Ant–plant interaction in the Neotropical savanna: direct beneficial effects of extrafloral nectar on ant colony fitness

Current evidence suggests that ant–plant relationships may influence species composition, abundance, and interactions at the community scale. The main resource that plants offer to ants is extrafloral nectar (EFN) and the major part of published studies shown benefits from ants to plants possessing EFNs. However, the complementary question of whether and how ants benefit from EFNs is rarely addressed. Here, we present the results of a long-term study to demonstrate whether EFN has a positive effect on ant colony fitness. We quantified colony growth rate, survival and the final weight of individuals as measures of benefit derived from EFN. Our results provide clear evidence that EFN can have a significant positive impact on the survivorship, growth and reproduction of the Myrmicinae Cephalotes pusillus. In fact, a diet rich in EFN (providing at least 30 cal per day) resulted in five times more individuals per colony, greater body weights, and more eggs. These results have shed new light on the relationships between ants and EFN-bearing plants such as in tropical and temperate systems. The ant C. pusillus is the first case in which we have firm evidence that EFN improves colony growth and development, corroborating more than 100 years of experimental evidence of benefits to plants in these widespread relationships.     

The adaptive bases of ant-mimicry in a neotropical aphantochilid spider (Araneae: Aphantochilidae)

The aphantochilid spider Aphanlochilus rogersi accurately mimics black ants of tribe Cephalotini, and is commonly found in the neighbourhood of its models' nests. The mimic seems to be a specialized predator of this type of ant, rejecting any insect offered as prey other than cephalotines. In the field, A. rogersi was observed preying on the model species Zacryptocerus pusillus. In captivity, the spider preyed on the models Z. pusillus and Z. depressus, as well as on the yellow non-model Z. clypeatus. Recognition of correct prey by A. rogersi appears to be based primarily on visual and tactile stimuli. Capturing ant prey from behind was the most common attack tactic observed in A. rogersi, and is probably safer than frontal attacks, as in this case the spider can be bitten on the legs before the ant is immobilized. Aphanlochilus rogersi, when feeding on the hard-bodied ant models, uses the ant corpses as a ‘protective shield’ against patrolling ants of the victim's colony and resembles an ant carrying a dead companion. Certain types of mimetic traits in A. rogersi (close similarity to ant models in integument texture and pilosity of body and legs), together with ‘shielding behaviour’, are thought to function as ant-deceivers, facilitating the obligatory intimate contact the mimic must make with cephalotines in order to capture a prey among other ants. The close similarity in the arrangement of dorsal spines, body shape, integument brightness and locomotion, together with antennal illusion, is regarded as a strategy of A. rogersi for deceiving visually-hunting predators that avoid its sharp spined ant models. It is proposed that ant-mimicry in A. rogersi has both an aggressive and a Batesian adaptive component, and evolved as a result of combined selective pressures exerted both by Cephalotini ant models (through defensive behaviour towards the mimics which attack them) and predators that avoid cephalotines (through predatory behaviour toward imperfect mimics). This suggestion is schematized and discussed in terms of two tripartite mimicry systems.     

Friend or foe? A behavioral and stable isotopic investigation of an ant–plant symbiosis

In ant–plant symbioses, the behavior of ant inhabitants affects the nature of the interaction, ranging from mutualism to parasitism. Mutualistic species confer a benefit to the plant, while parasites reap the benefits of the interaction without reciprocating, potentially resulting in a negative impact on the host plant. Using the ant–plant symbiosis between Cordia alliodora and its ant inhabitants as a model system, I examine the costs and benefits of habitation by the four most common ant inhabitants at La Selva Biological Station, Costa Rica. Costs are measured by counting coccoids associated with each ant species. Benefits include patrolling behavior, effectiveness at locating resources, and recruitment response. I also compare the diets of the four ant species using stable isotope analysis of nitrogen (N) and carbon (C). Ants varied in their rates of association with coccoids, performance of beneficial behaviors, and diet. These differences in cost, benefit, and diet among the ant species suggest differences in the nature of the symbiotic relationship between C. alliodora and its ants. Two of the ant species behave in a mutualistic manner, while the other two ant species appear to be parasites of the mutualism. I determined that the mutualistic ants feed at a higher trophic level than the parasitic ants. Behavioral and dietary evidence indicate the protective role of the mutualists, and suggest that the parasitic ants do not protect the plant by consuming herbivores.     

Effects of food rewards offered by ant–plant Macaranga on the colony size of ants

Myrmecophytes (ant–plants) have special hollow structures (domatia) in which obligate ant partners nest. As the ants live only on the plants and feed exclusively on plant food bodies, sap-sucking homopterans in the domatia, and/or the homopterans honeydew, they are suitable for the study of colony size regulation by food. We examined factors regulating ant colony size in four myrmecophytic Macaranga species, which have strictly species-specific association with Crematogaster symbiont ants. Intra- and interspecific comparison of the plants showed that the ant biomass per unit food biomass was constant irrespective of plant developmental stage and plant species, suggesting that the ant colony size is limited by food supply. The primary food offered by the plants to the ants was different among Macaranga species. Ants in Macaranga beccariana and Macaranga bancana relied on homopterans rather than food bodies, and appeared to regulate the homopteran biomass and, as a consequence, regulate the ants own biomass. In contrast, ants in Macaranga winkleri and Macaranga trachyphylla relied primarily on food bodies rather than homopterans, and the plants appeared to manipulate the ant colony size. Per capita plant investment in ants (ant dry weight plant dry weight^–1) was different among the four Macaranga species. The homoptera-dependent M. beccariana and M. bancana harbored lower biomass of ants than the food-body dependent M. winkleri, suggesting that energy loss is involved in the homoptera-interposing symbiotic system which has one additional trophic level. The plants investment ratio to the ants generally decreased as plants grew. The evolution of the plant reward-offering system in ant–plant–homopteran symbioses is discussed with an emphasis on the role of homopterans.     

Competition hierarchy and plant defense in a guild of ants on tropical Passiflora

In facultative ant–plant interactions, ants may compete with each other for food provided by extrafloral nectar (EFN) plants. We studied resource competition and plant defense in a guild of ants that use the same EFN resource provided by two species of Passiflora in a seasonal rain forest in tropical China. At least 22 ant species were recorded using the EFN resource, although some of those species were rare. Among these ants, Paratrechina sp.1 and Dolichoderus thoracicus were more aggressive than other species. Ant aggressiveness measured as ant behavioral dominance index (BDI) was positively correlated with ant abundance on the Passiflora species studied. Ant BDI was also positively correlated to the protection that ants provided against herbivory. In Passiflora siamica, the number of workers patrolling on the plants did negatively correlate with average leaf loss per plant. We conclude that in this facultative Passiflora–ant system, plant defense upon herbivore was indeed influenced by the total number of ants present on plant and the aggressiveness of these ants.     

Size matters: mole (Talpa europaea) hills and nest-site selection of the ant Formica exsecta

Moles are fossorial mammals that can act both as zoogeomorphic agents and species diversity drivers. These popular animals regularly push heaps of earth from their subterranean tunnel systems to the surface. Thereby they rearrange and improve the local microtopography for ant nesting. Here we use a strongly molehill (Talpa europaea) mediated nest system of the unicolonial wood ant Formica (Coptoformica) exsecta to test for ecological factors influencing nest-site selection at the microhabitat scale. Our results show that the size of molehills plays an important role in the multifactorial process of the ant’s nest-site choice with solar insolation as a paramount factor. The ants clearly favored larger and better sun-exposed molehills, suggesting that the coaction of a zoogeomorphic modified microrelief and solar insolation can drive the spatial colonization of F. exsecta.     

How a lack of choice can force ants to climb up waxy plant stems

The aim of this study was to analyse the ability of generalist ants to climb waxy stems and to understand whether the avoidance of such stems is solely based on their reduced ability to adhere to these surfaces. We performed observations of the locomotory behaviour and recorded the traversed distances of Lasius niger ants on flower stems of three plant species covered with epicuticular wax crystals: Anethum graveolens, Dahlia pinnata, and Tagetes patula. Using conventional and cryo-scanning electron microscopy, the surface micromorphology of plant stems and structure of ant attachment organs were examined. The present study demonstrates that the reason ants do not climb wax-covered stems is not because they are incapable of walking on such vertical surfaces, as previously assumed, but rather because they have a choice of other substrates. Presumably, additional locomotory efforts are needed to master climbing on “greasy” stems, and ants need a certain time to adapt to a new quality of the stem surface, and after this period, they can walk on these waxy stems.     

Yellow crazy ant (<Emphasis Type="Italic">Anoplolepis gracilipes</Emphasis>) invasions within undisturbed mainland Australian habitats: no support for biotic resistance hypothesis

Ants are highly successful invaders, especially on islands, yet undisturbed mainland environments often do not contain invasive ants, and this observation is largely attributed to biotic resistance. An exception is the incursion of Yellow crazy ant Anoplolepis gracilipes within northeast Arnhem Land. The existence of A. gracilipes within this landscape’s intact environments containing highly competitive ant communities indicates that biotic resistance is not a terminally inhibitory factor mediating this ant’s distribution at the regional scale. We test whether biotic resistance may still operate at a more local scale by assessing whether ecological impacts are proportional to habitat suitability for A. gracilipes, as well as to the competitiveness of the invaded ant community. The abundance and species richness of native ants were consistently greater in uninfested than infested plots but the magnitude of the impacts did not differ between habitats. The abundance and ordinal richness of other macro-invertebrates were consistently lower in infested plots in all habitats. A significant negative relationship was found for native ant abundance and A. gracilipes abundance. No relationships were found between A. gracilipes abundance and any measure of other macro-invertebrates. The relative contribution of small ants (<2.5 mm) to total abundance and relative species richness was always greater in infested sites coinciding with a reduction of the contribution of the larger size classes. Differences in the relative abundance of ant functional groups between infested and uninfested sites reflected impacts according to ant size classes and ecology. The widespread scale of these incursions and non-differential level of impacts among the habitats, irrespective of native ant community competitiveness and abiotic suitability to A. gracilipes, does not support the biotic resistance hypothesis.     

A native ant armed to limit the spread of the Argentine ant

Argentine ants (Linepithema humile) usually actively displace native ants through a combination of rapid recruitment, high numerical dominance and intense aggressive fights. However, in some cases, native ants can offer a strong resistance. In Corsica, a French Mediterranean island, local resistance by the dominant Tapinoma nigerrimum has been proposed as a factor limiting Argentine ant invasion. With the aim of evaluating the abilities of T. nigerrimum in interference and exploitative competition, this study tested in the laboratory the aggressive interactions between this native dominant ant and the invasive Argentine ant. We used four different assays between L. humile and T. nigerrimum: (1) worker dyadic interactions, (2) symmetrical group interactions, (3) intruder introductions into an established resident colony, and (4) a competition for space and food. This study confirms the ability of Argentine ants to compete with native species, by initiating more fights, using cooperation and simultaneously deploying physical and chemical defenses. However, despite Argentine ant fighting capabilities, T. nigerrimum was more efficient in both interference and exploitative competition. Its superiority was obvious in the space and food competition assays, where T. nigerrimum dominated food in 100% of the replicates after 1 h and invaded Argentine ant nests while the reverse was never observed. The death feigning behavior exhibited by Argentine ant workers also suggests the native ant’s superiority. Our study thus demonstrates that T. nigerrimum can offer strong competition and so may be able to limit the spread of Argentine ants in Corsica. This confirms that interspecific competition from ecologically dominant native species can affect the invasion success of invaders, notably by decreasing the likelihood of incipient colony establishment and survival.     

Dynamics of the association between a long-lived understory myrmecophyte and its specific associated ants

Myrmecophytic symbioses are widespread in tropical ecosystems and their diversity makes them useful tools for understanding the origin and evolution of mutualisms. Obligate ant–plants, or myrmecophytes, provide a nesting place, and, often, food to a limited number of plant–ant species. In exchange, plant–ants protect their host plants from herbivores, competitors and pathogens, and can provide them with nutrients. Although most studies to date have highlighted a similar global pattern of interactions in these systems, little is known about the temporal structuring and dynamics of most of these associations. In this study we focused on the association between the understory myrmecophyte Hirtella physophora (Chrysobalanaceae) and its obligate ant partner Allomerus decemarticulatus (Myrmicinae). An examination of the life histories and growth rates of both partners demonstrated that this plant species has a much longer lifespan (up to about 350 years) than its associated ant colonies (up to about 21 years). The size of the ant colonies and their reproductive success were strongly limited by the available nesting space provided by the host plants. Moreover, the resident ants positively affected the vegetative growth of their host plant, but had a negative effect on its reproduction by reducing the number of flowers and fruits by more than 50%. Altogether our results are important to understanding the evolutionary dynamics of ant–plant symbioses. The highly specialized interaction between long-lived plants and ants with a shorter lifespan produces an asymmetry in the evolutionary rates of the interaction which, in return, can affect the degree to which the interests of the two partners converge.     

Seed desiccation limits removal by ants

Ants collect and disperse seeds that bear an attractive nutritive body called the elaiosome. In mesic habitats, many myrmecochorous plant species have elaiosomes that are usually soft and desiccation-sensitive. The aim of this study was to link the desiccation rate of two species of seeds (Chelidonium majus and Viola odorata) to the seed-removing behaviour of the ant Myrmica rubra. In laboratory experiments seeds of both species lost one-third of their weight in 24 h. Concurrently, seed removal rates decreased sharply (92%) for Viola odorata over 24 h and slowly for Chelidonium majus, which retained one-third of its attractiveness after one month of desiccation. Seeds recovered their initial weight almost entirely after being soaked in water for 18 h. This rehydration partially (Viola odorata) or totally (Chelidonium majus) restored the attractiveness of the seeds. In this paper, we show that the window of seed attractiveness during which ant–plant interactions occur is desiccation-dependent and plant-specific.     

Estimation of mortality by entomophages on exotic Harmonia axyridis versus native Adalia bipunctata in semi-field conditions in northern Italy

A semi-field experiment was carried out in two peach orchards in northern Italy to assess mortality due to predators and parasitoids on the exotic coccinellid Harmonia axyridis (Pallas) (Coleoptera: Coccinellidae) in comparison with the native coccinellid Adalia bipunctata L. (Coleoptera: Coccinellidae). The experiments were conducted in cages to avoid the possible escape of the exotic ladybird (not yet established in Italy). Two kinds of cage experiments were included: ‘exclusion cages’ (access by walking predators impeded) and ‘free cages’ (walking predators free to enter). The cages, containing all the stages of the two ladybird species, were placed in two localities and left for 24 h. All ladybird stages used for the semi-field experiments came from a laboratory rearing. The eggs of H. axyridis experienced less mortality than those of A. bipunctata. The ant workers were the most frequent predators in ‘free cages’ but A. bipunctata cannibalism on eggs was also detected. Larvae of both coccinellid species were predated equally but larval predation of L1 and L2 was higher in comparison to predation of L3 and L4. Pupae and adults of both exotic and native ladybirds were never attacked by predators. Predation on younger larval stages was higher in the ‘free cages’ in comparison with ‘exclusion cages’. No ladybird parasitisation was observed. The ‘free cage’ technique seems to provide a standardised and realistic estimation of predation impact but more studies are needed to evaluate ladybird parasitisation in semi-field conditions.     

Patterns of host ant use by sympatric populations of <Emphasis Type="Italic">Maculinea alcon</Emphasis> and <Emphasis Type="Italic">M. ‘rebeli’</Emphasis> in the Carpathian Basin

Maculinea butterflies show social parasitism via obligatory myrmecophily as their larvae are adopted and raised to pupation by Myrmica ants. Suitable hosts differ for different Maculinea species, and host ant specificity can further differ at the population-level. Although early studies suggested single ant species as main hosts for each Maculinea species, it has recently become clear that their host ant specificity is more complex. Maculinea alcon and Maculinea ‘rebeli’ have variously been separated according to adult and larval morphology, phenology, and their use of different ecosystems, including host plant and host ant species. However, recent genetic evidence has questioned their separation as good species. Here we compare the use of host ants by M. alcon and M. ‘rebeli’ at the regional scale in NE-Hungary and Transylvania (Romania), where molecular studies have found no species-level separation between the two forms. We opened 778 nests of Myrmica ants and searched for Maculinea specimens (larvae, pupae and exuviae) shortly before imago emergence from the nest in seven M. alcon sites, six M. ‘rebeli’- sites and one site where both M. alcon and M. ‘rebeli’ are syntopic. In all, Maculinea caterpillars were found in the nests of seven different ant species (M. alcon was recorded mainly with Myrmica scabrinodis and occasionally with M. salina and M. vandeli; M. ‘rebeli’ used mainly M. scabrinodis, M. sabuleti and M. schencki and occasionally M. lonae and M. specioides). Myrmica scabrinodis was found to be a general host of both M. alcon and M. ‘rebeli’, which is the first record for a common host ant of these two closely related butterflies within the same region. However there were also differences in host ant use patterns between the sites occupied by the two Maculinea taxa, which reflect differences in Myrmica communities between the two types of habitat. Possible explanations for the similar but not identical host use patterns of M. alcon and M. ‘rebeli’, and their relevance for the question of whether they are separate species are discussed. Received 27 November 2007; revised 28 May 2008; accepted 11 June 2008.     

Interaction between flowers, ants and pollinators: additional evidence for floral repellence against ants

It has been commonly suggested that ants negatively affect plant pollination, particularly in the tropics. We studied ant–flower–pollinator interactions in a lowland rainforest in Borneo. Frequency and duration of pollinator visits were compared between flowers attended by ants and flowers from which ants were excluded. In all four plant species studied, the activity of ants decreased the rate and/or duration of the pollinators’ floral visits. For this and other reasons it is expected that plants repel ants from flowers during anthesis. We tested this prediction for a different set of plant species in which we observed the behaviour of Dolichoderus thoracicus ants when encountering flowers. In eight out of 18 plant species studied, ants showed a significantly higher rejection rate when they encountered flowers than when they encountered controls. Our results are thus consistent with the hypothesis that ants may negatively affect plant fitness by reduced intensity of pollinator visits and that ants are repelled from flowers of many tropical plant species, although this repellence is clearly not ubiquitous.     

Do extrafloral nectar resources, species abundances, and body sizes contribute to the structure of ant–plant mutualistic networks?

Recent research has shown that many mutualistic communities display non-random structures. While our understanding of the structural properties of mutualistic communities continues to improve, we know little of the biological variables resulting in them. Mutualistic communities include those formed between ants and extrafloral (EF) nectar-bearing plants. In this study, we examined the contributions of plant and ant abundance, plant and ant size, and plant EF nectar resources to the network structures of nestedness and interaction frequency of ant–plant networks across five sites within one geographic locality in the Sonoran Desert. Interactions between ant and plant species were largely symmetric. That is, ant and plant species exerted nearly equivalent quantitative interaction effects on one another, as measured by their frequency of interaction. The mutualistic ant–plant networks also showed nested patterns of structure, in which there was a central core of generalist ant and plant species interacting with one another and few specialist–specialist interactions. Abundance and plant size and ant body size were the best predictors of symmetric interactions between plants and ants, as well as nestedness. Despite interactions in these communities being ultimately mediated by EF nectar resources, the number of EF nectaries had a relatively weak ability to explain variation in symmetric interactions and nestedness. These results suggest that different mechanisms may contribute to structure of bipartite networks. Moreover, our results for ant–plant mutualistic networks support the general importance of species abundances for the structure of species interactions within biological communities.     

Contrasting effects of an invasive ant on a native and an invasive plant

When invasive species establish in new environments, they may disrupt existing or create new interactions with resident species. Understanding of the functioning of invaded ecosystems will benefit from careful investigation of resulting species-level interactions. We manipulated ant visitation to compare how invasive ant mutualisms affect two common plants, one native and one invasive, on a sub-tropical Indian Ocean island. Technomyrmex albipes, an introduced species, was the most common and abundant ant visitor to the plants. T. albipes were attracted to extrafloral nectaries on the invasive tree (Leucaena leucocephala) and deterred the plant’s primary herbivore, the Leucaena psyllid (Heteropsylla cubana). Ant exclusion from L. leucocephala resulted in decreased plant growth and seed production by 22% and 35%, respectively. In contrast, on the native shrub (Scaevola taccada), T. albipes frequently tended sap-sucking hemipterans, and ant exclusion resulted in 30% and 23% increases in growth and fruit production, respectively. Stable isotope analysis confirmed the more predacious and herbivorous diets of T. albipes on the invasive and native plants, respectively. Thus the ants’ interactions protect the invasive plant from its main herbivore while also exacerbating the effects of herbivores on the native plant. Ultimately, the negative effects on the native plant and positive effects on the invasive plant may work in concert to facilitate invasion by the invasive plant. Our findings underscore the importance of investigating facilitative interactions in a community context and the multiple and diverse interactions shaping novel ecosystems.     

Nocturnal resource defence in aphid‐tending ants of northern Patagonia

相似文献   

Behavioural traits mediating effects of two plant-ants on their specific myrmecophyte host

Summary. Symbiotic ant species vary in their effects on the same host plant, and some ant species are exploiters. These can destabilise the interaction. The behavioural repertoire of ant associates should be important in determining their effects on the host. We examined several traits of foraging behaviour of two ant species associated with the African understorey tree Leonardoxa africana africana – the protective mutualist Petalomyrmex phylax and the parasite Cataulacus mckeyi – to examine the mechanistic basis of the difference in their effects on the plant. These two ant species displayed significant differences not only in activity rhythms, P. phylax patrolling leaves day and night while C. mckeyi was strictly diurnal, but also in their predatory behaviour (P. phylax killed more herbivores) and recruitment dynamics (C. mckeyi recruited at a lower level). For all traits examined, C. mckeyi expressed behaviours reflecting an opportunistic exploitation of the nectar produced by the host plant, whereas those displayed by P. phylax contribute to conferring effective protection of the host plant. Our results suggest that prey capture may be important as a source of food for Petalomyrmex, in addition to its function in increasing plant-provided resources through protection.Received 27 August 2004; revised 11 November 2004; accepted 16 November 2004.     

How ecology shapes caste evolution: linking resource use,morphology, performance and fitness in a superorganism

Caste evolution is a central process in the adaptive diversification of insect superorganisms. Nevertheless, how ecology shapes adaptive caste evolution remains poorly understood. Recent work with the ant genus Cephalotes has provided new comparative evidence that ecological specialization may drive adaptive caste specialization. Here, three key predictions of this adaptive hypothesis are supported, using a representative of the highest level of ecological specialization and the most specialized soldier phenotype. First, soldier defensive performance was maximal for the specific nesting resource used most often in nature. Second, colonies only used a specialized subset of available nesting resources and preferred the specific resource that maximizes soldier performance. Third, soldier performance and its limitations on resource use were found to have both direct and indirect consequences for colony reproduction. These findings suggest that the most specialized soldier phenotype in Cephalotes is indeed an adaptation to ecological specialization on a narrowly defined subset of available nesting resources.     

Dispersal mechanisms of the narrow endemic <Emphasis Type="Italic">Polygala vayredae</Emphasis>: dispersal syndromes and spatio-temporal variations in ant dispersal assemblages

This study assesses the dispersal mechanisms of the narrow endemic Polygala vayredae, analysing the functioning of its dispersal syndromes (anemochory and myrmecochory), the spatio-temporal variability of the disperser assemblage, foraging behaviour and dispersal ability, and the role of the elaiosome in ant attraction and seed germination. The dispersion of diaspores begins when either (1) capsules or seeds fall beneath the mother plant (barochory) or (2) the seeds are directly collected in the suspended capsules by ants (myrmecochory). As capsules frequently open and expose/disseminate seeds before leaving the mother plant, the adaptation for anemochory appears to be reduced and rarely functional, possibly with only occasional events of long-distance dispersal (e.g. under extreme weather conditions). P. vayredae is essentially myrmecochorous and a diverse array of ant species are involved in seed manipulation, with the elaiosome playing a major role in ant attraction. From the plant’s perspective for dispersal, the majority of ant species had a positive interaction with the seeds, but negative and potential neutral interactions were also observed. Overall, dispersal distances were limited and were mainly determined by ant body size. The frequency of interactions and the ant assemblage varied significantly both spatially and temporally, and these factors may have an effect on directing or disrupting the selection of plant traits. Low seed predation and similar germination rates of intact seeds compared with seeds without elaiosome indicate that seed predator avoidance and seed germination improvement after ant manipulation are not among the selective advantages of myrmecochory operating at present. Dispersal mechanisms that enhance seed dispersal within the population and only occasionally lead to long-distance dispersal events, along with the rarity and patchiness of suitable habitats, may be the main factors explaining the actual density and narrow distribution of this species.