Plant evolution at the crossroads: the imprint of history, climate, and humans on the Mediterranean flora

I expect that most botanists, when they think of evolution in the world's Mediterranean-type environments, think of convergent evolution in response to the unusual climatic conditions. Mediterranean-type climates are not only distinctive, with a cool, wet winter and a hot, dry summer, they are also geographically disparate and geologically recent. This combination has led to the assembly of diverse floras on five continents with many shared features, as recognized by Schimper (1903) and others in the early works of plant geography based on the collections and observations of 18th- and 19th-century botanists. The single trait most strongly associated with Mediterranean-type vegetation is the development of sclerophylly, the hard, leathery, evergreen leaves of the dominant shrub vegetation, and the associated physiological strategies for survival and growth in the seasonal Mediterranean-type climate. Cooper (1922) in California, and later Loveless (1962) and Beadle (1966) in Australia, provided the foundation for understanding the physiological significance of the sclerophyll strategy in relation to seasonal drought and/or low nutrient availability. With this background in mind, I opened John D. Thompson's new book, Plant Evolution in the Mediterranean (PEM for short) thinking about convergence and adaptation to Mediterranean-type climates. I was in for a pleasant surprise. Thompson is a population biologist by training, with 15 yr of research experience in the Mediterranean. In his broad-ranging synthesis, he presents the Mediterranean flora poised in a particular moment against a background of geological complexity, climatic fluctuations, and profound human influences. He focuses primarily on contemporary evolutionary processes, in particular the differentiation of populations in the complex environmental and anthropogenic mosaic of the Mediterranean region. Convergence, while receiving due notice, does not play a central role, and as the book unfolds, the particularities of the Mediterranean come to the fore, set in the context of general principles of ecology and evolution. PEM exemplifies the maturation of evolutionary ecology. In its first incarnation, in the 1960s and 1970s, the synthesis of ecology with optimality theory and evolution generated the successful, though much-berated, “adaptationist” program. Cost-benefit models were widely employed to demonstrate the adaptive value of physiological and morphological traits. Studies of convergence were held up as empirical tests of these models, and research on the flora and fauna of Mediterranean-type ecosystems played a central role in this effort (Cody and Mooney, 1978). Spurred in part by Gould and Lewontin's (1979) critique, attention shifted in the 1980s to detailed studies of within-population selection on the one hand (e.g., Lechowicz, 1984), and rigorous phylogenetic approaches to test adaptive hypotheses on the other (e.g., Donoghue, 1989). The former have provided critical support for the effects of selection, combining analyses of heritability with phenotypic selection analysis. The latter led to an important shift away from the static analysis of present-day species as independent entities, to a focus on the historical context, including the frequency and direction of phenotypic change. The process of adaptation, in an historical view, must consider modern traits relative to their ancestral states. Unfortunately, “historical effects,” “phylogenetic inertia” and related ideas have too often been presented as alternatives to selective explanations for traits, as if selection operates only within modern populations and only leads to evolutionary change. The problem presented by the study of evolutionary stasis is to distinguish the roles of stabilizing selection on well-adapted trait complexes vs. the absence of genetic variation and other proposed “constraints,” and we still lack adequate methodology to meet this challenge. PEM illustrates an emerging synthesis in evolutionary ecology, viewing phenotypic differentiation against a backdrop of heterogeneous selection pressures, gene flow, historical climate change, and past selection. The book is laid out in six substantial chapters. The first is a general background to the historical geologic, climatic, and floristic history of the Mediterranean. This is followed by two chapters on the ecology and evolution of endemism, the “cornerstone of Mediterranean plant diversity.” Chapters four and five address functional and reproductive traits, respectively, and the final chapter highlights the evolution of domesticated and invasive plants. In a brief conclusion, Thompson recaps his integration of the evolution of diversity and endemism on the one hand, with population ecology and adaptation on the other, and their significance for conservation of the Mediterranean flora. In the introductory chapter, Thompson states his goals to present a “synthetic account of evolutionary ecology in which the central theme is differentiation, both within and among populations” (p. 1). Throughout the book, his emphasis on the role of spatially heterogeneous selective pressures interacting with dispersal, genetic variation, and population history form a strong underlying thesis. The wealth of examples addressing differentiation among populations is indeed impressive. Thompson captures the dynamic nature of population differentiation and the possibility that we may be glimpsing the early stages of speciation. His intimate knowledge of the flora and geography shines through, as he presents case studies from the extensive Mediterranean literature with close attention to details of natural history. “Species” are never presented as unified entities, as is often the case in comparative biology when details of intraspecific differentiation are difficult to incorporate in broader analyses. The remarkable studies of spatial variation, polymorphism, and adaptation in aromatic plants exemplify this perspective. Whether coincidence or not, John D. Thompson's emphasis on spatial heterogeneity, primarily in abiotic factors and disturbance history, closely parallels John M. Thompson's theory of geographic coevolution (Thompson and Cunningham, 2002). Both emphasize the importance of spatially contrasting selection pressures, together with dispersal, underlying the differentiation and cohesion of populations within a species. In this sense, PEM approaches a monograph in its approach to plant evolution. Thompson does not present a new theory of evolution, but his focus on the importance of spatial processes in evolution provides a valuable counterpoint to studies conducted locally within individual populations, on the one hand, and broad comparative studies on the other. It is at this scale, among populations, that the idiosyncrasies of evolution are perhaps seen most clearly. Evolutionary studies within populations often support general adaptationist thinking, by illustrating the efficacy of selection in shaping traits suited to the local environment. Similarly, comparative research may provide broad support for the match between organisms and their environments along environmental gradients. Between these two extremes, the complex dynamics of evolutionary differentiation become most apparent. Local selection is countered by metapopulation dynamics and gene flow, and adaptive processes are tightly linked to the genetic and developmental complexities underlying the phenotype. As a result, the study of population differentiation provides fertile ground for the synthesis of population ecology, genetics, and evolution, though the resulting complexity makes it hard to extract generalizations. As a regional treatment, Thompson frames his analysis in the context of the “historical triptych” of human activity, geology, and climate. Perhaps nowhere else in the world has the first of these played such an important role in shaping plant evolution, as humans have heavily modified the environment, manipulating plant populations intentionally and unintentionally, for over 10 000 years. Strong selection favoring weediness in agricultural settings and anthropogenically disturbed environments provides a compelling explanation for the geographic asymmetry of invasions; many more plants from the Mediterranean have become invasive in other regions, particularly the other Mediterranean-type environments such as California, rather than the other direction. The imprint of human activities over several millennia strongly influences spatial heterogeneity across the landscape, and clearly shapes the research program of plant population biology in the Mediterranean region. This is the first of several ways in which the Mediterranean seems distinctive, rather than illustrative of general patterns that are mirrored across the world's five Mediterranean-type ecosystems. The role of climate is also counter to initial expectations. Despite Thompson's focus on the Mediterranean region, the Mediterranean-type climate does not play a strong role in his synthesis. Of greatest importance, in keeping with the theme of the book, is the broad and fine-scale spatial heterogeneity in climate and soil arising from geography and topography. The geographic gradient of increasing aridity from west to east, wrapping around the Mediterranean Sea, is strikingly similar to California's north to south rainfall gradient that runs down both sides of the Central Valley. The range of precipitation that characterizes Mediterranean-type climates, from approximately 250 to 1000 mm/year (Dallman, 1998), has profound effects on vegetation. On geographic gradients, this range virtually spans the transition from forest to desert. At small scales, proximity to the ocean and steep topography create sharp moisture gradients due to orographic effects and rain shadows. Combined with edaphic heterogeneity, these conditions lead to a remarkable mosaic of vegetation. On Mt. Tamalpais, just north of San Francisco, grassland, chaparral, oak woodland, and redwood forest are arrayed in a fine-scale mosaic within an area of <20 km2; desert and forest biomes rarely present such dramatic variation in such a small area, and this spatial mosaic plays a strong role in shaping population differentiation. However, Thompson says little about the effect of the Mediterranean-type climate per se, the cool, wet winters and hot, dry summers sought out by vintners and vacationers alike. This stands in sharp contrast to the long tradition of research on physiological and functional convergence in Mediterranean-type environments. Other features arising from the combination of climate and vegetation, such as the overarching importance of fire as an ecological and evolutionary factor, receive only secondary attention in relation to dispersal and vegetation dynamics. The long-standing discussions of convergent evolution in Mediterranean-type environments focus on the effects of climate and fire on plant form and function. On first reading, I found Thompson's treatment frustrating, because I had expected much more attention to these phenomena. Yet, on further reflection, it may be appropriate, as much of the research on convergence has actually provided little insight into evolutionary processes. It is clear that there are important similarities in the fauna and flora of Mediterranean-type ecosystems. However, many of these similarities are evident at the community level (e.g., similarity in the spectrum of leaf size in relation to climate), and thus represent ecological convergence at best, with an uncertain evolutionary component. On the other hand, the paradox of Fisher's fundamental theorem of natural selection rears its head: those traits that are variable, and thus amenable to study at the population level, are likely under variable or weak selection. Traits that have been under strong selection, and represent shared adaptations to the overall Mediterranean-type climate, may be fixed and provide little material for contemporary studies of evolutionary dynamics. Finally, Thompson's treatment of geology outlines the critical, though idiosyncratic, setting for the evolutionary play. I found myself enmeshed in his narrative of the region's geologic history, similar to the experience of reading John McPhee's (1993) geologic history of California. Yet, if asked, I could barely summarize the salient features beyond its complexity and the strong imprint it leaves on the geography. One feature in particular that stood out was the importance of islands in all aspects of Thompson's discussion. It takes only a brief glimpse at a map of the Mediterranean to realize that islands will play a central role in any study of biogeography, endemism, and population differentiation. As in other areas, the islands of the Mediterranean have provided important natural experiments and laboratories for the study of evolution. In this sense, the region is a wonderful place for such research, but it may also set it apart from other regions with Mediterranean-type environments where islands are a minor geographic element. Having laid out his objectives to synthesize evolutionary ecology in relation to differentiation, and the specific context of the Mediterranean region, the main chapters of the book offer a far-ranging overview of plant evolutionary population biology. Thompson's command of the theoretical and empirical literature is impressive, and the citations are right up to date. The conceptual organization and broad coverage resemble a textbook style treatment, though the balance of topics clearly reflects both the author's research interests and the particularities of the Mediterranean flora. However, for readers who do not study or teach in the Mediterranean, the result may be somewhat unsatisfying. It is not clear whether certain topics receive little or no treatment because they are deemed conceptually unimportant to evolutionary biology as a whole, because Thompson has specifically set out to synthesize processes at the population level, or because there is little empirical work in the Mediterranean that exemplifies or tests the theories. And for topics that are addressed, it is equally uncertain whether the Mediterranean examples are critical studies that have advanced the field or are simply the best examples from the region. Only in a few cases, as in the excellent treatment of the evolution of heterostyly, does Thompson clearly state that Mediterranean plants have offered critical case studies, on a global scale, to test the relevant hypotheses. As a result of these uncertainties, PEM presents a bit of a conundrum. What sort of book is it? Is it a monograph, presenting a new approach or synthesis of plant evolution? Is it primarily a regional treatment, a synthesis of evolutionary history and differentiation in the Mediterranean that incidentally exemplifies key concepts of evolution? Or is it a textbook, where a reader may expect to come away with a balanced treatment of the major concepts of plant evolution, illustrated with examples from the Mediterranean? Somehow, the book is a bit of each of these, alternately giving the satisfaction of a broad and comprehensive view of evolutionary processes, but also some ambiguity of what is new and what is review. My own research background differs from Thompson's in a rather systematic fashion: I have focused my work on functional and physiological ecology of woody plants, most of them widespread and fairly common in the California chaparral, and used comparative phylogenetic approaches to ask evolutionary questions. In contrast, Thompson's research primarily addresses herbaceous plants, with a focus on population level differentiation in reproductive biology and a special interest in the biology of rare and endemic taxa. Not surprisingly, I learned a lot reading the book, in large part due to my ignorance of these areas. Chapters 2 and 3 on endemism provide a fascinating discussion of the bridge between differentiation of populations and the origin of new taxa. Thompson reviews the classification of endemics, introduced by Favarger and Contandriopoulos (1961) and Stebbins and Major (1965), in analyses of the Mediterranean and California floras, respectively. The distinction between paleo- and neo-endemics highlights the role of relictual lineages vs. recent radiations, both of which make important contributions in the Mediterranean and California. Within the neo-endemics, there are a number of groupings, including the concepts of patro- and apo-endemics, locally distributed diploid taxa with widespread polyploid descendents, or polyploid taxa derived from widespread diploid species. While these categories are especially important in the study of plant endemism, they are somewhat problematic as they rely on the idea of ancestor-descendent relationships among extant taxa. As molecular phylogenies become available for more groups, it will be useful to reevaluate patterns of endemism in the Mediterranean and other floras. In the subsequent two chapters, Thompson addresses the evolution of vegetative traits and dispersal (Chapter 4) and of reproductive traits and breeding systems (Chapter 5). A consistent theme in both cases is the inclusive view of the role of history in shaping current phenotypes and the complex between spatial differentiation and introduced a critical distinction in analyses of the Mediterranean flora between taxa present to the vs. those that have into the region the of the Mediterranean-type climate The two the former by and large and and the latter by dry and small patterns have been for California, and the the view that evergreen was likely an ancestral trait of many woody the of a Mediterranean-type climate. This view, long held by based on and floristic is also by phylogenetic analysis of ancestral states for leaf form in the California flora distinction between and which has been by a number of subsequent studies, is primarily on based on of and taxa. research and comparative to this problem has in the flora of the Mediterranean and the other Mediterranean-type environments. Despite the wealth of comparative and physiological on functional traits in Mediterranean plants, few studies have selection, or population differentiation in such and as a result Thompson has little material to work from on this As in other of the the historical between and population the of some of the and the large for genetic analysis represent that have the development of evolutionary and This is particularly of woody plants, where large and long present for and common experiments and virtually genetic studies and of in important for which have of variation in growth and functional traits (e.g., addressing variation and evolution in reproductive traits, presents a more comprehensive view, both in conceptual and in the range of empirical This reflects the strong theory in this area, including evolution of its heterostyly, is little in many of these traits, in studies of population differentiation, and the genetic of variation is in a of study The Mediterranean region is also to a number of in reproductive particularly M. and Thompson including D. and have been to the region to studies of particular questions. such as and have as systems for the study of and heterostyly, respectively. In to the current mosaic of ecological conditions and selective pressures, the variation in reproductive biology may both the of the Mediterranean-type climate, and the of this vegetation the of the Mediterranean as for a diversity of taxa the a level of genetic populations for many as the and geography of offered As have from these of populations have against a backdrop of climate and human influences. As a result, the Mediterranean may provide a glimpse of a flora and fauna in an of evolutionary as and taxa sort across the Other Mediterranean-type environments, such as California, much of the climatic history with the Mediterranean, but they were not close to the they may not have the role as the major for taxa In the final chapter, Thompson domesticated and invasive species as models for the study of plant evolution. The Mediterranean is the of the world's major important such as including and and aromatic plants such as and While these plants have been widely around the there is also an history of their from to within the Mediterranean. In for there is of independent in several areas, but from populations are dominant in the region. The long history of in the Mediterranean may also the invasive of plants from the region, as they have been for in anthropogenic for of years. the diversity of the Mediterranean, with local evolution of species to human may the low level of to the region. The of to regions, including from and and from exemplify that often represent that are to phylogenetic in a flora. PEM is an book and a broad of readers will the book, or particular of I expect it will as an important foundation for studies of plant evolution in the Mediterranean, though with the region will both and that I do not a similar book for California, I be to it as the for a The conceptual organization in plant evolution, with examples from the regional flora that or be to in combination with a selection of the the provide and other an excellent for development of new hypotheses and the selection of study readers the Mediterranean, PEM provides an excellent into the extensive literature from this region for a of in plant evolution do to the relevant of Thompson's book, both for his of many topics and for the case a of recent studies, as as to and that are not widely in the examples studies of endemism and and 1984), and and recent monograph on Mediterranean Thompson's effort to the of geology and climate with general principles of evolutionary biology is similar for other regions of the world provide a fascinating for broader synthesis, as as a foundation for evolutionary processes in the conservation of plant