Abstracts

Abstracts


Homology—history of a concept

Alec L. Panchen

Department of Marine Sciences, University of Newcastle upon Tyne, NE1 7RU, UK

The concept of homology is traceable to Aristotle, but Belon’s comparison in 1555 of a human skeleton with that of a bird expressed it overtly. Before the late 18th century, the dominant view of the pattern of organisms was the scala naturae—even Linnaeus with his divergent hierarchical classification did not necessarily see the resulting taxonomic pattern as a natural phenomenon. The divergent hierarchy, rather than the acceptance of phylogeny, was the necessary spur to discussion of homology and the concept of analogy. Lamarck, despite his proposal of evolution, attributed homology to his escalator naturae and analogy to convergent acquired characters. Significantly, it was the concept of serial homology that emerged at the end of the 18th century, although comparison between organisms became popular soon after, and was boosted by the famous Cuvier/Geoffroy Saint-Hilaire debate of the 1830s. The concepts of homology and analogy were well understood by the pre- (or anti-) evolutionary comparative anatomists before the general acceptance of phylogeny, and they were defined by Owen in 1843. The acceptance of evolution led to the idea that homology should be defined by common ancestry, and to the confusion between definition and explanation. The term ‘homoplasy’, introduced by Lankester in 1870, also arose from a phylogenetic explanation of homology.

Return to contents
©1999 The Novartis Foundation


 Homoplasy, homology and the problem of ‘sameness’ in biology

David B. Wake

Museum of Vertebrate Zoology, University of California, Berkeley, CA 94720-3160, USA

The reality of evolution requires some concept of ‘sameness’. That which evolves changes its state to some degree, however minute or grand, although parts remain ‘the same’. Yet homology, our word for sameness, while universal in the sense of being necessarily true, can only ever be partial with respect to features that change. Determining what is equivalent to what among taxa, and from what something has evolved, remain real problems, but the word homology is not helpful in these problematic contexts. Hennig saw this clearly when he coined new terms with technical meanings for phylogenetic studies. Analysis in phylogenetic systematics remains contentious and relatively subjective, especially as new information accumulates or as one changes one's mind about characters. This pragmatic decision making should not be called homology assessment. Homology as a concept anticipated evolution. Homology dates to pre-evolutionary times and represents late 18th and early 19th century idealism. Our attempts to recycle words in science leads to difficulty, and we should eschew giving precise modern definitions to terms that originally arose in entirely different contexts. Rather than continue to refine our homology concept we should focus on issues that have high relevance to modern evolutionary biology, in particular homoplasy—derived similarity—whose biological bases require elucidation.

Return to contents
©1999 The Novartis Foundation


Homology among divergent Paleozoic tetrapod clades

Robert L. Carroll

Redpath Museum, McGill University, 859 Sherbrooke St. West, Montreal, Canada H3A 2K6

A stringent definition of homology is necessary to establish phylogenetic relationships among Paleozoic amphibians. Many derived characters exhibited by divergent clades of Carboniferous lepospondyls resemble those achieved convergently among Cenozoic squamates that have elongate bodies and reduced limbs, and by lineages of modern amphibians that have undergone miniaturization. Incongruent character distribution, poorly resolved cladograms and functionally improbable character transformations determined by phylogenetic analysis suggest that convergence was also common among Paleozoic amphibians with a skull length under 3#60cm, including lepospondyls, early amniotes and the putative ancestors of modern amphibians. For this reason, it is injudicious to equate apparent synapomorphy (perceived common presence of a particular derived character in two putative sister-taxa) with strict homology of phylogenetic origin. Identification of homology by the similarity of structure, anatomical position and pattern of development is insufficient to establish the synapomorphy of bone and limb loss or precocial ossification of vertebral centra, which are common among small Paleozoic amphibians. The only way in which synapomorphies can be established definitively is through the discovery and recognition of the trait in question in basal members of each of the clades under study, and in their immediate common ancestors.

Return to contents
©1999 The Novartis Foundation


Generation, integration and autonomy: three steps in the evolution of homology

Gerd B. Müller and Stuart A. Newman*

Department of Anatomy, University of Vienna, Währingerstrasse 13, A-1090 Wien, Konrad Lorenz Institute for Evolution and Cognition Research, Adolf-Lorenz-Gasse 2, A-3422 Altenberg, Austria, *and Department of Cell Biology and Anatomy, New York Medical College, Valhalla, New York 10595, USA

The homology concept harbours implicit assumptions about the evolution of morphological organization. Homologues are natural units in the construction of organismal body plans. Their origin and maintenance should represent a key element of a comprehensive theory of morphological evolution. Therefore, it is necessary to understand the causation of homology and to investigate the mechanisms underlying its origination. The study of this issue cannot be limited to the molecular level, because there appears to exist no strict correspondence between genetic and morphological evolution. It is argued that the establishment of homology follows three distinct (if overlapping) steps: (a) the generation of morphological building elements; (b) the integration of new elements into a body plan; and (c) the autonomization of integrated construction units as lineage-specific homologues of phenotypic evolution. In contrast with traditional views, it is proposed that the mechanistic basis for steps (a) and (b) is largely epigenetic, i.e. a consequence of the inherent propensities of developmental systems under changing conditions. Step (c) transcends the proximate mechanisms underlying the establishment of homologues and makes them independent attractors of morphological organization at the phenotypic level.

Return to contents
©1999 The Novartis Foundation


On the homology of structures and Hox genes: the vertebral column

Frietson Galis

Institute for Evolutionary and Ecological Sciences, University of Leiden, PO Box 9516, 230ORA Leiden, The Netherlands

Research on expression patterns of Hox genes has revealed a surprisingly high conservation among vertebrates. In agreement with this conservation, a correlation has been found between the anterior limits of expression areas of certain Hox genes and the borders between morphological regions of the vertebral axis. These similarities are striking and important, but also counterintuitive, unless there are strong selection pressures to protect this conservatism. It is important to identify the selective forces that maintain these conservative networks. These selective forces can be due to pleiotropy or to internal selection. Discussed are the selective factors that are involved in the evolutionary constraint on the number of cervical vertebral numbers in mammals. Factors involved are due to internal selection and involve susceptibility to cancer, stillbirths and neuronal problems. It is intriguing how similar genetic networks can lead to fundamentally different animals. Clearly the same genes are used for different purposes. It is therefore important to try find these differences. The search for homology between organisms, and the enthusiasm about similarities that come with it, at times impedes the discovery of such differences. I have searched the literature for differences within vertebrates in the functioning and expression patterns of Hox genes during the development of the vertebral axis. The ensuing implications for homology of structures and genes are discussed. The vertebral column is a promising model system for the evaluation of the relationship between homologous Hox genes and homologous structures because of the large conservation of Hox gene expression patterns along the anterior axis. However, extensive remodelling of the vertebrate column indicates that important changes in the genetic basis must have taken place. A survey of the literature indicates that the correlation between Hox gene expression areas and vertebral regions is not such that one can predict the borders between vertebral regions on the basis of Hox gene expression patterns. The involvement of Hox genes in the development of identity of vertebrae is complex and the problems regarding the value of gene expression patterns for the determination of anatomical homology are discussed.

Return to contents
©1999 The Novartis Foundation


Developmental basis of limb homology in urodeles: heterochronic evidence from the primitive hynobiid family

J. R. Hinchliffe and E. I. Vorobyeva*

Institute of Biological Sciences, University of Wales, Aberystwyth, Wales, SY23 3DA, UK, and *Institute of Ecology and Evolution, 33 Leninsky Prospect, Moscow 117071, Russia

The vertebrate limb is a classic example of homology, long assumed to be underpinned by a developmental ‘bauplan’ of the type proposed in the Shubin/Alberch branching and segmenting model. In the anuran/amniote pattern skeletogenesis proceeds in a proximodistal direction with digits forming from the posterior to the anterior. But in free-living larvae of ‘advanced’ urodeles, the pattern of skeletogenesis is distinctly different with digits 1 and 2 and the basal commune developing early, in an anterior/distal position. This different pattern is cited as evidence for a diphyletic theory of tetrapod evolution. Reassessing this problem, we analysed the pattern of early skeletogenesis of three genera (Salamandrella, Ranodon, Onychodactylus) of the ‘basal’ family of Hynobiids, using immunofluorescence to localize chondroitin-6-SO4 in Salamandrella. Here the developmental sequence was more proximodistal (intermedium preceding basal commune; early formation of the digital arch). This pattern, also found in direct developing urodeles such as Bolitoglossa subpalmata, resembled that in anurans/amniotes. Uniquely amongst tetrapods, urodeles use their developing limbs for locomotion. We attribute the unusual pattern in ‘advanced’ urodeles to adaptive modification of the developing limb. Differences in the pattern between ‘basic’ and ‘advanced’ urodeles and between urodeles and anuran/amniotes are interpreted as heterochronic within an overall single tetrapod developmental bauplan.

Return to contents
©1999 The Novartis Foundation


 Larval homologies and radical evolutionary changes in early development

Rudolf A. Raff

Indiana Molecular Biology Institute, and Department of Biology, Indiana University, Bloomington, IN 47405, USA

Larval forms are highly conserved in evolution, and phylogeneticists have used shared larval features to link disparate phyla. Despite long-term conservation, early development has in some cases evolved radically. Analysis of evolutionary change depends on identification of homologues, and this concept of descent with modification applies to embryo cells and territories as well. Difficulties arise because evolutionary changes in development can obscure homologies. Even more difficult, threshold effects can yield changes in process whereby apparently homologous features can arise from new precursors or pathways. We have observed phenomena of this type in closely related sea urchins that differ in developmental mode. A species developing via a complex feeding larva and its congener, which develops directly have different embryonic cell lineages and divergent patterns of early development, but converge on the adult sea urchin body plan. Despite differences in embryonic developmental pathways, conserved gene expression territories are evident, as are territories whose homologies are in doubt. The highly derived development of the direct developer evidently arises from an interplay of novel organization of the egg, loss of expression of regulatory genes involved in production of feeding larval features, and changes in site and timing of expression of a number of genes.

Return to contents
©1999 The Novartis Foundation


A research programme for testing the biological homology concept

Günter P. Wagner

Department of Ecology and Evolutionary Biology, Yale University, New Haven, CT 06520-8106, USA

The classical homology concept has served as a heuristic principle for organizing the enormous wealth of information on comparative anatomical patterns across a wide range of organisms. However, the classical homology concept reaches its limit as knowledge of the evolutionary, genetic and developmental processes that underlie these anatomical patterns increases. The biological homology concept places the known anatomical patterns into a mechanistic context and asserts that character identity is based on common variational properties. In this chapter a research programme for testing the biological homology concept that involves the following steps is outlined: (1) identifying of two or more putative homologues in a clade; (2) determining the phylogenetic distribution of the putative homologues; (3) describing the intra- and interspecific variation patterns of each putative homologue; (4) describing the development of each putative homologue, and determining if modes of development and distribution of homologues are phylogenetically congruent; and (5) providing and testing a model of how differences in modes of development between putative homologues effect differences in variational tendencies. The goal is to demonstrate a link between developmental and variational differences of two homologues.

Return to contents
©1999 The Novartis Foundation


Homology and homoplasy: the retention of genetic programmes

Axel Meyer

Department of Biology, University of Konstanz, 78457 Konstanz, Germany

Homology describes the inevitable evolutionary phenomenon that the similarity of structures among different organisms is due to the commonality of their descent. This continuity of information is maintained in evolutionary lineages in terms of genes and developmental mechanisms and will retain ‘sameness’ and retard, funnel and direct evolutionary diversification. Analogous ‘sameness’ is said to be due to independent, convergent evolution, and also involves similarity of function; the latter is not a necessary condition for structures to be identified as homologous. Here, I suggest that the biological basis for these seemingly disparate kinds of ‘sameness’ in evolution may in some, or even most, instances not be all that different and may be based on the same principle—the long evolutionary retention of genes, gene interactions and developmental mechanisms. Evolution might recycle and re-recruit similar mechanisms repeatedly during its course, and it often makes do with what is already available to it rather than to newly evolve or reinvent many gene interactions and developmental mechanisms repeatedly. Apparently there is no, or only a negligible, ‘genomic cost’ or even a selective advantage to maintaining genes and developmental mechanisms for long evolutionary periods of time, even if they are not continuously used in all members along an evolutionary line. Therefore, the biological basis of both homologous traits (those that are evolutionarily always expressed) and homoplasious traits (those that are not always ‘on’, but are ‘re-awakened’ during evolution) might not be so different, and the distinction between homology and some forms of homoplasy may be somewhat artificial.

Return to contents
©1999 The Novartis Foundation


Homology in the nervous system: of characters, embryology and levels of analysis

Georg F. Striedter

Department of Psychobiology and Center for the Neurobiology of Learning and Memory, University of California at Irvine, Irvine, CA 92697, USA

The establishment of homologies is critically dependent upon the process of character identification. Valid characters must reliably appear in many individuals and be delimitable from other characters. They are not defined by any essential attributes, but rather by the formation of distinct clusters in a multidimensional morphospace. Features in two or more species can be considered possible homologues only if they are identifiable as the same character, for it would be nonsensical to homologize them as different characters. In order to confirm that a character is indeed homologous between species, one must examine its phylogenetic distribution to determine that it is unlikely to have evolved several times independently in the taxa being compared. This method of homologue identification can be applied to embryonic as well as adult characters and to characters at various levels of organization, including cell types and cellular aggregates. Difficulties arise, however, when one attempts to link the homology of adult characters to that of their embryonic precursors, or the homology of cellular aggregates to that of their constituent cell types. These efforts are misguided because different characters cannot be homologized to each other (as different characters). This perspective suggests that many neural characters may lack homologues, and therefore be truly novel, in other taxa.

Return to contents
©1999 The Novartis Foundation


Natural history and behavioural homology

Harry W. Greene

Museum of Vertebrate Zoology and Department of Integrative Biology, University of California, Berkeley, CA 94720-3101, USA

Although similarities and differences among animals have long inspired ethologists, three misconceptions haunt the literature on behavioural evolution: the absence of ‘ethofossils’ seriously hampers detection of behavioural homology; behaviour is more variable and subject to experiential modification than morphology; and behaviour is especially subject to convergence. As a backdrop to address these issues, I briefly survey parental care by amniotes and antipredator mechanisms in non-avian reptiles. Although those behaviours remain inadequately sampled taxonomically, they clearly vary at several cladistic levels; they are conservative across some major groups, innovative within subclades, and exhibit apparent homoplasy among and within groups. These behavioural examples also illustrate contextual influences on the expression of traits, as well as how behavioural context can shape other aspects of development. Enhanced understanding of behavioural evolution will follow from greater emphasis on how developmental context, including behaviour itself, shapes phenotypes; from integration of data for fossil and recent organisms; and from much denser ethological sampling among taxa. Phylogenetic analyses of behavioural similarity should in turn provide exciting insights into the evolutionary roles of behavioural shifts and constraints, as well as inform our aesthetic appreciation for the richness of nature.

Return to contents
©1999 The Novartis Foundation


Evolutionary dissociations between homologous genes and homologous structures

Gregory A. Wray

Department of Ecology and Evolution, State University of New York, Stony Brook, NY 11794-5425, USA

Phenotype is encoded in the genome in an indirect manner: each morphological structure is the product of many interacting genes, and most regulatory genes have several distinct developmental roles and phenotypic consequences. The lack of a simple and consistent relationship between homologous genes and structures has important implications for understanding correlations between evolutionary changes at different levels of biological organization. Data from a variety of organisms is beginning to provide intriguing glimpses of the complex evolutionary relationship between genotype and phenotype. Much attention has been devoted to remarkably conserved relationships between homologous genes and structures. However, there is increasing evidence that several kinds of evolutionary dissociations can evolve between genotype and phenotype, some of which are quite unexpected. The existence of these dissocations limits the degree to which it is possible make inferences about the homology of structures based solely on the expression of homologous genes.

Return to contents
©1999 The Novartis Foundation


Establishing homology criteria for regulatory gene networks: prospects and challenges

Ehab Abouheif

Department of Ecology and Evolution, State University of New York at Stony Brook, Stony Brook, NY 11794-5245, USA

One of the most remarkable discoveries to emerge from the field of developmental genetics is the observation that many regulatory genes and segments of their interactive networks (pathways) appear to have been conserved in several metazoan phyla. Determining whether these conserved regulatory networks are homologous, i.e. derived from an equivalent network in the most recent common ancestor, is critical to understanding comparisons between model system studies, and the evolution of metazoan body plans. To this end, I outline some of the evolutionary properties of regulatory networks, and propose both similarity and phylogenetic criteria that can be used to test the hypothesis that two regulatory networks are homologous. Furthermore, I propose that genetic networks can be treated as a distinct level of biological organization, and can be analysed together with other hierarchical levels, such as genes, embryonic origins and morphological structures, in a comparative framework. Examples from the literature, particularly the genetic regulatory networks involved in patterning arthropod and vertebrate limbs, are examined using the proposed criteria and hierarchical approach.

Return to contents
©1999 The Novartis Foundation


The effect of gene duplication on homology

Peter W. H. Holland

School of Animal and Microbial Sciences, The University of Reading, Whiteknights, Reading RG6 6AJ, UK

Genes related by gene duplication within an organism's evolutionary lineage are termed paralogues; genes related by speciation are orthologues. It is generally agreed that orthologous genes must be compared when using DNA sequences to reconstruct the evolutionary history of organisms. There is an important exception: information from paralogous genes can reveal the root position of a phylogenetic tree. The duplicated rDNA genes of arrow worms provide an example. Gene duplication is also relevant when comparing gene expression between taxa; for example, when trying to identify homologous roles of genes. When gene duplication occurred after lineage divergence, single orthologues no longer exist, and comparison is complicated. This is a particular problem when comparing roles of vertebrate and invertebrate genes. Amphioxus and ascidian genes can be useful in such situations, since they diverged before extensive gene duplication in the vertebrate lineage. Using Otx and Pax as examples, I show how examination of amphioxus or ascidian genes reveals patterns of gene divergence after duplication, assisting the identification of homologous gene functions. Given the problems of comparing duplicated genes between species, the time is ripe for the introduction of additional terminology to elaborate on the concepts of paralogy and orthology.

Return to contents
©1999 The Novartis Foundation