Search

The obligate mutualism between figs and their species-specific pollinator wasps has been thought to be relatively fragile in the face of population fluctuations of either mutualist. Here we report on the recovery of a Florida fig pollination mutualism devastated by Hurricane Andrew in August 1992. Damage to Ficus aurea included loss of all leaves and fruits and many branches, as well as the presumed local extinction of its pollinator Pegoscapus jimenezi. Within five months, however, fig flowering plenology and fig wasp abundance (measured by the number of pollinators entering inflorescences) had recovered to near prehurricane levels. Unusual phenological traits of F. aurea may have aided in the rapid reestablishment of pollinator populations; in addition, the wasps may have previously underappreciated capacities for long distance movements. This study suggests that obligate interactions can be surprisingly resilient to certain population-level catastrophes.

Larval development of the species-specific pollinators (Agaonidae, Hymenoptera) of the two native Florida fig species is markedly temperature-sensitive. Census data from a site near Miami indicate that Pegoscapus assuetus (the pollinator of Ficus citrifolia) and Pegoscapus jimenezi (the pollinator of Ficus aurea) take 7-8% longer to develop for each 1 C decrease in ambient temperature. However, development does not slow sufficiently to permit the wasps to remain within their sheltered larval habitat (the fig inflorescence) throughout winter, a season in which the short-lived adults have little chance of finding an oviposition site. The difficulty of wasp persistence through winter is probably a major factor limiting most fig-pollinator mutualisms to the tropics. The ability of a few species to persist at the northern limits of some subtropical regions, such as in S Florida, remains unexplained.

I describe aspects of the life histories of the 3 nonpollinating wasps (Torymidae) associated with a common fig tree in Costa Rica.Specifically I consider patterns of wasp arrival to trees, oviposition behavior, larval ecology and species specificity of the fig/non pollinator interaction.The purpose is: to contrast life histories of mutualist and non mutualist wasp dependent upon the same fig ; to examine how the fig/pollinator mutualism limits and is limited by the non pollinating wasps; to make some predictions about the evolution of the interaction between figs and their non pollinating wasps

Comparing populations that differ in access to mutualists can suggest how traits associated with these interactions have evolved. I discuss geographical and seasonal variation in the success of a primarily tropical mutualism (the fig/pollinator interaction), and evaluate some possible adaptations allowing it to persist at the edge of its range. Pollinators probably have difficulty in seasonal sites because 1) fig trees flower rarely in winter and 2) trees that do flower are less detectable and more difficult to reach. Fig biologists believe that seasonality must have selected for adaptations allowing pollinators to survive winter. However, geographical comparisons do not support two current ideas, the synchrony-breakdown hypothesis and the specificity-breakdown hypothesis. I pose two alternatives: plasticity of fruit and wasp developmental time, and adaptations of free-living fig wasps. I also distinguish between the impact of seasonality on monoecious versus dioecious figs; the latter group appear better adapted to reproduce in cool climates. A combination of comparative, observational, and experimental approaches has great potential for advancing our understanding of mutualisms.

Fruit maturation patterns of a Costa Rican fig tree, Ficus pertusa, were studied. Like other figs, F. pertusa is pollinated by a species-specific wasp (Agaonidae) whose female offspring transfer pollen between trees. Pollination success (the proportion of inflorescences pollinated) ranged from 1 to 100% among 21 F. pertusa crops studied over 2 yr and averaged only 65%. Resource availability evidently was limiting to fruit set as well: every crop abscised many inflorescences at a predictable point during growth. This abscission period usually preceded the brief but variably timed period of pollinator arrivals; in most cases every pollinated, undamaged inflorescence set fruit. In contrast, many resource-limited species reduce crop sizes after pollination. Numerous advantages of fruit abortion have been proposed. However, fruit abortion would not be expected in plants that evolved under conditions of pollen limitation; although pollinators have been thought to be overabundant in highly coevolved pollination mu- tualisms, figs' unique flowering phenology may in fact make pollen limitation common. The fact that future pollen carriers as well as seeds mature within fruits may also help explain these unusual fruit maturation patterns.

tical models. Regression, analysis of variance, and experimental designs. Richard D. Irwin, Homewood, Illinois, USA. Pielou, E. C. 1977. Mathematical ecology. John Wiley and Sons, New York, New York, USA. SAS Institute. 1985. SAS user's guide, statistics. 1985 edition. SAS Institute, Cary, North Carolina, USA. Schultz, J. C., and I. T. Baldwin. 1982. Oak leaf quality declines in response to defoliation by gypsy moth larvae. Science 217:149-151. Scott, H. M. 1924. Observations on the habits and life history of Galerucella nymphaeae (Coleoptera). Transactions of the American Microscopical Society 43:11-16. Strong, D. R., J. H. Lawton, and R. Southwood. 1984. Insects on plants, community patterns and mechanisms. Harvard University Press, Cambridge, Massachusetts, USA. Wallace, J. B., and J. O'Hop. 1985. Life on a fast pad: waterlily leaf beetle impact on water lilies. Ecology 66:15341544. Williams, A. G., and T. G. Whitham. 1986. Premature leaf abcission: an induced plant defense against gall aphids. Ecology 67:1619-1628.

The over 700 species of Ficus are thought to have co–speciated with their obligate pollinators (family Agaonidae). Some of these wasp species pollinate figs actively, while others are passive pollinators. Based on direct observations of mode of pollination in 88 species, we show that mode of pollination can confidently be predicted from fig traits only (anther–to–ovule ratio) or from wasp traits only (presence of coxal combs). The presence of pollen pockets is not a predictor of mode of pollination. Data, direct and indirect, on 142 species, demonstrate numerous cases of the loss of active pollination and suggest one or few origins of active pollination. Hence, active pollination, an impressive example of the sophisticated traits that may result from mutualistic coevolution, depends on selective forces that can be overcome in some species, allowing reversions. Despite frequent loss, active pollination remains the predominant mode of pollination in Ficus .

The origin of the deep pools of liquid held in the cuplike bracts of Heliconia imbricata (Heliconiaceae), in which flower bases and fruits are submerged, was examined. Strong evidence was obtained that bract fluid is primarily a plant secretion rather than accumulated rainwater. Drained bracts regained their original volumes by the next day regardless of whether they were protected from rain; an average drained bract accumulated over 11 ml of liquid overnight. Newly-opening bracts held on average over 9 ml of fluid, and evidence suggests that fluid continues to be secreted into bracts for weeks after they have opened. The average pH of the accumulated fluid was basic (averaging 7.8) relative to rainwater (pH 5.0). Bract fluid may function in defending flowers and fruits from herbivores, although physical factors of the bracts appear to be important as well. HELICONIA PLANTS (HELICONIACEAE) ARE ABUNDANT and conspicuous throughout the neotropics. All species have thick bracts that surround the flowers and developing fruits, although the arrangement and orientation of the bracts vary greatly within the group (Daniels & Stiles 1979). Many Heliconia species have erect inflorescences consisting of a series of brightly colored, alternately-arranged bracts, many of which at any time contain flowers, buds, and developing fruits. New bracts arise at the top of the rachis every 3-10 days. These erect, cuplike or spoutlike bracts often hold pools of liquid. Several researchers have suggested that the liquid is trapped rainwater, which protects the submerged fruits and flower bases from herbivory (e.g., Skutch 1933, Stiles 1979), although Seifert (1982) did state that in some Heliconia species part of the liquid is actively transported into the bracts. Neither the origin of the fluid nor its protective function have ever been verified experimentally. Here I report on the origin and nature of the bract fluid of Heliconia imbricata (Kuntze) Baker. Heliconia imbricata is one of the most common species of Heliconia in lowland Costa Rica. During its flowering period its dark-red, cuplike bracts hold large volumes of foul-smelling liquid known to contain a diverse microfauna (Vandermeer et al. 1972). Also Heliconia imbricata inflorescences support a small community of aquatic and semiaquatic insects, primarily fly and beetle larvae, that feeds in and around the liquid-filled bracts (Seifert 1975, Seifert & Seifert 1976a, b). Typical of most Heliconia species in Costa Rica, H. imbricata flowers and fruits most heavily during the wetseason months of July and August (Daniels & Stiles 1979). Dry spells are frequent, however, and small standing pools of water probably evaporate quickly. If the risks of floral and fruit herbivory were great and submergence did act ' Received 29 January 1985, accepted 20 June 1985. as protection, then H. imbricata might be expected to actively and regularly secrete, rather than passively and opportunistically trap, fluid in its bracts. Therefore I predicted that if H. imbricata bracts accumulated only rainwater or throughfall liquid, then bagged, drained bracts would remain dry, while exposed, drained ones would accumulate new fluid, and newly-opening bracts would contain no fluid. Additionally, I expected that if the liquid were rainwater, its pH would be equal to or more acidic than rain (since organic decay acidifies standing water). Finally, I examined patterns of liquid volume held in

Predators inflict high mortality on the four species of wasps associated with the fig Ficus pertusa L. in Monteverde, Costa Rica. Because one of these wasps is the obligate pollinator of the fig, predation may have a strong impact on successful pollen donation by the plant. The natural histories of several predators are described: an ant that feeds on wasps arriving to oviposit, moth and weevil larvae that destroy wasps as they develop within the fruits, a staphylinid beetle that feeds on mature wasps before they leave the fruits, and a group of birds that gleans wasps as they leave. The synchrony of arrival and departure of pollinators from the fig trees probably make them the species least vulnerable to predation. FIGS (Ficus SPP.) AND THEIR POLLINATOR WASPS form a "keystone mutualism" (Gilbert 1980), an interaction upon which other species depend and around which they accumulate. The dependence of frugivorous birds (Snow 1981, Beehler 1985), bats (Heithaus et al. 1975, Morrison 1978), and primates (Raemaekers 1979, Milton 1980, Terborgh 1983) on ripe figs is well known. Several insects (Smilanick & Ehler 1976, Janzen 1979, Frank 1983) and birds (Janzen 1981) are major seed predators when fruits are still on the tree, and a large group of lygaeid bugs feeds exclusively on seeds of fallen figs (Slater 1972). Such interactions have a major impact on the female reproductive function of seed production in figs. Less well known are species that interact with the mutualism via their impact on the pollinator wasps (Hymenoptera: Chalcidoidea: Agaonidae), and thus potentially limit the figs' male reproductive function of pollen donation. The larvae of certain chalcidoid wasps are known to kill developing agaonid larvae (Kuttamathiathu 1955, Galil & Eisikowitch 1968), and ants (Frank 1983), dragonflies, and birds (Janzen 1979) have been observed to prey upon adults. Here I describe the suite of insects and birds that are major predators of four wasp species associated with one Costa Rican fig, and speculate on the extent to which they specialize on fig wasps. I then consider how great an impact predation has on the pollinator species, and thus indirectly on the fig itself. STUDY SITE AND NATURAL HISTORY Ficus pertusa, a small (5-12 m), hemiepiphytic or freestanding tree, is one of the two most common Ficus species in Central America (Condit 1969). It is found from 0 to 2000 m in moist highland to lowland forests and savannas and in both disturbed and undisturbed habitats (Burger 1977). F. pertusa grows abundantly in pastures and remnant patches of primary and secondary forest throughout the community of Monteverde, Costa Rica, located just below the Continental Divide on the Pacific slope of the Cordillera de Tilaran (10?12'N, 84?42'W, elevation 13501450 m). Undisturbed forest in the area is transitional between lower montane wet and moist forest (Holdridge

While the one-to-one specificity of the fig pollinator mutualism is often discussed in reviews of coevolution, the means by which specificity of the interaction is enforced and the frequency of pollinator errors have not been examined. Specificity at different stages in the interaction was studied in a common neotropical fig, Ficus pertusa L., particularly in relation to its sympatric congener F. tuerckheimii Standley. 99% of pollinators arriving at sticky traps on flowering figs were the specialist species. Foreign pollinators virtually never entered F pertusa syconia to oviposit or emerged from mature fruits. Pollinators arrive at F pertusa trees in a one-day burst that is well timed with the presence of unpollinated syconia, providing evidence for the existence of a species-specific volatile attractant. Some nonpollinating wasps associated with F pertusa appear to use the same attractant to locate the tree.