research program. My critiques are meant to be constructive. By not overreaching they will ultimately be much more successful. So what are they trying to accomplish?
S&P are trying to bring to the forefront of evolutionary theory the importance of development and the environment in shaping phenotypes. As they put it, "...our ultimate goal is to understand how different forms have evolved" (p. 228). They do so by advancing a research program (sensu Lakatos, 1974). In this framework a research program has three main components: core concepts, auxiliary concepts, and research projections. The most important of the core concepts of S&P is that the DNR is the object of selection.
Here S&P make two mistakes. The first mistake is the one of overreaching. They propose the DNR as the object of selection rather than an object of selection. They want all of selection to be on complex ontogenies. Yet we have plenty of evidence that this need not be the case. From laboratory studies we know that selection can alter some portions of an organism while having little or no effect on the rest of the organism. From comparisons of natural populations we know that natural selection results in the fine-tuning of adaptations, again implying an ability to independently alter small parts of the phenotype. I agree with their claim that this is not always the case. S&P provide plenty of examples of selection on one portion of the phenotype having pervasive effects. But not always. S&P undercut their own thesis with a number of counterexamples.
A less critical mistake is identifying their ideas too closely with Lakatos. One aspect of Lakatos' model of science is that the core concepts are not up for debate. Contrary to what is stated by S&P, the core concepts are not falsifiable, merely discardable. Lakatos attempted to reconcile empiricist and social constructivist views of philosophy of science in a reaction to Popper (1959) and Kuhn (1962). None of these positions accurately reflects true scientific practice, which is actually much more fluid (Mayo, 1996).
Despite these caveats, S&P make several important contributions to evolutionary ideas. For the readers of this journal, they place botany much more front and center than is found in most evolution texts. Throughout the book they use botanical examples from the work of themselves and others. They emphasize the groundbreaking work of Clausen, Keck, and Heisey (1940). Here again, though, I must take issue. They suggest through the use of citations, or the lack of them, that Stebbins deliberately ignored the work of Clausen, Keck, and Heisey, which contributed to it being overlooked for many years (p. 49). I think that there is a much simpler explanation. The reason that plants were mostly left out of the Modern Synthesis was the absence of plant population biology as a recognizable discipline. Plant population biology arose in Great Britain in the 1960s with the efforts of Harper, Bradshaw, and their students. It was then imported to United States in the 1970s. Carl Schlichting and I are both children of that importation. Previous to that, botanical research in ecology and evolutionary biology focussed primarily on either agricultural genetics, taxonomy and systematics, community ecology, or physiology. See Allen, Mitman, and Hoekstra (1993) for an ecological perspective on botanical research in the first half of this century.
More important than giving proper prominence to plants, S&P move phenotypic plasticity to center stage. Plasticity has been a hot topic for the past 15 years. Interestingly, it was simultaneously discovered by a number of people. Besides the ones listed by S&P (Stearns, Scheiner, Schlichting, and Via, p. 260), add De Jong and Gavrilets. Of course research on this topic stretches much further back. A number of people did research on plasticity in the 1970s, including an extensive series of experiments by Jinks and his students (Jinks and Connolly, 1973; Jinks, Perkins, and Pooni, 1973; Jinks, Jayasekara, and Boughey 1977; Brumpton, Boughey, and Jinks, 1977). In the past dozen years several review articles have appeared (Schlichting, 1986; Sultan, 1987; West-Eberhard, 1989; Stearns, De Jong, and Newman, 1991; Scheiner, 1993; Via et al., 1995; DeWitt, Sih, and Wilson, 1998). The time is now ripe for broader treatments. Besides S&P, two books on phenotypic plasticity are scheduled to appear next year by Van Tienderen and Brakefield, and DeWitt and Scheiner.
The other topic that S&P move front and center is development. Ever since claims by Gould and others in the 1970s that developmental constraints are central to patterns of evolution, these ideas have been shouldering their way into evolutionary theory. Why did it take so long for development and plasticity to move to the forefront of evolutionary thinking? As S&P point out, these are not new ideas. They were certainly around at the dawn of the Modern Synthesis. Besides book-length treatments such as those of Schmalhausen (1949) and Goldschmidt (1940), these ideas appear in the writings of people such as Wright (1932). The answer is straightforward. On the evolutionary side it took 60 years just to work out the theory in the simplest cases of a single environment and multiple environments with constant genotypic expression. Only now have we progressed far enough that we can explore the more complicated case of genotype–environment interaction. On the development side we had to wait for the Molecular Synthesis to provide the necessary information about the molecular and cellular bases of development. Even now we are still struggling with relatively simple organisms such as Caenorhabdites (a fixed number of cells) and Arabidopsis (a short-lived annual). It will be awhile before we know as much about more complex organisms. More is known on the animal side where Drosophila is rather well examined and the desire to understand human biology pushes mammalian studies. On the plant side the other well-studied organism is corn, another annual. I suspect that it will be quite some time before we know nearly as much about any long-lived plant.
A central theme of the book is the interaction, or perhaps tension, between the internal and the external. This tension extends prior to the Modern Synthesis in the arguments between the Mendelists and the Biometricians (Provine, 1971). That debate was between those that saw evolution mainly determined by mutation—internal processes—vs. those that saw it determined by natural selection—external processes. While the Modern Synthesis was a reconciling of these positions, it also came down squarely on the side of external processes being the main shaper of evolutionary trajectories. In the 1970s the internal was brought back into the fray by Gould, Alberch, and others. S&P very nicely describe this renaissance of internalist concerns in their chapters on allometry and ontogeny. Phenotypic plasticity has widened this debate in two ways. First, the external now comes into play twice, both as a shaper of selection and as a factor in how development creates phenotype. The internal is also made more complex because feedback effects can come about and be affected by both internal and external signals.
In S&P this external/internal duality makes itself known in a number of ways. For example, S&P discuss the issue of constraints on evolution and how we should classify and understand them. They group them under two headings: genetic/epigenetic and selective (p. 168). These two headings could be thought of as internal and external, although I would split their category of developmental constraints, which confounds the duality. Sometimes the external/internal duality is explicit, other times it is not. I recommend keeping this duality in mind as you read this book. S&P are clearly of the mind that both together determine evolutionary patterns. I am less convinced. I tend to the position that internal constraints are only temporary, albeit that temporary may mean millions of years! Ultimately it is the external that shapes evolution.
In what other ways have S&P overextended themselves? One example is how they spread the definition of reaction norm too thinly. By their definition, "...reaction norm refers to the set of phenotypes that can be produced by an individual genotype that is exposed to different environmental conditions" (p. 51). This definition has problems. As they acknowledge in a footnote, the usual definition of reaction norm refers to a gradient of a single environmental factor. This definition has been the standard for at least the past several decades, not just a recent invention by Stearns as they imply. While they are correct that the original definition of Woltereck was more encompassing, his definition gave way to the more restrictive one in the 1950s. There was a good reason for this shift. By limiting what is meant by a reaction norm, it becomes tractable. We now have models of reaction norms (e.g., De Jong, 1990; Gibert et al., 1998) and theories about their evolution (e.g., Gomulkiewicz and Kirkpatrick, 1992; Gavrilets and Scheiner 1993).
The definition of S&P also ignores the distinction between discrete and continuous environmental variation, which may be critical. While S&P discuss in detail the evolution of discrete forms of plasticity, they ignore the question of the relationship between discrete shifts in phenotype (developmental conversion) and discrete environments. I agree with them that discrete environments can lead to developmental conversion (e.g., heterophylly). However, it is not at all clear whether this will be true for continuous environmental variation. They claim that "...developmental conversion is a more sophisticated and complex mode of plastic response compared with phenotypic modulation" (p. 336). Here I must disagree. The issue is one between adaptive and nonadaptive plasticity, which map roughly onto what is termed active and passive plasticity. Only roughly because it is possible for passive plasticity to be adaptive and active plasticity could be a nonadaptive holdover. An example of active plasticity is stem elongation under low light, an active developmental switch by the plant, while an example of passive plasticity is less growth under low nutrient conditions—a response that the plant has no control over. Active, adaptive responses to continuous environmental variation that result in continuous phenotypic variation can be just as complex as developmental conversion. If the optimal phenotype varies continuously, it may be more difficult for a plant to find the right match than under developmental conversion. We certainly have plenty of examples of adaptive plasticity of continuous traits (e.g., leaf thickness in Dicerandra [Winn, 1996], wing/thorax ratio in Drosophila [David et al., 1994], gall size in Eurosta [Weis and Gorman, 1990], offspring size in exothermic animals [Yampolsky and Scheiner, 1996]).
The domain of the theory of plasticity evolution is development under uncertainty. Just as the theory of natural selection does not apply to all traits, the theory of plasticity evolution does not apply to all forms of plasticity. The key to theory development in this case is to restrict it to instances where there is a lag between the decision by the organism and when selection occurs. Under such circumstances the primary limits and constraints on the evolution of plasticity are external and informational. Otherwise we are now in the domain of development, physiology, and genetics and the theories which pertain to them. By restricting our focus we are more likely to make progress. One result of this restriction is the realization that conditions for adaptive evolution of plasticity are likely more restrictive than implied by S&P (pp. 78–81). Plasticity evolution is hampered by low heritability and the fact that it will often depend on global fitness across demes (Scheiner, 1993, 1998). One surprising oversight of S&P is a failure to discuss this other set of environmental factors, demic structure. This oversight is even more surprising given their focus on Wright's Shifting Balance Theory. One of the key next steps in studies of plasticity evolution is to expand beyond two environments. S&P emphasize this with regard to laboratory studies. This shift is also needed for theoretical and field studies.
Another overextension problem with their definition of reaction norm is that it includes all traits at once. Again, by making the reaction norm everything, it becomes intractable. As much as I fight against the proliferation of jargon, in this case I must insist that if they want a term that is all-encompassing both environmentally and phenotypically, they need a new word. The same sort of problem occurs with their definition of allometry. Historically allometry has always referred to morphology. To extend the definition to all correlations (p. 99) makes it lose meaning, especially when a phrase like "changes in correlations during development" is simple and clear. Again, if S&P want a single term in place of that phrase, they should coin one.
While I praise S&P for their strong botanical emphasis, in their zeal to create an all-encompassing research program they never address a fundamental question. Are plants and animals different when it comes to development and evolutionary constraints? Animals have closed developmental systems, while plants have open systems with totipotent meristems. Plants do not have the clear distinction between soma and germ line that animals have. Of course my division here into plants and animals is too simplistic and in this regard there are plant-like animals and animal-like plants. It would be instructive to examine how these developmental differences create different evolutionary processes.
While S&P are to be commended for the extensive literature that they cover, sometimes they are careless with attributions and data. I have already mentioned the Stearns reaction norm definition. Another example is the effects of plasticity on genetic covariances (pp. 106–107), which should be credited to De Jong (1990), not the subsequent review paper (Stearns, De Jong, and Newman, 1991). In the case of one of my own papers, they state that we reported both broad- and narrow-sense genetic correlations for plasticity, when we only reported narrow-sense ones (p. 222; Scheiner, Caplan, and Lyman, 1991). The other correlations that we reported were mean-family correlations, which are not the same as broad-sense genetic correlations. Of course it is easy to spot errors about one's own work and I did not see much of this for material with which I was familiar. For the most part, errors tended to be ones of omission rather than commission.
In general S&P write in a clear and engaging style. They are not adverse to adding a touch of whimsy here and there. Sometimes, though, they get a bit too cute, for example calling Clausen, Keck, and Heisey "the original dream team" (p. 41). Their treatment of these three men as a single unit ignores the fact that their series Experimental Studies on the Nature of Plant Species did not have all of them as authors on all books and ignores their individual and separate contributions to botanical research.
Overall, despite my many caveats, I like this book. This review has already gone on too long and my copy is still full of Post-It(TM) notes containing reactions to various claims and statements. But that shows the value of this book. Any good book, especially one as ambitious as this, should provoke strong reactions. Often an extreme position helps to clarify the issues and to encourage research. The best science is not wishy-washy. S&P have done a fine job assembling a diverse literature and setting out their ideas cogently and clearly. I recommend that you read it.
FOOTNOTES
1 Phenotypic evolution: a reaction norm perspective. Carl D. Schlichting and Massimo Pigliucci. Sinauer Associates, Inc. 1998. 387 pages. $38.95 paper. ISBN 0-87893-799-4 
REFERENCES
Allen, T. F. H., G. Mitman, and T. W. Hoekstra.1993Synthesis mid-century: J. T. Curtis and the continuum concept. In J. S. Fralish, R. P. McIntosh, and O. L. Loucks [eds.], John T. Curtis Fifty years of Wisconsin plant ecology, 123–143. Wisconsin Academy of Sciences, Arts & Letters, Madison, WI.
Brumpton, R. J., H. Boughey, and J. L. Jinks.1977Joint selection for both extremes of mean performance and of sensitivity to a macroenvironmental variable. I. Family selection. Heredity 38: 219–226. [ISI]
Clausen, J. D., D. Keck, and W. M. Hiesey.1940Experimental studies on the nature of species. I. Effects of varied environments on Western North American plants. Carnegie Institure of Washington, Washington, DC.
David, J. R., B. Moreteau, J. R. Gauthier, G. Pétavy, J. Stockel, and A. Imasheva.1994Reaction norms of size characters in relation to growth temperature in Drosophila melanogaster: an isofemale lines analysis. Genetics, Selection, Evolution 26: 229–251. [CrossRef][ISI]
De Jong, G.1990Quantitative genetics of reaction norms. Journal of Evolutionary Biology 3: 447–468.
DeWitt, T. J., A. Sih, and D. S. Wilson.1998Costs and limits of phenotypic plasticity. Trends in Ecology and Evolution 13: 77–81.
Gavrilets, S., and S. M. Scheiner.1993The genetics of phenotypic plasticity. V. Evolution of reaction norm shape. Journal of Evolutionary Biology 6: 31–48.
Gibert, P., B. Moreteau, J. R. David, and S. M. Scheiner.1998Describing the evolution of reaction norm shape: body pigmentation in Drosophila. Evolution 52: 1501–1506. [CrossRef][ISI]
Goldschmidt, R. B.1940The material basis of evolution. Yale University Press, New Haven, CT.
Gomulkiewicz, R., and M. Kirkpatrick.1992Quantitative genetics and the evolution of reaction norms. Evolution 46: 390–411. [CrossRef][ISI]
Jinks, J. L., and V. Connolly.1973Selection for specific and general response to environmental differences. Heredity 30: 33–40. [ISI]
———, N. E. M. Jayasekara, and H. Boughey.1977Joint selection for both extremes of mean performance and of sensitivity to a macroenvironmental variable. II. Single seed descent. Heredity 39: 345–355. [ISI]
———, J. M. Perkins, and H. S. Pooni.1973The incidence of epistasis in normal and extreme environments. Heredity 31: 263–269. [ISI]
Kuhn, T. S.1962The structure of scientific revolutions. University of Chicago Press, Chicago, IL.
Lakatos, I.1974Popper on demarcation and induction. In P. A. Schilpp [ed.], The philosophy of Karl Popper, 241–273. Open Court Publishing Co., LaSalle, IL
Mayo, D. G.1996Error and the growth of experimental knowledge. University of Chicago Press, Chicago, IL.
Mayr, E., and W. B. Provine [eds.].1980The evolutionary synthesis. Harvard University Press, Cambridge, MA.
Popper, K. R.1959The logic of scientific discovery. Hutchinson & Co., London, UK.
Provine, W. B.1971The origins of theoretical population genetics. University of Chicago Press, Chicago, IL.
———.1986Sewall Wright and evolutionary biology. University of Chicago Press, Chicago, IL.
Scheiner, S. M.1992Grand syntheses in the making: review of Stearns, S. C., The Evolution of Life Histories. Science 258: 1820–1822.
———.1993Genetics and evolution of phenotypic plasticity. Annual Review of Ecology and Systematics 24: 35–68. [CrossRef][ISI]
———.1998The genetics of phenotypic plasticity. VII. Evolution in a spatially structured environment. Journal of Evolutionary Biology 11: 303–320. [CrossRef][ISI]
———, R. L. Caplan, and R. F. Lyman.1991The genetics of phenotypic plasticity. III. Genetic correlations and fluctuating asymmetries. Journal of Evolutionary Biology 4: 51–68.
Schlichting, C. D.1986The evolution of phenotypic plasticity in plants. Annual Review of Ecology and Systematics 17: 667–693. [CrossRef][ISI]
Schmalhausen, I. I.1949Factors of evolution. Blakiston, Philadelphia, PA.
Stearns, S. C., G. De Jong, and B. Newman.1991The effects of phenotypic plasticity on genetic correlations. Trends in Ecology and Evolution 6: 122–126.
Sultan, S. E.1987Evolutionary implications of phenotypic plasticity. Evolutionary Biology 21: 127–178. [ISI]
Via, S., R. Gomulkiewicz, G. De Jong, S. M. Scheiner, C. D. Schlichting, and P. Van Tienderen.1995Adaptive phenotypic plasticity: consensus and controversy. Trends in Ecology and Evolution 10: 212–217. [CrossRef]
Weis, A. E., and W. L. Gorman.1990Measuring selection on reaction norms: an exploration of the Eurosta-Solidago system. Evolution 44: 820–831. [CrossRef][ISI]
West-Eberhard, M. J.1989Phenotypic plasticity and the origins of diversity. Annual Review of Ecology and Systematics 20: 249–278.
Winn, A. A.1996The contributions of programmed developmental change and phenotypic plasticity to within-individual variation in leaf traits in Dicerandra linearifolia. Journal of Evolutionary Biology 9: 737–752. [CrossRef][ISI]
Wright, S.1932The roles of mutation, inbreeding, crossbreeding and selection in evolution. Proceedings of the Sixth International Congress of Genetics 1: 356–366.
Yampolsky, L. Y., and S. M. Scheiner.1996Why larger offspring at lower temperatures? A demographic approach. American Naturalist 147: 86–100. [CrossRef][ISI]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
