The Creativity of Natural Selection? Part II: The Synthesis and Since

R.A. Fisher, indeterminism, and the fundamental theorem of natural selection

This paper is about the relationship between R.A. Fisher's fundamental theorem of natural selection (FTNS) and the two major pieces that Fisher wrote on indeterminism, "Indeterminism and Natural Selection" (1934) and Creative Aspects of Natural Law (1950). I argue that the FTNS presents a picture of natural selection that is interestingly different from what we find in these two indeterminism pieces, pace some recent work that advances the opposite conclusion. I also identify as the source of this difference both the mathematical form of the FTNS (i.e., a differential equation) and Fisher's meta-scientific commitment to advancing "general" claims about evolution.

Environmental complexity is more important than mutation in driving the evolution of latent novel traits in E. coli

Recent experiments show that adaptive Darwinian evolution in one environment can lead to the emergence of multiple new traits that provide no immediate benefit in this environment. Such latent non-adaptive traits, however, can become adaptive in future environments. We do not know whether mutation or environment-driven selection is more important for the emergence of such traits. To find out, we evolve multiple wild-type and mutator E. coli populations under two mutation rates in simple (single antibiotic) environments and in complex (multi-antibiotic) environments. We then assay the viability of evolved populations in dozens of new environments and show that all populations become viable in multiple new environments different from those they had evolved in. The number of these new environments increases with environmental complexity but not with the mutation rate. Genome sequencing demonstrates the reason: Different environments affect pleiotropic mutations differently. Our experiments show that the selection pressure provided by an environment can be more important for the evolution of novel traits than the mutational supply experienced by a wild-type and a mutator strain of E. coli.

Novel traits without immediate fitness benefit evolve frequently but we don’t know whether mutation or environment-driven selection drives this evolution. Here, using experimental evolution of E. coli populations, the authors demonstrate the importance of selection in the evolution of latent novel traits.

The Synthesis and the Two Scenarios

On the occasion of the 75th anniversary of the Society for the Study of Evolution, I look back at the so-called "evolutionary synthesis," to which many of the Society's founders and prominent founding members were committed. An important plank in the synthesis platform had to do with the importance of selection relative to mutation. Of course, there is no evolutionary change without mutation, and no adaptive evolutionary change without selection. So how could selection be more important than mutation, or vice-versa? At issue was whether adaptive evolutionary change is initiated and directed by selection, or by the appearance of new advantageous variation. Proponents of the synthesis took the position that Darwin himself had defended, namely that adaptive evolutionary change is initiated and directed by natural selection on standing variation, no new variation, no mutation, required. Natural selection is, in this sense, not just "creative," but is the creative agent of evolutionary change. In taking this extreme position, proponents of the synthesis were reacting to the equally extreme position of Mendelian-mutationists, who held that adaptive evolutionary change always commences with, and is directed by, the appearance of new advantageous mutations, and for whom mutation is the creative agent of evolutionary change. I conclude with some comments on "relative significance" issues and controversies, and respects in which the relative significance issues at the heart of the synthesis persist. This article is protected by copyright. All rights reserved.

Eclipsing the Eclipse?: A Neo-Darwinian Historiography Revisited

Julian Huxley's eclipse of Darwinism narrative has cast a long shadow over the historiography of evolutionary theory around the turn of the nineteenth century. It has done so by limiting who could be thought of as Darwinian. Peter Bowler used the eclipse to draw attention to previously understudied alternatives to Darwinism, but maintained the same flaw. In his research on the Non-Darwinian Revolution, he extended this problematic element even further back in time. This paper explores how late nineteenth-century neo-Darwinian conceptualizations of Darwinism were later utilized by several advocates and detractors of the Modern Synthesis. John Beatty has shown how this continuity hinges at least partly on the perceived importance of the creativity of natural selection. The paper provides a more thorough look at Darwin's two conflicting accounts of variation, ascribed to struggles in explaining quantitative versus qualitative characters. In doing so, it suggests that other forms of Darwinism persisted, in both the non-Darwinian revolution and eclipse periods, because of tension between contingency and creativity in Darwin's own work. This tension is traced out from Darwin's conceptions of variation into the work of Alfred Russel Wallace, Hugo de Vries, and Thomas Henry Huxley. Based on this, the eclipse narrative is criticized for not considering the meaning of Darwinism in different geographical locations. Britain and the United States showed few signs of an eclipse. Rather, the rise of German debates about Haeckel's vision of Darwinism have been mistaken for a universal decline in support.

Heritable changeability: Epimutation and the legacy of negative definition in epigenetic concepts

Epigenetic concepts are fundamentally shaped by a legacy of negative definition, often understood by what they are not. Yet the function and implication of negative definition for scientific discourse has thus far received scant attention. Using the term epimutation as exemplar, we analyze the paradoxical like-but-unlike structure of a term that must simultaneously connect with but depart from genetic concepts. We assess the historical forces structuring the use of epimutation and like terms such as paramutation. This analysis highlights the positive characteristics defining epimutation: the regularity, oxymoronic temporality, and materiality of stable processes. Integrating historical work, ethnographic observation, and insights from philosophical practice-oriented conceptual analysis, we detail the distinctive epistemic goals the epimutation concept fulfils in medicine, plant biology and toxicology. Epimutation and allied epigenetic terms have succeeded by being mutation-like and recognizable, yet have failed to consolidate for exactly the same reason: they are tied simultaneously by likeness and opposition to nouns that describe things that are assumed to persist unchanged over space and time. Moreover, negative definition casts the genetic-epigenetic relationship as an either/or binary, overshadowing continuities and connections. This analysis is intended to assist practitioners and observers of genetics and epigenetics in recognizing and moving beyond the conceptual legacies of negative definition.

Population and organismal perspectives on trait origins

Some biologists and philosophers of biology claim selection can "create" novel traits. Others claim creativity is to be found only in development. I here endorse the former claim, but take seriously and address the concerns that underlie the latter. My discussion of these issues is informed by recent work that champions the "return of the organism" to mainstream evolutionary biology, and I suggest how population and organismal perspectives on trait origins can be reconciled.