xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
leader 00000nas 2200000Ka 4500
controlfield tag 008 000000c19749999pautr p s 0 0eng d
datafield ind1 8 ind2 024
subfield code a M39-00137
Dos polinizadores de Ficus tuerckheimii (Moraceae): la competencia del parasitismo y la competencia entre especies dentro de un mutualismo obligado
Two Agaonid pollinators of Ficus tuerckheimii (Moraceae): parasitism and interspecific competition within an obligate mutualism
The pollinating and non-pollinating wasp fauna of the strangler fig, Ficus tuerckheimii, provide a unique suite of biological relationships in which to observe potential species interactions. Nearly all of the 750 fig species worldwide are pollinated by a single, species-specific agaonid wasp. Pollinated by two congeneric
species of fig wasps, Ficus tuerckheimii is a rare exception. I investigated the relationship between the two foundress pollinators as well as the cohort of pollinating and non-pollinating (interloper) wasps that emerge from fig syconia on one tree in Monteverde, Costa Rica. The pollinating species appeared to colonize
syconia randomly (R2 = 0.0126, F = 1.234, df = 1.97, p = 0.269). Although interloper wasps were significantly less abundant than pollinating species, there was no association between interlopers and either
pollinator or fig seed production (ANOVA, F = 15.568, p < 0.001; LSD, p < 0.001). This study provides evidence to suggest that the overall strength of species interactions in the fig-pollinator-interloper system may not be as strong as ecological models suggest.
Las avispas polinizadoras y no polinizadoras de la especie de higuern Ficus tuerckheimii ofrecen una oportunidad nica para observar las interacciones entre especies. Casi todas las 750 especies de higuerones en el mundo estn polinizadas por la especie de avispa de la familia Agaonidae. Ficus tuerckheimii es una de las excepciones raras de especies de higuerones, porque es polinizada por dos especies de avispas congnitas. Yo investigue la relacin entre las dos polinizadoras y el grupo de avispas no polinizadoras que salen de sicionios de un individuo de F. tuerckheimii en Monteverde, Costa Rica.
Text in English.
Tropical Ecology 2007
Ecologa Tropical 2007
t Monteverde Institute : Tropical Ecology
Two Agaonid pollinators of Ficus tuerckheimii (Moraceae): parasitism and interspecific competition within an obligate mutualism Marissa Jones Department of Biology, University of Puget Sound ABSTRACT The pollinating and nonpollinating wasp fauna of the strangler fig, Ficus tuerckheimii provide a unique suite of biological relationships in which to observe potential species interactions. Nearly all of the 750 fig species worldwide are pollinated by a single, species specific agaonid wasp. Pollinated by t wo congeneric species of fig wasps, Ficus tuerckheimii is a rare exception. I investigated the relationship between the two foundress pollinators as well as the cohort of pollinating and nonpollinating (interloper) wasps that emerge from fig syconia on one tree in Monteverde, Costa Rica. The pollinating species appeared to colonize syconia randomly (R 2 = 0.0126, F = 1.234, df = 1.97, p = 0.269). Although interloper wasps were significantly less abundant than pollinating species, there was no association bet ween interlopers and either pollinator or fig seed production (ANOVA, F = 15.568, p < 0.001; LSD, p < 0.001). This study provides evidence to suggest that the overall strength of species interactions in the fig pollinator interloper system may not be as st rong as ecological models suggest. RESUMEN Las avispsas polinizadoras y no polinizadoras de la especie de higuer Â— n Ficus tuerckheimii ofrecen una oportunidad Âœ nica para observar las interacciones entre especies. Casis todas las 750 especies de higuerones en el mundo est Â‡ n polinizados por su propio, avispa especie especifica de la familia Agaonidae. Ficus tuerckhiemii es una de las excepciones raras de especies de higuerones, porque es polinizado por dos especies de avispa congÂŽnicas. Yo investigu ÂŽ la relac i Â— n entre las dos polinizadoras y el grupo de avispas no polinizadoras que salen de siconios de un individuo de F. tuerckheimii en Monteverde, Costa Rica. Las polinizadoras aparecieron colonizer siconios al azar (R 2 = 0.0126, F = 1.234, df = 1.97, p = 0.26 9). Aunque las avispas no polinizadoras fueron menos abundante que las que polinizan, no hubo asociaci Â— n entre las avispas que no polinizan, tambien ni entre polinizadoras ni producci Â— n de semillas (ANOVA, F = 15.568, p < 0.001; LSD, p > 0.001). Este estu dio provee informaci Â— n que sugiere que las interacciones entre especies no son tan fuerte como los mod ÂŽ los ecol Â— gicos normalmente predicen. INTRODUCTION The principle of competitive exclusion states that competitive coexistence of two species in the same habitat requires some form of ecological differentiation that allows the species to avoid direct competition and distinguish between conspecifics (members of the same species) and erospecifics (members of another species) (Chesson 1991 in Kerdelhue et al. 1997). Niche partitioning can occur physiologically, behaviorally, temporally, spatially, or geographically (Bronstein 1989, Kerdelhue et al. 1997, Michaloud et al. 1996 White 1997). The obligate mutualism between Ficus (Moraceae) and fig wasps (Hymenopter a, Chalcidiod: Agaonidae) is an ideal model system in which to study niche partitioning and possible mechanisms of speciation (Michaloud et al 1996, Bronstein 1988a). Nearly all of the 750 species worldwide are pollinated by a unique and coevolved
species of fig wasp, with only a few studied exceptions documented in the literature (e.g. Michaloud et al. 1996, Ramirez 1971). Figs and their agaonid mutualists share morphological, chemical, and physiological co adapted characteristics that work in concert to e xecute a unique system of pollination and reproduction (Michaloud et al. 1996). Fig inflorescences, or syconia, are comprised of hundreds of florets that line an enclosed cavity (Bronstein 1988b). The syconium is penetrated female wasps that are chemically attracted to volatile compounds released by their respective fig mutualist partners (Grison Pige et al. 2002). A female, pollen laden wasp enters the syconium through the ostiole in the apex of the fig, during which process she is stripped of her wings, a ntennae, and presumably many foreign contaminants. The wasp oviposits down the style of the florets and onto the fig ovary, where her offspring develop by feeding on the fig ovule. The florets, however, have styles of varying lengths so that a portion of t he ovaries are not accessible to pollinating wasps and remain intact for seed production. During the process of oviposition, the female actively delivers pollen to receptive stigma and dies shortly thereafter. Over the next month, both the fig seeds and wa sps develop within the syconium. Wingless male wasp progeny emerge from the fig ovules, locate females within the florets, and copulate with them. Males help the females exit the syconium by chewing an exit hole though the outer wall. Upon leaving the fig, the young females actively fill their pollen pockets and fly off in search of a new fig in which to deposit their eggs (Janzen 1983). Information about mutualisms, niche partitioning, and speciation can be gained through investigations of exceptions to th e one wasp one Ficus relationship (Michaloud et al. 2006). The strangler fig, Ficus tuerckheimii is a large canopy tree that is common in pastures and forests at 1300 1550 m in Monteverde, Costa Rica (Haber et al. 1996). In contrast to most other specie s in this genus, F. tuerckheimii is pollinated by two congeneric fig wasps, Pegoscapus carlosi and P. mariae (Chalcidiod: Agaonidae) (Ram Â’ rez 1970). These wasps exploit fruits of the same size and occur sympatrically, making them an ideal model system in w hich to observe mechanisms of interspecific competition and niche partitioning. Figs and fig wasps have received much commendation as examples of a coevolved system of pollination and reproduction, yet figs are also host to a rich assortment of nonpollina ting or interloper wasps whose offspring develop within the syconium without pollinating the florets (Bronstein 1991). In addition to the interactions between pollinating wasps of F. tuerckheimii more information can be gained about the potential interact ions between the congerneric pollinators by investigating the reproductive output of pollinating fig wasps, nonpollinating fig wasps and fig seed production. Bronstein (1991) censused the nonpollinating wasp fauna of Ficus pertusa another species of stran gler fig in Monteverde, Costa Rica, and found that interloper wasps discriminate between syconia and preferentially oviposit in those that had been pollinated. She also suggested that interlopers may not have a strong negative effect on either figs or poll inators. Ficus abort the majority of unpollinated syconia so interloper wasps must colonize a syconium and develop synchronously with pollinating species in order to ensure that the tree matures the fruit (Bronstein 1988a). These wasps presumably respond t o the same chemical cues that are released by the fig and utilized by the pollinators. It is unknown whether the pollinators are found synchronously and it is
therefore possible that there may be differences in the interlopers associated with the two polli nators of F. tuerckheimii Ficus sur in Africa is another example of a fig that is pollinated by two species of fig wasp. Kerdelhu ÂŽ et al. (1996) found that Certatolsen silvestrianus and C. flabellatus, the specific pollinators of this species, exploit fru its of approximately the same size and show no avoidance of fruits pollinated by the congener, suggesting that they are not competitively exclusive. The researchers suggest that niche partitioning may still occur by utilizing different lengths of flowers, a fraction of the available flowers, or over a large spatial or geographical distance. Nevertheless, species interaction models predict that these species cannot coexist without some form of niche differentiation (Begon et al. 1990). Investigating F. tuer ckheimii in Monteverde, Costa Rica, White (1997) set traps in trees and found Pegoscarpus mariae was nearly five times more abundant than P. carlosi and that both species of wasp were caught during the day and almost none were caught at night. This suggest s that temporal separation is not the primary mechanism of separation between these two species. Yule (1999) found that foundress abundance varied between trees at different elevations, with no discernable pattern, indicating that P. carlosi and P. mariae do not specialize on trees based on geographic location or altitudinal differences in habitat. Yule proposed that P. carlosi and P. mariae were actually two morphs and not distinct species. In the present study, I investigate competitive interactions betwe en two agaonid pollinators on one F. tuerckheimii individual in order to determine whether P. carlosi and P. mariae partition syconia, colonize randomly, or have a negative affect on the relative abundance of the congener within a syconium. I also observe the outcome of the syconia in regard to Ficus seed production, inhabitation by interloper wasps, and number of wasp progeny. It is possible that P. carlosi and P. mariae avoid syconia pollinated by erospecifics and that there will be one competitively domi nant species. I predict that syconia that are more densely occupied by pollinating wasps of either species will also be more exploited by parasites, as parasites rely on pollinated syconia for defelopment. It is also possible that one species of pollinat or may be more associated with non pollinating wasps. MATERIALS AND METHODS Background Developmental phases of monoecious figs have been divided into five phases: pre female, female, interfloral, male, and post floral (Galil and Eisikowitch 1968 in Bron stein 1989) (Table 1). In this study, I sample figs in the female (B) and male (D) stages, which can be identified by the size and color of the syconia (Kerdelhue et al. 1997). In Ficus tuerckheimii syconia in reproductive phase B are green and firm. Syco nia in reproductive phase D are slightly fleshy and yellowish in color. Phase A pre female Young syconium prior to the opening of the ostiole Phase B female Ostiolar scales loosen, female flowers ripen, pollinators penetrate into the syconium and ovipos it into the ovaries. Phase C interfloral Wasp larvae and fig embryos develop within their respective ovaries Phase D male Male flowers mature, fertilized female wasps leave the syconia via tunnels bored by the males. Phase E post floral Both the syconia and the seeds inside them ripen.
Table 1. Developmental phases of monoecious figs as defined by Galil and Eisikowitch 1968 in Bronstein 1989 Sampling Site Ficus tuerckheimii is a large canopy tree (20 40 m) that is common in pastures and forests at 1300 1550m in the Premontane Moist Lifezone (Haber et al. 2000, Holdridge 1941). Fruiting is sporadic within a population, but synchronous within a tree (Janzen 1983). I identified one Ficus tuerckheimii individual that contained fruit in both reproductiv e phase B and D at Hotel Belmar in Monteverde, Costa Rica. The tree was located in an open pasture adjacent to secondary forest. I collected fruit in April and May of 2007 by hand or using pole clippers. Relative Abundance of Foundress Wasps I collected 99 syconia from one F. tuerckheimii individual in reproductive phase B (female), when the syconia were receptive to foundress wasps, but the foundresses had not yet started to decompose within the inflorescence. Fruit were collected based on the abundance of receptive syconia and the accessibility of inflorescence bearing branches. I measured the diameter of the syconia by averaging the length, width and height, which were measured to the nearest millimeter using calipers. The height was not recorded for th e first 13 syconia and so the average height of the others was substituted to calculate the average value. I dissected the syconia under a dissecting microscope, counted the number of foundress wasps within each syconium, and identified the wasps to speci es. Pegoscapus carlosi has black legs and a black abdomen, making it easy to distinguish from P. mariae which has translucent, yellow legs and a yellow and black abdomen (Yule 1999). I also noted the presence of other structures, such as white, gall like masses and red sticky matter, within the syconia. I used a t test to compare the abundance of foundress wasps of each species and a correlation analysis to determine the effect P. carlosi had on P. mariae Wasp progeny I collected 25 fruit in reproductive phase D by climbing to accessible parts of the tree. Studying fruits of this age required collection by hand in order to verify that fruit were at the appropriate reproductive phase. Syconia become moderately swollen and light yellow when the wasp progeny are ready to emerge. Exit holes created by males are visible in the syconia if the female wasps have already emerged. I therefore selected fruit that were yellowish, but did not contain exit holes. I again recorded the average diameter of each syconium. T o verify that fruit contained wasps of the proper age and to induce synchronous emergence, I cut syconia in half before placing them in glass emergence chambers sealed with cotton. Wasps that were ready to emerge began to do so immediately following the di sturbance of the syconium interior. I allowed wasps to emerge for 48 hours. Specimens were killed with acetone and identified to species. Data were analyzed using a correlation analysis. Voucher specimens are stored at the Estaci Â— n Biol Â— gica Monteverde. I sectioned syconia into quarters and recorded the number of seeds within each quarter to estimate the total number of seeds within the entire fruit. The average error inherent in this method was 20% based on the comparison of observed and estimated number of seeds from 5 fruits. I also recorded the number of wasps remaining within the syconium after the 48 hour emergence period by dissecting the fig ovules. The number of
wasps encountered in one quarter of the fruit was not a good proxy for the number of wa sps remaining in the entire fruit so these values were not ncluded in statistical analyses. I compared the number of seeds to the average syconium and the number of wasp progeny using a correlation analysis. Nonpollinating wasp species were identified to g enus based on the following characteristics: Idarnes sp. is similar in size and color to P. carlosi but is distinguished by a straight ovipositor that is three times its own body length (Bronstein 1991). Torymus sp. is also similar in body size to Pegosca pus but the ovipositor is thicker, curved, and intermediate in length between Pegoscapus and Idarnes. Hansonita sp. has a metallic green body is considerably larger than any of the other interloper or pollinator species. The ovipositor to body length rati o is similar to that of Torymus Species were identified to genus using the online database, figweb.com. The relative abundances of wasps were compared using an ANOVA test. LSD post hoc analyses were performed to determine differences between groups. A co rrelation matrix was used to compare each wasp species and the number of seeds per syconium. RESULTS Foundress wasp abundance There was no significant difference in the abundance of Pegoscapus mariae (N = 259) and P. carlosi (N = 333) at the study tree (t test, p = 0.096, t = 1.67) (Figure 1). Both pollinators were present in about 4/5 of the 99 phase B syconia (77% P. carlosi 78% P. mariae ). Of the 99 syconia collectedNinety three percent contained foundress wasps and 63% contained both species of pollin ators. I found that 16% had been colonized by only P. carlosi and 13% by only P. mariae The number of foundresses of P. carlosi within a syconium did not appear to appear to be correlated with the abundance of P. mariae (Correlation test, R 2 = 0.0126, F = 1.234, df = 1.97, p = 0.269, beta = 0.112); on average, syconia were pollinated by 5.98 3.54 wasps and almost always fewer than 14 (Figure 3). 0 1 2 3 4 5 6 7 P. carlosi P. mariae Number of wasps per syconium
Figure 1. The average number of foundress wasps encountered in 99 syconia from Ficus tuerckheimii did not d iffer (t test, p = 0.096, t = 1.67). 0 2 4 6 8 10 12 0 2 4 6 8 10 12 14 16 P. carlosi P. mariae Figure 2. The relative abundance of Pegoscapus carlos i and Pegoscapus mariae per Ficus tuerckheimii syconium. Wasps appear to colonize syconia randomly (Correlation test, R 2 = 0.0126, F = 1.23 4, df = 1.97, p = 0.269, beta = 0.112). In addition to the pollinator wasp fauna, I observed non flower structures within syconia that have been implicated in association with interloper wasps. White, irregular, tissue masses were present in seven syconi a. One to five of the structures was present in syconiaThese occurred both in syconia that contained wasps and those that were vacant of foundress wasps. Seven syconia contained florets that were encased within the syconium. I dissected the structures, bu t did not encounter any visibly developing organisms. These occurred both in syconia that contained wasps and those that were vacant of foundress wasps. Seven syconia contained florets that were encased in reddish sticky matter. These were sometimes associ ated with the white, gall like structures, but also occurred when no wasps were present and also when foundresses were present. Wasp progeny Interloper and pollinator wasps were not equally abundant within syconia (ANOVA, p < 0.001, F = 15.568, df error = 119, df effect = 4) (Figure 2). There was no significant difference between the number of P. mariae and P. carlosi that emerged from syconia (LSD test, p = 0.0861), nor between Idarnes and Torymus (LSD test, p = 0.441), although pollinators were significan tly more abundant than interlopers (LSD test, all p < 0.001). Hansonita was omitted from these comparisons as it was only encountered in one syconium. Although I made no attempt to identify males to species, the sex ratio for
pollinating wasps was clearly skewed, and females were considerably more abundant than males (LSD test, p < 0.001). I made the assumption that the males I identified were Pegoscapus as they were present in syconia that did not contain interlopers. 0 20 40 60 80 100 120 140 P. carlosi P. mariae Idarnes Torymus Hansonita Males Seeds Mean number per syconium Figure 3. The mean number (S.D.) of emerging wasps and seeds per Ficus tuerckheimii syconium. Pollinating wasps were significantly more abundant that nonpollinating wasps ( ANOVA, p < 0.001, F = 15.568; LSD post hoc, p < 0.001). The number of wasps that emerged fro m each syconium was highly variable (Figure 2). Of the 25 syconia collected, P. carlosi was the most abundant (N = 1808) species, distributed in 67% of all syconia. Pegoscapus mariae (N = 1528) was less abundant overall, but present in 71% of all syconia. Idarnes was the most common interloper (N = 239) and was encountered as many figs as either of the pollinator species, whereas Torymus (N = 35) was less common and only appeared in one third of the syconia. In contrast to the relationship between foundre ss wasps, the emerging offspring of P. carlosi and P. mariae were weakly negatively correlated (correlation, R 2 = 0.128, F = 3.65, df = 1.25, p = 0.067, beta = 0.36). Although not statistically significant, data suggest that syconia that contained more P. carlosi contained fewer P. mariae and vice versa. The presence of P. carlosi explained about 13% of the variation in the abundance of P. mariae The effect on wasp abundance was stronger than either that of average syconium diameter or seed production.
0 20 40 60 80 100 120 140 160 180 200 0 50 100 150 200 250 P. carlosi P. mariae Figure 4 Correlation between the number of emerged, congeneric pollinating wasps of Ficus tuerckheimii. Data suggest that P. carlosi has a slight negative effect on the abundance of P. mariae (Correlation test, R 2 = 0.128, F = 3 .65, df = 1.25, p = 0.067, beta = 0.36). Interloper wasps did not appear to preferentially associate with either species of pollinator wasp ( P. carlosi : R 2 = 0.008759, F = 0.2209, df = 1.25, p = 0.6424; P. mariae : R 2 = 0.01342, F = 0. 3400, df = 1.25, p = 0.5650). Nevertheless, Idarnes and Torymus were found exclusively in the presence of Pegoscapus There were no significant correlations between any of the pollinating wasps, nonpollinating wasps, or seeds (Correlation matrix, all p > 0.079). Larger sy conia were not characterized by more wasp offspring (R 2 = 0.023, F = 0.541, df = 1.23, p = 0.467, beta = 0.152) or higher seed production (R 2 = 0.010, F = 0.241, df = 1.23, p = 0.628, beta = 0.102). I observed that the number of emerging wasps was more rel ated to the color of the syconium rather than the size. This observation is substantiated by Bronstein (1988a), who determined that differences in fruit size upon pollination were maintained through maturation. 0 20 40 60 80 100 120 140 13 14 15 16 17 18 19 20 21 Average syconium diameter (mm) Estimated number of seeds 0 50 100 150 200 250 300 13 14 15 16 17 18 19 20 21 Average syconium diameter (mm) Number of wasp progeny
Figure 5. The effect of average syconium diameter on wasp progeny and seed production. DISCUSSION The overall strength of species interactions in the fig pollinator interloper system may not be as strong as ecological models suggest. Competitio n is defined as an antagonistic interaction over a limiting resource that results in a decrease in fitness in one or both of the species involved (Begon et al. 1990). If P. carlosi and P. mariae were competing, theory predicts that the syconia that had bee n founded by more P. carlosi would contain fewer P. mariae or vice versa. This may occur by one species exploiting more of the florets within a syconium, or by foundress wasps directly fighting over or avoiding figs pollinated by the congener so as to pre emptively avoid the decrease in fitness inflicted by the presece of the congener (the ghost of competition past) (Begon et al. 1990). If the wasps were in competition, one would expect that there would be a negative correlation between the number of P. ca rlosi and the number of P. mariae in a syconium. In actuality, foundress wasps appear to colonize syconia randomly. Data suggested that P. carlosi are slightly negatively correlated with P. mariae as emerging offspring. Pegoscapus carlosi was more abundant overall and on average both in foundress wasps and emerging offspring, so it is possible that this relationship is the artifact small differences in foundress abundance that are amplified in emerged offspring due to each foundress ovipositing multiple tim es within a syconium. Pegoscapus carlosi was slightly more abundant than P. mariae but this difference in abundance was not sufficient to support the prediction that there would be a competitively dominant species or pollinating fig wasp. On the other ha nd, White (1999) found that P. mariae was more abundant than P. carlosi It is possible that the abundance of pollinating species varies between trees and is the result of chance and stochastic processes that determine the number of foundress wasps that en counter a Ficus with receptive syconia. Therefore, a census of pollinator populations in multiple trees across a diverse range of habitats would be necessary in order to determine whether there are differences in the overall abundance of P. carlosi and P. mariae within a population or between metapopulations. It is also possible that P. carlosi and P. mariae are not interacting under limiting conditions. It is widely accepted that figs and their respective fig wasps have coevolved, and therefore, F. tuerck heimii has most likely provided sufficient florets for both pollinators so that there would not be an evolutionary incentive for a pollinator to exploit flowers with styles that would be otherwise reserved for seed production. In her pioneer work on Ficus pertusa in Monteverde, Costa Rica, Bronstein (1988a) determined that figs had evolved in a pollen limiting environment. In contrast to most other fruit trees, figs are not aborted due to limiting resources, but rather, in response to pollination. The trees matured 95% of the 65% of fruits that had been pollinated. This ratio appears to be similar for F. tuerckheimii as 93% of the syconia collected contained foundress wasps, suggesting that at maximum, 7% had not been pollinated. The definition of competiti on states that competition can only occur when resources are limiting (Begon et al. 1990). Therefore, if F. tuerckheimii is similar to F. pertusa and initial fruit production exceeds the number that will eventually be pollinated, P. carlosi and P. mariae may not be interacting under limiting conditions as fruit set would have evolved to exceed the
number of syconia pollinated by the wasps. Therefore, competition may not be strong enough to force the two congeners into the exploitative or interference compe tition that would elicit drastic resource partitioning. If this is the case, F. tuerckheimii may simply benefit by having two species of pollinators. In a pollen limiting environment, increased pollination or pollination success is the most effective way to increase fruit production. Ficus conservation in fragmented forests requires consideration of fruiting phenology, duration of receptivity of the figs to pollinator wasps, and pollinator dynamics (Mawdsley et al. 1998). Anstett et al. (in Mawdsley et al. 1998) used these parameters to construct a minimum viable population size (MVP) for figs in fragmented habitats that projected at 99% survival over 1000 years in order to assess the likelihood of survival for long lived species with reduced population siz es. Ficus tuerckheimii would likely have a smaller MVP because the probability of being encountered by a pollinator is higher when there are two possible species, making it more likely to survive in highly fragmented environments. Given that F. tuerckheimi i is already common in both pastures and forests in Monteverde, and is the most common species of Ficus in the area, it is probable that this species has a high chance of survival even if habitat fragmentation increases (Haber et al. 1996). Little is known about the nonpollinating wasp fauna of F. tuerckheimii however, much can be inferred based on morphology. Torymus Hansonita and Idarnes have long ovipositors used to penetrate the tough outer wall of the syconium. Torymus and Idarnes are the same size as Pegoscapus and are thus unlikely predators of the pollinator species (Bronstein 1991). Neither species was negatively correlated with seeds or pollinators, suggesting that the interlopers may not decrease the fitness of the tree either by directly decr easing seed production or by indirectly reducing the number of pollinators. I therefore propose that the relationship between F. tuerckheimii and interloper wasps is commensalistic rather than parasitic, as was suggested for F. pertusa (Bronstein 1991). H ansonita however, is considerably larger than the pollinators and was encountered in a syconium that was devoid of other wasps. Although highly anecdotal, finding this suggests that the species may be predatory on other wasps within the syconium (Bronstei n 1991). White galls have been reported for other interloper species, and it is therefore possible that the syconia with white, gall like structures observed in the census of foundress abundance contained developing parasites or parasitoids (Bronstein 1991 ). The plausible species interactions observed between all wasp species is summarized in Table 2. Table 2. Summary of plausible interactions between pollinating and nonpollinating wasp species of Ficus tuerckheimii P. mariae P. carlosi Idarnes Troymus Hansonita P. mariae --------------None for foundress, weak negative emerging None None Negative (anecdotal) P. carlosi None for foundress, weak --------------None None Negative (anecdotal)
negative emerging Idarnes Necessary for colonization Neces sary for colonization --------------None Negative (anecdotal) Troymus Necessary for colonization Necessary for colonization None --------------Negative (anecdotal) Hansonita None None None None --------------The number of seeds was not explained by the abundance of emerging wasps or by the average diameter of the syconium. I initially predicted that the number of wasps would be negatively correlated with seed production, as wasps act as seed predators on fig ovules (Janzen 1983). Alternately, if seed production were a reflection of the number of flowers pollinated by foundress wasps, we would expect that it would be positively correlated with the number of foundresses and, indirectly, the number of emerging offspring. This was not supported by the data. A possible explanation for the variation in seeds encountered within syconia may be that the seeds had not fully developed, as Galil and Eisikowitch (1968, in Bronstein 1989) determine that seeds do not develop until reproductive phase E, when fruit s become fleshy and wasps have left the syconium. If a single wasp pollinates a syconium, its offspring will mate exclusively with siblings, thereby potentially reducing their fitness. Alternately, if too many wasps enter a syconium, the females may exper ience intraspecific competition for short styled ovules. Morton (2006) proposed that there is an optimum number of foundresses per syconium that mediates competition and avoids inbreeding depression. Presumably, P. mariae and P. carlosi experience increase d competition and a higher chance of encountering syconia that contain only one other conspecific foundress wasps. Although it was outside the scope of the present study, valuable information regarding seed set and wasp emergence could be gained by followi ng fruit through the entire course of their development. It would also be beneficial to increase the sample size for emerging wasps in order to determine the relative abundance and overall richness of interlopers. This study provides persuasive evidence th at foundress pollinating wasps do not avoid colonizing syconia that have been entered by the congener, but other possible mechanisms of resource partitioning remain. To address these questions, future studies may wish to sample replicate trees from across a wide range of habitats. Yule (1999) proposed that P. carlosi and P. mariae were not discrete species, but two color morphs of the same species. This would explain the apparent lack of niche differentiation exhibited by the congeners, but has yet to be su bstantiated in the literature. Genetic analysis of DNA could be used to determine the degree of difference between the two wasps. Alternately, the wasps currently defined as P. carlosi and P. mariae could be separated upon emerging from syconia and placed in pollination bags around maturing syconia. If the offspring that emerged from a syconium manipulated to contain only P. carlosi also contained yellow morphs, it could be inferred that P. carlosi and P. mariae were in fact one species. ACKNOWLEDGEMENTS
I would like to thank Tania Chavarr Â’ a for advising my study, and Alan and Karen Masters for providing the backdrop of tropical ecology that helped me place my study in context. Thanks to Camryn Pennington for sound advice and for keeping starvation and dis order at bay in the lower lab. I would like to thank Tom McFarland for the ingenious pole clipper extension, wire probes, and of course, the daily puzzles. Many thanks to everyone at Hotel Belmar, first for the use of their Ficus and also for their help, f riendliness, and general interest in the project. In addition to Hotel Belmar, I would also like to thank Hotel Sapo Dorado, Stella's Bakery, and Hotel El Bosque for letting me sample figs on their properties when I was initially choosing a study tree. Muc h Â’ sima gracias to Ana Ruth, Jairo, Andrey, and Alexandra for their hospitality and good company. Many thanks to Bonnie Waring and Sara Weinstein for helping me make sense and order out of my paper. Lastly, I would like to acknowledge the lower lab rats for their astounding intelligence, wit, determination, and microscopitory stamina. LITERATURE CITED Begon, M., J. L. Harper, and C. R. Townsend. 1990. Ecology: Individuals, Populations, and Communities, 2 nd Ed. Blackwell Scientific Publications, Cambridge, MA. pp. 191 195. Bronstein, J. L. 1988a. Limits of fruit production in a monoecious fig: consequences of an obligate mutualism. Ecology 69(1): 207 214. Bronstein, J.L. 1988b. Mutualism, antagonism, and the fig pollinator interaction. Ecology 69(4):1298 130 2. Bronstein, J. L. 1989. A mutualism at the edge of its range. Experientia 45: 622 637. Bronstein, J. L. The nonpollinating wasp fauna of Ficus pertusa : exploitation of a mutualism? Oikos 61: 175 186. Grison Pige, L., J. Hessiere, and M. Hossaert McKey. 2 002. Specific attraction of fig pollinating wasps: role of volatile compounds released by tropical figs. Journal of Chemical Ecology 28(2): 283 295. Haber, W., W. Zuchowski, and E. Bello. 1996. An Introduction to Cloud Forest Trees. Monteverde, Costa Rica. p. 153. Holdridge, L. R. 1967. Lifezone ecology. Tropical Science Center, San Jose, Costa Rica. Janzen, D. H. 1983. What sort of pollinators are Costa Rican insects? Fig pollination. In: Costa Rican Natural History D. H. Janzen, ed. The University of Ch icago Press, Chicago, IL, pp. 696 700. Kerdelhu ÂŽ C., M.E. Hochberg, and J. Rasplus. 1997. Active pollination of Ficus sur by two sympatric fig wasp species in West Africa. Biotropica 29(1): 69 75. Mawdsley, N. A., S. G. Compton, and R. H. Whittaker. 1998. Population persistence, pollination mutualisms, and figs in fragmented tropical landscapes. Conservation Biology 12 (6): 1416 1420. Michaloud, G., S. Carriere, and M. Kobbi. 1996. Exceptions to the one:one relationship between African fig trees and their figs wasp pollinators: possible evolutionary scenarios. Journal of Biogeography 23: 513 520. Morton, D. N. Syconia Selection Criteria of the fig wasps Pegoscapus mariae, P. carlosi, and a non pollinating species on Ficus tuerckheimii UCEAP Monteverde Tr opical Biology, Spring 2006. Ram Â’ rez, W. B. 1970. Host specificity of fig wasps (Agaonidae). Evolution 24: 680 691. White, A. 1997. Interactions of two sympatric species of agaonid wasps on Ficus tuerckheimii UCEAP Monteverde Tropical Biology, Spring 19 97. Yule, A. M. 1999. Pegoscapus carlosi and Pegoscapus mariae spatially differentiate while coexisting as Ficus tuerckheimii pollinators: species of morphs? UCEAP Monteverde Tropical Biology, Spring 1999.