ABSTRACT
The well-known orchid genus Paphiopedilum has attracted much attention from biologists because of its diverse floral traits. Although these traits have been thoroughly described, little is known about their evolutionary trajectory. In this study, we explored their evolutionary patterns and trajectory through phylogenetic analyses and close observations, and 10 characters in 21 Chinese species mapped onto an existing phylogenetic tree. Lip shape, staminode shape, petal shape, and petal width are relatively congruent with molecular phylogenies, thereby validating the existing traditional classification system. All four of those characters, along with flower number, are strongly conserved, and are significantly affected by phylogeny. By contrast, flower color (including that of the dorsal sepal, lip, and petal) is significantly convergent among those examined species and less affected by phylogeny. Therefore, this character is independent of evolution and mainly influenced by environmental factors. All of these characters are key, classical indicators when distinguishing among species within the subgenera Brachypetalum and Paphiopedilum.
KEYWORDS: Characters, Paphiopedilum, phylogenetic analysis, pollination syndrome, trait evolution
Introduction
Floral traits can change many times during speciation to enable flowering plants to adapt to fluctuating environments and continue to reproduce successfully.1,2 For example, this shift can be a consequence of exposure to different pollinator groups.1,3 Mapping these traits onto phylogenies is one of the most popular methods for studying potential pathways of adaptive evolution.1,4,5 This approach is particularly appropriate for orchid plants because fossil records have provided few insights into the origins of their specific relationships.6
Species from the genus Paphiopedilum, within Orchidaceae, vary in their leaf morphology and floral and ecological characters.7,8 Lip, staminode, petal shape are important traits of traditional classification among species of this genus.7 The remarkable diversity found in Paphiopedilum makes it an ideal plant group for testing evolutionary shifts. Traits have already been described for Chinese species of Paphiopedilum.7,9 However, analyses of pollination ecology have been conducted for only a few species within that genus.10-12 ) and phylogenetic studies on the evolution of their plant traits are scarce.8 This shortcoming is probably not due to a lack of examples for trait diversity within genera or families, but rather because botanists tend to study single species in detail, without a comparative perspective.
Bees and hoverflies are the main pollinators of Chinese orchids and drive the diversification of Paphiopedilum (Banziger, 1996; Shi et al., 2006, 2009; Liu et al. 2009; Tang et al. 2014). Most species of Paphiopedilum in China are reported as growing terrestrially or facultative eiphytically, fewer species are found to be lithophytes (P. villosum and P. parishii). Paphiopedilum species can survive well in well-ventilated, well drained and moist environments (Liu et al. 2009). The large, uniquely beautiful flowers of Paphiopedilum contribute significantly to floriculture industries, and these species have also attracted much attention from evolutionary biologists.13,14,15 However, their popularity as ornamental plants has led to over-collection from the wild for commercial purposes. Likewise, the destruction and loss of habitat has meant that all species of the genus are endangered and some even face extinction.16 Therefore, knowledge about the evolution of traits in Paphiopedilum is essential for their conservation and introduction into the ornamental trade. However, rigorous tests of the relationships between their reproductive/vegetative traits and evolutionary trends have not previously been available.
Here, we focused on 21 Chinese species of Paphiopedilum, a well-known slipper orchid genus. Our investigation utilized a molecular phylogeny of angiosperms for six gene regions: the chloroplast genes atpB, matK, trnK, trnL, rbcL, and nuclear ribosomal ITS from Gen Bank.17 that allowed us to map 10 characters onto a phylogenetic framework. The study objective was to examine existing classifications, phylogenetic relationships, and evolutionary patterns so that we could re-evaluate and analyze the pathways of trait evolution in this genus and determine how those traits are influenced by pollinator systems.
Results
For most of these 21 Paphiopedilum species (Fig. 1), color of the floral organs (i.e., dorsal sepal, lip and petal) is significantly convergent. Except for P. dianthum and P. parishii, which have multi-flowered inflorescences, Chinese species of this genus are characterized by single-flowered inflorescences, and are rarely twin-flowered.(Fig. 2)
Figure 1.
Twenty Paphiopedilum species: A, P. appletonianum; B, P. insigne; C, P. helenae; D, P. villosum; E, P. charlesworthii; F, P. tigrinum; G, P. tranlienianum; H, P. barbigerum; I, P. spicerianum; J, P. dianthum; K, P. parishii; L, P. purpuratum; M, P. wardii; N, P. bellatulum; O, P. emersonii; P, P. venustum;7 Q, P. micranthum; R, P. armeniacum; S, P. malipoense; T, P. hangianum; and U, P. delenatii.
Figure 2.
Evolution of flower number for 21 species of Paphiopedilum.
Based on shape and structure, the lip can be easily divided into two distinct types: 1) subglobose, ellipsoid, or ovoid; and 2) helmet- or slipper-shaped. The former represents a plesiomorphy of this genus, with the evolutionary pattern of lip shape going from subglobose → ellipsoid or ovoid → helmet-shaped or slipper-shaped (Fig. 3A). Lip shape among these species demonstrates significant phylogenetic conservatism (K = 3.491, P = 0.001).
Figure 3.
Evolution of lip shape (A) and staminode shape (B) for 21 species of Paphiopedilum.
Staminode shape can be divided into seven progressive types: suborbicular or subsquare, rhombic-ovate, conduplicate, obcordate, obcordate to lunate, lunate and obovate (Fig. 3B) that its evolutionary trajectory was from suborbicular or subsquare → rhombic-ovate → conduplicate → obcordate → obcordate to lunate → lunate → obovate, and the staminode shapes displayed strong phylogenetic conservatism (K = 5.249, P = 0.001).
Evenly shaped petals have evolved to an uneven shape (Fig. 4A). The broad-shaped petal is a plesiomorphy of this genus that has shifted to more slender petals (Fig. 4B). Both petal shape and width show significant phylogenetic conservatism (shape: K = 1.265, P = 0.001; width: K = 3.491, P = 0.001).
Figure 4.
Evolution of petal shape (A) and petal width (B) for 21 species of Paphiopedilum.
Leaves from these species generally are narrowly oblong or lorate. Based on the presence or absence of gridding spots on their surfaces, leaves can be either tessellated or solid green, respectively. Most species within subgenus Brachypetalum have gridding spots that indicate tessellation (the exceptions being P. hangianum and P. emersonii, which have plain green leaves). Within subgenus Paphiopedilum, the solid-green leaves are characteristic for the early diverging section Pardalopetalum but is absent in species from section Barbata. Reversions of this trait occur in all species of section Paphiopedilum. Overall, the tessellated leaf represents plesiomorphism, with appearance shifting from tessellated to solid green (Fig. 5A). Paphiopedilum species either have distinct rhizomes (tufted growth form) or do not have obvious rhizomes (scattered growth form). The tufted growth form represents plesiomorphism, with a general evolutionary shift from the tufted to scattered growth from.(Fig. 5B)
Figure 5.
Evolution of leaf type (A) and growth form (B) for 21 species of Paphiopedilum.
Our PCA indicated that species within subgenus Paphiopedilum clustered in the negative field on the PC1 axis while those in subgenus Brachypetalum clustered in the positive field on that axis (Fig. 6). The first axis accounted for 45.28% of the explained variance and separated the species between those two subgenera. The second axis explained 17.5% of the variance separated all tufted species (except P. spicerianum) from scattered-form species (except P. villosum). Therefore, the second axis appeared to provide a reasonable representation of differences in growth form for species within subgenus Paphiopedilum.
Figure 6.
Plot of principal component analysis for 21 species of Paphiopedilum.
Discussion
We analyzed the evolution of 10 traits from 21 Paphiopedilum species sampled in China, using these combined data to map the traits onto a phylogenetic framework. This approach enabled us to display their evolutionary patterns and identify the synapomorphy for different clades. Overall, the shape of the leaves, staminodes, and petals, as well as petal width and number of flowers per inflorescence show strong phylogenetic conservatism and distinct evolutionary shift. Flower color is significantly convergent among those examined species, the flower color character has a weak signal, showing strong convergence, possibly because of a departure from Brownian motion evolution, e.g., adaptive evolution, and less affected by phylogeny, rather than more affected by environment. Flowers of these slipper orchids in China are thought to be pollinated by deceit, usually relying upon morphology and scent to attract pollinators.11,12,18,19 Therefore, they may reduce their investment in flower color. Moreover, color as perceived by human observers is likely to be quite different than that perceived by pollinators.20 and what appears striking to the human eye may be meaningless to pollinators.21In contrast to the conservatism found with the LS, SS, PS, PW, and FN characters.
In contrast, the floral character of lip shape for Paphiopedilum flowers is very important for insect pollination.11,14 Here, two shapes occur: subglobose/ovate/subsquare or slipper-/helmet-shaped. The former is similar to that associated with the genus Cypripedium, and also occurs in the species of subgenus Brachypetalum examined here, i.e., P. bellatulum, P. armeniacum, P. delenatii, P. emersonii, P. hangianum, P. malipoense, and P. micranthum. These similarities between the flowers of Paphiopedilum and Cypripedium (for example, P. armeniacum and C. irapeanum), are a result of having similar pollination syndromes that involve bees as pollinators.13 However, ten species of Paphiopedilum are pollinated by hoverflies (Table 2.).10-12,22-25 The subglobose-shaped lip, with involute margins apically, may help prevent access by insects with larger bodies. Members of subgenus Paphiopedilum include P. parishii, P. dianthum, P. appletonianum, P. wardii, P. venustum, P. purpuratum, P. spicerianum, P. insigne, P. barbigerum, P. tranlienianum, P. tigrinum, P. charlesworthii, P. helenae, and P. villosum, all of which have helmet-shaped lips that have large, deep mouths but no involute margins. Obviously, this structure is beneficial because it prevents smaller-sized insects from escaping. The shift in morphology to helmet-shaped lips among species within the Paphiopedilum genus may, therefore, be an adaptation that makes these plants more accessible to pollinators with smaller bodies.7 Several studies have successfully demonstrated that variation in floral traits among groups is a consequence of adaptation to different pollinators.7,11,12 However, another research has found no evidence that pollinator specialization has driven speciation in the Orchidaceae.26
Table 2.
Matrix of morphological characters for 21 species of Paphiopedilum. Character numbers and values for trait states are defined in Table 1.
| Character | States and codings |
|---|---|
| 1. Lip shape | 0, sub-globose or ellipsoid or ovoid; 1, helmet-shaped |
| 2. Staminode shape | 0, sub-orbicular or sub-square; 1, rhombic-ovate; 2, conduplicate; 3, obcordate; 4, obcordate to lunate; 5, lunate; 6, obovate |
| 3. Petal shape | 0, even; 1, uneven |
| 4. Petal width | 0, broad; 1, narrow |
| 5. Dorsal sepal color | 0, light (green, yellow, or white); 1, not light |
| 6. Lip color | 0, light ( green, yellow, or white); 1, not light |
| 7. Petal color | 0, light (green, yellow, or white); 1, not light |
| 8. Leaf type | 0, tessellated; 1, solid green |
| 9. Growth form | 0, tufted; 1, scattered |
| 10. Flower number | 0, single-flowered inflorescence; 1, multi-flowered inflorescence |
Staminode shape exhibits high morphological diversity within Paphiopedilum, playing an important role in species identification.7 This character shows strong phylogenetic conservatism and may have an important role in luring pollinators. Whereas species within the original subgenus Brachypetalum have broad and even petals, species within subgenus Paphiopedilum have narrower, more uneven petals.
Variations in leaf types are considered an adaptation to different growing environments and a strategy for reducing damage by herbivores.27,28 However, the ecological significance of petal form has rarely been reported and is poorly understood. Thus, further investigation is needed to show whether the latter character has a function similar to that of leaf characters or other petal morphology with regard to pollinator attraction.
The tessellated leaf trait found in most of the early-diverging species within subgenus Brachypetalum is absent in most species of subgenus Paphiopedilum. Tessellated leaves are thought to serve as anti-herbivory camouflage in understory herbaceous plants that grow under sun-flecked light conditions.29 However, no evidence exists for the significance of such an adaptation in Paphiopedilum. Most of the species within that genus, whether their leaves are solid green or tessellated, occur in similar shady forest-floor habitats. However, a few species with plain green leaves have been found in open, sunny areas while some species with tessellated leaves live in deep shade.14
All species examined here in subgenus Brachypetalum exist in tufts while most species of subgenus Paphiopedilum are scattered. Chinese Paphiopedilum species generally occur on shrubby slopes or at forest edges, but can often be found as well in moist, warm environments with good drainage. Overall, our analytical results are in accord with those reported previous study.9 Nevertheless, we have not been able to determine the value of this evolutionary pattern in Paphiopedilum.
Our PCA analysis based on 10 characters revealed a clear division into two subgenera: Brachypetalum and Paphiopedilum. Axis 1 separated the species in two groups. Plants in the first group have tessellated leaves; one- or (rarely) two-flowered inflorescences; green, yellow, or white flowers; broad and even petals; ovoid-shaped lips with involuted margins; and a staminode that is uni- or tridentate at its apex. Plants in the second group have solid-green leaves; single- or multi-flowered inflorescences; long, tapering petals; and helmet-shaped lips. These traits are key classical features that describe the plant characters for subgenus Brachypetalum and subgenus Paphiopedilum.
In conclusion, our comprehensive investigation concentrated on reproductive and vegetative characters and examined the patterns of trait evolution for 21 Chinese Paphiopedilum species. Phylogeny has had a significant effect on these characters, with lip shape, staminode shape, petal shape, petal width, and flower number being strongly conserved. In contrast, flower color (including that of the dorsal sepal, lip, and petal) of these species is significantly convergent. Striking differences in these traits are found between plants within subgenus Brachypetalum and subgenus Paphiopedilum. Our findings suggest that these species have adapted to variations in environmental conditions and pollination syndromes. Further studies should focus on explaining and testing the significance of ecological adaptations as they relate to the evolutionary patterns presented here.
Material and methods
Study material
The genus Paphiopedilum within Orchidaceae comprises 79 species.7 Plants usually occur in limestone or mountainous forests of tropical and subtropical regions that range from India and southern China through Southeast Asia and the Malesian islands to the Solomon Islands.14 In China, 27 species are found, with most being distributed in southern Yunnan and southwestern Guanxi provinces.7 Most of the species are terrestrial, some are epiphytic or lithophytic.14 This genus is distinguished by high floral diversity and intricate adaptations to pollinators. China is one of diversity distribution centers,7 representing plant traits of the whole genus Paphiopedilum. Molecular phylogenetic studies have confirmed that Paphiopedilum is a monophyletic genus.15,30
Analyses of phylogenetics and trait evolution
We used the online version of Phylomatic17 to obtain a molecular phylogenetic tree of our sampled taxa. Data for 10 characters of 21 Paphiopedilum species in China (Table 2.) were compiled from our observations and the literature.7,9 Those characters included lip shape (LS), staminode shape (SS), petal shape (PS), petal width (PW), dorsal sepal color (DSC), lip color (LC), petal color (PC), leaf type (LT), growth form (GF), and flower number (FN). And this genus has high diversity in flower color, so type of flower color were divided into two states (light vs. no light) to be analyzed. To evaluate patterns of trait evolution, we developed a matrix of morphological characters for each species (Table 3.). A parsimonious analysis was then performed with Mesquite v. 2.6 software.31
Table 3.
Known pollinators of Paphiopedilum species.
| Character and coding value assigned for trait state |
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Species | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| appletonianum | 1 | 4 | 0 | 1 | 0 | 1 | 1 | 0 | 1 | 0 |
| armeniacum | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| barbigerum | 1 | 6 | 1 | 1 | 0 | 1 | 0 | 1 | 0 | 0 |
| bellatulum | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| charlesworthii | 1 | 6 | 1 | 1 | 1 | 0 | 0 | 1 | 1 | 0 |
| delenatii | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| dianthum | 1 | 3 | 1 | 1 | 0 | 0 | 0 | 1 | 0 | 1 |
| emersonii | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 |
| hangianum | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 | 0 | 0 |
| helenae | 1 | 6 | 1 | 1 | 0 | 0 | 0 | 1 | 0 | 0 |
| insigne | 1 | 6 | 1 | 1 | 0 | 0 | 0 | 1 | 0 | 0 |
| malipoense | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| micranthum | 0 | 2 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| parishii | 1 | 3 | 1 | 1 | 0 | 0 | 1 | 1 | 0 | 1 |
| purpuratum | 1 | 5 | 1 | 1 | 0 | 1 | 1 | 0 | 1 | 0 |
| spicerianum | 1 | 6 | 1 | 1 | 0 | 0 | 0 | 1 | 1 | 0 |
| tigrinum | 1 | 6 | 1 | 1 | 0 | 0 | 1 | 1 | 1 | 0 |
| tranlienianum | 1 | 6 | 1 | 1 | 0 | 0 | 0 | 1 | 1 | 0 |
| venustum | 1 | 5 | 1 | 1 | 0 | 0 | 0 | 0 | 1 | 0 |
| villosum | 1 | 6 | 1 | 1 | 0 | 1 | 1 | 1 | 0 | 0 |
| wardii | 1 | 5 | 0 | 1 | 0 | 0 | 0 | 0 | 1 | 0 |
Table 1.
Trait states and codings for orchid characters.
| Species | Pollinator (s) | Reference (s) |
|---|---|---|
| armeniacum | Hoverfly and bee | Liu et al., 2005 |
| barbigerum | hoverfly | Shi et al., 2009 |
| bellatulum | hoverfly | Banziger, 2002 |
| callosum | hoverfly | Banziger, 2002 |
| charlesworthii | hoverfly | Banziger, 1994; Cribb, 1998 |
| dianthum | hoverfly | Shi et al., 2006 |
| hirsutissimum | hoverfly | Shi, 2008 |
| malipoense | bee | Liu et al., 2009 |
| micranthum | bee | Shi, 2008 |
| purpuratum | hoverfly | Liu et al., 2004 |
| rothschildianum | hoverfly | Atwood, 1985 |
| villosum | hoverfly | Banziger, 1996 |
A phylogenetic signal (K) can be used to express the conservatism of traits. Cases where K < 1 indicate convergence,K = 1 implies that closely related species have trait values that agree completely with a Brownian model, and K > 1 represents traits that are more conserved than presumed from a Brownian expectation.32 Here, we calculated K values by using ‘picante’, based on the R package.33 A principal component analysis (PCA) was performed with the ‘prcomp’ function of the R ‘vegan’ to characterize the associations among Paphiopedilum species.
Disclosure of potential conflicts of interest
No potential conflicts of interest were disclosed.
Funding
This research was financially supported by the National Natural Science Foundation of China (31300200, 31370362) and the Natural Science Foundation of Yunnan Province (2013FA044).
References
- 1.Johnson SD, Linder HP, Steiner KE. Phylogeny and radiation of pollination systems in Disa (Orchidaceae). Am J Bot 1998; 85:402–11; PMID:21684924; http://dx.doi.org/10.2307/2446333 [PubMed] [Google Scholar]
- 2.Guan ZJ, Zhang SB, Guan KY, Guan KY, Li SY, Hu H. Leaf anatomical structures of Paphiopedilum and Cypripedium and their adaptive significance. J Plant Res 2011; 124:289–98; PMID:20711624; http://dx.doi.org/ 10.1007/s10265-010-0372-z [DOI] [PubMed] [Google Scholar]
- 3.Hapeman JR, Inoue K. Plant-pollinator interactions and floral radiation in Platanthera (Orchidaceae) In: Givnish TJ, Sytsma KJ, eds. Molecular Evolution and Adaptive Radiation. Cambridge, UK: Cambridge University Press; 1997; 433–54. [Google Scholar]
- 4.Armbruster WS. Evolution of plant pollination systems: hypotheses and tests with the neotropical vine Dalechampia. Evolution 1993; 47:1480–505; http://dx.doi.org/10.2307/2410162 [DOI] [PubMed] [Google Scholar]
- 5.Whittall JB, Hodges SA. Pollinator shifts drive increasingly long nectar spurs in columbine flowers. Nature 2007; 447:706–9; PMID:17554306; http://dx.doi.org/ 10.1038/nature05857 [DOI] [PubMed] [Google Scholar]
- 6.Ramirez SR, Gravendeel B, Singer RB, Marshall CR, Pierce NE. Dating the origin of the Orchidaceae from a fossil orchid with its pollinator. Nature 2007; 448:1042–45; PMID:17728756; http://dx.doi.org/doi: 10.1038/nature06039 [DOI] [PubMed] [Google Scholar]
- 7.Liu ZJ, Chen XQ, Chen LJ, Lei SP. The Genus Paphiopedilum in China. Beijing: Science Press; 2009. [Google Scholar]
- 8.Zhang SB, Guan ZJ, Sun M, Zhang JJ, Cao KF, Hu H. Evolutionary association of stomatal traits with leaf vein density in Paphiopedilum, Orchidaceae. PLoS One 2012; 7:e40080; PMID:22768224; http://dx.doi.org/ 10.1371/journal.pone.0040080 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Wang YQ. The geography of Chinese species of Paphiopedilum. Guihaia 2000; 20:289–94. [Google Scholar]
- 10.Banziger H. The mesmerizing wart: the pollination strategy of epiphytic lady slipper orchid Paphiopedilum villosum (Lindl.) Stein (Orchidaceae). Bot J Linn Soc 1996; 121:59–90; http://dx.doi.org/10.1111/j.1095-8339.1996.tb00745.x [Google Scholar]
- 11.Shi J, Cheng J, Luo D, et al. Pollination syndromes predict brood-site deceptive pollination by female hoverflies in Paphiopedilum dianthum (Orchidaceae). Acta Phytotaxonomica Sinica 2006; 45:551–60; http://dx.doi.org/ 10.1360/aps07025 [DOI] [Google Scholar]
- 12.Shi J, Luo YB, Bernhardt P, Ran JC, Liu ZJ, Zhou Q. Pollination by deceit in Paphiopedilum barbigerum (Orchidaceae): a staminode exploits the innate colour preferences of hoverflies (Syrphidae). Plant Biol 2009; 11:17–28; PMID:19121110; http://dx.doi.org/ 10.1111/j.1438-8677.2008.00120.x [DOI] [PubMed] [Google Scholar]
- 13.Cribb PJ. The Genus Paphiopedilum. London: Collingridge; 1987. [Google Scholar]
- 14.Cribb PJ. The Genus Paphiopedilum (ed. 2). Kota Kinabalu (Sabah, Malaysia: ): Natural History Publications (Borneo) Sdn. Bhd. & R.B.G. Kew; 1998. [Google Scholar]
- 15.Chochai A, Leitch IJ, Ingrouille MJ, Fay MF. Molecular phylogenetics of Paphiopedilum (Cypripedioideae); Orchidaceae) based on nuclear ribosomal ITS and plastid sequences. Bot J Linn Soc 2012; 170:176–96; http://dx.doi.org/ 10.1111/j.1095-8339.2012.01293.x [DOI] [Google Scholar]
- 16.Yuan L, Yang ZL, Li SY, Hu H, Huang JL. Mycorrhizal specificity, reference, and plasticity of six slipper orchids from South Western China. Mycorrhiza 2010; 20:559–68; PMID:20217434; http://dx.doi.org/ 10.1007/s00572-010-0307-5 [DOI] [PubMed] [Google Scholar]
- 17.Smith SA, Beaulieu JM, Stamatakis A, Donoghue MJ. Understanding angiosperm diversification using small and large phylogenetic trees. Am J Bot 2011; 98:404–14; PMID:21613134; http://dx.doi.org/ 10.3732/ajb.1000481 [DOI] [PubMed] [Google Scholar]
- 18.Liu ZJ, Zhang JY, Ru ZZ, Lei SP, Chen LJ. Conservation biology of Paphiopedilum purpuratum (Orchidaceae). Biodiversity Sci 2004; 12:509–16. [Google Scholar]
- 19.Liu KW, Liu ZJ, Lei SP, Li LQ, Chen LJ, Zhang YT. Study on pollination biology in Paphiopedilum armeniacum (Orchidaceae). Shenzhen Sci Technol 2005; 11:171–83. [Google Scholar]
- 20.Kevan P, Giurfa M, Chittka L. Why are there so many and so few white flowers? Trends Plant Sci 1996; 1:280–84; http://dx.doi.org/ 10.1016/1360-1385(96)20008-1 [DOI] [Google Scholar]
- 21.Hensel LE, Sargent RD. A phylogenetic analysis of trait convergence in the spring flora. Botany 2012; 90:557–64; http://dx.doi.org/ 10.1139/B2012-029 [DOI] [Google Scholar]
- 22.Banziger H. Smart alecks and dumb flies: natural pollination of some wild lady slipper orchids (Paphiopedilum spp., Orchidaceae). In Clark J, Elliott WM, Tingley G, Biro J eds, Proceedings of the 16th World Orchid Conference, Vancouver BC: Vancouver Orchid Society 2002; 165–69. [Google Scholar]
- 23.Banziger H. Studies on the natural pollination of three species of wild lady-slipper orchids (Paphiopedilum) in Southeast Asia. In: Pridgeon A. ed, Proceedings of the 14th World Orchid Conference, Edinburgh: HMSO 1994; 201–02. [Google Scholar]
- 24.Shi J. 2008. Pollination biology of Paphiopedilum (Orchidaceae) in China Ph.D Dissertation. Beijing: Graduate University of Chinese Academy of Sciences; 2008. [Google Scholar]
- 25.Atwood JT. Pollination of Paphiopedilum rothschildianum: brood-site deception. National Geogr Res Spring 1985:247–54. [Google Scholar]
- 26.Gravendeel B, Smithson A, Slik FJW, Schuiteman A. Epiphytism and pollinator specialization: drivers for orchid diversity? Philos Trans R Soc Lond B 2004; 359:1523–35; PMID: 15519970; http://dx.doi.org/ 10.1098/rstb.2004.1529. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Givnish TJ. On the adaptive significance of leaf form In: Solbrig OT, Jain S, Johnson GB, Raven PH, eds. Topics in Plant Population Biology. London: MacMillan, 1979; 375–407. [Google Scholar]
- 28.Brown VK, Lawton JH, Grubb PJ. Herbivory and the evolution of leaf size and shape. Philos Trans R Soc Lond B 1991; 333:265–72; http://dx.doi.org/ 10.1098/rstb.1991.0076. [DOI] [Google Scholar]
- 29.Givnish TJ. Leaf mottling: relation to growth form and leaf phenology and possible role as camouflage. Funct Ecol 1990; 4:463–74; http://dx.doi.org/ 10.2307/2389314 [DOI] [Google Scholar]
- 30.Cox AV, Pridgeon AM, Albert VA, Chase MW. Phylogenetics of the slipper orchids (Cypripedioideae, Orchidaceae): nuclear rDNA ITS sequences. Plant Syst Evol 1997; 208:197–223; http://dx.doi.org/ 10.1007/BF00985442 [DOI] [Google Scholar]
- 31.Maddison WP, Maddison DR. Mesquite: A modular system for evolutionary analysis. Version 2. 6. Computer Program Documentation Distributed by the Author 2009; http://mesquiteproject.org/mesquite/mesquite. [Google Scholar]
- 32.Blomberg SP, Garland T Jr, Ives AR. Testing for phylogenetic signal in comparative data: behavioral traits are more labile. Evolution 2003; 57:717–45; PMID:12778543; http://dx.doi.org/ 10.1554/0014-3820(2003)057 [DOI] [PubMed] [Google Scholar]
- 33.R Development Core Team R: A Language and Environment for Statistical Computing, Vienna, Austria: The R Project for Statistical Computing website; 2011; http://www.R-project.org. [Google Scholar]