| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
2Departamento de Ciencias Ecológicas, Facultad de Ciencias, Universidad de Chile, Casilla 653, Santiago, Chile; 3Instituto de Ecología y Biodiversidad, Facultad de Ciencias, Universidad de Chile, Casilla 653, Santiago, Chile
Received for publication November 25, 2005. Accepted for publication April 12, 2006.
ABSTRACT
Concerted changes in flower morphology and pollinators provide strong evidence on adaptive evolution. Schizanthus (Solanaceae) has zygomorphic flowers and consists of 12 species of annual or biennial herbs that are distributed mainly in Chile and characterized by bee-, hummingbird-, and moth-pollination syndromes. To infer whether flowers diversified in relation to pollinator shifts, we traced the evolutionary trajectory of flower traits and visitors onto a phylogeny based on sequence data from ITS, waxy, and trnF/ndhJ DNA. Maximum-likelihood ancestral reconstruction of floral traits suggests that ancestral Schizanthus had a bee-pollination syndrome. The hummingbird syndrome evolved in S. grahamii, a high elevation species in the Andes. The moth syndrome evolved in the ancestor of three species that inhabit the Atacama Desert. Results of mapping flower visitors onto the phylogeny show that the shift from bee to hummingbird pollination concurred with a shift in pollinators as predicted by the syndromes. However, the same pattern was not found for the moth syndrome. Visits by moths were observed only in one of the three moth-syndrome species, and at a very low rate. This mismatch suggests either anachronic floral characters or maintenance of rare, imperceptible moth pollination backed up by capacity for autonomous selfing. Overall, results suggest that diversification of flower traits in Schizanthus has occurred in relation to pollinator shifts.
Key Words: bees • Chile • floral evolution • hummingbirds • molecular phylogeny • moths • pollination syndromes • Solanaceae
One of the major features of flowering plants is that particular suites of floral characters are associated with specific groups of animal pollinators (Stebbins, 1974
; Fægri and van der Pijl, 1979
; Pellmyr, 2002
). The idea that different pollinator syndromes (flower phenotypes specialized for different groups of pollinators) evolve from the direct influence of different pollinator species has become a central tenet in flowering plant evolution (e.g., Grant and Grant, 1965
; Stebbins, 1974
; Fægri and van der Pijl, 1979
). Implicit in the syndrome concept is the assumption that pollinator-mediated selection is the main force driving floral diversification into highly specialized morphologies (e.g., Johnson et al., 1998
; Goldblatt et al., 2001
). However, several studies have suggested that historical and random processes unrelated to flower adaptation and developmental constraints can also influence the differential representation of flower morphologies across taxa (e.g., Armbruster, 2002
; Herrera et al., 2002
). These results altogether have stimulated a reevaluation of the validity of the syndrome concept to predict pollination systems (e.g., Waser et al., 1996
; Johnson and Steiner, 2000
). One way to examine whether well-defined floral trait combinations evolved in relation to particular pollination systems consists of evaluating their concomitant changes and joint evolutionary trajectories in a phylogenetic context (Fenster et al., 2004
and cited references). This approach allows the examination of historical effects in floral evolution and removal of the variation in floral traits introduced by evolutionary relatedness (Fenster et al., 2004
). Despite these evident advantages, studies of this kind have been undertaken in a mere 12 groups of plants (Fenster et al., 2004
). In this paper, we study the relationship between flower evolution and pollinator service in the solanaceous genus Schizanthus using a phylogenetic approach.
The genus Schizanthus Ruiz and Pavón, endemic to the southern South American Andean region, is distributed between 22–40° S (Fig. 1). Molecular (Olmstead and Palmer, 1992
; Martins and Barkman, 2005
) and morphological data (Grau and Grönbach, 1984
) indicate that Schizanthus diverged early from the rest of the family Solanaceae, constituting the monogeneric tribe Schizanthoideae (Olmstead and Palmer, 1992
). Extant Schizanthus comprise 12 species of annual to sometimes biennial herbs that grow to 1 m high. They occur in diverse habitats including the desert, coastal, high Andean, and mediterranean-climate regions of Chile, with two species reaching the Argentinean side of the Andes. Flowering in most species occurs during the austral spring-summer, although northern desert populations emerge only in high precipitation years concurring with El Niño events. The floral morphology in the genus Schizanthus is unusual among members of Solanaceae. The flowers are bilabiate and strongly zygomorphic, resembling the papilionaceous flower (Fig. 1) (Walters, 1969
; Grau and Grönbach, 1984
; Knapp, 2002
). The corolla consists of five petals fused into a tube. The three uppermost petals constitute the upper lip, comprising a banner and two dissected lateral sections. The two lower petals are deeply dissected, and their inner portion is fused to form a keel that in some species retains the stamens prior to explosive pollen discharge (Cocucci, 1989
). The outer portions of the lower petals constitute the wings. Moth- and bee-pollination syndromes have been previously described by Coccuci (1989). The moth-pollination syndrome consists of white corollas with long tubes; backward-reflexed and highly dissected, lateral sections; reduced lower lips; and the absence of explosive pollen discharge. The bee-pollination syndrome consists of pink-purple corollas, nectar guides, nonreflexed lateral sections, lower lips extended as landing platforms, and the presence of explosive pollen discharge. In addition, a hummingbird-pollination syndrome was detected during this study, in which the flowers (with the exception of the banner) are red and the tubes long. Such flowers lack a landing platform and have non-explosive pollen discharge.
|
MATERIALS AND METHODS
Phylogenetic reconstruction
Leaves and floral buds of all 12 species of Schizanthus were collected in the field in Chile between 25–37° S and dried with silica gel immediately after collection. Information on location and study sites are shown in Table 1. Genomic DNA was extracted from each species using the DNEasy Plant mini kit (Quiagen, Hilden, Germany). The trnF/ndhJ regions of the cpDNA and the ITS region of nDNA were amplified by PCR in two stages following Hershkovitz and Zimmer (1997)
, i.e., a double-strand amplification followed by separate asymmetric amplification of each strand using primers internal to the first. In the case of ITS, the primer pair TTTCTTTTCCTCCGCTTA and GAAGGAGAAGTCGTAACAAG was used for the double-strand amplification and the internal primers ITS4 (White et al., 1990
) or N18L18 (Hershkovitz and Zimmer, 1997
) for the asymmetric amplifications. In the case of the trnF/ndhJ spacer region, the primer pair GYTGGTAGAGCAGAGGACTG and TGGATAGGATGGCCCTTAC was used for the double-strand amplification and the internal primers, either GGACTGAAAATCCTCGTGTC or TGGATAGGATGGCCCTTAC for the asymmetrical amplifications. A region of waxy spanning introns 4–8 of the corresponding sequence of Solanum tuberosum (van der Leij et al., 1991
) was amplified in one stage using the primer pair CTAYAARMGAGGGGTTGATCG and GCRTTCATCCAGTTGATTTTC.
|
Purified products were sequenced using the ABI Prism BigDye Cycle Sequencing Ready Reaction kit (Applied Biosystems, Foster City, California, USA) on an ABI 3100 sequencer. In the case of ITS and trnF/ndhJ, the single strand PCR products of both directions were sequenced using the internal primers described. In the case of waxy, the double-strand PCR products were sequenced in both directions using the internal primers GGCACACTGCTCTACTTCC or GRTAKGCAATGTTGTGGATGC. Sequences from both strands of each PCR product were examined, compared, and corrected using the program Bioedit (Hall, 1999
) from which a consensus sequence was generated. Sequence data were manually aligned in Bioedit.
Bayesian inference of phylogeny with Markov chain Monte Carlo sampling was conducted for the concatenated sequences of the three DNA regions using MrBayes 3.04 (Huelsenbeck and Ronquis, 2001
). A general time-reversible model of DNA substitution and shape parameter of the gamma distribution (GTR + G) model was used with parameters partitioned across the genes. One cold chain and three heated chains were run simultaneously for one million generations, and one tree per 100 generations was sampled. The first 200 trees were discarded as burn-in, and Bayesian posterior probabilities were estimated on the 50% majority rule consensus of the remaining 9800 trees. Trees were also constructed by maximum parsimony in PAUP* version 4.0 (Swofford, 1999
), and bootstrap support at nodes was computed for 1000 replicates of data (Felsenstein, 1985
). The three DNA regions were analyzed separately and in combination. Equal-weighted parsimony analyses were carried out by heuristic search with tree-bisection-reconnection (TBR) branch swapping and random sequence addition replicates. To test for conflicts between the different DNA data sets, we conducted an incongruence length difference test (ILD, Farris et al., 1985
) with 500 random partitions.
Because Schizanthus is a strongly isolated genus in the Solanaceae (Grau and Grönbach, 1984
; Tétényl, 1987
; Olmstead and Palmer, 1992
; Martins and Barkman, 2005
), and the risk of homoplasy is high when a highly divergent outgroup from the ingroup is used to root the trees (Swofford et al., 1996
; Graham et al., 2002
), we used the midpoint-rooting criterion. The midpoint procedure places the root between the most distant taxa under the assumption that molecular evolution between these taxa is clock-like. To test this assumption, likelihood ratio tests (Felsenstein, 1988
) were performed for each DNA sequence to compare the likelihood scores of data with and without forcing the molecular clock. The likelihood ratio was assumed to be 
2 distributed with N taxa minus two degrees of freedom. The molecular clock-like assumption was not rejected for any of the three data sequences, therefore justifying the use of the midpoint rooting.
Floral characters and visitors
One flower per each of 10 plants per population (one population per species) was collected in the field. The flowers were preserved in 70% alcohol (Table 1) and dissected later to separate the upper and lower lip so as to flatten the corolla. The flattened flowers were scanned, and the following morphological traits recorded using SigmaScan Pro 5.0 (SPSS, 1998
): (1) corolla size, obtained as the sum of the lower and upper lip areas; (2) degree of corolla dissection, calculated as the ratio between the squared perimeter of the lips and corolla size; (3) relative size of the lower lip, calculated as the ratio between the lower lip and corolla size; and (4) relative length of corolla tube, calculated as the ratio between the lengths of the corolla tube and the major axis of the corolla. In addition, the following characteristics were measured directly in the field: (5) lateral section orientation of the upper lip, (backward-reflexed or non-reflexed orientation); (6) corolla color (white: without evident nectar guides when examined in daylight; pink-purple: light pink to purple lateral sections and/or keels, and nectar guides, often with dark-purple spots; or red: red lateral sections and keel); (7) type of pollen presentation (explosive or non-explosive); (8) nectar volume, obtained from 10 different female-phase bagged flowers at 1000, 1600, and 2300 using microcapillary tubes (1 µL and 5 µL). Nectar measurement times were selected so as to cover the range of daily times of nectar secretion found for different pollination syndromes.
Flower visitors were observed over three sunny days for nine species (one population per species). In the case of moth-syndrome species, observations were extended up to 6 d because of the low visitation rates. For two populations of S. hookeri and three of S. grahamii were observed. In the remaining species, only one population could be studied because most of these species only appear abundantly during El Niño years (in our case, the spring season of 2000 and 2002). Temporal considerations prevented direct observations of pollinators of S. parvulus, S. laetus, and S. litoralis. Flower visitors were observed directly or by using binoculars in 2-m2 patches for one 30-min period per hour between 0900 and 2400 hours. For the dark hours, observations were performed under a red light source. Total flowers observed per species per 30 min ranged from 207 to 617. Flower visitors were trapped for identification, grouped as bees, dipterans, and lepidopterans, and identified to the maximum resolution possible. The effectiveness of flower visitors for bees and dipterans was judged by observing the presence or absence of Schizanthus pollen on the insect body. We did not trap hummingbirds. However, the flower-handling pattern of the white-sided hillstar, Oreotrochilus leucopleurus, the only hummingbird observed, was consistent with effective pollination. Because of their extremely low visitation rates, we were unable to trap moths. The effectiveness of moths as pollinators on Schizanthus is less certain.
Ancestral reconstruction of floral traits and flower visitors
Maximum-likelihood ancestral reconstructions of floral characters and flower visitors were undertaken using the majority rule consensus tree recovered from the Bayesian analysis of the combined nuclear and chloroplast dataset. The only polytomy detected in the consensus tree was resolved by assigning a branch length of 0.00001. Ancestral states of qualitative floral traits were reconstructed using Mesquite (Maddison and Maddison, 2004
) based on a one-parameter model. Ancestral states of continuous floral characters were estimated in ANCML (Schluter et al., 1997
), which assumes a Brownian model of evolution. Ancestral reconstructions of the presence/absence of hummingbird and bees visitors were conducted separately using maximum parsimony in Mesquite (Maddison and Maddison, 2004
). Moths were not considered for ancestral reconstructions because of their unknown effectiveness and low visitation rates. The influence of phylogeny on the evolution of continuous floral traits was evaluated by fitting Pagel's (2002)
phylogeny-scaling parameter included in Continuous (Pagel, 2000
). When = 1.0, trait evolution is influenced strongly by phylogeny; when = 0.0, trait evolution is independent of phylogeny. Lambda parameters were fitted by maximum likelihood. Nested likelihood ratio tests were undertaken to determine whether the observed values differed significantly from 0.
RESULTS
Phylogenetic reconstruction
The alignment of the chloroplast and nuclear regions for the 12 species considered a total of 2223 positions. ITS varied the most with 53 of 644 variable sites (8.2%), of which 36 were parsimony-informative. This was followed by waxy with 64 of 856 variable sites (7.5%), of which 39 were informative. The chloroplast region was the most conserved with 34 of 723 variable sites (4.7%), with 14 sites informative. Figure 2A shows the majority rule consensus tree recovered from the Bayesian analysis of the combined nuclear and chloroplast data set. Three major clades within the genus Schizanthus were recovered. Clade A consisted of S. alpestris positioned sister to a second unresolved clade containing S. candidus, S. integrifolius, and S. lacteus. Clade B contained S. hookeri and S. grahamii. Clade C placed S. laetus with two sister subclades, one containing S. litoralis and S. porrigens, and the other S. tricolor, S. pinnatus, and S. parvulus. Clades A and B were strongly supported by the three DNA regions when analyzed separately, but clade C was only supported by the waxy data (Fig. 2B–D). Other differences among the data sets relate to the root position and the clustering of S. litoralis with S. porrigens. This subclade was strongly supported by the ITS data, but not by the more informative waxy data set. Despite these differences, the ILD test did not provide evidence of conflicts between the three data sets.
|
|
|
) did not differ from 0.0 for corolla size ( = 0, P = 1.0), relative lower lip size ( = 0.36, P = 0.07), corolla dissection ( = 0.47, P = 0.11), and relative corolla tube length ( = 0.76, P = 0.17), indicating that the different components of the flower have evolved independently of any phylogenetic effects in this genus.
|
|
Molecular phylogeny
The combined results for ITS, waxy, and trnF/ndhJ produced a well-resolved phylogeny. Our molecular phylogenetic hypothesis was partially congruent with the morphological classification of Grau and Grönbach (1984)
, who recognized four groups: (1) S. candidus, S. integrifolius, and S. lacteus; (2) S. hookeri and S. grahamii; (3) S. tricolor and S. pinnatus; (4) S. litoralis and S. porrigens. They suggested that S. laetus is most closely related to the fourth group and that S. alpestris and S. parvulus are isolated taxa with no clear affinities. Our molecular data support the recognition of the four groups as well as the suggested relationship of S. laetus. Schizanthus alpestris is also included in the first group, and S. parvulus forms part of the clade comprising the third group. Interestingly, multiple waxy clones were obtained from a single individual of S. alpestris. Within-individual variation of waxy has been observed in other Solanaceae (Peralta and Spooner, 2001
). This variation could be attributable to diverse causes, for example, allelic variation, presence of more than one waxy locus (Evans et al., 2000
), or polyploidy. Irrespective of the causes involved, the two waxy clones of S. alpestris belong to the same clade of the phylogeny.
Molecular and morphological data indicate that Schizanthus diverged early from the rest of the Solanaceae (Grau and Grönbach, 1984
; Olmstead and Palmer, 1992
; Martins and Barkman, 2005
), probably in the late Cretaceous to early Tertiary, as suggested by data on the origin and diversification of the Solanales (Magallón et al., 1999
) and the earliest fossil record for the Solanaceae (Eocene; Collinson et al., 1993
). At this time, tropical and subtropical conditions dominated the southern South American region (Hinojosa and Villagrán, 1997
). Despite the ancient origin of Schizanthus, the molecular data suggest that the three major clades of the genus diverged recently. Considering the evolutionary rates for ITS in other herbaceous plants (Richardson et al., 2001
) and the molecular clock, the age of the root would lie at about 5 million years (my). The major clades of Schizanthus thus diverged when the current desert and semidesert regions were already arid and the Andean mountains had reached considerable height (Hinojosa and Villagrán, 1997
).
Evolution of floral traits in association with flower visitors
Ancestral reconstruction of floral visitors suggests that the original flower of Schizanthus was bee-pollinated (Fig. 4). This is consistent with the floral trait reconstruction that gave ancestral flowers similar to those of extant bee-pollinated S. alpestris (Fig. 3, Table 3). Bee pollination (Fig. 4) and bee-syndrome traits (Fig. 3) are still conserved in clade C.
The only species showing a hummingbird-pollination syndrome was S. grahamii (Table 2). Its sister species, S. hookeri, lacks the suite of characters related to the hummingbird-syndrome, but has high nectar production (Table 2) and attracts some hummingbird attention (Table 3). The common ancestor of the two species presumably acquired hummingbird pollination (Fig. 4), and later S. grahamii evolved away from bees to specialize exclusively on birds (Figs. 3, 4). These two species are alpine species of the central Chilean Andes. Conditions for biotic pollination here tend to deteriorate with elevation (Arroyo et al., 1982
, 1985
). It is known that transitions to hummingbird pollination occur in environments with low insect activity (Cruden, 1972
), thus the appearance of hummingbird pollination is not unexpected at high elevations, as indeed is seen in the northern tropical Andes (Kay et al., 2005
).
Traits associated with moth pollination are found in S. candidus, S. lacteus, and S. integrifolius, which inhabit coastal and inland areas of the Atacama Desert. However, moth visitation was observed only in S. integrifolius and at a very low rate (only two visits in 30 h of observation during six consecutive nights) compared with bee visitation. Other plant species with a moth-pollination syndrome are visited by both diurnal and nocturnal visitors (e.g., Young 2002
). In some cases, the effectiveness of nocturnal pollinators is greater even when diurnal visits are more abundant (Groman and Pellmyr, 1999
). We do not have information on the effectiveness of bees and moths in S. integrifolius and thus cannot accept or reject the hypothesis that the moth syndrome evolved from the direct influence of the most effective pollinator. That the stigmas are receptive and pollen and nectar are available during the daylight hours suggests that effective diurnal pollination is not impossible (F. Pérez, unpublished data). The complete absence of flower visitors in S. lacteus and S. candidus is intriguing. In a pollinator exclusion experiment we found that 90–100% of the flowers in these species form fruits, indicating a high autonomous selfing capacity (F. Pérez, unpublished data). Darwin (1862)
was perplexed by the presence of adaptations for selfing in plants with specialized floral morphology. Zhang et al. (2005)
likewise recognized this paradox, suggesting that the retention of specialized pollination syndromes in plants with high selfing levels could represent an anachronism. Although we cannot rule out insufficient sampling, the persistence of the moth syndrome in S. lacteus and S. candidus could well be anachronic. Paleoecological studies in the Atacama Desert show that aridity has increased significantly over the last 3000 years (Latorre et al., 2002
, 2003
), which suggests that current conditions for moth pollination are perhaps less amenable than they were in the past, leading to a mismatch between floral morphology and current visitors. Chilean ecosystems are depauperate in terms of hawkmoth species richness (Ureta and Donoso, 1956
). Whether this pattern is replicated in other genera inhabiting the Atacama Desert is unknown at present and needs to be assessed in future studies. On other hand, the conservative nature of the moth syndrome seems counterintuitive, given that floral evolution in Schizanthus is not strongly influenced by phylogeny. Possibly El-Niño-related climatic fluctuations impinging on pollinator abundance in species with high specialized morphology have promoted the acquisition of a backup breeding system that allows seed set in years when pollinators are infrequent.
Transitions between pollination syndromes
Intermediate stages have been suggested to occur in the transition from one pollination system to another (Baker, 1963
; Stebbins, 1974
; Armbruster, 1993
; Wilson et al., 2006
). The combination of bee-syndrome traits and high nectar production in S. hookeri (reflecting a mixed syndrome with bees, long-tongued flies, and hummingbirds as pollinators) may represent a generalized intermediate stage in the bee-to-bird transition. This pattern agrees with Wilson et al. (2006)
who suggested that "despecialized" stages might occur in the transition from bee- to hummingbird-pollination syndrome in Penstemon. Here bee-pollinated species experienced a "despecialization" transition through the acquisition of traits attractive to hummingbirds (such as high nectar production), but at the same time maintained traits that allowed bee pollination (pink color, landing platform). Subsequent "respecialization" occurred when pollination by hymenoptera was discouraged due to the appearance of traits that favor hummingbird pollination (e.g., reduction of the lower lip and acquisition of red). However, the situation in S. hookeri does not necessarily parallel that in Penstemon. This species of Schizanthus may represent a legitimate generalist species, rather than a transitional species. Generalized pollination systems would be adaptive at high elevations in the Andes where conditions for biotic pollination deteriorate with elevation. Reduction in pollinator resources driven by specialization in the stringent high elevation conditions may have promoted the acquisition of a backup breeding system as revealed by the observation that Schizanthus grahamii, a species that "respecialized" on hummingbirds, unlike S. hookeri, is capable of fruit formation in the absence of pollinators (F. Pérez, unpublished data).
Lability of flower trait
Our results indicate that estimates did not differ from 0 for all continuous floral traits, indicating that floral evolution was not strongly influenced by phylogeny. Repeated and independent changes of floral traits during phylogeny were observed (Fig. 3). For example, reduction of the lower lip and loss of explosive pollen discharge occurred independently in the hummingbird-pollinated S. grahamii and in the ancestor of the clade showing moth-syndrome traits (Fig. 3C, G). These results are consistent with other studies documenting that floral characters tend to be labile and often show independent evolution and reversals along phylogenies (Armbruster, 1993
; Prather, 1999
; Kimball and Crawford, 2004
).
Conclusions
In Schizanthus, a number of floral traits have evolved in a concerted way with changes in floral visitors. While the maintenance of a suite of characters related to the bee syndrome in bee-pollinated species and the acquisition of characters related to the hummingbird syndrome in hummingbird-pollinated S. grahamii provide strong evidence for concerted evolution, the presence of traits related to a moth syndrome in the absence of moth pollinators fails to comply with this pattern. This mismatch suggests either anachronic floral characters possibly related to recent climate changes reducing the abundance of specialized pollinator taxa or maintenance of rare moth pollination backed up by capacity for autonomous selfing. The lack of close correspondence between floral morphology and pollinators in the moth-pollination syndrome reveals the importance of considering factors such as effectiveness of floral visitors and breeding system for a more complete understanding of patterns of floral evolution in Schizanthus. Overall, high floral diversification in the genus Schizanthus is a product of concerted changes associated with adaptation to different groups of pollinators in the mediterranean, high alpine, and desert ecosystems of Chile and adjacent Argentina.
|
1 The authors thank J. Armesto, D. Navea, L. San Martín, A. M. Humaña, C. G. Ossa, D. Rougier, A. Pérez, and C. Zavaleta for their help in different stages of this work and CONAF for permission to work in the Chilean National System of Protected Areas. This research was funded by grant FONDECYT 2010023 to F. P. and doctoral and postdoctoral fellowships from the Center for Advanced Studies in Ecology and Research in Biodiversity (PO2-051-F ICM) to F. P., M. T. K., and R. M. acknowledge support of the BBVA Prize in Conservation Biology Research. Laboratory equipment for phylogenetic analyses was provided by a Mellon Foundation grant. 
4 Author for correspondence (fefapt{at}yahoo.com
)
LITERATURE CITED
Armbruster W. S.. 1993. Evolution of plant pollination systems: hypotheses and tests with the neotropical vine Dalechampia. Evolution 47: 1480-1505.[CrossRef][ISI]
Armbruster W. S.. 2002. Can indirect selection and genetic context contribute to trait diversification? A transition-probability study of blossom-colour evolution in two genera. Journal of Evolutionary Biology 15: 468-486.[CrossRef][ISI]
Arroyo M. T. K Armesto J. J. Primack R.. 1985. Community studies in pollination ecology in the high temperate Andes of central Chile. II. Effect of temperature on visitation rates and pollination possibilities. Plant Systematics and Evolution 149: 187-203.[CrossRef][ISI]
Arroyo M. T. K. Primack R. Armesto J. J.. 1982. Community studies in pollination ecology in the high temperate Andes of central Chile. I. Pollination mechanism and altitudinal variation. American Journal of Botany 69: 82-97.[CrossRef][ISI]
Baker H. G.. 1963. Evolutionary mechanisms in pollination biology. Science 139: 877-883.
Cocucci A. A.. 1989. El mecanismo floral de Schizanthus. Kurtziana 20: 113-132.
Collinson M. E. Boulter M. C. Holmes P. L.. 1993. Magnoliophyta ("Angiospermae"). In M. J. Benton [ed.], The fossil record, vol. 2 809-841 Chapman and Hall, London, UK.
Cruden R. W.. 1972. Pollinators in high-elevation ecosystems: relative effectiveness of birds and bees. Science 176: 1439-1440.
Darwin C.. 1862. On the various contrivances by which British and foreign orchids are fertilized by insects Murray, London, UK.
Evans R. C. Alice L. A. Campbell C. S. Kellogg E. A. Dickinson T. A.. 2000. The granule-bound starch synthase (GBSSI) gene in Rosaceae: multiple putative loci and phylogenetic utility. Molecular Phylogenetics and Evolution 17: 388-400.[CrossRef][ISI][Medline]
Fægri K. van der Pijl L.. 1979. The principles of pollination ecology Pergamon, Oxford, UK.
Farris J. S. Källersjö M. Kluge A. G. Bult C.. 1985. Testing significance of incongruence. Cladistics 10: 315-319.[CrossRef]
Felsenstein J.. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783-791.[CrossRef][ISI]
Felsenstein J.. 1988. Phylogenies from molecular sequences: inference and reliability. Annual Review of Genetics 22: 521-565.[CrossRef][ISI][Medline]
Fenster C. B. Armbruster W. S. Wilson P. Dudash R. Thomson J. D.. 2004. Pollination syndromes and floral specialization. Annual Review of Ecology, Evolution, and Systematics 35: 375-403.[CrossRef][ISI]
Goldblatt P. Manning J. C. Bernhardt P.. 2001. Radiation of pollination systems in Gladiolus (Iridaceae: Crocoideae) in southern Africa. Annals of the Missouri Botanical Garden 88: 713-714.[CrossRef][ISI]
Grau J. Grönbach E.. 1984. Untersuchungen zur variabilität in der gattung Schizanthus (Solanaceae). Mitteilungen der Botanischen Staatssammlung München 20: 111-203.
Graham S. W. Olmstead R. G. Barrett S. C. H.. 2002. Rooting phylogenetic trees with distant outgroups: a case study from the commelinoid monocots. Molecular Biology and Evolution 19: 1769-1781.
Grant V. Grant K. A.. 1965. Flower pollination in the Phlox family Columbia University Press, New York, New York, USA.
Groman J. D. Pellmyr O.. 1999. The pollination biology of Manfreda virginica (Agavaceae): relative contribution of diurnal and nocturnal visitors. Oikos 87: 373-381.[CrossRef][ISI]
Hall T. A.. 1999. BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series 41: 95-98.
Herrera C. M. Cerda X. Garcia M. B. Guitian J. Medrano M. Rey P. J. Sanchez-Lafuente A. M.. 2002. Floral integration, phenotypic covariance structure and pollinator variation in bumblebee-pollinated Helleborus foetidus. Journal of Evolutionary Biology 15: 108-121.[CrossRef][ISI]
Hershkovitz M. A. Zimmer E. A.. 1997. On the evolutionary origins of the cacti. Taxon 46: 217-232.[CrossRef][ISI]
Hinojosa L. F. Villagrán C.. 1997. Historia de los bosques del sur de Sudamérica. I. Antecedentes paleobotánicos, geológicos y climáticos del Terciario del cono sur de América. Revista Chilena de Historia Natural 70: 225-239.[ISI]
Huelsenbeck M. A. Ronquist F.. 2001. MrBayes: Bayesian inference of phylogeny. Bioinformatics 17: 754-755.
Johnson S. D. Linder H. P. Steiner K. E.. 1998. Phylogeny and radiation of pollination systems in Disa (Orchidaceae). American Journal of Botany 85: 402-411.[Abstract]
Johnson S. D. Steiner K. E.. 2000. Generalization versus specialization in plant pollination systems. Trends in Ecology and Evolution 15: 140-143.
Kay K. M. Reeves P. A. Olmstead R. G. Schemske D. W.. 2005. Rapid speciation and the evolution of hummingbird pollination in neotropical Costus subgenus Costus (Costaceae): evidence from nrDNA ITS and ETS sequences. American Journal of Botany 92: 1899-1910.
Kimball R. T. Crawford D. J.. 2004. Phylogeny of Coreopsideae (Asteraceae) using ITS sequences suggests lability in reproductive characters. Molecular Phylogeny and Evolution 33: 127-139.
Knapp S.. 2002. Floral diversity and evolution in the Solanaceae. In Q. C. B. Cronk, R. M. Bateman, J. A. Hawkins [eds.], Developmental genetics and plant evolution 267-297 Taylor and Francis, London, UK.
Latorre C. Betancourt J. Rylander K. A. Quade J.. 2002. Vegetation invasions into absolute desert: a 45 000 yr rodent midden record from the Calama-Salar de Atacama basins, northern Chile (lat 22°–24°S). Geological Society of America Bulletin 114: 349-366.
Latorre C. Betancourt J. Rylander K. A. Quade J. Matthei O.. 2003. A vegetation history from the arid prepuna of northern Chile (22–23°S) over the last 13 500 years. Palaeogeography, Palaeoclimatology, Palaeoecology 194: 223-246.[CrossRef]
Maddison W. P. Maddison D. R.. 2004. Mesquite: a modular system for evolutionary analysis, version 1.01. Website http://mesquiteproject.org [Accessed April 2004].
Magallón S. Crane P. R. Herendeen P. S.. 1999. Phylogenetic pattern, diversity and diversification of eudicots. Annals of the Missouri Botanical Garden 86: 297-372.[CrossRef][ISI]
Martins R. T. Barkman T. J.. 2005. Reconstruction of Solanaceae phylogeny using the nuclear gene SAMT. Systematic Botany 30: 433-447.
Olmstead R. G. Palmer J. D.. 1992. A chloroplast DNA phylogeny of the Solanaceae. Subfamilial relationships and character evolution. Annals of the Missouri Botanical Garden 79: 346-360.[CrossRef][ISI]
Pagel M.. 2000. Continuous Computer program distributed by the author, website http://sapc34.rdg.ac.uk/meade/Mark [Accessed May 2005].
Pagel M.. 2002. Modelling the evolution of continuously varying characters on phylogenetic trees. In N. MacLeod, P. L. Forey [eds.], Morphology, shape and phylogeny 269-286 Taylor and Francis, London, UK.
Pellmyr O.. 2002. Pollination by animals. In C. M. Herrera, O. Pellmyr [eds.], Plant–animal interactions, an evolutionary approach 157-184 Blackwell, Oxford, UK.
Peralta I. E. Spooner D. M.. 2001. Granule-bound starch synthase (GBSSI) gene phylogeny of wild tomatoes (Solanum L. section Lycopersicon [Mill.] Wettst. subsection Lycopersicon). American Journal of Botany 88: 1888-1902.
Prather L. A.. 1999. The relative lability of floral vs. non-floral characters and a morphological phylogenetic analysis of Cobaea (Polemoniaceae). Botanical Journal of the Linnean Society 134: 433-450.
Richardson J. E. Pennington R. T. Pennington T. D. Hollingsworth P. M.. 2001. Rapid diversification of a species-rich genus of neotropical rain forest trees. Science 293: 2242-2245.
Schluter D. Price T. D. Mooers Ø. A. Ludwig D. D.. 1997. Likelihood of ancestor states in adaptive radiation. Evolution 51: 1699-1711.[CrossRef][ISI]
SPSS.. 1998. Sigma Scan Pro 5.0 SPSS Science, Chicago, Illinois, USA.
Stebbins G. L.. 1974. Flowering plants: evolution above the species level Harvard University Press, Cambridge, Massachusetts, USA.
Swofford D. L.. 1999. PAUP* 4.0: phylogenetic analysis using parsimony (*and other methods) Sinauer, Sunderland, Massachusetts, USA.
Swofford D. L. Olsen G. J. Waddell P. J. Hillis D. M.. 1996. Phylogenetic inference. In D. M. Hillis, C. Moritz, B. K. Mable [eds.], Molecular systematics 407-514 Sinauer, Sunderland, Massachusetts, USA.
Tétényi P.. 1987. A chemotaxonomic classification of the Solanaceae. Annals of the Missouri Botanical Garden 74: 600-608.[CrossRef][ISI]
Ureta E. Donoso R.. 1956. Revisión de la familia Sphingidae (Lep. Het.), en Chile. Boletín del Museo Nacional de Historia Natural 26: 237-256.
van der Leij F. R. Visser R. G. F. Ponstein A. S.. Jacobsen E. Feenstra W. J.. 1991. Sequence of the structural gene for granule-bound starch synthase of potato (Solanum tuberosum) and evidence for a single point deletion in the amf allele. Molecular Gene Genetics 228: 240-248.
Walters R. D.. 1969. A revision of the genus Schizanthus Ph.D. thesis, Indiana University, Terre Haute, Indiana, USA.
Waser N. M. Chittka L. Price M. Williams N. M. Ollerton J.. 1996. Generalization in pollination systems, and why it matters. Ecology 77: 1043-1060.[CrossRef]
White T. J. Lee S. Taylor J.. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In M. A. Innis, D. H. Gelfand, J. J. Sninsky, T. J. White [eds.], PCR protocols: a guide to methods and applications 315-322 Academic Press, San Diego, California, USA.
Wilson P. Castellanos M. C. Wolfe A. D. Thomson J. D.. 2006. Shifts between bee and bird pollination in Penstemon. In N. M. Waser, J. Ollerton [eds.], Plant–pollinator interactions: from specialization to generalization University of Chicago Press, Chicago, Illinois, USA.
Young H. J.. 2002. Diurnal and nocturnal pollination of Silene alba (Caryophyllaceae). American Journal of Botany 89: 433-440.
Zhang L. Barrett S. C. H. Gao J. Chen J. Cole W. W. Liu Y. Bai Z. Li Q.. 2005. Predicting mating patterns from pollination syndromes: the case of "sapromyophily" in Tacca chantrieri (Taccaceae). American Journal of Botany 92: 517-552.
This article has been cited by other articles:
|
|
 |
|
  Flowering Newsletter bibliography for 2006 J. Exp. Bot., April 20, 2007; (2007) erm028v2. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
