Distributions patterns of Cixiidae (Hemiptera, Fulgoroidea) in China highlight their high endemic diversity

The preprints are preliminary versions of works accessible electronically in advance of publication of the final version. They are not issued for purposes of botanical, mycological or zoological nomenclature and are not effectively/validly published in the meaning of the Codes. Therefore, nomenclatural novelties (new names) or other nomenclatural acts (designations of type, choices of priority between names, choices between orthographic variants, or choices of gender of names) should NOT be posted in preprints. The following provisions in the Codes of Nomenclature define their status:


Introduction
China covers an area of 9,634,057 km , which is quite similar to the entire area of Europe, and spans nearly 50 degrees of latitude from north to south, and more than 60 degrees of longitude from east to west in a world-renowned monsoon region (National Bureau of Statistics of the People's Republic of China, http://data.stats.gov.cn). Most regions have cold, dry winters and warm, rainy summers, but in combination with the varying topography and terrain conditions, the climate is actually very complex and locally diverse with a wide variety of temperature zones and precipitation gradients (Ren and Wen 2011). Most regions are located in the temperate zone (semi-tropical, warm, mid-range, and cold-2 have studied this group, mostly incidentally during their scientific travels. The first significant contribution was by Melichar (1902), who described 2 genera with 5 species from western China. Matsumura (1914) published 'Die Cixiinen Japans', describing 14 genus and 30 species, mostly from Taiwan. Kato 1932 focused on Northeastern China taxa, and published one new species. The first checklist of Cixiidae from the China mainland was provided by Hu (1935), who listed 11 species in 5 genera, which was updated by Metcalf (1936) in his 'Catalogue of the Homoptera'. Since then, many new species have been added. Jacobi (1944) reported 5 new species from the Fujian province. Fennah (1956) added 6 genera and 17 species from South China. Hori (1982) described 3 new Betacixius species from Taiwan. Chou et al. (1985) described 7 species in 4 genera in his "Economic Insect Fauna of China (Fulgoromorpha)". Tsaur provided a series of important contributions to the fauna from Taiwan, describing 155 species in 20 genera , Tsaur and Lee 1987, Tsaur et al. 1988, Tsaur et al. 1991a, Tsaur et al. 1991b, Tsaur 2009, Tsaur and Hsu 2003. Since then, several papers describing new recent taxonomic discoveries have been published (Wang 1991, Wang 1992, Liang 2001, Liang 2005a, Liang 2005b, Huang 1995, Hua 2000, Guo and Feng 2010, Guo and Wang 2007, Guo et al. 2009, Song and Liang 2013, Zhang and Chen 2011, Zhang and Chen 2013a, Zhang and Chen 2013b, Zhang and Chen 2011a, Zhang and Chen 2011b, Ren et al. 2014, Xing and Chen 2014, Bai et al. 2015, Luo et al. 2019a, Luo et al. 2019b, Zhi et al. 2019, Zhi et al. 2020a, Zhi et al. 2018b, Zhi et al. 2018a, Zhi et al. 2017, Zhi et al. 2021. All of these studies primarily focused on taxonomical issues, with limited ecological and geographical analyses. However, Cixiidae are obligatory phytophagous taxa and therefore directly linked to the distribution of their host plants. Both are patterned by the historical biogeography of the areas where they are distributed. How Cixiidae do follow the patterns of biogeogaphical distribution (major biological realms, biogeographical regions) already well established in China? Which boundaries can be identified for Cixiidae and at which taxonomical levels? The aim of this paper is to identify these correlations and to investigate how these zoogeographical regions are connected in China.
This current paper provides an updated species list of Chinese Cixiidae, their distribution in the main Chinese zoogeographical regions, and, also, which zoogeographical patterns they are distributed in. It also briefly proposes that an obligatory phytophagous insect group like the Cixiidae can help elucidate zoogeographical boundaries between the Palaearctic, Sino-Japanese and Oriental realms. Accordingly, the objectives of this paper are: (1) to provide a comprehensive species list of the Cixiidae from China, and their precise distribution within provinces. Although 191 Chinese species of Cixiidae have been currently recorded in FLOW (Fulgoromorpha Lists on The Web, Bourgoin 2021), an exhaustive review of the literature allows us to include in this review 253 species; (2) to compare Cixiidae species richness at the level of the Chinese zoogeographical regions and to document their distribution patterns and their endemism in each region, both at the tribal and generic level; (3) to investigate what biogeographical patterns the Cixiidae reflect: are they recognized effectively in a particular Sino-Japanese realm or a simple area of transition between the Palearctic and Oriental realms?
The distribution matrix includes 253 Chinese Cixiidae species. Among them, 3 species: Cixius narke Kramer, 1981, Oliarus splendidula Fieber,1876, andTachycixius (Tachycixius) pilosus (Olivier, 1791), were excluded from the analyses because we could not confirm their occurrence in China (no specimens observed during our survey) or because of uncertainties about where they were collected. Forty-eight additional Cixiidae species (Suppl. material 1) from adjacent areas based on literature and FLOW (Bourgoin 2021) were added for the cluster analysis. The species richness, distribution patterns and endemism, were calculated from our dataset and were used to generate a Venn diagram (Fig. 2) by TBtools Software .
The diversity of Cixiid species was analyzed in each of the 10 zoogeographical regions ( Fig. 1) and data were assembled into 1° × 1°grid-cells (~100 km × 100 km). Similarity between each pair of grid-cells was calculated using the Jaccard Coefficient of similarity ( Jaccard 1901), not involving the use of correspondence analysis to transform the data ( Smith and Bermingham 2005). Similarity of the genus and species levels were compared among the 10 zoogeographical regions using non-metric multidimensional scaling (NMDS) and an unweighted pairwise group method using arithmetic mean (UPGMA) clustering based on the Jaccard similarity coefficient matrix. The similarity coefficient matrix was imported into NTSYS Version 2.1 software (Rohlf 2000) for the cluster analysis.

Data resources
This checklist contains information updated up to April, 2021 compiled from scientific papers, book chapters, conference abstracts, theses, and the FLOW website (Bourgoin 2021). It also includes our own unpublished taxonomic data and original information from the following institutions: Shanghai Entomological Museum C.A.S (SEM), Museum of China Agricultural University (CAU), Entomological Museum of Northwest A&F University (NWAFU), Museum of Chinese Academy of Forestry (CAF) and Muséum National d'Histoire Naturelle (MNHN).

Kuvera vilbastei Anufriev, 1987
Nomenclature: Kuvera vilbastei Anufriev, 1987: 7| Anufriev & Emeljanov, 1988 to 161 species and 70.81% of endemic species in Taiwan region. In-between, species richness and endemism ratios are distributed in two groups: the Northeast China and the Qinghai-Tibet regions, respectively, with 8 and 10 species and 12.1% and 20% of endemism. The North, Central and South-West China regions, had comparable numbers of species and endemism, respectively, with 29 to 62 species and 35-40% of endemism. No significant differences in endemism among regions was observed. More than five-fourths of the species (205 species; 81.03%) are reported to occur in only in China, which illustrates the high level of endemism of Chinese fauna for this family (Table 1).

Distribution patterns of cixiid species in China
Based on the eight zoogeographic regions of China (Fig. 1), 38 main distribution patterns are observed (Fig. 2, Table 2). The number of species distributed only mono-regionally with a mono-regional distribution type (accounting for regional endemism) is highly variable among the different regions: Taiwan (45.10%), Central China (8.70%), Southwest China (6.72%), South China (5.14%), North China (3.95%), Qinghai-Tibet (0.79%) and Northeast China (0.40%). No endemic species were observed in the Nei Mongol-Xinjiang region ( Table 2). Nine bi-regional distribution patterns were observed, and among them the South China-Taiwan regional combination has the greatest number of species (13 species), which is 5.14% of the total number of Cixiidae species in China. Nine tri-regional distribution patterns were also observed, among which the largest number of species (11 species), is 4.35% of the total number of Chinese cixiids species in the Central-South China-Taiwan regions. The Southwest-South China-Taiwan combination has 6 species, which is 2.37% of the total number of cixiids in China. Five distribution patterns occur in 4 zoogeographic regions, among which the North-Southwest-Central-South China region and the Northeast-Central-South China-Taiwan region have two species, which is 0.79% of the total number of cixiids in China. All the remaining IV-, V-, VI-and VII-distribution patterns have only a single species, which represents 0.40% of the total number of cixiids in China (Table 2).

Cixiid patterns of distribution at the tribal level
Of the ten cixiid tribes distributed in China (Table 3), Pentastirini (20.55% of the cixiid species) and Semonini (17%) are the two most widely distributed tribes in all of China. Cixiini, which is the most species rich tribe with 45.85% of the species, is distributed in 7 regions of China, but has not been reported from the Northeast China region. Andini (5.14%) is not distributed in the Palaearctic realm in China; Eucarpiini (6.32%) and Borysthenini (1.98%) are distributed only in the Southwest, Central, South China, and in the Taiwan regions. The remaining tribes, Bennini (0.40%), Brixiini (0.79%), Oecleini (1.58%) and Stenophlepsini (0.40%) are only found in Taiwan (Table 3).

Cluster and Ordination Analysis
In both the generic and specific taxonomic levels, the dendograms clearly separate the northernmost regions (Rusian Far East, Nei Mongol-Xinjiang and Northeast China regions) from all other regionsand with the similar relationships for Chinese zoogeographical regions: (((((South-Central) + SouthWest) + North) + Taiwan) + Qinghai-Tibet). At the species-level the south adjacent China country region appears as sister to all of China. In contrast, at the generic level, this south adjacent China region sister to the central and south Chinese regions. In the northernmost regions, Russian Far East is closer to the Northeast China region at the species level and closer to the Nei Mongol-Xinjiang region at the generic level. In both analyses, the cophenetic correlation coefficient (r>0.8) is high, indicating close agreement between the cluster assignment and the original Jaccard similarity coefficient matrix.
The cluster analysis and the NMDS ordination generally showed similar interrelationships among regions (Fig. 3b, d). The stress values of 0.18 (generic level) and 0.30 (species level) demonstrate the accuracy of the projections in the matrix in the 2D ordination space. At the generic level (Fig. 3b), the Nei Mongol-Xinjiang and Russian Far East regions are closely related to each other, and the Northeast China, Nei Mongol-Xinjiang, and Russian Far East regions are clearly separated from the other 7 regions. The Southwest, Central, and South China regions are closely grouped together, and are also related to the North China and Taiwan regions, but the Qinghai-Tibet and VM regions are more separated. At the species level (Fig. 3d), a roughly similar pattern occurs and the Russian Far East is closer to the Northeast China region, but the VM region is clearly separated and more distant from all other regions.

Discussion
Host-plants are a key factor affecting the distribution of these obligatory, phytophagous planthoppers. The host-plants and the planthopper species complex are also affected by other complex topographic and climatic factors embedded in a long dynamic geological process. The high diversity of Chinese biotopes allows us to better understand and explain the resulting high diversity of Chinese Cixiidae with 253 species already described, that represent no less than 8.6% of the total species richness of the family.

Current Chinese Cixiidae diversity and distribution
More than 80% of the Cixiidae species are considered to be endemic to China. The highest endemism is found in Taiwan (45.1%), followed by the Central China (8.7%), Southwest China (6.72%) and South China (5.14%) regions. These figures are consistent with the species richness and endemism patterns observed in other Hemiptera groups, such as aphids (Huang et al. 2008, Gao et al. 2018), leafhoppers (Yuan et al. 2014, or more specifically for planthoppers (Zhao et al. 2020b, Zhao et al. 2020a). The South China-Taiwan distribution patterns pattern (5.14%), Central-South China-Taiwan (4.35%) and the South Western-South China-Taiwan (2.37%) are richer in distributions of species. This pattern probably results from the past interconnection of the island of Taiwan with the Asian continent during the Quaternary period, when the sea level fell, facilitating the species flow between these areas (Lei et al. 2003, Tang et al. 2006). Its subsequent geographical isolation after the Quaternary period explains its relatively independent pattern of speciation (Gao et al. 2018) and its high endemicity of species. More recently, the uplift of the Qinghai-Tibetan Plateau starting in the middle of the Eocene period (45-38 Ma), also had profound effects on the topography and watersheds of East Asia, the aridity of inland Asia, and the Asian monsoon system. These abiotic factors produced a three-stage pattern of species distribution, from high in the west to low in the east (Zhang et al. 2000). The vertical differentiation in plant distribution (Jin et al. 2003), affected their diversity and inceased the richness of local speciation events , Favre et al. 2015, and subsequently influenced the species distribution and speciation of the Cixiidae. During the late Oligocene to early Miocene periods (25-15Ma), the expansion of the Tibetan Plateau continued, and the East Asian monsoon and Indian monsoon prevailed in the Asian continent. This resulted in an increase of both temperature and sea levels , which allowed the northward propagation of fauna and flora. This area was pushed back southwards at the end of the Miocene period (10 Ma) by the uplift of the Hengduan mountains , which caused the climate to cool (Xie et al. 2019, Yu et al. 2020. Since the middle of the Holocene period (6Ka), rainfall declined and monsoon strength weakened, resulting in a dramatic decrease in precipitation in northern China, which affected the vegetative environment (Zhao et al. 2009, Huang et al. 2012. Quantitative precipitation reconstructions based on pollen collected from northern China indicated that a strong sealand pressure and temperature gradient caused by strong summer insolation in the northern hemisphere during the early Holocene period (0.14-0.07 Ma) caused enhanced monsoons (Chen et al. 2015, Zhao et al. 2009, Cook et al. 2011. Obviously the cixiid fauna diversity fluctuated at the same periods along with the diversity of their host-plants. However, without more robust phylogeny studies of Cixiidae, it remains difficult to better infer a more precise biogeographic historical scenario for the family and to link their distribution patterns to any of these important past events.

Biogeographical patterns of Chinese cixiids
Traditionally, the global biogeographical regionalization of China covers both the Oriental and Palearctic realms, which are bounded by the Qingling Mountain-Huai River, around 32-34N in the east of China (Sclater 1858. Wallace 1876, Zhen 1960, Zhang 1999, Cox 2001, Kreft and Jetz 2010, Morrone 2015, Song et al. 2016. In 2013, based on its zoological fauna, Holt added a Sino-Japanese realm standing between the Palearctic and Oriental realms, and from west of Tibet to east of the Japanese archipelago. He located the Palaearctic/Sino-Japanese boundary at about 40-41N, and the Sino-Japanese/Oriental boundary at 24-25N in southeast China (Fig. 1). Kreft and Jetz (2013) questioned the validity of this realm because they regarded it as just a biogeographical transition zone between the Palearctic and Oriental realms. According to their taxa ethoecological characteristics, the Sino-Japanese realm boundaries are generally clustered with the Oriental realm (Kreft and Jetz 2010, Song et al. 2016).
This result is also observed here for the Chinese Cixiidae divided into two major zoogeographic areas: the Nei Mongol-Xinjiang and Northeast China regions from the rest of China. This boundary corresponds to the Palearctic/Sino-Japanese north boundary and appears to be more well defined than the Palearctic/Oriental boundary. The Andini tribe serves as a landmark for the Palearctic/Sino-Japanese north boundary, while the Eucarpini and Borysthenini tribes are primarily concentrated south to the Qingling Mountain-Huai River point to the traditional Palaearctic/Oriental boundary as proposed by . Eucarpini and Borysthenini are landmarks for the south Sino-Japonese realm, clustering with the Oriental realm. Bennini, Brixini, Oecleini and Stenophlepsini, which are all distributed in Taiwan, may either indicate the northern limit of older and wider distributions of these tribes or might have resulted from occasional dispersions from neighbouring south regions.
At the genus level, the south parts of China cluster with the Indochina region in our analyses, but at the species level all of China forms a unique group. This may be related to the late Eocene uplift of the Himalayas and recent uplift of the Himalayan-Hengduan Mountains in the late Miocene, with a peak before the late Pliocene (Harrison et al. 1992. These geographical uplifts resulted in the formation of large topographic barriers isolating South china fauna and favorizing recent speciation events and endemism as already shown in several other taxa such as frogs (Che et al. 2010), insects (Ye et al. 2016;), birds (Cai et al. 2018, Liu et al. 2016, Dong et al. 2020), mammals (Ge et al. 2017) and plants (Ebersbach et al. 2017, Feng et al. 2013. Moreover, Quaternary (2.6 Ma) tectonic movements and the influence of the Indian and Pacific monsoons greatly contributed also to the segregation, dispersal and speciation of Cixiidae in southern China and Southeast Asia (Shi et al. 1998, Liang 2003. The South China region is usually included in the Oriental realm in other studies (Zhang 1999, but our analysis indicates that for Cixiidae the South China region is closer to the Central, Southwest, and North China region (Sino-Japonese realm). This is consistent with the results of a quantitative analysis of terrestrial mammals in China and adjacent regions by Xiang et al. (2004), where clustering analysis showed the proximity of South China region to Central and Southwest China regions, and they suggested these regions as the South China Division.

Conclusions
This study is the first zoogeographic analysis based on grid cells of Cixiidae in China and adjacent areas. However, since collecting efforts are stronger into southern China and taxonomic studies clearly advanced in the Taiwan region because of studies by Tsaur over the past three decades , Tsaur and Lee 1987, Tsaur et al. 1988, Tsaur et al. 1991a, Tsaur et al. 1991b, Tsaur 2009, Tsaur and Hsu 2003, the results reported from those areas may be based on more data than those in other regions of China. Previous studies of Cixiidae in countries adjacent to China are very limited in their numbers and scope of the investigations. Despite studies by Distant (1911), Anufriev (1987), Anufriev and Emeljanov (1988), , Fennah 1978, Hoch (2013 a large number of species are still undescribed, particularly in the Indochina peninsula.
With the current available data, the observed distribution patterns show that an intercalary Sino-Japanese realm is recognizable between the Palaearctic and Oriental realms. At the regional level, the South China region cluster is more closely with the Southwest, Central and North China regions. Taiwan is clearly separated from the South China region and mainland China, but is more closely related to the Qinghai-Tibet region and Indochina countries. The Central and South China regions are close to each other, but the Qinghai-Tibet region is singularly different. In the future, a more comprehensive analysis in combination with a more reliable phylogenetic analysis of the Cixiidae famiy in certain areas of China and countries surrounding China, will be essential to develop a more complete biogeographical history of the evolution, development, and distribution of Cixiidae in China.