Utilization of an unpredictable food source by Melanterius ventralis, a seed-feeding biological control agent of Acacia longifolia in South Africa

A seed-feeding weevil, Melanterius ventralis (Coleoptera: Curculionidae), has been introduced into South Africa to supplement a gall wasp,Trichilogaster acaciaelongifoliae (Hymenoptera: Pteromalidae), in the biological control programme against an alien invasive tree, Acacia longifolia (Mimosaceae), from Australia. The gall wasp debilitates most of the flower buds on A. longifolia andreduces seed-set by >95%. The intended rolefor M. ventralis is to destroy theresidual seeds. To achieve this, the gravidfemales need to be able to locate a food sourcethat is both heterogeneously dispersed andfrequently scarce due to damage caused by T.acaciaelongifoliae. Observations showedthat M. ventralis females are meticulousin choosing sites to oviposit so that larvae donot become overcrowded and food limited. Cagetests and field observations revealed thatfemales located pods regardless of density andposition, and that the duration of time spenton branches was proportional to the number ofpods on the branches and to the condition ofpods. The females spent little time onbranches of acacia species other than A.longifolia, but some time was spent onbranches with pods of A. melanoxylon thathad been coated with juice extracted from A.longifolia. All indications are that M.ventralis has the attributes needed to bean excellent supplementary biological controlagent to T. acaciaelongifoliae and theprogress of the weevil continues to be monitored.     

Why a gall former can be a good biocontrol agent: the gall wasp Trichilogaster acaciaelongifoliae and the weed Acacia longifolia

ABSTRACT.

• 1 The pteromalid Trichilogaster acaciaelongifoliae (Froggatt) causes galls to develop in the place of inflorescences during the reproductive phase of its host Acacia longifolia (Andr.) Willd. (Fabaceae). After being introduced to South Africa during 1982 and 1983, this wasp reduced the reproductive potential of A. longifolia by >89% when only 50% of the branches on a tree were galled. In addition, vegetative growth of galled branches was reduced by 53% when the vacated galls had desiccated.
• 2 The wasp is able to manipulate its host so effectively because: (a) the dry mass of developing galls was significantly greater than that of the corresponding reproductive organs, peaking in spring (September) when galls were 25 times the mass of unfertilized inflorescences. (b) The large biomass diverted to gall production is accounted for by a high proportion (66–73%) of multiple-chambered galls. The dry mass of multiple-chambered galls was significantly greater than that of both single-chambered galls and pods. (c) The extent of biomass diversion to gall production was relatively greater during the earlier part of the reproductive season, enhancing stress on the host. (d) The wasp sometimes forced the plant to produce up to 200% more galls per branch than the normal quota of inflorescences. This phenomenon, called forced commitment, further increases stress on the host plant. (e) Galls constituted up to 21% and 40% of the dry and wet biomass of above-ground parts of infested trees, respectively. This caused breakage and mortality of large branches and stems. (f) Reproduction in A. longifolia has been shown elsewhere to be so energy consuming that vegetative growth is strongly inversely related to pod production. Because the stress from galling by T. acaciaelongifoliae coincides with and replaces reproduction with a greater stress, successful reduction of both reproduction and vegetative growth are achieved.
• 3 This biological control programme is the first in which a gall-forming hymenopteran has been used to control a weed. Since the effects of gall-formers are indirect compared with those of insects attacking vegetative plant parts, their potential as biocontrol agents has been underrated. This research provides ecologically based guidelines for the future selection of such agents.

     

Sexually dimorphic gall structures correspond to differential phytohormone contents in male and female wasp larvae

Sexually dimorphic galls are rare among gall‐inducing insects and the reason for their occurrence is unknown. The pteromalid wasp Trichilogaster acaciaelongifoliae, which induces galls on Acacia longifolia, is one such species. In the present study, the anatomical and physiological attributes of male and female galls of T. acaciaelongifoliae are examined and compared. Histological preparations are used to characterize anatomical differences between male and female gall chambers. Bioassays, high‐performance liquid chromatography‐mass spectrometry and an enzyme immunoassay are used to measure concentrations of auxin and cytokinin in normal buds, galled tissues, and larvae of both sexes. Female chambers are found to be 3.3‐fold larger, and are associated with 1.5‐fold more storage tissue and 3.5‐fold more vascular tissues than male chambers. Tissues from female chambers induce stronger cytokinin‐like bioactivity than tissues from male chambers. Female larvae have considerably higher concentrations of cytokinin free bases, ribosides, glucosides and monophosphates than male larvae; higher auxin‐like bioactivity than in normal or galled plant tissues; and almost twice the concentration of auxin than male larvae. Both male and female larvae contain much higher auxin concentrations than either galled or normal plant tissues. These findings suggest that differing levels of phytohormones are involved in the development of sexual dimorphism of gall structures in this species.     

Nine new species of Dasineura (Diptera: Cecidomyiidae) from flowers of Australian Acacia (Mimosaceae)

Abstract. Thirteen species of Australian acacias are invasive plants in agricultural and native vegetation areas of South Africa. Biological control programmes for Australian acacias in South Africa have been implemented and are aimed at suppressing reproductive vigour and, in some cases, vegetative growth of these weeds. Gall-forming midges are under consideration as potential biological control agents for invasive acacias in South Africa. Entomological surveys in southern Australia found a diverse cecidomyiid fauna associated with the buds, flowers and fruits of Acacia species. Nine new Dasineura species are described and two species, D. acaciaelongifoliae (Skuse) and D. dielsi Rübsaamen, are redescribed. The newly described taxa are D. fistulosa sp.n. , D. furcata sp.n. , D. glauca sp.n. , D. glomerata sp.n. , D. oldfieldii sp.n. , D. oshanesii sp.n. , D. pilifera sp.n. , D. rubiformis sp.n. and D. sulcata sp.n. All eleven species induce galls on ovaries and prevent the formation of fruit. Two general types of gall are caused. Type A comprises woody, tubular galls with larvae living inside ovaries (D. acaciaelongifoliae, D. dielsi, D. fistulosa, D. furcata, D. glauca, D. glomerata, D. oldfieldii). Type B includes soft-tissued, globose galls that belong to four subtypes: inflated, baglike, hairy galls with larvae living between ovaries (D. pilifera); pyriform, pubescent swellings with larvae living inside ovaries (D. rubiformis); globose, hairy, swellings with larvae living superficially on ovaries in ovoid chambers (D. oshanesii); and inconspicuous, glabrous swellings with larvae living superficially on ovaries in shallow groovelike chambers (D. sulcata). The gall types are associated with a particular pupation pattern. In type A galls, larvae pupate within larval chambers in galls, whereas in type B galls pupation takes place between ovaries in galls or in the soil beneath the host tree. Gall midges responsible for the same general gall type are morphologically related and differ from species causing the other gall type. Phylogenetic analysis of a 410 bp fragment of the mitochondrial cytochrome b gene supports the division of the gall midge species into two groups except for D. sulcata, which appears as a subgroup of the group causing type A galls. The interspecific divergence values in group A species were between 0.5 and 3.9% with intraspecific divergence estimates of 0–0.2%. Gall midges causing type B galls had interspecific divergence values of 4.6–7.3% and intraspecific divergence values of 0–3.7%. Closely related biology and morphology together with low cytochrome b divergence estimates suggest a more recent speciation in group A when compared with species of group B. Dasineura rubiformis and D. dielsi are proposed as potential biological control agents for Acacia mearnsii De Wild. and Acacia cyclops A. Cunn. ex G. Don, respectively, in South Africa due to their narrow host range and ability to form high population densities that reduce seed formation. Both species produce galls with low biomass, which makes them compatible with commercial exploitation of their host species in Africa.     

Benefits of photosynthesis for insects in galls

Insect-induced plant galls are predominantly reputed to act as strong carbon sinks, although many types of galls contain chlorophyll and have the potential to photosynthesize. We investigated whether the photosynthetic capacity of bud galls induced by a Pteromalid wasp, Trichilogaster acaciaelongifoliae, in Acacia longifolia subsidises carbon budgets or provides O₂ to the larvae while concurrently consuming CO₂ in the dense gall tissue, thereby maintaining (O₂) and (CO₂) within the range of larval tolerance. Low (O₂) (<5?%?v/v) were found within the internal tissues of galls, and these concentrations responded only marginally to light, suggesting that the photosynthetic activity within the gall is inconsequential in the provision of O₂ to the larvae. The metabolic response of larvae to reduced (O₂) and elevated (CO₂) indicated that larvae were tolerant of hypoxia/hypercarbia and also capable of reducing their respiratory rates to cope with hypercarbia. The low mortality of larvae in galls shaded with Al-foil for 20?days showed that photosynthesis was not vital for the survival of the larvae, although growth of shaded galls was substantially reduced. Gas exchange measurements confirmed that, while photosynthesis never fully compensated for the respiratory costs of galls, it contributed substantially to the maintenance and growth, especially of young galls, reducing their impact as carbon sinks on the host. We conclude that, although photosynthesis may contribute to O₂ provision, its main role is to reduce the dependence of the insect-induced gall on the host plant for photosynthates, thereby reducing intra-plant, inter-gall competition and enhancing the probability that each gall will reach maturity.     

Trichilogaster sp. [Hymenoptera: Pteromalidae], a potential biocontrol agent for the weedAcacia pycnantha [Fabaceae]

The gall wasp,Trichilogaster sp., was imported from Australia to assess its potential as an agent for the control of the invasive shrub/treeAcacia pycnantha Benth. in South Africa. Host specificity tests indicate safety for release; of 19 tree/shrub species tested, including 16 species closely related toA. pycnantha, galls developed only onA. pycnantha. However, galling intensity remained consistently low on the host plant; only 21–29% of the branches exposed to the wasp were galled during 3 years of rearing. Neither the prolonged presence of males in test cages (someTrichilogaster species are thelytokous) nor the stage of maturity of reproductive buds exposed to oviposition affected the percentage of branches galled. It is not recommended thatTrichilogaster sp. be released before the possibility of insect-plant homeostasis or mis-matching of wasp and host plant populations/strains/subspecies is investigated, especially since galling intensities of 30% were ineffective in reducing seed production of a relatedTrichilogaster species/Acacia association.      

Relative investment in egg load and poison sac in fig wasps: Implications for physiological mechanisms underlying seed and wasp production in figs

Fig pollinating wasps and most non-pollinator wasps apply secretions from their poison sacs into oviposited flowers that appear necessary to the formation of the galls that their developing offspring consume. Thus, both eggs and poison sac secretions appear to be essential for wasp reproduction, but the relative investment in each is unknown. We measured relative investment in poison sac and egg production in pollinating and non-pollinating wasps associated with seven species of monoecious Panamanian figs representing both active and passive pollination syndromes. We then collected similar data for four fig hosts in China, where some wasp species in the genus Eupristina have lost the ability to pollinate (“cheaters”). All wasps examined possessed large poison sacs, and we found a strong positive correlation between poison sac size and absolute egg production. In the Panamanian species, the relative poison sac to egg investment was highest in the externally ovipositing non-pollinator wasps, followed by active pollinators, then by passive pollinators. Further, pollinator wasps of fig species with demonstrated host sanctions against “cheating” wasps showed higher investment in the poison sac than wasps of species without sanctions. In the Chinese samples, relative investment in the poison sac was indistinguishable between pollinators and “cheaters” associated with the same fig species. We suggest that higher relative investment in poison sac across fig wasp species reflects higher relative difficulty in initiating formation of galls and subsequently obtaining resources from the fig. We discuss the implications for the stability of the fig–wasp mutualism, and for the ability of non-pollinators to exploit this mutualism.     

Factors Influencing Realized Sex Ratios in Fig Wasps: Double Oviposition and Larval Mortalities

Pollinator fig wasps (Agaonidae) are a model system for studies of sex ratio evolution. They lay their eggs in galled ovules within figs. Only one adult emerges from each gall, suggesting that only one egg is always laid per ovule, but if double oviposition occurs then the assumption that adult (realised) sex ratios of fig wasps are representative of primary sex ratios may be violated. Many galls also fail to produce any wasps. If they initially contained eggs then differential mortality rates may also modify realized sex ratios. We investigated whether Kradibia (= Liporrhopalum) tentacularis foundresses in Ficus montana figs avoid laying in ovules that already contain eggs. Comparisons of oviposition frequencies and wasp emergence frequencies showed that most galls that failed to produce wasps will have had eggs laid in them, but few occupied ovules contained two eggs. Realised sex ratios therefore do not necessarily reflect primary sex ratios in this species, but double oviposition is not responsible.     

Finding hidden females in a crowd: Mate recognition in fig wasps

Multi-species mating aggregations are crowded environments within which mate recognition must occur. Mating aggregations of fig wasps can consist of thousands of individuals of many species that attain sexual maturity simultaneously and mate in the same microenvironment, i.e, in syntopy, within the close confines of an enclosed globular inflorescence called a syconium – a system that has many signalling constraints such as darkness and crowding. All wasps develop within individual galled flowers. Since mating mostly occurs when females are still confined within their galls, male wasps have the additional burden of detecting conspecific females that are “hidden” behind barriers consisting of gall walls. In Ficus racemosa, we investigated signals used by pollinating fig wasp males to differentiate conspecific females from females of other syntopic fig wasp species. Male Ceratosolen fusciceps could detect conspecific females using cues from galls containing females, empty galls, as well as cues from gall volatiles and gall surface hydrocarbons.In many figs, syconia are pollinated by single foundress wasps, leading to high levels of wasp inbreeding due to sibmating. In F. racemosa, as most syconia contain many foundresses, we expected male pollinators to prefer non-sib females to female siblings to reduce inbreeding. We used galls containing females from non-natal figs as a proxy for non-sibs and those from natal figs as a proxy for sibling females. We found that males preferred galls of female pollinators from natal figs. However, males were undecided when given a choice between galls containing non-pollinator females from natal syconia and pollinator females from non-natal syconia, suggesting olfactory imprinting by the natal syconial environment.     

Nematodes Associated with Fig Wasps,Pegoscapus spp. (Agaonidae), and Syconia of Native Floridian Figs (Ficus spp.)

Syconia in successive developmental phases from Ficus laevigata Vahl (F. citrifolia Miller sensu DeWolf 1960) (Moraceae) and successive life stages of its fig wasp pollinator, Pegoscapus sp. (P. assuetus (Grandi) sensu Wiebes 1983) (Agaonidae) were dissected to elucidate their association with two undescribed species of nematodes. Parasitodiplogazter sp. (Diplogasteridae) are transported by female Pegoscapus sp. into the cavity of a phase B syconium as third-stage juveniles (J3), where they molt to the J4 stage and greatly increase in size in the hemocoel of the fig wasp after it begins to pollinate and oviposit in female florets. The J4 exit the wasp cadaver in a phase B or early phase C syconium, and molt to adults that mate and lay eggs. New J3 infect the next generation of female or male wasps as they emerge from their galls in phase D figs. Mated entomogenous females of Schistonchus sp. (Aphelenchoididae) are transported in the hemocoel of female wasps to the fig cavity of a phase B syconium. Female Schistonchus sp. exit the wasp and parasitize immature male florets causing an exudate, the development of hypertrophied epidermal cells of the anther filaments and anthers, and aberrations of the anther filament, anthers, and pollen. At least one generation of Schistonchus sp. occurs in the male florets. Entomogenous females appear at about the time that fig wasps molt to adults in their galls in late phase C syconia. Another Schistonchus sp. was recovered from females of P. mexicanus (Ashmead) (P. jimenezi (Grandi) sensu Wiebes 1983) and from the syconia of F. aurea Nuttall and appears to have a life cycle similar to that described for the Schistonchus sp. from F. laevigata.     

Two invasive acacia species secure generalist pollinators in invaded communities

Exotic entomophilous plants need to establish effective pollinator interactions in order to succeed after being introduced into a new community, particularly if they are obligatory outbreeders. By establishing these novel interactions in the new non-native range, invasive plants are hypothesised to drive changes in the composition and functioning of the native pollinator community, with potential impacts on the pollination biology of native co-flowering plants. We used two different sites in Portugal, each invaded by a different acacia species, to assess whether two native Australian trees, Acacia dealbata and Acacia longifolia, were able to recruit pollinators in Portugal, and whether the pollinator community visiting acacia trees differed from the pollinator communities interacting with native co-flowering plants. Our results indicate that in the invaded range of Portugal both acacia species were able to establish novel mutualistic interactions, predominantly with generalist pollinators. For each of the two studied sites, only two other co-occurring native plant species presented partially overlapping phenologies. We observed significant differences in pollinator richness and visitation rates among native and non-native plant species, although the study of β diversity indicated that only the native plant Lithodora fruticosa presented a differentiated set of pollinator species. Acacias experienced a large number of visits by numerous pollinator species, but massive acacia flowering resulted in flower visitation rates frequently lower than those of the native co-flowering species. We conclude that the establishment of mutualisms in Portugal likely contributes to the effective and profuse production of acacia seeds in Portugal. Despite the massive flowering of A. dealbata and A. longifolia, native plant species attained similar or higher visitation rates than acacias.     

Foraging behavior of Brachygastra lecheguana (Hymenoptera: Vespidae) on Banisteriopsis malifolia (Malpighiaceae): Extrafloral nectar consumption and herbivore predation in a tending ant system

In the Brazilian savanna many plant species bear regular associations with patrolling ants that are aggressive towards insect herbivores. However, not only ants but also several species of predatory wasps are attracted to plants due to the extrafloral nectaries (EFNs). Such wasps feed on both herbivores and plant exudates. In this study we describe the foraging behavior of the social Polistinae wasp Brachygastra lecheguana in the extrafloral nectaried shrub Banisteriopsis malifolia, and investigated the influence of patrolling ants Camponotus blandus on the activity of the wasp. Brachygastra lecheguana fed on the endophytic larvae of Anthonomus (Curculionidae) beetles that developed inside flower buds. The wasp lacerated the bud layers to reach the beetle larvae located at the bud core. The wasp visits to Ba. malifolia were statistically related to the abundance of flower buds and beetles. Ant exclusion experiments revealed that the hunting behavior of B. lecheguana on beetles was not related to the absence of C. blandus. However we found that wasps spent more time consuming extrafloral nectar on branches where ants were excluded. This is the first study reporting extrafloral nectar consumption by B. lecheguana, as well as the predation on herbivores in natural areas. In cerrado vegetation, ants benefit the plant by reducing insect herbivores, and our study provides evidence that the B. lecheguana – Ba. malifolia system represents a potential interaction where the wasp may also benefit the host plant. The value of this wasp species as a plant‐guard is discussed.     

Foundress Fig Wasps are More Likely to Re-emerge From Older Figs

The mutualistic interaction between Ficus spp. and their pollinating fig wasps (Agaonidae) centres on the plants’ unique inflorescences—their figs. Each Ficus species is pollinated by foundresses of host-specific fig wasps which enter figs to lay eggs in the female flowers. Most foundresses are trapped in the first figs they enter, but in some species wingless foundresses can re-emerge and subsequently enter and oviposit into further figs. We investigated whether number of potential oviposition sites, age of the fig and age of the wasp influence the likelihood of re-emergence of lone foundresses of the Asian fig wasp Kradibia (=Liporrhopalum) tentacularis from previously un-entered figs of Ficus montana. Likelihood of re-emergence was not influenced by wasp age or flower numbers (resource abundance), but was more frequent from older figs that had waited longer to be pollinated. Laying eggs in several figs offers clear advantages, but foundresses often failed to re-emerge despite being unable to lay all their eggs. Resource quality not quantity appears to be the main influence on the fig wasp’s oviposition decisions. The physical difficulty that the wasps experience when trying to re-emerge may prevent it, even when re-emergence would be advantageous for both the insect and its host plant, but older fig wasps were not detectably ‘weaker’ than younger individuals.     

Volatiles associated with different flower stages and leaves of Acacia cyclops and their potential role as host attractants for Dasineura dielsi (Diptera: Cecidomyiidae)

Acacia cyclops (Fabaceae) is an Australian species which was introduced into South Africa in the nineteenth century. Because of its invasive status in South Africa, a gall midge, Dasineura dielsi (Diptera: Cecidomyiidae), was released in 2001 in order to impact its reproduction by inducing galls on the flowers and thereby preventing seed set. Nothing is known about the cues used by D. dielsi for locating its host flowers. As part of an initial investigation into whether or not chemical cues might play a role in host finding, we analysed headspace samples of Acacia cyclops volatiles from leaves and reproductive parts at different stages (early bud, late bud, early flowering, and senescing flowering stages) using gas chromatography–mass spectrometry (GC–MS). In total, 72 different compounds were detected of which 62 were identified. The analyses showed that open flowers, the stage used by D. dielsi for oviposition, and yellow buds had similar odour compositions with (Z)-3-hexen-1-ol acetate, 4-oxoisophorone, (Z)-β-ocimene, an unknown aliphatic compound, heptadecane, and nonadecane dominating in open flowers. Leaf volatiles were distinct from those in the reproductive plant parts by their high relative amount of (Z)-β-ocimene. (Z)-3-Hexen-1-ol acetate had its maximum relative amount in the green bud samples and was much lower in the later floral stages. In contrast, 4-oxoisophorone peaked in yellow buds and open flowers with little or none of it found in younger or older stages. The volatile compounds of the different flower stages and leaves are discussed in relation to their potential role as attractants used by the biocontrol agent D. dielsi to locate its host plant.     

Analysis of Tiphia parasitoids preovipositional behaviors and of their scarab host defensive responses

Tiphia vernalis Rohwer and Tiphia popilliavora Rohwer were introduced as biocontrol agents against Japanese beetles (Popillia japonica Newman) and oriental beetles (Anomala orientalis Waterhouse). Studies have shown that under field conditions, T. vernalis parasitize Japanese beetles more often than oriental beetles. This study was done to understand how tiphiid wasps handle the two different host species and the influence of host defensive behaviors on the oviposition process of tiphiid wasps. The preovipositional behaviors performed by Tiphia wasps included: stinging, examining, moving soil, kneading, host-feeding, and host scraping. The frequency, sequence, and total time spent on each behavior before oviposition were scored and compared between two host species. The sequence and frequency of preovipositional behaviors performed by both Tiphia wasps did not show a difference between the two host species. However, female T. vernalis spent significantly longer time trying to sting oriental beetles than Japanese beetles in order to paralyze them. The time T. popilliavora spent on prestinging behaviors did not show a difference between Japanese and oriental beetles. The defensive behaviors performed by Japanese and oriental beetle grubs included: vigorous movements, rubbing their abdomen or head against the wasp’s abdomen, and biting at the attacking wasp. The frequency and total time spent on each defensive behavior was scored and compared between two host species. Overwintered, third instar oriental beetle grubs spent significantly longer time on defensive behaviors when they were attacked by T. vernalis which likely cause wasps to spend longer time trying to sting oriental beetle grubs. The active host resistance gained through behavioral defenses could make oriental beetle grubs less susceptible to T. vernalis attack than Japanese beetle grubs, especially under field conditions. Younger grubs attacked by T. popilliavora did not exhibit these differences.     

Between‐species facilitation by male fig wasps in shared figs

1. Facilitation is recorded from diverse plant–insect interactions, including pollination and herbivory. 2. The significance of facilitation resulting from the behavior of males of multiple fig wasp species inside figs was investigated. Female fig wasps emerge from natal figs via exit holes dug by males, especially male pollinators. When no males are present, the females struggle to escape and may die. 3. Ficus microcarpa L. is a widely‐established invasive fig tree from Southeast Asia. Its pollinator is absent in South Africa, so the tree cannot reproduce, but two Asian non‐pollinating fig wasps (NPFW) Walkerella microcarpae and Odontofroggatia galili occupy its figs. Abundance patterns of the two NPFW and the proportion of male‐free figs in South Africa, Spain (where the pollinator is introduced), and in China, where the native fig wasp community is diverse, were compared to determine the consequences of reduced species richness for insect survival. 4. Female fig wasps in male‐free figs were found to be trapped, and small clutch sizes contributed to the absence of males in both species. The presence of pollinators in Spain allowed most NPFW to develop in figs containing males. Far more male‐free figs were present in South Africa, elevating mortality rates among female NPFW. Facilitation of female release by males of other NPFW species nonetheless benefitted the rarer species. 5. Selection pressures in South Africa currently favour greater aggregation of NPFW offspring and/or less female biased sex ratios.     

Secondary galling: a novel feeding strategy among ‘non‐pollinating’ fig wasps from Ficus curtipes

相似文献   

Diversification in the use of resources by Idarnes species: bypassing functional constraints in the fig–fig wasp interaction

Mutualisms such as the fig–fig wasp mutualism are generally exploited by parasites. We demonstrate that amongst nonpollinating fig wasps (NPFWs) parasitic on Ficus citrifolia, a species of Idarnes galls flowers and another species feeds on galls induced by other wasps killing their larvae. The galling wasp inserts its ovipositor through the fig wall into the fig cavity. The ovipositor then follows a sinuous path and is introduced through the stigma and style of the flower. The egg is deposited between the integument and nucellus, in the exact location where the pollinating mutualistic wasp would have laid its egg. Gall induction is a complex process. In contrast, the path followed by the ovipositor of the other species is straightforward: attacking a larva within a developed gall poses different constraints. Shifts in feeding regime have occurred repeatedly in NPFWs. Monitoring traits associated with such repeated evolutionary shifts may help understand underlying functional constraints. © 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, 106 , 114–122.     

Mutualism from the inside: coordinated development of plant and insect in an active pollinating fig wasp

Recent studies on the obligate interaction between fig trees and their pollinating agaonid wasps have focused on population aspects and wasp?Cseed exploitation at the level of the inflorescence. Detailed studies on larval and gall development are required to more fully understand how resources are exploited and adaptations fine-tuned by each partner in nursery pollination mutualisms. We studied the larval development of the active pollinating fig wasp, Pegoscapus sp., and the galling process of individual flowers within the figs of its monoecious host, Ficus citrifolia, in Brazil. The pollinator development is strongly dependent on flower pollination. Figs entered by pollen-free wasps were in general more likely to abort. Retained, unpollinated figs had both higher larval mortality and a lower number of wasps. Pegoscapus sp. larvae are adapted to plant development, with two contrasting larval feeding strategies proceeding alongside gall development. The first two larval stages behave as ovary parasites. Later larval stages feed on hypertrophied endosperm. This indicates that a successful galling process relies on endosperm, and also reveals why pollination would be a prerequisite for the production of high-quality galls for this Pegoscapus species.     

Utilisation of chemical signals by inquiline wasps in entering their host figs

The fig tree, Ficus curtipes, hosts an obligate pollinating wasp, an undescribed Eupristina sp., but can also be pollinated by two inquiline (living in the burrow, nest, gall, or other habitation of another animal) wasps, Diaziella yangi and an undescribed Lipothymus sp. The two inquilines are unable to independently induce galls and depend on the galls induced by the obligate pollinator for reproduction and, therefore, normally enter receptive F. curtipes figs colonised by the obligate pollinators. However, sometimes the inquilines also enter figs that are not colonised by the pollinators, despite consequent reproductive failure. It is still unknown which signal(s) the inquilines use in entering the colonised and non-colonised figs. We conducted behavioural experiments to investigate several possible signals utilised by the inquilines in entering their host receptive figs. Our investigation showed that both inquiline species enter the receptive F. curtipes figs in response to the body odours of the obligate wasps and one of the main compounds emitted by the figs, 6-methyl-5-hepten-2-one. The compound was not found in the pollinator body odours, suggesting that the two inquiline wasps can utilise two signals to enter their host figs, which is significant for the evolution of the fig-fig wasp system. These inquilines could evolve to become mutualists of the figs if they evolve the ability to independently gall fig flowers; there is, however, another possibility that a monoecious Ficus species hosting such inquilines may evolve into a dioecious one if these inquilines cannot evolve the above-mentioned ability. Additionally, this finding provides evidence for the evolution of chemical communication between plants and insects.