AOBPreview originally published online on June 28, 2008
Annals of Botany 2008 102(3):305-316; doi:10.1093/aob/mcn106
Flower Morphology, Pollination Biology and Mating System of the Complex Flower of Vigna caracalla (Fabaceae: Papilionoideae)
Cátedra de Botánica, Laboratorio de Biología Reproductiva, Facultad de Ciencias Naturales, Universidad Nacional de Salta, Calle Buenos Aires 177, 4400 Salta, Argentina
* For correspondence. E-mail angelaetcheverry{at}salnet.com.ar
Received: 9 January 2008 Returned for revision: 5 March 2008 Accepted: 6 June 2008 Published electronically: 28 June 2008
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Background and Aims Vigna caracalla: has the most complex flower among asymmetrical Papilionoideae. The objective of this study was to understand the relationships among floral characteristics, specialization, mating system and the role of floral visitors under different ecological contexts.
Methods: Five populations were studied in north-western Argentina, from 700 to 1570 m a.s.l. Anthesis, colour and odour patterns, stigmatic receptivity, visitors and pollination mechanism were examined and mating-system experiments were performed.
Key Results: The petals are highly modified and the keel shows 3·75–5·25 revolutions. The sense of asymmetry was always left-handed. Hand-crosses showed that V. caracalla is self-compatible, but depends on pollinators to set seeds. Hand-crossed fruits were more successful than hand-selfed ones, with the exception of the site at the highest elevation. Bombus morio (queens and workers), Centris bicolor, Eufriesea mariana and Xylocopa eximia trigger the pollination mechanism (a ‘brush type’). The greatest level of self-compatibility and autonomous self-pollination were found at the highest elevation, together with the lowest reproductive success and number of pollinators (B. morio workers only).
Conclusions: Self-fertilization may have evolved in the peripheral population at the highest site of V. caracalla because of the benefits of reproductive assurance under reduced pollinator diversity.
Key words: Fabaceae, Vigna caracalla, asymmetry, breeding system, complex flowers, elevational gradient, pollination
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Plants in the Papilionoideae (Fabaceae) present complex flowers. The petals are modified into a vexillum, wings and keel, with the reproductive structures being enclosed by the keel. The typical papilionoid flower has a zygomorphic symmetry, but there is a group of non-related taxa, among them Phaseoleae (in part), Robinieae and Fabeae (Yeo, 1993), whose flowers diverge from the ‘flag blossom’ pattern and are asymmetric. These flowers are strongly transformed and complex. For example, some species within the tribe Phaseoleae have highly modified petals with co-adaptive assemblages between the wings and keel, and a narrow pipe-like rostrum extending from the keel that causes it to coil (Einsiedel, 1976). Because of the elaborate architecture of Papilionoideae flowers, the application and reception of pollen are achieved only through very specialized mechanisms (Delpino, 1868/9, reviewed in Westerkamp, 1997). The most common mechanism of pollination within the family is known as a ‘brush type’ (Lavin and Delgado, 1990; Yeo, 1993), in which the stigma and the pollen-laden stylar brush emerge from the keel in response to the insect's flower handling.
Whereas the reproductive biology, floral morphology and pollination biology of Papilionoideae plants with zygomorphic flowers are well studied, little is known about these aspects in plants with asymmetric flowers. It has been proposed that asymmetric flowers are effective and precise in pollen placement and retrieval on the body of pollinators (Brizuela et al., 1993; Westerkamp, 1993; Etcheverry et al., 1998, 1999, 2001; Hoc and Amela García, 1998, 1999). Westerkamp (1997) suggested that the very specific sites for pollen deposition would prevent or minimize interspecific pollination and pollen loss by insect grooming. With respect to the breeding system of asymmetrical Papilionoideae, it has been reported that they are predominantly autogamous or even cleistogamous, probably as a consequence of their complicated morphology, for example Macroptilium panduratum (Etcheverry et al., 2001); Vigna minima (Gopinathan and Babu, 1987) and Macroptilium fraternum (Drewes and Hoc, 2000). There are other species of asymmetrical Papilionoideae that are xenogamous or facultatively xenogamous, such as Phaseolus coccineus (Búrquez and Sarukhán, 1984), Macroptilium bracteatum (Brizuela et al., 1993), Macroptilium lathyroides (Etcheverry et al., 1998), Macroptilium erythroloma (Etcheverry et al., 1999) and Phaseolus vulgaris var. aborigeneus (Hoc and Amela García, 1999), and even obligately xenogamous, such as Phaseolus augustii (Hoc and Amela García, 1998). However, there is a lack of information about the reproductive biology and floral visitors of most Papilionoideae species with asymmetrical flowers. Within this group, Vigna caracalla (subgen. Cochliasanthus) is considered to be the species with the most complex flowers (Lindman, 1902; Troll, 1951). It has been speculated that the extreme complexity of V. caracalla may lead to a total loss of functionality of the flower, requiring that it be completely autogamous (Faegri and van der Pijl, 1979). Empirical data are needed to determine (1) if V. caracalla still outcrosses despite the complexity of its flowers, (2) if pollinators are available in natural populations, and (3) how pollinators manipulate the flowers in order to access floral rewards.
Moreover, multipopulational studies on the reproductive biology of V. caracalla are required in order to understand the variability and relationships among floral morphology, mating system and floral visitors under different ecological contexts. Based on the geographic mosaic theory (Thompson, 1994, 1997), plant–pollinator interactions may be expected to vary among populations, generating a complex pattern of differential adaptation. For example, Fausto et al. (2001) found variation in plant and flower densities, pollinator densities and pollinator visitation rates among populations of Clarkia xantiana (Onagraceae). Herrera et al. (2001) studied the geographic variation in autonomous self-pollination levels in Helleborus foetidus (Ranunculaceae), showing that it was lowest in regions with the lowest pollinator service. Silva-Montellano and Eguiarte (2003) reported a latitudinal pattern in floral traits, abundance of pollinators and fecundity of Agave lechuguilla (Agavaceae). In Papilionoideae, Búrquez and Sarukhán (1980) showed that the highest-elevation population of Phaseolus coccineus was exclusively hummingbird-pollinated, while in the lower-elevation populations the pollinators were hummigbirds and bumblebees. They also reported differences in floral colours, floral phenology and nectar offer. Given its broad distribution, V. caracalla is an excellent system to study plant–pollinator interactions in a geographic context.
In this study, an analysis and comparison was made of (1) the morphological traits, (2) the pollination mechanism, (3) pollinator assemblage, and (4) the mating system of Vigna caracalla through an altitudinal gradient in north-western Argentina. It was expected that the highest (peripheral) population would exhibit a tendency towards autogamy because pollinators and/or mates are scarce (Jain, 1976; Wyatt, 1986; Busch, 2005).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Study species
Vigna caracalla (L.) Verdc. is a tuberous, winding, perennial plant native to South America (Brazil, Bolivia, Paraguay, Argentina, Peru, Ecuador and Colombia) and Central America (Guatemala, Nicaragua, Costa Rica, Mexico and Panama; Maréchal et al., 1978). The plant is widely cultivated as an ornamental species (Piper, 1926; Burkart, 1943) and may be used for forage (Fernández et al., 1988).
Vigna caracalla flowers are the largest among the American Phaseolinae (Lackey, 1983) and are arranged in axillary pendent inflorescences with acropetal maturation. They are compound racemes of 4–21 modified simple racemes (‘pseudoracemes’, Tucker, 1987). Each simple raceme has a contracted rachis with two (rarely three) flowers. The other floral buds degenerate in early stages of development and transform into an extra-floral nectary with a swollen aspect (Etcheverry, 2006).
Study sites
Five populations of V. caracalla were studied in the provinces of Salta and Jujuy, Argentina. The populations are distributed over an elevation gradient of 870 m (700–1570 m a.s.l.; Table 1). Four of the population sites (Campo Quijano, La Caldera, Vaqueros and Calilegua) are located in an Eastern Andes seasonal rainforest (Yungas), whereas the fifth site (Betania) is located in a dry forest (Chaco) that includes small trees, thorn shrubs, cacti, herbs, epiphytes and vines (Cabrera, 1971). Voucher specimens are deposited in the Museo de la Facultad de Ciencias Naturales de la Universidad Nacional de Salta (MCNS). Campo Quijano is the highest and most western population, and is peripheral for the distribution of the species in the region studied (L. Novara, University of Salta, Argentina, pers. comm.).
|
Fieldwork was carried out from January through to the end of March (rainy season) for five consecutive years (2001 to 2005).
Floral traits
Floral growth of ten selected individuals was monitored in the Vaqueros population during January 2000. Twenty flower buds were marked on each plant and observed twice a day until they opened.
Floral morphology was studied with fresh flowers using a dissecting microscope and an environmental scanning electronic microscope (Philips XL30 ESEM). Twenty-five randomly selected flowers were analysed from eight-to-ten different plants in each population. For these flowers, the number of revolutions of the keel from the starting point of turning to its tip were counted and measurements were made of the length of the pedicel, the corolla and the nectary depth. The fresh colour of petals was recorded using a Munsell chart (Munsell Color, 1977).
For histological studies, flowers from the Vaqueros population were fixed in FAA, dehydrated through an ethyl alcohol/xylol series, and embedded in paraffin. Cross- and longitudinal sections were stained with safranin and fast-green.
For purposes of this work, the left and right sides of the flower were described from the point of view of the floral visitors. Drawings of fresh flowers were made using a camera lucida.
Stigmatic receptivity throughout anthesis was determined by using hydrogen peroxide (Kearns and Inouye, 1993).
The location of scent emission was determined by staining intact flowers with an aqueous solution of neutral red:water (1 : 1000) for 5 h (Kearns and Inouye, 1993). In addition, floral parts were separated and put into small glass containers with snap fasteners, which were presented to ten volunteers to detect their odour.
Visitors
Throughout the 5-year study, flower visitors were collected and identified in all populations. Vouchers are deposited at the Museo de Ciencias Naturales Bernardino Rivadavia, Buenos Aires, Argentina (MACN). Measurements of body and proboscis length were taken using a digital caliper. The foraging behaviour of visitors was analysed by photography and video recording. For each population an estimate was made of the number of visits per flower per observation period (pollinator visitation rate) during one year, as follows: La Caldera, 2001; Vaqueros, 2002; Campo Quijano, 2003; Betania, 2004; and Calilegua, 2005. Flowers were observed for periods of 15 min, from 0700 h to 1900 h, totalling ten to 12 observation-hours per site (overall total = 54 h). All observations were made from a fixed location from which 23–44 flowers could be monitored. The duration of each flower visit was recorded at the Vaqueros site.
An index of proportional similarity (PSI; Feinsinger, 2004) was calculated in order to compare the proportion of pollinators among all populations. The index takes values between 1 and 0; when PSI = 1 a pair of populations is similar in composition and proportion of pollinator species. The PSI values obtained were correlated with geographic distances between populations.
Pollen carried externally by floral visitors was identified using the techniques of Buchmann and Shipman (1990) and Genise et al. (1990). In species where pollen of V. caracalla was present, a quantification was made using a hemocytometer (Kearns and Inouye, 1993). For each pollinator species, eight-to-ten bees were examined. Pollinators were distinguished from nectar- or pollen robbers by their behaviour, and by the presence of V. caracalla pollen on body sites where the stigma makes contact (see Dafni, 1992).
Pollinator efficiency
In order to assess pollinator efficiency of the main pollinators, the procedure described by Herrera (1987) was followed. For the Vaqueros population, 20 inflorescences were bagged with nylon mesh. These exclosures precluded access of all floral visitors, as demonstrated by frequent checks. Every few days from mid-January to February 2005, one of the bags was removed and the enclosed inflorescence was watched at close range. Bombus morio workers or Xylocopa eximia coming to flowers were allowed to visit 1–2 individual flowers, and were then chased away. The visited flower was immediately removed and placed in a glass vial; separate vials were used for each pollinator. The collected flowers were dissected individually, the style was removed and the number of pollen grains on stigmas was counted under a microscope.
Mating system
Assessment of the mating system involved ten randomly selected plants in each population (see Table 4 for the number of flowers used in each treatment). The treatments were as follows: (1) free pollination (control) – flowers were exposed to the natural agents of pollination; (2) autonomous self-pollination – buds were bagged throughout their flowering period; (3) hand self-pollination – bagged buds were self-pollinated by hand immediately after anthesis; (4) apomixis – anthers of bagged buds were clipped; and (5) outcrossing by hand – bagged flowers were emasculated the day before anthesis and pollinated with pollen of recently opened flowers of other individuals (50–100 m away). Prior to pollen deposition the stigmatic surface was rubbed with a brush in order to release stigmatic fluids. The number of fruits was recorded after 1 week and fruit set was estimated as the proportion of flowers setting fruits. For each population, a self-compatibility index (SCI) was estimated as the ratio between fruit set obtained by hand self-pollination and that obtained by cross-pollination by hand. When SCI was <0·2, the population was considered to be self-incompatible (Bawa, 1974; Arroyo and Uslar, 1993).
Fruit and seed production, and relative reproductive success
Fruit set and seed set were estimated on entire inflorescences under natural pollination that were not used for any of the pollination treatments above. Fruit set is here defined as the proportion of flowers that developed into fruits in each inflorescence. Seed set is defined as the proportion of ovules that developed into seeds in all the mature pods within an inflorescence; this assessment was possible because of the linear arrangement of the ovules in the ovary. The relative reproductive success (RRS) of an inflorescence was calculated by multiplying fruit set by seed set (Kasagi and Kudo, 2003). For each population, 35 randomly selected inflorescences were analysed from different plants.
Data analysis
The results were analysed using SYSTAT (1992) for Windows. Chi-square tests were performed in order to compare the frequencies of developed and aborted fruits between mating-system treatments. Flowers, fruits and seed traits were compared by one-way analysis of variance (ANOVA). A non-parametric method (Kruskal–Wallis) was used when the assumptions required for parametric methods were not met (Sokal and Rohlf, 1995).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Floral morphology
The flower of Vigna caracalla (Fig. 1) is markedly asymmetric except for the calyx, which is zygomorphic. The total length of flowers and pedicels varied between 4–7 cm and 5–15 mm, respectively.
|
Calyx
The calyx has a tubular form and tightly encloses the petal bases, avoiding disarticulation during insect visits.
Corolla
The corolla has five petals that form a functional tube enclosed by the elongated tubular calyx. All populations showed similar petal colours.
Vexillum
The vexillum is pale purple (Munsell chart: 5 RP 7/4) and has a yellow mark band (2·5 Y 8/8) of approximately 9 mm in length at the base of the blade; the yellow markings on flowers at the Vaqueros site have blue-green margins (5 BG 3/4; see Fig. 1). In the same area, a tactile guide is present. This guide consists of two prominent ridges (Fig. 2A, arrows) that firmly clasp the free stamen, closing the nectar entrance. This configuration blocks access to the nectar from the sides (Fig. 2B).
|
Wings
The wings show a curvature that positions them in a transversal orientation (Fig. 1, rw–lw). They are purple (5 RP 6/4) on the outermost sides and white (5 RP 3/8) on the inner ones. The left wing lies at the lower side of the flower where it functions as a landing platform for visitors. On the proximal part of the blades, both wings have transversal puckerings (Fig. 2D). Histologically, the epidermal cells located on the crests have alternate cuticular striations, which are absent in the ‘valleys’ (Fig. 3A, B). This structure provides a foothold for the visitors and also a tactile guide to the nectar.
|
Keel
The keel (Fig. 1K) is formed by two white-coloured petals. The distal portion of the keel petals narrow abruptly, enclosing the style and the distal part of the androecium. This portion is coiled and forms the rostrum of the keel (Fig. 2C, E). The mean number of revolutions differed significantly among populations (H = 22·86, P = 0·001; Fig. 4). The populations at Betania (4·7 revolutions) and Calilegua (4·9 revolutions) showed significantly higher values than the other populations. Quijano had the highest variation in revolutions (3·75–5·25).
|
The keel petals are joined along their lower margins. From above the claws, the upper margins of the keel petals show rows of unicellular trichomes (Fig. 2C), which interdigitate in the distal part (last revolution of the keel). These trichomes close the keel except at the very tip, leaving a small opening (Fig. 2E). In this way, a secluded space is created for the accumulation of pollen, which is exposed only during an insect's visit. During their visits, bees lower the left wing, inducing the emergence of the stigma and the stylar brush charged with pollen (Fig. 2E).
Androecium
The androecium has ten stamens; nine of them are fused proximally and the tenth remains free. The filament of the tenth stamen is basally widened, and has a protuberance that fits into the vexillar channel (Fig. 2B). The base of the filament is narrow, leaving space for two entrances to the nectar chamber (Fig. 2F). Anther dehiscence occurs at or near anthesis and all of the pollen is trapped between the trichomes of the stylar brush.
Gynoecium
The distal segment of the style is coiled with the same number of revolutions as the keel. Totally expanded, the style reaches up to 85 mm in length. There is a ring of non-receptive unicellular hairs surrounding the stigma (Fig. 3C), which probably prevents contact between a flower's own pollen mass and the stigma. SEM micrographs show the stigmatic surface to be covered by a ‘membrane’ (the cuticle of the stigmatic epidermis), which prevents self-pollen germination by holding the secretory products of the stigma (Fig. 3C, arrow). This membrane is disrupted only by tripping, allowing the pollen to contact the stigmatic fluid and germinate. In unvisited flowers the peroxidase test was negative throughout the floral cycle; the stigma reacted only after rubbing the membrane artificially. The pollen is presented to pollinators by the stylar trichomes (9 mm long) located below the stigma, which are exposed during visits (‘brush’ pollination mechanism; Fig. 3D).
Phenology of flowering and visitation
The floral span was 12 h with anthesis beginning just before sunrise (0700 h). All available pollen is deposited in the stylar brush and many visits are needed for the complete removal of pollen, because of the dispensing mechanism. Flower received a mean of 15 visits by Bombus morio workers (the main pollinator of V. caracalla in all populations) during its anthesis. These visits occurred from 0700 h to 18–1900 h. At the end of anthesis the flowers turned dull yellow, no rewards remained and the stigmas were pollinated. In the La Caldera population, the stigmatic pollen load of open-pollinated flowers was 206·7 ± 54·88 (± s.e., n = 25). The visitors systematically rejected these flowers.
A strong, sweetish scent was released during the entire floral life with a peak between 1200 h to 1500 h. Olfactory tests pointed to the vexillum and (with less intensity) the wings as the sites of odour emission; neutral red solution stained only the margin of the vexillum blade (4–5 mm width).
Visitor assemblage
Ten different visitors were recorded (Table 2), among which only Bombus morio (queen and workers; Fig. 5A), Xylocopa eximia (Fig. 5B), Centris bicolor (Fig. 5C) and Eufriesea mariana (Apidae) activated the pollination mechanism and always carried pollen of V. caracalla on the upper-right area of the pronotum. Since these visitors touched the stigma with their body parts where pollen of V. caracalla was deposited, they were considered as pollinators. Xylocopa eximia carried a much higher quantity of pollen grains when compared to B. morio workers (78 125 ± 6467 vs. 11 458 ± 1577, respectively). Differences were also found in pollinator efficiency: the number of pollen grains deposited on stigmas in a single visit was 25·3 ± 2·07 (n = 9) for X. eximia and 11·9 ± 0·78 (n = 11) for B. morio workers.
|
|
Apis mellifera and one species of Meliponini (Fig. 5D) were pollen robbers, and two species of Lepidoptera were nectar robbers. The robbers were small, unable to trigger the pollination mechanism and never touched the stigma. Two species of ants (Formicidae) were present, one species of which was observed in all populations visiting the extrafloral nectary situated in the secondary axis of the inflorescence. These visits were observed from anthesis to fructification. The other ant species was a floral predator present in all populations.
All of the pollinators cited have a large body size (14·8–24·5 mm) and large proboscis (10·9–15·5 mm; Table 3). The biggest pollinators were B. morio queens and Xylocopa eximia, and the smallest was C. bicolor. Among all the observed pollinators, B. morio queens and X. eximia showed the best correspondence with the nectary depth of the flowers of V. caracalla (overall mean 18·06 ± 3·39)
|
All five species of pollinators were present at the Calilegua population. Three species were present at La Caldera and Betania, two in Vaqueros, and only one species was recorded at Campo Quijano. Bombus morio queens were observed only at Calilegua. Xylocopa eximia was present in all populations except Campo Quijano. Bombus morio workers were the only pollinator present in all populations (Table 2).
Pollinator activity
The most frequent pollinators in all populations were B. morio workers (Table 2). The proportions of visits of B. morio workers were: Campo Quijano, 100 %; Calilegua, 90·6 %; La Caldera, 91·6 %; Vaqueros, 95·6 %; and Betania 93·2 %. The next most important pollinator was Xylocopa eximia, with 4·7 to 6·8 %. Bombus morio queens, C. bicolor and E. mariana were present at very low proportions (1·3–2 %).
The values of PSI varied between 0·906 (Campo Quijano–La Caldera, 150 km) and 0·960 (Vaqueros–La Caldera, 12 km), indicating that the pairs of populations analysed were similar in composition and proportion of pollinator species.
Visitation rates for B. morio at each population site were significantly different (H = 13·48, d.f. = 4, P = 0·009) and consistently decreased with decreasing site elevation (Fig. 6).
|
The mean duration of flower visits by X. eximia was approximately twice that of B. morio workers within the Vaqueros population (37·44 ± 7·77 vs. 18·72 ± 1·63; t = –2·35, d.f. = 87, P = 0·044).
Mating system
The results of the mating system experiments are presented in Table 4. In general, the control treatment (free pollination) yielded less fruit than the other treatments. In the apomixis treatment none of the flowers set fruit; therefore, the possibility of asexual reproduction was discarded. In Campo Quijano, some flowers set fruits through autonomous self-pollination, although in a very low proportion (Table 4). In the other populations, plants were not capable of autogamous pollination and were entirely dependent upon pollinators for reproductive success. The lowest and highest reproductive success for both control and hand-cross-pollination treatments were recorded at Campo Quijano and La Caldera, respectively. Cross-pollination by hand set between 2·9 (Campo Quijano) and 5·3 (Vaqueros) times more fruits than their respective controls.
|
Fruit set from cross-pollination was significantly higher than that obtained through self-pollination by hand, except at Campo Quijano where self-pollination set more fruits than outcrossing (Table 5).
|
Self-compatibility indices (SCI) indicated that all of the populations were self-compatible (Table 5); Campo Quijano exhibited the highest SCI.
Relative reproductive success (RRS)
There were significant differences in RRS (H = 22·16; d.f. = 4, P < 0·0001) among the five populations, although the differences were not related to site elevation (Table 6). Relative reproductive success was significantly lower for Campo Quijano in comparison with the other populations, which were statistically similar.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The flower of Vigna caracalla has remarkable and unique morphological features making it arguably the most complex flower in the Papilionoideae. These features are closely matched to those of large-bodied, long-tongued bees, while forming impediments to some small visitors such as Apis mellifera and Meliponini. This finding supports the hypothesis that floral complexity may reflect selection to narrower functional groups of pollinators (Fenster et al., 2004).
The thickness and shape of sepals and petals, and the joint function of keel-wings and the vexillum–androecium, are directly related to the size of the pollinators of V. caracalla. Wing sculpturing, also present in V. vexillata (Hedstrom and Thulin, 1986), provides a resistant surface to visitors and also functions as a nectar-guide because it is orientated towards the nectary, as suggested by Etcheverry (2001a) for Crotalaria spp. (Papilionoideae).
In relation to floral asymmetry, the results support the hypothesis of Endress (1999, 2001) that large bees are the main pollinators of asymmetrical flowers. This conclusion has also been drawn for some asymmetrical flowers of the Tribe Phaseoleae, for example Phaseolus coccineus (Búrquez and Sarukhán, 1980), Macroptilium erythroloma (Etcheverry et al., 1999), V. vexillata (Hedstrom and Thulin, 1986) and V. longifolia (Hoc et al., 1993).
With respect to the revolutions of the keel, Faegri and van der Pijl (1979) speculated that more revolutions and complexity in the flowers of V. caracalla would result in loss of floral functionality and a tendency towards autogamy. However, in all of the populations in the present study the flowers functioned independently of the number of revolutions of the keel, and the site of pollen deposition was always in the upper-right area of the pronotum of the insects. The revolutions of the keel are also observed in the style; as a consequence the style is very long (85 mm), exceeding the maximum values reported by Shivanna and Owens (1989) in their review of Papilionoideae (max. 30–60 mm in Arachis). Mulcahy (1979) noted that the angiosperm style provides a selective arena in which a ‘sieving’ mechanism occurs, that is, deleterious mutations might retard pollen-tube growth and thereby be eliminated from the gene pool. For example, Malti and Shivanna (1985) and Snow and Spira (1991) have presented experimental evidence for such a selective function in Crotalaria retusa and Hibiscus moscheutos, respectively.
The pollination mechanism employed by V. caracalla is complex and precise, and constitutes a significant example of adaptation. Considering the site of pollen placement, V. caracalla is nototribic, and as such constitutes a variation from sternotriby, which is typical of zygomorphic Papilionoideae. This characteristic has been reported for other asymmetric species of Vigna with spiral keels, such as V. vexillata (Hedstrom and Thulin, 1986) and V. longifolia (Hoc et al., 1993). The dominant mode of pollination of Papilionoideae at the Vaqueros site was sternotribic (A. V. Etcheverry, unpubl. res.), for example in Cologania ovalifolia, Crotalaria spp., Desmodium spp., Galactia latisiliqua, Rynchosia edulis and Zornia contorta. These sympatric taxa bloom simultaneously with V. caracalla and several of them share visits by Bombus morio workers. The observed variation in pollen placement is probably related to a mechanism for minimizing the deposition of heterospecific pollen, as Nilsson (1987) and Pauw (2006) suggested for sympatric orchids in Madagascar and in South Africa, respectively.
Considering the relationship between floral morphology and breeding system, Lloyd and Schoen (1992) concluded that families characterized by specialized pollination mechanisms, such as Papilionoideae, have undergone natural selection promoting cross-pollination. Vigna caracalla is self-compatible, but depends on pollinators to set seeds. Two structural features of the gynoecium are related to the mating system: the stigmatic ‘membrane’ and the collar of peristigmatic hairs. The stigmatic membrane prevents autonomous self-pollination, and as a consequence V. caracalla requires ‘tripping’ (sensu Arroyo, 1981) for effective pollination. This membrane was reported to be present in seven tribes of Papilionoideae, among them Phaseoleae (Shivanna and Owens, 1989). Within Vigna, this membrane has been reported in V. adenantha (Castro and Agulló, 1998). In the Campo Quijano population, autonomous self-pollination may occur by spontaneous rupture of the stigmatic membrane. Other Papilionoideae such as Vicia faba (Lord and Heslop-Harrison, 1984) and Medicago scutellata (Kreitner and Sorensen, 1985) have a similar structure that prevents self-fertilization, but autofertile lines have thin cuticles allowing spontaneous disruption and self-fertilization. The occurrence of peristigmatic hairs may maximize outcrossing by preventing self-pollen deposition on the stigmas during the first stages of anthesis. Similar observations have been reported for other Papilionoideae legumes, such as Phaseolus coccineus (Ibrahim and Coyne, 1975), Lupinus nanus subsp. latifolius (Juncosa and Webster, 1989), Crotalaria stipularia (Etcheverry, 2001b) and Crotalaria micans (Etcheverry et al., 2003).
In terms of pollinator activity, Campo Quijano had the highest rate of visits by B. morio workers, but this was not reflected in the production of fruit and seeds. Visitation rate alone is not a good indicator of reproductive success; McDade and Davidar (1984) found a correlation of only 0·065 between visitation rate and seed production for Pavonia dasypetala (Malvaceae), and Schemske and Horvitz (1984) and Horvitz and Schemske (1990) showed that the most abundant pollinator of Calathea ovandensis was not the most efficient. These latter two authors concluded that fruit production was related to efficiency in floral manipulation.
Regarding the efficiency of the most frequent pollinators, Xylocopa eximia carried and deposited more pollen after single visits than did B. morio workers. In addition, the duration of visits of X. eximia was longer, a characteristic that Harder (1990) and Thomson (1986) suggest increases pollen deposition on stigmas. The most effective pollinator is determined by the frequency and ability to affect fruit set (Stebbins, 1970). In all populations the most abundant pollinators were B. morio workers, which are generalists and have the shortest proboscides of the observed pollinators. Thomson (2003) showed that different members of a functional group of pollinators can vary greatly in their effectiveness because of differences in behaviour and morphology. Moreover, some may actually decrease the fitness of a plant by wasting pollen or by decreasing the number of visits by more effective pollinators. We conclude that the activity of X. eximia is important to the reproductive success of V. caracalla and this species is probably the most reliable pollinator.
Low reproductive success seems to be a general phenomenon in V. caracalla, because it was markedly low in the five studied populations. The maximum fruit^:^flower ratio observed in this study (14 %, La Caldera) is well below the mean value for self-compatible hermaphroditic species (72 %; Sutherland and Delph, 1984). Lee (1988) and Guitián et al. (1996) proposed several factors that may limit fruit production, including extrinsic causes (pollen limitation, herbivory, frost, etc) and intrinsic causes (genotypes, stored resource content, etc.). The fact that fruit set is limited by one or more of these factors is relevant to several aspects of reproductive biology, including the evolution of breeding systems and floral displays (Knight et al., 2005, and references therein). The most common factor reported in the literature is resource limitation (e.g. Stephenson, 1981; Berjano et al., 2006). However, the distinction between pollen and resource limitation is not clear because these factors are complex and interrelated (Lee, 1988, and references therein). Several hypotheses have been proposed to assign a role to these ‘extra’ flowers (Sutherland, 1987; Bertin, 1988; Ehrlén, 1991). These include pollinator attraction, bet-hedging against temporal or spatial variation in pollinator abundance, selective abortion, reserves of ovaries and pollen export. Bearing in mind that these hypotheses are not mutually exclusive (Stephenson, 1981; Guitián, 1993) our results are compatible with the reserve-ovary hypothesis (Ehrlén, 1991). This postulates that surplus flowers constitute an ovary reserve in case of flower mortality. Inflorescences of V. caracalla in which a proportion of flowers are lost due to extrinsic factors (e.g. herbivory by ants) are able to maintain fruit-set levels by allowing the development of flowers that would otherwise be aborted. Guitián et al. (1996) arrived at similar conclusions in their study of Cornus sanguinea (Cornaceae).
The low reproductive success of V. caracalla in the studied populations may reflect the lack of the primary specialized pollinator(s), as has been observed in other species with showy, complex flowers, for example the orchid Oeceoclades maculata (González-Díaz and Ackerman, 1988) and the ginger Caulokaempferia coenobialis (Wang et al., 2004).
The greatest level of self-compatibility and autonomous self-pollination, and the lowest reproductive success and number of pollinators (B. morio workers only) was found in the highest-elevation population. The evolution of self-fertilization from outcrossing, one of the most common evolutionary transitions in plants, has occurred in many taxonomic groups (Stebbins, 1970; Wyatt, 1983; Barrett, 2002). Recently, Fenster and Martén-Rodriguez (2007) demonstrated in a survey of 38 plant families, including Papilionoideae, that pollination specialization is often associated with floral traits that facilitate autonomous selfing.
Abiotic and biotic ecological factors can influence mating-system evolution and the degree of self-pollination expressed by individual plants. Pollinator limitation, in numbers or in species, favour shifts from outcrossing to autonomous self-fertilization because selfed-seeds provide reproductive assurance (Baker, 1955; Wyatt, 1986; Inoue et al., 1996; Kalisz and Vogler, 2003; Berjano et al., 2006; but see Herrera et al., 2001). We consider this transition to be occurring in the peripheral population of V. caracalla.
Concluding remarks
Regional variation in population characteristics may set the stage for evolutionary differentiation in reproductive traits across the geographic range of a species. As expected, the peripheral population had the greatest level of self-compatibility and autonomous self-pollination, and the lowest reproductive success and number of pollinators. There is a considerable interest in understanding geographic variation in floral visitors, reproductive traits and their interactions based on the geographic mosaic model proposed by Thompson (1994, 1997); however, to date few empirical studies have attempted to test this model.
|
ACKNOWLEDGMENTS |
|---|
We thank Stefan Vogel, Christian Westerkamp, Ramón Palacios and Patricia Hoc for inspiration; Dulce Figueroa Castro and three anonymous reviewers for comments on the manuscript; Peter Mitchel and Robert Earl Espinoza for helping with our English; and Julio César Quinteros for field assistance.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
-
Arroyo MTK. Breeding systems and pollination biology in Leguminosae. In: Advances in legume systematics—Polhill RM, Raven PH, eds. (1981) London: Royal Botanic Gardens, Kew. 723–769. Part 2.
Arroyo MTK, Uslar P. Breeding systems in temperate mediterranean-type climate montane sclerophyllous forest in central Chile. Botanical Journal of Linnean Society (1993) 111:83–102.[Web of Science]
Baker HG. Self-compatibility and establishment after ‘long-distance’ dispersal. Evolution (1955) 9:347–348.[CrossRef][Web of Science]
Barrett SCH. The evolution of plant sexual diversity. Nature Reviews Genetics (2002) 3:274–284.[CrossRef][Web of Science][Medline]
Bawa KS. Breeding systems of tree species of a lowland tropical community. Evolution (1974) 28:85–92.[CrossRef][Web of Science]
Berjano R, De Vega C, Arista M, Ortiz PL, Talavera S. A multi-year study of factors affecting fruit production in Aristolochia paucinervis (Aristolochiaceae). American Journal of Botany (2006) 93:599–606.
Bertin RI. Paternity in plants. In: Plant reproductive ecology—Lovett Doust J, Lovett Doust L, eds. (1988) New York: Oxford University Press. 30–59.
Bianchi AR, Yáñez CE. Las precipitaciones del Noroeste Argentino (1992) 2da. edn. Salta, Argentina: Salta, Instituto Nacional de Tecnología Agropecuaria (INTA).
Brizuela MM, Hoc PS, Di Stilio VS, Agulló MA, Palacios RA. Biología floral de Macroptilium bracteatum (Leguminosae, Phaseoleae). Darwiniana (1993) 32:41–57.
Buchmann SL, Shipman CW. Pollen harvesting rate for Apis mellifera L. on Gossypium (Malvaceae) flowers. Journal of the Kansas Entomological Society (1990) 63:92–100.[Web of Science]
Burkart A. Las leguminosas argentinas silvestres y cultivadas (1943) 1ra. edn. Buenos Aires: Acme Agency.
Búrquez JA, Sarukhán J. Biología floral de poblaciones silvestres y cultivadas de Phaseolus coccineus L. I. Relaciones planta-polinizador. Boletín de la Sociedad Botanica de México (1980) 39:5–27.
Búrquez JA, Sarukhán J. Biología floral de poblaciones silvestres y cultivadas de Phaseolus coccineus L. II. Sistemas reproductivos. Boletín de la Sociedad Botanica de México (1984) 46:3–12.
Busch JW. The evolution of self-compatibility in geographically peripheral populations of Leavenworthia alabamica (Brassicaceae). American Journal of Botany (2005) 92:1503–1512.
Cabrera AL. Fitogeografía de la República Argentina. Boletín de la Sociedad Argentina de Botánica (1971) XIV:1–2.
Castro MA, Agulló MA. Anatomy of the stigma of Vigna adenantha (G. F. Mayer) Marechal, Mascherpa and Stainer (Leguminosae, Papilionoideae). Biocell (1998) 22:9–18.[Web of Science]
Dafni A. Pollination biology: a practical aproach (1992) Oxford, UK: Oxford University Press.
Delpino F. Ulteriori osservazioni e considerazioni sulla dicogamia nel regno vegetale. 1. Atti della Società Italiana di Scienze Naturali in Milano. 11:265–332. 1868/9.
Drewes SI, Hoc PS. Morfología y desarrollo de flores cleistógamas en Macroptilium fraternum(Fabaceae). Kurtziana (2000) 28:229–238.
Ehrlén J. Why plants produce surplus flowers. A reserve ovary model. American Naturalist (1991) 138:918–933.[CrossRef][Web of Science]
Einsiedel B. Vergleichende Morphologie der asymmetrischen Schmetterlingsblüten. (1976) Berlin: Freie Universität. Diploma thesis.
Endress PK. Symmetry in flowers: diversity and evolution. International Journal of Plant Sciences (1999) 160:S3–S23.[CrossRef][Web of Science][Medline]
Endress PK. Evolution of floral symmetry. Current Opinion in Plant Biology (2001) 4:86–91.[CrossRef][Web of Science][Medline]
Etcheverry AV. Wing morphology in the flower of some American species of Crotalaria. Beiträge zur Biologie der Pflanzen (2001) a 72:181–197.
Etcheverry AV. Role of staminal growth in delayed self-pollination of Crotalaria stipularia (Fabaceae: Papilionoideae). Beiträge zur Biologie der Pflanzen (2001) b 72:215–228.
Etcheverry AV. Biología Reproductiva de Vigna caracalla (Fabaceae: Papilionoideae). (2006) Argentina: University of Buenos Aires. Doctoral Thesis.
Etcheverry AV, Pérez de Bianchi SM, Martín Montiel de López D. Fenología floral de Macroptilium lathyroides (L.) Urban (Fabaceae) y características reproductivas. In: Aportes Botánicos de Salta. Serie Misceláneas. (1998) 1. Argentina: Herbario MCNS, Facultad de Ciencias Naturales, Universidad Nacional de Salta. 7.
Etcheverry AV, Pérez de Bianchi SM, Martín Montiel de López d. Floral biology of Macroptilium erythroloma (L.) Urban (Fabaceae). Beiträge zur Biologie der Pflanzen (1999) 71:1–16.
Etcheverry AV, Pérez de Bianchi SM, Martín Montiel de López D. Reproductive biology in the amphicarpic legume Macroptilium panduratum (Fabaceae, Papilionoideae). Beiträge zur Biologie der Pflanzen (2001) 72:181–197.
Etcheverry AV, Protomastro JJ, Westerkamp C. Delayed autonomous self-pollination in the colonizer Crotalaria micans (Fabaceae: Papilionoideae): structural and functional aspects. Plant Systematics and Evolution (2003) 239:15–28.[CrossRef][Web of Science]
Faegri K, van der Pijl L. The principles of pollination ecology (1979) 3rd edn. Oxford, UK: Pergamon Press.
Fausto JA, Eckhart V, Gebber MA. Reproductive assurance and the evolutionary ecology of self-pollination in Clarkia xantiana (Onagraceae). American Journal of Botany (2001) 88:1794–1800.
Feinsinger P. El Diseño de Estudios de Campo para la Conservación de la Biodiversidad. (2004) Santa Cruz de la Sierra, Bolivia: Ed. FAN.
Fenster CB, Armbruster WS, Wilson P, Dudash MR, Thomson JD. Pollination syndromes and floral specialization. Annual Review of Ecology, Evolution and Systematics. (2004) 35:375–403.[CrossRef][Web of Science]
Fenster CB, Martén Rodriguez S. Reproductive assurance and the evolution of pollination specialization. International Journal of Plant Science (2007) 168:215–228.[CrossRef][Web of Science]
Fernández JG, Benítez CA, Pizzio RM, Royo Pallares O. Leguminosas forrajeras nativas del Este de la Provincia de Corrientes. Serie Técnica (1988) 26. Mercedes, Corrientes, Argentina: EEA.
Genise J, Palacios RA, Hoc PS, Carrizo R, Moffat L, Mom MP, et al. Observaciones sobre la biología floral de Prosopis (Leguminosae, Mimosoideae). II: Fases florales y visitantes en el distrito chaqueño serrano. Darwiniana (1990) 32:27–39.
González-Díaz N, Ackerman JD. Pollination, fruit set, and seed set production in the orchid. Oeceoclades maculata. Lindleyana (1988) 3:150–155.
Gopinathan MC, Babu CR. Breeding systems and pollination in Vigna minima (Leguminosae, Papilionoideae). Plant Systematics and Evolution (1987) 156:117–126.[CrossRef][Web of Science]
Guitián J. Why Prunus mhaleb (Rosaceae) produces more flowers than fruits. American Journal of Botany (1993) 80:1305–1309.[CrossRef][Web of Science]
Guitián P, Guitián J, Navarro L. Fruit set, fruit reduction and fruiting strategy in Cornus sanguina (Cornaceae). American Journal of Botany (1996) 83:744–748.[CrossRef][Web of Science]
Harder L. Pollen removal by bumblebees and its implications for pollen dispersal. Ecology (1990) 71:1110–1125.[CrossRef][Web of Science]
Hedstrom I, Thulin M. Pollination by a hugging mechanism in Vigna vexillata (Leguminosae: Papilionoideae). Plant Systematics and Evolution (1986) 154:275–283.[CrossRef][Web of Science]
Herrera CM. Components of pollinator: ‘quality’: comparative analysis of a diverse insect assemblage. Oikos (1987) 50:79–90.[CrossRef][Web of Science]
Herrera CM, Sánchez-Lafuente AM, Medrano M, Guitián J, Cerdá X, Rey P. Geographical variation in autonomous self-pollination levels unrelated to pollinators service in Helleborus foetidus (Ranunculaceae). American Journal of Botany (2001) 86:1025–1032.
Hoc P, Amela García MT. Floral biology and reproductive system of Phaseolus augusti (Fabaceae). Beiträge zur Biologie der Pflanzen (1998) 70:121–140.
Hoc P, Amela García MT. Biología floral y sistema reproductivo de Phaseolus vulgaris var. aborigeneus (Fabaceae). Revista de Biología Tropical (1999) 47:59–67.[Web of Science]
Hoc P, Di Stilio V, Agulló M, Brizuela M, Palacios R. Biología floral de Vigna longifolia (Leguminosae, Phaseoleae). Darwiniana (1993) 32:27–39.
Horvitz C, Schemske D. Spatio-temporal variation in insect mutualists of a neotropical herb. Ecology (1990) 71:1085–1097.[CrossRef][Web of Science]
Ibrahim A, Coyne D. Genetics of stigma shape, cotyledon position and flower color in reciprocal crosses between Phaseolus vulgaris and Phaseolus coccineus. Journal of the American Society for Horticultural Science (1975) 100:622–626.
Inoue K, Maki M, Masuda M. Evolution of Campanula flowers in relation to insect pollination on islands. In: Floral biology. Studies on floral evolution in animal-pollinated plants—Lloyd DG, Barrett SCH, eds. (1996) New York: Chapman and Hall. 377–400.
Inouye DW. The terminology of floral larceny. Ecology (1980) 61:1251–1253.[CrossRef][Web of Science]
Jain SK. The evolution of inbreeding in plants. Annual Review of Ecology and Systematics (1976) 7:469–495.[CrossRef][Web of Science]
Juncosa AM, Webster BD. Pollination in Lupinus nanus subsp. latifolius (Leguminosae). American Journal of Botany (1989) 76:59–66.[CrossRef][Web of Science]
Kalisz S, Vogler DW. Benefits of autonomous selfing under unpredictable pollinator environments. Ecology (2003) 84:2928–2942.[CrossRef][Web of Science]
Kasagi T, Kudo G. Variations in bumble bee preference and pollen limitation among neighboring populations: comparisons between Phyllodoce caerulea and P. aleutica (Ericaceae) along snowmelt gradients. American Journal of Botany (2003) 90:1321–1327.
Kearns CA, Inouye DW. Techniques for pollination biologists (1993) Niwot, CO: University Press of Colorado.
Knight TM, Steets JA, Vamosi JC, Mazer SJ, Buurd M, Campbel DR, et al. Pollen limitation of plant reproduction: pattern and process. Annual Review of Ecology, Evolution and Systematics (2005) 36:467–497.[CrossRef][Web of Science]
Kreitner GL, Sorensen EL. Stigma development and the stigmatic cuticle of Medicago scutellata. Canadian Journal of Botany (1985) 63:813–818.
Lackey JA. A review of generic concepts in American Phaseolinae (Fabaceae, Faboideae). Iselya (1983) 2:21–64.
Lavin M, Delgado S. Pollen brush of Papilionoideae (Leguminosae): morphological variation and systematic utility. American Journal of Botany (1990) 77:1294–1312.[CrossRef][Web of Science]
Lee TH. Patterns of fruit and seed production. In: Plant reproductive ecology—Lovett Doust J, Lovett Doust L, eds. (1988) Oxford, UK: Oxford University Press. 179–202.
Lindman CAM. Die Blüteneinrichtungen einiger südamerikanischer Pflanzen, Bihang Till K (1902) Svenska Vetensk. Akad. Föerhandlingar Stockholm, 27 Afd. III No. 14.
Lloyd DG, Schoen DJ. Self- and cross-fertilization in plants. I. Functional dimensions. International Journal of Plant Science (1992) 153:358–369.[CrossRef][Web of Science]
Lord E, Heslop-Harrison Y. Pollen and stigma organization in Leguminosae: stigma organization and the breeding system of Vicia faba. Annals of Botany (1984) 54:827–836.
Malti Shivanna KR. The role of the pistil in screening compatible pollen. Theoretical and Applied Genetics (1985) 70:684–686.[CrossRef][Web of Science]
Maréchal R, Mascherpa J, Stainer F. Etude taxonomique d'un groupe complexe d'especies des genres Phaseolus, Macroptilium et Vigna (Papilionaceae) sur la base de donnés morphologiques et polliniques, traitées par l'analyse informatique. Boissiera (1978) 28:1–273.
McDade L, Davidar P. Determinants of fruit set in Pavonia dasypetala (Malvaceae). Oecologia (1984) 64:61–67.[CrossRef][Web of Science]
Mulcahy DL. The rise of angiosperms: a genecological factor. Science (1979) 206:20–23.
Munsell Color. Color charts for plant tissues (1977) New Windsor, NY: Macbeth Division of Kollmorgen Instruments Corporation.
Nilsson LA. Angraecoid orchids and hawkmoths in central Madagascar: specialized pollination systems and generalists foragers. Biotropica (1987) 19:310–318.[CrossRef][Web of Science]
Pauw A. Floral syndromes accurately predict pollination by a specialized oil-collecting bee (Rediviva peringueyi, Melittidae) in a guild of South African orchids (Coryciinae). American Journal of Botany (2006) 93:917–926.
Piper CV. Studies in American Phaseolinae. Contributions from the United States National Herbarium (1926) 22:663–701.
Schemske D, Horvitz C. Variation among floral visitors in pollination ability: a precondition for mutualism specialization. Science (1984) 225:519–521.
Shivanna KR, Owens SJ. Pollen pistil interactions (Papilionoideae). In: Advances in legume biology. Monographs of Systematic Botany of the Missouri Botanical Garden—Stirton CH, Zarucchi JL, eds. (1989) 29:157–182.
Sutherland S. Why hermaphroditic plants produce many more flowers than fruits: experimental tests with Agave mackelveyana. Evolution (1987) 41:750–759.[CrossRef][Web of Science]
Sutherland S, Delph LF. On the importance of male fitness in plants: patterns of fruit-set. Ecology (1984) 65:1093–1104.[CrossRef][Web of Science]
Silva-Montellano A, Eguiarte LE. Geographic patterns in the reproductive ecology of Agave lechuguilla (Agavaceae) in the Chihuahuan desert. I. Floral characteristics, visitors, and fecundity. American Journal of Botany (2003) 90:377–387.
Snow AA, Spira TP. Pollen vigour and the potential for sexual selection in plants. Nature (1991) 352:796–797.[CrossRef][Web of Science]
Sokal RR, Rohlf FJ. Biometry: the principles and practice of statistics in biological research (1995) 3rd edn. New York: W. H. Freeman.
Stebbins GL. Adaptive radiation in Angiosperms. I. Pollination mechanisms. Self-fertilization and population variability in the higher plants. Annual Review of Ecology and Systematics (1970) 1:307–326.[CrossRef]
Stephenson AG. Flower and fruit abortion: proximate causes and ultimate functions. Annual Review of Ecology and Systematics (1981) 12:253–279.[CrossRef][Web of Science]
SYSTAT. SYSTAT version 5·01 for Windows (1992) Evanston IL: Systat, Inc.
Thomson JD. Pollen transport and deposition by bumble bees in Erythronium: influences of floral nectar and bee grooming. Journal of Ecology (1986) 74:329–341.[CrossRef][Web of Science]
Thomson JD. When is it a mutualism? American Naturalist (2003) 162:S1–S9.[CrossRef][Web of Science][Medline]
Thompson JN. The coevolutionary process (1994) Chicago, IL: University of Chicago Press.
Thompson JN. Evaluating the dynamics of coevolution among geographically structured population. Ecology (1997) 78:1619–1623.[CrossRef][Web of Science]
Troll W. über die Blüte von Phaseolus caracalla L. Botanische Notizen (1951) II:37–48.
Tucker SC. Pollination in papilionoid legumes: their nature, development, and variation. Biological Journal of the Linnean Society (1987) 95:181–206.[CrossRef]
Wang YQ, Zhang DX, Renner SS, Chen ZY. A new self- pollination mechanism. Nature (2004) 431:39–40.
Westerkamp C. The co-operation between the asymetric flower of Lathyrus latifolius (Fabaceae – Vicieae) and its visitors. Phyton (1993) 33:127–137.
Westerkamp C. Keel blossoms: bee flowers with adaptations against bees. Flora (1997) 192:125–132.
Wyatt R. Pollinator–plant interactions and the evolution of breeding systems. In: Pollination biology—Real L, ed. (1983) Orlando, FL: Academic Press. 51–95.
Wyatt R. Ecology and evolution of self-pollination in Arenaria uniflora (Caryophyllaceae). Journal of Ecology (1986) 74:403–418.[CrossRef][Web of Science]
Yeo PF. Secondary pollen presentation. Form, function and evolution (1993) New York: Springer-Verlag.
CiteULike 
Connotea 
Del.icio.us What's this?
Related articles in Ann Bot:
- ContentSnapshots
Ann Bot 2008 102: iii.[Extract] [Full Text]
This article has been cited by other articles:
|
|
 |
|
  D. L. Marshall, J. J. Avritt, S. Maliakal-Witt, J. S. Medeiros, and M. G. M. Shaner The impact of plant and flower age on mating patterns Ann. Bot., January 1, 2010; 105(1): 7 - 22. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Compiled by, F. Tooke, T. Chiurugwi, and N. Battey Flowering Newsletter bibliography for 2008 J. Exp. Bot., June 23, 2009; (2009) erp154v1. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||