Global distribution patterns of organisms have become a hot research topic in recent years due to increasing concerns about the global loss of species richness (Gaston 2000). The cogent statement of the increase in species diversity from polar to equatorial regions was confirmed by plants and vertebrate animals (Willig et al. 2003), but little is known about global pattern in species level of soil invertebrates, although these are recognized as one of the most species rich groups on earth (Wardle et al. 2004; Fierer et al. 2009). Maraun et al. (2007) investigated the latitudinal diversity gradient in a soil taxon, oribatid mites, and made a negative conclusion that diversity increases from the boreal to the warm temperate regions, but did not increase further in the tropics. Recently, Wu et al. (2011) used a molecular approach to analyze samples from 11 locations worldwide and suggested that there may be an inverse relationship between above-ground plant biodiversity and soil invertebrate animal biodiversity.

Litter-dwelling thrips is a group of thysanopteran insects that have adopted the habitat of forest litter or surface soil where they feed only on either fungal hyphae or fungal spores during the early stages of leaf decay (Mound 2005; Wang and Tong 2012). The diversity of these litter thrips is usually related to environmental factors, including temperature and humidity of the soil, the plant species that produce the litter (Tree and Walter 2012) and the species of fungi involved in decomposition (Ananthakrishnan 1996). This diversity is a potential indicator to assess changes in the forest environment (Mound 1977). A majority of litter–dwelling thrips species are wingless, usually have weak migratory ability, and can well provide materials for analyzing the fauna and zoogeography (Mound 1970). Another important role was played by some species which could be effective in the natural control particularly those efficient spore feeders of plant pathogenic fungi (Ananthakrishnan 1981).

Most publications of litter-dwelling thrips species were descriptions of new genera and species, sometimes with little information on their vegetation or microhabitat associations (Dang et al. 2013; Haga 1973; Mound 1970, 1976, 1977, 2002, 2013; Mound and Palmer 1983; Mound et al. 2013; Okajima 1981, 1994, 1998; Okajima and Urushihara 1992; Wang and Tong 2011; Wang et al. 2007, 2013). Mound (2002) indicated that litter–dwelling thrips are particularly diverse in the subtropics and tropics, with up to 50% of thysanopteran species in these regions. For example, almost 50 species of litter thrips were described from a single area of forest 50 km in diameter in southern Brazil (Mound 1977). Few quantitative field studies have been published on litter thrips. The only published quantitative field studies of litter thrips were conducted in subtropical China (Wang et al. 2008; Wang and Tong 2012) and in Australia (Tree and Walter 2012). In China, Wang and Tong (2012) found Thysanoptera constituted more than 1% of total litter–dwelling macroinvertebrate individuals extracted with modified Tullgren funnels. Species richness and abundance of litter-dwelling thrips gradually increased from July to December, and then declined rapidly. Litter-dwelling thrips were found only in the litter layer and upper soil layer (0-5 cm in depth) and were entirely absent in deeper soil layers. Tree and Walter (2012) found leaf-litter thrips were much more common and diverse in dry sclerophyll forest than in wetter forest types in subtropical southeast Queensland, Australia. The species diversity in soil fauna has been studied in temperate regions for more than 50 years, but with scarcely any mention of thrips (Wallwork 1976; Hattenschwiler et al. 2005). This lack of reference to thrips raises the question whether or not litter-dwelling thrips are distributed only in tropical and sub tropical regions.

To determine whether or not species diversity of litter-dwelling thrips alters at higher latitudes, we collected litter samples from six natural reserves which are located respectively in the temperate, subtropical and tropical zones, along a 4100 km latitudinal gradient in East China. The observations are also interesting from the point of view of understanding geographical scale differences in ecology, and responses of ecosystems to global warming.

Extant insects of the order Thysanoptera include approximately 6000 described species worldwide, classified into nine families (ThripsWiki 2014). At least 2500 species are placed in the families Phlaeothripidae and Merothripidae, and many of these are found in forest litter where they feed on fungi (Mound 2005). The systematics of litter-dwelling thrips has been well studied by Mound and Palmer (1983), Mound (2013) and Mound and O'Neill (1974). These litter thrips are especially diverse in some areas of the tropics and subtropics, which were proved by some qualitatively extensive surveys with the purpose of extracting abundant specimens to describe new taxa and study morphological variation. For example, almost 50 species in eight genera of litter thrips were described from a single area of forest 50 km in diameter in southern Brazil (Mound 1977). Two merothripid species and 15 phlaeothripid species were recorded from Jinmuji forest, Kanagawa Prefecture, Japan (Okajima and Urushihara 1992). Fifty thrips species, including twenty-six undescribed species, were recovered from various forest types and microhabitats in a single locality of subtropical eastern Australia (Tree and Walter 2012).

In recent years, we conducted a series of investigations by quantitative sampling methods to survey the species diversity of leaf-litter thrips in China. We found these litter-dwelling thrips to be a common group of litter macro-invertebrates, with high species diversity and relative abundance in Guangdong Province of subtropical China. For example, the numbers of these thrips accounted for 3% to 13.5% of total litter macroinvertebrate individuals caught in four different forest types of two natural reserves (Li et al. 2004a, b). In another natural forest, the relative abundance of Thysanoptera was 5.0%, representing 10 genera and 12 species (Wang et al. 2008). In an urban forest remnant, 25 species of 19 genera of leaf-litter thrips were collected, and these constituted 6.5% of total litter macro-invertebrate individuals caught (original totals 1413 thrips out of 21817 individual macro-invertebrates) (Wang et al. 2007; Wang and Tong 2012). In the present work, the relative abundance of Thysanoptera was over 3.0% in four natural reserves from the subtropical and tropical zones, and reached 5.9% in the tropical zone of Wuzhishan Natural Reserve (Hainan Province), only less than Acarina and Collembola. In contrast, it was only 0.3% in the warm temperate zone of Yunmengshan National Forest Park (Beijing) and no individual was collected in mid temperate zone of Changbaishan Natural Reserve (Jilin Province). 14 species and 12 genera of fungus-feeding thrips were collected from leaf litter of tropical China, 10 genera and 12 species in the subtropics, but only 2 species in the warm temperate zone. No species was collected in mid temperate zone. The results indicate that species number and relative abundance increase with decreasing latitude.

In the New World, the fauna of litter thrips is represented primarily by three genera, Eurythrips, Terthrothrips and Tylothrips (Mound 1976, 1977). In contrast, samples taken in the Old World rarely include these genera, whereas species of Apelaunothrips and Adraneothrips are often abundant, with species of Psalidothrips and Zemiathrips found commonly in the litter of sclerophyll forests in Australia (Mound 2002; Tree and Walter 2012). A total of 128 species in 43 genera belonging to 2 families of litter thrips were documented from China. Fauna and Zoogeographical analyses indicated that fungus-feeding thrips are diverse in China. Psalidothrips and Apelaunothrips are the dominant genera, with 9 species and 8 species, respectively. Relative abundance also can provide assemblage composition and fauna characteristics, and refers to the total number of specimens for a particular species divided by the total number of all litter thrips. Psalidothrips ascitus and Hyidiothrips japonicus were dominant species in samples from the tropics, accounting for 48.6% and 17.1% of the total adult thrips individuals, respectively Heliothripoides reticulates and Terthrothrips palmatus were collected only in this reserve. In subtropics, Preeriella parvula, Psalidothrips ascitus, Apelaunothrips lieni, Stephanothrips japonicus were dominant species, accounting for 33.5%, 19.9%, 15.1% and 13.5% of the total adult thrips individuals, respectively. In the warm temperate zone, a few individuals were collected of Preeriella parvula and Holothrips.

A total of five distribution patterns can be recognized among the 15 genera of litter-dwelling thrips in six nature reserves. Genera with a Pan-tropical distribution are most abundant, including the following: Acallurothrips, Adraneothrips, Baenothrips, Holothrips, Karnyothrips, Preeriella, Stephanothrips and Terthrothrips. These genera are found in the tropical and subtropical areas of Asia-Africa-America. Apelaunothrips is distributed in the tropical and subtropical areas of Asia-Africa. Hyidiothrips and Psalidothrips occur in the warmer areas of Asia and America. Heliothripoides, Holurothrips and Mystrothrips are distributed across tropical or subtropical Asia, being found in China, Japan, Korea and the India-Malaya area. Excluding the genera with cosmopolitan distributions (Hoplothrips), almost all litter-dwelling genera are found in tropical and subtropical areas. Differing from geographic distribution patterns at genus level, the 18 species of fungus-feeding thrips were divided into four distribution patterns, according to species currently distribution areas. Endemic to South China distribution: Apelaunothrips lieni, Adraneothrips chinensis and Terthrothrips palmatus. Eastern Asia distribution: Adraneothrips russatus, Holurothrips morikawai, Hyidiothrips japonicus, Psalidothrips simplus, Stephanothrips japonicus and Mystrothrips flavidus recorded in Japan and South China. Tropical Asia distribution: Heliothripoides reticulates, Psalidothrips ascitus and Preeriella parvula recorded in India-Malaya area or Japan. Tropical Asia, Africa and America disconnected distribution: Baenothrips asper and Karnyothrips flavipes. These zoogeographic analyses indicate that litter-dwelling thrips possess tropical and subtropical characteristics. Litter-dwelling thrips inhabiting forest litter usually have weak flight ability, some species are even wingless, but individual genera or even species are sometimes found in two disconnected continents. The causes have been discussed by Wang and Tong (2012).

The species richness of litter-dwelling thrips in our quantitative investigation at six locations across a broad latitudinal gradient reveal an increase with decreasing latitude, as reported for many other taxonomic groups of invertebrates, prominent examples include termites (Collins 1989), butterflies (Sime and Brower 1998) and springtails (Sun et al. 2013). However, we also found that the density of litter-dwelling thrips peaked at our mid-latitude sites (Shennongjia Natural Reserve), with no individual collected at the northern extreme. Individuals of oribatid mites, springtails and soil nematodes were richest in the warm temperate regions, and they could be found from polar to equatorial regions (Bardgett and Wardle 2010); a similar conclusion was made concerning leaf-associated aquatic hyphomycetes (Jabiol et al. 2013). The mechanisms underlying such patterns are still not fully understood, but several explanations have been proposed, such as habitat relationship, body size of organisms, solar energy, temperature and precipitation, etc (Turner 2004; Willig et al. 2003). We think that temperature might be a limiting factor for the distribution of litter-dwelling thrips in the northern temperate zone, particularly areas with snow cover in winter. In tropical regions, particularly complicated climatic conditions lead to dramatic fluctuation among different seasons (Wang and Tong 2012), and this could decrease number of individuals of litter thrips.