Pattern and consequences of floral herbivory in four sympatric Ipomoea species.
Flowers of animal-pollinated plants should be under strong selection to be attractive to pollinators owing to their central role in reproduction (Waser, 1978; Johnson and Dafni, 1998; Wolfe and Krstolic, 1999; Kawagoe and Suzuki, 2004). Features such as flower color, size, number and type of reward all participate in attracting pollinators for pollen dissemination and receipt (Waser and Price, 1981; Schemske and Horovitz, 1984; Nilsson, 1988; Harder, 1990; Herrera, 1993; Hodges, 1995; Johnson et al., 1995; Campbell et al., 1996; Conner and Rush, 1996; Galen, 1996). Angiosperms that utilize the same pollinator or pollinator guild (e.g., hummingbirds) often have flowers that have similar suites of floral traits referred to as pollination syndromes (Faegri and van der Pijl, 1979; Fenster et al., 2004). Yet, traits that serve to differentially attract pollinators may at the same time differentially attract organisms that feed on flowers (Armbruster, 1997; Matter et al., 1999; Galen and Cuba, 2001; Leege and Wolfe, 2002; Canela and Sazima, 2003; Held and Potter, 2004). Flowers, therefore, act as resources for both mutualistic pollinators and antagonistic predators (Armstrong and Marsh, 1997; Wolfe, 1997; Irwin and Brody, 1998; Krupnick and Weis, 1999; Galen and Cuba, 2001; Irwin et al., 2004).
Florivores can have obvious direct effects on plant fitness if they damage the reproductive structures (stamens and pistils), resulting in the loss of gametes (Breedlove and Ehrlich, 1968; Cunningham, 1995; Krupnick and Weis, 1999). Damage to attractive structures (corolla and calyx) can also indirectly affect reproductive success if these damaged flowers receive fewer pollinator visits, or visits of lower quality (Irwin and Brody, 1999; Krupnick et al., 1999; Mothershead and Marquis, 2000; McCall, 2006; Pohl et al., 2006). Despite the impact on plant reproductive success, little is known about how florivores respond to variation in floral traits that clearly function in pollinator attraction (Leege and Wolfe, 2002; Irwin et al., 2004).
The goal of this study was to evaluate whether sympatric Ipomoea (morning glory) species with similar pollination syndromes also experience similar florivory intensities (proportion of flowers damaged), and what role native vs. introduced status plays in shaping florivory intensity. In the southeast US, we observed lepidopteran larvae (predominately Spodoptera spp.) causing damage to flowers, but not harming foliage, of four common Ipomoea species. These vining plants produce flowers that differ in appearance and can be grouped into two different pollination syndromes (Faegri and van der Pijl, 1979). Ipomoea hederifolia and L quamoclit have red-orange and red flowers, respectively, with narrow tubular corollas, dilute nectar and are predominately visited by hummingbirds and sulfur butterflies (hereafter referred to as 'hummingbird-pollinated' Ipomoea). Ipomoea hederacea and L cordatotriloba have blue and purple flowers, respectively, with larger flared corolla openings, concentrated nectar and are predominately visited by bumble bees (hereafter referred to as 'bumble bee-pollinated' Ipomoea) (Wolfe and Sowell, 2006). In addition to differences in floral morphology, the four Ipomoea species differ in their origins. Whereas Ipomoea cordatotriloba is native to the southeast US, the remaining three species were introduced from tropical America (Godfrey and Wooten, 1981) and are considered noxious weeds by the United States Department of Agriculture (2001). A popular idea in the invasion biology literature is that invasiveness results when introduced species escape their enemies (e.g., Elton, 1958; Crawley, 1987; Wolfe, 2002). Thus, these four species provide an excellent system to evaluate not only the possible antagonistic effects of attraction and attack, but whether species are native or introduced, because individuals of the four Ipomoea species grow intermingled at several local sites.
Specifically, we addressed the following questions: (1) Does the intensity and pattern of florivory differ between the Ipomoea species? If so, does pollination syndrome or origin (native/introduced) explain variation in florivory among the focal Ipomoea species? Are these patterns consistent between years? (2) What flower parts, attractive (corolla) or reproductive structures (stamens, pistil), are damaged or consumed by florivores? (3) What is the relative effect of damage to attractive vs. reproductive structures on fruit and seed set?
Study system.--This study was conducted on four Ipomoea species common in southeast Georgia (Bulloch County) in 1999, 2000 and 2005: L cordatotriloba Dennst., L hederifolia L., I. quamoclit L. and I. hederacea (L.) Jacq., with nomenclature following Austin and Huaman (1996), with subgeneric delineation following Miller et al. (1999). A recent molecular stud), (Miller et al., 1999) upholds previous morphologically-defined systematic placement of the bumble bee-pollinated L hederacea and L cordatotriloba into different subgenera, Ipomoea and Eriospermum, respectively. The hummingbird-pollinated I. hederifolia and I. quamoclit, are closely related and belong to subgenus Quamoclit.
Ipomoea cordatotriloba is a native perennial, whereas the remaining three species are nonnative annuals (Godfrey and Wooten, 1981). In southeast Georgia, the species flower from Jul. to first frost (typically early Nov.) and exhibit strongly congruent seasonal flowering phenologies (Wolfe and Sowell, 2006). Individual flowers remain open for less than one day, with anthesis beginning at or shortly before daybreak (Wolfe and Sowell, 2006). Fruits mature in 3-4 wk, and the dehiscent fruit capsules contain a maximum of six seeds in I. hederacea, and four seeds in I. hederifolia, L quamoclit and L cordatotriloba.
Measuring the magnitude of florivory.--Our goal was to determine the magnitude and consequences of florivory for each of the four Ipomoea species. We assessed florivory intensity (proportion of flowers with damage) at approximately weekly intervals throughout each flowering season at seven frequently disturbed roadsides. All sites were separated by at least several miles, and were alongside roads or railways. Each site contained up to several hundred flowers of each Ipomoea species daily. The specific sites used varied among years owing to availability of flowering material because some sites were mown or received herbicide application from state transportation workers before or shortly after our surveying began. These sites were eliminated from further surveying within the study year and not included in analyses. Sites surveyed by year were: 1999-Hunter's Pointe, Hwy 24, Hwy 67; 2000-Hunter's Pointe, Hopeulikit, Hwy 80E; 2005-Mill Creek, Railroad. Each year surveying started in early Aug. and continued until late Oct. or early Nov., encompassing peak flowering and the majority of the seasonal flowering period of the four Ipomoea species (Wolfe and Sowell, 2006).
Although a slightly different sampling method was used each year, the overall approach provided a robust measure of florivory intensity. In 1999 we performed exhaustive censuses of all intact and damaged Ipomoea flowers throughout the entire site (no subsampling within sites) on each survey date to obtain site-level florivory intensity. In 1999 and 2005, we distinguished between three types of floral damage: (1) attractive structures (corolla only), (2) reproductive structures (pistil and stamens only) and (3) whole flower (attractive and reproductive structures) to determine the patterns of floral damage among Ipomoea species.
In 2000 and 2005, we counted the number of intact and damaged flowers within permanent plots distributed regularly throughout the sites. In 2000, 5 [m.sup.2] plots were established in early Jul., just before flowering began, but after growing vines were clearly visible, and were distributed approximately every 10 m linearly throughout the sites. In 2005, 1 [m.sup.2] plots were established along transects approximately every 2 m throughout the sites. Within plots, all flowers of each Ipomoea species were counted, and recorded as being intact or damaged. We summed intact and damaged flowers across plots within a site to yield an estimate of site-level florivory intensity (proportion of surveyed flowers with damage) for each Ipomoea species. Over the course of the three study years, we surveyed approximately 48,000 flowers among the four Ipomoea species (Table 1).
For all years, each census was performed at midday, after florivore activity ceased (Sowell and Wolfe, pers. obs.), to provide a realistic assessment of florivory intensity on a daily cohort of flowers.
Measuring the consequences of florivary.--We determined the consequences of florivory by comparing fruit and seed production in damaged and undamaged flowers. Our protocol was to locate and tag naturally damaged flowers along with adjacent undamaged flowers (within 30 cm) around midday after florivore activity had ceased. Small paper tags placed on the pedicel identified flowers as damaged or undamaged. We returned 3-4 wk later to collect the mature fruit. Flowers that did not set fruit retained tags on their pedicels and were recorded as unsuccessful flowers.
We used three sites in 1999 (Hunter's Pointe, Hwy 24, Hwy 67) and one site in 2005 (Mill Creek) to quantify fruit and seed set. Fruit set for Ipomoea cordatotriloba was excluded from analysis in 1999 because too few flowers were produced on days when we tagged flowers to follow through fruit set, most likely due to drought conditions. In 1999, we did not differentiate type of floral damage (attractive, reproductive organs) in our damage treatment, tagging 109 damaged and 109 undamaged flowers across five days in early Sept. and early Oct. In 2005, we tagged 105 damaged flowers and 347 undamaged flowers across 14 d from early Sept. through early Nov. and recorded the type of floral damage (attractive, reproductive structures) to determine the fate of flowers that had their reproductive organs damaged relative to those that received damage only to attractive structures.
Statistical analyses.--All analyses were conducted with JMP version 3.2.1 (SAS, 1997). Nonparametric analyses were used when assumptions of normality and heteroscedasticity could not be met by data transformation. We utilized ranks of florivory intensity (proportion of flowers damaged) as our response variable, in a Scheirer-Ray-Hare extension of Kruskal-Wallis test (Sokal and Rohlf, 1995).
For analyses on florivory intensity, we included the following in the model: species, site, species by site interaction and number of flowers available to florivores of each Ipomoea species on each census day as a covariate. Florivory intensity did not vary by sampling date within a season, and we have excluded sampling date from the analyses. Separate analyses, utilizing the same model stated above, were conducted for each of the 3 y because a different suite of sites was sampled each year. We tested for differences in rank florivory intensity between the individual Ipomoea species with post-hoc analyses, examining the affects of origin and pollination syndrome, with significance values adjusted by a sequential Bonferroni correction (Rice, 1989). The frequency of damage to different floral tissues (attractive, reproductive structures) in the four species was analyzed with a chi-square test.
Fruit and seed set in three Ipomoea species in 1999, and four Ipomoea species in 2005, was analyzed with the effect of species, flower damage (yes, no) and the species--flower damage interaction as effect variables. For fruit set, we utilized a two-way logistic regression, with fruit maturation (yes, no) as the response variable. For seed set, we utilized an ordinal logistic regression, with seed number as the response variable. For both the logistic and ordinal regression we report likelihood ratio chi-squares ([chi square]). Throughout the manuscript we present the mean [+ or -] SE.
MAGNITUDE OF FLORIVORY
Florivory was ubiquitous in the three study years. Pooling across the three study years, we observed florivory intensity (percentage of flowers damaged) that ranged from a low of 0.96 [+ or -] 0.25% in Ipomoea hederifolia, to a high of 6.44 [+ or -] 0.91% in L cordatotriloba (Fig. 1). There were significant differences in florivory intensity among the Ipomoea species in 1999 and 2000, but not significantly so in 2005, following sequential Bonferroni correction (Table 1). Post-hoc analyses revealed that levels of florivory did not differ between the two pollination syndromes. However, the native L cordatotriloba suffered higher florivory rates compared to the three non-native focal Ipomoea (Table 2).
The degree of damage we observed was consistent among sites within years, except for 1999, where sites were significantly different for flower damage levels after Bonferroni correction (Table 2). Furthermore, the florivory intensity within each Ipomoea species was not affected by the number of flowers available in each site on a given day (Table 1).
The most frequently attacked floral structure was the corolla (Fig. 2), with over 60% of damaged flowers receiving damage that varied from patchy removal of corolla tissue to the loss of the entire corolla (pers. obs.). The pattern of damage to floral parts within each species was similar between 1999 and 2005. Overall, approximately 50% of damaged flowers had their reproductive organs damaged. Stamens and pistils, when damaged, were usually completely consumed by florivores. In both 1999 and 2005, the frequency of damage to the different floral structures varied among the four Ipomoea species (Fig. 3; 1999: [chi square] = 17.26, df = 6, P < 0.008; 2005: [chi square] = 18.54, df = 6, P < 0.005).
REPRODUCTIVE CONSEQUENCES OF FLORIVORY
Fruit set.--Overall, flower damage significantly reduced fruit set in the Ipomoea species (Fig. 3). In 1999, only 22% of damaged flowers produced fruit compared to 52% of undamaged, control flowers and this reduction was significant ([chi square] dam = 4.94, df = 1, P < 0.03). In 1999, with the native L cordatotriloba excluded from fruit set analysis (see Methods), there were significant differences in fruit set between the three remaining Ipomoea species in 1999 ([chi square]spp = 33.47, df = 2, P < 0.0001). A second analysis, collapsing the three non-native species into their two pollination syndromes, reveals no difference in fruit set between the bumble bee and two hummingbird-pollinated species ([chi square]syn = 0.05, df = 1, P > 0.82) after Bonferroni correction. Thus, differences in fruit set were not explained by pollination syndromes in 1999.
[FIGURE 1 OMITTED]
The species-by-treatment interaction was not significant in either year (1999: [([chi square]int = 1.72, df = 2, P > 0.42; 2005: [([chi square]int = 0.98, df = 3, P > 0.81), indicating that the Ipomoea species experienced similar reductions of fruit set when flowers were damaged.
In 2005, with all four focal Ipomoea species in the fruit set analysis, there were significant dirt>fences in fruit set between the species ([chi square]spp = 28.05, df = 3, P < 0.0001). The differences in fruit set among the Ipomoea species was not driven b x pollination syndrome ([chi square]syn = 2.71, df = 1, P > 0.10), but rather by origin. The native I. cordalotriloba had the lowest fruit set among the four focal species, 15.0 [+ or -] 3.1%, relative to the three non-native Ipomoea, 32.9 [+ or -] 2.6% ([chi square]oxi = 3.90. df = 1, P < 0.05), although this difference is only marginally significant following Bonferroni correction.
Because of the way we tagged flowers in 2005, we were able to determine whether the type of damage suffered by a flower (attractive vs. reproductive organs) influences subsequent reproductive success. Overall, percentage fruit set in undamaged flowers was 28.0% compared to 21.8% in flowers with damage only to the corollas, but this difference was not significant ([chi square]dmg = 0.04, df = 1, P > 0.83). However, not surprisingly, the consequences of florivory damage to the reproductive organs were severe, relative to undamaged flowers, with fruit set reduced to 8.3% ([chi square]dmg = 7.97, df = 1, P < 0.005), significant following sequential Bonferroni correction.
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
Seed set.--The pattern of flower damage on seed production was similar to that observed on fruit set. In 1999, florivory had a marginal effect on seed set ([chi square]trt = 3.39, df = 1, P > 0.06). There were significant differences between the species for fruit set ([chi square]spp = 25.93, df = 2, P < 0.0001), yet this variation in seed set was not explained by pollination syndromes ([chi square]syn = 0.35, df = 1, P > 0.55) among the three non-native Ipomoea.
In 2005, there were significant differences between the species for seed set ([chi square]spp = 22.34, df = 3, P < 0.0001). Again, differences in seed set were not driven by pollination syndrome ([chi square]syn = 2.95, df = 1, P < 0.09), but rather by origin. The native Ipomoea cordatotriloba yielded 3.2 [+ or -] 0.3 seeds per fruit, whereas the non-native Ipomoea average 2.6 [+ or -] 0.3 seeds per fruit ([chi square]ori = 3.90, df = 1, P < 0.05), although this difference is only marginally significant following Bonferroni correction.
There was no significant reduction in the average seeds per fruit between undamaged flowers (1.58 [+ or -] 0.14), and flowers experiencing damage only to the corolla (1.61 [+ or -] 0.38; [chi square]trt = 0.14, df = 1, P > 0.70). However, seed set was approximately four times greater in undamaged flowers (1.58 [+ or -] 0.14) compared to those flowers that received damage to their reproductive organs (0.42 [+ or -] 0.23): [chi square]trt = 8.23, df = 1, P < 0.005.
Flowers and floral characteristics in a variety of species have been shown to be the result of multiple, conflicting selection pressures from pollinators and antagonists such as florivores, nectar robbers and seed predators (Brody, 1997; Strauss, 1997; Galen, 1999; Irwin and Brody, 1999; Pilson, 2000; Galen and Cuba, 2001). Florivores may also drive the evolution of sexual specialization in plants (Ashman, 2000, 2002; Leege and Wolfe, 2002). Recently, Cariveau et al. (2004) revealed that some floral traits in Castilleja linariaefolia are under stronger selection pressure from pre-dispersal seed predators than pollinators. The influence of florivores perhaps is not surprising, given the close association between insects and flowering plants likely developed as an initially antagonistic relationship, due to the increased attractiveness of flowers as food items of early angiosperm flowers relative to gymnosperm strobili (Frame, 2003). Even though these plants likely suffered negative effects due to loss of reproductive tissues, the reliable movement of pollen by florivores, relative to wind pollination in gymnosperm strobili, likely led to the evolution of mutualistic pollination system (Frame, 2003).
The main finding of this study was that all four of the Ipomoea species experienced insect-mediated damage to their flowers, and florivory intensity was associated not with pollination syndrome as we first hypothesized, but florivory intensity varied with whether the species was native or introduced. The non-native, hummingbird-pollinated L hederifolia and I. quamoclit consistently had the lowest florivory intensity, while the native, bumble bee-pollinated L cordatotriloba experienced the greatest florivory intensity. In two of three study years, the non-native, bumble bee-pollinated L hederacea tended to have intermediate, albeit not significantly different, florivory intensities. Florivores not utilizing I. cordatotriloba may be more likely to host-shift onto L hederacea, due to similar floral traits (esp. flora pigments, see below). Florivory intensity for the four Ipomoea were fairly consistent among years, and were comparable to florivory intensities reported for other plant species (Ackerman and Montalvo, 1990; Bishop and Schemske, 1998; Breadmore and Kirk, 1998; Malo et al., 2001; Leege and Wolfe, 2002). Our results support the notion that the non-native Ipomoea are experiencing lower floral herbivory and that this is due to either to higher resistance to florivores, or that florivores prefer the native L cordatotriloba for intrinsic reasons not known.
Florivory will clearly have direct effects on fitness via male and female function, with the most obvious impact being the loss of gametes (Louda, 1982; Bertness and Shumway, 1992; Louda and Potvin, 1995; Lohman et al., 1996; Juenger and Bergelson, 1997; Krupnick and Weis, 1999; Maron et al., 2002). In our study, approximately 50% of all damaged Ipomoea flowers experienced the loss of the pistil and stamens, and this had a direct impact on female reproductive success. In 2005, few of the flowers receiving damage to their reproductive organs produced any fruit. Florivory may negatively impact male fitness through loss of pollen-bearing stamens. Plants with damaged flowers should experience reduced male reproductive success via decreased pollen export (Krupnick and Weis, 1999; Mothershead and Marquis, 2000). Whether florivory causes pollen limitation, though, at the community level in our Ipomoea sites is unknown (e.g., Ashman et al., 2004), though, florivory has been shown to induce pollen limitation in other studies (Bertness and Shumway, 1992; Cunningham, 1995; Krupnick and Weis, 1999).
Interactions with enemies that consume flowers or render flowers dysfunctional (e.g., disease-Alexander and Antonovics, 1988; galls-Wolfe, 1997; Wolfe and Rissler, 1999) can impact plant reproductive success even before pollination can occur. The most obvious influence of floral damage may be via its indirect effect on pollinator-mediated attraction. Studies on other plant species have found reduced pollinator visitation rates when flowers appear damaged (Johnson et al., 1995; Lohman et al., 1996; Lehtila and Strauss, 1999; Krupnick et al., 1999; Mothershead and Marquis, 2000; Malo et al., 2001; Canela and Sazima, 2003). This can influence male fitness if pollen removal rates are reduced and female fitness if pollen deposition rates are reduced. In 2005, female reproductive success (fruit set) was depressed by approximately 30% in flowers that had only their corolla damaged, but organs left intact, suggesting that removal of attractive structures indirectly affected fruit set. Yet, decreased attraction and reduced pollen transfer in these Ipomoea will likely not result in complete reproductive failure since the four species are self-compatible (Ennos, 1981; Stucky, 1985; Murcia, 1990), and can set fruit in the absence of pollinators (Wolfe and Sowell, pers. obs.). In these Ipomoea, autogamous pollen grains may be deposited on the stigma before or shortly after anthesis begins. Since the florivores are rarely active before anthesis (Sowell and Wolfe, pers. obs.), damage by florivores may not occur until pollen tubes have already passed down the style. Thus, fruit set could still occur via selfing before the onset of floral damage. The mean number of seeds per fruit (seed set) was also not affected by damage to the corolla, but was reduced when reproductive organs were damaged by florivores.
Three of the four Ipomoea examined in this study are non-native and considered invasive in the southeast U.S. (USDA, 2001). An often-cited explanation for the success of invasiveness in the introduced range is the Enemy Release Hypothesis (Keane and Crawley, 2002) which proposes that colonization in the introduced range usually results in the escape from enemies in the native range, allowing the introduced species to exhibit enhanced growth and reproductive output, resulting in population expansion. We would therefore expect enemy attack to be lower in non-native species relative to either native congeners or the species in their native range (Maron and Vila, 2002; Wolfe, 2002; Torchin et al., 2003; Agrawal et al., 2005). Although the current study was not designed as a direct test of the Enemy Release hypothesis, it is interesting that our results were consistent with expectation, since the florivory rate was greatest on the native L cordatotriloba compared to the non-native Ipomoea.
It is important to point out a limitation of a comparative study such as ours that examines multiple species. It is entirely possible that the similar patterns of florivory (or any measurement taken on a species) between Ipomoea hederifolia and I. quamoclit is a result of the fact that they are closely related and placed in the same subgenus (Miller et al., 1999), and not due to the fact that they display the same pollination syndrome. Thus, the ideal study would be one that utilizes replicate Ipomoea: (1) within subgenera that exhibit different pollination syndromes, or (2) that exhibit the same pollination syndrome in different subgenera. Similarly, we must be cautious to conclude that the uniqueness of L cordatotriloba's greater florivory intensity is due to its being a native species since there was no replication at the species level for native species while we did have multiple (3) introduced Ipomoea species.
Why then, should floral herbivores exhibit a preference for the bumble bee-pollinated Ipomoea cordatotriloba, relative to the hummingbird-pollinated Ipomoea species? It is possible that the pattern of florivore choice is the result of a pleiotropic effect. In Ipomoea, the production of specific plant secondary defensive compounds and specific classes of floral pigments are not independent. Anthocyanins and secondary flavonoids are derived from the same precursor compounds in a common biosynthetic pathway in Ipomoea purpurea, a purple bumble bee-pollinated morning-glory (Simms and Bucher, 1996; Fineblum and Rausher, 1997). Species with reduced anthocyanin pigmentation in their flowers, such as L hederifolia and L quamoclit, may devote more precursor favonoids to secondary defensive compounds throughout the plant (Fineblum and Rausher, 1997), and consequently, they may be better defended against foliar herbivory and florivory. Recent work on Nemophila menziesii has also shown that damage to flowers can induce increased resistance to florivory later in the season (McCall, 2006). Future work should determine if the orange and red-flowered, hummingbird-pollinated L hederifolia and I. quamoclit are better defended via secondary flavonoids, relative to the blue and purple-flowered, bumble bee-pollinated Ipomoea, and if the pattern of florivory that we observed is in part the result of a difference in secondary flavonoids among the four Ipomoea species.
Acknowledgments.--The authors thank C. O'Neil, B. Paige, B. Penna and IL Wollett for help conducting fieldwork; M. McAloon for identification of lepidopteran larvae; C. Barr, A. Blair, S. Freedberg, S. Keller, M. Neiman, V. Panjeti, D. Sloan, D. Taylor and two anonymous reviewers for valuable comments that improved the manuscript. This work was supported by grants from the National Science Foundation (DEB4)349553) (LMW) and Georgia Southern University (Graduate College [DRS]; College of Science and Technology Academic Excellence [DRS]; Irene Burr Boole Scholarship [DRS]; Catalyst Grant [LMW]).
SUBMITTED 7 JULY 2008 ACCEPTED 21 JANUARY 2009
ACKERMAN, J. D. AND A. M. MONTALVO. 1990. Short- and long-term limitations to fruit production in a tropical orchid. Ecology, 71:263-272.
AGRAWAL, A. A., P. M. KOTANEN, C. E. MITCHELL, A. G. POWER, W. GODSOE AND J. KLIRONOMOS. 2005. Enemy release? An experiment with congeneric plant pairs and diverse above-and belowground enemies. Ecology, 86:2979-2989.
ALEXANDER, H.. AND J. ANTONOVICS. 1988. Disease spread and population dynamics of anther-smut infection of Silene alba caused by the fungus Ustilago violacea. J. Ecol., 76:91-104.
ARMBRUSTER, W. S. 1997. Exaptations link evolution of plant-herbivore and plant-pollinator interactions-a phylogenetic inquiry. Ecology, 78:1661-1672.
ARMSTRONG J. E. AND D. MARSH. 1997. Floral herbivory, floral phenology, visitation rate, and fruit set in Anaxagorea crassipetala (Annonaceae), a lowland rain forest tree of Costa Rica. J. Torrey Bot. Soc., 124:228-235.
ASHMAN, T-L. 2000. Pollinator selectivity and its implications for the evolution of dioecy and sexual dimorphism. Ecology, 81:2577-2591.
--. 2002. The role of herbivores in the evolution of separate sexes from hermaphroditism. Ecology, 83:1175-1184.
--, T. M. KNIGHT, J. A. STEETS, P. AMARASEAARE, M. BURD, D. R. CAMPBELL, M. R. DUDASH, M. O. JOHNSTON, S.J. MAZER, R.J. MITCHELL, M. T. MORGAN AND W. G. WILSON. 2004. Pollen limitation of plant reproduction: ecological and evolutionary consequences. Ecology, 85:2408-2421.
AUSTIN, D. F. AND Z. HUAMAN. 1996. A synopsis of Ipomoea (Convolvulaceae) in the Americas. Taxon, 45:3-38.
BERTNESS, M. D. AND S. W. SHUMWAY. 1992. Consumer driven pollen limitation of seed production in marsh grasses. Am. J. Bot., 79:288-293.
BISHOP, J. G. AND D. W. SCHEMSKE. 1998. Variation in flowering phenology and its consequences for lupines colonizing Mount St. Helens. Ecology, 79:534-546.
BREADMORE, K. N. AND W. D.J. KIRK. Factors affecting floral herbivory in a limestone grassland. Acta Oecol., 19:501-506.
BREEDLOVE, D. E. AND P. R. EHRLICH. 1968. Plant-herbivore coevolution: lupines and lycaenids. Science, 162:671-672.
BRODY, A. IL 1997. Effects of pollinators, herbivores, and seed predators on flowering phenology. Ecology, 78:1624-1631.
CAMPBELL, D. R., N. W. WASER AND M. V. PRICE. 1996. Mechanisms of hummingbird-mediated selection for flower width in Ipomopsis aggregata. Ecology, 77:1463-1472.
CANELA, M. B. F. AND M. SAZIMA. 2003. Florivory by the crab Armases angustipes (Grapsidae) influences hummingbird visits to Aechmea pectinata (Bromeliaceae). Biotropica, 35:289-294.
CARIVEAU, D., R. E. IRWIN, A. IC BRODY, L. S. GARCIA-MAYEYA AND A. VON dER OHE. 2004. Direct and indirect effects of pollinators and seed predators to selection on plant and floral traits. Oikos, 104:15-26.
CONNER, J. IL AND S. RUSH. 1996. Effects of flower size and number on pollinator visitation to wild radish, Raphanus raphanistrum. Oecologia, 105:509-516.
CRAWLEY, M.J. 1987. The population biology of invaders. Phil. Trans. R. Soc. Lond. B., 314:711-731.
CUNNINGHAM, S. A. 1995. Ecological constraints on fruit initiation by Calyptrogyne ghiesbreghtiana (Arecaceae): floral herbivory, pollen availability, and visitation by pollinating bats. Am. J. Bot., 82:1527-1536.
ELTON, C. S. 1958. The ecology of invasions by animals and plants. The University of Chicago Press, Chicago, Illinois.
ENNOS, R. 1981. Quantitative studies of the mating system in two sympatric species of Ipomoea (Convolvulaceae). Genetica, 57:93-98.
FAEGRI, K. AND L. VAN DER PIJL. 1979. The principals of pollination ecology, 3rd edition. Pergamon Press, Oxford New York Toronto Sydney Paris Frankfurt.
FENSTER, C. B., W. S. ARMBRUSTER, P. WILSON, M. R. DUDASH AND J. THOMSON. 2004. Pollination syndromes and floral specialization. Annu. Ecol. Evol. Syst., 35:375-403.
FINEBLUM, W. L. AND M. D. RAUSHER. 1997. Do floral pigmentation genes also influence resistance to enemies? The W locus in Ipomoea purpurea. Ecology, 78:1646-1654.
FRAME, D. 2003. Generalist flowers, biodiversity and florivory: implications for angiosperm origins. Taxon, 52:681-685.
GALEN, C. 1996. Rates of floral evolution: adaptation to bumblebee pollination in an alpine wildflower, Polemonium viscosum. Evolution, 50:120-125.
--. 1999. Flowers and enemies: predation by nectar-thieving ants in relation to variation in floral form of an alpine wildflower, Polemonium viscosum. Oikos, 85:426-434.
--AND J. CUBA. 2001. Down the tube: pollinators, predators, and the evolution of flower shape in the alpine skypilot, Polemonium viscosum. Evolution, 55:1963-1971.
GODFREY, R. IL AND J. w. WOOTEN. 1981. Aquatic and wetland plants of southeastern United States: Dicotyledons. The University of Georgia Press, Athens, Georgia, USA.
HARBER, L. D. 1990. Behavioral responses by bumble bees to variation in pollen availability. Oecologia, 85:41-47.
HELD, D. W. AND D. A. POTTER. 2004. Floral affinity and benefits of dietary mixing with flowers for a polyphagous scarab, Popillia japonica Newman. Oecologia, 140:312-320.
HERRERA, C. M. 1993. Selection on floral morphology and environmental determinants of fecundity in a hawk moth-pollinated violet. Ecol. Monogr., 63:251-275.
HODGES, S. A. 1995. The influence of nectar production on hawkmoth behavior, self pollination, and seed production in Mirabilis multiflora (Nyctaginaceae). Am. J. Bot., 82:197-204.
IRWIN, R. E. AND A. K. BRODV. 1998. Nectar robbing in Ipomopsis aggregata: effects on pollinator behavior and plant fitness. Oecologia, 116:519-527.
--AND--. 1999. Nectar-robbing bumble bees reduce the fitness of Ipomopsis aggregata (Polemoniaceae). Ecology, 80:1703-1712.
--, L. S. ADLER arm A. K. BRODV. 2004. The dual role of floral traits: pollinator attraction and plant defense. Ecology, 85:1503-1511.
JOHNSON, S. D. AND A. DAFNI. 1998. Response of bee-flies to the shape and pattern of model flowers: implications for floral evolution in a Mediterranean herb. Funct. Ecol., 12:289-297.
JOHNSON, S. G., L. F. DELPH AND C. L. ELDERKIN. 1995. The effect of petal-size manipulation on pollen removal, seed set, and insect-visitor behavior in Campanula americana. Oecologia, 102:174-179.
JUENGER, T. AND J. BrRGEtSON. 1997. Pollen and resource limitation of compensation to herbivory in scarlet gilia, Ipomopsis aggregata. Ecology, 78:1684-1695.
KAWAGOE, T. AND N. SUZUKI 2004. Cryptic dioecy in Actinida polygama: a test of the pollinator attraction hypothesis. Can. J. Bot., 82:214-218.
KEANE, R. M. AND J. M. CRAWLEY. 2002. Exotic plant invasions and the enemy release hypothesis. Trends Ecol. Evol., 17:164-170.
KRUPNICK, G. A. AND A. E. WEIS. 1999. The effect of floral herbivory on male and female reproductive success in Isomeris arborea. Ecology, 80:135-149.
--. AND D. R. CAMPBELL. 1999. The consequences of floral herbivory for pollinator service to Isomeris arborea. Ecology, 80:125-134.
LEEGE, L. M. AND L. M. WOLFE. 2002. Do floral herbivores respond to variation in flower characteristics in Gelsemium sempervirens (Loganiaceae), a distylous vine? Am. J. Bot., 89:1270-1274.
LEHTILA, K. AND S. Y. STRAUSS. 1999. Effects of floral herbivory on male and female reproductive traits of wild radish, Raphanus raphanistrum. Ecology, 80:116-124.
LOHMAN, D.J., A. R. ZANGERL AND M. R. BERENBAUM. 1996. Impact of floral herbivory by parsnip webworm (Oecophoridae): Depressaria pastinacella Duponchel on pollination and fitness of wild parsnip (Apiaceae: Pastinaca sativa L.). Am. Midl. Nat., 136:407-412.
LOUDA, S. M. 1982. Limitation of the recruitment of the shrub Haplopappus squarrosus (Asteraceae) by flower- and seed-feeding insects. J. Ecol., 70:43-53.
--AND M. A. POTWN. 1995. Effect of inflorescence-feeding insects on the demography and lifetime fitness of a native plant. Ecology, 76:229-245.
MALO, J. E., J. LEIRANA-ALCOCER AND V. PARRA-TABLA. 2001. Population fragmentation, florivory, and the effects of flower morphology alterations on the pollination success of Myrmecophila tibicinis (Orchidaceae). Biotropica, 33:529-534.
MARON, J. L., J. K. COMBS AND S. M. LOUDA. 2002. Convergent demographic effects of insect attack on related thistles in coastal vs. continental dunes. Ecology, 83:3382-3392.
--AND M. VILA. 2001. When do herbivores affect plant invasions? Evidence for the natural enemies and biotic resistance hypotheses. Oikos, 95:361-373.
MATTER, S. F., J. B. LANDRY, A. M. GRECO AND C. D. LACOURSE. 1999. Importance of floral phenology and florivory for Tetraopes tetraopthalmus (Coleoptera: Cerambycidae): tests at the population and individual level. Environ. Entomol., 28:1044-1051.
MCCALL, A. C. 2006. Natural and artificial floral damage induces resistance in Nemophila menziesii (Hydrophyllaceae) flowers. Oikos, 112:660-666.
--AND R. E. IRWIN. 2006. Florivory: the intersection of pollination and herbivory. Ecol. Lett., 9:1351-1365.
MILLER, R. E., M. D. RAUSHER AND P. S. MANOS. 1999. Phylogenetic systematics of Ipomoea (Convolulaceae) Based on ITS and waxy sequences. Syst. Bot., 24:209-227.
MOTHERSHEAD, K. AND R.J. MARQUIS. 2000. Fitness impacts of herbivory through indirect effects on plant-pollinator interactions in Oenothera macrocarpa, Ecology, 81:30-40.
MURCIA, C. 1990. Effect of floral morphology and temperature on pollen receipt and removal in Ipomoea trichocarpa. Ecology, 71:1098-1109.
NILSSON, L. A. 1988. The evolution of flowers with deep corolla tubes. Nature, 334:147-149. PILSON, D. 2000. Herbivory and natural selection on flowering phenology in wild sunflower, Helianthus annuus. Oecologia, 122:72-82.
POHL, N., G. CARVALLO, C. BoTTO-MAHAN AND R. MEDEL. 2006. Nonadditive effects of flower damage and hummingbird pollination on the fecundity of Mimulus luteus. Oecologia, 149:648-655.
RICE, W. R. 1989. Analyzing tables of statistical tests. Evolution, 43:223-225.
SAS. 1997. JMP, version 3.2.1. SAS Institute, Cary, North Carolina, USA.
SCHEMSKE, D. W. AND C. C. HORVITZ. 1984. Variation among floral visitors in pollination ability: a precondition for mutualism specialization. Science, 225:519-521.
SIMMS, E. L. AND M. A. BUCHER. 1996. Pleiotropic effects of flower-color intensity on herbivore performance on Ipomoea purpurea. Evolution, 50:957-963.
SOKAL, R. R. AND F.J. ROHLF. 1995. Biometry, third edition. W. H. Freeman and Company, New York, New York, USA.
STRAUSS, S. Y. 1997. Floral characters link herbivores, pollinators, and plant fitness. Ecology, 78:1640-1645.
STUCKY, J. M. 1985. Pollination systems of sympatric Ipomoea hederacea and L purpurea and the significance of interspecific pollen flow. Am. J. Bot., 72:32-43.
TORCHIN, M. E., K. D. LAFFERTY, A. P. DOBSON, V. J. MCKENZIE AND A. M. KURIS. 2003. Introduced species and their missing parasites. Nature, 421:628-630.
UNITED STATES DEPARTMENT OF AGRICULTURE, NATIONAL RESOURCES CONSERVATION SERVICE. 2001. The PLANTS Database, Version 3.1 [online], National Plant Data Center, Baton Rouge, Louisiana, USA [online]. Available from http://plants.usda.gov [accessed 6 Jun. 2008].
WASER, N. M. 1978. Interspecific pollen transfer and competition between co-occurring plant species. Oecologia, 36:223-236.
--AND M. V. PRICE. 1981. Pollinator choice and stabilizing selection for flower color in Delphinium nelsonii. Evolution, 35:376-390.
WOLFE, L. M. 1997. Differential flower herbivory and gall formation on males and females of Neea psychotrioides, a dioecious tree. Biotropica, 29:169-174.
--. 2002. Why alien invaders succeed: support for the escape-from-enemy hypothesis. Am. Nat., 166:705-711.
--AND J. KRSTOLIC. 1999. Floral symmetry and its influence on variance in flower size. Am. Nat., 154:484-488.
--AND L. J. RISSLER. 1999. Reproductive consequences of the gall-inducing fungal pathogen (Exobasidium vaccini) on flame azalea (Rhododendron calendulaceum). Can. J. Bot., 77:1454-1459.
--AND D. R. SOWELL. 2006. DO pollination syndromes partition the pollinator community? A test using four sympatric morning glory species. Int. J. Plant Sci., 167:1169-1175.
DEXTER R. SOWELL (1) AND LORNE M. WOLFE
Department of Biology, Georgia Southern University, P.O. Box 8042, Statesboro 30460
(1) Corresponding author present address: Department of Biology, University of Virginia, P.O. Box 400328, Charlottesville 22904; email: email@example.com; Telephone: (434) 982-5218; FAX: (434) 982-5628
Table 1.--Number of flowers surveyed across Ipomoea species and study years Species 1999 2000 2005 Species Total I. hederifolia 17,718 2375 805 20,898 I. quamoclit 12,439 862 1362 14,663 I. hederacea 2493 233 142 2868 I. cordatotriloba 6143 2769 648 9560 Year total 38,793 6239 2957 47,989 Table 2.--The analysis of florivory intensity on four Ipomoea species in southeastern Georgia. Two-way ANCOVA of ranks (Scheirer-Ray-Hare extension of the Kruskal-Wallis test) testing the effects of species, site, species by site interaction, and daily number of flowers available within species as a covariate, on the proportion of flowers damaged daily by florivores. Post-hoc analyses with species pooled by their origin (native, introduced) and pollination syndrome (hummingbird-, bumble bee-pollinated) follow below the species entry. For each year, P values significant following a sequential Bonferroni correction are bolded Sources df SS H P< 1999 Species 3 112,128.30 63.35 <0.0001# Origin 1 46,965.85 26.53 <0.0001# Pollination syndrome 1 394.87 0.22 0.64 Site 2 17,326.65 9.79 0.01# Species X site 6 45,356.30 25.62 <0.0002# Flower number 1 2776.51 1.57 0.21 Error 126 52,467.42 2000 Species 3 50,935.17 45.01 <0.0001# Origin 1 39,612.82 35.01 <0.0001# Pollination syndrome 1 1684.74 1.49 0.23 Site 2 6227.36 5.50 0.07 Species X site 6 6952.86 6.14 0.41 Flower number 1 1523.04 1.35 0.25 Error 100 52,735.04 2005 Species 3 984.40 4.62 0.03 Origin 1 906.81 4.26 0.04 Pollination syndrome 1 216.77 1.02 0.32 Site 1 46.79 0.22 0.64 Species X site 3 148.69 0.70 0.87 Flower number 1 555.55 2.61 0.11 Error 38 7606.80 Note: For each year, P values significant following a sequential Bonferroni correction are bolded is indicated with #.
|Printer friendly Cite/link Email Feedback|
|Author:||Sowell, Dexter R.; Wolfe, Lorne M.|
|Publication:||The American Midland Naturalist|
|Date:||Jan 1, 2010|
|Previous Article:||Allelopathic effects of goldenrod species on turnover in successional communities.|
|Next Article:||Variation in endophyte symbiosis, herbivory and drought tolerance of ammophila breviligulata populations in the Great Lakes region.|