We synthesized the information on fish diversity in the Yangtze River Basin, documenting 378 species/subspecies that had been recorded and described from the basin, belonging to 14 orders, 32 families and 144 genera. Of these, 338 species/subspecies are freshwater fishes, of which 269 species/subspecies are from the Cypriniformes. Of the total, 11 species are migratory fishes, and 29 are species of brackish water of the estuary. A total of 162 species/subspecies are endemic to the river and 69 are threatened. We divided the Yangtze River Basin into 19 sub-basins. Except for the two sub-basins of the headwaters and the upper and middle reaches of Jinshajiang River, the other 17 sub-basins showed similar values in species richness and G-F diversity indices, although the values were higher in the upper reaches than in the middle and lower reaches. However, the proportion of endemic species decreased gradually from the headwater to the estuary of the basin in parallel with the gradient of elevation. Jaccard similarity analysis showed that the 19 sub-basins were clustered into three groups: (1) the headwaters and the upper and middle reaches of Jinshajiang River, located in the eastern Qinghai-Xizang Plateau and Hengduan Mountains; (2) the other sub-basins of the upper reaches, located in Western Sichuan Plateau, Yungui Plateau, Sichuan Basin, and Qinling-Daba Mountains; (3) the middle and lower reaches, belonging to the Huaiyang Mountains, Jiangnan Hills and the plain along the middle and lower reaches of the Yangtze River. This grouping reflects the environmental characteristics of the basins and the three large topographic plat-forms of the Chinese mainland.

Tacca integrifolia is a tropical herb with a core distribution in southeast Asia and disjunct populations in the Brahmaputra valley in China. Significant morphological differences exist between the two regions. Genetic variation within and among four populations, three from the Brahmaputra valley, and one from Malaysia, was assessed using 19 Inter-Simple Sequence Repeat (ISSR) primers. The Malaysian population was more than 2,000 km from the other populations. A total of 165 discernible loci were obtained, of which 111 were polymorphic (67.68%). Genetic diversity was relatively high at the species level (PPB = 67.68%, H_T =0.185, and H_sp = 0.292), but was low within populations (PPB = 12.81%, H_E = 0.065 and H_pop = 0.044). Extremely high levels of population genetic differentiation were detected based on Nei’s genetic diversity analysis (G_ST = 0.777) and AMOVA analysis (F_ST =0.9206). Genetic differentiation among the three Brahmaputra populations (0.28%) and within these populations (7.94%) was low (P < 0.001). Restricted gene flow (N_m = 0.1435) among populations may result from self-pollination, inefficient seed dispersal, populationisolation, and habitat fragmentation. The extremely low genetic diversity in Brahmaputra populations was probably due to the unique landforms and vicariance of the Brahmaputra valley.

Patterns of biodiversity along environmental gradients is one of the basic issues in biodiversity research. Furthermore, the relationship between species richness and aboveground biomass is one of the most controversial subjects in ecology. In this paper, we studied the patterns of species richness along environmental gradients and the relationship between species richness and aboveground biomass in two major ecosystem types on the Tibetan Plateau, alpine meadow and alpine steppe. The results are summarized as follows: (1) 267 species were recorded in 250 plots of 50 sites, 179 of which occurred in alpine meadows and 135 in alpine steppe. The number of species ranged from 3 to 32 in alpine meadow, while it ranged from 2 to 18 in alpine steppe. (2) Species richness increased with latitude and longitude, and decreased with altitude. A stepwise regression analysis of species richness and environmental factors showed that the species richness was significantly correlated with growing season precipitation and warmth index. (3) Species richness was positively correlated with aboveground biomass.

In order to understand the distribution pattern of the resources of arbuscular mycorrhizal (AM) fungi in arid zones in northwestern China, we investigated AM fungal diversity in five vegetation types: desert, grassland, forest, meadow, and oasis farmland. Forty AM fungal species belonging to six genera were identified from the rhizospheres of 60 plant species. Of these fungi, 35 were identified to species and five were identified to genus. Glomus claroideum was the dominant species throughout the sampling area, but the occurrence and distribution pattern of AM fungi in the five vegetation types were quite different. Entrophospora was only observed in desert while Paraglomus was found in oasis farmland and desert, Archaeospora existed in grassland, desert and oasis farmland, and Scutellospora existed in meadow, forest and grassland. The biodiversity index in oasis farmland and meadow were higher than other vegetation types, indicating soil moisture may be the most important factor affecting the distribution of AM fungi in arid zones.

The study of the relationship between biodiversity and ecosystem functioning (BEF) is a hot topic in the field of terrestrial ecosystem ecology, and is of great significance for the efficient use and management of ecosystems. Furthermore, it plays an important role in the restoration of degraded ecosystems and biodiversity conservation. Alpine grassland is the main ecosystem type found in the Qinghai-Tibet Plateau. In recent years, progress has been made on species diversity and ecosystem functioning and their mutual relationship in alpine grasslands. This paper analyzes existing problems in the research of grassland biodiversity and ecosystem functioning in terms of the study of underlying ecological processes and the impacts on ecosystem multi-functionality under global change. The effects of different grassland types, grassland degradation, grazing disturbance, simulated climate change, mowing, fertilization, enclosure, and replanting on the relationship between biodiversity and ecosystem functioning in alpine grasslands are also thoroughly reviewed in this paper. Moreover, deficiencies and future research directions of alpine grassland BEF are identified: carrying on the BEF research of alpine grasslands based on the functional diversity of species, comprehensively considering the effects of abiotic factors such as resource supply levels, disturbance intensity and scale, and environmental fluctuation on the relationship between species diversity and ecosystem function, and paying attention to the effect of scale and element coupling on BEF research of alpine grasslands under global climate change. Finally, based on research progress and conclusions of BEF in alpine grasslands, we put forward suggestions to improve the utilization rate of alpine grassland resources and biodiversity conservation, including strengthening grazing management, protecting biodiversity, improving governance of degraded grasslands, maintaining biodiversity function, strengthening innovation and protection concepts and enhancing ecosystem functioning that has been seriously weakened by climate change and human disturbance.

The relationship between plant species diversity and primary productivity has long been a hot topic in biodiversity research. We examined the relationship between species richness and productivity in four typical grasslands of northern China at different spatial scales. At the community scale, a positive correlation was found for six of seven communities, and a unimodal pattern was found only for one community (Stipa glareosa community). At large scale (vegetation type or landscape/region), a significant positive relationship was found. Species richness ranged from 4 to 35 species, and community above-ground productivity from 13 to 368 g·m^–2·yr^–1. The highest species richness and above-ground productivity were found in the alpine meadow community, followed by meadow steppe, typical steppe and desert steppe.

Pollinator specificity has traditionally been considered as the main mechanism of reproductive isolation in orchids. Cypripedium smithii and C. tibeticum are two closely related species with similar characters and sympatric distributions. With the knowledge of reproductive biology of C. tibeticum, we studied the reproductive biology of C. smithii and compared the morphological characters between C. tibeticum and C. smithii. The results showed that a series of intermediate types occurred between typical C. smithiiand typical C. tibeticum, indicating that no distinct taxonomic boundary existed between C. tibeticum and C. smithii. Pollination experiment showed that both C. smithii and C. tibeticum were pollinated by bumble queens. The hand-pollination experiments indicated that C. tibeticum and C. smithii were cross-fertilized, so that they could hybridize in the natural conditions and produced transitional types. It is better to combine them as one identical species rather than separate them as two distinct species.

The genetic diversity of ten populations of Sophora moorcroftiana in the middle reaches of the Yarlung Zangbo River, Tibet, China was assessed using allozyme analysis by horizontal sliceable starch gel electrophoresis. Twenty four loci (46 alleles) of 13 enzyme systems demonstrate low levels of genetic variation within populations, with the value of P_p＝25.0%～37.5%,A_p＝1.3～1.7 and H_ep=0.112～0.169. At the species level, the genetic diversity of S. moocroftiana (P_s=37.5%, A_s=1.9, H_es=0.171) was lower than the mean value of angiosperms of long lived woody species (P_s=59.5%,A_s=2.10,H_es=0.183). Wright′s F statistics analysis indicated that F_IS, a measure of the deviation from random mating within the 10 populations, was -0.0071, suggesting deviation from Hardy-Weinberg equilibrium and a slight heteroygote excess in some populations. The higher level of differentiation (F_ST＝0.1748) among populations than those of other long lived woody plants may result from habitat fragmentation and low levels of gene flow (N_m=1.1802) caused by environmental deterioration and human disturbance, including over felling and over-grazing. It was worth noting that populations H2 (Xietongmen), H31 (Jiangdang1), H32 (Jiangdang2), and H5 (Langsailing) harbored the majority of alleles and had high levels of genetic diversity, suggesting that these populations in particular should be conserved in situ.

Lichens are present in a wide range of habitats throughout the world. Due to their unique biodiversity, we explored the species composition of floor lichens under different vegetation types in Two-river Source Nature Reserve in Altay Mountains, Xinjiang. The sampling sites covered six vegetation types including mountain desert, mountain grassland, mixed coniferous and broadleaved forest, coniferous forest, subalpine meadow, and alpine meadow. The results showed that the lichen flora in this area is rich and complex. A total of 46 floor lichen species belonging to 6 genera of 5 families were recorded. Of these, Cladoniaceae with 32 species is the dominant family. The species similarity index between the mixed coniferous and broadleaved forest and the coniferous forest was the highest of all pairs (0.739), whereas that between the mixed coniferous and broadleaved forest and the alpine meadow belt was the lowest (0.200). Mountain forests, having the maximum species diversity, were the distribution center for lichen diversity and key areas for their conservation in Two-river Source Nature Reserve in Altay Mountains, Xinjiang. We also discuss conservation strategies for lichens.

In order to probe into the dynamics of butterfly species and quantities in different habitats at high elevation, from 1993 to 1997, we surveyed butterfly diversity in four habitat types at the Huzhu Northern Mountain National Forest Park, Qinghai. We selected a plain among valleys, a temperate steppe, a mixed coniferous and broad-leaved forest, and an alpine shrub meadow as our study sites. We trapped 4,745 butterflies, belonging to eight families, 57 genera and 86 species. The genera with the most abundant species included Colias (6 species), Pieris (5 species), Aporia (4 species), Pontia (3 species), and Melitaea (3 species). The most individual number occurred in the families Pieridae and Lycaenidae. In terms of butterfly species composition and quantities, significant differences existed among different habitats. The number of butterfly species was the largest in the mixed coniferous and broad-leaved forest, while the individual number peaked in the plain among valleys. The species diversity of butterflies was highest in the mixed coniferous and broad-leaved forest and the lowest in alpine shrub meadow, while the evenness index was highest in mixed coniferous and broad-leaved forest and lowest in the plain among valleys. Dominance was the highest in the alpine shrub meadow. The mixed coniferous and broad-leaved forest had the most endemic genera/species (20 genera, 24 species), followed by the alpine shrub meadow (8 genera, 8 species), the plain among valleys (6 genera, 7 species), and the temperate steppe (5 genera, 6 species). The similarity coefficients of butterfly species among different habitats ranged from 0 to 0.4259. Using cluster analysis to identify the species composition of butterflies, we discovered that the plain among valleys had the highest similarity with the temper-ate steppe, and secondarily with the mixed coniferous and broad-leaved forest, and third with the alpine shrub meadow. In order to protect the butterfly diversity of this area, we suggest establishing a natural protected zone to better balance tourism and biodiversity protection.

To explore the altitudinal differentiation of flora of Mt. Gongga and its relationship with those of surrounding areas, we investigated eight vertical vegetation zones on the eastern slope of Mt. Gongga, and analyzed the altitudinal distribution, the floristic components at the level of family, genus and species, and the contrast between different vegetation zones. Major results are as follows： (1) temperate flora primarily dominates the vegetation along the altitudinal gradient, but tropical and temperate components are equally abundant at the arid valley bottom area. The composition and altitudinal distribution of the tropical elements indicate the remnant effects of the Paleotropical and PaleoMediterranean flora; (2) East Asia and East Asia-North America arealtypes have the greatest influence on the forest flora in the middle part of Mt. Gongga, which are chiefly ancient temperate components; (3) North Temperate arealtype is the main body of the flora of Mt. Gongga, and comprises the bulk of the advanced and endemic components in the flora since the rise of the Tibet Plateau. It represents the young components of the flora, and (4) a wide range of Chinese endemic species comprise about 40%－65% of the flora of different vertical vegetation zones, increasing in dominance at higher elevations. At middle elevations, the flora of Mt. Gongga most closely relate to the flora of central China; at the alpine zone, the connection between Mt. Gongga and Tibet Plateau and East Himalayan Mountains takes precedence. The biogeographical significance of Gongga Mountain and the properties of the flora endemism at Mt. Gongga and Hengduan Mountain Range are discussed.

Understanding macro-scale spatial patterns in species diversity and their underlying mechanisms is central to macroecology and biogeography. In this study, we explored geographic patterns of species richness and their environmental determinants for overall terrestrial mammals and each major mammalian order in China, using datasets of species distribution, climate, topography and vegetation. Species richness of terrestrial mammals exhibited significant latitudinal gradients, decreasing from south to north. High species richness generally occurred in tropical and subtropical mountains, whereas low species richness was found in the eastern plains, the arid areas of northwest regions, and central areas of the Qinghai-Tibet Plateau. Geographic patterns of species richness varied among mammalian orders. The best model, which included a remote sensing-based vegetation index (NDVI), number of ecosystems, and annual range of temperature, accounted for 66.2% of variation in overall mammal species richness, with NDVI being the most important determinant. This suggests that patterns of mammal richness in China are governed by the integrated effects of different environmental predictors, with vegetation productivity playing a major role. The best models for various orders of mammals identified different combinations of determinants, possibly reflecting differences in evolutionary history and physiological tolerances.

The relationship between species diversity and ecosystem function is one of the core problem of biodiversity research, and productivity is an important modality of ecosystem function. The species diversity of natural communities is often strongly related to their productivity. The grassland is an important component of terrestrial ecosystems and plays a significant role in the biodiversity and ecosystem function. Understanding of how to use natural vegetation to restore deteriorated grasslands is critical for protection of grassland ecosystems. Moreover, the relationship between species richness and productivity relies on scales. At some scales, productivity affects diversity, and at other scales, species diversity also affects productivity. Thus, it is necessary to know the relationship between plant functional diversity and productivity in different grassland types. In 2003 the differences in plant functional diversity, functional composition and productivity in different grassland types of alpine meadow were analyzed. Humans are modifying both the identities and the numbers of species in ecosystems. However, the impacts of such changes on ecosystem processes are still controversial. The effects of functional diversity and composition on plant community productivity were analyzed in four types of alpine meadows. The research results show:(1) the effects of the legume functional group on productivity was greater than other functional groups in Kobresia humlis meadow and Potentilla froticosa shrub. (2) Although the biomass of forbs, C3 plants and sedge functional groups account for higher proportion of community productivity in Kobresia pygmaca meadow and K. tibetica swamp meadow, differences were not significantly different, indicating that the factors influencing community productivity were not only species diversity, but also the identity of species present and environmental resources. These factors were in turn mainly affected by species diversity and evenness among functional groups. That is, differences in functional group composition have a larger effect on ecosystem processes than functional group richness alone. (3) A significant correlation was found between the coverage per functional group and the primary productivity of communities in the four types of alpine meadows. (4) Changes in productivity with number of species per functional group showed a unimodal pattern, such that the maximum community productivity occured at an intermediate level of number of species per functional group.

Testacea (or testate amoebae, thecamoebians) are free-living amoeboid protozoa inhabiting a shell or test and they play an important role in material cycle and energy flow in terrestrial and aquatic ecosystems. The fauna of Testacea was analyzed from three polar regions of the Earth (Arctic, Antarctic and Tibet). In total, 315 species from 62 genera were recorded in the polar regions, i.e. 232 species (51 genera) in the Arctic, 131 species (30 genera) in the Antarctic, and 173 species (42 genera) in Tibet. In each polar region, the most diverse genera were Arcella, Centropyxis,Difflugia, Euglypha, Nebela; they accounted for 51.3%, 63.4% and 60.1% of the total species number in the Arctic, Antarctic and Tibet, respectively. Seventy-three species (23.2% of all species) and twenty-four genera (38.7% of all genera) were common to the three polar regions. One hundred and sixty-seven species (53.0% of species) were found only in one of the polar regions. Both Assulina muscorum and Centropyxis aerophila were widely distributed with the highest frequency (90%) in 40 subregions from the Arctic, Antarctic and Tibet. Cluster analysis revealed that the highest species-level similarity of Testacea was between the Arctic and Tibet (56.3%). Further, species similarity was the highest between the Arctic and Tibet based on Arcellinida species data, but the lowest based on filose Testacea species. Distinct differences in Testacea fauna indicate that some species are not ubiquitously distributed in spite of better passive long-distance dispersal than macro-organisms. Geographic distributional patterns of Testacea diversity are closely related to body size, habitat type and historical events, and our perception of these patterns are strongly influenced by taxonomic resolution (morphological criteria), sampling effort and spatial scales. We propose that study of genetic diversity among and within common Testacea morphospecies in relation to ecological and historical factors will elucidate geographic distributional patterns within this interesting group.

The Tibetan gazelle (Procapra picticaudata) is an endangered species inhabiting theQinghai-Tibetan Plateau. From October 2002 to December 2003, we observed 924 groups (3643 individuals) of Tibetan gazelles in Hoh Xil region. These groups could be classified into five types.Female groups (525) were the major type, accounting for 56.8%. Solitary gazelle (a special group type), male groups, mother-calf groups and mixed groups made up 26.1%, 11.0%, 3.0% and 3.0%, respectively. Group size ranged from one to 17 individuals. Groups of 2-10 individuals, solitary gazelle groups and larger groups (>10 individuals) accounted for 70.0%, 26.1% and 3.9%, respectively. 72.9% of gazelles were observed in group with sizes of 2-8 individuals, while 20.5% were observed in group sizes > 8 and 6.6% were solitary. A chi-square test showed that the frequencies of individuals in different group sizes were significantly different (P<0.001). Group sizes varied significantly among different group types (Kruskal-Wallis H test,P<0.001). There were significantly more males (68.1%) than females (P<0.001) in solitary groups. To sum up, female group was the main group type of Tibetan gazelles, and the optimal group size was 2-8 individuals.

The plateau pika (Ochotona curzoniae) is a keystone species in alpine meadow ecosystems on the Tibetan Plateau. Therefore, ecological studies of the relationship between plateau pika habitat use and the stability of alpine meadows are important. However, little information is available about large-scale spatial relationships between plateau pika habitat use and vegetation distribution. Therefore, we studied this topic using Geographic Information System (GIS) analyses. The number and position of active plateau pika burrows were sampled along seven line transects totaling 2,423 sampling units, 19 km in length, in a 44 km² study area in Yunbo Gou, Shiqu County, Sichuan Province during September 2008. Historically, domestic livestock grazing and other human activities were rare in this area. A no-grazing policy has been in effect since 2006, thereby removing livestock competition in the study area. Normalized Difference Vegetation Index (NDVI) values within the study area were extracted from a CBERS-02 remote sensing image taken September 18th, 2008. A digital elevation model (DEM) of the study area was established based on a digitized topographical map of Shiqu County, and slope values within the study area were subsequently extracted using the DEM. NDVI and slope values were compared with the number of plateau pika burrows in each sampling unit. A Bonferroni z test revealed that numbers of pika burrows were lower than expected in units with low NDVI values (NDVI of 0.1-0.3), not different than expected in middle NDVI groups (0.3-0.4), and higher than expected in high NDVI groups (0.4-0.6). A linear regression model found a significant positive correlation between NDVI value and pika burrow count (y = 16.50 x + 0.87; R² = 0.78, P < 0.001). In contrast, no linear relationship was detected between slope and pika burrow number (linear regression, R²= 0.04, P = 0.441). This study showed that plateau pikas use areas with higher vegetation coverage more than areas with lower vegetation coverage in the absence of livestock competition. Our results lend some support to the conclusion that the pika’s natural use of vegetation does not accelerate the destruction of the alpine meadow vegetation in this region.

Gahai Salt Lake, which is situated in the Qaidam Basin of the Qinghai-Tibet Plateau at an elevation of 2849.6 m above sea level, is a saline lake with high salinity and abundant biological resources of Artemia (brine shrimp). An investigation on the characteristics of community diversity of plankton with relation to salinity was conducted in the summer of 1997. Forty-six species of phytoplankton and twelve species of zooplankton have been identified in the major body of water in the in-lake area with 22 sampling sites at 5 sections. The differences of the species composition, richness , individual abundance and biomass of plankton were compared between two sampling area with a big change of salinity in the waters. The results show that the salinity is an important limiting factor in determining the structure of biological communities in the inland saline lakes.

Spruce forest is a typical alpine forest type near timberline in the eastern Tibetan Plateau. However, little is known about this ecosystem. In the present study we investigated community structure and vascular plant composition of such forests in northern Zamtang County, Sichuan Province, China. We sampled six plots (40 m×50 m) in primary forests in a range of 3,700-3,800 m elevation. For all trees ≥ 3 m in height in each plot, height and diameter at 50 cm height were measured. A total of 180 subplots (1 m×1 m), 30 for each plot, were sampled to estimate the coverage of shrubs, herbs and bryophytes, and species composition of shrubs and herbs. Four sub-layers could be recognized within the canopy, suggesting that the forest is uneven-aged and multi-storied. Structure analysis of tree height and diameter showed that the arborous layer was stable and maintained a certain regenerative ability. Shrubs showed a patchy spatial pattern with simple structure and low richness. Compared to low coverage (15.4±20.9%) of shrubs, coverage of herbs and bryophytes (49.4±21.9% and 36.6±21.2%, respectively) were higher, forming obvious herb and bryophyte layers. A total of 124 vascular species (109 herbs and 15 shrubs) were recorded in all six plots, belonging to 68 genera and 30 families. Temperate areal-type genera and hemicryptophyte species were dominant. We concluded that the forest was a moss-spruce forest type. Although the species composition and richness were quite similar to other forests in Tibetan Plateau, community structure differed significantly. Our results provide insights into management and biodiversity conservation in forests near the timberline in the eastern Tibetan Plateau.

The arid regions of China include Mongolian- Xinjiang warm-temperate arid region and Qinghai-Xizang frigid arid region, located in western China, host 120 species of small mammals including Largomorpha and Rodentia. We collected all of the available distributional data for these species in these regions. Based on GIS models of species range which were predicted by a wildlife-habitat model, we researched the distribution patterns of rodent diversity there. A total of 766 equal-area quadrangle grids that each had an area of 6,470 square kilometers were used as geographic units to collect distributional data of the species, supported by MapInfo Professional Version 4.0. Based on these grid data we obtained the number of genera and families, and calculated the small mammal species diversity of genera and families by Shannon-Winner index and G-F index. In order to explain the species differentiation tendency of small mammals, we introduced number ratios of species to genera and of genera to families as differentiation indices. The results indicate that more genera occurred in mountains around the Junggar Basin, Qilian Mountains, Helan Mountains and the west slopes of Da Hinggan Mountains, where species richness was higher, as might be explained by an edge effect. In contrast, richness was least on the Ordos Plateau and Zhangbei Plateau. There were more families on the south slopes of the Altay Mountains, Tianshan Mountains of Ili and East Qilian Mountains, but fewer on the Ordos Plateau and Zhangbei Plateau. High species diversity often occurred in ecotones such as those among or between frigid desert and warm-temperate desert, desert and mountains, plateaus and basins, desert and steppes, and mountains and steppes. In the Qinghai-Xizang Plateau, number ratios of genera to families were much lower than in the Mongolian-Xinjiang region. In contrast,, there were higher ratios of species to genera in the former region. The G-F index was higher in Qinghai-Xizang Plateau, where a great species differentiation was implied. Since small mammals can be used as indicators of biodiversity, the distributional patterns of their species diversity could be helpful for assessment of biodiversity in arid regions of West China.

A total of 13 plots were set along a transect from the altitude of 5176 m to 5390 m in Pushila, Everest-Choyu region, Tibet, to investigate community structure of alpine sparse vegetation and effects of micro-topography. A total of 80 higher plant species, belonging to 47 genera, were identified. The most prevalent species were Kobresia pygmaea, Lagotis humilis, Potentilla cuneata, Pedicularis oederi var. sinensis, Potentilla polyschista, and Arenaria densissima. Several genera were highly species rich, such as Saxifraga, Saussurea, Kobresia, Potentilla, Gentiana, and Draba. As altitude decreased, richness of both species and genera tended to increase. Vegetation coverage was found to be a good indicator for habitat conditions. Cluster analysis supported this finding. With increasing vegetation coverage, transformed species richness (TSR) and transformed genus richness (TGR) increased, while Shannon-Wiener index decreased. Four habitat types that reflected a good correlation between community structure and micro-topography were identified based on CCA analysis using species matrix and five variables (altitude, slope, aspect, soil depth, and vegetation coverage).

Bumblebees are important pollinators of many wild flowers and crops and play a significant role in maintaining natural and agricultural ecosystems. The varied geomorphology and vegetation of China makes it the greatest hotspot of bumblebee diversity in the world. However, the bumblebee fauna of China has been insufficiently studied. Here, we report the results of systematic field surveys and the application of bumblebees to pollination over the last two decades in China. The results showed the following: (1) More than 50,000 bumblebee specimens were collected during 2002-2017. The taxonomic status of some difficult taxa was revealed by integrating morphology with DNA barcoding. A total of 125 bumblebee species have been identified, which represents 50% of the total number of bumblebee species worldwide. (2) We report the first compiled list of the bumblebee species of China, which includes 22 species that are endemic to China. The transitional zone from the eastern Qinghai-Tibetan Plateau to the Loess Plateau, the Qinling Mountains, and the Sichuan Basin is the centre of bumblebee diversity worldwide. (3) Six native bumblebee species, including Bombus lucorum, B. patagiatus, B. ignitus, B. pyrosoma, B. picipes and B. lantschouensis from Northern China, were selected for rearing between 1998-2017. Furthermore, B. patagiatus and B. lantschouensis, which had traits that favoured domestication, are now used as pollinators of crops in greenhouses within China. We propose a multi-pronged strategy to conserve the native bumblebees of China, which includes protecting their habitats and food resources and controlling invasive alien species and pesticide use. We hope that this study will help inform the conservation and the sustainable use of wild pollinators across the globe, but especially bumblebees of China.

Sclater′s monal (Lophophorus sclateri) is an endangered species in the Galliformes, with restricted distribution in Tibet and Yunnan. In Yunnan, the bird formerly has only been recorded at three sites. A survey on the distribution and habitat of this species was undertaken from October 1999 to August 2003, resulting in confirmation of the present occurrence of the species in eleven localities in seven counties in Yunnan, together with an additional possible three localities to be further confirmed. Sclater′s monals were most frequently observed in three types of habitats: alpine meadow, alpine-subalpine bamboo, and subalpine coniferous forestry. Threats to the survival of this pheasant arise from the low carrying capacity of the habitat, low growth rate of the population, human disturbance, and low public awareness. Based on this study, we recommend pilot programmes of captive breeding, regulation of poaching, and encouragement of basic biological research for effective protection.

China’s mammalian fauna is unique: it possesses the endemic species and genera of the Qinghai-Tibetan Plateau; it has the relic families, Ailuropodidae and Lipotidae, of the Tertiary Period; more than half of the species of the Ochotonidae in the country are endemic; furthermore China is also the country with world most abundant ungulate diversity. In the new millennia, changes to the taxonomy of mammals of world were proposed. New mammalian species and new mammals were reported in China. Thus it is necessary to renew the inventory of China’s mammal diversity. During the compiling of The Red List of Endangered Species in China: Mammals, we collected the available data on China’s mammals (including those of Taiwan) and formed The Checklist of China’s Mammal Species 2015. For taxonomy of orders, we followed the existing theory using molecular phylogeny whereas for the classification of families and the taxa below family level, we followed the A Complete Checklist of Mammal Species and Subspecies in China: A Taxonomic and Geographic Reference by Wang (2003) and Mammal Species of the World: A Taxonomic and Geographic Reference by Wilson & Reeder (2005). We also adopted Ungulate Taxonomy by Groves & Grubb (2011) to classify the ungulates in China. The initial list included species recorded in taxonomic books and field guides, new species and new records in literature before March 31, 2015. After five evaluation meetings and two rounds of evaluation by correspondence, we finalized The Checklist of China’s Mammal Species 2015, which includes 12 orders, 55 families, 245 genera and 673 species. Among those species, 18 new species (including 11 discovered with molecular biology methods, the validity of the species are still in dispute among zoologists), 18 new species records in country, and 60 subspecies were elevated to the status of species were added to the checklist. According to literature and expertise, we deleted 52 species which are not found in China. Of them, specimens of some species were collected only once, and then never been found again during following surveys. Taxonomists hold different opinions on 20 Glires, and those species need further investigation. Compared with the data of IUCN Red List (2014), the number of mammal species in China recorded in the study is surpassed that of Indonisia (670), China became the country with the most abundant mammal species in the world. The mammals in China accounted for 12.3% of the world total numbers of mammalian species. There are 150 endemic mammals in China, which accounted for 22.3% of the national total. 43% of the species in Lagomorpha are endemics to the country. The endemic ratio of Ochotonidae is even higher, 52% of the total. Endemic ratio of Eulipotyphla is 35%. About one fifth of Primates, Chiroptera and Rodentia in China are also endemics, including 12 new bat species discovered in the country during recent decade. The Checklist of China’s Mammal Species 2015 provides the most updated baseline information for biodiversity research and conservation.

The impacts of grassland fencing on rangeland plant communities during the plant-growing season were studied in the Ketu area, Haiyan County, Qinghai Province, China in 2001. The Ketu area is one of the main distribution ranges of a rare and critically endangered gazelle, the Przewalski gazelle ( Procapra przewalskii ). Grassland fences were built for the purpose of vegetation restoration in the desert in 1999. Domestic animals at high density grazed outside the fenced range, while the Przewalski's gazelle at low density grazed inside the fenced ranges. Above ground biomass, plant height, plant cover, and plant species diversity of the plant communities outside and inside the fenced ranges were randomly sampled and analyzed in the laboratory to determine the impact of different grazing rates outside and inside fenced ranges. (1) The above-ground biomass outside the fences in June was significantly lower than that inside the fences (p <0.05), however, the above ground biomass outside the fences in both July and August were significantly higher than those inside the fences ( p <0.05). (2) The range of plant heights outside the fences (mean: 11.2～37.0 cm) was broader than that inside the fences (mean: 20.6 ～28.0 cm). Plant height outside the fences in June was significantly lower than that inside the fences ( df= 39, p <0.05). There was no significant difference between plant heights outside and inside the fences ( df =39, p >0.05). (3) The plant coverage during both June and July were significantly lower outside the fences than inside the fences (June: df = 19, p<0.05; July: df = 19, p < 0.05), but during August plant coverage outside the fences was similar to that inside the fences ( df =19,p > 0.05). (4) The animal edible herbage outside the fences was significantly lower than that inside the fences ( df = 6, p<0.05). The percentage of Chinese Stellera among total plants outside the fences was significantly higher than that inside the fences ( df =6, p<0.05). (5) The biodiversity index of the plant community outside the fences was significantly higher than that inside the fences ( df=6, p <0.05). Based on the above results, we discuss the impacts of fencing on the range community and test the hypothesis that grazing affects the productivity of the plant community. We suggest solutions for range, endangered species and integrated biodiversity management.

We studied the availability of cereal grains as food resources for win tering Black-necked Cranes in south-central Tibet Autonomous Region, People's Republic of China during January 1996. Barley, spring wheat, and winter wheat were the three major crops. Fecal samples revealed that Black necked cranes fed extensively on waste cereal grain, especially wheat. Winter wheat seedlings comprised only a minor portion of the crane diet. Post-harvest farming practices including plowing and livestock grazing influenced surface residue cover (seeds, leaves, stems, or roots) and waste cereal grain (kernels, spikelets, or chaffy bracts) available for cranes. Plowing buried most crop residue. In unplowed fields, proportion of waste grain residue cover was similar in harvested barley and harvested wheat fields. Harvested wheat produced the greatest density of surface waste grain. We recommend providing unplowed, harvested wheat and harvested barley fields near rivers and cultivating winter wheat closer to primary roads in order to maximize grain availability and minimize potential disturbance to wintering Black necked Cranes.

The movements of four species of breeding waterbirds were studied using color marking, radio-tracking, and satellite-tracking at Qinghai Lake between April and September, 2006. Forty five bar-headed geese (Anser indicus) were captured with foot traps, including six individuals tagged with radio transmitters in April and another six birds with satellite transmitters in July. A total of 104 brown-headed gulls (Larus brunnicephalus) were captured with hand nets, and six of these were radio-tagged in April. Fifty one great black-headed gulls (L. ichthyaetus) were captured with foot traps, including two birds that were radio-tagged in April. Seventy five great cormorants (Phalacrocorax carbo) were captured using the spotlight method, including six birds radio-tagged in May and June, and four individuals tagged with satellite transmitters in August. Data showed three distinct movement routes for the bar-headed goose, one for the brown-headed gull, four for the great black-headed gull, and two for the great cormorant. Furthermore, one of these routes was shared by all four waterbird species—from Luci Island and Egg Island to Quanwan along the shore near the Buhahekou and Tiebuqiahekou. These areas are also important as foraging and resting sites for many other waterbird species during migration periods.

Nectar guides are special floral structures that are assumed to attract and/or direct pollinators towards a concealed nectar reward. Nectar guides are commonly found in many angiosperm plants, especially entomophilous flowers, but their functional significance and impacts on the reproductive success of plants are poorly understood. For example, Delphinium caeruleum of the family Ranunculaceae possesses two barbate staminodes, between which is the nectar entrance. To test the effects of these barbs on male and female fitness, we experimentally investigated the effects of barb removal on seed production and pollen exportation of D. caeruleum at the Haibei Station of Qinghai Province. Our results showed that flowers of D. caeruleum were dichogamous and herkogamous, which prevented autonomous selfing in this species. The mating system of D. caeruleum belongs to obligate xenogamy and thus pollinators are necessary for the successful pollination. Compared to control flowers, the number of pollen grains exported was reduced significantly after removing the barbs, but fruit set and seed number was not affected. Our primary results indicate that the barbs on the staminode act as nectar guides, which improve the male fitness of D. caeruleum, but have little effect on female fitness. Our results suggest that selection through male fitness might have played an important role in the floral evolution of D. caeruleum at our study site. These results need to be further investigated at a larger number of sites to evaluate selection mechanisms on floral traits of D. caeruleum.

Qinghai Lake is the largest inland saline lake in China, but the diversity of halophiles within the lake remains unknown. We isolated 35 bacterial strains from different water samples in Qinghai Lake using an Oesterhelt-Stoeckenius medium (OSM). The majority of halophiles in this lake were moderate halophiles (grew at 0.4 to 3.5 M NaCl), accounting for about 62.9% (22 strains) of total bacteria, followed by the slight halophiles (22.9%, 8 strains) which could grow at 0 to 1.0 M NaCl, whereas the halotolerant (grew at 0 to more than 1.0 M NaCl) and nonhalophile (grew at 0 to 1.0 M NaCl) represented 11.4% (4 strains) and 2.9% (1 strain) of total bacteria, respectively. We investigated and analyzed the population diversity and evolutionary relationship based on the 16S rDNA sequences. A total of 35 bacterial strains were sequenced, and homology analysis showed that a majority was affiliated with γ-Proteobacteria and Bacilli, which accounted for 68.6% (24 strains) and 17.1% (6 strains) of total bacteria, respectively, whereas Actinobacteridae (3 strains), α-Proteobacteria (1 strain) and Eurotiomycetidae (1 strain) represented a small portion of total bacteria. Based on phylogenetics, 35 strains belonged to 14 different genera. Ten strains of Halomonas in the Oceanospirillales were the most dominant species, whereas 4 strains of Marinomonas represented a minor species component. Overall, Halomonas was the dominant group of moderate halophile, which can likely be attributed to its ability to adapt to relatively low salinity environments.

This paper reported 106 species of butterflies blonging to 52 genera , seven families collected in south slope of Altai Mountains in northwest Xinjiang , China. Of which , the family Papilionidae has 1 species/ 1 genus , Parnassiidae 6/ 1 , Pieridae 22/ 8 genera , Satyidae 23/ 16 , Nymphalidae 32/ 16 , Lycaenidae 15/ 12 and Hesperiidae 10/ 7. According to their distribution characteristics , four vertical zones were divided : (1) half2desert in low relief terrain ; (2) mountain bush2herb zone ; (3) mountain prairie2timber zone ; (4) subalpine steppe2 meadow steppe2alpine vegetation with cushion plants.

The faunal composition of butterflies in different habitat types in the Qilian area of Qinghai Province , northwest China , was surveyed f rom 1997 to 1999. Five habitat types were classified based on altitude , climate , soil and vegetation differences : hilly cultivated land , mountain steppe , forest steppe , alpine shrub meadow and bare rock. Of the total collection of 4 ,367 specimens , 53 species of 35 genera in six families were identified. Species richness , species composition similarity (J accard index) and species diversity index ( Shannon-Wiener index , H′) were calculated for the five habitat types. The species richness increased in the following sequence : bare rock (6 spp. ) , mountain steppe (13 spp. ) , forest steppe (14 spp. ) , alpine shrub meadow (22 spp. ) and hilly cultivated land (23 spp. ) . In cont rast , the species diversity increased in a different sequence : bare rock ( H′= 1.4629) , forest steppe ( H′= 2.2398) , mountain steppe ( H′= 2.2676) , hilly cultivated land ( H′= 2.7071) and alpine shrub meadow ( H′= 2.7734) . The butterfly species composition between forest steppe and hilly cultivated land had the highest similarity (J accard index = 0.3704) , with the second highest similarity between mountain steppe and alpine shrub meadow (J accard index = 0.2500) , and the lowest similarity between bare rock and other habitat types.

We measured floral traits of Trollius ranunculoides at different flowering time, observed the taxa of pollinators and pollinator visiting frequencies, and collected the seeds in the end of growing season to estimate pollinator preferences and the overall phenotypic selection gradient. We estimated phenotypic selection on floral traits at different flowering time using female fitness (seed number). We found significant differences of floral traits among different flowering time, with the corresponding preference, taxa, and visiting frequencies of pollinators also different. During early flowering time, visiting frequencies of bees were higher, flower calxy and petal size were wider, and total stalk length was shorter on flowers that bees preferred; however, there was no significant flower preference by different fly species. Furthermore, our estimates of phenotypic selection through female fitness differed at variable flowering time. The results of our study indicate that divergent flowering time and corresponding variation in pollinators might lead to different phenotypes, ultimately resulting in the divergence of floral traits. However, pollinator preference was dissimilar from phenotypic selection estimated by female fitness due to reasons not sufficiently evaluated in this study.

Large carnivores play an important role in the regulation of food-web structure and ecosystem functioning. However, large carnivores face serious threats that have caused declines in their populations and geographic ranges due to habitat loss and degradation, hunting, human disturbance and pathogen transmission. Conservation of large carnivore species richness and population size has become a pressing issue and an important research focus of conservation biology. The western Sichuan Plateau, located at the intersection of the mountains of southwest China and the eastern margin of the Tibetan Plateau, is a global biodiversity hotspot and has high carnivore species richness. However, increasing human activities may exacerbate the destruction of local flora and fauna, thereby threatening the survival of wild carnivores. Information on species composition and dietary habits can improve our understanding of the structure and function of the ecosystem and food-web relationships in the study area. In addition, species composition and dietary habits are of great significance for understanding multi-species coexistence mechanisms and preserving biodiversity. This study collected carnivore fecal samples from Xinlong and Shiqu counties in the Ganzi Tibetan Autonomous Prefecture, Sichuan Province. DNA was then extracted from the samples and the species was identified based on DNA sequences and DNA barcoding techniques. Seven carnivores were identified, including five large carnivores (Canis lupus, Ursus arctos, Panthera pardus, P. uncia and Canis lupus familiaris) and two medium and small-sized carnivores (Prionailurus bengalensis and Vulpes vulpes). Using fecal DNA, high-throughput sequencing and metabarcoding, we conducted diet analysis for the seven carnivores and found 28 different food molecular operational taxonomic units (MOTUs), including 19 mammals, eight birds and one fish species. The predominant prey categories of wolves, dogs and brown bears were ungulates. The domestic yak (Bos grunniens) was the most frequently identified prey species. Small mammals such as rodents and lagomorphs accounted for a significant proportion in the diets of leopard cats and red foxes, The most frequent prey of this category of carnivore were the Chinese scrub vole (Neodon irene) and plateau pika (Ochotona curzoniae). In addition, leopards and snow leopards mainly fed on the Chinese goral (Naemorhedus griseus) and blue sheep (Pseudois nayaur), respectively. Our study highlights the utility of fecal DNA and metabarcoding technique in fast carnivore surveys and high-throughput diet analysis, and provides a technical reference and guidance for future biodiversity surveys and food-web studies.

The Qinghai-Tibetan Plateau is a unique physical geographical unit, rich in habitat types and species, and a hot spot in biodiversity and global environmental change research. Unique fauna evolves on the Qinghai-Tibetan Plateau. Among these wild animals are the endemic hoofed animals which possess unique characteristics, such as Tibetan antelope (Pantholops hodgsonii), wild yak (Bos mutus), Tibetan gazelle (Procapra picticaudata), Przewalski’s gazelle (P. przewalskii), and white-lipped deer (Przewalskium albirostris). In this study, we explored the following questions: How many ungulate species are there on the Qinghai-Tibetan Plateau? How many endemic ungulate species are there? What is the distribution pattern of these species? What is the status of their survival? What is their conservation status? We firstly identified the geographic boundaries of the Qinghai-Tibetan Plateau. On the plateau, there are 28 ungulate species, 10 of which are endemic to the plateau. The ungulates of the Qinghai-Tibetan Plateau account for 42% of China’s extant ungulate species whereas the density of the hoofed species per unit area on the plateau is 62% higher than that of the whole country. However, the distribution of ungulate species on the Qinghai-Tibetan Plateau is uneven, and shows a pattern with low species density in the western region but high species density in the eastern part of the plateau, in contrast to the high species density of the endemic ungulates in the hinterland of the plateau. However, the ratio of threatened ungulates on the Qinghai-Tibetan Plateau is high, among them, 71% of the ungulates are threatened species, 54% are included in CITES Appendix I or II. The Red List Index of ungulate species on the Qinghai-Tibetan Plateau has continued to decline from 1998 to 2015, and this trend is compounded and worsened by the trend of global change, indicating that the living status of ungulates on the plateau is continuously deteriorating. Through four decades of conservation in the country, some important populations and habitats of the ungulates on the plateau are not protected by nature reserves and the newly established the Three-River-Source National Park. Therefore, to achieve the win-win goal of human social, and economic development and nature conservation on the plateau, we should manage the grassland ecosystems sustainably, save wildlife populations and habitats, and protect biodiversity on the plateau.

Studies of the distribution of biodiversity are hotspots in ecology and biogeography. Mountain ecosystems, which contain high habitat heterogeneity and biodiversity, play an important role in biodiversity conservation. The Himalayas are located in the southern rim of the Qinghai-Tibetan Plateau. The complicated geological topography, vast elevational gradients (100-8,844 m), significant vertical climatic zonation, and diverse biodiversity make it an ideal study area for biodiversity distribution research. By combining field data and data from museum records and literature, 313 mammal species were found for each 100 m elevational band along the gradient from 100 to 6,000 m above sea level (a.s.l.) in the Himalayas. Most of the elevational species richness patterns were hump-shaped. Species richness of the overall mammals peaked at an elevation of 900-1,400 m. Species composition along the elevational gradient in the Himalayas can be classified into five groups (100-1,500 m, 1,500-2,000 m, 2,000-3,000 m, 3,000-4,200 m and 4,200-6,000 m a.s.l.) using UPGMA clustering of elevation-based species assemblages, which was similar to the divisions of the vertical vegetation zone. The distribution of mammals is uneven, with low diversity found in the northern slope of the Himalayas and high diversity in the southern slope; the valleys on south slope of the Himalayas that contain higher biodiversity are biological corridors connecting the northern and southern fauna. To maintain the exchange of biodiversity, we should sustainably strengthen protection to the valley ecosystems.

The diversity and abundance of below-ground microorganisms and animals play an important role in shaping above-ground biodiversity and helps maintain ecosystem function. Yet, we have a limited understanding of belowground biodiversity, e.g. its spatial/temporal patterns, driving factors and responses to global change and human activities. This knowledge gap is particularly acute for the Tibetan alpine grassland that is sensitive to climate change and occupies 60% of the area of the Tibetan Plateau. Here, we first review recent studies that reported the drivers of patterns in five major soil organism groups, including fungi, bacteria, archaea, nematodes and arthropods on Tibetan alpine grassland. We then focus on the responses of soil biodiversity to climate change and human activities. Finally, we highlight some open questions for future research of soil diversity on the Tibetan alpine grassland. Specifically, we recommend that future studies examine (1) The mechanisms underlying distribution patterns of belowground biodiversity; (2) Links between aboveground and belowground biodiversity; (3) Effects of belowground biodiversity on the health and functioning of ecosystems; (4) Manipulative experiments of belowground biodiversity.

Sinochasea trigyna Keng (Family Poaceae) is endemic to the Qinghai-Tibet Plateau, and is listed as a second class national protected wild plant. The data about this monotypic genus is very scanty since its material difficult to acquire. According to the criterion of IUCN, the endangered level of S. trigyna is evaluated as VU B2ab(iii) based on the data from field investigations, herbarium records and the published literature. In this paper, we also reported the chromosome number (i.e. 2n = 22) and the type microchromosome of S. trigyna, which are both novel findings.