Soil and Biota of Serpentine: A World View 2009 Northeastern Naturalist 16(Special Issue 5):121–130 Climate Gradients, Climate Change, and Special Edaphic Floras Susan Harrison1,*, Ellen Damschen2, and Barbara M. Going1 Abstract - Serpentine endemics and other soil-restricted taxa may be presumed to face extraordinarily high risk from climate change because their narrow edaphic niches limit their possibilities to adapt through migration. However, their distinctive life-history traits and their competitive relationships with faster-growing soil generalists may complicate this picture and produce unexpected outcomes. Here we propose a conceptual framework for how serpentine endemics will fare under climate change, together with three potential tests of its predictions. We believe climate change should be embraced by serpentine plant ecologists as a critical area for greater study. Introduction To date, surprisingly little attention has been given to how serpentine floras, and other special edaphic floras, may be affected by global climate change. This issue is important for two reasons. First, the world’s special edaphic floras are major contributors to global biodiversity. Examples include the rich floras of limestone grasslands in southern Europe, dolomite glades in the Ozarks, shale barrens in Appalachia, or serpentine outcrops in the Mediterranean, Cuba, New Caledonia, and California (Anderson et al. 1999, Kruckeberg 2005). In California, one of the world’s botanical diversity hotspots (Myers et al. 2000, Stein et al. 2000), 612 of 1742 rare plants are associated with serpentine, limestone, volcanic outcrops, vernal pools, or other special substrates (Skinner and Pavlik 1994), and plants restricted to serpentine comprise >10% of species unique to the state even though serpentine is <2% of the state’s area (Kruckeberg 1984, Safford et al. 2005). Second, their naturally fragmented distributions make special edaphic floras potentially exceptionally vulnerable. Plants confined to outcrops of special soils might be expected to have far lower chances of successful migration to suitable new sites and thus far greater risks of extinction in the face of climate change, than plants that are soil generalists. This problem is of general importance. Forecasts of biotic change over the coming decades are typically made by modeling current versus future climatic envelopes of species, and these models assume that even if a species is incapable of migration, it will at least survive in the geographic areas of overlap between its present and future climatic envelopes (Loarie et al. 2008, Schwartz et al. 2006, Thomas et al. 1Department of Environmental Science and Policy, University of California – Davis, Davis, CA 95616. 2Department of Biology, Washington University, 1 Brookings Drive, Campus Box 1137, St. Louis, MO 63130. *Corresponding author - spharrison@ ucdavis.edu. 122 Northeastern Naturalist Vol. 16, Special Issue 5 2004). Edaphic endemics are perfect examples of why this assumption is too optimistic; it underestimates the risks faced by species whose geographic distributions are constrained by other niche requirements besides climate. Nonetheless, in an informal survey at the Sixth International Serpentine Conference, only about 15 of the 80 participants agreed with the statement that they had given thought to the possible effects of climate change on serpentine floras. Of these 15, only one agreed with the statement that serpentine plants are likely to be at higher risk than other plants, while the others agreed that the risks faced by serpentine plants ought to be lower. We believe this view base on the perception that because serpentine plants are less sensitive to temperature and water availability than other plants, as we describe below. We first propose a conceptual framework for considering the possible effects of climate change on special edaphic floras. We then discuss ongoing work to test this model. As we will explain, we believe that the geographic variation in serpentine versus nonserpentine (or other special edaphic versus “normal”) plant communities is one important source of evidence that is readily available. Other sources of evidence include field manipulations of water availability, and comparisons of historical and modern vegetation data. Conceptual Model Assumption 1: Edaphic restriction is relative We begin by assuming that few species are 100% specialized on any particular substrate. For example, in the unusually well-studied Californian serpentine flora, the 200+ taxa that are considered strong “endemics” are those with roughly >85% of their known occurrences on serpentine; another several hundred taxa show lesser degrees of affinity, and have been called “indicators” (Kruckeberg 1984, Safford et al. 2005). Similarly, species that are “endemic” to limestone glades may only be restricted to glades within a limited portion of their range and species that are “characteristic” to limestone glades are not necessarily confined to glade habitats (Nelson and Ladd 1983). It follows, then, that any given species may be more or less “endemic” to a special soil depending on the ecological circumstances, including the climate and the surrounding plant community. For example, species that specialize on serpentine in some parts of their ranges, but are more generalized elsewhere, have been termed “regional endemics” (Rune 1953) or “regional indicators” (Kruckeberg 1984). The reasons for geographic variability in substrate restriction have never been examined, to our knowledge. The review of observational studies below suggests a pattern, such that species and assemblages are more soil-restricted in more climatically favorable regions. Assumption 2: Edaphic restriction is influenced by competition We next assume that competition with other plant species is one of the major reasons why species show restriction to special soils. When freed from competitors that grow faster and taller on more fertile soils, but that are less tolerant of serpentine, it has been found that serpentine specialists grow equally well (in some cases, better) on the more fertile soils (Kruckeberg 2009 S. Harrison, E. Damschen, and B.M. Going 123 1954). Similar patterns have been observed in species endemic to other soils (e.g., Anderson et al. 1999, Baskin and Baskin 1998, Kruckeberg 2005, Sharitz and McCormick 1973, Tansley 1917, Ware 1991). A widely held paradigm is that edaphic endemics are tolerant of either resource-poor or excessively cation-rich (“toxic”) soils, but are incapable of fast growth on more fertile soils, while many soil generalists (“bodenvag” species) show the opposite set of traits . If this is true, we would expect that as the climate becomes more favorable (i.e., when water or degree-day limitations on plant growth are relaxed), edaphic endemics will receive relatively modest direct benefits, because of their limited capacity for resource uptake and growth. Their major competitors, soil generalists, will benefit much more from climatic favorability, but only on fertile soils, whereas their growth will remain strongly limited by scarce nutrients or toxic elements on special soils. Assumption 3: The degree of edaphic restriction may vary with climate The degree of restriction to special soils may vary along climatic gradients. The most common pattern is for species to be more confined to serpentine in more climatically favorable regions. The general explanation that has been proposed for this pattern (Brooks 1987, Raven and Axelrod 1978, Rune 1953, Whittaker 1960) is that, as Brooks (1987) stated, “competitive pressure restricts some plants either to the edaphically harsh environment of serpentine, or to climatically harsh environments,” which corresponds to our conceptual model below. For example, a number of species are serpentine specialists in the mesic coastal mountains of California, although they are soil generalists in colder or more arid zones; examples include Pinus jeffreyi Balf. (Jeffrey Pine; Pinaceae), Calocedrus decurrens Torr. (Incense Cedar; Cupressaceae), Arctostaphylos nevadensis A. Gray (Pinemant Manzanita; Ericaceae), and Quercus vaccinifolia Kell. (Huckleberry Oak; Fagaceae). Likewise, several species from southern Californian deserts reach their northern range limits on serpentine in the mesic coastal mountains, e.g., Juniperus californica Carrière (California Juniper; Cupressaceae), Eriogonum wrightii Torr. ex Benth. (Wright’s Buckwheat; Polygonaceae), Fremontodendron californicum (Torr.) Coville (California Flannelbush; Malvaceae), and Arctostaphylos glauca Lindl. (Bigberrry Manzanita; Ericaceae) (Kruckeberg 1984, Whittaker 1960). Similar examples have been noted in eastern North America, northern and central Europe, Japan, and the neotropics (Borhidi 1991, Brooks 1987, Rune 1953). Boreal and subarctic species may reach their southern limits on serpentine (Brooks 1987, Rune 1953). Montane species often extend their elevational distributions downward on serpentine, sometimes by 1000 m or more (Borhidi 1991, Brooks 1987, Whittaker 1960). An interesting historical example comes from Sweden, where Arenaria norvegica Gunnerus (Arctic Sandwort; Caryophyllaceae) is restricted to serpentine only south of 66°N latitude. Rune (1953) explained that this small herb was outcompeted on more favorable substrates as forest invaded 124 Northeastern Naturalist Vol. 16, Special Issue 5 southern Sweden in late postglacial times. At higher latitudes, A. norvegica remains widespread on other soils, presumably because competitive pressure is lower where forest is sparse or absent. The model and its predictions We predict that more favorable (e.g., wetter) conditions will favor the net success of edaphic endemics on special soils, where the growth of their generalist (bodenvag) competitors is limited by other factors, but these conditions will reduce the net success of edaphic endemics on normal soils, because of increased competition from generalists. Less favorable (e.g., drier) conditions will have the opposite effects. These interactive effects will lead both individual species and entire assemblages to show greater edaphic specialization in more favorable climates (Fig. 1), which is the general pattern that is observed. We think this three-way interaction between climate, competition, and soils provides a testable ecological explanation for biogeographic patterns in the floras of special soils. We also propose that this model provides a basis for predictions about human-caused climate change. Our research aims to test the following hypotheses: 1. Special edaphic floras will show higher distinctiveness from the floras of nearby normal soils in more favorable climates (those with higher mean rainfall, higher minimum or lower maximum temperatures, etc.). Figure 1. Conceptual Model. Bold arrows and multiple ++ or -- signs indicate very strong effects; dashed arrows and signs in parentheses (+ or -) indicate weak or absent effects. Our hypothesized effect—that more favorable climates lead to more edaphically specialized floras—depends on the differences in the strengths of the four arrows on the left side of the figure. 2009 S. Harrison, E. Damschen, and B.M. Going 125 2. Field water manipulations will have lesser effects (additions beneficial, reductions harmful) on serpentine endemics than related non-endemics. Water manipulation effects will also depend on whether competitors are present or absent, and whether the background community is a serpentine or nonserpentine grassland. 3. The serpentine flora of the Siskiyou Mountains will show lesser effects of climatic warming than the diorite flora, when we compare vegetation data collected in 2007 with data collected by ecologist Robert Whittaker in 1949–1951. Testing the Conceptual Model 1. Observational studies of Californian serpentine and Ozark glade floras In California, serpentine is found almost entirely within the boundaries of the California Floristic Province, i.e., the zone of Mediterranean climate. Its vegetation ranges from conifer woodland and forest through chaparral (shrub) to grassland, along latitudinal and elevational moisture gradients (Alexander et al. 2006, Grace et al. 2007, Harrison et al. 2006, Kruckeberg 1984). The species richness of serpentine endemics declines sharply along California’s north-to-south gradient of decreasing precipitation, both in absolute terms and as a percentage of total species richness (Harrison et al. 2000). Interestingly, serpentine endemics also diminish to the north and south of the California Floristic Province; there are almost none in the remainder of western North America, despite an abundance of serpentine (Alexander et al. 2006). In previous work, we sampled species and >70 environmental variables at 109 serpentine sites across California (Harrison et al. 2006). The sites spanned 1200 km in latitude, 0–2750 m in elevation, 21–257 cm in average annual precipitation, and 28.9–38.6 °C in maximum July temperature. For this study, we will sample a paired non-serpentine site for every serpentine site in the previous study. Sites will be chosen by examining geologic, vegetation, and road maps and identifying an accessible non-serpentine area that is within a relatively constant distance (e.g., 0.5–5 km) from each serpentine site, and represents the typical vegetation for the area. To test Hypothesis 1 at the assemblage level, we will calculate assemblage- level dissimilarity for each pair of sites (one minus the Jaccard similarity coefficient) and regress it on mean annual rainfall and other climatic variables. We will use partial Mantel tests to ask if such effects are significant after spatial and environmental distances (e.g., based on soil variables) are taken into account. To test Hypothesis 1 at the individual species level, we will use our previous data and Safford et al. (2005) to divide the species we find into avoiders, tolerators, and weak and strong endemics. For each species, we will use combined data from the present and previous study to calculate the proportion of its total abundance that is on serpentine. We will regress this measure on mean annual rainfall and other climatic variables, across all sites where the species was found. We want to generalize beyond the flora of serpentine, and the glades of the Ozark Plateaus Province provide an ideal opportunity. Ozark glades are open 126 Northeastern Naturalist Vol. 16, Special Issue 5 herbaceous communties found on several different substrates (limestone, dolomite, sandstone, rhyolite) within the generally forested Ozark Plateaus Province, in the southern half of Missouri and portions of Arkansas, Illinois, Kansas, and Oklahoma. Experimental evidence suggests that glade endemics fit our conceptual model. They may grow well off of glade soils in the absence of competition (Baskin and Baskin 1988, Sharitz and McCormick 1973, Ware 1991). The limestone glade endemic Talinum calcaricum Ware (Limestone Fameflower; Portulacaceae)was much less affected by drought stress than its generalist competitors when growing together on limestone, supporting our model (Ware 1991). Increasing water availability shifted competitive dominance from the shallow-soil granite endemic Minuartia uniflora (Walt.) Mattf. (One-flower Stichwort; Caryophyllaceae) to the deeper soil endemic Sedum smallii (Britt. ex Small) Ahles (Elf Orpine; Crassulaceae) to the larger species that form the dominant community of normal soils (Sharitz and McCormick 1973). We will employ the same sampling protocols used in California to sample species and environmental variables at 75 paired glade and non-glade sites across the Ozark Plateaus Province. We will test the same hypotheses using the same techniques. We will also determine whether the relationship between climate and edaphic endemism is the same or different for dolomite, limestone, and sandstone. We expect our model to fit best for dolomite, the harshest substrate, and worst for sandstone, the most benign of the three soils. 2. Experimental study of 3 serpentine endemics and 3 non-endemics Serpentine grasslands are ideal for experimental tests of the climatecompetition- soil interaction, because they are dominated by annual species that are amenable to manipulation. Dominant functional groups in both serpentine and other grasslands in California are annual grasses (nearly all exotics) and forbs (native and exotic) (Harrison 1999). At our study site in northern California (the University of California’s McLaughlin Reserve), the vernal annual flora of serpentine grasslands include a small number serpentine endemics, all of which are annual forbs; they include Clarkia gracilis ssp. tracyi (Jepson) Abdel-Hamee &R. Snow (Tracy’s Clarkia; Onagraceae), Navarretia jepsonii V. Bailey ex Jepson (Jepson’s Pincushionplant; Polemoniaceae), and Calycadenia pauciflora A. Gray (Smallflower Western Rosinweed; Asteraceae). Each has a close relative in the nearby nonserpentine flora (Clarkia purpurea (Curtis) Nelson & J.F. Macbr. [Winecup Fairyfan; Onagraceae], Navarretia pubescens (Benth.) Hook & Arn [Downy Pincushionplant; Polemoniaceae], and Hemizonia congesta DC. [Hayfield Tarweed; Asteraceae]). In this part of the study, we will focus on precipitation, because it is well known that the amount and timing of rainfall plays a central role in determining the composition and dynamics of Mediterranean-climate annual systems such as Californian grasslands. To test Hypothesis 2, we will use rainout shelters, supplemental watering, and competition manipulations in the field. We will set up 60 plots in serpentine and non-serpentine grasslands. Each of the three endemic species 2009 S. Harrison, E. Damschen, and B.M. Going 127 and three non-endemic species will be planted into one 30- x 30-cm subplot that is cleared of competing species by clipping all aboveground material, and one 30- x 30-cm subplot that is uncleared. We will impose 2 levels of total water availability: 50% of normal and 200% of normal. These will be compared to a control with normal rainfall. We will impose each of these treatments in combination with two levels of soil (serpentine, non-serpentine) and competition (with, without) in our field experiment. Competitor removal will be repeated monthly by carefully clipping all aboveground material except the focal plants. Plant performance will be assessed by measuring plant height, seed number, seed weight, and root and shoot dry biomass. The experiment will be repeated for two years to ensure that the results are robust against variability in annual climate. We anticipate setting up similar experiments in the Ozark Glades. 3. Resampling Robert Whittaker’s historic Siskiyou sites Our most direct test of our conceptual model’s applicability to modern climate change is our resampling of sites studied by ecologist Robert Whittaker in the Siskiyou Mts of Oregon in 1949–1951 (Whittaker 1954, 1960). The Klamath-Siskiyou region includes North America’s largest exposure of ultramafic rock, including serpentine and peridotite. The size if the area and its great age (>50 ma), high rainfall, and topographic complexity of ultramafic rocks combine to make it the continent’s leading hotspot of serpentine endemism; of 246 plant taxa confined to serpentine and peridotite in California, 97 are found here (Alexander et al. 2006, Coleman and Kruckeberg 1999, Safford et al. 2005). Climatic warming and reduced snowpack, leading to drier conditions in the spring and summer growing season, have been documented throughout the Pacific Northwest (Mote et al. 2003). Flowering phenology has become significantly earlier since the 1950s at the Oregon Caves National Monument at the center of the study region (J. Roth, Oregon Caves National Monument, National Park Service, Cave Junction, OR, unpubl. data). Robert Whittaker’s goal was to quantify plant community variation along three environmental gradients: elevation, soil type, and the local variation from xeric to mesic microsites that he called the “topographic moisture gradient.” A very mesic slope is a streamside or a shallow north slope, while a very xeric site is a steep south-facing slope. We used the topographic moisture gradient as a way to interpret climate-related change. Specifi- cally, we predicted that the plant species composition of any site sampled in 2007 would resemble that of a warmer (more xeric, south-facing) site in 1949–1951. Whittaker collected vegetation data from 290 plots on diorite soils in 1949 and from 55 plots on serpentine and 50 plots on gabbro soils in 1950–1951. At each 50- x 20-m plot, he obtained an estimate of percent areal cover for each herb species, a total herb species list, counts of tree individuals by diameter at breast height (dbh) and species, and counts of shrubs by species. 128 Northeastern Naturalist Vol. 16, Special Issue 5 We entered Whittaker’s data into a database with herb count, herb cover, shrub count, and small and large tree counts, by species and plot number. To locate sites as similar as possible to his (e.g., his plot number 268), we followed the same road or trail on the same substrate (e.g. Wimer Road, serpentine), stopped at the same elevation (e.g., 702 m [2400 ft]), and sought the nearest place with the same slope and aspect (e.g., 25°E). In 2007, we sampled all of Whittaker’s serpentine plots, and 53 plots representing a subset of Whittaker’s diorite that match the serpentine plots in elevation. We followed Whittaker’s sampling methods exactly. We are just beginning to analyze the data. Our preliminary results show that floras have shifted, such that the flora of any given site in 2007 resembles the flora of a warmer slope and aspect in 1949–51. But our preliminary results do not support Hypothesis 3, because this shift to a more “xeric” species composition (i.e., toward identities and abundances of species characteristic of a warmer slope/aspect) appears to have occurred equally in serpentine and diorite floras. Conclusions Special edaphic floras are important contributors to both regional and global biodiversity, but their potentially extreme vulnerability to climate change has been little studied. If one ignores any life-history differences that might distinguish edaphic endemics from other species, as well any role ecological interactions may play in producing edaphic endemism, then the fate of these floras looks very bad. Restricted to tiny edaphic islands, they are much likelier to go extinct under climate change, and much less likely to successfully migrate to suitable new habitats, than plant species that are soil generalists. Our conceptual model, in contrast, considers two additional aspects of edaphic endemism that may be very important in predicting climate-change responses. The first is that endemics may be intrinsically slow-growing, even when adequate resources are available, and thus they may be less sensitive to temperature and water availability than other species regardless of what soils they are growing on. The second is that the restriction of edaphic endemics to special soils may be the outcome of soil-dependent competition with faster-growing generalist species, whose growth is more strongly influenced by temperature and water availability than that of edaphic endemics, but which can only grow well on fertile soils. Putting these two pieces together, our model predicts that in regions where climate change leads to a warmer and wetter environment, serpentine or other edaphic endemics should be outcompeted by soil generalists on all but the harshest “special” soils, and thus endemics may become less common. However, in regions where climate change leads to a warmer and drier environment, edaphic endemics will be less adversely affected than other species, and may even expand their ranges into marginal (less harsh) soil habitats at the expense of generalist species. In the latter case, shifts in the relative competitive abilities of edaphic endemics and generalists could, to some extent, offset the extra risks faced by endemics under climate change. 2009 S. Harrison, E. Damschen, and B.M. Going 129 We are still far from knowing whether our conceptual model provides a good framework for understanding climate-driven changes in either space or time, in serpentine and other edaphic floras. However, we hope it will at least encourage other scientists studying these floras to begin addressing the greatest environmental challenge of our time: human-induced climate change. Acknowledgments The authors thank Nishanta Rajakaruna and Bob Boyd for organizing the Sixth International Serpentine Conference; David Ackerly, Bruce Baldwin, and Rod Myatt for reviewing this paper; and anonymous reviewers for helpful and consistently favorable comments on the proposed work. Literature Cited Alexander, E.A., R.G. Coleman, T. Keeler-Wolf, and S. Harrison. 2006. Serpentine Geoecology of Western North America. Oxford University Press, New York, NY, USA. 512 pp. Anderson, R.C., J.S. Fralish, and J.M. Baskin (Eds.). 1999. Savannas, Barrens, and Rock Outcrop Communities of North America. Cambridge University Press, New York, NY, USA. Baskin, J.M., and C.C. Baskin. 1988. Endemism in rock outcrop plant communities of unglaciated eastern United States: An evaluation of the roles of edaphic, genetic, and light factors. Journal of Biogeography 12(5/6):829–840. Borhidi, A. 1991. Phytogeography and Vegetation Ecology of Cuba. Akademiai Kiado, Budapest, Hungary. Brooks, R.R. 1987. Serpentine and its Vegetation: A Multidisciplinary Approach. Dioscorides Press, Portland, OR, USA. Coleman, R.G., and A.R. Kruckeberg. 1999. Geology and plant life of the Klamath- Siskiyou Mountain Region. Natural Areas Journal 19:320–340. Grace, J.B., H.D. Safford, and S. Harrison 2007. Large-scale causes of variation in the serpentine vegetation of California. Plant and Soil 293:121–132. Harrison, S. 1999. Native and alien species diversity at the local and regional scales in a grazed Californian grassland. Oecologia 121:99–106. Harrison, S., J.L. Viers, and J.F. Quinn, 2000. Climatic and spatial patterns of diversity in the serpentine plants of California. Diversity and Distributions 6:153–161. Harrison, S., H.D. Safford, J.B. Grace, J.H. Viers and K.F. Davies. 2006. Regional and local species richness in an insular environment: Serpentine plants in California. Ecological Monographs 76:41–56. Kruckeberg, A.R. 1954. The ecology of serpentine soils: A symposium. III. Plant species in relation to serpentine soils. Ecology 35:267–274. Kruckeberg, A.R. 1984. California Serpentines: Flora, Vegetation, Geology, Soils and Management Problems. University of California Press, Berkeley, CA, USA. Kruckeberg, A.R. 2005. Geology and Plant Life. University of Washington Press, Seattle, WA, USA. Loarie S.R., B.E. Carter, K. Hayhoe, S. McMahon, R. Moe, C.A. Knight, and D.D. Ackerly 2008. Climate change and the future of California’s endemic flora. PloS One 3:e2502. 130 Northeastern Naturalist Vol. 16, Special Issue 5 Mote, P.W., E.A. Parson, A.F. Hamlet, W.S. Keeton, D. Lettenmaier, N. Mantua, E.L. Miles, D.W. Peterson, D.L. Peterson, R. Slaughter, and A.K. Snover 2003. Preparing for climatic change: The water, salmon, and forests of the Pacific Northwest. Climatic Change 61:45–88. Myers, N., R.A. Mittermeier, C.G. Mittermeier, G.A.B. da Fonseca, and J. Kent. 2000. Biodiversity hotspots for conservation priorities. Nature 403:853–858. Nelson, P.W. 1987. The terrestrial natural communities of Missouri. Missouri Department of Conservation, Jefferson City, MO, USA. Nelson, P., and D. Ladd. 1983. Preliminary report on the identification, distribution, and classification of Missouri glades. Pp. 59–76, In C.L. Kucera (Ed.). Proceedings of the Seventh North American Prairie Conference. Southwest Missouri State University, Springfield, MO, USA. Raven, P.J., and D. Axelrod. 1978. Origin and relationships of the California Flora. University of California Publications in Botany, No. 72. Berkeley, CA, USA. Rune, O. 1953. Plant Life on Serpentine and Related Rocks in the North of Sweden. Acta Phytogeographica Suecica 31, Uppsala, Sweden. Safford, H.D., J.H. Viers and S. Harrison. 2005. Serpentine endemism in the Calfornia flora: A database of serpentine affinity. Madroño 52:222–257. Schwartz, M.W. Iverson, L.R., Prasad, A.M., Matthews S.N., and R.J. O’Connor. 2006. Predicting extinctions as a result of climate change. Ecology 87:1611–1615. Sharitz, R.R., and J.F. McCormick. 1973. Population dynamics of two competing annual species. Ecology 54:723–740. Skinner, M.W., and B.A. Pavlik 1994. California Native Plant Society’s inventory of rare and endangered plants in California. California Native Plant Society special Publication No. 1, 5th edition, Sacramento, CA, USA. Stein B.A., L.S. Kutner, and J.S. Adams. 2000. Precious Heritage: The Status of Biodiversity in the United States. Oxford University Press, Oxford, UK. 399 pp. Tansley, A.G. 1917. On competition between Galium saxatile L. and Galium sylvestre Poll. on different types of soil. Journal of Ecology 5:173–179. Thomas, C.D., A. Cameron, R.E. Green, M. Bakkenes, L.J. Beaumont, Y.C. Collingham, B.F.N. Erasmus, M.F.d. Siqueira, A. Grainger, L. Hannah, L. Hughes, B. Huntley, A.S.v. Jaarsveld, G.F. Midgley, L. Miles, M.A. Ortega-Huerta, A.T. Peterson, O.L. Phillips, and S.E. Williams. 2004. Extinction risk from climate change. Nature 427:45–148. Ware, S. 1991. Influence of interspecific competition, light, and moisture levels on growth of rock outcrop Talinum (Portulacaceae). Bulletin of the Torrey Botanical Club 118(1):1–5 Whittaker, R.H. 1954. The ecology of serpentine soils. IV. The vegetation response to serpentine soils. Ecology 35:275–288. Whittaker, R.H. 1960. Vegetation of the Siskiyou Mountains, Oregon and California. Ecological Monographs 30:279–338.