**1. Introduction**

It is a fascinating issue for ecologists to develop a general theory or principle to interpret the mechanisms of global gradients and stabilization of biodiversity. This question has perplexed biogeographers and ecologists for about 100 years, and the diverse theories and hypotheses have been put forward to account for latitudinal gradients in biodiversity (Wright 1983; Rohde, 1992; Waide *et al*., 1999; Colwell & Lees, 2000; Gaston, 2000; Allen *et al*. 2002; Hawkins *et al*., 2003; Willig *et al*. 2003; Ricklefs, 2004; Evans & Gaston, 2005; Evans *et al*., 2005; Mittelbach *et al*. 2007; Gillooly& Allen, 2007; Storch *et al.* 2007; Cardinale, *et al*., 2009), Recent decade, the metabolic theory of biodiversity (MTB) is developed and attracting a lot of attentions of ecologists as a novel hypothesis based on metabolic theory of ecology (MTE) and the energeticequivalence rule (West *et al*. 1997, 1999; Enquist *et al*. 1998; Allen *et al*. 2002, 2007; Brown *et al*. 2004; Deng *et al*. 2006, 2008). The MTB is recognized as a general principle that can quantify relationships between the dynamic processes of population and biodiversity patterns in ecosystem, and between species richness and environmental factors (see also Allen *et al*. 2003, 2006; Allen & Gillooly 2006; Gillooly & Allen 2007). The metabolic eco-evolutionary model of biodiversity, the most recent extension of the MTB, has been developed by Stegen *et al*., (2009).

The MTB considered that species richness, *S*, in plots of fixed area, *A*, should be described by a form of equation as following <sup>k</sup> <sup>0</sup> ( )( )e *E T <sup>T</sup> S JA B B* . In this expression, *E* is the activation energy of metabolism, -0.6 to -0.7 eV, *k* is Boltzmann's constant (8.62×10-5 eV K-1, where *K* is degrees Kelvin), *T* is environmental temperature in degrees kelvin, 3 4 0 0 *B bM* , where *b*0 is a normalization constant that varies by taxonomic group, *M* is individual body size, *BT* is the total energy flux of a population per unit area, varying with taxonomic group and plot area *A*, and *J/A* is the total density of individual per unit area. Apparently, the species richness should vary as a function of abundance, body size and environmental temperature. So, when

© 2012 Deng and Zhang, licensee InTech. This is an open access chapter distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2012 Deng and Zhang, licensee InTech. This is a paper distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

abundance and size are both presumed constants across geographical space, the relationship between the natural logarithm of species richness (ln*S*) and the inverse temperature (1/k*T*) will be re-expressed by a linear equation: <sup>1</sup> ln (k ) *S ET C* . In this equation, the intercept term ( <sup>0</sup> ln[( )( )] *C B B JA <sup>T</sup>* ) incorporates the effect of mean body size of the study taxon, area and total community abundance (Allen *et al*., 2002; Gillooly & Allen, 2007).

Species Distribution Patterns, Species-Area and Species-Temperature Relationships in Eastern Asian Plants 99

Compositae 9 104 655 293.78 163.66 1.34 2.50 Poaceae 9 86 597 254.44 155.37 1.29 2.52 Rosaceae 9 49 406 197.33 122.44 0.45 -0.64 Liliaceae 9 35 164 93.22 41.32 0.11 -0.32 Labiatae 9 28 253 130.33 79.02 0.07 -1.10 Angiosperm 11 1009 7891 4363.45 2513.16 -0.02 -1.47 Gymnosperm 10 4 63 31.90 19.53 0.27 -0.73 Seed plant 11 1019 7954 4392.45 2524.83 -0.03 -1.47

MAT (°C) 11 -2.80 20.88 12.34 7.48 -0.98 0.002

Mean latitude (°) 11 23 50 35.33 9.54 0.35 -1.49 Mean longitude (°) 11 100 132 115.1 9.94 0.25 -0.44 Area (km2) 11 52000 960000 401660 273070 0.64 0.73

Compositae 71 11 324 78.66 44.51 2.55 12.34 Poaceae 70 8 131 63.13 28.75 0.40 -0.16 Rosaceae 71 2 185 54.93 33.76 1.27 2.24 Liliaceae 49 1 85 31.7 19.52 1.08 0.93 Labiatae 56 3 96 30.41 17.51 1.31 2.99 Angiosperm 255 79 3893 1186.44 705.08 0.84 0.91 Gymnosperm 234 1 110 14.21 12.33 3.32 19.32 Pteridophyte 189 1 594 108.86 92.75 1.41 3.51 Vascular plant 193 138 4543 1388.18 809.92 0.86 0.99

MAT (°C) 270 -2.8 29 13.93 5.71 -0.61 0.44

Mean latitude (°) 270 18.4 51.6 30.74 6.55 0.71 0.37 Mean longitude (°) 270 95 130.6 111.88 7.04 0.12 -0.37 Area (km2) 270 0.64 6698 490.89 922.28 3.94 18.69

**Table 1.** Summary statistics of species richness in different plant groups, climate variables, and areas of

Empirical evaluations of how well observed richness patterns fit the central predictions of the MTB are now appearing in several literature (Allen *et al*., 2002; Kaspari *et al*., 2004; Hunt *et al*., 2005; Algar *et al*., 2007; Cassemiro *et al*., 2007; Latimer, 2007; Hawkins *et al*., 2007a, b; Sanders *et al*., 2007). Although to date the observed patterns in biodiversity have been taxonomically and geographically limited (Ellison, 2007), the data sets for the detailed plant groups are relatively absent. Wang et al. (2009) showed that magnitude of temperature dependence (i.e. *E*) of tree species richness in both eastern Asia and North America increases with spatial scale at the large scale of 50×50km to 400×400km. Therefore, we conjectured that the species richness inherent dependence of spatial scale may influence on the successful tests for the predictions of the MTB (also see Zhang et al. 2011). However, it is unclear (i) how the species richness responds to temperature at the variant spatial scale, especially at

the small scale level. (ii) what spatial scale range is appropriate to the MTB.

Sample type Variable N Minimm Maximum Mean SD Skewness Kurtosis

**Floristic region Species richness**

**Nature reserve Species richness**

**Climatic variables**

**Climatic variables**

11 floristic regions and 270 nature reserve used in this paper.

**Location and area**

**Location and area**

The intense and continuous controversies for the MTB have been focusing on two primary predictions: 1) whether ln-transformed species richness is linearly associated with an inverse rescaling of ambient temperature or not, and 2) if so, whether the slope of the relationship is encompassed in the theoretical value range of -0.6 to -0.7. The proponents argue that this theory accounts for diversity gradients over a range of spatial scales from mountain slopes to continental and global gradients, and for many groups of plants and ectothermic animals (Allen *et al*. 2002, 2003;Brown *et al*., 2003, 2004; Gillooly & Allen, 2007). Kaspari *et al*.'s (2004) and Hunt *et al*.'s (2005) analyses using respectively ant communities and deep-sea communities datasets were also in favor of the Allen *et al*.'s (2002) predictions. Concomitantly, the disagreements for MTB emerged in some literature. Hawkins *et al*. (2007a) tested the predictions of this theory with 46 different data sets compiled from a variety of terrestrial plants, invertebrates, and ectothermic vertebrates, and found that the results were partly deviated from the predictions of the MTB (Allen *et al*. 2002; Brown *et al*. 2004), Accordingly, they considered that MTB was a poor predictor for the observed diversity gradients in most terrestrial system. Latimer (2007) subsequently reanalyzed some Hawkins *et al*.'s (2007a) data sets using a Bayesian approach and supported their conclusions. Algar *et al.* (2007) recently showed that the relationship between richness and temperature was actually curvilinear for several data sets in North America, and slopes varied systematically in geographical space, which were strongly consistent with Cassemiro *et al* (2007) analyzing for New World amphibians. As a consequence, they claimed that Allen *et al*.'s (2002) model did not give an adequate fit to the data.

**Figure 1.** Location of the eleven floristic regions and 270 nature reserves of China used in this study. Floristic regions was marked with Arabic numerals as following (1, Daxinganling; 2, NE China; 3, NE China plain; 4, North China region; 5, East China; 6, Lingnan region China; 7, Tsinling Mountains; 8, The region of Hengduan Mountain; 9, Central China; 10, Dian-qian-gui region; 11, The region of Yunnan Plateau)(also see Zhang et al. 2011).


abundance and size are both presumed constants across geographical space, the relationship between the natural logarithm of species richness (ln*S*) and the inverse temperature (1/k*T*) will be re-expressed by a linear equation: <sup>1</sup> ln (k ) *S ET C* . In this equation, the intercept term ( <sup>0</sup> ln[( )( )] *C B B JA <sup>T</sup>* ) incorporates the effect of mean body size of the study taxon,

The intense and continuous controversies for the MTB have been focusing on two primary predictions: 1) whether ln-transformed species richness is linearly associated with an inverse rescaling of ambient temperature or not, and 2) if so, whether the slope of the relationship is encompassed in the theoretical value range of -0.6 to -0.7. The proponents argue that this theory accounts for diversity gradients over a range of spatial scales from mountain slopes to continental and global gradients, and for many groups of plants and ectothermic animals (Allen *et al*. 2002, 2003;Brown *et al*., 2003, 2004; Gillooly & Allen, 2007). Kaspari *et al*.'s (2004) and Hunt *et al*.'s (2005) analyses using respectively ant communities and deep-sea communities datasets were also in favor of the Allen *et al*.'s (2002) predictions. Concomitantly, the disagreements for MTB emerged in some literature. Hawkins *et al*. (2007a) tested the predictions of this theory with 46 different data sets compiled from a variety of terrestrial plants, invertebrates, and ectothermic vertebrates, and found that the results were partly deviated from the predictions of the MTB (Allen *et al*. 2002; Brown *et al*. 2004), Accordingly, they considered that MTB was a poor predictor for the observed diversity gradients in most terrestrial system. Latimer (2007) subsequently reanalyzed some Hawkins *et al*.'s (2007a) data sets using a Bayesian approach and supported their conclusions. Algar *et al.* (2007) recently showed that the relationship between richness and temperature was actually curvilinear for several data sets in North America, and slopes varied systematically in geographical space, which were strongly consistent with Cassemiro *et al* (2007) analyzing for New World amphibians. As a consequence, they claimed that Allen

**Figure 1.** Location of the eleven floristic regions and 270 nature reserves of China used in this study. Floristic regions was marked with Arabic numerals as following (1, Daxinganling; 2, NE China; 3, NE China plain; 4, North China region; 5, East China; 6, Lingnan region China; 7, Tsinling Mountains; 8, The region of Hengduan Mountain; 9, Central China; 10, Dian-qian-gui region; 11, The region of

area and total community abundance (Allen *et al*., 2002; Gillooly & Allen, 2007).

*et al*.'s (2002) model did not give an adequate fit to the data.

Yunnan Plateau)(also see Zhang et al. 2011).

**Table 1.** Summary statistics of species richness in different plant groups, climate variables, and areas of 11 floristic regions and 270 nature reserve used in this paper.

Empirical evaluations of how well observed richness patterns fit the central predictions of the MTB are now appearing in several literature (Allen *et al*., 2002; Kaspari *et al*., 2004; Hunt *et al*., 2005; Algar *et al*., 2007; Cassemiro *et al*., 2007; Latimer, 2007; Hawkins *et al*., 2007a, b; Sanders *et al*., 2007). Although to date the observed patterns in biodiversity have been taxonomically and geographically limited (Ellison, 2007), the data sets for the detailed plant groups are relatively absent. Wang et al. (2009) showed that magnitude of temperature dependence (i.e. *E*) of tree species richness in both eastern Asia and North America increases with spatial scale at the large scale of 50×50km to 400×400km. Therefore, we conjectured that the species richness inherent dependence of spatial scale may influence on the successful tests for the predictions of the MTB (also see Zhang et al. 2011). However, it is unclear (i) how the species richness responds to temperature at the variant spatial scale, especially at the small scale level. (ii) what spatial scale range is appropriate to the MTB.

Here we aimed to evaluate how the relationship between species richness and temperature predicted by MTB varied with respect to sampling scales, as well as with respect to different plant taxonomic group using an extensive plant data sets including three divisions in vascular plant at two different sample scales including nature reserve grain and floristic grain.

Species Distribution Patterns, Species-Area and Species-Temperature Relationships in Eastern Asian Plants 101

not linear and rejected the entire MTB (Fig3-g; Table3). The slope values estimated by RMA regression for all taxonomic groups were significantly exclusive from the second prediction

The species-area relationships for all taxonomic divisions at both the floristic and nature reserve special scales indicated that the area size of community have more impact on the species richness for subdivision (e.g. family) than for division (Fig. 4 and 5). Moreover, the observed slope values were close to or encompass (95% CI) the theoretical values predicted by MBT at the spatial scale range of 50- 6698 km2, excluding the size of area class less than 50

**Figure 2.** The relationship between natural logarithm of species richness (ln*S*) and inverse temperature (1/k*T*) for seven groups in 11 floristic regions: two divisions (gymnosperm and angiosperm) and five

Group Figure N R2 P RMA slope(95%CI) Compositae 2-a 9 0.64 0.009 **-0.48(-0.74– -0.23)**  Poaceae 2-b 9 0.82 <0.001 **-0.55(-0.76– -0.34)**  Rosaceae 2-c 9 0.73 0.003 **-0.66(-0.97– -0.35)**  Liliaceae 2-d 9 0.67 0.007 **-0.46(-0.69– -0.22)**  Labiatae 2-e 9 0.89 <0.001 **-0.70(-0.92– -0.50)**  Angiosperm 2-f 11 0.86 <0.001 **-0.64(-0.83– -0.46)**  Gymnosperm 2-g 10 0.71 0.002 **-0.80(-1.15– -0.45)**  Seed plant 2-h 11 0.89 <0.001 **-0.64(-0.80– -0.48) Table 2.** Summary of regressions testing Model II (RMA) slopes of richness-temperature relationships for cases with linear relationship between inverse scaled temperature and ln-transformed richness in 11

families of angiosperm (Compositae, Poaceae, Rosaceae, Liliaceae and Labiatae).

of MTB (Table2).

km2 (Fig. 6; Table 4).

floristic regions.
