DOI: 10.13102/sociobiology.v61i2.136-144Sociobiology 61(2): 136-144 (June, 2014) Open access journal: http://periodicos.uefs.br/ojs/index.php/sociobiology ISSN: 0361-6525 Response of Ants to the Leafhopper Dalbulus quinquenotatus DeLong & Nault (Hemiptera: Cicadellidae) and Extrafloral Nectaries Following Fire G. Moya-Raygoza¹, K.J. Larsen² Introduction Ants (Hymenoptera: Formicidae) often live in mutua- listic relationships with trophobiont insects that excrete honey- dew, or with plants bearing extrafloral nectaries (EFNs) that produce nectar (Hölldobler & Wilson, 1990). Some species of aphids, whiteflies, scale insects, mealybugs, treehoppers, and leafhoppers (Hemiptera) live in facultative or obligatory mutualistic relationships with ants (Way, 1963; Buckley, 1987; Blüthgen et al., 2006; Gibb & Cunningham, 2009; Fagundes et al., 2013). In these associations, the insect provides honey- dew, a sugary excretion of carbohydrates, amino acids, and water for the ants, whereas ants protect the hemipterans from natural enemies (Delabie, 2001; Heil & Mckey, 2003; Zhang et al., 2012; Zhang et al., 2013). Plants in over one hundred families bear EFNs that produce secretions rich in sugars, Abstract Previous investigations of mutualistic associations between ants and plants bearing extrafloral nectaries (EFNs) or between ants and trophobiont leafhoppers have studied these relationships separately, but nothing is known on how ant abundance responds to these two food resources occurring in the same habitat when that habitat is disturbed by fire. The objectives of this study are to document ant abundance with the tropho- biont five-spotted gamagrass leafhopper, Dalbulus quinquenotatus DeLong & Nault, and with EFNs on trees of Acacia pennatula (Schlecht & Cham.) Benth. (Fabaceae) that occur in the same habitat, and how ant abundance in both of these mutualisms is affected after disturbance by fire. This study was performed at several sites in central Mexico where the perennial gamagrass Tripsacum dactyloides L. (Gramminae) and A. pennatula both occur. More ants were collected in association with the leafhopper D. quinquenotatus than with EFNs of A. pennatula. At sites where dry season fire occurred, new green leaves were produced by both T. dactyloides and A. pennatula after the burn. On these new leaves after fire, significantly more ants tended D. quinquenotatus leafhoppers on T. dactyloides than visited EFNs on A. pennatula. In burned sites the ants Anoplolepis gracilipes Smith, Brachymyrmex obscurior Forel and Pheidole sp. live in association with the leafhoppers, whereas EFNs on A. pennatula were associated with the ants A. gra- cilipes, B. obscurior, Camponotus sp., Crematogaster sp. and Solenopsis sp. Sociobiology An international journal on social insects 1 - Universidad de Guadalajara, CUCBA, Jalisco, Mexico. 2 - Department of Biology, Luther College, Decorah, USA. Article History Edited by Gilberto M M Santos, UEFS, Brazil Received 16 March 2014 Initial acceptance 24 April 2014 Final acceptance 16 May 2014 Keywords Acacia pennatula, five-spotted gamagrass leafhopper, Tripsacum dactyloides, mutualism Corresponding author Gustavo Moya-Raygoza, Ph.D. Departamento de Botánica y Zoología CUCBA, Universidad de Guadalajara Km 15.5 carretera Guadalajara-Nogales Las Agujas, Zapopan, C.P. 45110 Apdo. Postal 139, Jalisco, México E-mail: moyaraygoza@gmail.com amino acids, and lipids that attract ants, and in return these ants protect those plants from herbivores (González-Teuber & Heil, 2009; Byk & Del-Claro, 2011; Marazzi et al., 2013; Weber & Keeler, 2013). Ants are attracted to high quality sugar resources as food (Heil & McKey, 2003). Previous studies have shown that when both honeydew and extrafloral nectar are offered to ants, ants are more abundant at the honeydew rather than at exudates of EFNs (Fiala, 1990; Rashbrook et al., 1992; Del- Claro & Oliveira, 1993; Blüthgen et al., 2000; Katayama et al., 2013). Ants were more abundant tending the hemipterans, particularly when greater numbers of hemipterans are present be- cause of the larger quantities of honeydew produced (Kataya- ma & Suzuki, 2010). Blüthgen et al. (2000) found greater numbers of ants at honeydew resources as opposed to EFN resources because honeydew is apparently a higher quality RESEARCH ARTICLE - ANTS Sociobiology 61(2): 136-144 (June, 2014) 137 food resource, rich in amino acids. Moreover, Katayama and Suzuki (2003) demonstrated that if an aphid colony increases in size, ants stop using EFNs and strengthen their mutualistic association with aphids. Fire affects the growth of plants because some peren- nial species such as grasses and plants bearing EFNs quickly re-grow after disturbance occurs. New leaves formed after the plants burn are ready to be colonized directly or indirectly by ants, often attracted to food resources such as the honeydew produced by leafhoppers that feed on young grasses (Moya- Raygoza, 1995) or from nectar produced by EFNs (Alves- Silva & Del-Claro, 2013). Ants respond to burned plants with EFNs or hemipterans in the same way. The abundance of ants increased on the shrub Banisteriopsis campestris (A. Juss.) which bears EFNs after fire, mainly because of concentrated extrafloral nectar (Alves-Silva & Del Claro, 2013). Alves- Silva (2011) and Koptur et al. (2010) also found a richer ant community guarding plants from herbivory after fire because of the production of extrafloral nectar. Similarly, higher num- bers of ants were found tending the honeydew-producing fivespotted gamagrass leafhopper, Dalbulus quinquenotatus DeLong & Nault, after its host plant, the perennial gamagrass Tripsacum dactyloides L. (Gramminae), was burned (Moya- Raygoza, 1995). Mutualisms between ants and EFNs-bearing plants and ants and trophobiont hemipterans have been investigated separately after disturbance by fire, but little is known how ant abundance responds to these two food resources when present in the same habitat. This study was performed in central Mexico, where the perennial gamagrass T. dactyloides hosts D. quinquenotatus leafhoppers and trees of Acacia pen- natula (Schlecht & Cham.) Benth. (Fabaceae) with EFNs oc- cur together in the same habitats (Fig. 1a). These sites often are accidentally burned, and the fire often kills or drives away insects living on those plants. Dalbulus quinquenotatus lives on the basal leaves of T. dactyloides in an obligatory mutua- lism with tending ants (Larsen et al., 1991). Ants tending D. quinquenotatus receive honeydew and protect this leafhop- per from natural enemies (Moya-Raygoza & Nault, 2000). In contrast, A. pennatula have EFNs and live in a mutualistic re- lationship with ants (Moya-Raygoza, 2005), providing nectar for the ants in return for protection from herbivory. Fire is an important abiotic factor in mutualisms be- cause it affects plant re-growth and the abundance of ants that depend on exudates produced indirectly by trophobiont insects and directly by EFNs. When fire consumes the foliage of both T. dactyloides and A. pennatula, the mutualisms involving ants with both species are temporarily disrupted. However, only a few days after being burned, new leaves of both plant species begin to re-grow (Fig. 1b) and are soon colonized by D. quinquenotatus and ants in the case of T. dactyloides, or by ants visiting EFNs in the case of A. pennatula. Measur- ing the total abundance of ants collecting honeydew from D. quinquenotatus and visiting EFNs resources before and after the host plants are burned helps us understand the ecological importance of mutualisms that can be strong driving forces for community organization (Wimp & Whitham, 2001). The objectives of this study are to document ant abundance with D. quinquenotatus leafhoppers and EFNs in the same habi- tat, and how ant abundance in both of these mutualisms is affected after disturbance to their habitat by fire. Materials and Methods Study system Nine field sites containing both T. dactyloides and A. pennatula were chosen for this study. Each site had both spe- cies of plant present and covered an area of 0.05-0.25 ha. All sites were in the state of Jalisco in Central Mexico. The sites were: 1) El Arenal: 1,501 m elev, 20°46.032´N, 103°40.766´W; 2) Los Chorros: 1,371 m elev, 20°41.211´N, 103°41. 558´W; 3) San Isidro: 1,266 m elev, 20°49.014´N, 103°20. 262´W; 4) Agua Caliente: 1,385 m elev, 20°25.770´N, 103°41.485´W; 5) Cocula: 1,273 m elev, 20°25.595´N, 103°44.601´W; 6) San Agustin: 1,638 m elev, 20°30.682´N, 103°28.796´W; 7) La Mimila: 1,649 m elev, 20°44.411´N, 103°37.686´W; 8) El Molino: 1,608 m elev, 20°23.938´N, 103°32.760´W; and 9) Zapopan: 1,631 m elev, 20°44.283´N, 103°30.805´W (Fig. 1c). The closest sites were 5.45 km apart (Agua Caliente and Cocula) while the most distant sites (San Isidro and Cocula) were 60.44 km apart. All sites had similar habitat character- istics and belong to pine-oak ecosystem (Rodríguez-Trejo & Myers, 2010). Plants of both species live on steep slopes or beside roadways and grow on limestone soils (Wilkes, 1972). The sites had similar vegetation consisting of a plant commu- nity containing T. dactyloides interspersed with A. pennatula trees and few other plants such Lysilona sp. Each T. dactyloides population was composed of scattered clumps consisting of clusters of stems. Tripsacum dactyloides can use rhizomes to spread across the landscape and does not possess extrafloral nectaries. Moreover, ants are present on T. dactyloides only when the plants are hosts for D. quinquenotatus leafhoppers as compared with plants without D. quinquenotatus (Larsen, et al. 1991). All sites were sampled to determine the numbers of ants when leafhoppers and EFNs were available. Acacia pennatula has actively secreting extrafloral nectaries on young leaves primarily from April to June (McVaugh, 1987; Moya-Raygoza, 2005), whereas leafhoppers are present on T. dactyloides pri- marily during the wet season from June to September (Moya- Raygoza, 1995) when these habitats are not burned. We ob- served that when the habitats were burned, both plant species started to produce new green leaves within several days, and this altered the food resources available for visiting ants. Fires generally occur from March to May towards the end of each dry season. The dry season in Jalisco generally occurs from October to May and is characterized by lower rainfall, lower Moya-Raygoza and Larsen - Ants, leafhoppers and EFNs after fire138 temperatures and shorter days as compared with the wet sea- son which typically lasts from June to September (Mosino- Aleman & Garcia, 1974). After burning, both honeydew and EFN nectar food resources for ants are found in May and June within the same plant community. The highest nectar secretion rates have been documented from EFNs on young leaves of damaged plants (Heil et al., 2004), while high numbers of D. quinquenotatus leafhoppers have been found on T. dactyloides after fire (Moya-Raygoza, 1995). No data were collected be- tween October and April because ants do not visit either of these food resources during that time. The wet season begins in June, and no fires occur once the rains begin to fall. Sampling We confirmed the presence of ants associated with EFNs of A. pennatula and D. quinquenotatus at each site. Once these fires took place, we sampled ants on burned and unburned sites. We selected A. pennatula trees at each site and neighboring clumps of T. dactyloides. Ten terminal branches on each selected A. pennatula tree and one basal leaf from each of ten different T. dactyloides clumps were randomly selec- ted. Terminal branches of A. pennatula were selected because the highest concentration of EFNs occurs on these branches, whereas basal leaves of T. dactyloides were selected because this is where the highest numbers of D. quinquenotatus are found. We collected all nymphs and adults of D. quinquenotatus leaf- hoppers and all tending ants from the basal 10 cm of each selected T. dactyloides stem. All EFNs were counted and ants collected from the ter- minal 10 cm of each selected A. pennatula branch. Therefore ant abundance at each resource was quantified on one stem or branch for each of 10 separated plants of T. dactyloides and A. pennatula by site. We selected the same 10 cm surface on both plant species to have approximately the same area of food resource available for the ants. Sampling at all sites was performed between 09:00 and 14:00 h, one site per day during the last week of May 2007, first week of June 2012, and the second week of September 2012. The Arenal and Los Chorros sites were burned in May 2007, while the Zapopan and Los Chorros sites were burned in June 2012. Dalbulus quinquenotatus, EFNs and ants were sampled approximately one month after each fire. Ants were sampled at these times because both extrafloral nectar produced by A. pennatula and honeydew produced by D. quinquenotatus was present. All collected insects were stored in 70% ethanol and returned to the lab for identification. Analysis of Deviance, using R.3.1.0 for Windows (R Project), was performed to evaluate the interaction (resource for ants, honeydew-extrafloral nectar × disturbance, fire-with- out fire) on the number of ants. This comparison included the ant abundance obtained on the three sampling dates. Further- more, the total number of ants tending D. quinquenotatus on T. dactyloides was compared vs the total number of ants on A. pennatula bearing EFNs with a Wilcoxon test using SPSS 12 for Windows. Therefore a comparison of ant abundance at leafhoppers vs EFNs was conducted when combining both burned and unburned resources in the three sample dates. Average and standard error were determined for the number of D. quinquenotatus nymphs and adults, tending ants, and EFNs for each burned and unburned site. Fig 1. Tritrophic interaction trophobiont five-spotted gamagrass leafhopper-Ants-Acacia pennatula. A) Hillside in Jalisco, Mexico late in the dry season covered with T. dactyloides hosting D. quin- quenotatus leafhoppers, interspersed with trees of A. pennatula bearing EFNs in unburned site. B) Young green leaves growing on A. pennatula (left) and T. dactyloides (right) several days after being burned by fire. C) Location of field sites containing both T. dacty- loides and A. pennatula from the state of Jalisco in central Mexico. A B C Sociobiology 61(2): 136-144 (June, 2014) 139 Results Ant species collected from burned T. dactyloides as- sociated with the leafhopper D. quinquenotatus were Anop- lolepis gracilipes Smith, Pheidole sp., and Brachymyrmex ob- scurior Forel. Ants found on unburned T. dactyloides tending D. quinquenotatus included A. gracilipes and B. obscurior. Greater ant species richness was associated with A. pennatula. Anoplolepis gracilipes, B. obscurior, Camponotus sp., Cre- matogaste sp. and Solenopsis sp. were found at EFNs when A. pennatula was burned. Ant taxa found visiting the EFNs of unburned A. pennatula included A. gracilipes, B. obscurior, Crematogaster sp., Dorymyrmex sp. and Pheidole sp. We ob- served these ants differed in body size and likely collect and store honeydew or extrafloral nectar differently. New green leaves were produced by both T. dactyloides and A. pennatula after they were burned. Disturbance by fire does not have the same effect on the numbers of ants tending D. quinquenotatus and visiting EFNs. We found an interaction between fire and plant species, and significantly more ants were found tending D. quinquenotatus leafhoppers on T. dactyloides than visiting EFNs (Z = 7.63; P = 0.001). Rapid colonization of new growth on T. dactyloides by ants and leafhoppers was ob- served after burning in the last week of May 2007. At this time only adult leafhoppers were observed in the two burned sites tended by a great number B. obscurior ants, while EFNs were visited by few ants of Solenopsis sp. at the two burned sites (Table 1 and Fig. 2). In June 2012, leafhoppers were tended by Pheidole sp. and a great number of nymphs were tended by great numbers of B. obscurior ants at the two 2012 burned sites (Table 2 and Fig. 3). Near the end of the wet season in September 2012, four months after the June fire, a large number of leafhopper nymphs were tended by larger numbers of B. obscu- rior ants, while low numbers of A. gracilipes, Camponotus sp., Fig 2. Average number (± standard error) of leafhoppers, ants tend- ing D. quinquenotatus leafhoppers on T. dactyloides, EFNs, and ants visiting EFNs on A. pennatula from burned and unburned sites in Jalisco, Mexico in the last week of May 2007. Crematogaster sp. and B. obscurior ants visited the EFNs at the two burned sites (Table 3 and Fig. 4). The number of ants tending leafhoppers was signifi- cantly higher than the number of ants found visiting EFNs of A. pennatula when combining both burned and unburned resources in the three sample dates (Wilcoxon = 299.50; Z = 3.04; P = 0.002). Leafhoppers and ants were found together at the end of the dry season in May 2007 on the six unburned sites, while only in two of the six unburned sites ants visited the EFNs of A. pennatula (Table 1 and Fig. 2). In June 2012, at the end of the dry season, no ants or leafhoppers were found on the leaves of unburned T. dactyloides plants that were dried out (Table 2 and Fig. 3). In September 2012, at the end of the wet season, only in one of the four unburned sites ants visited the EFNs of A. pennatula, whereas in these four unburned sites ants tended the leafhoppers (Table 3 and Fig. 4). Table 1. Average number (± standard error) of Dalbulus quinquenotatus nymphs, adults, and tending ants (and species of tending ant), Acacia pennatula EFNs, and ants on 10 stems and 10 branches of T. dactyloides and A. pennatula respectively in burned (in May 2007) and unburned sites at locations in Jalisco, Mexico at the end of the dry season in May 2007. Site Ant/Leafhopper interaction on Tripsacum dactyloides Ant/Acacia interaction Dalbulus quinquenotatus Ants Ant species A. pennatula EFNs Ants Ant SpeciesNymphs Adults Both resources burned 1. Arenal 0 9.2 ± 1.7 7.7 ± 1.4 B. obscurior 5.6 ± 0.1 1.0 ± 0.4 Solenopsis sp. 2. Los Chorros 0 1.4 ± 0.3 20.4 ± 5.7 B. obscurior 6.9 ± 0.3 0 - Both resources unburned 3. San Isidro 3.9 ± 2.2 1.7 ± 0.7 15.1 ± 5.7 B. obscurior 5.8 ± 0.2 0 - 4. Agua Caliente 0.7 ± 0.5 0.5 ± 0.4 1.7 ± 1.3 B. obscurior 5.4 ± 0.1 1.0 ± 0.2 Pheidole sp. 5. Cocula 10.2 ± 1.2 1.2 ± 0.4 4.3 ± 0.8 B. obscurior 5.4 ± 0.1 0 - 6. San Agustin 6.5 ± 3.3 1.9 ± 1.1 13.0 ± 5.1 B. obscurior 6.6 ± 0.1 0 - 7. La Mimila 3.7 ± 1.9 1.1 ± 0.5 5.7 ± 4.3 B. obscurior 5.5 ± 0.2 0 - 8. El Molino 2.0 ± 0.6 3.9 ± 1.2 12.1 ± 2.9 B. obscurior 6.6 ± 0.1 5.8 ± 1.4 B. obscurior Moya-Raygoza and Larsen - Ants, leafhoppers and EFNs after fire140 Table 2. Average number (± standard error) of Dalbulus quinquenotatus nymphs, adults, and tending ants (and species of tending ant), Acacia pennatula EFNs, and ants on 10 stems and 10 branches of T. dactyloides and A. pennatula respectively in burned (in June 2012) and unburned sites at locations in Jalisco, Mexico in June 2012. Site Ant/Leafhopper interaction on Tripsacum dactyloides Ant/Acacia interaction Dalbulus quinquenotatus Ants Ant species A. pennatula EFNs Ants Ant Species Nymphs Adults Both resources burned 1. Zapopan 4.4 ± 2.3 2.4 ± 0.7 4.8 ± 1.8 Pheidole sp. 5.6 ± 0.1 1.9 ± 0.2 A. gracilipes Camponotus sp. 2. Los Chorros 37.0 ±17.9 1.1 ± 0.5 38.9 ± 16.2 B. obscurior 5.8 ± 0.3 1.5 ± 0.4 Crematogaster sp. B. obscurior Both resources unburned 3. San Isidro 0 0 0 - 6.5 ± 0.4 2.5 ± 0.5 Dorymyrmex sp. Crematogaster sp. 4. San Agustin 0 0 0 - 6.2 ± 0.4 0.3 ± 0.1 B. obscurior 5. La Mimila 0 0 0 - 1.4 ± 0.7 1.1 ± 0.5 A. gracilipes 6. El Arenal 0 0 0 - 6.3 ± 0.4 3.3 ± 0.9 A. gracilipes Discussion The exudates honeydew and extrafloral nectar are key factors determining the abundance of ants when both food resources for ants are present (Buckley, 1983; Fiala, 1990; Rashbrook et al., 1992; Del-Claro & Oliveira, 1993; Blüt- hgen et al., 2006; Katayama et al., 2013). Considering the abundance of ants tending the leafhopper D. quinquenotatus compared with the abundance of ants visiting EFNs, more ants were collected in association with D. quinquenotatus than with EFNs on A. pennatula. This finding is similar to the results of other studies (Fiala, 1990; Rashbrook et al., 1992; Del-Claro & Oliveira, 1993; Blüthgen et al., 2000; Katayama & Suzuki, 2003; Katayama & Suzuki, 2010; Katayama et al., 2013) comparing ant abundance at honeydew-producing in- sects with plants with EFNs in non-disturbed conditions. In the rainforest canopy, ants are usually more abundant at honey- dew than extrafloral nectar, as honeydew is apparently a more valuable resource to ants than nectar from EFNs (Blüthgen et al., 2000). Ants (Camponotus sp.) also did not stop tending the honeydew-producing membracids (Guayaquila xiphias Fabricius) when an alternative EFN sugar source was avail- able on Didymopanax vinosum (Cham. & Schltdl.), their host plant (Del-Claro & Oliveira, 1993). Recently Katayama et al. (2013) demonstrated that the ant Lasius japonicus Santsci switches from visiting EFNs on the bean plant Vicia faba L. to the aphid Aphis craccivora Koch, because the density and total food reward to ants from the aphids exceed that from EFNs. We ascribe the difference in abundance between ants visiting the leafhopper D. quinquenotatus and EFNs on A. pennatula to several factors. First, D. quinquenotatus leaf- Fig 3. Average number (± standard error) of leafhoppers, ants tend- ing D. quinquenotatus leafhoppers on T. dactyloides, EFNs, and ants visiting EFNs on A. pennatula from burned and unburned sites in Jalisco, Mexico in the first week of June 2012. Fig 4. Average number (± standard error) of leafhoppers, ants tend- ing D. quinquenotatus leafhoppers on T. dactyloides, EFNs, and ants visiting EFNs on A. pennatula from burned and unburned sites in Jalisco, Mexico in the second week of September 2012. Sociobiology 61(2): 136-144 (June, 2014) 141 hoppers produce honeydew at a consistent rate (Larsen et al.,1992), whereas EFNs are highly variable in nectar produc- tion over the course of a day, resulting in a less predictable re- source for the ants. For example, nectar production is highly variable in the plant Macaranga tanarious (L.) Muell. Arg. (Heil et al., 2000). Second, D. quinquenotatus is sedentary and gregarious (Heady & Nault 1985), resulting in a higher density of both nymphs and adult leafhoppers on the basal leaves of T. dactyloides. At higher leafhopper densities, more honeydew is produced in a concentrated area allowing easy collection by the ants. Third, D. quinquenotatus responds to the stroking of their abdomen by antennae of tending ants by excreting and holding honeydew droplets until droplets are removed by ants (Larsen et al., 1992). Ant-tended Dalbulus quinquenotatus leafhoppers secrete three to six times the volume of honeydew compared with other species of non- myrmecophilous Dalbulus leafhoppers (Larsen et al., 1992), increasing the availability of honeydew for tending ants. In contrast, EFNs of A. pennatula do not respond to antennation by ants by increasing extrafloral nectar secretions. However, this is not universal as Inga plants have been shown to increase nectar production in response to tending ants (Bixenmann et al., 2011). Fourth, D. quinquenotatus leafhoppers and their tend- ing ants often live together in mud shelters made by tending ants on the basal leaves of the gamagrass. Within these shelters, high densities of ants and leafhoppers occur and parasitism is reduced (Moya-Raygoza & Larsen, 2008). These shelters help to increase the quantity of honeydew for tending ants by concentrating the leafhoppers, whereas A. pennatula does not provide shelters for ants in the form of big thorns as is found on other Acacia species. Providing shelter for members of the mutualism is important in establishing obligatory relation- ships (Speight et al., 1999). Fifth, the honeydew of myrmeco- philous hemipterans contains melezitose that provide nitrogen and is a higher quality nectar than nectar from EFNs (Cook & Davidson, 2006). Sixth, excess D. quinquenotatus leafhoppers are sometimes eaten by tending ants (Moya-Raygoza & Nault, 2000), making the leafhoppers a high quality source of pro- tein. Ant colony growth and reproduction requires substantial quantities of protein (Davidson et al., 2003). Moreover, this D. quinquenotatus leafhopper-ant associa- tion is an obligate and highly specialized mutualism as compared with the more general and facultative ant-A. pennatula mutua- lism. Moya-Raygoza (2005) found that the ant B. obscurior visits active EFNs of A. pennatula but does not protect this species of Acacia from herbivores. Lack of protection by ants against herbivores is common among plants with EFNs (Buckley, 1983; Heads, 1986; Oliveira et al., 1999; Ruhren, 2003). In contrast, both Moya-Raygoza and Nault (2000) and Larsen et al. (2001) have shown that tending ants protect both nymph and adult D. quinquenotatus from predators. Thus, this mutualism between D. quinquenotatus and ants is obligatory, as these leafhoppers apparently cannot live without tending ants. Post-fire response Both T. dactyloides and A. pennatula respond quickly to a fire event with new growth, producing young leaves ready to be colonized by herbivorous insects. Previous studies con- ducted in the tropics have found that some species of plants respond to fire with vigorous growth, which can be colonized rapidly by herbivores (Prada et al., 1995; Vieira et al., 1996). We found that ants are adapted to colonize plants quickly af- ter fire, taking advantage of new resources such as honeydew offered by D. quinquenotatus feeding on T. dactyloides and ex- trafloral nectar produced by EFNs of A. pennatula, resulting in the reestablishment of these mutualistic interactions only a few days after fire. We found more ants tending leafhoppers than visiting EFNs at burned sites where both T. dactyloides and A. pennatula were found. Fire does not kill T. dactyloides, but instead stimulates the growth of new stems from T. dactyloides rhizomes. These new stems are the first food resources that appear within the Table 3. Average number (± standard error) of Dalbulus quinquenotatus nymphs, adults, and tending ants (and species of tending ant), Acacia pennatula EFNs, and ants on 10 stems and 10 branches of T. dactyloides and A. pennatula respectively in burned (in June 2012) and unburned sites at locations in Jalisco, Mexico at the end of the wet season in September 2012. Site Ant/Leafhopper interaction on Tripsacum dactyloides Ant/Acacia interaction Dalbulus quinquenotatus Ants Ant species A. pennatula EFNs Ants Ant Species Nymphs Adults Both resources burned 1. Zapopan 0.9 ± 0.7 0.5 ± 0.2 1.1 ± 0.5 A.gracilipes 5.5 ± 0.1 0.9 ± 0.2 A. gracilipes Camponotus sp. 2. Los Chorros 9.5.0 ±2.1 1.0 ± 0.2 10.0 ± 1.3 B. obscurior 5.0 ± 0.2 0 - Both resources unburned 3. San Isidro 15.1± 5.4 1.5± 0.5 14.3± 6.1 B. obscurior 5.6 ± 0.1 0 - 4. San Agustín 8.9± 1.3 1.9± 0.3 1.4± 0.4 B. obscurior 5.4± 0.1 0 - 5. La Mimila 1.2± 0.3 0.9± 0.3 1.4± 0.3 A. gracilipes 4.4 ± 0.2 0 - 6. El Arenal 6.9± 2.8 3.9± 1.3 3.9± 1.4 A. gracilipes 4.9 ± 0.1 2.8 ± 0.6 A. gracilipes Moya-Raygoza and Larsen - Ants, leafhoppers and EFNs after fire142 community and are quickly recolonized by D. quinquenotatus. These leafhoppers may come from contiguous unburned sites. These immigrant leafhoppers start to feed and produce large quantities of honeydew that attract large numbers of ants. The numbers of ants revealed this fast recolonization by leafhop- pers and ants at sites where fire occurred in either May 2007 or June 2012. In contrast, in June 2012 no leafhoppers or ants were found on T. dactyloides leaves at unburned sites because those leaves were dried out. Although EFNs at unburned sites were actively producing extrafloral nectar at that time, few ants were present. No previous studies have compared the ant abundance at leafhoppers and EFNs on fire-disturbed habitats when both resources are available at the same time. Schowalter (2006), reported that ants and sap-sucking insects such as leafhop- pers dominate early-successional tropical forests as they con- tain an abundance of young, succulent leaf tissue that favor sap-sucking hemipterans and tending ants. In North Ameri- can grasslands, populations of some leafhopper species are significantly greater following fire due to immigration from unburned areas into rapidly growing burned areas (Warren et al., 1987). Previously, Moya-Raygoza (1995) found that D. quinquenotatus leafhoppers were found in larger numbers and tended by a greater number of ants in burned than unburned T. dactyloides colonies, because recently burned plants produce new young leaves with higher concentrations of nitrogen. Similar results have been found in the interaction be- tween ants and EFN-bearing plants in other systems after dis- turbance. For example, pruned plants (Conocarpus erectus L.) grew faster and produced higher numbers of extrafloral nectaries and attracted a higher density of ants (Piovia-Scott, 2011). Leaf damage also increases the production of extraflo- ral nectar in different plants (Heil et al., 2001). In another case, higher abundance of ants was found in the shrub B. campes- tris after fire because of a high concentration of extrafloral nectar (Alves-Silva & Del-Claro, 2013). Similarly Alves- Silva (2011) and Koptur et al. (2010) found a more diverse ant fauna guarding plants from herbivory after fire occurred due to the high production of extrafloral nectar. This is not surprising as ants are attracted to high quality sugar resources produced by plants with EFNs (Heil & McKey, 2003). Therefore, the availability of honeydew and extrafloral nectar to ants after fire is important because it can regulate ecological dominance, affecting the ant trophobiont and plant communities. Greater numbers of ants tending leafhoppers may result in better protection of these honeydew producers by ants compared with the ant protection of plants with EFNs that can also occur in these fire-prone sites. Moreover, coloni- zation by ants after fire is important to initiate these mutual- isms with both hemipterans and EFNs. Our results highlight the importance of investigating mutualisms not only in paired species, but also among multiple mutualisms involving ants when a system is disturbed. Acknowledgments We are grateful to Miguel Vasquez Bolaños for the identification of some of the ant taxa. We also appreciate the comments and suggestions of two anonymous reviewers. References Alves-Sila, E. (2011). Post fire resprouting of Banisteriopsis mallifolia (Malpighiacea) and the role of extrafloral nectaries on the associated ant fauna in a Brazilian Savanna. Sociobiology, 58: 327-340. Alves-Silva, E. & Del-Claro, K. (2013). Effect of post-fire resprouting on leaf fluctuating asymmetry, extrafloral nectar quality, and ant-plant-herbivore interactions. Naturwissen- schaftlen, 100: 525-532. doi: 10.1007/s00114-013-1048-z Bixenmann, R.J., Coley, P.D. & Kursar, T.A. (2011). Is ex- trafloral nectar production induced by herbivores or ants in a tropical facultative ant-plant mutualism? Oecología 165: 417-̶ 425. doi: 10.1007/S00442-010-1787-x Blüthgen, N., Verhaagh, M., Goitía, W., Jaffé, K., Morawetz, W. & Barhlott, W. (2000). How plants shape the ant commu- nity in the Amazonian rainforest canopy: the key role of extra- floral nectaries and homopteran honeydew. Oecologia, 125: 229-240. doi: 10.1007/s004420000449 Blüthgen, N., Mezger, D. & Linsenmair, K.E. (2006). Ant- hemiptera trophobiosis in a Bornean rainforest - diversity, specificity and monopolization. Insectes Sociaux, 53: 194 -203.doi: 10.1007/s00040-005-0858-1 Buckley, R.C. (1983). Interactions between ants and mem- bracid bugs decreases growth and seed set of host plant bear- ing extrafloral nectaries. Oecologia, 58: 132-136. Buckley, R.C. (1987). Interactions involving plants, Homoptera, and ants. Annual Review of Ecology and Systemtics, 18: 111- 135.doi:0066/4162/87/1120-0111 Byk, J. & Del-Claro, K. (2011). Ant-plant interaction in the Neotropical savanna: direct benefical effects of extrafloral nectaries on an ant colony fitness. Population Ecology, 53: 327–332. doi: 10.1007/s10144-010-0240-7 Cook, S.C. & Davidson, D.W. (2006). Nutritional and func- tional biology of exudate-feeding ants. Entomologia Experi- mentalis et Applicata, 118: 1-10. Davidson, D.W., Cook, S.C., Snelling, R.R. & Chua, T.H. (2003). Explaining the abundance of ants in lowland tropical rainforest canopies. Science, 300: 969-972. Delabie, J. (2001). Trophobiosis between Formicidae and Hemiptera (Sternorrhyncha and Auchenorrhyncha): an overview. Neotropical Entomology, 30: 501-516. doi: 10.1590/51519- 566x2001000400001. Sociobiology 61(2): 136-144 (June, 2014) 143 Del-Claro, K. & Oliveira, P.S. (1993). Ant-homoptera interac- tions: do alternative sugar sources distract tending ants? Oikos, 68: 202-206. Fagundes, R., Riveiro, S.P. & Del-Claro, K. (2013). Tending- ants increase survivorship and reproductive success of Cal- loconophora pugionata Dietrich (Hemiptera, Membracidae), trophobiont herbivore of Myrcia obovata O. Berg (Myrtales, Myrtaceae). Sociobiology, 60: 11-19. Fiala, B. (1990). Extrafloral nectaries vs ant-Homoptera mu- tualism: a comment on Becerra and Venable. Oikos, 59: 281- 282.doi: 10.2307/3545545 Gibb, H. & Cunningham, S.A. (2009). Does the availability of arboreal honeydew determine the prevalence of ecologically dominant ants in restored habitats? Insectes Sociaux, 56: 405- 412. doi: 10.1007/s00040-009-0038-9 González-Teuber, M. & Heil, M. (2009). Nectar chemistry is tailored for both attraction of mutualists and protection from exploiters. Plant Signaling and Behavior, 4: 809-813. Heads, P.A. (1986). Bracken, ants and extrafloral nectaries. IV. Do wood ants (Formica lugubris) protect the plant against insect herbivores? Journal of Animal Ecology, 55: 795-809. Heady, S.E. & Nault, L.R. (1985). Escape behavior of Dal- bulus and Baldulus leafhoppers (Homoptera: Cicadellidae). Environmental Entomology, 14: 154-158. Heil, M. & McKey, D. (2003). Protective ant-plant interac- tions as model systems in ecological and evolutionary research. Annual Review of Ecology, Evolution and Systematics, 34: 425-453. doi: 10.1146/annurev.ecolsys.34.011802.132410 Heil, M., Fiala, B., Baumann, B. & Linsenmair K.L. (2000). Temporal, spatial and biotic variation in extrafloral nectar secretions by Macaranga tanarius. Functional Ecology, 14: 749-757. Heil, M., Koch, T., Hilpert, A., Fiala, B., Boland, W. & Lin- senmair K.L. (2001). Extrafloral nectar production of the ant- associated plant, Macaranga tanarius, is an induced, indirect, defensive response elicited by jasmonic acid. PNAS, 98: 1083-1088.doi: 10.1073/pnas.031563398 Heil, M., Greiner, S., Meimberg, H., Krüger, R., Noyer, Jean- Louis,, Heubl, G., Linsenmair, K.E. & Boland W. (2004). Evolutionary change from induced to constitutive expression of an indirect plant resistance. Nature, 430: 205 ̶-208. Hölldobler, B. & Wilson, E.O. (1990). The Ants. Cambridge: Harvard University Press, 732 p. Katayama, N. & Suzuki, N. (2003). Changes in the use of ex- trafloral nectaries of Vicia faba (Leguminosae) and honeydew of aphids by ants with increasing aphid density. Annals of the Entomological Society of America, 96: 579-584. doi: 0013- 8746/03/0579-0584 Katayama, N. & Suzuki, N. (2010). Extrafloral nectaries indi- rectly protect small aphid colonies via ant-mediated interac- tions. Applied Entomology and Zoology, 45: 505-511. Katayama, N., Hembry, D.H., Hojo, M.K. & Suzuki, N. (2013). Why do ants shift their foraging from extrafloral nec- tar to aphid honeydew? Ecological Research, 28: 919-926. doi: 10.1007/s11284-013-1074-5 Koptur, S., William, P. & Olive, Z. (2010). Ants and plants with extrafloral nectaries in fire successional habitats on Andros (Ba- hamas). Florida Entomologist, 93: 89-99. Larsen, K.J., Vega, F.E., Moya-Raygoza, G. & Nault, L.R. (1991). Ants (Hymenoptera: Formicidae) associated with the leafhop- per Dalbulus quinquenotatus (Homoptera: Cicadellidae) on gamagrasses in Mexico. Annals of the Entomological Society of America, 84: 498-501. doi: 0013- 8746/91/0498-0501 Larsen, K.J., Heady, S.E. & Nault, L.R. (1992). Influence of ants (Hymenoptera: Formicidae) on honeydew excretion and escape behaviors in a myrmecophile, Dalbulus quinquenotatus (Homoptera: Cicadellidae), and its congeners. Journal of Insect Behavior, 5: 109-122. doi: 0892-7553/92/0100-0109 Larsen, K.J., Staehle, L.M. & Dotseth, E.J. (2001). Tending ants (Hymenoptera: Formicidae) regulate Dalbulus quinquenotatus (Homoptera: Cicadellidae) population dynamics. Environmental Entomology, 30: 757–762. doi: 0046-225x/01/0757-0762 Marazzi, B., Bronstain, J.L. & Koptur, S. (2013). The diversity, ecology and evolution of extrafloral nectaries: current perspec- tives and future challenges. Annals of Botany, 111: 1243-1250. doi: 001.10.1093/aob/mct109 McVaugh, R. (1987). Flora Nova-Galiciana (Leguminosae), Vol 5. University of Michigan Press, Ann Arbor, 786 p. Mosino-Aleman, P.A. & Garcia, E. (1974). The climate of Mexico. In R.A. Bryson & F.K. Hare (Eds). Climate of North America, vol. 11. World Survey of Climatology (pp 345-404). Elsevier Scientific, New York. Moya-Raygoza, G. (1995). Fire effects on insects associated with the gamagrass Tripsacum dactyloides in Mexico. Annals of the Entomological Society of America, 88: 434-440. doi: 0013-8746/95/0434-0440 Moya-Raygoza, G. (2005). Relationships between the ant Brachymyrmex obscurior (Hymenoptera, Formicidae) and Acacia pennatula (Fabaceae). Insectes Sociaux, 52: 105-107. doi:10.1007/s00040-004-0777-6 Moya-Raygoza, G. & Nault, L.R. (2000). Obligatory mutua- lism between Dalbulus quinquenotatus (Homoptera: Cicadel- lidae) and attendant ants. Annals of the Entomological Soci- ety of America, 93: 929–940. doi: 0013-8746/00/0929-0940 Moya-Raygoza, G. & Larsen, K.J. (2008). Positive effects of shade and shelter construction by ants on a leafhopper-ant mutualism. Environmental Entomology, 37: 1471-1476. doi: 0046- 225x/08/1471-1476 Moya-Raygoza and Larsen - Ants, leafhoppers and EFNs after fire144 Oliveira, P.S., Rico-Gray, V., Diaz-Castelazo, C. & Castillo- Guevara, C. (1999). Interactions between ants, extrafloral nectaries and insect herbivores in Neotropical coastal sand dunes: herbivore deterrence by visiting ants increases fruit set in Opuntia stricta (Cactaceae). Functional Ecology, 13: 623–631.doi:10.1046/j.1365-2435.1999.00360.x Piovia-Scott, J. (2011). The effect of disturbance on ant-plant mutualism. Oecologia, 166: 411-420. doi:10.1007/s00442- 010-1851-6 Prada, M., Marini-Filho, O.J. & Price, P.W. (1995). Insects in flower heads of Aspilia foliacea (Asteraceae) after a fire in a Central Brazilian Savanna: Evidence for the plant vigor hypothesis. Biotropica, 27: 513-518. Rashbrook, V.K., Compton, S.G. & Lawton, J.H. (1992). Ant- herbivore interactions: reasons for the absence of benefits to a fern with foliar nectaries. Ecology, 73: 2167-2174. Rodríguez-Trejo, D.A. & Myers, R.L. (2010). Using oak cha- racteristics to guide fire regime restoration in Mexican pine- oak forests. Ecological Research, 28: 304 -̶323. doi: 10.3368/ er.28.3.304 Ruhren, S. (2003). Seed predators are undeterred by nectar- feeding ants on Chamaescrista nictitans (Caesalpineaceae). Plant Ecology, 166: 189-198. Schowalter, T.D. (2006). Insect Ecology: An Ecosystem Ap- proach. Academic Press, Elseiver, New York, 350 p. Speight, M.R., Hunter, M.D. & Watt, A.D. (1999). Ecology of Insects Concepts and Applications. Blackwell Science, Ox- ford, UK, 572 p. Vieira, E.M., Andrade, I. & Price, P.W. (1996). Fire effects on a Palicaurea rigida (Rubiaceae) gall midge: A test of the plant vigor hypothesis. Biotropica, 28: 210-217. Warren, S.D., Scifres, C.J. & Teel, P.D. (1987). Response of grassland arthropods to burning: a review. Agriculture, Eco- systems and Environment, 19: 105-130. doi:10.10167-8809- (87)90012-0 Way, M.J. (1963). Mutualism between ants and honeydew- producing Homoptera. Annual Review of Entomology, 8: 307- 344. Weber, M.G. & Keeler, K.H. (2013). The phylogenetic dis- tribution of extrafloral nectaries in plants. Annals of Botany, 111: 1251-1261. doi: 10.1093/aob/mcs225 Wilkes, H.G. (1972). Maize and its wild relatives. Science, 117: 1071-1077. Wimp, G. M. & Whitham, T.G. (2001). Biodiversity con- sequences of predation and host plant hybridization on an aphid-ant mutualism. Ecology, 82: 440-452. Zhang, S., Zhang, Y. & Ma, K. (2012). Distribution of ant- aphid mutualism in canopy enhances the abundance of beetles on the forest floor. PLoS ONE, 7 (4): e35468. doi: 10.1371/ journal.pone.0035468. Zhang, S., Zhang, Y. & Ma, K. (2013). The ecological effects of ant-aphid mutualism on plants at a large spatial scale. Sociobio- logy, 60: 236-241. doi: 10.13102/sociobiology.v60i3.236-241.