Digest: Process‐based phylogenetic models provide unique insights into the evolution of mutualistic networks*

Figs and their pollinating fig wasps are a classic example of long‐term obligate associations between different species. Satler et al. use a process‐based model adopted from molecular evolution to identify the major processes that affect cophylogenetic matching between figs and fig wasps. They find that host‐switching is one of the most important evolutionary processes contributing to current cophylogenetic patterns, illustrating the value of probabilistic approaches to studying the evolutionary history of mutualisms.     

Genome‐wide sequence data suggest the possibility of pollinator sharing by host shift in dioecious figs (Moraceae,Ficus)

The obligate mutualism of figs and fig‐pollinating wasps has been one of the classic models used for testing theories of co‐evolution and cospeciation due to the high species‐specificity of these relationships. To investigate the species‐specificity between figs and fig pollinators and to further understand the speciation process in obligate mutualisms, we examined the genetic differentiation and phylogenetic relationships of four closely related fig‐pollinating wasp species (Blastophaga nipponica, Blastophaga taiwanensis, Blastophaga tannoensis and Blastophaga yeni) in Japan and Taiwan using genome‐wide sequence data, including mitochondrial DNA sequences. In addition, population structure was analysed for the fig wasps and their host species using microsatellite data. The results suggest that the three Taiwanese fig wasp species are a single panmictic population that pollinates three dioecious fig species, which are sympatrically distributed, have large differences in morphology and ecology and are also genetically differentiated. Our results illustrate the first case of pollinator sharing by host shift in the subgenus Ficus. On the other hand, there are strict genetic codivergences between allopatric populations of the two host–pollinator pairs. The possible processes that produce these pollinator‐sharing events are discussed based on the level and pattern of genetic differentiation in these figs and fig wasps.     

Molecular phylogenies of fig wasps: partial cocladogenesis of pollinators and parasites

Figs (Ficus spp., Moraceae) and their pollinating wasps form an obligate mutualism, which has long been considered a classic case of coevolution and cospeciation. Figs are also exploited by several clades of nonpollinating wasps, which are parasites of the mutualism and whose patterns of speciation have received little attention. We used data from nuclear and mitochondrial DNA regions to estimate the phylogenies of 20 species of Pleistodontes pollinating wasps and 16 species of Sycoscapter nonpollinating wasps associated with Ficus species in the section Malvanthera. We compare the phylogenies of 15 matched Pleistodontes/Sycoscapter species pairs and show that the level of cospeciation is significantly greater than that expected by chance. Our estimates of the maximum level of cospeciation (50 to 64% of nodes) are very similar to those obtained in other recent studies of coevolved parasitic and mutualistic associations. However, we also show that there is not perfect congruence of pollinator and parasite phylogenies (for any substantial clade) and argue that host plant switching is likely to be less constrained for Sycoscapter parasites than for Pleistodontes pollinators. There is perfect correspondence between two terminal clades of two sister species in the respective phylogenies, and rates of molecular evolution in these pairs are similar.     

Cophylogenetic assessment of New World warblers (Parulidae) and their symbiotic feather mites (Proctophyllodidae)

Host–symbiont relationships are ubiquitous in nature, yet evolutionary and ecological processes that shape these intricate associations are often poorly understood. All orders of birds engage in symbioses with feather mites, which are ectosymbiotic arthropods that spend their entire life on hosts. Due to their permanent obligatory association with hosts, limited dispersal and primarily vertical transmission, we hypothesized that the cospeciation between feather mites and hosts within one avian family (Parulidae) would be perfect (strict cospeciation). We assessed cophylogenetic patterns and tested for congruence between species in two confamiliar feather mite genera (Proctophyllodidae: Proctophyllodes, Amerodectes) found on 13 species of migratory warblers (and one other closely related migratory species) in the eastern United States. Based on COI sequence data, we found three Proctophyllodes lineages and six Amerodectes lineages. Distance‐ and event‐based cophylogenetic analyses suggested different cophylogenetic trajectories of the two mite genera, and although some associations were significant, there was little overall evidence supporting strict cospeciation. Host switching is likely responsible for incongruent phylogenies. In one case, we documented prairie warblers Setophaga discolor harboring two mite species of the same genus. Most interestingly, we found strong evidence that host ecology may influence the likelihood of host switching occurring. For example, we documented relatively distantly related ground‐nesting hosts (ovenbird Seiurus aurocapilla and Kentucky warbler Geothlypis formosa) sharing a single mite species, while other birds are shrub/canopy or cavity nesters. Overall, our results suggest that cospeciation is not the case for feather mites and parulid hosts at this fine phylogenetic scale, and raise the question if cospeciation applies for other symbiotic systems involving hosts that have complex life histories. We also provide preliminary evidence that incorporating host ecological traits into cophylogenetic analyses may be useful for understanding how symbiotic systems have evolved.     

Should I stay or should I go? Pollinator shifts rather than cospeciation dominate the evolutionary history of South African Rediviva bees and their Diascia host plants

Plant–pollinator interactions are often highly specialised, which may be a consequence of co‐evolution. Yet when plants and pollinators co‐evolve, it is not clear if this will also result in frequent cospeciation. Here, we investigate the mutual evolutionary history of South African oil‐collecting Rediviva bees and their Diascia host plants, in which the elongated forelegs of female Rediviva have been suggested to coevolve with the oil‐producing spurs of their Diascia hosts. After controlling for phylogenetic nonindependence, we found Rediviva foreleg length to be significantly correlated with Diascia spur length, suggestive of co‐evolution. However, as trait correlation could also be due to pollinator shifts, we tested if cospeciation or pollinator shifts have dominated the evolution of Rediviva–Diascia interactions by analysing phylogenies in a cophylogenetic framework. Distance‐based cophylogenetic analyses (PARAFIT, PACo) indicated significant congruence of the two phylogenies under most conditions. Yet, we found that phylogenetic relatedness was correlated with ecological similarity (the spectrum of partners that each taxon interacted with) only for Diascia but not for Rediviva, suggesting that phylogenetic congruence might be due to phylogenetic tracking by Diascia of Rediviva rather than strict (reciprocal) co‐evolution. Furthermore, event‐based reconciliation using a parsimony approach (CORE‐PA) on average revealed only 11–13 cospeciation events but 58–80 pollinator shifts. Probabilistic cophylogenetic analyses (COALA) supported this trend (8–29 cospeciations vs. 40 pollinator shifts). Our study suggests that diversification of Diascia has been largely driven by Rediviva (phylogenetic tracking, pollinator shifts) but not vice versa. Moreover, our data suggest that, even in co‐evolving mutualisms, cospeciation events might occur only infrequently.     

Cophylogenetic patterns are uncorrelated between two lineages of parasites on the same hosts

Free‐living organisms are often host to multiple lineages of closely related parasites. Different lineages of obligate parasites living on the same hosts might potentially be expected to display similar cophylogenetic patterns. However, there are also reasons why these lineages might have different evolutionary histories (e.g. host switching, host geography). In the present study, we use mitochondrial and nuclear DNA sequence data to evaluate the cophylogenetic patterns between doves and their wing and body lice. Previous studies have found that the wing and body lice of doves have different levels of congruence between their phylogenetic histories. However, these studies are limited in scope, either taxonomically or geographically. We used both new and existing data to generate a worldwide and taxonomically diverse data set for doves and two independent groups of lice: wing and body lice. Using event and topology‐based methods, we found that cophylogenetic patterns were not correlated between wing and body lice, even though both groups showed evidence of cospeciation with their hosts. These results indicate that external factors vary in their impact on different groups of parasites and also that broad sampling is critical for identifying patterns in cophylogenetic analyses.     

Pollinator sharing in dioecious figs (Ficus: Moraceae)

As one of the most specialized pollination syndromes, the fig (Ficus)–fig wasp (Agaonidae) mutualism can shed light on how pollinator behaviour and specificity affect plant diversification through processes such as reproductive isolation and hybridization. Pollinator sharing among species has important implications for Ficus species delimitation and the evolutionary history of the mutualism. Although agaonid wasp pollinators are known to visit more than one host species in monoecious figs, pollinator sharing has yet to be documented in dioecious figs. The present study investigated the frequency of pollinator sharing among sympatric, closely‐related dioecious figs in Ficus sections Sycocarpus and Sycidium. Molecular and morphological species identification established the associations between pollinating agaonid wasp species and host fig species. Cytochrome oxidase I was sequenced from 372 Ceratosolen pollinators of Ficus section Sycocarpus and 210 Kradibia pollinators of Ficus section Sycidium. The association between fig species and morphologically distinct clades of pollinator haplotypes was predominantly one‐to‐one. In Ceratosolen, six of 372 pollinators (1.5%) visited fig species other than the predominant host. No pollinator sharing was detected between the two Sycidium host species, although a rare hybrid shared Kradibia pollinators with both parental species. These findings point to low rates of pollinator sharing among closely‐related dioecious fig species in sympatry, and perhaps lower rates than among monoecious figs. Such rare events could be evolutionarily important as mechanisms for gene flow among fig species. Differences in rates of pollinator sharing among fig lineages might explain the conflicting phylogenetic patterns inferred among monoecious figs, dioecious figs, and their respective pollinators. © 2011 The Linnean Society of London, Biological Journal of the Linnean Society, 2011, 103 , 546–558.     

Can chemical signals,responsible for mutualistic partner encounter,promote the specific exploitation of nursery pollination mutualisms? – The case of figs and fig wasps

In nursery pollination mutualisms, where pollinators reproduce within the inflorescence they pollinate, floral scents often play a major role in advertizing host location and rewards for the pollinator. However, chemical messages emitted by the plant that are responsible for the encounter of mutualist partners can also be used by parasites of these mutualisms to locate their host. Each species of Ficus (Moraceae) is involved in an obligatory nursery pollination mutualism with usually one pollinating fig wasp (Hymenoptera: Chalcidoidea: Agaonidae). In this interaction, volatile compounds emitted by receptive figs are responsible for the attraction of their specific pollinator. However, a large and diverse community of non-pollinating chalcidoid wasps can also parasitize this mutualism. We investigated whether the chemical message emitted by figs to attract their pollinator can promote the host specificity of non-pollinating fig wasps. We analysed the volatile compounds emitted by receptive figs of three sympatric Ficus species, namely, Ficus hispida L., Ficus racemosa L., and Ficus tinctoria G. Forster, and tested the attraction of the pollinator of F. hispida ( Ceratosolen solmsi marchali Mayr), and of one species of non-pollinating fig wasp [ Philotrypesis pilosa Mayr (Hymenoptera: Chalcidoidea: Pteromalidae)] to scents emitted by receptive figs of these three Ficus species. Analysis of the volatile compounds emitted by receptive figs revealed that the three Ficus species could be clearly distinguished by their chemical composition. Behavioural bioassays performed in a Y-tube olfactometer showed that both pollinator and parasite were attracted only by the specific odour of F. hispida . These results suggest that the use by non-pollinating fig wasps of a specific chemical message produced by figs could limit host shifts by non-pollinating fig wasps.     

Molecular phylogeny of fig wasp pollinators (Agaonidae, Hymenoptera) of Ficus section Galoglychia

The obligate mutualism between fig trees and their fig wasp pollinators, together with the general tendency for each host species to be pollinated by one fig wasp species, led to the hypothesis that these two lineages have cospeciated. The pollinators of African figs of section Galoglychia form a diverse group of genera whose species seem to be less constrained to a specific host than other pollinating fig wasp genera. Various authors have suggested remarkably different phylogenetic relationships between the seven genera associated with section Galoglychia. These uncertainties concerning the classification make it difficult to understand the historical patterns of association between these wasps and their hosts. The phylogenetic tree for the pollinators was reconstructed with 28S, COI and ITS2 DNA sequence data and compared with morphological classification of the hosts. Pollinator genera were monophyletic in all analyses. However, the relative position of some genera remains unresolved. Investigation of host−fig association suggests that there have been frequent host jumps between host subsections. This indicates that cospeciation between fig trees and fig wasps is not as stringent as previously assumed. In addition, pollinators of the genus Alfonsiella associated with three host figs (Ficus craterostoma, F. stuhlmannii and F. petersii) are morphologically very similar in South Africa. We investigated the possibility that these pollinators form a complex of species with host‐based genetic differentiation. Molecular analyses supported the distinction of the pollinator of F. craterostoma as a good species, but the pollinators of F. stuhlmannii and F. petersii clustered within the same clade, suggesting that these two host species share a single pollinator, Alfonsiella binghami. Based on both molecular data and morphological re‐evaluation, a new Alfonsiella species is described, Alfonsiella pipithiensis sp. nov., which is the pollinator of F. craterostoma in southern Africa. A key to both females and males of all described species of Alfonsiella is provided.     

One fig to bind them all: host conservatism in a fig wasp community unraveled by cospeciation analyses among pollinating and nonpollinating fig wasps

The study of chalcid wasps that live within syconia of fig trees (Moraceae, Ficus ), provides a unique opportunity to investigate the evolution of specialized communities of insects. By conducting cospeciation analyses between figs of section Galoglychia and some of their associated fig wasps, we show that, although host switches and duplication have evidently played a role in the construction of the current associations, the global picture is one of significant cospeciation throughout the evolution of these communities. Contrary to common belief, nonpollinating wasps are at least as constrained as pollinators by their host association in their diversification in this section. By adapting a randomization test in a supertree context, we further confirm that wasp phylogenies are significantly congruent with each other, and build a "wasp community" supertree that retrieves Galoglychia taxonomic subdivisions. Altogether, these results probably reflect wasp host specialization but also, to some extent, they might indicate that niche saturation within the fig prevents recurrent intrahost speciation and host switching. Finally, a comparison of ITS2 sequence divergence of cospeciating pairs of wasps suggests that the diversification of some pollinating and nonpollinating wasps of Galoglychia figs has been synchronous but that pollinating wasps exhibit a higher rate of molecular evolution.     

Parasitism of a pollinator fig wasp: mortalities are higher in figs with more pollinators,but are not related to local densities of figs

相似文献   

Seasonal change in the structure of fig-wasp community and its implication for conservation

Figs (Moraceae) and their pollinating wasps (Agaonidae) constitute a famous reciprocal mutualism in which figs provide some female flowers for the development of fig wasp offspring while the fig wasps pollinate fig flowers. However, figs also host many non-pollinating wasps which are either parasitoids or resource competitors of pollinators, and bring no benefit for figs and are detrimental to fig’ fitness. Our data onFicus racemosa in Xishuangbanna showed that the numbers of non-pollinators and the mature syconia without pollinator wasps increase in rainy season, especially in the highly fragmented forest. This might be because of the longer developing time of the syconia and thereby longer oviposition time to non-pollinators in the dry season. The galled flower and the viable seed percentages in dry seasons are also larger than in rainy seasons in both primary forest and fragmented forest, and the development of non-pollinators is mainly at the expense of pollinator wasps. Our results showed that there exists a discriminative seasonal impact of non-pollinators and fragmentation effects on population size of fig’s pollinators. This implies that fig/fig wasp mutualism is more fragile in dry season, and that the critical population size and breeding units of figs in seasonal area might be larger than previously estimated without considering the seasonal change of pollinator population.     

Pollination and parasitism in functionally dioecious figs

Fig wasps (Agaonidae: Hymenoptera) are seed predators and their interactions with Ficus species (Moraceae) range from mutualism to parasitism. Recently considerable attention has been paid to conflicts of interest between the mutualists and how they are resolved in monoecious fig species. However, despite the fact that different conflicts can arise, little is known about the factors that influence the persistence of the mutualism in functionally dioecious Ficus. We studied the fig pollinator mutualism in 14 functionally dioecious fig species and one monoecious species from tropical lowland rainforests near Madang, Papua New Guinea. Observations and experiments suggest that (i) pollinating wasps are monophagous and attracted to a particular host species; (ii) pollinating and non-pollinating wasps are equally attracted to gall (male) figs and seed (female) figs in functionally dioecious species; (iii) differing style lengths between gall figs and seed figs may explain why pollinators do not develop in the latter; (iv) negative density dependence may stabilize the interaction between pollinating wasps and their parasitoids; and (v) seed figs may reduce the search efficiency of non-pollinators. This increased pollinator production without a corresponding decrease in seed production could provide an advantage for dioecy in conditions where pollinators are limiting.     

Interactions between pollinator and non‐pollinator fig wasps: correlations between their numbers can be misleading

Ficus and their species–specific pollinator fig wasps represent an obligate plant–insect mutualism, but figs also support a community of non‐pollinating fig wasps (NPFWs) that consist of phytophages and parasitoids or inquilines. We studied interactions between Kradibia tentacularis, the pollinator of a dioecious fig tree species Ficus montana, and an undescribed NPFW Sycoscapter sp. Members of Sycoscapter sp. oviposited 2–4 weeks after pollinator oviposition, when host larvae were present in the figs. No negative correlation was found between the numbers of the two wasp species emerging from figs in a semi‐natural population. However, in experiments where the numbers of pollinator foundresses entering a fig were controlled, Sycoscapter sp. significantly reduced the numbers of pollinator offspring. Consequently, it can be concluded that Sycoscapter sp. is a parasitoid of K. tentacularis (which may also feed on plant tissue). Sycoscapter females concentrate their oviposition in figs that contain more potential hosts, rendering invalid conclusions based on simple correlations of host and natural enemy numbers.     

Shift to mutualism in parasitic lineages of the fig/fig wasp interaction

The interaction between Ficus and their pollinating wasps (Chalcidoidea, Agaonidae) represents a striking example of mutualism. Figs also host numerous non-pollinating wasps belonging to other chalcidoid families. We show that six species of Ficus that are passively pollinated by the agaonid genus Waterstoniella also host specific wasps belonging to the chalcidoid genera Diaziella (Sycoecinae) and Lipothymus (Otitesellinae). Both belong to lineages that are considered as parasites of the fig/fig wasp mutualism. We show that these wasps are efficient pollinators of their hosts. Pollen counts on wasps of a species of Diaziella hosted by Ficus paracamptophylla show that Diaziella sp. transports more pollen than the associated pollinator when emerging from its natal fig. Further, the number of pollinated flowers in receptive figs is best explained by the number of Diaziella plus the number of Waterstoniella that had entered it. Figs that were colonised by Diaziella always produced seeds: Diaziella does not overexploit its host. Similarly, figs of Ficus consociata that were colonised solely by a species of Lipothymus produced as many seeds as figs that were colonised only by the legitimate pollinator Waterstoniella malayana . Diaziella sp. and Lipothymus sp. seem to pollinate their host fig as efficiently as do the associated agaonid wasps. Previous studies, on actively pollinated Ficus species, have found that internally ovipositing non-agaonid wasps are parasites of such Ficus species. Hence, mode of pollination of the legitimate pollinator conditions the outcome of the interaction between internally ovipositing parasites and their host.     

Are nematodes costly to fig tree–fig wasp mutualists?

Most mutualisms are exploited by parasites, which must strike an evolutionary balance between virulence and long‐term persistence. Fig‐associated nematodes, living inside figs and dispersed by fig wasps, are thought to be exploiters of the fig–fig wasp mutualism. The life history of nematodes is synchronized with the fig development and adapted to particular developmental characteristics of figs. We expect host breeding systems (monoecious vs. gynodioecious figs) and seasonality to be central to this adaptation. However, the details of the adaptation are largely unknown. Here, we conducted the first field surveys on the prevalence of nematodes from monoecious Ficus microcarpa L.f. (Moraceae), gynodioecious Ficus hispida L.f., and their pollinating fig wasps in two seasons and two developmental stages of figs in Xishuangbanna, China. We followed this up by quantifying the effects of nematodes on fitness‐related traits on fig wasps (e.g., egg loads, pollen grains, and longevity) and fig trees (seed production) in gynodioecious F. hispida. The magnitude of nematode infection was compared between pre‐ and post‐dispersal pollinators to quantify the probability of nematodes being transported to new hosts. Our results showed that Ficophagus microcarpus (Nematoda: Aphelenchoididae) was the only nematode in F. microcarpa. In F. hispida, Martininema guangzhouensis (Nematoda: Aphelenchoididae) was the dominant nematode species, whereas Ficophagus centerae was rare. For both species of Ficus, rainy season and inter‐floral figs had higher rates of nematode infection than the dry‐hot season and receptive figs. Nematodes did not affect the number of pollen grains or egg loads of female wasps. We did not detect a correlation between seed production and nematode infection. However, carrying nematodes reduced the lifespan and dispersal ability of pollinator wasps, indicating higher rates of post‐emergence mortality in infected fig wasps. Severely infected fig wasps were likely ‘filtered out’, preventing the overexploitation of figs by wasps and stabilizing the interaction over evolutionary time.     

Complementary fruiting phenologies facilitate sharing of one pollinator fig wasp by two fig trees

Aims Most pollinator fig wasps are host plant specific, with each species only breeding in the figs of one fig tree species, but increasing numbers of species are known to be pollinated by more than one fig wasp, and in rare instances host switching can result in Ficus species sharing pollinators. In this study, we examined factors facilitating observed host switching at Xishuangbanna in Southwestern (SW) China, where Ficus squamosa is at the northern edge of its range and lacks the fig wasps that pollinate it elsewhere, and its figs are colonized by a Ceratosolen pollinator that routinely breeds in figs of F. heterostyla .Methods We recorded the habitat preferences of F. squamosa and F. heterostyla at Xishuangbanna, and compared characteristics such as fig size, location and colour at receptive phase. Furthermore, the vegetative and reproductive phenologies in the populations of F. squamosa and F. heterostyla were recorded weekly at Xishuangbanna Tropical Botanical Garden for 1 year.Important findings Ficus squamosa is a shrub found near fast-flowing rivers, F. heterostyla is a small tree of disturbed forest edges. Although preferring different habitats, they can be found growing close together. Both species have figs located at or near ground level, but they differ in size when pollinated. Fig production in F. squamosa was concentrated in the colder months. F. heterostyla produced more figs in summer but had some throughout the year. The absence of its normal pollinators, in combination with similarly located figs and partially complementary fruiting patterns appear to have facilitated colonization of F. squamosa by the routine pollinator of F. heterostyla. The figs probably also share similar attractant volatiles. This host switching suggests one mechanism whereby fig trees can acquire new pollinators and emphasizes the likely significance of edges of ranges in the genesis of novel fig tree–fig wasp relationships.     

Culture-Free Survey Reveals Diverse and Distinctive Fungal Communities Associated with Developing Figs (Ficus spp.) in Panama

The ancient association of figs (Ficus spp.) and their pollinating wasps (fig wasps; Chalcidoidea, Hymenoptera) is one of the most interdependent plant–insect mutualisms known. In addition to pollinating wasps, a diverse community of organisms develops within the microcosm of the fig inflorescence and fruit. To better understand the multipartite context of the fig–fig wasp association, we used a culture-free approach to examine fungal communities associated with syconia of six species of Ficus and their pollinating wasps in lowland Panama. Diverse fungi were recovered from surface-sterilized flowers of all Ficus species, including gall- and seed flowers at four developmental stages. Fungal communities in syconia and on pollinating wasps were similar, dominated by diverse and previously unknown Saccharomycotina, and distinct from leaf- and stem endophyte communities in the same region. Before pollination, fungal communities were similar between gall- and seed flowers and among Ficus species. However, fungal communities differed significantly in flowers after pollination vs. before pollination, and between anciently diverged lineages of Ficus with active vs. passive pollination syndromes. Within groups of relatively closely related figs, there was little evidence for strict-sense host specificity between figs and particular fungal species. Instead, mixing of fungal communities among related figs, coupled with evidence for possible transfer by pollinating wasps, is consistent with recent suggestions of pollinator mixing within syconia. In turn, changes in fungal communities during fig development and ripening suggest an unexplored role of yeasts in the context of the fig–pollinator wasp mutualism.     

Pollinators entering female dioecious figs: why commit suicide?

In the dioecious fig/pollinator mutualism, the female wasps that pollinate figs on female trees die without reproducing, whereas wasps that pollinate figs on male trees produce offspring. Selection should strongly favour wasps that avoid female figs and enter only male figs. Consequently, fig trees would not be pollinated and fig seed production would ultimately cease, leading to extinction of both wasp and fig. We experimentally presented pollinators in the wild (southern India) with a choice between male and female figs of a dioecious fig species, Ficus hispida L. Our results show that wasps do not systematically discriminate between sexes of F. hispida. We propose four hypotheses to explain why wasp choice has not evolved, and how a mutualism is thus maintained in which all wasps that pollinate female figs have zero fitness.     

Parasites and mutualism function: measuring enemy‐free space in a fig–pollinator symbiosis

Mutualisms involve cooperation between species and underpin several ecosystem functions. However, there is also conflict between mutualists, because their interests are not perfectly aligned. In addition, most mutualisms are exploited by parasites. Here, we study the interplay between cooperation, conflict and parasitism in the mutualism between fig trees and their pollinator wasps. Conflict occurs because each fig ovary can nurture either one seed or one pollinator offspring and, while fig trees benefit directly from seeds and pollinator offspring (pollen vectors), pollinators only benefit directly from pollinator offspring. The mechanism(s) of conflict resolution is debated, but must explain the widespread observation that pollinators develop in inner, and seeds in outer, layers of fig flowers. We recently suggested a role for non‐pollinating figs wasps (NPFWs) that are natural enemies or competitors of the pollinators and lay their eggs through the fig wall. Most NPFW offspring develop in outer and middle layer flowers, suggesting that inner flowers provide enemy‐free space for pollinator offspring. Here, we test the hypothesis that NPFWs cannot reach inner flowers, by measuring wasp and fig morphology at the species‐specific times of NPFW attack in the field. We found that three species of Sycoscapter and Philotrypesis wasps that parasitise pollinators could reach 34–73%, 75–92% and 82–97% of fig ovaries, respectively. Meanwhile, Eukobelea and Pseudidarnes gall‐formers, despite having shorter ovipositors, can access almost all fig flowers (93–99% and 100%), because they attack smaller (younger) fig fruits. Our mechanistic results from ovipositing wasps support spatial patterns of wasp offspring segregation within figs to suggest that inner ovules provide enemy‐free‐space for pollinators. This may contribute to mutualism stability by helping select for pollinators to avoid laying eggs where they are likely to be parasitised. These outer flowers then remain free to develop as seeds, promoting mutualism persistence.