Combining words into new formulas is an all too easy exercise. But in our case it could turn into a dangerous trick, if evolutionary developmental biology (evo-devo) does not prove to be a fruitful new adventure in science. The question is increasingly acute, as a rapidly rising number of researchers are lured by the new flag, more and more resources are put into experimental and theoretical efforts under this banner, and evo-devo is finally getting public acknowledgment in the form of dedicated university chairs, specialised journals, workshops, and the launch of new professional societies.

Over the past few years, the nature, or the identity, of evo-devo has been passionately debated. However, it would be unwise to attempt to crystallise this discipline's content in a brief formula. Taking a historical perspective, hardly any facet of contemporary science would recognise itself in what in the past would have seemed an adequate definition of a research field under the same name. Scientific disciplines change along the years. In biological parlance, one could say that scientific disciplines develop, or that they evolve.

In the past decade or so, there has been a significant increase in the available data on the developmental mechanisms underlying the process of segmentation in a wide range of arthropod taxa. This large body of data makes it possible to attempt, albeit cautiously, a comparative analysis of the various aspects of the segmentation process, and to try to find which of its features and components may have been present in the arthropod common ancestor. A recent review (Peel et al. 2005) covers much of what is known about the diversity of segmentation processes in arthropods, although even at the time of this writing, less than a year later, there is already a substantial amount of newly published data not covered therein. My aim in this chapter is not to repeat the review and synthesis presented in Peel et al. (2005), but to build on it, adding the most recent data, and expand the discussion into the more speculative domain of evolutionary reconstructions. The reader is encouraged to refer to that review for more details of the currently available data and for a more complete bibliography.

When addressing a large-scale evolutionary question, such as that suggested in the title of this chapter, it is important to define the boundaries of the problem discussed. In this review, I will focus only on the mechanisms of trunk segmentation, ignoring the differentiation and segmentation of the head region, and the posterior unsegmented region.

THE INVASION OF LAND

Land plants evolved from aquatic algal ancestors. The algae are a polyphyletic group from which the transition to land, and acquisition of developmental features associated with land plants, have occurred many times (Lewis and McCourt 2004). Recent phylogenetic evidence points to the charophyte algal lineage as the sister group to the land plants (Figure 16.1). Developmental features shared by charophytes and land plants are cell cleavage by phragmoplasts, plasmodesmatal connections between cells, and a placental link between haploid and diploid phases of growth (Marchant and Pickett-Heaps 1973). These and other features of derived charophytes, in particular growth from an apical cell in the gametophyte (Graham 1996, Graham et al. 2000), suggest that many of the cellular characteristics required for the development of land plants may have evolved within their common stem group.

FROM HAPLOID TO DIPLOID

The major character that distinguishes land plants from charophyte algae is the development of a diploid embryo. In charophytes, the majority of the life cycle is represented by the haploid gametophyte and only the unicellular zygote is diploid, undergoing meiosis immediately after formation. Embryo development represents a major growth transition in that meiotic division of the zygote is delayed and diploid sporophytic cells divide by mitosis, giving rise to a multicellular body. Although in the earliest land plants the gametophyte generation remained dominant, this transition in growth pattern led to a dramatic change in life history such that the sporophyte generation gradually became the dominant form (Graham 1985, Kenrick 1994, Graham and Wilcox 2000).

Are there clades whose particular origin or evolutionary history are more adequately explained when considering the possibilities offered by changes at the level of developmental processes, instead of thinking in terms of the unceasing interplay between gene mutation and natural selection? Yes, this is exactly the field where evo-devo offers its best performances. These are stories rooted deep in time, where one must also consider the possibility that in due course even the ‘rules of the game’, such as the role of Hox genes, or more generally the genotype–phenotype relationships, have evolved along with their products. Or, they may be stories where adaptive explanations are unsatisfactory, and the course of evolution can appear to be driven more by the nature of variation that is produced at each generation than by adaptive necessities.

Jaume Baguñà, Pere Martinez, Jordi Paps and Marta Riutort (Chapter 12) address the problem of early bilaterian evolution. Their work is based on the most recent molecular phylogenies and on new data on Hox/ParaHox and microRNA sets that identify acoelomorphs as the earliest branching extant bilaterians. Evidence for axial homologies in gene expression between cnidarians and bilaterians and the evidence that cnidarians were at their origin bilaterally symmetric all point to an older last common ancestor for bilaterians. Thus, what under different phylogenetic hypotheses appear to be a number of phylogenetically coincident character changes (the complex Urbilateria hypothesis) turn into a series of nodes connected by stem ancestors along which new characters were progressively acquired.

Spiral cleavage is a characteristic feature of several protostomian taxa, sometimes united as Spiralia (Dohle 1996), but its presence in a number of these groups has been debated. This could be the result of too vague definitions, so I will emphasise here the presence of both a spiral pattern with shifting direction of the spindles in the early cleavages and a cell lineage including prototroch cells (trochoblasts) differentiating from cells along the border between first and second micromere quartet. This automatically excludes the non-ciliated groups, but their cleavage types could be discussed in the light of the conclusions reached here.

The cleavage pattern defines two regions of the larvae: the episphere, consisting of cells from the first micromere quartet, including the primary and accessory trochoblasts, and the hyposphere ‘below’ the prototroch (Figures 21.1 and 21.2). The origin of different parts of the central nervous systems from these two regions has been documented sporadically in a number of older papers on embryology of various species, and some more recent studies provide information obtained by modern methods including cell labelling. The literature on cell lineage up to about 2004 has been summarised earlier (Nielsen 2004, 2005a). Here I will try to update the information and to incorporate new information obtained from studies of Hox genes (see also Nielsen 2005b), with special emphasis on the origin of the nervous system.

What is evo-devo? Undoubtedly this is a shorthand for evolutionary developmental biology. There, however, agreement stops. Evo-devo has been regarded as either a new discipline within evolutionary biology or simply a new perspective upon it, a lively interdisciplinary field of studies, or even necessary complement to the standard (neo-Darwinian) theory of evolution, which is an obligate step towards an expanded New Synthesis. Whatever the exact nature of evo-devo, its core is a view of the process of evolution in which evolutionary change is the transformation of (developmental) processes rather than (genetic or phenotypic) patterns. Thus our original question could be more profitably rephrased as: What is evo-devo for? This section contributes many-faceted insights into the identity and scope of evo-devo.

According to Gerd Müller (Chapter 1), evo-devo is a discipline in its own right, because it asks a specific set of questions, solves biological problems that could not be solved by other approaches, and affects our understanding of evolutionary theory. After a short reflection on evo-devo history, the chapter examines in detail a set of evo-devo big questions. All these have at their core two interrelated components, namely how evolution affects development, and how the properties of developmental systems affect the course of evolution. Finally the author considers current evo-devo research programs, and discusses the impact of evo-devo on the theory of evolution.

EVO-DEVO'S IDENTITY

Evo-devo studies the evolution of development, and how changes in development influence phenotypic evolutionary change. The evolution of novelties and body plans are considered as the most distinctive research areas of evo-devo (Wagner 2000, 2001, Wagner et al. 2000, Müller and Newman 2005). Nevertheless, there seems to be little consensus about evo-devo's disciplinary identity. It has been regarded as a branch of developmental biology, part of evolutionary biology, a revision of evolutionary theory or an independent new synthetic discipline (Gilbert et al. 1996, Arthur 2000, 2002, 2004a, b, Hall 2000, Raff 2000, Wagner 2000, Robert et al. 2001, Gould 2002, Wilkins 2002, Baguñá and Garcia-Fernàndez 2003, Gilbert 2003, Kutschera and Niklas 2004, Amundson 2005, Müller and Newman 2005). Similarly, there has been skepticism about evo-devo's promise in both the literature (Wagner 2000, 2001, Richardson 2003, Wagner and Larsson 2003, Coyne 2005) and at meetings such as the one in 2006 in Venice, at which the present book was conceived.

Although various factors are at play, I think that current skepticism partly results from a failure to articulate evo-devo's conceptual foundation properly. This issue comes into focus when it is observed that the papers outlining evo-devo's research agenda almost exclusively link the promise of evo-devo to discovering general concepts and rules. Arthur (2002: 757), for example, expresses concern when he writes that we are currently in a situation ‘where it almost seems that anything goes, that is, any developmental gene, its expression pattern and the resultant ontogenetic trajectory can evolve in any way.

SETTING THE PROBLEM

The emergence of dramatic morphological differences (disparity) and the ensuing bewildering increase in the number of species (diversity) documented in the fossil record at key stages of animal and plant evolution have defied, and still defy, the explanatory powers of Darwin's theory of evolution by natural selection. Among the best examples that have captured the imagination of the layman and the interest of scores of scientists for 150 years are the origins of land plants from aquatic green plants, of flowering plants from seed plants, of chordates from non-chordates and of tetrapod vertebrates from non-tetrapods; and the conquest of the land by amphibians; the emergence of endotherms from ectotherm animals; the recurrent invention of flight (e.g. in arthropods, birds and mammals) from non-flying ancestors; and the origin of aquatic mammals from four-legged terrestrial ancestors.

Key morphological transitions pose a basic difficulty: reconstruction of ancestral traits of derived clades is problematic because of a lack of transitional forms in the fossil record and obscure homologies between ‘ancestral’ and derived groups. Lack of transitional forms, in other words gaps in the fossil record, brought into question one of the basic tenets of Darwin's theory, namely gradualism, as Darwin himself acknowledged. Since Darwin, however, and especially in the past 50 years, numerous examples that may reflect transitional stages between major groups of organisms have accumulated.

It has been well established that considerable differences exist in the developmental pattern among animal taxa, for instance with respect to how blastomeres perform their early cleavages, how they acquire different fates or how symmetry is formed (Gilbert and Raunio 1997). Even among relatively closely related species, for instance within sea urchins or tunicates, impressive differences can be found in the pattern of development (Jeffery et al. 1999, Raff 1999).

Nematodes appear to be excellent candidates for a comparative study of early embryogenesis (Schierenberg 2005a). The phylum Nematoda is very old, its origin dating back to the Cambrian (Douzery et al. 2004), and has many different species (estimates range from tens of thousands to several millions); eggs can develop outside the mother from the first cleavage onward, they are transparent (although to a variable degree), the freshly hatched juveniles appear to have essentially invariant species-specific cell numbers of around 600 cells (for those species tested so far), many strains can be cultured in the laboratory on simple agar plates, and, last but not least, one of them, Caenorhabditis elegans, has become one of the best-studied model systems.

In this chapter, selected aspects of the early embryogenesis of five representatives from different branches of the phylogenetic tree are compared with C. elegans and the impact of the observed differences for evolutionary considerations are discussed.

Here is a question of the utmost importance for our understanding of what has been called the ‘big picture’ of evolution (Simpson 1944, 1953): are the divergences that lead ultimately to high-level sister groups, such as those that would typically be labelled as orders, classes and phyla, qualitatively or quantitatively different from those that lead to low-level sister groups, such as races, species and genera? In other words, is mega-evolution more than just accumulated micro/macro-evolution, or alternatively is evolution effectively ‘scale-independent’ (Leroi 2000)?

This question can be approached in three ways. We can choose to compare the magnitude of changes involved in high- and low-level divergences, the type of changes, or the timing (in development) of changes. Here, I argue that previous work on the first of these has been unproductive and has generated more heat than light; but that the second and third offer better prospects for shedding light on this important issue. However, in an unusual strategy, I also play devil's advocate with my own argument at the end of the chapter. This helps to take us in an interesting, final (for now) direction.

Because the designation of high-level sister groups as, for example, orders or classes, is a subjective rather than an objective process, I will, wherever possible, use specific examples rather than general levels of taxon.

Two important events marked the year 2006 in the still short history of evolutionary developmental biology. The first European Workshop on Evolutionary Developmental Biology, held in Venice on 5–6 May 2006, offered some 30 researchers from most of the European teams active in this field a timely perspective on key issues in the discipline, and opened a lively discussion on where to move next, in terms of problems, model organisms, and levels of investigation. The second event was the Founding Congress of the European Society of Evolutionary Developmental Biology, held in Prague on 16–19 August 2006, which was attended by more than 300 biologists from all over the world.

This book is based on a selection from the papers contributed to the Venice workshop, plus five additional essays expressly written for this work.

The Venice workshop was generously sponsored by the Istituto Veneto di Scienze Lettere ed Arti and hosted in the wonderful Palazzo Cavalli-Franchetti. We are very grateful to Leopoldo Mazzarolli, the President of the Istituto, for sympathetically offering this academy's spaces for the first evo-devo event at European scale; our sincere thanks are also extended to Alessandro Franchini, Antonio Metrangolo, and to the whole technical staff of the Istituto for steadily helping in the organization of the meeting.

The book has benefited from the enormous help provided by numerous colleagues in reviewing more or less advanced drafts of the chapters.

The morphological gap between Euarthropoda (the crown group that contains all extant arthropods) and living arthropod-like animals such as onychophorans, tardigrades and pentastomids is bridged by a number of fossils known primarily from rocks some 520 to 490 million years old (e.g. Fuxianhuia, Chengjiangocaris, Shankouia, see Figure 15.1; cf. Waloszek et al. 2005). These centimetre-scale fossil animals illuminate critical steps in early arthropod evolution (particularly head and limb development) but provide a limited amount of developmental information because of a lack of early ontogenetic stages. Small individuals that might represent pre-mature stages are scarce or absent, and the degree of allometry among the available individuals is generally modest. A limited number of early arthropod taxa do show more substantial ontogenetic information (Waloszek and Maas 2005). This chapter reviews the morphological development of early arthropods from two perspectives. The first is that provided by ontogenetic series based on the well-preserved biomineralised exoskeletons of trilobites, the best represented arthropod taxon in Palaeozoic rocks, but one whose development is seldom considered in broader comparative context. The second is that provided by ‘Orsten’-type preserved faunas, in which the entire cuticle of numerous post-embryonic specimens of various species, mainly representatives of the crustacean evolutionary lineage (stem derivatives and Labrophora with phosphatocopines and members of the Eucrustacea; see Maas and Waloszek 2005) was replaced with spectacular fidelity by calcium phosphate in the absence of any compaction (Müller 1985).

The morphology of the appendages of the arthropods has been adapted to a large number of life styles that is virtually unparalleled in any other organ in the Metazoa. Different appendage types exist e.g. for walking, swimming, jumping, prey-capture, chewing, biting, mating, egg-laying, breathing in air, fresh water and salt water, and sensory perception (see Figure 20.1 for examples). Very specialised appendage types exist for specialised modes of life: for example, the spinnerets in spiders, brush legs for the distribution of pheromones (e.g. some moths) or stings for defence (e.g. bees and wasps). In many cases, appendages from a single segment or from several segments unite and form an entirely new structure capable of tapping into new resources, e.g. the labium of insects, formed by the fusion of an appendage pair, or the proboscis of ticks, mosquitoes and flies, all of which are composed of the appendages of at least two head segments.

A number of different appendage types can be present on a single individual. The number of different appendage types and their specific morphology depend on the species' life style, but in most cases at least three different types are present: appendages for sensory perception, feeding and locomotion (Figure 20.1).

The appendages of the arthropods thus have been a prime target of adaptive evolution. They are unparalleled in their sheer number of novel forms and functions. Therefore, they are an excellent model for the study of the principles of adaptive evolution and morphological change and innovation.