Evolutionary developmental biologists categorize many different kinds of things, from ontogenetic stages to modules of gene activity. The process of categorization—the establishment of “kinds”—is an implicit part of describing the natural world in consistent, useful ways, and has an essentially practical rather than philosophical basis. Kinds commonly serve one of three purposes: they may function (1) as practical tools for communication; (2) to support prediction and generalization; or (3) as a basis for theoretical discussions. Beyond the minimal requirement that classifications (...) reflect biological reality, what sorts of kinds or classification will be useful in advancing a research program depends on the epistemological context. Thus, the important meaning of “natural” in “natural kinds” is not “natural with respect to nature,” but “natural with respect to the question.” This conclusion arises from the recognition that the proper role of concepts (e.g. natural kind, module, homology, model) is not to answer biological questions, but rather to help frame them. From a scientist’s perspective, arguing about the wording (or existence) of a single definition of “natural” or “kind” is beside the point: we get more work done by letting the question at hand determine what kinds of kinds are natural, on the basis of their ability to help answer it. We should be content to let “natural kinds” remain vague, multivalent, and–therefore broadly useful. For a philosopher like Hacking, the diverse, disparate, and ultimately incommensurable uses of the term “natural kind” have diluted its value so far that it loses all meaning. For practicing scientists, however, defining useful kinds in the context of particular questions and systems remains a productive epistemological strategy. (shrink)

The traditional practice of establishing morphological types and investigating morphological organization has found new support from evolutionary developmental biology (evo-devo), especially with respect to the notion of body plans. Despite recurring claims that typology is at odds with evolutionary thinking, evo-devo offers mechanistic explanations of the evolutionary origin, transformation, and evolvability of morphological organization. In parallel, philosophers have developed non-essentialist conceptions of natural kinds that permit kinds to exhibit variation and undergo change. This not only facilitates a construal of species (...) and higher taxa as natural kinds, but also broadens our perspective on the diversity of kinds found in biology. There are many different natural kinds relevant to the investigative and explanatory aims of evo-devo, including homologues and developmental modules. (shrink)

Selectionist evolutionary theory has often been faulted for not making novel predictions that are surprising, risky, and correct. I argue that it in fact exhibits the theoretical virtue of predictive capacity in addition to two other virtues: explanatory unification and model fitting. Two case studies show the predictive capacity of selectionist evolutionary theory: parallel evolutionary change in E. coli, and the origin of eukaryotic cells through endosymbiosis.

A flurry of technological advances in molecular, cellular and developmental biology during the past decade has provided a clearer understanding of mechanisms underlying phenotypic diversification. Building upon such momentum, a recent paper tackles one of the foremost topics in evolution, that is the origin of species‐specific beak morphology in Darwin's finches.1 Previous work involving both domesticated and wild birds implicated a well‐known signaling pathway (i.e. bone morphogenetic proteins) and one population of progenitor cells in particular (i.e. cranial neural crest), as (...) primary factors for establishing beak size and shape. But these results were limited in their ability to explain fully the morphogenetic bases of patterned outgrowth. So in a quest to identify novel genes whose expression correlated with differences in beak anatomy among Darwin's finches, a DNA microarray approach was undertaken using tissues harvested from the Galápagos Islands. The results are striking and point to a protein called calmodulin, which is a mediator of cellular calcium signaling, as a key determinant of beak length. BioEssays 29: 1–6, 2007. © 2006 Wiley Periodicals, Inc. (shrink)

Studies of chimeras and in vivo development demonstrate that cell lineages are often quite variable, apparently in response to chance perturbations. This points to an apparent contradiction: although individual cells are the units of genetic information and differentiation, not all cellular events need be precise for the development of functional organisms. The social organization of ants can serve as a metaphor that helps understand the mechanisms that underlie such development. Ants suggest that continued cellular interactions and environmental conditions could specify (...) the proportion and general location of specialized units. Leaf venation is used as a concrete example of this general principle. A signal produced continuously by all cells specifies a requirement for vein differentiation. The cells that respond by differentiation then transport the signal away from the leaf; this removal acting as a feedback indicating that the requirement is being met. Because transport increases during vein differentiation, early initiation occurs in excess and vein ‘competition’ for the signal assures an acceptable outcome. Such specification would be robust since it does not depend on events in any single cell, and chance events, rather than being corrected or reversed, may be built upon in reaching an expected, collective phenotype. The absence of detailed information preceding development distinguishes this hypothesis from the common alternatives of a program or blueprint. Collective specification would have important implications for developmental plasticity and evolution. BioEssays 25:897–903, 2003. © 2003 Wiley Periodicals, Inc. (shrink)

Editors' introduction to the special issue on homology (Biology and Philosophy Vol. 22, Issue 5, 2007).

Many scientists and philosophers of science are troubled by the relative isolation of developmental from evolutionary biology. Reconciling the science of development with the science of heredity preoccupied a minority of biologists for much of the twentieth century, but these efforts were not corporately successful. Mainly in the past fifteen years, however, these previously dispersed integrating programmes have been themselves synthesized and so reinvigorated. Two of these more recent synthesizing endeavours are evolutionary developmental biology and developmental systems theory. While the (...) former is a bourgeoning and scientifically well-respected biological discipline, the same cannot be said of DST, which is virtually unknown among biologists. In this review, we provide overviews of DST and EDB, summarize their key tenets, examine how they relate to one another and to the study of epigenetics, and survey the impact that DST and EDB have had on biological theory and practice. (shrink)

In 1956, in his Principles of Embryology, Conrad Hal Waddington explained that the word “epigenetics” should be used to translate and update Wilhelm Roux’ German notion of “Entwicklungsmechanik” to qualify the studies focusing on the mechanisms of development. When Waddington mentioned it in 1956, the notion of epigenetics was not yet popular, as it would become from the 1980s. However, Waddington referred first to the notion in the late 1930s. While his late allusion clearly reveals that Waddington readily associated the (...) notion of epigenetics with the developmental process, in the contemporary uses of the notion this developmental connotation seems to have disappeared. The advent and success of molecular biology have probably contributed to focusing biologists’ attention on the “genetic” or the “non-genetic” over the “developmental”. In the present paper, I first examine the links that exist, in Waddington’s work, between the classical notion of epigenesis in embryology and those of epigenetics that Waddington proposed to connect, and even synthesize, data both from embryology and genetics. Second, I show that Waddington’s own view of epigenetics has changed over time and I analyze how these changes appear through his many representations of the relationships between genetic signals and developmental processes. (shrink)

The concept of plasticity has always been present in the history of developmental biology, both within the theory of epigenesis and within morphogenesis studies. However this tradition relies also upon a genetic conception of plasticity. Founded upon the concepts of ‘‘phenotypic plasticity’’ and ‘‘reaction norm,’’ this genetic conception focuses on the array of possible phenotypic change in relation to diversified environments. Another concept of plasticity can be found in recent publications by some developmental biologists (Gilbert, West-Eberhard). I argue that these (...) authors adopt a ‘‘broad conception of plasticity’’ that is closely related to a notion of development as something that is ongoing throughout an organism’s lifecycle, and has no clear-cut boundaries. However, I suggest that given a narrow conception of plasticity, one can define temporal boundaries for development that are linked to specific features of the morphological process, which are different from behavioral and physiological processes. (shrink)

Embryonic development and ontogeny occupy whatis often depicted as the black box betweengenes – the genotype – and the features(structures, functions, behaviors) of organisms– the phenotype; the phenotype is not merelya one-to-one readout of the genotype. Thegenes home, context, and locus of operation isthe cell. Initially, in ontogeny, that cell isthe single-celled zygote. As developmentensues, multicellular assemblages of like cells(modules) progressively organized as germlayers, embryonic fields, anlage,condensations, or blastemata, enable genes toplay their roles in development and evolution.As modules, condensations are (...) fundamentaldevelopmental and selectable units ofmorphology (morphogenetic units) that mediateinteractions between genotype and phenotype viaevolutionary developmental mechanisms. In ahierarchy of emergent processes, gene networksand gene cascades (genetic modules) link thegenotype with morphogenetic units such ascondensations, while epigenetic processes suchas embryonic inductions, tissue interactionsand functional integration, link morphogeneticunits to the phenotype. To support theseconclusions I distinguish units of heredityfrom units of transmission and discussepigenetic inheritance by tracing the historyof relationship between embryology andevolution, especially the role(s) assigned tocells or to cellular components in generatingtheories of morphological change in evolution.The concept of cells as modular morphogeneticunits is modeled and illustrated using themammalian dentary bone. (shrink)

Philosophical discussions of biological classification have failed to recognise the central role of homology in the classification of biological parts and processes. One reason for this is a misunderstanding of the relationship between judgments of homology and the core explanatory theories of biology. The textbook characterisation of homology as identity by descent is commonly regarded as a definition. I suggest instead that it is one of several attempts to explain the phenomena of homology. Twenty years ago the ‘new experimentalist’ movement (...) in philosophy of science drew attention to the fact that many experimental phenomena have a ‘life of their own’: the conviction that they are real is not dependent on the theories used to characterise and explain them. I suggest that something similar can be true of descriptive phenomena, and that many homologies are phenomena of this kind. As a result the descriptive biology of form and function has a life of its own—a degree of epistemological independence from the theories that explain form and function. I also suggest that the two major ‘homology concepts’ in contemporary biology, usually seen as two competing definitions, are in reality complementary elements of the biological explanation of homology. (shrink)

The paper analyzes the early theory building process of Thomas Hunt Morgan from the 1910s to the 1930s and the introduction of the invisible gene as a main explanatory unit of heredity. Morgan’s work marks the transition between two different styles of thought. In the early 1900s, he shifted from an embryological study of the development of the organism to a study of the mechanism of genetic inheritance and gene action. According to his contemporaries as well as to historiography, Morgan (...) separated genetics from embryology, and the gene from the whole organism. Other scholars identified an underlying embryological focus in Morgan’s work throughout his career. Our paper aims to clarify the debate by concentrating on Morgan’s theory building—characterized by his confidence in the power of experimental methods, and carefully avoiding any ontological commitment towards the gene—and on the continuity of the questions to be addressed by both embryology and genetics. (shrink)

Evolutionary psychologists tend to view the mind as a large collection of evolved, functionally specialized mechanisms, or modules. Cosmides and Tooby (1994) have presented four arguments in favor of this model of the mind: the engineering argument, the error argument, the poverty of the stimulus argument, and combinatorial explosion. Fodor (2000) has discussed each of these four arguments and rejected them all. In the present paper, we present and discuss the arguments for and against the massive modularity hypothesis. We conclude (...) that Cosmides and Tooby's arguments have considerable force and are too easily dismissed by Fodor. (shrink)

Evo-Devo exhibits a plurality of scientific “cultures” of practice and theory. When are the cultures acting—individually or collectively—in ways that actually move research forward, empirically, theoretically, and ethically? When do they become imperialistic, in the sense of excluding and subordinating other cultures? This chapter identifies six cultures – three /styles/ (mathematical modeling, mechanism, and history) and three /paradigms/ (adaptationism, structuralism, and cladism). The key assumptions standing behind, under, or within each of these cultures are explored. Characterizing the internal structure of (...) the cultures is necessary for understanding how they collaborate or compete, and how they are fragmented or integrated, in the rich interdisciplinary /trading zone/ (Galison 1997) of Evo-Devo. Evo-Devo is an important example of how science can progress through a radical plurality of perspectives and cultures. (shrink)

It has become somewhat of a platitude to call Aristotle the first epigenesist insofar as he thought form and structure emerged gradually from an unorganized, amorphous embryo. But modern biology now recognizes two senses of “epigenesis”. The first is this more familiar idea about the gradual emergence of form and structure, which is traditionally opposed to the idea of preformationism. But modern biologists also use “epigenesis” to emphasize the context-dependency of the process itself. Used in this sense development is not (...) simply the unfolding of a pre-determined sequence of changes specified in advance by the organism’s genotype. It is also sensitive to inputs from the internal and external environment, which help determine in real-time which of the many potential developmental pathways are actualized during the process. Within this paradigm developing embryos are viewed as dynamic and responsive systems that react to inputs from the internal and external environment ‘on the fly’. In this paper I argue that, while Aristotle was an epigenesist in the first sense, he would have rejected epigenesis in the more modern sense. First, Aristotle’s model of choice for a developing embryo is the automaton that executes a set of preset movements (GA 734b9-13, 741b7-15). The automatons he has in mind are not dynamic AI systems capable of modifying their behaviour on the fly in response to environmental cues but completely deterministic mechanisms whose movements are fixed by their original design. Second, given Aristotle’s views about the different kinds of causal powers there are, it looks like only intentional agents endowed with actual decision-making powers could be capable of the sort of plasticity at the core of a more dynamic epigenesis. For that kind of epigenesis requires powers for alternative outcomes, and Aristotle is explicit that such powers require rational desires (προαίρεσις) that control which of those alternatives to bring about. If I am right, then he could not have made sense of the idea of a developing embryo (as a non-intentional system) making adjustments to its phenotype on the fly in response to emerging problems and opportunities, given the conceptual resources available to him. (shrink)