| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
Center for Applied Tropical Ecology and Conservation (CATEC), University of Puerto Rico, Río Piedras Campus, P.O. Box 23360, San Juan, Puerto Rico 00931-1910 USA; Department of Biology, University of Puerto Rico, Río Piedras Campus, P.O. Box 23360, San Juan, Puerto Rico 00931-3360 USA; Institute for Tropical Ecosystem Studies, University of Puerto Rico, Río Piedras Campus, P.O. Box 21910, San Juan, Puerto Rico 00931-1910 USA
Received for publication June 22, 2006. Accepted for publication January 16, 2007.
ABSTRACT
The effectiveness of flower visitors as pollinators will determine their potential role as selective agents on flower traits. Pitcairnia angustifolia has floral characters that would fit pollination by long-billed hummingbirds, and they should be the most effective pollinators for this plant. To test this prediction, we characterized the behavior of visitors toward flowers and their pollination effectiveness. Coereba flaveola (bananaquits) was the most frequent flower visitor and acted as a primary nectar robber; however, they pollinated incidentally and deposited pollen on stigmas. The endemic short-billed hummingbird Chlorostilbon maugaeus behaved as a secondary robber and did not pollinate flowers. As expected, the long-billed hummingbird, Anthracothorax viridis, was the most efficient visitor in terms of pollen deposition; however, it was the least frequent flower visitor. Introduced Apis mellifera (honeybees) were second in efficiency at depositing pollen and performed one third of the flower visits. Estimates of the expected rate of pollen deposition by each pollinator did not identify a single most effective pollinator. For P. angustifolia at least three flower visitors including an exotic bee and a nectar robber may be equally important to reproductive success. While these results limit our ability to make predictions on the role of hummingbird-pollination on current flower evolution, they do suggest the potential for pollination redundancy among flower visitors for P. angustifolia populations.
Key Words: bananaquits • honeybees • hummingbirds • nectar robbing • Pitcairnia angustifolia • pollination syndromes • Puerto Rico
The process of pollination is often viewed as a mutualistic relationship in which plants offer a reward to flower visitors that transport pollen among conspecific plants resulting in plant reproduction (Faegri and van der Pijl, 1971
; Feinsinger, 1983
; Waser, 1983
). Traditionally, plant evolutionary ecologists have given plant–pollinator interactions important roles on the diversification of angiosperms (Grant and Grant, 1965
; Stebbins, 1970
). The long-standing concept of pollination syndromes is centered on the idea that unrelated plant species have converged on a suite of characteristics that enable them to exploit the morphology, sensory mechanisms, nutritional needs, and overall behavior of their animal visitors (Faegri and van der Pijl, 1971
; Feinsinger, 1983
; Mayfield et al., 2001
). Under this concept, flowers evolve suites of traits that are associated with the attraction and utilization of particular pollinator types. If flower specialization is achieved, then flowers can only be pollinated by a subset of potential animal visitors (Grant and Grant, 1965
; Stebbins, 1970
; Faegri and van der Pijl, 1971
; Feinsinger, 1983
). This concept of specialized pollinator syndromes is now being reexamined given that more generalized pollination systems (i.e., those where flowers may be pollinated by multiple species of pollinators) may be beneficial under certain conditions (Waser et al., 1996
).
The role of animal visitors in the floral evolution of plant species where reproductive success is pollen-limited will ultimately depend on their pollination effectiveness (Herrera, 1988
; Gómez, 2004
). Pollination effectiveness refers to the total contribution to plant fitness by animals that pollinate flowers (Herrera, 1987
, 1989
; Ivey et al., 2003
). This fitness contribution can be estimated in terms of the animal's pollination efficiency and its frequency of flower visits (Stone, 1996
; Herrera, 1987
, 1989
; Gómez and Zamora, 1999
). One way to quantify pollination efficiency is to measure the amount of compatible pollen grains (i.e., stigma pollen loads) deposited on stigmas after a single visit (Stebbins, 1970
; Spears, 1983
; Schemske and Horvitz, 1984
; Herrera, 1987
, 1989
; Inouye et al., 1994
; Mayfield et al., 2001
). Thus, the combined effects of pollination efficiency of flower visitor and its flower visitation rate will determine its pollination effectiveness and therefore its potential role as a selective agent on floral traits (Herrera, 1988
, 1989
; but see Aigner, 2001
).
For some plant species, the most frequent animal visitor cannot be predicted by floral morphological characters (Fishbein and Venable, 1996
; Ollerton, 1998
; Fleming et al., 2001
; Rivera-Marchand and Ackerman, 2006
). For example, various cacti species in the Sonoran desert have flowers with an apparent bat pollination syndrome, but diurnal bird pollinators (e.g., hummingbirds, woodpeckers, finches) are responsible for most seed production in those species (Fleming et al., 2001
). Similarly, queens of the long-tongued bumblebee Bombus oppositus deposit pollen more efficiently on stigmas of flowers of scarlet gilia (Ipomopsis aggregata), despite the flowers' apparent hummingbird syndrome. In years of low hummingbird abundance, the bees are the most common visitors to flowers of this species (Mayfield et al., 2001
). In the ocotillo (Fouquieria splendens) plant with flowers apparently adapted for hummingbird pollination, there is a strong positive relationship between seed set and the abundance of nectar-robbing carpenter bees (Xylocopa), that act as incidental pollinators in years with few hummingbirds (Waser, 1979
). These types of results are common in plant–pollinator studies and have been used to emphasize that the most effective pollinator may not be predicted from flower morphology of the plant. Further, less efficient flower visitors may under certain conditions become important pollinators and thus may be critical to the persistence of plant species (Waser et al., 1996
).
Despite the potential for mutual fitness benefits, plant–pollinator interactions are also viewed as cases of reciprocal exploitation between two interacting species (Brostein, 2001
). Often they are the targets of "cheaters" that forage for floral rewards without providing apparent compensating benefits to the plants that they visit (Morris, 1996
; Irwin and Brody, 1998
, 1999
, 2000
; Maloof and Inouye, 2000
). Nectar robbing, usually achieved by perforations made near the base of the corolla tube is common among hummingbird-pollinated flowers (Maloof and Inouye, 2000
; Faegri and van der Pijl, 1971
; Feinsinger, 1983
). A flower visitor is considered a primary nectar robber when it perforates the corolla or calyx of flowers to obtain nectar without contacting the reproductive parts of the flowers (Inouye, 1980
, 1983
). In contrast, a flower visitor is considered a secondary robber when it uses a perforation already made by primary nectar robbers to get access to floral nectar (Inouye, 1980
, 1983
). Nectar robbers are traditionally viewed as illegitimate visitors with little or no pollination capabilities (Inouye, 1980
, 1983
). At least three mechanisms have been proposed by which nectar robbers may have negative consequences for plants fitness. First, nectar robbers are expected to have direct negative consequences for plant fitness if they damage the flowers and do not pollinate (Irwin and Brody, 1998
, 1999
, 2000
). Second, indirect negative effects on plant fitness may follow when legitimate pollinators avoid robbed flowers and finally nectar robbers may directly interfere with legitimate pollinators and exclude them from flowers (Roubik, 1982
; Irwin et al., 2001
). However, despite the potential for negative effects by nectar robbers on plant fitness, recent reviews have indicated that their effects on plant fitness are quite variable and may even be positive in some cases (Maloof and Inouye, 2000
; Irwin et al., 2001
).
Pitcairnia angustifolia is a bromeliad with floral morphology that suggests pollination by long-billed hummingbirds (i.e., nectar rich, long tubular, red-orange flowers). In previous work on the characterization of the hummingbird flower communities in Puerto Rico, a very close morphological fit was found between hummingbird species and the flowers they visited (Kodric-Brown et al., 1984
; Ricart, 1992
). A close correspondence between floral morphology of P. angustifolia (mean corolla length 32.9 mm; J. J. Fumero-Cabán, unpublished data) and its long-billed hummingbird visitor, Anthracothorax viridis (mean bill length = 24.4 mm, Kodric-Brown et al., 1984
) was interpreted as evidence of a mutualistic relationship between these two species. Kodric-Brown and co-workers (1984) also reported visits by the endemic short-billed hummingbird (Chlorostilbon maugaeus; mean bill length = 13.5 mm, Kodric-Brown et al., 1984
) and bananaquits (Coereba flaveola; Order Paseriformes: mean bill length = 10–11 mm; J. M. Wunderle, International Institute for Tropical Forestry USDA-FS, personal communication), but the visits were illegitimate and resulted in nectar robbing (Kodric-Brown et al., 1984
). We were also aware of the possibility of visits by Apis mellifera (max proboscis length in Puerto Rico = 6 mm; B. Rivera-Marchand, Interamerican University-Bayamón, unpublished. data), but their role in pollination of this species had not been considered. Although a close morphological fit between A. viridis and P. angustifolia may promote a specialized pollination system, testing of this hypothesis also requires data on the role of each flower visitor in the pollination success of this species.
In this study, we investigated the capacity of different flower visitors to deposit pollen on stigmas of P. angustifolia flowers on Puerto Rico. We determined whether these animal species visited flowers legitimately, how frequently they visited the flowers, and how much pollen they deposited on stigmas. We tested the hypothesis that long-billed hummingbirds were the most effective pollinator given their closer morphological fit with flowers relative to other potential flower visitors. We also tested whether or not bananaquits had negative effects on flower pollination given their nectar-robbing activities. Information on the pollination capacity of flowers visitors with very diverse morphologies, origins (native, exotics), and functional roles (robbers and pollination) is useful for evaluating the role of the visitors as future agents of natural selection on flowers traits. They are also valuable for appraising the degree of redundancy of plant–pollinator interactions to plant population success and as a result, the potential resilience of the plant population to changes in the pollinator communities.
MATERIALS AND METHODS
Study site
This study was conducted at the Río Abajo State Forest Reserve located in the north-central part of Puerto Rico (18°19' N, 66°43' W). This reserve is the largest (2275 ha) and highest (300 m above sea level) biological reserve on limestone soils in Puerto Rico. It was established in 1935 and was partly managed for timber plantations. The reserve has a mean annual temperature of 25.3°C and a mean annual precipitation of 1993 mm (Ewel and Whitmore, 1973
).
Study system
Pitcairnia angustifolia Aiton, is a perennial, large terrestrial bromeliad usually located on rocky substrates and often found in large colonies. Plants are clonal with a dense rosette of coriaceous leaves, and they produce racemose or paniculate inflorescences (Acevedo-Rodríguez, 1996
). This species has nectar-rich, perfect flowers with long red-orange corolla tubes that open during daylight hours and last 1 day. Plants are self-compatible, but not autogamous (J. J. Fumero-Cabán, unpublished data). Flowering occurs from late May until late September with peak flowering in approximately the second week of August at the Río Abajo site (J. J. Fumero-Cabán, personal observation). Within Puerto Rico, the distribution of P. angustifolia ranges from 300 to 900 m above sea level, and plants may occur on karst, serpentine, or volcanic soils of moist and wet forests (Kodric-Brown et al., 1984
; Acevedo-Rodríguez, 1996
). The geographical range of this species includes Puerto Rico, the Virgin Islands, and the Lesser Antilles (Acevedo-Rodríguez, 1996
). A voucher specimen was deposited at UPRRP (Fumero-Cabán 1; UPRRP 037264).
Flower visitors and visitation frequency
Observations were conducted from 13 July to 30 August 2002 to identify flower visitors and to determine their visitation frequency and roles as pollinators or robbers. We monitored 60 plants in six patches (10 plants per patch), and each patch was observed on two nonconsecutive days within the season, with the exception of patch 2, which was observed on three different nonconsecutive days. Observation periods were from 0600 hours to 1200 hours and from 1400 hours to 1800 hours. The periods were determined by the flower's life span (corollas opened at dawn and closed at dusk) and included peak hours of pollinator activity. A total of 90 h of field observations were made, but for logistic reasons, the number of observation hours varied slightly among patches. During observation periods, we recorded the number of open flowers per plant, the identity of each flower visitor, the number of visits made by each flower visitor, and whether visits were legitimate or involved nectar robbing. A visit was considered legitimate when the visitor used the corolla opening to access the reward, a behavior that is likely to put them in contact with reproductive parts. A visitor was considered to engage in nectar robbing when it used perforations in the corolla to access the reward. When a flower visitor behaved as a nectar robber, we also recorded whether or not it was a primary robber (i.e., made perforations on flowers) or a secondary robber (i.e., used existing corolla perforations). In addition, we recorded any antagonistic interactions between flower visitors. An antagonistic interaction occurred when an animal actively prevented others from approaching open inflorescences or showed aggressive behaviors that resulted in the eviction of animals that were visiting flowers. Such behaviors are particularly common in territorial hummingbird species (e.g., Wolf, 1975
; Temeles et al., 2004
). For each plant observed, we calculated the visitation rate by each animal species as the number of visits x flower –1·hr–1·plant–1 and then averaged the value for each pollinator type for plants within each patch (N = 6). Differences in mean plant visitation rates among flowers visitors were analyzed using a one-way ANOVA and a Tukey test of multiple comparisons to detect differences among visitor types. Based on literature reports (Kodrick-Brown et al., 1984) and preliminary observations of flower visitors of P. angustifolia, the potential flower visitors of this species were the long-billed hummingbird A. viridis (green mango), A. mellifera (honeybee), the passerine bird C. flaveola (bananaquit), and the short-billed hummingbird C. maugaeus (Puerto Rican emerald).
Pollination efficiency of flower visitors
Pollination efficiency of each flower visitor was estimated as the number of pollen grains deposited on a virgin stigma in a single visit (i.e., stigma pollen load, Inouye et al., 1994
). On different dates, during the flowering season in 2002, we marked and covered elongated flower buds the day before they opened with caps constructed from plastic drinking straws to prevent flower visitation. The next day the straw caps were removed, and each flower was exposed for a single visit. Upon visitation, the identity of the visitor was recorded, and the stigma was collected for pollen counts. In the field, stigmas were mounted on individual microscope slides in a glycerin–gelatin medium, and each stigma was then squashed with a cover slip (Kearns and Inouye, 1993
). Stigma preparations were placed in a plastic bag and transported in a cooler with ice for processing in the laboratory. All pollen grains deposited on the stigma were counted under a microscope at 200x. Twenty stigmas were collected per visitor species, and 20 unvisited flowers were collected for the control (controls). These control flowers were subjected to the same treatment as the visited flowers except that no visits were permitted. The marked color contrast between the stigma (red) and the pollen grains (yellow) facilitated pollen counts without the need of staining or fixation techniques. Differences in mean pollen load deposited on stigmas among flower visitors were analyzed using a one-way ANOVA followed by a Tukey test of multiple comparisons to detect differences among visitor types. To meet the normality requirements for this analysis of variance and due to multiple zeros in the data set, the data were square-root transformed using the following transformation:
|
| (1) |
).
Overall pollination effectiveness
The relative contributions to plant fitness by different flower visitors consist of two multiplicative components of pollination effectiveness (Herrera, 1987
, 1989
): pollination efficiency, estimated here as the amount of pollen deposited on stigmas per visit (i.e., stigma pollen loads, Inouye et al., 1994
) and the pollinator's visitation rate. To estimate the overall pollination effectiveness of each pollinator species, we multiplied the pollination efficiency by visitation rate for each plant within each floral patch observed. The product of these two components represents the rate of pollen grains deposited per stigma per plant per patch by each type of flower visitor. This index of pollination effectiveness allowed us to evaluate the relative contribution of different flower visitors to plant fitness. Differences among flower visitors in pollination effectiveness per patch were analyzed using a Welch's modify -stat, which tests the hypothesis of equality of group means without assuming variance homogeneity among groups (Welch, 1951
; Quinn and Keough, 2002
). All statistical analyses for this study were done with the statistical software package JMP V 4.0.4 (SAS Institute, Cary, North Carolina, USA).
RESULTS
Flower visitors and visitation frequency
Flowers of P. angustifolia were visited by individuals of four animal species, which made 1146 flower visits in a total of 90 h of observations. The long-billed hummingbird Anthracothorax viridis tended to visit the flowers legitimately. Apis mellifera also visited the flowers legitimately, but unlike A. viridis, they collected pollen and were unable to reach the nectar at the base of the corolla tube. Individuals of Coereba flaveola behaved as primary nectar robbers perforating flowers of P. angustifolia at the base with no further damage; Chlorostilbon maugaeus, the short-billed hummingbird, behaved as a secondary robber. In terms of total visits to flowers, C. flaveola was the most frequent visitor and made 57.9% (664) of total flower visits. Apis mellifera was the second most frequent, making 29.5% (338) of total flower visits. Flower visits by hummingbirds were much less frequent. Individuals of long-billed A. viridis made only 9.7% (111) of the flower visits, whereas short-billed C. maugaeus made only 33 flower visits, accounting for only 2.9% of total flower visits. Average plant visitation rates by C. flaveola were nine times higher relative to A. mellifera. In addition, average plant visitation rates by individuals C. maugaeus and A. viridis were not significantly different from each other but were fewer than those of C. flaveola and A. mellifera (one-way ANOVA, F3,23 = 22.64, P < 0.0001; Fig. 1a).
|
Pollination efficiency of flower visitors
There were significant differences in the mean pollen load deposited on stigmas of virgin flowers by the different flowers visitors of P. angustifolia (one-way ANOVA F4,95 = 172.67, P = 0.001; Fig. 1b). As expected, the long-billed hummingbird A. viridis, was the most efficient pollinator in terms of pollen deposition on single flowers. Apis mellifera was the second most efficient at depositing pollen but had only one third of the efficiency of A. viridis. The nectar robber C. flaveola deposited some pollen upon visitation, but had only about 6.7% of the efficiency of A. viridis and 20% of the efficiency of A. mellifera. Pollen deposition on single flowers by the secondary robber C. maugaeus was not statistically different from the mean number of pollen grains found in control flowers (Fig. 1b).
Overall pollination effectiveness
When both components of pollination effectiveness (i.e., pollen deposition and visitation frequency) were combined, there were significant differences in pollination effectiveness among flower visitors of P. angustifolia (Fig. 1c; Welch F9,8 = 20.38, P = 0.0002). Chlorostilbon maegaeus had a much lower effectiveness than did A viridis, A. mellifera, and C. flaveola. These last three had very similar pollination effectiveness.
DISCUSSION
Flowers of P. angustifolia in the Rio Abajo Forest Reserve were visited by three species of pollinators with different visitation behaviors (long-billed hummingbird, honeybee, bananaquit). These pollinators were not equally efficient at depositing pollen after one flower visit. However, once their visitation rates were taken into account, all pollinators were equally effective at depositing pollen because flowers on a plant were expected to receive the same amount of pollen grains from each pollinator type over time.
As expected based on floral morphology, A. viridis was the most efficient pollinator per flower visit. With its long bill, A. viridis was the only visitor capable of legitimately reaching the nectar located at the base of the tubular corolla. Interestingly, when one considers only the effective pollinators, the two components of pollination effectiveness (pollination efficiency and visitation rates) were not correlated. Although the long-billed hummingbird was the most efficient visitor, it made many fewer visits per flower-hour-plant than bananaquits or honeybees (Fig. 1c).
There are at least three possible explanations for this incongruence between pollination efficiency and visitation rates by A. viridis. First, our observation periods showed that C. maugaeus, a visitor that acted as a secondary robber, effectively excluded A. viridis at least 17 times from P. angustifolia patches. Hummingbirds have high metabolic demands and thus require high rates of energy intake (Mongomerie and Gass, 1981
). Aggressive interactions with other flower visitors may reduce the levels of net energy intake expected by A. viridis from P. angustifolia flowers and prevent these birds from visiting. A second possibility, although, not necessarily mutually exclusive, is that the nectar depletion caused by visits of the two species of nectar robbers (i.e., C. flaveola and C. maugaeus), leads to low visitation rates by long-billed hummingbirds. Given that flower preferences by hummingbirds are at least partially based on nectar availability (Meléndez-Ackerman et al., 1997
; Meléndez-Ackerman and Campbell, 1998
; Blem et al., 1997
; Biernaskie et al., 2002
), low visitation rates by long-billed hummingbirds should also be expected in P. angustifolia patches to the extent that nectar robbing reduces the nectar standing crop in this species (Irwin and Brody, 1998
). A third alternative is the availability of other local flowering species that may be competing with P. angustifolia for pollinator service. Indeed, at this site long-billed hummingbird visits to flowers of the common timber tree Hibiscus elatus were observed although not quantified (J. J. Fumero-Cabán, personal observation).
Apis mellifera is a generalist forager and visits flowers with different pollination syndromes (Roubik, 1980
; Aizen and Feinsinger, 1994
). In agreement with other studies, A. mellifera had lower pollination efficiency than the native pollinator (Vaughton, 1996
; Gross and Mackay, 1998
; Hansen et al., 2002
). Nevertheless, they visited flowers much more frequently than did A. viridis. Once visitation rates were taken into account, A. mellifera was as effective as A. viridis at depositing pollen. These results were unexpected given that there is a better morphological correspondence between the bills of A. viridis and P. angustifolia flowers relative to all remaining flower visitors. Moreover, where introduced, Apis mellifera is often considered to be detrimental to populations of native pollinators and thought to disrupt specialized relationships between native bee pollinators and their plants (Roubik, 1980
). That was not the case in our study, where the recently introduced A. mellifera (now mostly feral) were found to be good pollinators of P. angustifolia and as effective as the bird pollinators of this species.
Flower visits by C. maugaeus, a secondary nectar robber indeed did not result in pollen transfer (Fig. 1a). Nevertheless, despite its behavior as primary nectar robbers, we found that C. flaveola did have pollination capabilities albeit incidental. We observed that flowers of P. angustifolia bent and struck against the bananaquit's body during the process of nectar robbing and that this bending motion of flowers most likely causes self-pollen to fall on the stigma. On a per visit basis this mode of pollination was much less efficient at pollen deposition than that of legitimate pollinators (i.e., A. viridis and A. mellifera, Fig. 1b). However, due to their relatively high visitation rates, bananaquits are still expected to deposit as much pollen per flower as the legitimate pollinators (Fig. 1c). The end result is that P. angustifolia flowers are visited by three animal species (two natives, one exotic) that are equally effective at depositing pollen even when long-billed hummingbirds can be much more efficient at pollen deposition on a per flower basis.
How do we reconcile the presence of a floral phenotype that seems adaptive to hummingbird pollination but yet maintains multiple pollinators? First, honeybees are a recently introduced species in Puerto Rico (fewer than 400 yr, Cox, 1994
). Pitcairnia angustifolia plants are long-lived perennials; thus one possibility is that there has not been enough time for honeybees to drive radical evolutionary changes in floral phenotypes. Second, recent models addressing fitness trade-offs between pollinators suggest that simply ranking pollinators by their relative effectiveness on the mean floral phenotype may not be enough to predict floral specialization (Aigner, 2001
). These models suggest that factors such as the shape of fitness functions related to pollination service by individual pollinators as well as the strength of fitness interactions among available pollinators could be more important than pollination effectiveness in predicting floral specialization. The models argue that data on the fitness contributions by different pollinators over a variety of floral phenotypes would be required in order to address hypotheses on the potential for floral specialization (Aigner, 2001
).
One potential limitation of our study is that we only considered the maternal component of pollination effectiveness (i.e., pollen deposition). Pollinators may also differ in their pollination effectiveness by affecting the rate of pollen transfer among flowers (Thomson and Goodell, 2001
). Pollen grooming by pollen-collecting bees may lead to considerable pollen waste from the perspective of plants that depend on animal-mediated pollination (Stanton et al., 1992
; Holsinger and Thomson, 1994
). Indeed, Apis mellifera has shown lower transfer effectiveness relative to other bee pollinators under certain conditions (Thomson and Goodell, 2001
). Similarly, pollination by C. flaveola is indirect and thus they may not be able to effect any pollination if flowers have already been visited and anthers have already been depleted of pollen. Thus, all other things being equal, long-billed hummingbirds could still be more effective pollinators if they circulate larger portions of pollen relative to honeybees and bananaquits.
Recent studies have shown that patterns of flower visitation in pollination systems are often generalized (Waser et al., 1996
), especially for native and endemic plants on oceanic islands (Carlquist, 1974
; Olesen et al., 2002
). Local populations of P. angustifolia were effectively visited by three species (two native, one exotic) that differ functionally, yet no single flower visitor was statistically more effective than the other two (Fig. 1c). For selection to favor a generalized pollination system in P. angustifolia populations, plants must be exposed to consistent conditions of pollen limitation. Under this assumption, honeybee pollination could be favored under conditions of nectar robbing because the bees do not discriminate between robbed and intact nonrobbed flowers (Fumero-Cabán, 2004
); they are frequent visitors and are not chased away by competitors. Less efficient but more abundant pollinators may compensate for the lack of main pollinators (Waser, 1979
; Fleming et al., 2001
; Mayfield et al., 2001
; Rivera-Marchand and Ackerman, 2006
). Coereba flaveola is a primary nectar robber, yet it still deposits a significant amount of pollen and thus behaves as a robber-like pollinator (Navarro, 2000
, 2001
). This type of activity can be particularly favorable to flowering plants during periods of low abundance of primary pollinators (Waser, 1979
; Navarro, 2000
, 2001
). Conditions leading to temporal variation in the abundance of pollinators may be common on tropical islands like Puerto Rico that are subjected to frequent hurricanes (Weaver, 1986
; Rathcke, 2000
). Such large disturbances may also lead to unpredictable changes in the abundance of pollinators and may reduce the abundance of nectarivorous bird populations (Wunderle, 1995
; Rathcke, 2000
, 2001
). To the extent that fluctuations in pollinator abundance are indeed dramatic at this site, a pollination system with multiple pollinators (and even some redundancy) could also be favored over time. Long-term studies are needed to determine the temporal stability of the interaction between P. angustifolia and its pollinators and their long-term role in floral evolution on this plant species.
Conclusions
In summary, we have shown that despite its suite of traits typical of a hummingbird pollination syndrome, P. angustifolia flowers in Puerto Rico were visited by several pollinator species (long-billed hummingbird, honeybee, bananaquit). Whereas long-billed hummingbirds were indeed the most efficient visitors at depositing pollen, once visitation rates were considered, estimates of pollination effectiveness that take into account the maternal component of fitness (i.e., pollen deposition) indicate that they were no more effective than the remaining pollinators. On the other hand, these estimates are only a partial approximation of pollination effectiveness because they do not include data on pollen removal and donation, which may provide additional information on plant fitness. Overall, our results indicate that the interaction between P. angustifolia and its pollinators and nectar robbers is complex. Whether or not the presence of multiple pollinators of P. angustifolia selects against or in favor of floral specialization would ultimately be determined by trade-offs between negative and positive effects of their activities on total plant reproduction.
FOOTNOTES
1 The authors thank J. D. Ackerman, L. McDade, J. M. Wunderle, J. K. Zimmerman, and two anonymous reviewers for comments on the manuscript, and the Department of Natural Resources and the Environment for permits to do research work at the site. This research was funded by NSF-Luquillo LTER (DEB-900002456), NSF-CREST (HRD-0206200) through the Center for Applied Tropical Ecology and Conservation of the University of Puerto Rico and the Research Experience for Undergraduates Program in Tropical Ecology and Evolution (NSF-REU DEB-9912316). 
2 Author for correspondence (e-mail: josejfumero{at}yahoo.com
)
LITERATURE CITED
Acevedo-Rodríguez P.. 1996. Flora of St. John, U.S. Virgin Islands New York Botanical Garden Press, Bronx, New York, USA.
Aigner P. A.. 2001. Optimality modeling and fitness trade-off: when should plants become pollinator specialists?. Oikos 95: 177-184.
Aizen M. A. Feinsiger P.. 1994. Forest fragmentation, pollination, and plant reproduction in a chaco dry forest, Argentina. Ecology 75: 330-351.
Biernaskie J. M. Cartar R. V. Hurly T. A.. 2002. Risk-averse inflorescence departure in hummingbirds and bumble bees: could plants benefit from variable nectar volumes?. Oikos 98: 98-104.
Blem C. R. Blem L. B. Cosgrove C. C.. 1997. Field studies of rufous hummingbird sucrose preference: does source height affect test results?. Journal of Field Ornithology 68: 245-252.
Bronstein J. L.. 2001. Mutualisms. In C. W. Fox, D. A. Roff, D. J. Fairbairn [eds.], Evolutionary ecology: concepts and case studies Oxford University Press, New York, New York, USA.
Carlquist S.. 1974. Island biology Columbia University Press, New York, New York, USA.
Cox R.. 1994. Africanized honeybees in Puerto Rico. American Bee Journal 134: 668.
Ewel J. J. Whitmore J. L.. 1973. The ecological life zones of Puerto Rico and the US Virgin Islands. Forest Service Research Paper ITF-18 Institute of Tropical Forestry, Rio Piedras, Puerto Rico.
Faegri K. van der Pijl L.. 1971. The principles of pollination ecology Pergamon Press, New York, New York, USA.
Feinsinger P.. 1983. Coevolution and pollination. In D. J. Futuyma, M. Slatking [eds.], Coevolution 282-310 Sinauer, Sunderland, Massachusetts, USA.
Fishbein M. Venable D. L.. 1996. Diversity and temporal change in the effective pollinators of Asclepias tuberosa. Ecology 77: 1061-1073.
Fleming T. H. Sahley C. T. Holland J. N. Nason J. D. Hamrick J. L.. 2001. Sonoran desert columnar cacti and the evolution of generalized pollination systems. Ecological Monographs 71: 511-530.
Fumero-Cabán J. J.. 2004. Effects of nectar robbing by bananaquits on the reproductive success of Pitcairnia angustifolia (Bromeliaceae) Master's thesis, University of Puerto Rico, Río Piedras, Puerto Rico.
Gómez J. M.. 2004. Generalización en las interacciones entre plantas y polinizadores. Revista Chilena de Historia Natural 75: 105-116.
Gómez J. M. Zamora R.. 1999. Generalization vs. specialization in the pollination system of Hormathophylla spinosa (Cruciferae). Ecology 80: 796-805.
Grant K. Grant V.. 1965. Flower pollination in the phlox family Columbia University Press, New York, New York, USA.
Gross C. L. Mackay D.. 1998. Honeybees reduce fitness in the pioneer shrub Melastoma affine (Melastomataceae). Biological Conservation 86: 169-178.
Hansen D. M. Olesen J. M. Jones C. G.. 2002. Trees, birds and bees in Mauritius: exploitative competition between introduced honeybees and endemic nectarivorous birds. Journal of Biogeography 29: 721-734.
Herrera C. M.. 1987. Components of pollination "quality": comparative analysis of a diverse insects assemblage. Oikos 50: 79-90.
Herrera C. M.. 1988. Variation in mutualisms: the spatiotemporal mosaic of a pollination assemblage. Biological Journal of the Linnean Society 35: 95-125.
Herrera C. M.. 1989. Pollinator abundance, morphology, and visitation rate: analysis of the "quantity" component in a plant–pollinator system. Oecologia 80: 241-248.
Holsinger H. E. Thomson J. D.. 1994. Pollen-discounting in Erythronium grandiflorum: mass-action estimates from pollen transfer dynamics. American Naturalist 144: 799-812.
Inouye D. W.. 1980. The terminology of floral larceny. Ecology 61: 1251-1252.
Inouye D. W.. 1983. The ecology of nectar robbing. In B. Bentley, T. Elias [eds.], The biology of nectarines 153-173 Columbia University Press, New York, New York, USA.
Inouye D.W. Gill D. E. Dudash M. R. Fenster C. B.. 1994. A model of lexicon for pollen fate. American Journal of Botany 81: 1517-1530.
Irwin R. E. Brody A. K.. 1998. Nectar robbing in Ipomopsis aggregata: effects on pollinator behavior and plant fitness. Oecologia 116: 516-527.
Irwin R. E. Brody A. K.. 1999. Nectar-robbing bumble bees reduce the fitness of Ipomopsis aggregata (Polemoniaceae). Ecology 80: 1703-1712.
Irwin R. E. Brody A. K.. 2000. Consequences of nectar robbing for realized male function in a hummingbird-pollinated plant. Ecology 81: 2637-2643.
Irwin R. E. Brody A. K. Waser N. M.. 2001. The impact of floral larceny on individuals, populations, and communities. Oecologia 129: 161-168.
Ivey C. T. Martínez P. Wyatt R.. 2003. Variation in pollinator effectiveness in swamp milkweed, Asclepias incarnata (Apocynaceae). American Journal of Botany 90: 214-225.
Kearns C. A. Inouye D. W.. 1993. Techniques for pollination biologists University Press of Colorado, Niwot, Colorado, USA.
Kodric-Brown A. Brown J. H. Byers G. S. Gori D. F.. 1984. Organization of a tropical island community of hummingbirds and flower. Ecology 65: 1358-1368.
Maloof J. Inouye D. W.. 2000. Are nectar robbers cheaters or mutualists?. Ecology 81: 2651-2661.
Mayfield M. M. Waser N. M. Price M. V.. 2001. Exploring the "most effective pollination principle" with complex flowers: bumblebees and Ipomopsis aggregata. Annals of Botany 88: 591-596.
Meléndez-Ackerman E. Campbell D. R.. 1998. Adaptive significance of flower color and inter-trait correlations in an Ipomopsis hybrid zone. Evolution 52: 1293-1303.
Meléndez-Ackerman E. Campbell D. R. Waser N. M.. 1997. Hummingbird behavior and mechanisms of selection on flower color in Ipomopsis. Ecology 78: 2532-2541.
Montgomerie R. D. Gass C. L.. 1981. Energy limitation in hummingbird populations in tropical and temperate communities. Oecologia 50: 162-165.
Morris W. F.. 1996. Mutualism denied? Nectar-robbing bumblebees do not reduce female or male success of bluebells. Ecology 77: 1145-1162.
Navarro L.. 2000. Pollination ecology of Anthyllis vulneraria subsp. vulgaris (Fabaceae): nectar robbers as pollinators. American Journal of Botany 87: 980-985.
Navarro L.. 2001. Reproductive biology and effects of nectar robbing on fruit production in Macleania bullata (Ericaceae). Plant Ecology 152: 59-65.
Olesen J. M. Eswkildsen L. I. Venkatasamy S.. 2002. Invasion of pollination networks on oceanic islands: importance of invader complexes and endemic super generalist. Diversity and Distributions 8: 181-192.
Ollerton J.. 1998. Sunbird surprise for syndromes. Nature 394: 726-727.
Quinn G. P. Keough M. J.. 2002. Experimental design and data analysis for biologists Cambridge University Press, Cambridge, UK.
Rathcke B. J.. 2000. Hurricane causes resource pollination limitation of fruit set in a bird-pollinated shrub. Ecology 81: 1951-1958.
Rathcke B. J.. 2001. Pollination and predation limit fruit set in a shrub, Bourreria succulenta (Boraginaceae), after hurricanes on San Salvador Island, Bahamas. Biotropica 32: 330-338.
Ricart C. M.. 1992. Feeding ecology of nectar-feeding birds in the lower montane wet forest life zone, Maricao, Puerto Rico. Acta Científica 6: 41-48.
Rivera-Marchand B. Ackerman J. D.. 2006. Bat pollination breakdown in the Caribbean columnar cactus Pilosocereus royenii. Biotropica 38: 635-642.
Roubik D. W.. 1980. Foraging behavior of competing Africanized honeybees and stingless bees. Ecology 61: 836-845.
Roubik D. W.. 1982. The ecological impact of nectar-robbing bees and pollinating hummingbird on a tropical shrub. Ecology 63: 354-360.
Schemske D. W. Horvitz C. C.. 1984. Variation among floral visitors in pollination ability: a precondition for mutualism specialization. Science 225: 519-521.
Sokal R. Rohlf F. J.. 1995. Biometry, 3rd ed W. H. Freeman, New York, New York, USA.
Spears E. E.. 1983. A direct measure of pollinator effectiveness. Oecologia 57: 196-199.
Stanton M. L. Ashman T. L. Galloway L. F. Young H.. 1992. Estimating male fitness of plants in natural populations. In R. Wyatt [ed.], Ecology and evolution of plant reproduction: new approach 62-90 Chapman and Hall, New York, New York, USA.
Stebbins G. L.. 1970. Adaptive radiation of reproductive characteristics in angiosperms. I. Pollination mechanisms. Annual Review of Ecology and Systematics 1: 307-326.[Medline]
Stone J. L.. 1996. Components of pollination effectiveness in Psychotria suerrensis, a tropical distylous shrub. Oecologia 107: 504-512.
Temeles E. J. Muir A. B. Slutsky E. B. Vitousek M. N.. 2004. Effect of food reductions on territorial behavior of purple-throated caribs. Condor 106: 691-695.
Thomson J. D. Goodell K.. 2001. Pollen removal and deposition by honeybee and bumblebee visitors to apple and almond flowers. Journal of Applied Ecology 38: 1032-1044.
Vaughton G.. 1996. Pollination disruption by European honeybees in the Australian bird-pollinated shrub Grevillea barlkyana (Proteaceae). Plant Systematics and Evolution 200: 89-100.
Waser N. M.. 1979. Pollinator availability as a determinant of flowering time in ocotillo (Flouquieria splendens). Oecologia 39: 107-121.
Waser N. M.. 1983. The adaptive nature of floral traits: ideas and evidence. In L. A. Real [ed.], Pollination biology 241-285 Academic Press, New York, New York, USA.
Waser N. M. Chittka L. Price M. V. Williams N. W. Ollerton J.. 1996. Generalization in pollination systems, and why it matters. Ecology 77: 1043-1060.
Weaver P. L.. 1986. Hurricane damage and recovery in the montane forests of the Luquillo Mountains of Puerto Rico. Caribbean Journal of Science 22: 53-70.
Welch B. L.. 1951. On the comparison of several mean values: an alternative approach. Biometrika 38: 330-336.[Medline]
Wolf L. L.. 1975. Prostitution behavior in a tropical hummingbird. Condor 77: 140-144.
Wunderle J. M.. 1995. Responses of bird populations in a Puerto Rican forest to hurricane Hugo: the first 18 months. Condor 97: 879-896.
This article has been cited by other articles:
|
|
 |
|
  C. de Vega, M. Arista, P. L. Ortiz, C. M. Herrera, and S. Talavera The ant-pollination system of Cytinus hypocistis (Cytinaceae), a Mediterranean root holoparasite Ann. Bot., May 1, 2009; 103(7): 1065 - 1075. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
