Pollination syndromes of Chinese gesneriaceae: a comparative study between Hainan Island and neighboring regions.
China is rich in plant species and diverse in topography (Lopez-Puyol et al., 2006; Wei et al., 2010), providing a suitable study region to evaluate potential drivers of evolutionary change in floral syndromes and pollination syndromes between mainland and islands. Among China's mega-biodiversity, the family Gesneriaceae shows a diversification center in Southwest China, including Hainan Island (Li & Wang, 2005). Gesneriaceae can be divided into two subfamilies, i.e. Cyrtandroideae (superior ovary with two unequal cotyledons; ~ 1700 spp.) and Gesnerioideae (inferior ovary with two equal cotyledons; ~ 1000 spp.) (Li & Wang, 2005). Cyrtandroideae occurs mostly in Asia and Africa, while Gesnerioideae is restricted to tropical America (Li & Wang, 2005; Perret et al., 2013). An up-to-date phylogeny (Perret et al., 2013) indicated that the two subfamilies probably separated at ~40 Mya and that most narrowly-endemic species are newly evolved due to recent speciation rather than relicts surviving from the ice ages (Wang et al., 2010; Wei et al., 2010; Perret et al., 2013).
All the Chinese Gesneriaceae belong to Cyrtandroideae, and about 59 genera and 460 species are recognized (Li & Wang, 2005; Wei et al., 2010), among which 27 genera and 375 species are endemic (Li & Wang, 2005; Wei et al., 2010). Such striking species diversity and endemism in both mainland China and Hainan Island provides an opportunity to examine how Gesneriaceae diversity evolved and is maintained in tropical Asia. Compared with the mainland, islands normally have lower species richness (Bascompte et al., 2006; Lord et al., 2013; Marten-Rodriguez et al., 2009, 2015) with fewer pollinator species (Carlquist, 1974; Barrett et al., 1996). Consequently, plants on an island are thought to have generalized pollination modes, i.e., they depend on more than one type of pollinators or are wind pollinated, and many have floral mechanisms than enable autonomous self-pollination (Barrett et al., 1996). Therefore, island-mainland systems can provide a useful context to evaluate interactions between pollination and mating systems in response to changed pollination services, such as variation in pollinator availability and visitation rates (Marten-Rodriguez et al., 2015; Traveset et al., 2015). Several studies have shown that floral traits can lead to accurate predictions about the effective pollinators of plant species (Hargreaves et al., 2004; Pauw, 2006; Armbruster et al., 2011; Rosas-Guerrero et al., 2014). The predictability of pollination mode from floral syndromes is often greater in tropical species (Rosas-Guerrero et al., 2014).
In this study, we compared patterns of floral diversity and inferred pollination syndromes for Gesneriaceae in Hainan Island with those of neighboring Chinese regions, i.e. Yunnan, Guizhou, Guangxi, Guangdong. We evaluated distribution patterns of Gesneriaceae within this globally important biodiversity hotspot; we predicted that the species endemism ratio would be higher on Hainan Island than in neighboring mainland regions and that the species on Hainan would include a higher proportion of species with floral traits that promote selfing or have generalist pollination mechanisms.
Materials and Methods
The species diversity of Gesneriaceae in Hainan Island and its neighboring regions were obtained from Floral of China, Li and Wang (2005), Wei et al. (2010). When there was any mismatching information about taxonomy and distribution, we referred to the latest one of the above literature.
Besides the data on total numbers of species and endemic species for a given region, the geographical area and elevation range, i.e. difference between the highest and the lowest elevation (estimation of habitat heterogeneity) of each region were also collected according to Tang et al. (2006) and Ren (2015). Habitat heterogeneity is a major structuring agent of ecological assemblages involving all environmental factors such as temperature and humidity. It affects plant diversity at both local and regional scales and ultimately contributing to overall biodiversity of a region. So we take into account of habitat heterogeneity, together with area, when we estimate species density and endemic ratio.
To determine species relationships of Gesneriaceae from Hainan Island and nearby regions, we searched in GeneBank for nuclear ITS (ITS1/ITS2) and chloroplast trwL-F for all species on Hainan Island and all available Gesneriaceae species from nearby regions. The data obtained were edited using Sequence Matrix software (Vaidya et al., 2011) and compared in MEGA software (Kumar et al., 2008). Four species with known speciation time (-24 Mya) (Moller et al., 2010; Weber et al, 2011; Perret et al., 2013), i.e. Primulina dryas, Bournea sinensis, Liebigia asperifolia, Chirita gemella were also included into the phylogeny tree reconstruction to estimate the relative time of origin for Hainan-endemic species. The phylogenetic tree was drawn using Beast software (Drummond & Rambaut, 2007).
Species Diversity and Endemism Ratio
The total number of species and endemic species were counted for Hainan Island and its neighboring regions. The data for area and habitat heterogeneity (approximated by altitudinal range) were log-transferred to ensure data are comparable among regions and to minimize their impacts on the calculation of species density (Tang et al., 2006; Ren, 2015). This area- and habitat-adjusted species density (SD) was calculated by SD = NT/[ln(A) + ln(E)], where NT is the species richness (the total number of species in the region), A is the area of the region ([km.sup.2]), and E is the habitat heterogeneity (Table 1). To estimate the relative proportion of endemic species, the area- and habitat-adjusted endemism index (EI) was calculated as EI = [NE/(NT - NE)]/[ln(A) + ln(E)], according to methods described in Tang et al. (2006) and Ren (2015), where NT is the total number of species, NE is the number of endemic species and A and E are as above. Endemism ratio (ER) was also calculated as ER = NE/NT x 100% for each region to be used as a complementary index of EI to estimate the relative level of endemism of a region.
Floral Phenotypes and Pollination Syndromes
To characterize floral phenotypes,, we measured 8 floral traits and recorded habit and habitat information for all species on Hainan Island according to methods in Marten-Rodriguez et al. (2009). These floral traits are (i) corolla color; (ii) corolla constriction; (iii) corolla curvature; (iv) corolla shape; (v) corolla ornamentation (i.e. dots or stripes on corolla); (vi) floral symmetry (actinomorphic or zygomorphic); (vii) stamen number (two, four or five); (viii) anther position (determined by the anther location relative to the corolla tube, i.e. included, gular, or exserted). These traits were chosen because they are main factors for pollinator attracting and possess potential effect on pollination efficacy and accuracy. For example, floral color and shape had been shown to be fundamental floral signals to pollinator perception (Fenster et al., 2004; Rosas-Guerrero et al., 2014). Floral symmetry, i.e. zygomorphic versus actinomorphic corolla, is crucial in selecting pollinator types and determining their forage behavior (Sargent, 2004; Gong & Huang, 2009). Corolla constriction and curvature can promote efficient contact of sexual organs with bird beak or moth tongue and may prevent visitation of illegal pollinators (Campos et al, 2015). In this paper, we considered a floral tube as constrict when the diameter of corolla throat is obviously narrower than corolla mouth and corolla curvature was defied as the curvature of corolla tube is greater than 5 [degrees], measured by a protractor according to Campos et al. (2015) and Marten-Rodriguez et al. (20099). Corolla ornamentation (spots or stripes on petals) may act as honey guide and so we also recorded whether there is corolla ornamentation or not. Anthers number and position directly determine the possibility of pollen pickup by pollinators and the particular pollinator's body part for pollen carrying.
For each species, we measured 10-15 fresh flowers collected from at least three individuals. For corolla color, seven types, i.e., purple, white, blue, orange, red, yellow and pink, were recognized. Five corolla shapes (tubular, slender tubular, tubular-funnel, campanulate and urceolate forms (Fig. 3) were identified in this study. For the species in mainland China, their floral traits were determined mainly by literature surveys such as Li and Wang (2005) and Wei et al. (2010) and the references in Table 2.
The pollinators of Gesneriaceae on Hainan Island were mainly obtained from our field studies in 2014-2015. Observations were made of floral visitors, and only visitors that touched sexual organs were recorded as pollinators. For most Gesneriaceae from the Chinese mainland, data from published literature was used to determine their pollinators. For species without any empirical studies on pollination, their pollination syndromes and their potential pollinators were predicted from corolla traits and reward types, according to Fenster et al. (2004), Hargreaves et al. (2004), Zhang (2004) and Rosas-Guerrero et al. (2014).
Species Diversity and Endemism
Based on the surveys for all species of Gesneriaceae in Hainan Island and its neighboring regions, 24 species were recognized on Hainan Island, of which eight were endemic. Yunnan and Guangxi regions had the highest species richness (236 and 211 species) and number of endemic species (106 and 120 species). If geographic area was taken into account, Guangxi and Hainan Island were highest in endemic index (Table 1), with values around two times those in other regions, despite Hainan having low species density (Table 1).
For the 24 species in Hainan Island, ITS 1/2 and Trn L-F sequence data were unavailable for four species, i.e. Stauranthera umbrosa, Oreocharis flavida Merr., Oreocharis dasyantha, Paraboea changjiangensis. The remaining 20 species from Hainan Island (five were endemic) and eight closest relatives from nearby regions were used to construct a phylogenetic tree. The phylogenetic tree (Fig. 1) showed that most Hainan Gesneriaceae were more closely related to species from Southeast Asia rather than mainland China. Most Hainan-endemic species evolved recently and in a relatively short time (Fig. 1).
Corolla Color and Ornament
A purple corolla was common in species from Hainan Island and the Chinese mainland, followed by white and blue corollas ([x.sup.2](HN) = 29.084, P < 0.0001; [x.sup.2](YN) - 116.412, P < 0.0001; [x.sub.2](GX) = 359.662, P < 0.0001; [x.sup.2] (GZ) = 142.807, P < 0.0001; [x.sup.2](GD) = 119.778, P < 0.0001). Yunnan Province had many species with red and yellow corollas (Fig. 2).
Corolla ornamentation was nearly as common as corollas without ornament in GX ([X.sup.2] = 0.168, P = 0.6817) and GZ ([x.sup.2] = 0.093, P = 0.7607)) (Fig. 2), while in HN, YN and GD, most species had corollas without dots, stripes or any other ornaments or appendages (Fig.2).
Corolla Shape and Type
Five corolla types were identified in this study: tubular, slender tubular, tubular-funnel, campanulate and urceolate forms (Fig. 3). The campanulate and tubular-funnel form was the most abundant corolla types in HN ([x.sup.2] = 31, P < 0.0001), GX ([x.sup.2] = 60.439, P < 0.0001) and GZ ([x.sup.2] = 38.928, P < 0.0001) (Fig. 3). Although zygomorphic flowers showed overwhelming dominance inall regions (Fig. 2), HN had a much higher percentage (21%) of actinomorphic species than other regions (all <10%) ([X.sup.2] = 8.167, P< 0.0001).
Most species in Hainan Island and its neighboring regions had two or four stamens, rarely five. The number of species with two stamens or four stamens in Hainan Island and Yunnan were nearly equal (Fig. 4). In GX ([x.sup.2] = 86.43, P < 0.0001), GZ ([X.sup.2] = 12.629, P = 0.0004) and GD ([x.sup.2] = 15.254, P < 0.0001), species with two stamens were more common, especially in Guangxi, where there were 172 species with two stamens, while four stamens were found in only 39 species (Fig. 4). In HN, species with exserted, gular or included stamens were fairly equal to each other ([chi square]= 0.25, P = 0.8825), while most species in other regions possessed stamens deeply included in the corolla tube (Fig. 4).
Habitat and Habit
In YN ([chi square] = 292.447, P < 0.0001), GD ([chi square] = 56.857, P < 0.0001) and GZ ([chi square] = 52.688, P < 0.0001), most species grew on non-limestone habitat, while in Guangxi and Hainan Island, nearly half of the species were limestone-endemic (Fig. 5). For all regions, herb is the most common habit for Gesneriaceae ([chi square] (HN) = 15.75, P = 0.0004; [chi square] (YN) = 123.5, P < 0.0001; [chi square] (GX) = 277.724, P < 0.0001; [chi square] (GZ) = 115.413, P < 0.0001; [chi square] (GD) = 98, P < 0.0001)., but the liana habit was relatively widespread in Yunnan (Fig. 5).
There are few detailed empirical studies on pollination of Gesneriaceae in China; we found only eight cases (Table 2). Bees were the only pollinator functional group for all these species (Table 2), including bumblebee, carpenter bees, Apidae and Amegilla. Our field studies on Hainan Gesneriaceae also found that Bombus and Amegilla were the most frequent pollinators, with no bird- and moth-pollinations were observed (Shao-Jun Ling & Ming-Xun Ren, unpublished data).
When pollinators were predicted from floral syndromes, mainly using corolla color and reward type (Fenster et al., 2004; Hargreaves et al., 2004; Rosas-Guerrero et al., 2014), bee-pollination was widely predicted. In Yunnan, however, where red corollas and curved floral tubes were widespread (Figs. 2, 4), bird pollination was predicted to be fairly common.
We also recognized a highly-specialized pollination system, mirror-image flowers, in three genera: Paraboea (Gao et al., 2006), Gyrocheilos and Didymocarpus (Wei et al., 2010). Mirror-image flowers (enantiostyly) are a sexual polymorphism in which the style is deflected away from the floral axis either to the left (left-styled flower) or the right (right-styled flower) (Jesson & Barrett, 2002; Gao et al., 2006). Such complex floral syndromes have been widely recognized as a highly specialized mechanism for promoting cross-pollination between left- and right-styled flowers via pollen transfers on either the left or right sides of the abdomens of bees (Jesson & Barrett, 2002; Ren et al., 2013). No mirror-image flowers were found in the species on Hainan Island, which include two Paraboea species.
Species Diversity and Relationships
Based on Burtt (1998) and Perret et al. (2013), the primitive taxa of Chinese Gesneriaceae have closer affinities to taxa in South Asia, including the Indo-China Peninsula. The ancestors of the South Asian gesneriads were hypothesized to come from neotropical America via two independent long-distance dispersals during the Eocene and Oligocene (Perret et al., 2013). After this original dispersal, long-distance dispersals were very few (Wang et al., 2010; Perret et al., 2013), and species diversification within Asia occurred largely in situ, especially on limestone areas (Wei et al., 2010).
It is not surprising that Hainan Island has the lowest species diversity compared to other regions of China, since it is very small, with an area of approximately 3.4 x [10.sup.4] [km.sup.2]. Twenty-four Gesneriaceae species are found on the island, among which eight species and two genera are endemic. The two endemic genera, Metapetrocosmea and Cathayanthe, are both monotypic (Li & Wang, 2005; Xing et al., 2014). The endemism index is high (0.27) in Hainan Island, only slightly lower than Guangxi (0.336), which is widely accepted as an evolutionary hotspot for Asian Gesneriaceae (Li & Wang, 2005; Wang et al., 2010; Wei et al, 2010). All endemic taxa from Hainan Island are found on late-derived branches on the phylogenetic tree and evolved in a short period, starting at -27 Mya (Fig. 1). The burst of speciation was likely triggered by the tropical monsoon in SE Asian (first formed at about 28 Mya), which was caused by the collisions of the Indian Plate with Eurasian Plate at about 34 Mya. Not only the propagula of tropical plants can be brought by the warm pacific currents of monsoon in SE Asian to northward migration, but Hainan Island which located in north margin of the tropics possessed abundant rainfall and hyperthermal and humid climates, and made it a paradise that lots of tropical plants of northward migration can repreduce freely in Hainan Island. At that time, Hainan Island was separated from mainland China about 100 km, and isolated effection is obvious strong on species differentiation because of mountains and rivers in island (Zhu, 2016). These geographical and climatic factors is the point that Generiaceae endemics was generated centrally then in Hainan Island.
Hainan Island was linked with Leizhou Peninsula of Guangdong Province two times because of sea recession at the end of Piocene and decrease of sea level in late Peistocene (Liang, 2013; Zhu, 2016). In this process, Generiaceae species (ie. Aeschynanthus bracteatus, Chirita sinensis, Beccarinda tonkinensis and so on) of south China and Guangdong province got into Hainan Island, resulted in many species exchanges between Hainan Island and its' nearby regions.
Two scenarios can explain why the Gesneriaceae of Hainan Island have a higher level of the area-adjusted endemism index than that of Yunnan, Guizhou and Guangdong (Table 2). First, Hainan is an island, which leads to reproductive isolation from congeners on the mainland and produces greater opportunities for lineage divergence. Secondly, the highly fragmented landforms produced by limestone weathering in the south-central part of Hainan probably further facilitates speciation which has been hypothesized to be the reason for Gesneriaceae diversification in China (Wei et al., 2004). The isolated environments of islands surrounded by ocean, combined with the limestone topography in south-central Hainan, generate habitat isolation and can lead to species differentiation.
Corolla color is an important indicator of pollinator type (Rodriguez-Girones & Santamaria, 2004; Zhang, 2004). Normally, red and orange flowers are attractive to birds (Rodriguez-Girones & Santamaria, 2004) and purple, yellow and blue flowers are mainly visited by bees (Grant, 1950). Our data show that purple and white Gesneriad flowers dominate all regions, including Hainan Island, indicating Gesneriaceae in these regions are mainly visited by bees, including Bombus, Chalcididae, Amegilla and Lasioglossum (Gao et al., 2006; Wen et al., 2012). In Yunnan, red and orange flowers are also fairly common, which suggests that the Yunnan flora is more diverse in pollinator types and birds may be common pollinators in addition to bees, although bird visitors have not yet been reported in this region.
Hainan Island and Yunnan have significantly more species with four stamens, while diandrous flowers are more common in other regions, especially in Guangxi (Fig. 4). A reduced number of stamens is a derived character normally associated with increased pollination efficiency in Chinese Gesneriaceae, according to phylogenetic studies that used both morphological and molecular data (Li & Wang, 2005; Wei et al., 2010). Such patterns indicate that the Gesneriaceae flora in Hainan Island and Yunnan may be more generalized in pollination mechanisms.
The most distinct floral trait separating Hainan Island from other regions is the location of the anthers. Hainan species normally protrude their separated stamens well out of the floral tube or place their anthers at the floral throat, while most species in other regions hide their stamens deeply in the floral tube (Fig. 5). Compared to included stamens, which require that the pollinators enter the floral tube for pollen to be picked up, separated and exserted stamens can contact most visitors approaching the flower and thus are available to generalized pollinators, such as hoverflies and quickly-visit bees (e.g. Amegilla). Therefore, generalist pollination is suggested to be more common on Hainan Island as compared with the other regions, which probably is an evolutionary outcome of the pollinator-depauperate environment on the island (Marten-Rodriguez et al., 2015; Traveset et al., 2015).
In Antillean Gesneriaceae, pollinator-mediated selection on floral form plays an important role in species diversification, and such natural selection provides support for shifts between hummingbird vs. bat pollination or specialized vs. generalized pollination (Marten-Rodriguez et al., 2009). In other words, the floral phenotypes could be used to predict the pollination syndromes and pollinators in this family (Marten-Rodriguez et al., 2009; Rosas-Guerrero et al, 2014).
Although most floral traits show a similar pattern of variation in Hainan Island and other regions, relatively more species in Hainan Island have a corolla without ornamentation, which may reflect a generalized pollination system. Bee pollination is normally treated as a relatively more general pollination mechanism than bird pollination (Fenster et al, 2004; Rodriguez-Girones & Santamaria, 2004). Hainan Island is similar to Guangxi, Guizhou and Guangdong, with bee pollination being the most common strategy.
Floral symmetry is an important trait both in taxonomy and in pollination systems, and normally zygomorphy is thought to be more specialized in pollination adaptations, since it restricts pollinator behaviors and can therefore increase pollination efficiency (Sargent, 2004; Gong & Huang, 2009). Many studies have proposed that actinomorphic flowers are primitive and zygomorphy evolved from an actinomorphic ancestor under selection to increase pollination accuracy (Sargent, 2004; Gong & Huang, 2009). Gesneriaceae is largely zygomorphic and its sister families are the exclusively zygomorphic Scrophulariaceae and Calceolariaceae (APG, Angiosperm
Phylogeny Group III, 2009). In Asian Gesneriaceae, actinomorphic genera Ramonda, Conandron, Bournea, Thamnocharis, and Tengia are on the posterior branches of phylogenetic trees, and all are clustered respectively with their zygomorphic ancestors (Wang et al., 2010). This result indicates that actinomorphy is probably a derived trait in Asian Gesneriaceae. Wang et al. (2010) proposed that the evolution from zygomorphy to actinomorphy in these genera was possibly due to shifts in pollination strategies, such as a switch from nectar- to pollen-rewards and/or from specialist to generalist pollinators or vice versa. In our study, zygomorphy is generally more common in all regions. However, more than 20% of species in Hainan Island are actinomorphic, while less than one tenth of species in other regions are actinomorphic (Fig. 2). Such a higher proportion of actinomorphy on Hainan Island may represent an evolutionary trend associated with a switch to generalist pollinators (Gong & Huang, 2009; Wang et al., 2010).
Pollinators for mirror-image flowers are normally large-bodied bees such as Bombus andXylocopa (Jesson & Barrett, 2002; Gao et al., 2006), which can touch the deflected style and stamen simultaneously with different sides of their abdomens (Jesson & Barrett, 2002; Ren et al., 2013). Therefore, mirror-image flowers are a highly-specialized mechanism to increase the proficiency of cross-pollination by insect pollinators and limit sexual interference (Jesson & Barrett, 2002; Gao et al, 2006; Ren et al, 2013). All the three genera with mirror-image flowers are found on mainland China but only Paraboea is found on Hainan Island, where two Paraboea species occur. These two species are the widespread P. dictyoneura and Hainan-endemic P. changjiangensis; both lack mirror-image flowers (Xing et al., 2014; M.X.R., personal observations). These results further indicate the Hainan flora has less specialized pollination mechanisms that can be found in mainland relatives.
In summary, although Gesneriaceae species in Hainan Island are similar in most floral traits and pollination syndromes to the neighboring Chinese regions, species on this tropical island are more generalized in floral structures and pollination mechanisms, which might be an evolutionary outcome of a lower density of animal pollinators. The high species endemism in Hainan Island suggests this small island is a possible evolutionary hotspot for the Chinese Gesneriaceae, which is supported by recent phylogenetic studies (Wang et al., 2010; Wei et al., 2010; Perret et al., 2013).
Acknowledgements We thank Dr. Jordi Lopez-Pujol and You-Hai Shi for their help in early discussion of ideas and data collection. Financial supports arc provided by the National Natural Science Foundation of China (grants no. 31670230, 41661010) to M.X.R. and L.T.
APG (Angiosperm Phylogeny Group) III. 2009. An update of the angiosperm phylogeny group classification for the orders and families of flowering plants: APG III. Botanical Journal of Linnean Society, 161, 105-121.
Armbruster, W. S., Y. B. Gong, & S. Q. Huang. 2011. Are pollination "syndromes" predictive? Asian Dalechampia fit neotropical models. The American Naturalist, 178(1), 135-143.
Barrett, S.C.H., B. Emerson & J. Mallet. 1996. The reproductive biology and genetics of island plants. Philosophical Transactions of the Royal Society of London Series B: Biological Sciences, 351, 725-733.
Bascompte, J., P. Jordano & J.M. Olesen. 2006. Asymmetric coevolutionary networks facilitate biodiversity maintenance. Science, 312, 431-433.
Burtt, B.L. 1998. Climatic accommodation and phytogeography of the Gesneriaceae of the old world. In: Mathew P, Sivadasan M ed(s). Diversity and taxonomy of tropical flowering plants. Calicut, India: Mentor Books pp. 1-27.
Campos, E.O., D. B. Harvey & L.D. Thomas. 2015. Shape matters: corolla curvature improves nectar discovery in the hawkmoth Manduca sexta. Functional Ecology, 29, 462-468.
Carlquist, S. 1974. Island biology. Columbia University Press: New York & London.
Drummond, A. J. & A. Rambaut. 2007. BEAST: Bayesian evolutionary analysis by sampling trees. BMC evolutionary biology, 7(1), 214.
Fenster, C. B., W. S. Armbruster, P. Wilson, M. R. Dudash & J. D. Thomson. 2004. Pollination syndromes and floral specialization. Annual Review of Ecology, Evolution, and Systematics, 35,375-403.
Gao, J.Y., P.Y. Ren, Z.H. Yang & Q.J. Li. 2006. The pollination ecology of Paraboea rufescens (Gesneriaceae), a buzz-pollinated tropical herb with mirror-image flowers. Annals of Botany, 97,371-376.
Gong, Y.B. & S.Q. Huang. 2009. Floral symmetry: pollinator-mediated stabilizing selection on flower size in bilateral species. Proceedings of the Royal Society of London, Serial B, 276, 4013-4020.
Grant, V. 1950. The flower constancy of bees. The Botanical Review, 16(7), 379-398.
Guo, Y. F. & Y.Q. Wang. 2014. Floral ecology of Oreocharis pumila (Gesneriaceac): a novel case of sigmoid corolla. Nordic Journal of Botany, 32(2), 215-221.
--, --& A. Weber. 2013. Floral ccology of Oreocharis acaulis (Gesncriaceae): an exceptional case of "preanthetic" protogyny combined with approach herkogamy. Flora, 208(1), 58-67.
Hargreaves, A. L., S. D. Johnson & E. Nol. 2004. Do floral syndromes predict specialization in plant pollination systems? An experimental test in an 'ornithophilous' African Protca. Oecologia, 140(2), 295-301.
Jesson, L. K. & S. C. H. Barrett 2002. Enantiostyly: solving the puzzle of mirror-image flowers. Nature 417,707.
Kumar, S., M. Nei, J. Dudley & K. Tamur. 2008. MEGA: a biologist-centric software for evolutionary analysis of DNA and protein scquences. Briefings in Bioinformatics, 9(4), 299-306.
Li, Z. Y. & Y. Z. Wang. 2005. Plants of Gesneriaceae in China. Henan Science and Technology Publishing House, Zhengzhou.
Liang, G.H. 2013. Eight evidences on Hainan Island separating from China's Beibuwan Gulf with rotation. Acta Geologica Sinica, 87, 73-76.
Lopez-Pujol, J., F.M. Zhang & S. Ge. 2006. Plant biodiversity in China: richly varied, endangered, and in need of conservation. Biodiversity & Conservation,15(12), 3983-4026.
Lord, J.M., L. Huggins, L. M. Little & V.R. Tomlinson. 2013. Floral biology and flower visitors on subantarctic Campbell Island. New Zealand Journal of Botany, 51(3), 168-180.
Marten-Rodriguez, S., A. Almarales-Castro & C. B. Fenster. 2009. Evaluation of pollination syndromes in Antillean Gesneriaccae: evidence for bat, hummingbird and generalized flowers. Journal of Ecology, 97(2), 348-359.
--, M. Quesada, A. A. Castro, M. Lopezaraiza-Mikel & C. B. Fenster. 2015. A comparison of reproductive strategics between island and mainland Caribbean Gesneriaceae. Journal of Ecology, 103(5), 1190-1204.
Moller, M., A. Forrest, Y.G. Wei & A. Weber. 2010. A molecular phylogenetic assessment of the advanced Asiatic and Malcsian didymocarpoid Gesneriaceae with focus on non-monophyletic and monotypic genera. Plant Systematics and Evolution, 292, 223-248.
Pauw, A. 2006. Floral syndromes accurately predict pollination by a specialized oil-collecting bee (Rediviva Peringueyi, Melittidac) in a guild of south African orchids (Coryciinae). American Journal of Botany, 93(6), 917-926.
Perret, M., Chautems, A., de Araujo, A.O., Salamin, N. 2013. Temporal and spatial origin of Gesncriaceae in the new world inferred from plastid DNA sequences. Botanical Journal of the Linnacan Socicty, 171, 61-79.
Pu, G.Z., Y.M. Pan, S.C.Tang, C.Q. Wei & Q.X. Chen. 2009. Pollination biology and reproductive allocation of Chirita gueilinensis (Gesneriaceae). Bulletin of Botanical Research, 29(2), 169-175.
Ren, M.X. 2015. The upper reaches of the largest river in southern China as an 'evolutionary front' of tropical plants: evidences from Asia-endemic genus Hiptage (Malpighiaccae). Collectanea Botanica, 34, c003.
--, Y.F. Zhong & X.Q. Song. 2013. Mirror-image flowers without buzz pollination in the Asian endemic Hiptage benghalensis (Malpighiaceac). Botanical Journal of the Linnean Society, 173, 764-774.
Rodriguez-Girones, M.A & L. Santamaria. 2004. Why are so many bird flowers red? PLoS biology, 2(10), e350.
Rosas-Guerrero, V., R. Aguilar, S. Marten-Rodriguez, L. Ashworth, M. Lopezaraiza-Mikel, J. M. Bastida & M. Quesada. 2014. A quantitative review of pollination syndromes: do floral traits predict effective pollinators? Ecology letters, 17(3), 388-400.
Sargent, R.D. 2004. Floral symmetry affects speciation rates in angiosperms. Proceedings of the Royal Society of London, Serial B, 271, 603-608.
Tang, Z., Wang, Z., Zheng, C. & Fang, J. 2006. Biodiversity in China's mountains. Frontiers in Ecology and the Environment 4: 347-352.
Tang, S.C., G.Z. Pu, Pan, Y.M. R. Zhou, & C.Q. Wei. 2009. Pollination biology of Chirita lutea Y. Liu et YG Wei in China (Gesneriaceae). Journal of Tropical and Subtropical Botany, 17(4), 328-333.
Traveset, A., C. Tur, K. Trajelsgaard, R. Heleno, R. Castro-Urgal & J. M. Olesen. 2015. Global patterns of mainland and insular pollination networks. Global Ecology and Biogeography, 1-11.
Vaidya, G., D.J., Lohman & R. Meier. 2011. SequenccMatrix: concatenation software for the fast assembly of multi-gene datasets with character set and codon information. Cladistics, 27(2), 171-180.
Wang, C. N., M. Moller & Q.C. Cronk. 2004. Aspects of sexual failure in the reproductive processes of a rare bulbiliferous plant, Titanotrichum oldhamii (Gesneriaceae), in subtropical Asia. Sexual Plant Reproduction, 17(1), 23-31.
Wang, Y.Z., R.H. Liang, B.H. Wang, J.M. Li, Z.J. Qiu, Z.Y. Li & A. Weber. 2010. Origin and phylogenetic relationships of the old world Gesneriaceae with actinomorphic flowers inferred from ITS and trnL-trnF sequences. Taxon, 59(4), 1044-1052.
Wang, Y. B., H.W. Liang, N.B. Mo, K.P. Qin & G.G. Tang. 2011. Flower phenology and breeding system of rare and endangered Dayaoshania cotinifolia. Acta Bot Boreal Occident Sin, 31, 861-867.
Weber, A., D.J. Middleton, A. Forrest, R. Kiew, C.L. Lim, A.R. Rafidah, S. Susanne, T. Pramote, Y.G. Wei, T.L. Yao & M. Moller. 2011. Molecular systematics and remodelling of Chirita and associated genera (Gesneriaceae). Taxon, 60(3), 767-790.
Wei, Y.G., S.H. Zhong & H.Q. Wen. 2004. Studies of the flora and ecology Gesneriaceae in Guangxi Province. Acta Botanica Yunnanica, 26, 173-182
Wei, Y.G., F. Wen, M. Moller, A. Monro, Q. Zhang, Q. Gao, H.F. Mou, S.H. Zhong & C. Cui. 2010. Gesneriaceae of South China. Guangxi Sciences and Technology Publishing House, Nanning.
Wen, F., L.F. Fu, & Y.G. Wei. 2012. Pollination biology of Primulina repanda Var. guilinensis and P. glandulosa Var. yangshuoensis. Guihaia, 32(5), 571-578.
Xing, F.W., Chen, H.F X.S. Qin, R.J. Zhang, J.S. Zhou. 2014. Illustrated handbook of plants in tropical rainforest area of China: plants of Hainan (volume 3). Huazhong University of Science and Technology Publishing House, Wuhan.
Zhang, D.Y. 2004. Plant reproductive ecology and life-history evolution. Science Press, Beijing.
Zhu, H. 2016. Biogeographical evidences help revealing the origin of Hainan Island. PLoS ONE, 11(4), e0151941. doi:l 0.1371/journal
Shao-Jun Ling (1) * Qian-Wan Meng (1) * Liang Tang (1) * Ming-Xun Ren (1,2)
(1) Hainan Key Laboratory of Sustainabe Utilization of Tropical Bioresources, Institute of Tropical Agriculture and Forestry, Hainan University, Haikou 570228, China
(2) Author for Correspondence; e-mail: email@example.com
Published online: 2 March 2017
Caption: Fig. 1 Phylogcnetic relationship of representative Gesneriaceae species from Hainan Island and nearby regions. The phylogeny tree was drawn in BEAST software based on ITS and trn L-F sequences. The inset is the distribution map of Asian Gesneriaceac specics from http://www.gbif.org/. The mapped traits on the phylogenetic tree branches are: [DELTA] Zygomorphic corolla; [??] Actinomorphic corolla (the color corresponding to the corolla color); [arrow up] Exeretcd stamen; [down arrow]. Included stamen; [theta] Gular stamen (anther at the corolla throat). B (Bumble bees); H (Honeybees); D (bird); A (Amegilla). Branch color indicates geographic distribution
Caption: Fig. 2 Corolla color, ornamentation and symmetry of Gesneriaceae species on Hainan Island (HN) and mainland China (including Yunnan Province (YN), Guangxi Province (GX), Guizhou Province (GZ) and Guangdong Province (GD))
Caption: Fig. 3 Corolla types a and their occurrenccs b of Gesneriaceae species on Hainan Island (HN) and mainland China (see maintcxt and Fig. 2 for region abbreviations)
Caption: Fig. 4 Corolla constriction, curvature, stamen number, and anther location of Gesneriaceae species between Hainan Island (HN) and mainland China (see maintext and Fig. 2 for region abbreviations)
Caption: Fig. 5 habit (herb, shurb, liana) and Habitat (limestone, nonlimestone, both) of Gesneriaceae species between Hainan Island (HN) and mainland China (see maintext and Fig. 2 for region abbreviations)
Table 1 Geographic and biodiversity data for Gesncriaccae in Hainan Island, Yunnan, Guangxi, Guizhou and Guangdong provinces Region Area ([10.sup.4] x Elevation [km.sup.2]) range (m) Hainan Island (HN) 3.54 1800 Yunnan Province (YN) 39.00 6664 Guangxi Province (GX) 23.67 2141 Guizhou Province (GZ) 17.62 2793 Guangdong Province (GD) 17.98 1882 Region NT NE En SD EI (%) Hainan Island (HN) 24 8 33 12.96 0.27 Yunnan Province (YN) 236 106 45 42.44 0.147 Guangxi Province (GX) 211 120 57 53.75 0.336 Guizhou Province (GZ) 97 28 29 24.90 0.104 Guangdong Province (GD) 60 20 33 16.99 0.142 NT is the number of total species; SD (species density) = NT/ [ln(area) + ln(elevation range)]; NE is the number of endemic species; EI (endemism index) = [NE/(NT--NE)]/[ln(area) + ln(elevation range)]; En is Endemic ratio (percentage of endemic species) Table 2 Data for Chinese Gesneriaeeae with known pollinators Species Pollinator Titanotrichum oldhamii Ceratina flavipes, Notocrypta curvifascia, Xylocopa sp. Paraboea rufescens Amegilla malaccensis, Nomia sp. Chirita gueilinensis Amegilla albigena Chirita lutea Apidae sp., Crabro sp., Amegilla sp. Primulina repanda Chalcididae sp., Lasioglossum sp. \dtguilinensis Primulina glandulosa Chalcididae sp., Lasioglossum sp. var yangshuoensis Davaoshania cotinifolia Chalcididae sp" Eumenes sp., Sataspes tagalica Oreocharis acaulis Bombus sp. Oreocharis pumUa Bombus sp. Chirita heterotricha # Amegilla sp. Metapetrocosmea peltata # Amegilla sp. Oreocharis dasyantha Bombus sp. var. ferruginose # Paraboea hainanensis # Bombus sp. Species with # arc Hainan-endemic Species Family Locality Titanotrichum oldhamii Apidae, Hesperiidae Taiwan Paraboea rufescens Apidae, Halictidae Yunnan Chirita gueilinensis Halictidae Guangxi Chirita lutea Crabronidae, Apidae Guangxi Primulina repanda Chalcididae, Halictidae Guangxi \dtguilinensis Primulina glandulosa Chalcididae, Halictidae Guangxi var yangshuoensis Davaoshania cotinifolia Chalcididae, Guangxi Vcspidac, Sphingidae Oreocharis acaulis Apidae Guangdong Oreocharis pumUa Apidae Guangdong Chirita heterotricha # Apidae Hainan Island Metapetrocosmea peltata # Apidae Hainan Island Oreocharis dasyantha Apidae Hainan Island var. ferruginose # Paraboea hainanensis # Apidae Hainan Island Species with # arc Hainan-endemic Species Reference Titanotrichum oldhamii Wang et al., 2004 Paraboea rufescens Gao et al.. 2006 Chirita gueilinensis Pu et al., 2009 Chirita lutea Tang et al., 2009 Primulina repanda Wen et al., 2012 \dtguilinensis Primulina glandulosa Wen et al., 2012 var yangshuoensis Davaoshania cotinifolia Wang et al., 2011 Oreocharis acaulis Guo et al., 2013 Oreocharis pumUa Guo & Wang, 2014 Chirita heterotricha # Shao-Jun Ling and Qian-Wan Meng, personal observations Metapetrocosmea peltata # Shao-Jun Ling and Qian-Wan Meng, personal observations Oreocharis dasyantha Shao-Jun Ling and var. ferruginose # Qian-Wan Meng, personal observations Paraboea hainanensis # Shao-Jun Ling and Qian-Wan Meng, personal observations Species with # arc Hainan-endemic
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Hainan Island, China|
|Author:||Ling, Shao-Jun; Meng, Qian-Wan; Tang, Liang; Ren, Ming-Xun|
|Publication:||The Botanical Review|
|Date:||Mar 1, 2017|
|Previous Article:||Phytogeography and floristic affinities of the limestone flora of Mt. Exianling, Hainan Island, China.|
|Next Article:||Impacts of host trees and sowing conditions on germination success and a simple ex situ approach to generate symbiotic seedlings of a rare epiphytic...|