Beijing's international symposium on figs and fig wasps (2004)

September 15-19, 2004, Beijing, China

Institute of Zoology, Chinese Academy of Sciences

Chinese Entomological Society

ABSTRACTS

The evolutionary history of the fig wasp mutualism: origins and beyond the cospeciation paradigm (by Carlos A. Machado)

Carlos A. Machado

Dept. Ecology and Evolutionary Biology, University of Arizona 1041 E Lowell St., BSW 310, Tucson, AZ 85721 cmachado@email.arizona.edu

Abstract: I will review our current knowledge of the evolutionary history of the fig/fig wasp mutualism, focusing on recent attempts to compare molecular phylogenies of figs and wasps. I will discuss the reasons why those attempts are inadequate to test the hypothesis of cospeciation in the mutualism. Those studies have generally sampled one (or few) individual taxa representing the distal branch tip of very ancient, distantly related taxonomic subdivisions within the genus Ficus, and usually rely on analyses of only one or two genes. Such sampling provides only weak phylogenetic support, and will differentially emphasize ancient outcomes rather than the processes that ultimately generated them. Further, previous attempts have assumed a one-to-one fig/pollinator relationship, which has been shown not to hold for several Neotropical taxa. I will discuss the potential consequences that the presence of multiple pollinators may have on patterns of speciation and divergence in the figs and how that can affect tests of cospeciation. I will discuss more fruitful approaches for testing the cospeciation hypothesis and will present new resources that we have developed to conduct such tests.

Co-relationship between pollinator entry and ostiole close mechanism: a case study of F. microcarpa ― Eupristina verticillata mutualism (by Tong-Xin Zhang)

Tong-Xin ZHANG^1,2, Da-Wei HUANG ^1,3*, Yue-Guang FU⁴, Zheng-Qiang PENG⁴

¹Institute of Zoology, Chinese Academy of Sciences, Beijing, 100080, China.

² Graduate School of the Chinese Academy of Sciences, Beijing, 100039, China

³ Plant Protection College, Shandong Agricultural University, Tai’an, Shandong, 271018, China

⁴ Southern China University of Tropical Agriculture

^*Corresponding author: huangdw@ioz.ac.cn

Abstract: In the mutualism of F. microcarpa Linn. and its pollinator Eupristina verticillata Waterston, both partners completely depend on each other for their effective reproduction. The syconial ostiole plays a crucial role in letting pollinators get into the fig cavity and preventing other insects outside. The open and close of ostiole functionally determine the pollinator entry and further affect trade-off between pollinator offspring and fig seeds. To test the factors initiating the ostiole close and their intensities, experiments are conducted in Hainan Island, China. We have worked out measurement for the feedbacks of the ostiole to different entry numbers and entry times of pollinators. We recognized the diagnostic morphological characters for the ostiole open ready. Without pollinators, the ostioles keep open for 49-57h. When pollinator/pollinators enter, the ostioles close in relative shorter time (5-13h).

We introduced 1, 2 or 3 pollinators respectively at twenty replicates into figs one hour after the ostiole open-ready. The results showed that ostioles close within 12h when one pollinator entered, in 9h when 2 or 3 pollinators entered. We introduced one pollinator into figs respectively 1h, 6h, 11h, 16h and 21h after the ostiole open- ready. The ostioles close within14h, 17h, 20h, 24h and 26h respectively. After the first foundress’ entry, the later the successive pollinators enter, the more possible of their death on its way in. We concluded that pollinators’ entry can initiate the ostiole close mechanism and accelerate the close process. By this means, the ostiole allows the appropriate foundress enter the fig cavity and maintain the optimal tradeoff of the fig seeds and fig wasps.

Key words: Ficus microcarpa, Eupristina verticillata, ostiole open-ready, ostiole close, pollinator foundress

The trade-off in fig/fig wasp mutualism and the reciprocal system maintenance dynamic (by Rui-Wu Wang)

Rui-Wu WANG,

Kunming Institute of Zoology, Chinese Academy of Sciences, Kunming, Yunnan 650223. China. E-mail: ruiwukiz@hotmail.com

ABSTRACT: Whether the conflicts among mutualists exist in any systems of the reciprocal mutualism is in a heated argument. And in what situations the partners cooperate or conflict and why the mutualists do not exploit more of the available resource at expenses of their partners and thereby destabilize the mutualistic system are still not clear. Our work on F. racemosa shows whether the cooperation or conflict exists among the mutualists is mainly determined by the availability of the public resource (the unutilized female flowers). In fig/fig wasp system, both of the fig and its species-specific pollinator fig wasp (Agaonidae) utilize the same female flowers by which fig produce viable seeds and the pollinator wasp produce wasp offspring. In the cases that the proportion of vacant female flowers (unutilized female flowers) is low, such as in January, the number of viable seeds is negatively correlated with the number of wasp offspring and the foundress number is also negatively correlated with the number of viable seeds. The conflicts obviously exist in such system. While the number of viable seeds is positively correlated with the number of wasp offspring when the proportion of vacant female flowers is high, such as in the period from March to June, and the foundresses number is positively correlated with both of the viable seeds and wasp offspring. They might tend to cooperate when the public resource is not in limitation.

Our work on F. racemosa shows the style length of female flowers, the space of fig cavity, the regulation mechanism of figs to the abundance of foundresses entered in the cavities and the climate can directly or indirectly influence the trade-off between the viable seed and the wasp offspring. In most of the cases, because of the evolutionary and ecological constraints, the mutualists are not able to maximize the utilization of their public resource (female flowers) and thereby they cooperate rather than conflict. These data and results present the first empirical evidences to Axelrod & Hamiltom's model on the maintenance dynamics of reciprocal mutualism. Obviously, the net benefit to maintain the system stable to each partner in reciprocal mutualism is larger than that of defection in the cases that conflicts exist among them. It is "Prisoner Dilemma", and thereby is in a stable state.

Because the public resource (female flowers) is only provided by the plants, and the benefit to plant to maintain the system stable is much larger than that of their pollinators, and thereby the plants develop the mechanism to maintain the system stable (Boxed Pigs). The asymmetry between the mutualists implies that the reciprocal mutualism might evolve from the antagonistic interaction not from the commensalistic interaction. The altruistic mutants of the antagonistic partner can increase their fitness by the more resource reward from the hosts' discrimination against the exploiters and thereby the reciprocal mutualism could be evolved (by direct reciprocity).

Key words: mutualism; cooperation; conflict; fig; fig wasp; trade-off; coevolution

Patterns of diversification of afrotropical otiteselline fig wasps: evolution of host use and ecological niches (by Emmanuelle Jousselin)

Emmanuelle Jousselin¹*, Simon van Noort², Jean-Yves Rasplus³, Jaco Greeff¹

¹Department of Genetics University of Pretoria, Pretoria 0002, South Africa.

²Natural History Division, South African Museum, Iziko Museums, PO Box 61, Cape Town 8000, South Africa.

³Institut National de la Recherche Agronomique,Centre de Biologie et de Gestion des Populations, Campus International de Baillarguet, CS-30 016, 34 988 Montferrier sur Lez, France

Abstract: We studied the phylogenetic relationships of Otiteselline fig wasps associated with Ficus in the Afrotropical region using rDNA sequences. African fig species usually host two species of otiteselline fig wasps. Phylogenetic analyses reveal that this pattern of association results from the radiation of two clades of wasps superimposed on the fig system. Within each clade, wasps collected from different localities, but from the same host fig species group together, and species generally cluster according to their host taxonomy. The phylogenies of the two clades are also more congruent than expected by chance. Altogether, these results suggest that Otiteselline wasp speciation is largely constrained by the diversification of their hosts. Finally we show a difference in ovipositor length between the two Otiteselline species coexisting in the same Ficus species which probably corresponds to ecological differences. The diversification of ecological niches within the fig is probably, with cospeciation, one of the key factors explaining the diversification and maintenance of species of parasites of the fig/ pollinator system.

Keywords: Adaptive radiation, coevolution, cospeciation, Bayesian inference, Ficus, oviposition, phylogeny, plant/ insect interaction.

Figs and the diversity of tropical rain forests (by Rhett D. Harrison)

Rhett D. Harrison

Smithsonian Tropical Research Institute, Tupper Building, Unit 0948 APO, AA 34002 USA

Abstract: Explaining the diversity of tropical rain forests is a fundamental goal of tropical ecology. One approach to this problem is to examine the biology of characteristic tropical forest species and thereby derive an understanding of the traits that permit the co-existence of so many species. Ficus (Moraceae) is arguably the most important tropical plant genus. Figs are diverse. They are found in all lowland tropical rain forests and possess a variety of habits characteristic of tropical rain forest plants. Less well appreciated, however, is the extent to which figs contribute to the diversity of species in tropical rain forests. Here I review 17 tropical lowland florulas (Asia-pacific 7, Africa 4, Neotropics 6) to demonstrate that Ficus is invariably one of the most species-rich genera in lowland tropical forest. The variety of plant habits and very low densities and broad ranges of many species have lead to underestimation of fig alpha-diversities in the past. Further examination of florulas from the Asia-Pacific region reveals that (1) Ficus exhibits a steep latitudinal gradient, (2) figs are especially important component of regenerating forests on volcanic islands, and (3) in highly urbanised and otherwise disturbed environments fig assemblages are impoverished to a great extent than other genera. In the Neotropics figs, where relative fig diversity was more variable, figs were most speciose compared to other genera at sites with the highest overall plant diversity. Four basic attributes of fig biology have most likely interacted for their success, (1) Ficus is has evolved to encompass a phenomenal range of life-histories; (2) pre-adaptation to pioneer habits, in particular small seed size and diverse rooting habits, has enabled figs to colonise a broad variety of high energy environments; (3) high assimilation rates support high growth rates and production of latex, which protects, especially the inflorescences, against herbivory; and (4) figs possess a highly efficient – long range pollination system, which enables them to occupy rare niches untenable to other rain forest plants.

Keywords: Ficus, alpha-diversity, plant life-history, pioneer, pollination

The fig wasps in syconia of Ficus racemosa (by Lei Xu)

Lei XU ^*1,2, Da-Rong YANG ¹

1 Xishuangbanna Tropical Botanical Garden, China Academy of Sciences, Kunming 655023.

2 Graduate school of the China Academy of Sciences, Beijing 100039. *E-mail: ziboxul@sohu.com

Abstract: In the tropical rainforests of Xishuangbanna there has about 70 fig species, the most species of Ficus in China. The fig is considered to be the pre-eminent group of keystone plant resources in the ecosystem of tropical rainforests. Each species of Ficus is pollinated by an unique fig-pollinating wasp to complete its sexual reproduction, and the pollinator only depends on the florets ovaries inside host inflorescences to foster its offspring. Figs and their species-specific pollinating wasps (Agaonidae) form a remarkable plant-insect mutualism. Besides pollinating wasp in the syconia, there also exist several functional groups of non-pollinating wasps. They are gall makers that attack syconia from the exterior, gall makers that enter syconia as do the pollinators, and parasitoids that attack other fig wasp larva. Ficus racemosa Linn. is monoecious, which is an important plant owing to their heavy fruit production all year round to support a broad spectrum of vertebrate furgivores during times of food scarcity in the ecosystem of Xishuangbanna tropical rainforests.

From 2000 we observed the phenology characters of F. racemosa in the Menglun town of the Xishuangbanna Dai Autonomous Prefecture. And we also studied the fig wasps in the figs of F. racemosa by observing their behaviors and conducting quantitative introduction experiments. In the syconium of F. racemosa there are six wasp species: Ceratosolen fusciceps Mary, Apocrypta westwoodi Grandi, Apocrypta sp., Platyneura testacea Motschulsky, Platyneura mayri Rasplus and Platyneura agraensis Joseph. They belong to Agaonidae, Caliimomidae and Pteromalidae of Chalcidoidea respectively, among which only Ceratosolen fusciceps Mary of Agaonidae pollinates for F. racemosa by entering the figs from ostiole, Apocrypta westwoodi Grandi and Apocrypta sp. of Pteromalidae are the parasites of other wasp species. The Platyneura testacea Motschulsky, Platyneura mayri Rasplus and Platyneura agraensis Joseph of Caliimomidae are the parasites of fig seeds or the parasites of pollinators. They could result in the ovaries of the long-styled florets, which had been destined to produce seeds, turning into galls. Meanwhile, some might lay eggs into the ovaries of short-styled florets, which had contained the larva of fig-pollinating wasps, thus the pollinators would die for lacking of space and food resources. All these non-pollinating wasps oviposit outside the figs. The behavior observation results are following: P. testacea oviposit in the pre-female phase at first, and then Apocrypta sp. do after 5-7 days. P. mayri oviposit 2-3 days before and after the entering of C. fusciceps. After the pollinating wasp C. fusciceps enter fig 7-8 days, A. westwoodi begin to oviposit. After two weeks, P. agraensis oviposit at last. The wasps of Apocrypta are the parasites of the pollinating wasps in the primary view. But our observation and experiment results showed Apocrypta sp. parasite P. testacea, A. westwoodi can parasite P. mayri. In the conducting quantitative introduction experiments, we introduced P. mayri at first, then A. westwoodi to oviposit on the figs, in these figs there appear the offspring of A. westwoodi. While introducing the pollinator C. fusciceps firstly and then A. westwoodi, no A. westwoodi offspring have been found. But these results need to be validated by more experiments. And the relationships between pollinating and non-pollinating wasps and among these six wasp species also need to be studied further.

Keywords: Ficus racemosa ，pollinator，non-pollinating wasp， parasite

Interaction between nonpollinators and pollinator mutualism in the Ficus hispida (by Da-Rong Yang)

Da-Rong YANG ¹, Qiu-Yan WANG ¹, Yan-Qiong PENG ^1,2 , Xu LEI^1,2

1 Xishuangbanna Tropical Botanical Garden, China Academy of Sciences Kunming 655023

2 Graduate school of the China Academy of Sciences, Beijing 100039. *E-mail: yangdr@xtbg.ac.cn

Abstracts: The coevolution of figs (Ficus spp., Moraceae) and its associated pollinating fig wasps (Hymenoptera, Chalcidoidea, Agaonidae) are some of the more studied and storied relationships of obligate mutualism in the science world. Over 700 species of figs and their respective species-specific pollinating fig wasp have been described (Bronstein, 1991). While the act of pollination/reproduction of all figs and their fig wasp partners is remarkably conserved behaviorally, many wasps that do not act as pollinators (but are still considered fig wasps because of their use of the syconia to develop their offspring) use several different methods to successfully reproduce within the developing fig syconia. Complex inter- and intraspecies behavior and morphology within the non-pollinating wasp community have so far spurred several observational, behavioral and theoretical studies that have tried to place these non-pollinating fig wasp species into their proper ecological niche, but they are mainly based on the field observation and the analysis of natural wasp community and subsequent deduction. We conducted quantitative introduction experiments to clear the trade-off between pollinators and nonpollinators. Ficus hispida L. is a kind of pioneer species in the secondary woodland of tropical rainforests zone. Only being pollinated by its symbiotic pollinator wasp (Family Agaonidae) can it bears seeds and sustains its race. To its pollinator wasp (Ceratosolen solmsi marchali Mayr), only relying on the ovaries of florets dedicated by the males of F. hispida can it completes its life cycle. Except the symbiotic mutualistic pollinator wasp, there are three other species of non-pollinator wasps competing in the same crowded syconium of F. hispida, Philotrypesis pilosa, Philotrypesis sp. and Apocrypta bakeri. They are different kinds of wasp from a pollen delivery wasp and they insert their ovipositors from outside the syconium to lay eggs, and therefore, they don’t participate in the delivery of pollen. They inhabit the syconia of F. hispida at different developing time. Our Research work was conducted on the behavior and mechanics of the three non-pollinator wasp parasitism from August, 2002 to August, 2003, results show, all three non-pollinator wasps life cycles were much more variable than pollinating wasps, Philotrypesis pilosa was the first one who exploit the system of Ficus and its pollinators as soon as the pollinator entered into the fig cavity and oviposited, so the oviposition of Philotrypesis pilosa and Ceratosolen solmsi was performed simultaneously and their ovipositon could keep 2d at most, 1d commonly, Philotrypesis sp. was the second one who exploit the symbiotic system, and when the pollinator had been into the syconia for 6d to 15d Philotrypesis sp. loaded on it, their oviposition could keep 7d at most; the last one who exploit the syconia was the Apocrypta bakeri, their ovipositon came up during the period of the pollinator being into the syconia for 19d to 23d and kept 7d at most as well; the time spot had subject impact on the offspring dimension of Philotrypesis sp. when introduction at the different time spot of the period of Philotrypesis sp. ovipositon(significance level <0.001); field Apocrypta bakeri and Apocrypta bakeri of inside syconia were got and introduced into the syconia of the same sample in the same number respectively, results show that the offspring quantity of former was significantly higher than the latter; in the natural case pollinators occupied 87%, Philotrypesis pilosa 3%, Philotrypesis sp. 9% and Apocrypta bakeri 1%; correlation analysis, path analysis and regression analysis were carried out on the natural composition, results show, all three nonpollinators had negative impact on the pollinator, Philotrypesis sp. had the most decision coefficient on the pollinator, next to it was Philotrypesis pilosa, the last one was Apocrypta bakeri; when all the three nonpollinators were absent the pollinator scale was smaller, the occurrence of Philotrypesis pilosa and Philotrypesis sp. could reduce the number of pollinators significantly; the wasp composition was different at the same time but different sites, and at the site where jamming and devastation was serious the nonpollinators species was poor; the wasp number was different in the different time even in the same site, in the rainy season was the biggest and the foggy cool season was the least.

Key words: Ficus hispida, Ceratosolen solmsi marchal, Pollinator, Nonpollinator

More pollen-free pollinators, less their offspring: pollen effect on pollinator reproduction (by Wen-Quan Zhen)

Wen-Quan ZHEN ^1,2,3, Da-Wei HUANG ^1,4*, Jin-Hua XIAO ^1,3, Da-Rong YANG ²

¹Institute of Zoology, Chinese Academy of Sciences, Beijing, 100080, China.

² Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences, Yunnan, 666303, China

³ Graduate School of the Chinese Academy of Sciences, Beijing, 100039, China

⁴ Plant Protection College, Shandong Agricultural University, Taian, Shandong, 271018, China

^*Corresponding author: huangdw@ioz.ac.cn

Abstract: Pollen is vital to seed production of flowering plants. In fig and fig wasp mutualism, however, pollen can affect both the production of pollinator offspring and the seed reproduction. The studies with one pollinator were conducted before. We here studied the effect of multiple pollen-free/pollen-loaded pollinators on its offspring production for the first time. 1-5 the pollen-free/pollen-loaded pollinators were introduced into syconia of monoecious fig, Ficus racemosa L. 1, 2, 4, 6, and 8 the pollen-loaded/pollen-free pollinators were introduced into syconia of dioecious fig, F. hispida L. The results showed that (1) pollinator larvae could survive for lack of endosperm; (2) the production of pollinator offspring significantly decreased when foundresses were pollen-free; (3) the more pollen-free foundresses entered the syconium, the less their offspring were. The possible explanations were made. First, The pollen brought in by pollinator fertilizes the fig to make the endosperm which is best fit for pollinator reproduction. Second, pollinators already in the cavity might feed on the pollen for reproduction behavior. Third, when no pollen brought in, the pollinators might compete much more seriously each other in the process of oviposition.

Key words: pollen, pollen-free pollinator, pollen-loaded pollinator, foundresses

Speciation in fig wasps parasitoids: a phylogenetic approach (by George D. Weiblen)

George D. Weiblen and Summer I. Silvieus

Department of Plant Biology, University of Minnesota, 250 Biological Sciences Building, 1445 Gortner Avenue, Saint Paul, Minnesota, 55108, USA

ABSTRACT: The relationship between pollinating fig wasps (Agaoninae: Hymenoptera) and Ficus (Moraceae) is a well known obligate mutualism, but nonpollinating Agaonidae are also an integral part of the fig community. Nonpollinating parasitoid wasps and galler wasps depend on the mutualism to complete their life at the expense of figs and fig pollinators. We examined patterns of historical association among three trophic levels in the fig microcosm. Autotrophs included species of Ficus subgenus Sycomorus from New Guinea. Herbivores included Ceratosolen pollinators (Agaoninae: Agaonidae) and Platyneura gallers (Sycophaginae: Agaonidae). Parasitoids included Sycoscapter species attacking pollinators and Apocrypta species attacking gallers.

We compared molecular phylogenies of cytochrome oxidase I for the parasitoid trophic level to host phylogenies of figs, pollinators and gallers to address three questions. What are the host ranges of nonpollinating wasps? Have nonpollinators and hosts cospeciated? If so, is evidence of cospeciation between nonpollinators and their hosts stronger or weaker than for figs and pollinators? Two main ecological predictions were supported by phylogenetic reconciliation analyses. The evolution of galler associations appears less constrained than for pollinators, and parasitoids of gallers evidently have not tracked host phylogeny as closely as the parasitoids of pollinators, who appear closely cospeciated with their host figs. Ecological diversity in the fig microcosm provides a unique opportunity to investigate the impact of life history variation on evolutionary patterns of association ranging from cospeciation to host switching.

Keywords: Apocrypta, Ceratosolen, coevolution, cospeciation, DNA bardcoding, Platyneura, Ficus subgenus Sycomorus, Sycoscapter

Bayesian analysis: new insights into the phylogeny of fig pollinators (by Zi-Feng Jiang)

Zi-Feng Jiang^1,2, Da-Wei HUANG^1,3*, Wen-Quan ZHEN ^1,2

¹Institute of Zoology, Chinese Academy of Sciences, Beijing, 100080, China.

² Graduate School of the Chinese Academy of Sciences, Beijing, 100039, China

³ Plant Protection College, Shandong Agricultural University, Tai’an, Shandong, 271018, China

^*Corresponding author: huangdw@ioz.ac.cn

Abstract: The interaction between figs and fig-pollinating wasps is one of the most species-specific pollination mutualisms. Recently, phylogeny of both partners based on molecular data provided insights into a wide spectrum of coevolutionary questions. However, for the phylogeny of pollinators, there are some discrepancies between different studies and left some relationships unresolved, especially for deep nodes. Bayesian analysis mitochondrial gene cytochrome oxidase subuitI (COI) retrieved from Genbank and our data provided our current estimated phylogeny of fig pollinators. This study also clarified some discrepancies between previous studies. After rooting with Tetrapus, other pollinators fell into two clades. Wiebesia and Blatophaga are at basal positions in respective clade. Ceratosolen is not monophyletic because Kradibia and Liporrhopalum fell inside this group. Three subgenera of Ceratosolen: subgen. Ceratosolen, subgen Rothropus, and subgen Strepitus were not supported. Therefore, Ceratosolen is suggested to be redivided into three groups. Urostigma pollinators (including Dolichoris ) are clustered together. The monophyly of Wiebesia, Blatophaga, Dolichoris, are not supported in this analysis. However, Significant incongruence exists between mitochondrial data and morphology data. Possible reasons for those incongruence are discussed.

Keywords: COI; fig wasp; molecular phylogeny; Bayesian analysis

A Platyneura species of exploiting female syconia in dieocious Ficus auriculata (by Yan-Qiong Peng)

Yan Qiong PENG ^{1, 2} , Da Rong YANG ¹, Qiu Yan WANG ¹

¹ Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences, Kunming, Yunnan 650223, China

²Graduate School of the Chinese Academy of Sciences, Beijing 100039, China. Email : pengyq@xtbg.ac.cn OR yanqiongpeng@hotmail.com

ABSTRACT: Fig trees (Ficus spp) are pollinated by small hymenopteran wasps that develop within the fig. In dioecious species, female wasps enter and pollinate female figs that produce only seeds and within which the wasp is unable to reproduce. Inside a male fig, the fig-pollinating wasp oviposits into flowers that allow the development of larval wasps, and several non-pollinating fig wasps species compete for same pool of flowers in the male figs, which they are there at the expense of the mutualism, either indirectly by draining resources from the tree, or directly by attacking pollen vectors. Non-pollinating fig wasps generally do not attack female syconia. However, Platyneura sp., which is a gall-maker, can oviposit into male and female syconia of Ficus auriculata. This study shows that the population dynamic of Platyneura sp. within female syconia ties up the flowering phenology of F. auriculata. The flowering of Ficus auriculata shows synchrony within male trees and asynchrony within female trees. Each female tree always born syconia around the years, but most of male trees have not syconia during a long time. The peak periods of bearing fruits showed once on the male trees and twice on the female trees, which peak periods do not overlap. The reproduction of Platyneura sp. will shift to female syconia when male trees bear few figs. On the contrary, it shifts to male syconia again with increasing the number of male syconia. When same number of platyneura sp. is introduced to a single female or male syconium in controlled experiment, the offspring numbers of Platyneura sp. within female syconium are notably lower than that of within male syconium. Moreover, the offspring numbers of platyneura sp. increase as mother numbers increase, and the offspring numbers show remarkable difference among five‑, ten‑, and fifteen‑foundress broods within male syconia, but this characteristic is not appeared within female syconia. The results confirmed that Platyneura sp. prefers male syconia rather than female syconia. Its male is able to chew an exit hole through ostiolar bracts within female syconia, but it still depends on the pollinators’ male offspring to create an escape passage out of the mature syconium within male syconia.

Keywords: Non-pollinating fig wasps, fig tree, exploitation, mutualism, flowering phenology

Chemical attraction of fig volatiles to pollinating fig wasps （by Chun Chen）

Chun CHEN, Qi-Shi SONG, Da-Rong YANG

(Kunming Section, Xishuangbanna Tropical Botanic Garden, Chinese Academy of Sciences, Kunming, Yunnan 650223, China)

Abstract: The maintenance of the fig-fig wasp mutualism system is strongly dependent on the chemical orientation of pollinating fig wasps using fig volatiles. The pollinating fig wasps trace the fig volatiles of their particular host and enter the figs at receptive phase (female flower phase) only, where they oviposit for themselves or pollinate for their partners. Most studies focus on the specificity of volatile compounds in different fig species, but few does experiment to examine the attracting role of those chemicals, especially in olfactometer.

Some compounds were used to be found universal in most fig species, such as linalool, benzyl alcohol, geraniol, farnesol, etc. Research on chemical analysis of different fig developing phase showed that volatiles chemicals changed greatly in compound ratios between receptive floral phase and other ones. In Ficus hispida, linalool, geraniol and α-terpeneol have large amount in receptive figs of both male and female trees and decrease as interfloral phase coming. But some compounds was just one main constituent in either of male and female fig volatiles, such as farnesol tested in male receptive figs and β-pinene, limonene, γ-terpene tested in female ones. Benzyl alcohol and Benzeneacetaldehyde both were high constituents of F. hispida, but they were a little lower in receptive figs comparing to interfloral ones. Phenethyl alcohol was just found in male postparasitized fig volatiles. All the results had been published in the paper of Song Q S et al (J. Chem. Ecol., 2001).

The tropical gynodiecious pioneer tree: Ficus hispida and the corresponding pollinating fig wasp Ceratosolen solmsi marchali were the materials of this Research. 11 chemicals, which analyzed by GC-MS from fig volatiles of different phases either male or female Ficus hispida, were chosen for examination. A two-choice olfactometer was used to test the responses of the pollinating fig wasps to 11 fig volatile compounds of 2-3 different concentrations. The trapped wasp masses in the arm treated with volatile compound and in the control arm were recorded respectively.

Different concentrations usually result in changes of significant effects, but these changes hardly ever went to the opposite outcome except for benzyl alcohol, which showed stimulation in high concentration, deterrent in low one and no significant effect in medium one. The chemicals identified from receptive figs all show stimulation at least under one concentration, except for β-pinene. And the responses to most stimuli occur at the level of 100~1000ppm concentrations. Other two chemicals, increasing or appearing in interfloral figs showed expellant, like benzeneacetaldehyde, or had no significant effect, like phenethyl alcohol.

It is suited that the results of bioassays and the situation of chemicals lying in fig floral phases. Whenever it found in male or female fig volatiles, chemicals released in receptive phase showed attraction to pollinators. And chemicals mainly found in interfloral phase most had no significant effect or showed repellant. The results confirm the phenomenon that wasps are specially attracted to host fig trees at the time when figs are ready to be pollinated.

Data were analyzed by the Wilcoxon test, a nonparametric paired test.

Key words: Ficus hispida; Ceratosolen solmsi marchali; chemical attraction; field trap test; two-choice laboratory test.

Labile male morphology and intraspecific male polymorphism in the Philotrypesis fig wasps. (by Simon van Noort)

Emmanuelle Jousselin¹*, Simon van Noort², Jaco M. Greeff¹

¹Department of Genetics University of Pretoria, Pretoria 0002, South Africa

²Natural History Division, South African Museum, Iziko Museums, PO Box 61, Cape Town 8000, South Africa.

^*Current address : Institut National de la Recherche Agronomique,Centre de Biologie et de Gestion des Populations, Campus International de Baillarguet, CS-30 016, 34 988 Montferrier sur Lez, France, fax : 33 4 99 62 33 45: ejousselin@yahoo.com.

Abstract: We investigate the evolution of male morphology in the fig wasps belonging to the genus Philotrypesis (Chalcidoidea, Sycoryctinae). We first reconstruct the phylogenetic relationships of Philotrypesis associated with African figs using nuclear and mitochondrial DNA. We then determine male morphotypes in the species included in our phylogeny and show that intraspecific polymorphism is common. Most species present two types of males and some species have up to three types. These morphotypes are believed to represent alternative mating tactics: some males show morphological adaptations to fighting, others are winged dispersers and others are small sneakers. Mapping out these variations onto our phylogeny reveals that the combination of morphs changes randomly along the branches of the tree. Both parsimony and likelihood approaches indicate that there has been at least one transition from dimorphism to trimorphism, several gains and losses of the small morph and two independent acquisitions of the winged morph. Using maximum likelihood analyses of character evolution, we estimate transition rates for each morph and show that the evolution of each type of morph are not correlated and that forward and backward transition rates are not significantly different. Our results altogether suggest that male morphology is evolutionary labile, it responds quickly to selection imposed by the mating environment. This study, also suggests that seemingly complex phenotypes, such as winged males, can evolve several times and can even be recreated after having been lost.

Key words: discrete characters; evolutionary lability; macroevolution; male polymorphism; mating strategies; phenotypic variation; sexual selection

Ant prevent non-pollinating wasp from ovipositing on figs: implications for the stability of the fig-wasp mutualism (by Da-Rong Yang)

Zuo-dong WEI ^{1, 2}, Da-Rong YANG ¹

1 Kunming Branch, Xishuangbanna Tropical Botanical Garden, the Chinese Academy of Science, Kunming 650223, China.

2 Graduate School of the Chinese Academy of Sciences, Beijing 100039, China. Email: zuodongwei@hotmial.com

Abstract: Ficus racemosa is monoecious fig species which is species-specifically pollinated by Ceratosolen fusciceps. Besides the pollinators there are five species of non-pollinating fig wasps that also breed in the fig, which attack figs from the exterior. In addition, F. racemosa attract suit of ants which attending homopterans on the figs. An exclusion experiment was performed to evaluate whether ants tending homopterans affect the success of oviposition of fig wasp or not. Forging workers of ants attack non-pollinating wasps severely when which oviposits on the fruits. But, they can not interfere with pollinators, which can enter the fig from the ostioles in a few seconds. In absence of ants, offspring of non-pollinating wasps hold 99.83% of whole gall resource; meanwhile, the offspring of pollinators only hold 0.17% of these galls and there are 9.5 seeds in syconium. However, nearly 75 percent of wasps’ offspring is pollinator and there are about 2400 seeds in a syconium when attended by ants; only 25% of galls were hold by the offspring of non-pollinating wasps. The non-pollinators’ offspring depend on the pollinators’ male offspring to create an escape passage out of the mature syconium. So if there are too few male offspring of the pollinators, the offspring of pollinators and non-pollinators will die in the syconium. The stability of fig-wasp mutualism won’t maintain if too many non-pollinators attack figs, that is no ants protect figs from attacking by non-pollinators in Ficus racemosa. Despite this, our results suggest that offspring of non-pollinating wasps can hold whole gall resource in syconium, which will shed light on studies of non-pollinators’ effect on fig-wasp mutualism.

Key words: ant tending, fig-wasp mutualism, fig wasp, stability, pollination

The feeding and oviposition ecology of the non-pollinators in Ficus racemosa

Cheng-Yun YANG ^{1, 2}, Rui-Wu WANG^{1, 2*} and Gui-Fang ZHAO^2*

¹ Kunming Institute of Zoology, Chinese Academy of Sciences. Kunming, Yunnan 650223. China.

² College of Life Science, Northwest University. Xi’an, Shaanxi 710069, China)

Abstract: Ficus (Moraceae) and their species-specific pollinator wasps (Agaonidae) form a remarkable plant-insect obligate mutualism. Each species of figs also shelters a community of non-pollinating chalcids, composed of both gall-makers and parasitoids. One fig species can host up to 30 species of non-pollinating fig wasps. Previous studies mainly dealing with Ficus have long focused on the fig-pollinator relationships alone and have first aimed to understand pollinator dynamics, seed production, pollen dispersal, and stability of fig-pollinator system. However, the non-pollinating fig wasps have received poor attention so far, even though some authors have looked for the impact they have on the fig-pollinator mutualism.

What the impact of the non-pollinators on fig/fig wasp mutualism is mainly determined by the diets of the non-pollinators. Although there existed many reports that the non-pollinators might be inquilines and parasitoids of pollinators or gall-maker, qualitative experiments on the diets of the non-pollinator species are seldom reported and the effect of these non-pollinating fig wasps on mutualistic symbiosis system still opened to questions. In this study, we have tried to determine the diets of the non-pollinating fig wasps in Ficus racemosa and to analysis the exact impact of the non-pollinators on the fig/fig wasp mutualism by the data from the observation and qualitative experiments.

F. racemosa (Sycomorus) is monoecious. It is pollinated by Ceratosolen fusciceps (Agaonidae: Blastophagini) and coexisted with five species of non-pollinator fig wasps. All its non-pollinators have wingless males and winged females. They oviposit through the fig wall, and their offspring depend on the exit tunnel chewed by the pollinator males to leave the syconium. We observed their oviposition process from the phase pre-female to phase male with 22 fruits in one crop, and recorded the wasp number ovipositing on the observed fruits in one minute. In each day, we sampled three times.

0ur data shows that the sequence of the non-pollinating wasps oviposited: P. testacea is the first, then A. sp2, P. mayri, the pollinator, A. westwoodi in turn, and P. agraensis is the last. When the number of P. mayri oviposting increases, the pollinators start to enter the figs. The sustaining days per specie wasps ovipositing also can be got: P. testacea and P. agraensis can persist a week; A. sp2, P. mayri, and A. westwoodi can persist approximately ten days.

According to the ovipositing sequence and length of time, we controlled the fruits with organdies bag and thereby the fruits can be oviposited by different species within the same crop. Only P. testacea and P. mayri of the five non-pollinator species can independently be developed to be adult wasps, but P. testacea, P. mayri, A. sp2 and A. westwoodi can be developed to adult wasps in the un-pollinated fruits (field observation). Considering the ovipositing sequence and the data of the controlled experiments, A. sp2 might be the inquilines or parasitoids of P. testacea, and A. westwoodi might be the inquilines or parasitoids of pollinators or other four species that have oviposited before A. westwoodi. P. agraensis can’t oviposit alone, but there was a significant positive association between P. agraensis and the pollinator, Ceratosolen fusciceps. We also could not find P. agraensis in the unpollinated figs. Therefore we speculated that P. agraensis as seed-predators, or the inquilines/parasitoids of C. fusciceps.

Our work shows that two species might be gall-makers, namely, P. testacea and P. mayri, and A. sp2 and A. westwoodi might be inquilines or parasitoids in the five species of non-pollinating fig wasps in the figs of Ficus racemosa in Xishuangbanna (China), i.e., P. testacea and P. mayri are phytophagous; A. sp2 and A. westwoodi are raptatorial, at least functionally parasites (Kerdelhué and Rasplus 1996). As to P. agraensis, further studies will be done to validate their diets.

Our study indicates that the two species of Apocrypta are not gall-makers but inquilines or parasitoids. It is also noteworthy that in the three species of Platyneura, P. testacea and P. mayri feed on nourishing tissue that they induce within fig ovaries that are unexploited by the pollinators, like Idarnes, but P. agraensis feed on seeds or developing pollinators. It may be different strategies of exploitation based on selective pressure.

Non-pollinating parasitoids and gall-makers have negative impacts on the mutualism through predation of pollinator larvae and through competition with pollinators for seed resources. If host-parasitoid interactions are not stable, theory predicts ever-increasing population cycles leading to the eventual extinction of the parasitoid and its host and, therefore, to the extinction of the mutualism. How can mutualism persist, for organisms must have mechanisms by which they can deter pure exploitation and reinforce mutualistic actions of their partners? Predictions from theory suggest that a cryptic mechanism, such as selective abortion of flowers with heavy egg loads, could stabilize these interactions. Pellmyr and Huth (1994) confirmed such a mechanism exists in the yucca/yucca moth interaction. But in Ficus racemosa, the figs that only host the non-pollinating wasp larvae don’t be aborted. The reason that prevents these figs from abortion is not clear, further studies on non-pollinators may help to facilitate the question to be solved.

Ovipositor length of three Apocrypta species: effect on oviposition behaviors and correlation with syconial thickness

Wen-Quan ZHEN ^1,2,3, Da-Wei HUANG ^1,4*, Jing-Hua XIAO ^1,3, Da-Rong YANG ², Chao-Dong ZHU¹ and Xiao Hui¹

¹ Institute of Zoology, Chinese Academy of Sciences, Beijing, 100080, China.

² Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences, Yunnan, 666303, China

³ Graduate School of the Chinese Academy of Sciences, Beijing, China

⁴ Plant Protection College, Shandong Agricultural University, Taian, Shandong, 271018, China

^*Corresponding author: HUANG Da-Wei. e-mail: huangdw@ioz.ac.cn

Abstract: We investigated oviposition behavior in three non-pollinating fig wasps: Apocrypta bakeri Joseph on Ficus hispida L., A. westwoodi Grandi on F. racemosa L. and A. sp. on F. semicordata Buch.-Ham. For these three sympatric Apocrypta species, their oviposition behaviors are significantly different between one pair of species ( A. bakeri and A. westwoodi) and the other species ( A. sp. on F. semicordata). A. bakeri and A. westwoodi showed similarity in the following aspects: the posture of the abdomen and the action of the hind legs before penetration, and the bending ovipositor sheath during the penetration. In contrast, A. sp. behaves in quite different ways regarding to the oviposition mentioned above. This difference can be interpreted by the significant correlation between the ovipositor length and syconial thickness. Apocrypta sp. has shorter ovipositor than the other two species, which correlates with the thinner thickness of syconial wall of its host fig Ficus semicordata. We can deduce that the ovipositor length adapts to the syconial thickness and induces the oviposition behaviors to diverge. For all three Apocrypta species, the mid-leg length and hindleg length are significantly correlated to their ovipositor lengths. It might be explained by that the body movement adjusting the hind-legs and midlegs up and down, or forward and backward, are also influenced by the ovipositor length.

KEYWORDS: Oviposition behavior, adaptation, Apocrypta, non-pollinating fig wasp.

Oviposition timing: a strategy for non-pollinator fig wasps in same syconium

Wen-Quan ZHEN^1,2, Da-Wei HUANG^1,4,*, Chao-Dong ZHU¹, Hui XIAO¹ and Da-Rong YANG³

¹Institute of Zoology, Chinese Academy of Sciences, Beijing, 100080, China.

² Xishuangbanna Tropical Botanical Garden, Chinese Academy of Sciences, Yunnan, 666303, China

³ Graduate School of the Chinese Academy of Sciences, Beijing, 100039, China

⁴ Plant Protection College, Shandong Agricultural University, Taian, Shandong, 271018, China

^*Corresponding author: huangdw@ioz.ac.cn

ABSTRACT: All fig wasps associated with Ficus racemosa, including three non-pollinating fig wasps of Platyneura spp., demonstrate the oviposition timing. The different wasps sequentially lay their eggs in the syconia at different stages of fig development. The ovipositor lengths of three Platyneura spp are significantly different. With the shortest ovipositor, P. testacea obviously oviposits in small figs. It has been observed to appear on the fig at around one week earlier than the pollinators Ceratosolen fusciceps. P. mayri has a longer ovipositor and oviposits on the syconia unpollinated, being pollinated or pollinated. However, it prefers the pollinated figs to the unpollinated figs. With the longest ovipositor, P. agraensis lays eggs at the fig Phase C in which all fig ovaries are fully developed into galls or seeds. The oviposition timing of Platyneura species is resulted from the thickness of syconial wall and the thickness of ovary layer. Basing on the oviposition timing and ovipositor lengths of fig wasps, as well as the syconial traits, the possible feeding styles of fig wasps could be deduced. P. testacea and P. mayri are gall makers, while P. agraensis is either parasitoid or inquiline of other fig wasps. Ovipositing in different phases of the syconia is a strategy for the NPFWs to reduce the competition among them and to avoid the predation by the ants. This strategy may make sense for multiple species of fig wasps co-existing in the fig microhabitat.

Keywords: Ficus racemosa L, morphological constraint, Platyneura,

Scientific Program

Symposium time: 8:00-18:00

Tea time: 10:00-10:30 15:30-16:00;

Breakfast: 7:00;

Lunch: 12:30-13:30;

Dinner: 18:30-20:00

SEPTEMBER 15

All day registration

18:00 Reception

SEPTEMBER 16

Morning (Chairperson: George D Weiblen)

8:00-8:05 Welcome and brief introduction by Da-Wei Huang

8:05-9:05 The evolutionary history of the fig wasp mutualism: origins and beyond the cospeciation paradigm (by Carlos A. Machado)

9:05-10:05 Co-relationship between pollinator entry and ostiole close mechanism: a case study of F. microcarpa ― Eupristina verticillata mutualism (by Tong-Xin Zhang)

Tea time: 10:05-10:30

10:30-11:30 The trade-off in fig/fig wasp mutualism and the reciprocal system maintenance dynamic (by Rui-Wu Wang)

11:30-12:30 Patterns of diversification of afrotropical otiteselline fig wasps: evolution of host use and ecological niches (by Emmanuelle Jousselin)

Lunch: 12:30-13:30

Afternoon (Chairperson: Da-rong Yang)

13:30-14:30 Figs and the diversity of tropical rain forests (by Rhett D. Harrison)

14:30-15:30 The fig wasps in syconia of Ficus racemosa (by Lei Xu)

Tea time: 15:30-16:00

16:00-17:00 Interaction between nonpollinators and pollinator mutualism in the Ficus hispida (by Da-Rong Yang)

17:00-18:00 More pollen-free pollinators, less their offspring: pollen effect on pollinator reproduction (by Wen-Quan Zhen)

Dinner: 18:30-20:30

SEPTEMBER 17

Morning (Chairperson: Simon van Noort)

8:00-9:00 Speciation in fig wasps parasitoids: a phylogenetic approach (by George D. Weiblen)

9:00-10:00 Bayesian analysis: new insights into the phylogeny of fig pollinators (by Zi-Feng Jiang)

Tea time: 10:00-10:30

10:30-11:30 A Platyneura species of exploiting female syconia in dieocious Ficus auriculata (by Yan-Qiong Peng)

11:30-12:30 Chemical attraction of fig volatiles to pollinating fig wasps（by Chun Chen）

Lunch: 12:30-13:30

Afternoon (Chairperson: Emmanuelle Jousselin)

13:30-14:30 Labile male morphology and intraspecific male polymorphism in the Philotrypesis fig wasps (by Simon van Noort)

14:30-15:30 Ant prevent non-pollinating wasp from ovipositing on figs: implications for the stability of the fig-wasp mutualism (by Da-Rong Yang)

Tea time: 15:30-16:00

16:00-16:30 Chinese fig wasps on the way into the international fig cavity (by Da-Wei Huang)

16:30-17:30 Round Table Discussion

17:30-17:40 Closing remarks by Carlos A. Machado

18:30-20:30 Banquet

20:30 Chinese representatives leave for ZhongGuanCun

SEPTEMBER 18

Sightseeing: Great Wall, Forbidden City

SEPTEMBER 19

8:00 Move to the Guest House, Chinese Academy of Sciences

9:00 Visiting Da-Wei’s Lab in Institute of Zoology

12:30 Banquet with Da-Wei’s group

PARTICIPANTS LIST

1. Chen, Chun, Xishuangbanna Tropical Botanical Garden, China Academy of Sciences. Kunming 655023, China. E-mail: chenchun@xtbg.ac.cn.

2. Dang, Xiao-Dong, Beijing Normal University.

3. Ge, Peng, Scientific Instrument Software Company, Chinese Academy of Sciences. E-mail: gepeng@publicb.bta.net.cn. Phone: 13801237876

4. Harrison, Rhett D., Smithsonian Tropical Research Institute. Tupper Building, Unit 0948 APO, AA 34002 USA. E-mail: harrisonr@tivoli.si.edu.

5. Huang, Da-Wei, Institute of Zoology, Chinese Academy of Sciences. Beijing, 100080, China. E-mail: huangdw@ioz.ac.cn. Phone: 13910256670

6. Jiang, Zi-Feng, Institute of Zoology, Chinese Academy of Sciences. Beijing, 100080, China. E-mail: jiangzf@ioz.ac.cn.

7. Jousselin, Emmanuelle, Department of Genetics University of Pretoria. Pretoria 0002, South Africa. E-mail: jousseli@ensam.inra.fr.

8. Lei, Xu, Xishuangbanna Tropical Botanical Garden, China Academy of Sciences. Kunming 655023, China. E-mail: ziboxul@sohu.com.

9. Liu, Da-Jun, Institute of Zoology, Chinese Academy of Sciences. Beijing, 100080, China.

10. Machado, Carlos A., The University of Arizona. E-mail: cmachado@email.arizona.edu.

11. Peng, Yan-Qiong, Xishuangbanna Tropical Botanical Garden, China Academy of Sciences. Kunming 655023, China. E-mail: pengyq@xtbg.ac.cn.

12. van Noort, Simon, Natural History Division, South African Museum. Iziko Museums, PO Box 61, Cape Town 8000, South Africa. E-mail: svannoort@iziko.org.za.

13. Wang, Rui-Wu, Kunming Institute of Zoology, Chinese Academy of Sciences. Kunming, Yunnan 650223. China.. E-mail: ruiwukiz@hotmail.com. Phone: 13013300156

14. Weiblen, George D, Bell Museum of Natural History, University of Minnesota. 250 Biological Sciences, 1445 Gortner Avenue, Saint Paul, MN 55108, USA. E-mail: gweiblen@umn.edu.

15. Xiao, Hui, Institute of Zoology, Chinese Academy of Sciences. Beijing, 100080, China. E-mail: xiaoh@ioz.ac.cn

16. Xiao, Jin-Hua, Institute of Zoology, Chinese Academy of Sciences. Beijing, 100080, China. E-mail: xiaojh@ioz.ac.cn.

17. Yang, Da-Rong, Xishuangbanna Tropical Botanical Garden, China Academy of Sciences. Kunming 655023, China. E-mail: yangdr@xtbg.ac.cn. Phone: 13700668786

18. Zhang Tong-Xin, Institute of Zoology, Chinese Academy of Sciences. Beijing, 100080, China. E-mail: zhangtx@ioz.ac.cn.

19. Zhang, Da-Yong, Beijing Normal University. E-mail: zhangdy@bnu.edu.cn. Phone: 13683662799

20. Zhang, Yan-Zhou, Institute of Zoology, Chinese Academy of Sciences. Beijing, 100080, China. E-mail: zhangyz@ioz.ac.cn.

21. Zhen, Wen-Quan, Institute of Zoology, Chinese Academy of Sciences. Beijing, 100080, China.