Correcting Darwin’s Other Mistake (M.R. Rose & T. Garland Jr., 2009; see below)
We are taught early in our education as evolutionists that Charles Darwin got the mechanism of heredity wrong. He supposed that there are an arbitrary number of ductile transmissible gemmules that migrate from the organs to the gonads, allowing the possibility of a kind of blending inheritance along with the inheritance of acquired characters. Of course, Gregor Mendel’s discrete “hard” model for inheritance, which we now call genetics, turned out to be the correct mechanism for inheritance in eukaryotes. Furthermore, Darwin’s mistake about inheritance probably cost the field of evolutionary biology some decades of delay. Genetics wasn’t properly incorporated into evolutionary biology until the work of Fisher, Haldane, Wright, and Dobzhansky, in the period from 1910 to 1940 (Provine 1971; Mayr and Provine 1980). Regrettably, the one person who realized that genetics supplied the mechanism of heredity that Darwinian evolution needed was Mendel, a humble monk who died unappreciated in 1884, some twenty years after he had worked out the basic principles of inheritance in plants. If Darwin had read Mendel with understanding in the 1860s, it is conceivable that modern evolutionary biology would have developed some fifty years earlier than it did, although such counterfactual speculation is of course essentially an idle exercise.
One of the common themes in the classroom presentation of Darwin’s erroneous reasoning concerning heredity is the influence of Darwin’s gradualist prejudices. It is well-known that Darwin was in many respects a disciple of Charles Lyell, the leading gradualist geologist of 19th Century England. Lyell essentially founded modern scientific geology. Darwin was Secretary of the Geological Society early in his career, a scientific society dominated by Lyell’s thinking, particularly his methodological strictures. The cardinal axiom in Lyell’s geology was the idea that change in nature proceeds by gradual, observable, concrete mechanisms. In geology, such mechanisms are illustrated by erosion, subsidence, deposition, and the like. Darwin imported this style of thinking into biology. This led him to disparage the importance of discrete heritable variants, which he called “sports.” That in turn prevented Darwin from giving appropriate attention to the hypothesis of discrete inheritance, leading evolutionary biology up a blind alley of blending inheritance. This was the famous mistake that is a key motif in the education of beginning evolutionary biologists.
Darwin’s other mistake also came from his gradualist preconceptions. He repeatedly emphasized that natural selection acts only by slow accretion (Zimmer 2006). Darwin expected the action of selection within each generation to be almost imperceptible, even if thousands of generations of selection could evidently produce large differences between species:
“. . . natural selection will always act very slowly, often only at long intervals of time, and generally on only a very few of the inhabitants of the same region at the same time. I further believe, that this very slow, intermittent action of natural selection accords perfectly well with what geology tells us of the rate and manner at which the inhabitants of this world have changed.”
(Darwin, Origin of Species, 1st edition, Chapter IV)
Notably, the word “slowly” appears dozens of times in the Origin.
For modern scientists, at least, the problem with this assumption is that it implies that the action of natural selection will normally be very difficult to observe. Indeed, Darwin himself made no significant attempt to study natural selection in the wild. Instead, he studied the systematics of barnacles, bred pigeons, and crossed plants. He was certainly interested in both the long-term effects of evolution and the short-term effects of crosses, but he did not apparently seek out opportunities to study the process of natural selection itself. The closest he came to this was collecting an abundance of information on artificial selection from breeders, both agricultural and hobbyist, and discussions of their various results figure prominently in the Origin.
This dereliction did not persist, fortunately. At the start of the 20th Century, W.F.R. Weldon (1901) published a pioneering study of selection on the morphology of estuarine crabs. In 1915, W.E. Castle published reasonably quantitative data on the response to “mass selection” on coat coloration in rats. In the 1930s, animal breeders such as Jay L. Lush took up the quantitative genetics theory developed initially by R.A. Fisher to implement well-designed breeding programs. Theodosius Dobzhansky started the “Genetics of Natural Populations” series of articles in the 1930s, studying selection on the chromosomal inversions of Drosophila in both wild and laboratory populations, often enlisting the aid of Sewall Wright. Ecological geneticists such as E.B. Ford and H.B.D. Kettlewell studied industrial melanism, one of our best examples of natural selection in the wild (Clarke 2003). Starting from this wide range of groundbreaking work, evolutionary biology has developed into a substantial body of empirically founded knowledge.
But there remains a tendency to adopt unthinkingly Charles Darwin’s bias that natural selection is typically slow and difficult to observe. Very old patterns of research have persisted: studies of phylogenetics, genetic variation within and among populations, and occasional dramatic instances of natural selection in the wild have featured prominently in evolutionary research. As these research paradigms have persisted, and indeed dominated within evolutionary biology, experimental evolution has been slow to develop as a research strategy. In 1976, MRR did not consider trying selection for slowed aging in Drosophila because of the expectation that it would take too long to yield observable results. Reading about the results of an inadvertent and misinterpreted selection experiment by a neoLamarckian (Wattiaux 1968) in 1977 was the trigger that enabled MRR to overcome his typical Darwinian inhibitions, leading to a deliberate test of Hamilton’s (1966) analysis of the evolution of aging using laboratory evolution (Rose et al. 2004; Rose 2005). Now, of course, such laboratory evolution experiments on life-history characters are common in evolutionary biology. It was the pioneering work of Carol B. Lynch (1979; review in Lynch 1994) which convinced TG that selection experiments were actually practical for addressing classic questions in physiological ecology. Now, selection experiments of various types are common in evolutionary physiology (Bennett 2003; Garland 2003; Swallow and Garland 2005; Swallow et al. this volume). Yet as recently as 2005, TG observed colleagues discourage graduate students from undertaking selection experiments in this area. Old biases die lingering deaths.
It is our conviction that the Darwinian inhibition about experimental research on evolution should now be resolutely discarded.
Indeed, experimental evolution is key to the ongoing effort to foster biology’s reincarnation as a fully scientific field. It is only when evolutionary histories are known, controlled, and replicated that we can fairly claim to be performing rigorous experimental work. The biology of character X in inbred or mutant strain Y is like a beautiful painting: unique, intriguing, but of uncertain provenance or meaning. Any result with arbitrary strain Y may not be true of other strains or outbred populations of that species. And it will often be unclear how to sort out this situation. Strains M, Q, X, and Z might or might not have the same features. Individual outbred populations are marginally better, since they should have a broader set of genotypes, but they are still unique biological examples, of less reliability than postage stamps that are mass-produced to well-defined standards. If Ernest Rutherford could declare that science can be divided into physics and stamp-collecting, much of biology doesn’t even rise to the level of stamp-collecting. In its emphasis on quantitative trajectories, replication, and reproducibility, experimental evolution resembles physics more than it resembles most research in biology. We can only hope that both Darwin and Rutherford would have approved.
Definitions and Concepts
What is experimental evolution? We use the term to mean research in which populations are studied across multiple generations under defined and reproducible conditions, whether in the laboratory or in nature (for recent overviews, see Bennett 2003; Garland 2003; Swallow and Garland 2005; Chippindale 2006; Garland and Kelly 2006). This intentionally general definition subsumes various types of experiments that involve evolutionary (cross-generational, genetically based) changes. At one end of the continuum, the study of evolutionary responses to naturally occurring events (e.g., droughts, fires, invasions, epidemics) may constitute a kind of adventitious experimental evolution, especially if these events occur repeatedly and predictably enough that the study can be replicated, either simultaneously or in subsequent years. One might also include “adaptations to the humanized landscape” (Bell 2008a), such as industrial melanism in moths (Clarke 2003). Next, we have “invasive species,” which often invade repeatedly, thus allowing study of replicated events (Huey et al. 2005; Gilchrist and Lee 2007; Lee et al. 2007). Intentional “field introductions” involve populations placed in a new habitat in the wild, or cases in which a population’s habitat is altered by adding a predator, a pesticide, a food source, fertilizer, etc. The experimental population is then monitored across generations and compared with an unmanipulated control population …
“Laboratory natural selection” denotes experiments in which the environment of a laboratory-maintained population is altered (e.g., change of temperature, culture medium, food) as compared with an unaltered control population. “Laboratory culling” involves exposing an experimental population to a stress that is lethal (or sublethal) and then allowing the survivors (or the hardiest) to become the parents of the next generation. In all of the foregoing types of experiments, the investigator does not specifically measure and select individuals based on a particular phenotypic trait or combination of traits. Rather, selection is imposed in a general way, and the population has relatively great freedom to respond across multiple levels of biological organization (e.g., via behavior, morphology, physiology). “Multiple solutions” (different adaptive responses among replicate lines) are possible and even probable, depending on the kind of organism and experimental design.
Domestication is an interesting (and ancient) type of experimental evolution that generally involves some amount of intentional selective breeding. In some cases, the process has been replicated enough times that general principles might be discerned (e.g., several species of rodents have been domesticated). Of course, whenever organisms are brought from the wild to the laboratory or agricultural setting, some amount of adaptation to the new conditions will occur, and this may be studied. Once domesticated, organisms may be the subject of additional selective breeding programs, with varying degrees of control and replication, leading to multiple breeds or lines.
What we are terming “experimental evolution” clearly covers a broad range of possible experiments. Historically and at present, different methodologies for experimental evolution have been and are being applied unequally across topics (e.g., behavior, physiology, morphology) and across kinds of organisms (e.g., bacteria, Drosophila, rodents). …
In any case, to qualify as experimental evolution, we require most if not all of the following fundamental design elements: maintenance of control populations, simultaneous replication, observation over multiple generations, and the prospect of detailed genetic analysis. In short, experimental evolution is evolutionary biology in its most empirical guise.
Experimental evolution is becoming a mainstream part of the biological sciences, beyond the confines of evolutionary biology, narrowly construed. For example, the journal Integrative and Comparative Biology recently published a symposium on “Selection experiments as a tool in evolutionary and comparative physiology: Insights into complex traits” (Swallow and Garland 2005). In 2007, the journal Physiological and Biochemical Zoology published a “Focused Issue” on “Experimental Evolution and Artificial Selection.” The response to the call for papers was so great that they ended up publishing papers in parts of three successive issues, including several by contributors to this volume. In 2008, the journal Heredity published a collection of six short reviews on microbial studies using experimental evolution (Bell 2008b).
In closing, we would appeal to the words of our colleagues regarding the importance of selection experiments and experimental evolution:
“Ultimately, laboratory systems provide the best opportunity for the study of natural selection, genetic variation, and evolutionary response in the same population. … we suggest that the study of natural selection in a laboratory setting is the best method of making the link between natural selection and evolution and may thus permit predictive and rigorous study of adaptation.”
(Houle and Rowe 2003, pp. 50-51)
“… selection experiments are irreplaceable tools for answering questions about adaptation and the genetic basis of adaptive trait clusters (i.e., repeated evolution of suites of traits in particular environments).”
(Fuller et al. 2005, p. 391)
Bell, G. 2008a. Selection: the mechanism of evolution. 2nd ed. Oxford University Press, Oxford, U.K. xiii + 553 pp.
Bell, G. 2008b. Experimental evolution. Heredity 100:441-442.
Bennett, A. F. 2003. Experimental evolution and the Krogh Principle: generating biological novelty for functional and genetic analyses. Physiological and Biochemical Zoology 76:1-11.
Charlesworth, B., R. Lande, and M. Slatkin. 1982. A neo-Darwinian commentary on macroevolution. Evolution 36:474-498.
Chippindale, A. K. 2006. Experimental evolution. In Evolutionary Genetics, C. Fox and J. Wolf (eds.). Oxford University Press.
Clarke, B. 2003. The art of innuendo. Heredity 90:279-280.
Cooke, S. J., C. D. Suski, K. G. Ostrand, D. H. Wahl, and D. P. Philipp. 2007. Physiological and behavioral consequences of long-term artificial selection for vulnerability to recreational angling in a teleost fish. Physiological and Biochemical Zoology 80:480-490.
Fuller, R. C., C. F. Baer, and J. Travis. 2005. How and when selection experiments might actually be useful. Integrative and Comparative Biology 45:391-404.
Garland, T., Jr. 2003. Selection experiments: an under-utilized tool in biomechanics and organismal biology. Pages 23-56 in V. L. Bels, J.-P. Gasc, A. Casinos, eds. Vertebrate biomechanics and evolution. BIOS Scientific Publishers, Oxford, U.K.
Garland, T., Jr., and S. A. Kelly. 2006. Phenotypic plasticity and experimental evolution. Journal of Experimental Biology 209:2234-2261.
Garland, T., Jr., A. F. Bennett, and E. L. Rezende. 2005. Phylogenetic approaches in comparative physiology. Journal of Experimental Biology 208:3015-3035.
Gavrilets, S., and A. Vose. 2005. Dynamic patterns of adaptive radiation. Proceedings of the National Academy of Sciences, USA 102:18040-18045.
Gilchrist, G. W., and C. E. Lee. 2007. All stressed out and nowhere to go: does evolvability limit adaptation in invasive species? Genetica 129:127-132.
Hamilton, W. D. 1966. The moulding of senescence by natural selection. Journal of Theoretical Biology 12:12-45.
Hard, J. J.. M. R. Gross, M. Heino, R. Hilborn, R. G. Kope, R. Law, and J. D. Reynolds. 2008. Evolutionary consequences of fishing and their implications for salmon. Evolutionary Applications 1:388-408.
Houle, D., and L. Rowe. 2003. Natural selection in a bottle. American Naturalist 161:50-67.
Huey, R. B., G. W. Gilchrist, and A. P. Hendry. 2005. Using invasive species to study evolution. In: Species invasions: insights to ecology, evolution and biogeography. Pp. 139-164 in D. F. Sax, S. D. Gaines, and J. J. Stachowicz, eds. Sinauer Associates, Sunderland, MA.
Lee, C. E., J. L. Remfert, and Y. M. Chang. 2007. Response to selection and evolvability of invasive populations. Genetica 129:179-192.
Lynch, C. B. 1980. Response to divergent selection for nesting behavior in Mus musculus. Genetics 96:757-765.
Lynch, C. B. 1994. Evolutionary inferences from genetic analyses of cold adaptation in laboratory and wild populations of the house mouse. Pages 278-301 in C. R. B. Boake, ed. Quantitative genetic studies of behavioral evolution. University of Chicago Press, Chicago.
Mayr, E., and W. B. Provine, eds. 1980. The evolutionary synthesis. xi + 487 pp.
Provine, W. B. 1971. The origins of theoretical population genetics. Chicago, University of Chicago Press.
Rose, M. R. 2005. The Long Tomorrow; How evolution can help us postpone aging. Oxford University Press, New York.
Rose, M. R., H. B. Passananti, and M. Matos, eds. 2004. Methuselah flies: A case study in the evolution of aging. World Scientific Publishing, Singapore.
Swallow, J. G., and T. Garland, Jr. 2005. Selection experiments as a tool in evolutionary and comparative physiology: insights into complex traits – An introduction to the symposium. Integrative and Comparative Biology 45:387-390.
Wattiaux, J. M. 1968. Cumulative parental age effects in Drosophila subobscura. Evolution 22:406-421.
Weldon, W. F. R. 1901. A first study of natural selection in Clausilia laminata (Montagu). Biometrika 1:109-124.
Zimmer, C. 2006. Evolution in a petri dish. http://www.yalealumnimagazine.com/issues/2006_05/evolution.html.
“Darwin’s Other Mistake”
Excerpted from Chapter 1 for
EXPERIMENTAL EVOLUTION: Concepts, Methods, and Applications of Selection Experiments
Theodore Garland, Jr. and Michael R. Rose
University of California Press
Michael R. Rose
Dept. of Ecology and Evolutionary Biology,
University of California, Irvine, CA 92697
Theodore Garland, Jr.
Dept. of Biology,
University of California, Riverside, CA