Plant Classification for Ecological Purposes: is there a Role for Genome Size?

Annals of Botany 82 (Supplement A): 117-120, 1998 Article No. bo980723 Plant Classification for Ecological Purposes: is there a Role for Genome Size? J. P. GRIME Unit of Comparative Plant Ecology, Department of Animal and Plant Sciences, The University of Sheffield, Sheffield S10 2TN, UK Received: 19 November 1997 Returned for revision: 26 March 1998 Accepted: 17 June 1998 Key words: Genome size, nuclear DNA amount, plant functional types, global change. INTRODUCTION The relatively new science of ecology depends upon a much older one-taxonomy. Ecological investigations are of doubtful value if they refer to plants or animals which have not been accurately identified. Inadequate taxonomy undermines the usefulness of ecological studies mainly because it deprives other scientists of the opportunity to conduct independent tests of published results. Although taxonomy and ecology are closely linked it is necessary to recognize that the two activities have very different objectives. Taxonomy attaches particular importance to genetic and evolutionary relationships whereas ecology seeks to recognize affinities between organisms that perform similar functions or exhibit parallel responses in contemporary ecosystems but may have quite different evolutionary origins. Over recent years there has been growing recognition (Southwood, 1977; Pugh, 1980; Grime, 1988; Smith, Shughart and Woodward, 1997) that, for ecological purposes, we require an alternative system to complement that already provided by classical taxonomy. In this new system organisms will be classified according to their functional characteristics and strong emphasis will be placed upon those features which are most consistently correlated with success in particular types of habitat and failure in others. Already, various schemes have been put forward and functional classifications have been applied to organisms as disparate as seaweeds, phytoplankton, butterflies and fungi. Perhaps the most surprising development from this research activity has been the extent to which similar criteria have been found to be useful predictors of ecology in a wide range of animals and plants. Arguments between various protagonists seem destined to continue for several more years before the objective-a general functional classification of all organisms-can be E-mail 0305-7364/98/OA0117 + 04 $30.00/0 achieved. However, there are pressing reasons why we should proceed as rapidly as possible to such a unified system. Chief among these is the need to develop sound predictive models of the consequences of regional and global impacts of man's activities on the abundance of individual populations of animals and plants and on the functioning and sustainability of ecosystems. At present one of the most severe impediments in formulating these computer models is that we do not have a coherent basis upon which to characterize the functional biology of the organisms which inhabit even our commonest ecosystems. In seeking opportunities to contribute to the development of an ecological classification of organisms, two considerations are of paramount importance. First, it is essential that we should recognize morphological, physiological or biochemical traits or sets of traits that are reliable predictors of ecological responses. Second, it is necessary to establish large databases documenting patterns of variation in the selected traits across taxa and throughout the world. The aim of this essay is to comment briefly on the potential use of determinations of nuclear DNA amount in vascular plants in current attempts to predict the responses of vegetation to changes in landuse and climate. In order to pursue this objective, it is necessary to consider the measures required to establish the link between variation in genome size and ecological patterns and processes. It is also essential to identify the kinds of data collection and the form of database which would permit testable predictions of vegetation responses to climate and landuse at various geographical scales. First, however, it may be useful to ask a naive question 'Why has it taken so long to understand the ecological implications of variation in genome size?' THE LONG QUEST In common with several other subjects which matured late on the scientific scene, ecology has drawn heavily upon older neighbouring disciplines when formulating its ob© 1998 Annals of Botany Company Genome size is a strong candidate for inclusion in the list of traits needed to devise a functional classification of plants. To facilitate modelling and prediction of vegetation responses to regional and global changes in landuse and climate, a distinctively ecological agenda is recommended for future research on inter-specific variation in nuclear DNA © 1998 Annals of Botany Company amount. 118 Grime-PlantClassification: a Role for Genome Size? but, with rare exceptions, these studies refer exclusively to the capture and allocation of carbon, energy, mineral nutrients and water, or the tolerance of tissues to environmental stresses. To a remarkable extent reference is omitted to the construction of plant tissue by the processes of cell division and cell expansion. In our present state of knowledge of this essential component of growth it is not possible to assess its importance in plant ecology. We do know, however, that meristematic activity is sensitive to environmental controls (Kinsman et al., 1996) and there must be a strong suspicion that interspecific differences in DNA amount, cell size and length of the cell cycle (the three attributes are inextricably linked-see, for example, Olmo, 1983; Bennett, 1987) have major consequences for growth responses in plants and in cold-blooded animals and inevitably therefore will have been subject to climatic selection. Several investigations have been reported in which strong correlations have been detected between genome size, climatic variation and plant distribution (Bennett, 1976; Levin and Funderburg, 1979; Wakamiya et al., 1993), between genome size and the timing of leaf growth (Grime, Shacklock and Band, 1985), between genome size and responsiveness to year to year variation in climate (Grime et al., 1994), and in temperate grassland species a relationship has been established between genome size and frost sensitivity (MacGillivray and Grime, 1995). There is also some evidence (Thompson, 1990) that variation in genome size coincides with differences in seed dormancy and germination behaviour. In view of the low penetration of cell biologists and physiologists into problems associated with variation in genome size the time seems ripe to set a distinctively ecological agenda in this field of research. The concluding section of this paper explores the measures required for such a development. AN ECOLOGICAL AGENDA FOR RESEARCH ON GENOME SIZE In seeking a model for the procedures necessary to develop and validate an approach to utilizing genome size as an ecological (...truncated)