Impacts of Argentine ants on mealybugs and their natural enemies in California's coastal vineyards

Abstract 1. The Argentine ant, Linepithema humile, tends honeydew‐excreting homopterans and can disrupt the activity of their natural enemies. This mutualism is often cited for increases in homopteran densities; however, the ant’s impact on natural enemies may be only one of several effects of ant tending that alters insect densities. To test for the variable impacts of ants, mealybug and natural enemy densities were monitored on ant‐tended and ant‐excluded vines in two California vineyard regions. 2. Ant tending increased densities of the obscure mealybug, Pseudococcus viburni, and lowered densities of its encyrtid parasitoids Pseudaphycus flavidulus and Leptomastix epona. Differences in parasitoid recovery rates suggest that P. flavidulus was better able to forage on ant‐tended vines than L. epona. 3. Densities of a coccinellid predator, Cryptolaemus montrouzieri, were higher on ant‐tended vines, where there were more mealybugs. Together with behavioural observations, the results showed that this predator can forage in patches of ant‐tended mealybugs, and that it effectively mimics mealybugs to avoid disturbance by ants. 4. Ant tending increased densities of the grape mealybug, Pseudococcus maritimus, by increasing the number of surviving first‐instar mealybugs. Parasitoids were nearly absent from the vineyard infested with P. maritimus. Therefore, ants improved either mealybug habitat or fitness. 5. There was no difference in mealybug distribution or seasonal development patterns on ant‐tended and ant‐excluded vines, indicating that ants did not move mealybugs to better feeding locations or create a spatial refuge from natural enemies. 6. Results showed that while Argentine ants were clearly associated with increased mealybug densities, it is not a simple matter of disrupting natural enemies. Instead, ant tending includes benefits independent of the effect on natural enemies. Moreover, the effects on different natural enemy species varied, as some species thrive in the presence of ants.     

Do additional sugar sources affect the degree of attendance of Dysmicoccus brevipes by the fire ant Solenopsis geminata?

Mutualistic interactions between ants and Hemiptera are mediated to a large extent by the amount and quality of sugar‐rich honeydew produced. Throughout the neotropics, the predaceous fire ant Solenopsis geminata (Fabricius) (Hymenoptera: Formicidae) is found in association with colonies of the pineapple mealybug, Dysmicoccus brevipes (Cockerell) (Hemiptera: Pseudococcidae), which they actively tend and protect from attack by natural enemies. In this study, we evaluate the effects of access to a sucrose solution on the mutualistic association between S. geminata and D. brevipes. Ten colonies of either species were established, with D. brevipes maintained on pumpkin, Cucurbita maxima Duchesne (Cucurbitaceae), in screen cages. Five of the S. geminata colonies were permitted access to vials with 20% sucrose solution and a pumpkin with 20 adult mealybugs. The remaining ant colonies were allowed access to mealybug‐infested pumpkins. Ant colonies with access to the sucrose solution attended mealybugs significantly less than those without additional sugar sources. Mealybug survival rates were similar under both treatments. Total body sugars and fructose were nearly twice as high in ants with access to honeydew and sucrose vs. those with access to honeydew and water. Fructose accumulated on the pumpkins over time in both treatments, suggesting that honeydew was not fully exploited by the ants. In conclusion, D. brevipes enjoy lower degrees of ant attendance when S. geminata have alternative sources of carbohydrates. We further discuss the significance of these findings for the conservation of predaceous ants and mealybug biological control.     

Ant–hemipteran association decreases parasitism of Phenacoccus solenopsis by endoparasitoid Aenasius bambawalei

1. Mutualism between ants and honeydew-excreting hemipterans is ubiquitous in the ecosystem. It is widely accepted that ant tending facilitates the colony growth of hemipterans by protecting them from predators and parasitoids. However, few studies have explored how ant tending helps defend against natural enemies. 2. Ghost ant Tapinoma melanocephalum and the invasive mealybug Phenacoccus solenopsis have a close mutual relationship. Previous studies have shown that ghost ant tending can definitely reduce parasitism and visit frequency of Aenasius bambawalei, the dominant endoparasitoid of P. solenopsis. However, the ghost ant workers seldom attack the parasitoids. It is still unclear how the ghost ant adversely affects parasitoids. This study explored the mechanism underlying the impacts of ants on natural enemies of the mealybugs. 3. Honeydew produced by P. solenopsis was an attractant to A. bambawalei. Parasitoids exhibited less searching activity, shorter longevity and lower parasitism when supplied with less honeydew. Aenasius bambawalei showed significant avoidance of pygidial gland secretions and visual cues of ghost ants. Parasitism in plants treated with 6-methyl-5-hepten-2-one, actinidine, and gland extracts was significantly lower than that in plants treated only with solvents (paraffin oil or double-distilled water). 4. It is concluded that honeydew consumption by ghost ants could negatively influence the performance of parasitoids. The pygidial gland secretions and visual cues of ghost ants also significantly inhibit the parasitism. These results may contribute to a better understanding of the regulation mechanism in ant–hemipteran–enemy interactions.     

Density-Dependent Benefits in Ant-Hemipteran Mutualism? The Case of the Ghost Ant Tapinoma melanocephalum (Hymenoptera: Formicidae) and the Invasive Mealybug Phenacoccus solenopsis (Hemiptera: Pseudococcidae)

Although density-dependent benefits to hemipterans from ant tending have been measured many times, few studies have focused on integrated effects such as interactions between ant tending, natural enemy density, and hemipteran density. In this study, we tested whether the invasive mealybug Phenacoccus solenopsis is affected by tending by ghost ants (Tapinoma melanocephalum), the presence of parasitoids, mealybug density, parasitoid density and interactions among these factors. Our results showed that mealybug colony growth rate and percentage parasitism were significantly affected by ant tending, parasitoid presence, and initial mealybug density separately. However, there were no interactions among the independent factors. There were also no significant interactions between ant tending and parasitoid density on either mealybug colony growth rate or percentage parasitism. Mealybug colony growth rate showed a negative linear relationship with initial mealybug density but a positive linear relationship with the level of ant tending. These results suggest that benefits to mealybugs are density-independent and are affected by ant tending level.     

Effects Of Tending By Solenopsis Invicta (Hymenoptera: Formicidae) On The Sugar Composition And Concentration In The Honeydew Of An Invasive Mealybug,Phenacoccus Solenopsis (Hemiptera: Pseudococcidae)

Mutualistic interactions between ants and hemipterans are mediated by the honeydew produced by the hemipterans. Previous works have demonstrated that the invasive mealybug Phenacoccus solenopsis produces abundant honeydew and attracts a large number of workers of the fire ant Solenopsis invicta. Mealybugs exhibit higher fecundity when tended by fire ants. The honeydew produced by P. solenopsis plays an important role in interactions between these two species. However, relatively few studies have focused on whether there is a cost to P. solenopsis mealybugs of being tended by S. invicta through changes in their excretion behavior and the quantity of honeydew produced. Our results indicated that the honeydew of P. solenopsis contains xylose, fructose, sucrose, trehalose, melezitose, and raffinose. The sugar concentration in the mealybug honeydew changed in an ant‐tended treatment. When tended by fire ants, the mealybugs generated honeydew with a significantly decreased xylose concentration. In contrast, the droplets showed a considerable increase in the melezitose concentration. P. solenopsis excreted honeydew more frequently when tended by S. invicta, but the weights of the droplets excreted by the ant‐tended mealybugs were significantly lower. In addition, S. invicta exhibited a significant preference for different sugars. Melezitose was visited more intensively than the other sugars in two choice tests. These results may suggest that, to attract more tending ants, mealybugs adjust their carbohydrate metabolism.     

Ant tending impairs performance of Aenasius bambawalei by manipulating the honeydew composition produced by Phenacoccus solenopsis

Honeydew produced by hemipterans is known as a possible kairomonal resource for parasitoids. The application of artificial honeydew effectively improves the performance of natural enemies. Aenasius bambawalei is a particularly dominant and aggressive endoparasitoid of the invasive mealybug Phenacoccus solenopsis. Our previous study showed that tending by the ghost ant Tapinoma melanocephalum significantly reduced the parasitism of A. bambawalei. We hypothesize that ghost ant tending influences host location of parasitoids by manipulating the composition of the honeydew produced by mealybugs. In this study, we tested whether the honeydew composition differs between treatments with and without ant attendance and whether changes in the honeydew influence the performance of A. bambawalei. Our results show that the sucrose concentration increased significantly in the ant‐attendance treatment but decreased when ant attendance was switched to an ant‐exclusion treatment; the inverse was true for the glucose concentration. Compared with the plastic honeydew treatment (mealybug with ant attendance), parasitoids spent much more time searching, had longer lifespans and showed higher parasitism on filter papers treated with natural honeydew (mealybug without any pre‐treatment) and those treated with convalescent honeydew (mealybug having experienced ant attendance and then switched to ant exclusion). These results support the hypothesis that ant tending influences the performance of parasitoids by manipulating honeydew composition.     

Parasitoids of obscure mealybug,Pseudococcus viburni (Hem.: Pseudococcidae) in California: establishment of Pseudaphycus flavidulus (Hym.: Encyrtidae) and discussion of related parasitoid species

To improve natural suppression of the obscure mealybug, Pseudococcus viburni (Signoret), the parasitoids Pseudaphycus flavidulus (Brèthes) and Leptomastix epona (Walker) (Hymenoptera: Encyrtidae) of Chilean origin were released in California's Central Coast vineyards from 1997 to 1999. A survey for parasitoids of P. viburni was conducted in the Edna Valley appellation wine grape region from 2005 to 2007, 6–8 years after classical biological control releases were discontinued. Two survey methods were used. First, field collections of obscure mealybugs from commercial vineyard blocks (2005–2007) and, second, placement of “sentinel mealybugs” on potted (1 L) grape vines (2006 only). From both survey methods, P. flavidulus was recovered, albeit levels of parasitism were low (less than 0.6%). We also placed longtailed mealybug, Pseudococcus longispinus (Targioni Tozzetti), on potted plants concurrent with placement of sentinel obscure mealybugs in the vineyard in order to measure parasitoid activity on this closely-related mealybug species. No P. flavidulus were recovered from P. longispinus. Other encyrtid parasitoids reared from either P. viburni or P. longispinus were Anagyrus pseudococci (Girault), Leptomastix dactylopii Howard, Leptomastidea abnormis (Girault), Coccidoxenoides perminutus Girault, and Tetracnemoidea peregrina (Compere). A hyperparasitoid, Chaetocerus sp., was also reared. The data are discussed with respect to biological control of vineyard mealybugs and newly developed controls for the Argentine ant, Linepithema humile (Mayr) (Hymenoptera: Formicidae). Because Pseudaphycus species reared from mealybugs are superficially very similar a taxonomic key and discussion of host relationships for selected Pseudaphycus species are provided.     

Host‐plant dissections reveal contrasting distributions of Crematogaster ants and their symbionts in two myrmecophytic Macaranga species

1. Ant–plant mutualisms are among the most widespread and ecologically important insect–plant interactions in the tropics. The multitrophic mutualism involving Macaranga plants (Euphorbiaceae) and Crematogaster ants (Formicidae) is the most diverse in Southeast Asia. This interaction also includes trophobiotic scale insects (Coccidae) and nematodes inhabiting ant refuse piles. 2. Here two myrmecophytic systems were compared, Macaranga trachyphylla with Crematogaster captiosa (Mt + Cc) and Macaranga beccariana with Crematogaster decamera (Mb + Cd), using a fine‐scale dissection of the stems. For the two plant species, for each internode, both contents (ants, coccids, refuse piles) and structure (internode height, numbers of open and occluded ant holes) were recorded. 3. There were significant patterns in the vertical distribution of ant colonies and their symbionts in the plant stems. Most coccids were kept in the highest sections of both systems, although Mb + Cd hosted a broader range of coccid species than Mt + Cc. Three nematode species were recorded, but with a rather low specificity to plant or ant species. Furthermore, the fine‐scale distribution showed aggregation of closed holes with ant brood and separation of nematode‐infested refuse piles from eggs. 4. The results of this study indicate that ants manipulate spatial colony structure via distribution of brood, holes and the symbionts. It is suggested that ants optimise the location of refuse piles and occluded holes via spatial heterogeneity in their distribution among internodes. This paper discusses the protective role of occluded holes and demonstrates some general interactions with other symbiotic fauna.     

Apparent facilitation of an invasive mealybug by an invasive ant

Summary In the southeast United States, the invasive ant Solenopsis invicta is known to derive important carbohydrate (honeydew) resources from mealybugs utilizing grasses. Most important appears to be an invasive mealybug, Antonina graminis. We studied whether this mealybug and a similar native species also benefit from association with S. invicta. We found that mealybug occurrence increases significantly with increasing proximity to S. invicta mounds, suggesting that mealybugs benefit as well. Mutual benefits derived by S. invicta and A. graminis are consistent with a hypothesis proposing that associations among invasive species can be important in their success at introduced locations.     

红火蚁入侵对棉花粉蚧近距离扩散的促进作用

[背景]红火蚁与棉花粉蚧入侵到同一地区,因侵入生境重叠而相遇,进而产生互惠关系。这种互惠关系对红火蚁、棉花粉蚧的生存、扩散传播、入侵的意义和作用规律、机制等是需要解释的生态学问题。[方法]采用田问试验生态学的方法,通过迁移红火蚁蚁巢、向扶桑上接粉蚧等观察研究了红火蚁入侵对棉花粉蚧近距离扩散的影响。[结果]随着离蚁巢距离的增大,扶桑上工蚁数量逐渐减少,距离为1．0、2．0m时数量较多;发生该蚁区域距蚁巢2．0、3．0m扶桑感染粉蚧比率（75％、45％）显著高于无红火蚁区（25％、10％）,其感染比率与工蚁数量呈显著正相关,符合方程Y=0．0042X＋0．1992。[结论与意义]红火蚁入侵促进了棉花粉蚧的近距离扩散,扩散范围在2—3m。研究结果可为深入了解红火蚁与棉花粉蚧协同入侵规律等提供支持。     

Does the host plant affect the benefits from mutualisms? The invasive mealybug and ghost ant association

1. The benefits to trophobionte hemipterans are affected by the ant tending level, which is a widely accepted statement. The ant tending level is closely related to multiple factors. It is clear that the ant tending level can be affected by the temporal factor, age‐specific, the density of the hemipterans, and quantity and quality of honeydew produced by hemipterans. 2. Few studies of ant–hemipteran mutualisms have reported the patterns of host plants‐dependent effects, and whether host plants influence the ant tending level that is also unclear. As such, laboratory experiments were conducted to test whether the colony growth rate of an invasive mealybug Phenacoccus solenopsis Tinsley, parasitism of Aenasius bambawalei Hayat, an dominant parasitoid of P. solenopsis, are affected by tending by ghost ants (Tapinoma melanocephalum(Fabricius)], host plants (tomato and cotton), and interactions between the two factors. The difference in the ant tending level between the host plants was also determined. 3. The results showed that mealybug colony growth and parasitism were significantly affected by ant tending and host plant separately. There were significant interactions between the independent factors on the mealybug colony growth rate and percentage parasitism. These results suggest that benefits to mealybugs are host plant‐dependent.     

Host location and discrimination mediated through olfactory stimuli in two species of Encyrtidae

The role of olfactory stimuli in host detection and evaluation was studied in two encyrtid Hymenoptera. The first, Epidinocarsis lopezi De Santis, is a monophagous parasitoid of the cassava mealybug Phenacoccus manihoti Matile-Ferrero, itself feeding exclusively on cassava, Manihot esculenta Crantz. The second, Leptomastix dactylopii Howard, is a monophagous parasitoid of the Citrus mealybug, Planococcus citri Risso, but this latter is highly polyphagous. The behaviour of females of both parasitoids (attaction and locomotion) was compared in a tubular olfactometer for the odours of their respective hosts on cassava and poinsettia. Tests were made using: 1) healthy host-plant alone; 2) host-plant infested with unparasitized mealybugs; 3) unparasitized mealybugs only; 4) host-plant infested with parasitized mealybugs and 5) parasitized mealybugs only. Only E. lopezi was attracted by the odour of the host-plant alone, but both species were attracted by the odour of an infested host-plant and that of unparasitized mealybugs. The odour of parasitized mealybugs, alone or on host-plant, induced an undirected activity. The attraction of E. lopezi to the odour of the host-plant alone could be linked to the monophagous diet of its host, whereas the attraction of the two species of parasitoids to the odours of infested host-plants and unparasitized mealybugs could be due to the fact that both parasitoids are specialists. The behavioural response of both species to the odour of parasitized mealybugs revealed a new aspect in host discrimination: the identification of parasitized hosts could be partly mediated through olfactory stimuli, and not only through gustatory stimuli.     

Success of aphid parasitoids and their hosts varies with ant attendance: A field study

Aphid–ant associations are often described as mutually beneficial interactions in which honeydew is traded for protection from predators and parasitoids. The aim of the present study was to determine parasitization avoidance in ant‐tended aphid colonies. Field experiments were carried out on two host plants: hoary cress Lepidium draba (Brassicaceae) and Canadian teasel Cirsium arvense (Asteraceae). Lepidium was host to Acyrthosiphon gossypii (Aphididae) tended by the ant Lasius turcicus (Formicidae) and attacked by two parasitoids, Trioxys asiaticus (Braconidae) and Lysiphlebus fabarum (Braconidae). Cirsium was host to Brachycaudus cardui tended by Crematogaster sordidula and attacked by L. fabarum and Aphidius colemani (Braconidae). The per capita population growth rate of A. gossypii was significantly higher in the presence of ants, while B. cardui was negatively affected, albeit non‐significantly. The parasitism rate of A. gossypii decreased significantly when tended by Lasius turcicus, but the presence of Crematogaster sordidula in colonies of B. cardui significantly increased parasitism. Our results indicate that the effects of ant attendance vary between different aphid–ant interactions. Moreover, parasitoids can benefit from the presence of ants under some conditions.     

A quantified ethogram for oviposition behavior and oviposition preference in the hemipterophagous butterfly Spalgis epius (Westwood) (Lepidoptera: Lycaenidae)

In this study, we examined the oviposition behavior and preference of Spalgis epius, a potential predator of mealybug crop pests. An ethogram of oviposition behavior was constructed based on observations made in an oviposition cage. Ovipositional behavioral acts were catalogued and separated into two behavioral repertoires: searching and egg laying. Gravid females of S. epius oviposited similar numbers of eggs on three mealybug species. Females preferred eggs and adults to nymphs of mealybugs for oviposition. Among three species of mealybugs attended by ants, females laid fewer eggs in the mealybug mass attended by Oecophylla smaragdina than on mealybugs attended by Tapinoma melanocephalum and Camponotus variegatus. Females preferred mealybug masses already containing conspecific eggs to mealybug masses containing conspecific larvae or Cryptolaemus montrouzieri larvae for egg deposition. Gravid females laid larger numbers of eggs under bright sunlight than in diffused sunlight or shade. The results of this study showed that S. epius can effectively attack any species of mealybugs, avoid intra- and interspecific competition, and co-exist with some species of ants attending mealybugs. With the knowledge of these behaviors, this predator can be effectively utilized as a major biological control agent of mealybugs.     

Surrogate Habitats Demonstrate the Invasion Potential of the African Pugnacious Ant

Many ant species are highly invasive and are a significant component of disturbed ecosystems. They can have a major suppressive effect upon indigenous invertebrates, including other ants. Despite overwhelming circumstantial evidence for the ecological resourcefulness of many ants, there appears to be no experimental evidence illustrating the habitat breadth of a potentially invasive ant species. We demonstrate here that a particularly opportunistic and locally dominant ant Anoplolepis custodiens, which is a major indigenous African pest, overrides habitat structure to maintain its population level. We compared A. custodiens activity, morphology, foraging behaviour and ant species diversity in artificially established surrogate habitats (cover crops) in a vineyard containing an ample food resource in the form of the honeydew-producing mealybug Planococcus ficus. These cover crops were chosen so as to create highly altered habitats. The ant's ability to overcome these potentially suppressive habitat conditions hinged on its tight mutualism with the mealybug, and on its chasing away mealybug parasitoids. This ant species is predicted to be a latent invasive beyond Africa. It is unlikely to be impeded once it has established a foothold in a variety of novel habitats. It could locally invade to obtain food resources in a wide range of habitat types. Furthermore, in agricultural systems, cover crops are unlikely to control such an ant. Potential invasives such as this ant should be flagged as important quarantine suspects.     

Interactions between the ant Tapinoma nigerrimum (Hymenoptera: Formicidae) and the main natural enemies of the vine and citrus mealybugs (Hemiptera: Pseudococcidae)

Honeydew-excreting hemipterans, such as mealybug pests, can be protected from their natural enemies by tending ants in return for honeydew, thereby compromising the aims of biological control. In this respect, antagonistic interactions between the ant Tapinoma nigerrimum, native to the Mediterranean basin, and the main natural enemies of both the vine mealybug (VMB), Planococcus ficus, and the citrus mealybug (CM), Planococcus citri, were assessed in laboratory conditions. Parasitism of vine and CMs by their respective parasitoids, Anagyrus sp. nr. pseudococci and Leptomastix dactylopii, was negatively affected by the ant T. nigerrimum. Similarly, T. nigerrimum was shown to significantly disrupt the predatory potential of ladybird larvae, Cryptolaemus montrouzieri, when foraging on host CMs. By contrast, the presence of the ant did not negatively influence the predatory activity of C. montrouzieri adults when feeding on CMs. Consequently, the encyrtid parasitoids A. sp. nr. pseudococci and L. dactylopii and the larval stage of the predator C. montrouzieri may be considered as T. nigerrimum-sensitive, whereas the adults of C. montrouzieri may be regarded as T. nigerrimum-resistant predators. Accordingly, the ant T. nigerrimum constitutes a threat to the biological control of mealybugs by either the encyrtids A. sp. nr. pseudococci and L. dactylopii or the larval stage of the ladybird C. montrouzieri. Hence, adequate control of T. nigerrimum is highly recommended before any release of these mealybugs' natural enemies.     

Classical biological control program for the mealybug Oracella acuta in Guangdong Province,China

Abstract The mealybug Oracella acuta, native to the southeastern US, was accidentally introduced into slash pine plantations in Guangdong Province in China in 1988. A classical biological control program was initiated in 1995, and the parasitoids Allotropa oracellae, Acerophaus coccois, and Zarhopalus debarri were imported from the US. A total of 19 972 parasitized mealybugs were shipped to China from 1996–2004, from which 15 430 wasps emerged, 12 933 of which were the three target species. Efforts to establish a mass-rearing program for the parasitoids in China failed. Five field release sites were established, and 6 020 parasitoids were released. Only 118 individuals of the three imported species were collected during establishment checks, although several wasps were collected 1–2 years after the last parasitoid release. Over 2 000 Anagyrus dactylopii, a cosmopolitan parasitoid, emerged from the parasitized mealybugs collected, a majority from the Taishan area near the site of the original introduction of O. acuta. To date the imported parasitoids have failed to establish, and natural enemies have not noticeably reduced mealybug populations.     

Behavioural and chemical mechanisms behind a Mediterranean ant–ant association

1. Interspecific competition among ants is common, and so is competitive exclusion among dominant ant species. In contrast, specific associations between non‐parasitic ant species are rare, especially in the temperate zones. As an exception, the subordinate ant Camponotus lateralis frequently co‐occurs with the dominant Crematogaster scutellaris but rarely with other dominant ants. 2. This association is one of various associations between Camponotus and Crematogaster species across the world. However, the mechanisms behind these co‐occurences are largely unknown. 3. In the present study, we therefore investigated the association of Ca. lateralis and Cr. scutellaris. We studied the spatial association of the nests, interspecific aggression, both species' cuticular hydrocarbon profiles, and their propensity to follow the other species' pheromone trails. 4. Crematogaster scutellaris usually attacked and displaced the generally submissive Ca. lateralis, but was significantly less aggressive at jointly used trails. Camponotus nests were always in close proximity to Crematogaster nests. 5. The cuticular hydrocarbons of both species consisted of alkanes with chain lengths between C21 and C35. The two species had 25 hydrocarbons in common, including mono‐, di‐, and tetramethyl alkanes. Despite this qualitative similarity, however, the quantitative hydrocarbon composition differed between the two species. 6. Camponotus lateralis followed artificial trails containing trail pheromones of Cr. scutellaris, but the latter did not follow Ca. lateralis trail pheromones. Interspecific trail‐following by Camponotus, but not vice versa, has been observed in another Camponotus–Crematogaster association and may be a more general mechanism that facilitates associations between the two ant genera.     

Aggressiveness of ants attracted to the extrafloral nectary‐bearing plant,Mallotus japonicus,and temporal fluctuations in their abundance

Many plants that bear extrafloral nectaries (EFNs) attract various ant species that can exclude herbivores. The aggressiveness of the attracted ants and their temporal activity patterns are important factors that can affect the efficiency of herbivore exclusion from the plant. However, the characteristics of this mutualistic relationship between EFN‐bearing plants and ants have not been sufficiently elucidated. We investigated the aggressiveness of six ant species against the common armyworm, Spodoptera litura Fabr., and temporal fluctuations in the abundance of four aggressive ant species on an EFN‐bearing plant, Mallotus japonicus (L.f.) Müll. Arg. Workers of Crematogaster teranishii Santschi, Pheidole noda Smith, Pristomyrmex punctatus Smith and Formica japonica Motschoulsky were observed to be highly aggressive. In contrast, workers of Camponotus vitiosus Smith showed low aggressiveness. Paratrechina flavipes Smith workers did not attack the herbivore. The activity patterns of the four aggressive ant species greatly differed. Crematogaster teranishii and Ph. noda workers were constantly active throughout the day and night. In contrast, F. japonica was diurnal. Pristomyrmex punctatus was principally nocturnal. Formica japonica workers foraged solitarily, whereas workers of the other three species foraged in a group or recruited nestmates. Our results suggest that the efficacy of the indirect defense in M. japonicus depends principally on the attracted ant species.     

Spatio‐temporal patterns and interactions with honeydew‐producing Hemiptera of ants in a Mediterranean citrus orchard

• 1 The role of ants in the citrus agro‐ecosystem is controversial and understanding their ecology may help to clarify their function. The present study determined the daily and seasonal foraging patterns, the spatial distribution, the feeding sources and the associations with honeydew‐producing Hemiptera of three ant species that forage in citrus canopies.
• 2 The dominant ants Pheidole pallidula (Nylander) (Myrmicinae) and Lasius grandis Forel (Formicinae) foraged in mutually exclusive territories within the field, although they both shared their territory with the subordinate Plagiolepis schmitzii Forel (Formicinae), a distribution pattern known as ‘ant mosaic’.
• 3 The observed mean overlap for the spatial distribution was significantly lower than the generated by null models, providing strong evidence of spatial interspecific competition, especially between the two dominants.
• 4 Ants ascended to the canopies from April until November. Colony nutritional requirements and temperature probably shape their seasonal foraging patterns. The daily activity pattern of P. schmitzii was strictly diurnal, whereas L. grandis and P. pallidula were active during the entire day.
• 5 The ants' diet in the canopies consisted principally of hemipteran honeydew, whereas citrus nectar and predation/scavenging did not represent important food sources. More than 60% of the total honeydew sources and 100% of the citrus mealybug Planococcus citri colonies were tended by ants during spring and summer.