AOBPreview originally published online on August 31, 2006
Annals of Botany 2006 98(5):1073-1084; doi:10.1093/aob/mcl192
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Phytogeographical Analysis of Seed Plant Genera in China
1 Research and Collections Center, Illinois State Museum 1011 East Ash Street, Springfield, IL 62703, USA
2 Center for Forest Ecology and Forestry Eco-engineering, Institute of Applied Ecology, Chinese Academy of Sciences 72 Wenhua Road, Shenyang, Liaoning 110016, China
3 Department of Ecology, College of Environmental Sciences, Peking University Beijing 100871, China
4 School of Life Sciences, Sun Yatsen University Guangzhou 510275, China
5 Research Center of Systematic and Evolutionary Botany, Institute of Botany, Chinese Academy of Sciences Beijing 100093, China
6 Department of Renewable Resources, University of Alberta Edmonton, Alberta T6G 2H1, Canada
7 Research Center of Plant Ecology and Biodiversity Conservation, Institute of Botany Chinese Academy of Sciences, Beijing 100093, China
*For correspondence. E-mail hqian{at}museum.state.il.us
Received: 16 May 2006 Returned for revision: 14 June 2006 Accepted: 25 July 2006 Published electronically: 31 August 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
• Background and Aims A central goal of biogeography and ecology is to uncover and understand distributional patterns of organisms. China has long been a focus of attention because of its rich biota, especially with respect to plants. Using 290 floras from across China, this paper quantitatively characterizes the composition of floristic elements at multiple scales (i.e. national, provincial and local), and explores the extent to which climatic and geographical factors associated with each flora can jointly and independently explain the variation in floristic elements in local floras.
• Methods A study was made of 261 local floras, 28 province-level floras and one national-level flora across China. Genera of seed plants in each flora were assigned to 14 floristic elements according to their worldwide geographical distributions. The composition of floristic elements was related to climatic and geographical factors.
• Key Results and Conclusions Variations in percentages of cosmopolitan, tropical and temperate genera among local floras tend to be greater at higher latitudes than at lower latitudes. Latitude is strongly correlated with the proportions of 13 of the 14 floristic elements. Correlations of the proportions of floristic elements with longitude are much weaker than those with latitude. Climate represented by the first principal component of a principal component analysis was strongly correlated with the proportions of floristic elements in local floras (|r| = 0·75 ± 0·18). Geographical coordinates independently explained about four times as much variation in floristic elements as did climate. Further research is necessary to examine the roles of water–energy dynamics, geology, soils, biotic interactions, and historical factors such as land connections between continents in the past and at present in creating observed floristic patterns.
Key words: Biogeography, climate, floristics, latitudinal gradient, regionalization, seed plants
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
A central goal of biogeography and ecology is to uncover and understand distributional patterns of organisms. China has long been a focus of attention because of its rich biota, especially with respect to plants (Wu, 1991; Axelrod et al., 1998). The composition of the flora of China has been well documented through the compilation of the Flora Reipublicae Popularis Sinicae (an 80-volume set of 125 books, all having been published during 1959–2004), the Flora of China (Wu and Raven, 1994 to present; 11 volumes having been published), and many regional, provincial and local floras. Although China is nearly identical in size and of similar latitudinal breadth to the United States (9·6 vs. 9·4 million km2), the vascular plant flora of China is richer than that of the USA by a factor of 1·6 (Qian and Ricklefs, 1999). Of the world's estimated 260 140 species of vascular plants (Mabberley, 1997), approximately 30 000 (or 11·5 %) occur in China. The earliest (at least 124·6 million years old) angiosperm megafossil known to date, from the Archaefructaceae, is found in China (Sun et al., 2002). China is also the only country in the world that supports vegetational continuity from tropical, subtropical, temperate to boreal forests (Axelrod et al., 1998). This unbroken latitudinal gradient of forest vegetation, together with many north–south-orientated, highly rugged mountain ranges in south-western China, presumably reduced rates of extinction when advances and retreats of Pleistocene glaciations forced plant species to migrate southward and northward in response to climate change. Furthermore, the collision of the Indian subcontinent with the Asian continent commencing about 50 million years ago (Ma) created highly dissected and elevated landscapes in China, which made the region not only an important centre of survival (i.e. a ‘living museum’ or ‘refugia’), but also an important centre of speciation and evolution (i.e. a ‘cradle’) for vascular plants (Axelrod et al., 1998; Qian, 2002). Many genera and families of vascular plants can be found nowhere else (Ying et al., 1993). The flora of China is considered to be a mixture of Laurasian, Gondwanan and Tethyan elements (Wu, 1988), thus providing unique opportunities for addressing numerous important issues concerning global biogeography.
A major approach used to characterize a flora or fauna and to determine its biogeographical relationships with other regions is to group the taxa that comprise the flora or fauna of a given geographical area into geographical elements or types, defined according to the geographical range of taxa (Stott, 1981). Zhengyi Wu (C. Y. Wu) used this approach to characterize the flora of China at the generic level, based on floristic relationships between China and the rest of the world. He grouped the genera of Chinese seed plants into 15 areal types (floristic or phytogeographical elements), based on their worldwide distributions, and the results of his studies were reported in several publications (e.g. Wu, 1980, 1991; Wu and Wang, 1983). Following his classification, several hundred regional, provincial and local floras in China have been analysed. Each analysis tabulated the number of genera in each of the 15 areal types. However, none has analysed multiple floras except for a few studies that focused on a particular region (e.g. Shen and Zhang 2000). Little is known about how floristic patterns are related to climatic conditions and geographical factors.
The objectives of this study were to characterize quantitatively the composition of floristic elements in floras across China at multiple scales and to explore the extent to which geographical variables (e.g. latitude, longitude, area, elevation) and climatic variables (e.g. temperature and precipitation) can jointly as well as independently explain the variation in each of the different floristic elements in local floras. In addition, the present paper updates tabulations of genera in each of Wu's floristic elements for all of China and each of its provinces.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
Data sets
A database was compiled that included all genera of seed plants known in China primarily based upon the Flora Reipublicae Popularis Sinicae, Flora of China, regional and provincial floras, and journal articles pertinent to the flora of China. Electronic data from the Flora of China Checklist (http://mobot.mobot.org/W3T/Search/FOC/projsfoc.html) and the electronic version of the Flora of China included in eFlora (http://www.efloras.org/) were frequently accessed during this compilation. The information for the presence/absence and native/exotic status of each Chinese seed plant genus was documented for each of the 28 Chinese provinces and autonomous regions (all provinces and autonomous regions are referred to as provinces hereafter for convenience of discussion; Fig. 1).
|
Following Wu (1991, 1993), each genus of the Chinese seed plants was assigned to one of the following floristic elements based on the native distributions of each genus: cosmopolitan, pantropical, tropical Asian–tropical American, palaeotropical, tropical Asian–tropical Australian, tropical Asian–tropical African, tropical Asian, holarctic, eastern Asian–North American, temperate Eurasian, temperate Asian, Mediterranean, western to central Asian, central Asian, eastern Asian, and Chinese endemic (Table 1). Because the geographical limit of Chinese endemics was set by the national border of China and because the nature of this floristic element is by and large shared with the eastern Asian floristic element, as discussed in Qian et al. (2003), the Chinese endemics were combined with the eastern Asian floristic element in data analyses. Geographical delineations of the 14 floristic elements are described in Table 1. The proportions (in percentages) of genera for each of the 14 floristic elements in China as a whole and in each of 28 provinces were tabulated.
|
Data of floristic element composition in local floras were collected primarily from literature that was exclusively published in Chinese. The literature to 2003 was searched, including floras, books, monographs, proceedings, theses and journal articles. Data were found on the floristic elements defined by Wu for over 300 local floras. The following were excluded: those floras that were atypical to a region (e.g. floras from dry-hot valleys in south-west China, small limestone areas, wetlands), that applied to a particular type of vegetation (e.g. evergreen plant communities, alpine tundra), that did not include all seed plant genera in a flora (e.g. only for woody plants or halophytes), or that did not provide geographical coordinates. As a result, 232 local floras were selected. In addition, there are 29 floras for which the number of genera for each of Wu's floristic elements is not available in the original literature, and for these seed plant genera were tabulated in each of Wu's 15 floristic elements. As a result, a total of 261 local floras were included in the present study. These floras sampled every province and the full geographical range of China (Fig. 1). Most characterized vegetation in nature reserves. Chinese endemic genera were combined with eastern Asian genera in data analyses as discussed above.
For each flora, latitude and longitude (°) of the geographical midpoint and, if available, area (km2), maximum elevation (m) and topographic relief (elevation range; m) were recorded. Of the 261 local floras, 215 have area size data, and 151 have area size data plus maximum elevation and topographic relief. The mean area of the 215 floras is 15 327 km2, which is equivalent to a circle with a diameter of approximately 140 km. Several climatic variables were also obtained for the latitude–longitude half-degree grid point closest to the geographical midpoint of each flora using data in the International Institute of Applied System Analysis (IIASA) climatic database (Leemans and Cramer, 1991). This database provides values for each 0·5 ° of latitude and longitude. The climatic variables were mean annual temperature (°C), mean January temperature (°C), difference between mean January temperature and mean July temperature (°C), annual precipitation (mm) and summer (May to August) precipitation (mm). Two derived climatic indices were included: actual evapotranspiration (AET, mm) and potential evapotranspiration (PET, mm), which is proportional to the drying power of the environment, primarily a function of temperature. AET and PET were calculated following the approach developed by Cramer and Prentice (Cramer and Prentice, 1988; Prentice et al., 1992, 1993). These variables are often considered the most important climatic factors generating large-scale biotic patterns (e.g. Turner et al., 1988; O'Brien, 1998; O'Brien et al., 1998; Qian, 1998; Badgley and Fox, 2000; Rahbek and Graves, 2001; Diniz-Filho et al., 2003; Hawkins and Porter, 2003; Qian and Ricklefs, 2004; Ricklefs et al., 2004).
Data analyses
To minimize the effect of area size on analytical results, the percentage contribution of each floristic element to a flora was used to compare the compositions of floristic elements between floras. The total number of seed plant genera and the proportions (%) of genera for each floristic element were tabulated for the flora of China as a whole and for each province. The proportion of genera for each of the 14 floristic elements in the 261 local floras was summarized for their mean and standard deviation (s.d.) for each of the 28 provinces in China.
Spearman's rank correlations were used to determine correlations between proportions of floristic elements and latitudes and longitudes using the 261 local floras. To examine the relationships between floristic element compositions and climatic conditions, multivariate analyses were first performed to extract fewer synthetic variables for each data set and these resulting variables were then related to each of the original variables of the two data sets. A principal component analysis (PCA) was performed on the climatic data set based on a correlation matrix. The statistical significance of principal components was evaluated using the broken-stick model of Jackson (1993). A global non-metric multidimensional scaling (NMDS; Minchin, 1987) was used to extract highly informative dimensions from the data set of floristic elements. NMDS was chosen because it is well suited to floristic data (McCune and Mefford, 1999) and because it makes no assumptions about the data and provides a robust solution. The NMDS used Sørensen's index as a measure of distance with the maximum number of iterations set at 200. Significance of extracted dimensions was evaluated by a Monte Carlo test (30 runs with randomized data).
Using the 261 local floras, partial regressions were performed to determine the variation in floristic elements explained by geographical coordinates and climate. In each partial regression, three coefficients of determination were obtained using three general linear models: one combining both geographical coordinates and climate, one including only geographical coordinates, and the other including only climate. By comparing the three coefficients of determination, the proportions of the variance explained (1) by geographical coordinates independently, (2) by climate independently and (3) by the two jointly can be determined (Legendre and Legendre, 1998). The procedure of Borcard et al. (1992; see also Legendre, 1993) was followed by including all terms in a third-order polynomial of geographical coordinates (e.g. x, y, x2, xy, y2, x3, x2y, xy2, y3, where x and y represent latitude and longitude, respectively). This ensures not only linear relationships but also more complex (e.g. quadratic and cubic) relationships between floristic elements and geographical coordinates (Borcard et al., 1992). Bivariate plots of the relationships of the proportions of floristic elements and the seven climatic variables used in this study were examined to determine the relationships between these variables; no evidence for non-linear relationships was found.
Because area and elevation vary among the local floras used, which may have effects on the proportional composition of floristic elements in local floras, it is necessary to examine the amount of variation in floristic elements that can be explained by these two geographical variables in addition to the variation explained by latitude and longitude. The same partial regression models mentioned above were used except that climatic variables were replaced with log10-transformed area, maximum elevation and topographic relief.
SYSTAT version 7 (Wilkinson et al., 1992) was used for statistical summaries, correlation analyses and partial regression analyses, and PC-ORD version 4 (McCune and Mefford, 1999) was used for multivariate analyses (i.e. PCA and NMDS).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
According to the data, China has 3143 genera of native seed plants. Of these, 105 (3·3 %) are cosmopolitan, 1512 (48·1 %) are tropical in nature (FE2–FE7 in Table 2) and 1526 (48·6 %) are primarily temperate (FE8–FE14 in Table 2). In total, 590 (18·8 %) genera are restricted to eastern Asia, of which 270 are endemic to China. The total number of seed plant genera and the percentages of genera for each floristic element in each of the 28 provinces are shown in Table 2. The proportion of genera for each of the 14 floristic elements for the 261 local floras was summarized according to province (Table 3).
|
|
Latitude strongly and positively correlates with the proportions of genera of the cosmopolitan, holarctic, temperate Eurasian, temperate Asian, Mediterranean, western to central Asian, and central Asian floristic elements, and it strongly and negatively correlates with the proportions of genera in all six tropical floristic elements and negatively correlates with eastern Asian genera weakly but significantly (Table 4). Correlations of floristic elements with longitude are much weaker than those with latitude (Table 4). The strongest correlation is with the proportion of the eastern Asian–eastern North American genera (rs = 0·44), which is followed by the correlations with the proportions of the central Asian genera (rs = – 0·37) and Mediterranean, western to central Asian genera (rs = –0·35) (Table 4).
|
The first and second principal components of the PCA had eigenvalues of 5·404 and 0·965, and explained 77·2 and 13·8 % of the variation in the climatic data set, respectively. A broken-stick test indicated that only the first principal component (PC1) was significant. This principal component is highly correlated with all the seven climatic variables (|r| = 0·82 ± 0·07; Table 5), and is correlated highly to moderately with the floristic elements (|r| = 0·75 ± 0·18). The autopilot mode of an NMDS analysis running on the data set of floristic elements recommended a one-dimensional solution, and this dimension (NMDS1) is statistically significant (Monte Carlo test: P < 0·05, n = 30). Correlations between NMDS1 and the seven climatic variables and 14 floristic elements are all significant (P < 0·05), albeit weak in some cases (Table 5). The correlation between PC1 and NMDS1 is high (r = –0·89, P < 0·001), indicating strong relationships between climate variables and floristic elements. Both PC1 and NMDS1 are highly correlated with latitude (r = 0·902 for PC1, r = –0·871 for NMDS1). When the 261 local floras were ordinated by PC1 and NMDS1, the floras south of 30 °N tended to be completely separated from the floras north of it, and the floras north of 40 °N tended to intermingle with those located between 30 ° and 40 °N (Fig. 2).
|
|
The geographical coordinates of local floras explained, on average, 79·6 % of the variation in floristic elements (fractions [a] + [b] in Table 6). When climatic variables were added to the models, the two sets of variables together explained, on average, 82·9 % of the variation in floristic elements (Table 6). Partial regressions indicated that large amounts (66 % on average) of the variation in floristic elements were explained by both geographical coordinates and climatic variables (fraction [b] in Table 6). In other words, approximately four-fifths (or 80 %) of the explained variation was shared by the two sets of the explanatory variables. Of 16·9 % of the variance to which independent effects could be attributed, geographical coordinates accounted for 4·1 times as much as did climatic variables (the average of fraction [a] vs. the average of fraction [c] in Table 6, i.e. 13·6 vs. 3·3 %). When individual floristic elements were examined separately, geographical coordinates explained more variation than did climatic variables for 13 of the 14 floristic elements (compare fractions [a] with [b] in Table 6; binomial test: variation explained by geographical coordinates > variation explained by climatic variables, 13 of 14, P < 0·001). For the holarctic genera, the amounts of variation explained independently by geographical coordinates and by climatic variables were nearly equal (4·0 vs. 4·4 %; Table 6).
|
When using a smaller data set with 151 local floras that have both area size and elevational data (see Materials and methods for details), geographical coordinates explained approximately the same amount of the variation in floristic elements as they did in the full data set with 261 local floras (80·9 ± 11·2 vs. 79·6 ± 9·3 %, P = 0·808). When the proportion of a floristic element was regressed on the geographical coordinates (including all terms in a third-order polynomial) plus log10area, maximum elevation and topographic relief, they together explained, on average, 83·4 % of the variation in floristic elements (Table 7). The largest amount of the explained variation (95·4 %) is with the model for the holarctic genera (Table 7). The amount of the variation in floristic elements explained by geographical coordinates independent of area and elevation variables is much larger, by a factor of 18·3, than that explained by area and elevation variables independent of geographical coordinates (compare fractions [a] and [c] in Table 7; 51·2 ± 11·9 vs. 2·8 ± 2·0 %; ANOVA: F1,26 = 224, P < 0·001). Approximately 29 % of the variation in floristic elements is explained jointly by geographical coordinates and by log10area, maximum elevation and topographic relief (fraction [b] in Table 7).
|
When all tropical and temperate genera were pooled from each of the 261 local floras, the equilibrium latitude between tropical and temperate genera is approximately 27·5 °N (Fig. 3). The proportion of tropical genera decreases drastically and almost linearly from the southernmost latitude (approx. 18 °N) northward to approx. 35–37 °N, from where it decreases slowly and tended to level off. The proportion of temperate genera primarily mirrors, but inversely, the pattern for the tropical genera (Fig. 3). Variation in the percentage of the three major floristic groups (i.e. cosmopolitan, tropical and temperate genera) among local floras tended to be greater at higher latitudes than at lower latitudes (Fig. 3).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
A major objective of floristic analysis is to identify floristic patterns among the distributions of taxa (Stott, 1981; McLaughlin, 1989). A traditional approach of identifying distribution affinities among the taxa of a flora involves the recognition of floristic elements (phytogeographical elements or areal types) based upon the native worldwide distributional ranges of the taxa. Although floristic elements may be defined in different ways and the assignment of taxa to floristic elements may be more confident for some taxa than others, the floristic element approach has been used widely. For example, Weber (1965) used a hierarchical system of floristic elements to describe the flora of the southern Rocky Mountains; van Balgooy (1976) recognized 15 elements in a study of the phytogeography of New Guinea; Thorne (1972) recognized 16 elements that included all major disjunct distributions; Lausi and Nimis (1985) classified plant species from the Yukon Territory (north-west Canada) into eight elements; Qian (1999, 2001) recognized ten floristic elements in the native flora of North America north of Mexico; and Wu and Wang (1983; Wu, 1991) segregated indigenous Chinese seed plant genera into 15 elements, which they refer to as areal types. Wu's approach has been used in studies of over 400 regional, provincial and local floras across China. Following Wu's approach, except that Chinese endemic and eastern Asian elements were combined, Qian et al. (2003) analysed the phytogeography of East Asia that includes the eastern forested part of China, the Korean peninsula, Japan and the Russian Far East. The latitudes of the study area ranged from the southernmost land area of China (at about 18 °N) northward to 73 °N. The floras used in Qian et al. (2003) were at the provincial level or equivalent. Thus, the present study extended this by analysing local-scale floras distributed across the entire ranges of latitudes and longitudes in China.
The present study shows that the majority of the variation in floristic elements explained by climatic variables can also be explained by geographical coordinates, and vice versa. However, the amount of the variation uniquely explained by geographical positions is more than that uniquely explained by climatic variables, by a factor of 4·2 on average (Table 6). This suggests that geographical position, represented by latitude and longitude, of a flora has played a more important role in determining the composition of floristic elements than climate. This may be in part because floristic elements are defined by their geographical, rather than ecological, distributions. For example, the genera of the Mediterranean, western to central Asian element and the central Asian element are both centred in the areas west of China. As a result, the proportion of genera belonging to these two elements in a flora decreases from west to east. By contrast, the proportion of genera belonging to the eastern Asian–North American and tropical Asian–tropical American elements increases from west to east.
Variation in the proportional contribution of each of the floristic elements to a flora increases with latitude (Fig. 2). Floras north of 40 °N tend to intermingle with those located between 30 ° and 40 °N (Fig. 3). One possible explanation for this pattern may be that the penetration of some tropical genera into northern floras is determined more by the availability of particular suitable habitats in northern floras than by latitude, which is highly correlated with temperature (MacArthur, 1972). Some more northern localities may hold suitable habitats for certain tropical genera, and these habitats may be absent from some less northern localities. Variation in the proportion of genera belonging to tropical elements ultimately results in variation in the proportion of genera belonging to temperate elements. Testing this hypothesis requires additional data that include information about both plant distributions and local habitats.
The flora of China is strongly linked to the rest of the world. Of the 3143 seed plant genera in China, 1733 (55·1 %) are distributed in two or more of the world's continents (i.e. all those genera other than in the floristic elements FE7, FE11, FE13 and FE14 in Table 2). Although counts of genera shared between China and other continents are not available, we may make reasonable estimates based on the number of genera in each of the 14 floristic elements. China shares about 740 genera of seed plants with Europe (i.e. all genera in the floristic elements FE1, FE8, FE10 and FE12 in Table 2), shares no fewer than 650 genera with Africa (i.e. all genera in FE1, FE2 and FE4, and some genera in FE12), shares about 770 genera with Australia (i.e. all genera in FE1, FE4 and FE5 plus those genera in FE2 that reach the Australian continent), shares about 790 genera with North America (i.e. all genera in FE1, FE8 and FE9, and about 260 genera in FE2 and FE3), and shares about 500 genera with Central and South America (i.e. genera in FE1, FE2 and FE3). The floristic relationships between China and other regions of the world may be traced back to a time when all the major landmasses were united in the single supercontinent, Pangaea. About 180 Ma (McLoughlin, 2001), Pangaea broke into Gondwana, which became Australia, Antarctica, India, South America and South Africa, and Laurasia, which consisted of the present-day continents of Asia, Europe and North America (including Greenland). Asia and Europe have been on the same landmass since the breakup of Pangaea, although the floras of the two continents were once separated by the Eurasian north–south epicontinental seaway, the Turgai Strait, in western Siberia and eastern Europe before the end of the Eocene (approx. 35 Ma) (Wolfe, 1997). North America was connected to Europe until at least the early Eocene (50 Ma) through the North Atlantic Land Bridge (Tiffney, 1985, 2000), and was connected to Asia by land between Alaska and eastern Siberia in the mid Cretaceous (100 Ma) (Sanmartín et al., 2001). The connection between Asia and North America persisted via the Bering Land Bridge until the Pliocene. Because of the breakup of Pangaea, the terrestrial connection between North and South America was broken in the early Cretaceous (Emery and Uchupi, 1984) but the two continents were reunited during the Pliocene (3·7–3·0 Ma) via the Isthmus of Panama, the Panamanian Land Bridge (Hallam, 1994; Coates, 1996). All these land connections have undoubtedly contributed to the current floristic relationships between China and the rest of the world, through providing passages for plant migration between continents. According to palaeobotanical data, the Bering Land Bridge was open to migration of deciduous taxa throughout most of the Tertiary and was possibly open to evergreen taxa in the early Eocene; the North Atlantic Land Bridge, which lay at a lower latitude (with the southern tip of Greenland perhaps extending south to 45 °N; Smith et al., 1981; Tiffney and Manchester, 2001) in the early Tertiary, was open to both deciduous and evergreen taxa in the early Eocene (Tiffney, 2000). The floristic relationship between China and Africa was established in part because of the current terrestrial connection between Africa and Eurasia and in part because India brought the Gondwanan elements with it when it broke away from Gondwana, to which both India and Africa belonged, and then collided with Asia in the early Tertiary. Although India underwent remarkable latitudinal, and hence climatic and vegetational, changes (McLoughlin, 2001) after it rifted from Madagascar about 90 Ma and extensive volcanism, approximately 65 Ma, erased many taxa from central western India (Officer et al., 1987; Conti et al., 2002), fossil and molecular data demonstrate that some of the current elements in the Asian flora and fauna were brought by India from Gondwana (Conti et al., 2002).
Based upon the above estimated numbers of genera shared between China and other continents, China shares more genera with North America than with Europe, despite the fact that China and Europe are on the same Eurasian landmass. This pattern is concordant with the pattern reported by Qian (1999): of the 908 genera shared between North America (north of Mexico) and Eurasia, 93·6 % occurred in Asia and 81·7 % in Europe. This pattern could have arisen for several reasons. First, North America is geographically closer to Asia than to Europe. Although Asia and Europe are connected, they were separated by the Turgai Strait during the early Tertiary as mentioned above. This epicontinental seaway—which extended from the Arctic Ocean to the Tethys Seaway, was closed at its northern end in the Palaeocene and retreated in the late Oligocene (Tiffney and Manchester, 2001)—is believed to have acted as a barrier for floristic interchanges between the two continents. Second, the Bering Land Bridge persisted for much longer than the North Atlantic Land Bridge (Tiffney, 2000). The Bering Land Bridge was open to terrestrial migrations from at least the early Palaeogene until the opening of the Bering Strait between about 7·4 and 4·8 Ma (Marinkovich and Gladenkov, 1999; Tiffney and Manchester, 2001). During Pleistocene glaciations, terrestrial connections between Asia and North America were re-established (Sanmartín et al., 2001). By contrast, the southern (Thulean) route of the North Atlantic Land Bridge was broken in the early Eocene (50 Ma) and the northern (De Geer) route of the bridge was broken in the late Eocene (39 Ma) (Sanmartín et al., 2001). Third, fossil evidence demonstrates that more genera of seed plants went extinct in Europe than in Asia and North America during late Tertiary climate cooling and Pleistocene glaciations. For example, Aralia, Carya, Castanopsis, Catalpa, Hamamelis, Illicium, Lindera, Liriodendron, Liquidambar, Lithocarpus, Magnolia, Nyssa, Sassafras, Stewartia, Thuja and Torreya are all well-known eastern Asian–North American disjuncts; they occurred in Europe during the Tertiary (Latham and Ricklefs, 1993). Their extinction from Europe but persistence in both Asia and North America contributes to the closer floristic relationship of eastern Asia with North America than either with Europe.
Of the 1410 genera in China that are restricted to Asia (i.e. all genera belonging to floristic elements FE7, FE11, FE13 and FE14 in Table 2), many, if not most, are Tertiary relicts and hence palaeoendemics (Ying et al., 1993). Some were once distributed in one or more of the other continents, usually Europe and/or North America, based on fossil evidence. These Tertiary relicts include all endemic gymnosperm genera such as Cathaya, Cephalotaxus, Cunninghamia, Ginkgo, Glyptostrobus, Keteleeria, Metasequoia and Pseudolarix, and many angiosperm genera such as Cercidiphyllum, Cyclocarya, Emmenopterys, Engelhardtia, Eucommia, Euptelea, Fortunearia, Hemiptelea, Paulownia, Phellodendron, Platycarya, Pteroceltis, Sargentodoxa, Sinowilsonia and Tapiscia (Benton, 1993; Latham and Ricklefs, 1993, and references therein; Manchester, 1999). These once widely distributed but now narrowly restricted eastern Asian endemic genera provide links between the modern flora of eastern Asia and the Tertiary floras of other continents. Neoendemic genera probably account for a very small proportion of the total number of the eastern Asian endemics (Ying et al., 1993). These genera include Ajania, Anemoclema, Diplazoptilon, Dipoma, Formania, Isometrum, Metanemone, Sinadoxa, Skapanthus and Smithorchis (Wu, 1988; Ying et al., 1993). Some may have evolved probably as a result of the Himalayan–Tibetan uplift. One such example is Ajaniopsis, which is a close relative of Ajania and is distributed between 4600 and 5000 m in elevation in Xizang (Tibet) (Ying et al., 1993); another example is Sinadoxa distributed between 3900 and 4800 m in Qinghai (Wu, 1988; Liang and Wu, 1995). A low level of divergence in the internal transcribed spacer of the nuclear ribosomal DNA sequence of Sinadoxa from its progenitor Adoxa suggests a recent origin of this genus (Liu et al., 1999, 2000).
In conclusion, floristic patterns are strongly associated with geographical (particularly latitudinal) factors in local floras across China. Climate alone explains a small proportion of the variance in the composition of floristic elements. Further research is necessary to examine the roles of water–energy dynamics, geology, soils, biotic interactions, and historical factors such as land connections between continents in the past and present in creating observed floristic patterns.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
We are grateful to Shijian Deng, Sanhong Gao, Mei Han, Chang-Fu Hsieh, Guangyan Ni, Zehao Shen, Hongda Zhang (Hungta Chang) and Biao Zhu for help with botanical data, to Changhui Peng and Guofan Shao for climate data, and to William E. McClain and two anonymous reviewers for helpful comments. We would like to dedicate this paper to Prof. Zhenyi Wu (C. Y. Wu), who has been a leading botanist in China for decades and has developed a floristic analysis approach that has been widely used in China. This project is supported in part by a grant from the Institute of Applied Ecology, Chinese Academy of Sciences.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSLITERATURE CITED |
|---|
-
Axelrod DI, Al-Shehbaz I, Raven PH. (1998) History of the modern flora of China. In Zhang A and Wu S (Eds.). Floristic characteristics and diversity of East Asian plants(China Higher Education Press, Beijing) pp. 43–55.
Badgley C and Fox DL. (2000) Ecological biogeography of North American mammals: species density and ecological structure in relation to environmental gradients. Journal of Biogeography 27:1437–1467.[CrossRef][Web of Science]
van Balgooy MMJ. (1976) Phytogeography. In Paijmans K (Ed.). New Guinea vegetation(Australian National University Press, Canberra) pp. 1–22.
Benton MJ. (1993) The fossil record 2(Chapman and Hall, London).
Borcard D, Legendre P, Drapeau P. (1992) Partialling out the spatial component of ecological variation. Ecology 73:1045–1055.[CrossRef][Web of Science]
Cramer WP and Prentice IC. (1988) Simulation of regional soil moisture on a European scale. Norsk Geografisk Tidsskrift 42:149–151.
Coates AG. (1996) The geologic evolution of the Central American isthmus. In Obando JA, Jackson JBC, Budd AF, Coates AG (Eds.). Evolution and environment in tropical America(University of Chicago Press, Chicago) pp. 21–56.
Conti E, Eriksson T, Schönenberger J, Sytsma KJ, Baum DA. (2002) Early Tertiary out-of-India dispersal of Crypteroniaceae: evidence from phylogeny and molecular dating. Evolution 56:1931–1942.[CrossRef][Web of Science][Medline]
Diniz-Filho JAF, Bini LM, Hawkins BA. (2003) Spatial autocorrelation and red herrings in geographical ecology. Global Ecology and Biogeography 12:53–64.
Emery KO and Uchupi E. (1984) The geology of the Atlantic Ocean(Springer, New York).
Hallam A. (1994) An outline of Phanerozoic biogeography(Oxford University Press, Oxford).
Hawkins BA and Porter EE. (2003) Does herbivore diversity depend on plant diversity? The case of California butterflies. American Naturalist 161:40–49.
Jackson DA. (1993) Stopping rules in principal components analysis: a comparison of heuristical and statistical approaches. Ecology 74:2204–2214.[CrossRef][Web of Science]
Latham RE and Ricklefs RE. (1993) Continental comparisons of temperate-zone tree species diversity. In Ricklefs RE and Schluter D (Eds.). Species diversity in ecological communities(University of Chicago Press, Chicago) pp. 294–314.
Lausi D and Nimis PL. (1985) Quantitative phytogeography of the Yukon Territory (NW Canada) on a chorological–phytosociological basis. Vegetatio 59:9–20.[CrossRef]
Leemans R and Cramer WP. (1991) The IIASA database for mean monthly values of temperature, precipitation and cloudiness on a global terrestrial grid(International Institute for Applied Systems Analysis, Laxenburg, Austria) Research Report RR-91-18.
Legendre P. (1993) Spatial autocorrelation: trouble or new paradigm? Ecology 74:1659–1673.[CrossRef][Web of Science]
Legendre P and Legendre L. (1998) Numerical ecology 2nd edn (Elsevier, Amsterdam).
Liang H-X and Wu C-Y. (1995) On the taxonomic system, phylogeny and distribution in Adoxaceae. Acta Botanica Yunnanica 17:380–390.
Liu JQ, Zhou GY, Ho TN, Lu AM. (1999) Karyomorphology of Sinadoxa and its systematic significance. Caryologia 52:159–164.
Liu JQ, Chen ZD, Lu AM. (2000) The phylogenetic relationships of Sinadoxa, revealed by the ITS data. Acta Botanica Sinica 42:656–658.
Mabberley DJ. (1997) The plant-book: a portable dictionary of the vascular plants 2nd edn (Cambridge University Press, Cambridge).
MacArthur RH. (1972) Geographical ecology: patterns in the distribution of species(Harper and Row, New York).
Manchester SR. (1999) Biogeographical relationships of North American Tertiary floras. Annals of the Missouri Botanical Garden 86:472–522.[CrossRef]
Marinkovich L and Gladenkov AY. (1999) Evidence for an early opening of the Bering Strait. Nature 397:149–151.[CrossRef]
McCune B and Mefford MJ. (1999) PC-ORD—multivariate analysis of ecological data (version 4·0)(MjM Software Design, Gleneden Beach, OR).
McLaughlin SP. (1989) Natural floristic areas of the western United States. Journal of Biogeography 16:239–248.[CrossRef]
McLoughlin S. (2001) The breakup history of Gondwana and its impact on pre-Cenozoic floristic provincialism. Australian Journal of Botany 49:271–300.[CrossRef]
Minchin PR. (1987) An evaluation of the relative robustness of techniques for ecological ordination. Vegetatio 69:89–107.[CrossRef][Web of Science]
O'Brien EM. (1998) Water–energy dynamics, climate, and prediction of woody plant species richness: an interim general model. Journal of Biogeography 25:379–398.[CrossRef]
O'Brien EM, Whittaker RJ, Field R. (1998) Climate and woody plant diversity in southern Africa: relationships at species, genus and family levels. Ecography 21:495–509.[CrossRef]
Officer CB, Hallam A, Drake CL, Devine JD. (1987) Late Cretaceous and paroxysmal Cretaceous/Tertiary extinctions. Nature 326:143–149.[CrossRef]
Prentice IC, Cramer W, Harrison SP, Leemans R, Monserud RA, Solomon AM. (1992) A global biome model based on plant physiology and dominance, soil properties and climate. Journal of Biogeography 19:117–134.[CrossRef]
Prentice IC, Sykes MT, Cramer W. (1993) A simulation model for the transient effects of climate change on forest landscapes. Ecological Modelling 65:51–70.[CrossRef]
Qian H. (1998) Large-scale biogeographical patterns of vascular plant richness in North America: an analysis at the generic level. Journal of Biogeography 25:829–836.[CrossRef]
Qian H. (1999) Spatial pattern of vascular plant diversity in North America north of Mexico and its floristic relationship with Eurasia. Annals of Botany 83:271–283.
Qian H. (2001) A comparison of generic endemism of vascular plants between East Asia and North America. International Journal of Plant Sciences 162:191–199.[CrossRef]
Qian H. (2002) A comparison of the taxonomic richness of temperate plants in East Asia and North America. American Journal of Botany 89:1818–1825.
Qian H and Ricklefs RE. (1999) A comparison of the taxonomic richness of vascular plants in China and the United States. American Naturalist 154:160–181.[CrossRef]
Qian H and Ricklefs RE. (2004) Geographical distribution and ecological conservatism of disjunct genera of vascular plants in eastern Asia and eastern North America. Journal of Ecology 92:253–265.[CrossRef]
Qian H, Song J-S, Krestov P, Guo Q-F, Wu Z-M, Shen X-S, Guo X-S. (2003) Large-scale phytogeographical patterns in East Asia in relation to latitudinal and climatic gradients. Journal of Biogeography 30:129–141.[CrossRef]
Rahbek C and Graves GR. (2001) Multiscale assessment of patterns of avian species richness. Proceedings of the National Academy of Sciences of the USA 98:4534–4539.
Ricklefs RE, Qian H, White PS. (2004) The region effect on mesoscale plant species richness between eastern Asia and eastern North America. Ecography 27:129–136.[Medline]
Sanmartín I, Enghoff H, Ronquist F. (2001) Patterns of animal dispersal, vicariance and diversification in the Holarctic. Biological Journal of the Linnean Society 73:345–390.[CrossRef][Web of Science]
Shen Z-H and Zhang X-S. (2000) A quantitative analysis on the floristic elements of the Chinese subtropical region and their spatial patterns. Acta Phytotaxonomica Sinica 38:366–380.
Smith AG, Hurley AM, Briden JC. (1981) Phanerozoic paleocontinental world maps(Cambridge University Press, Cambridge).
Stott P. (1981) Historical plant geography(George Allen and Unwin, London).
Sun G, Ji Q, Dilcher DL, Zheng S, Nixon KC, Wang X. (2002) Archaefructaceae, a new basal angiosperm family. Science 296:899–904.
Thorne RF. (1972) Major disjunctions in the geographical ranges of seed plants. Quarterly Review of Biology 47:365–411.[CrossRef]
Tiffney BH. (1985) The Eocene North Atlantic land bridge: its importance in Tertiary and modern phytogeography of the Northern Hemisphere. Journal of the Arnold Arboretum 66:243–273.
Tiffney BH. (2000) Geographic and climatic influences on the Cretaceous and Tertiary history of Euramerican floristic similarity. Acta Universitatis Carolinae—Geologica 44:5–16.
Tiffney BH and Manchester SR. (2001) The use of geological and paleontological evidence in evaluating plant phylogeographical hypotheses in the Northern Hemisphere tertiary. International Journal of Plant Sciences 162:S3–S17.[CrossRef]
Turner JRG, Lennon JJ, Lawrenson JA. (1988) British bird species distributions and energy theory. Nature 335:539–541.[CrossRef]
Weber WA. (1965) Plant geography in the southern Rocky Mountains. In Herbert J, Wright E, Frey DG (Eds.). The Quaternary of the United States(Princeton University Press, Princeton) pp. 453–468.
Wilkinson L, Hill M, Welna JP, Birkenbeuel GK. (1992) SYSTAT for Windows: statistics(SYSTAT Inc, Evanston, IL).
Wolfe JA. (1997) Relations of environmental change to angiosperm evolution during the late Cretaceous and Tertiary. In Iwatsuki K and Raven PH (Eds.). Evolution and diversification of land plants(Springer, Tokyo) pp. 269–290.
Wu C-Y. (1980) The vegetation of China(Science Press, Beijing).
Wu C-Y. (1988) Hengduan Mountain flora and her significance. Journal of Japanese Botany 63:297–311.
Wu C-Y. (1991) The areal-types of Chinese genera of seed plants. Acta Botanica YunnanicaSuppl IV, pp. 1–139.
Wu C-Y. (1993) Addenda et corrigenda ad typi arealorum generorum spermatophytorum sinicarum. Acta Botanica YunnanicaSuppl IV, pp. 141–178.
Wu C-Y and Raven PH. (1994–2003) Flora of China(Science Press, and St. Louis: Missouri Botanical Garden Press, Beijing).
Wu C-Y and Wang H-S. (1983) Physical geography of China: phytogeography (I)(Science Press, Beijing).
Ying T-S, Zhang Y-L, Boufford DE. (1993) The endemic genera of seed plants of China(Science Press, Beijing).
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C. Ezcurra, N. Baccala, and P. Wardle Floristic Relationships Among Vegetation Types of New Zealand and the Southern Andes: Similarities and Biogeographic Implications Ann. Bot., June 1, 2008; 101(9): 1401 - 1412. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||