Academy of Sciences of the USA, 106, 8267–8272.
Wilson, E.O. (ed.) (1988). Biodiversity. National Academies Press, Washington.
Introduction written by Philippe GRANDCOLAS and Marie-Christine MAUREL.
1
Symmetry of Shapes in Biology: from D’Arcy Thompson to Morphometrics
1.1. Introduction
Any attentive observer of the morphological diversity of the living world quickly becomes convinced of the omnipresence of its multiple symmetries. From unicellular to multicellular organisms, most organic forms present an anatomical or morphological organization that often reflects, with remarkable precision, the expression of geometric principles of symmetry. The bilateral symmetry of lepidopteran wings, the rotational symmetry of starfish and flower corollas, the spiral symmetry of nautilus shells and goat horns, and the translational symmetry of myriapod segments are all eloquent examples (Figure 1.1).
Although the harmony that emanates from the symmetry of organic forms has inspired many artists, it has also fascinated generations of biologists wondering about the regulatory principles governing the development of these forms. This is the case for D’Arcy Thompson (1860–1948), for whom the organic expression of symmetries supported his vision of the role of physical forces and mathematical principles in the processes of morphogenesis and growth. D’Arcy Thompson’s work also foreshadowed the emergence of a science of forms (Gould 1971), one facet of which is a new branch of biometrics, morphometrics, which focuses on the quantitative description of shapes and the statistical analysis of their variations. Over the past two decades, morphometrics has developed a methodological framework for the analysis of symmetry. The study of symmetry is today at the heart of several research programs as an object of study in its own right, or as a property allowing developmental or evolutionary inferences. This chapter describes the morphometric characterization of symmetry and illustrates its applications in biology.
Figure 1.1. Diversity of symmetric patterns in the living world. For a color version of this figure, see www.iste.co.uk/grandcolas/systematics.zip
COMMENTARY ON FIGURE 1.1. – a) Centipede (C. Brena); b) corals (D. Caron); c) orchid (flowerweb); d) spiral aloe (J. Cripps); e) nautilus (CC BY 3.0); f) plumeria (D. Finney); g) Ulysses butterfly (K. Wothe/Minden Pictures); h) capri (P. Robles/Minden Pictures); i) Angraecum distichum (E. hunt); j) desmidiale (W. Van Egmond); k) diatom (S. Gschmeissner); l) sea turtle (T. Shultz); m) radiolarian (CC BY 3.0); n) starfish (P. Shaffner).
1.2. D’Arcy Thompson, symmetry and morphometrics
A century has passed since the original publication of D’Arcy Thompson’s major work, On Growth and Form (1917). On the occasion of this centenary, several articles have celebrated his magnum opus, highlighting the originality and enduring influence of some of Thompson’s ideas in disciplines such as mathematical biology, physical biology, developmental biology and morphometrics (see Briscoe and Kicheva (2017), Keller (2018) and Mardia et al. (2018) for specific reviews of these various contributions).
Thompson’s main thesis is that “physical forces”, such as gravity or surface tension phenomena, play a preponderant role in the determinism of organic forms and their diversity within the living world. His structuralist conception of the diversity of forms is accompanied by a critique of the Darwinian theory of evolution, but this critique is in fact based on an erroneous interpretation of the causal context (efficient cause vs. formal cause) presiding over the emergence of forms (Medawar 1962; Gould 1971, 2002, p. 1207).
The notion of symmetry is present in the backdrop throughout the book. The successive chapters go through the different orders of magnitude of the organization of life and the physical forces that prevail at each of these scales. The skeletons of radiolarians, the spiral growth of mollusk shells and the diversity of phyllotactic modes are some of the examples illustrating the pivotal role of symmetry in the architecture of biological forms. This interest in symmetry is in line with the work of Ernst Haeckel, whose book Kunstformen der Natur (Haeckel 1899) offers bold anatomical representations emphasizing (and sometimes idealizing) the exuberance and sophistication of symmetry patterns found in nature. Thompson sees the harmony and regularity of symmetric forms as the geometric manifestation of the mathematical principles that establish a fundamental basis for his theory of forms.
This emphasis on geometry finds its clearest expression in the last and most famous chapter of the book, “On the theory of transformation, or the comparison of related forms” (Arthur 2006). Thompson proposes a method for comparing the forms between related taxa, based on the idea of (geometric) transformation from one form to another, by means of continuous deformations of varying degrees of complexity. The morphological differences (location and magnitude) are then graphically expressed by applying the same transformation to a Cartesian grid placed on the original form. In spite of his admiration for mathematics, Thompson’s approach nevertheless remains qualitative and without a formal mathematical framework for its empirical implementation. These graphical representations will, however, have a considerable conceptual impact on biologists working on the issues of shape and shape change. Several more or less elegant and operational attempts at quantitative implementation were made during the 20th century (see Medawar (1944), Sneath (1967) and Bookstein (1978), for example), up until the formulation of deformation grids using the thin-plate splines technique proposed by Bookstein (1989), which is still in use today.
Beyond the questions of symmetry, Thompson’s idea of combining geometry and biology to study shapes remains at the heart of the principles and methods of modern morphometrics (Bookstein 1991, 1996).
1.3. Isometries and symmetry groups
In this section, we propose an overview of the mathematical characterization of the notion of symmetry, as it can be applied to organic forms. The physical environment in which living organisms evolve is comparable to a three-dimensional Euclidean space. It is denoted by E3. A geometric figure is said to be symmetric if there are one or more transformations which, when applied to the figure, leave it unchanged. Symmetry is thus a property of invariance to certain types of transformations. These transformations are called isometries.
Formally, an isometry of the Euclidean space E3 is a transformation T: E3 → E3 that preserves the Euclidean metric, that is, a transformation that preserves lengths (Coxeter 1969; Rees 2000):
for all x and y points belonging to E3.
The different isometries of E3 are obtained by combining rotation and translation (x ↦ α Rx + t, where R is an orthogonal matrix of order 3 and t is a vector of
