Introduction

Mutual interactions between ants and hemipterans occur commonly in ecosystems (Holway et al. 2002). Ants protect hemipterans against their predators and parasitoids. In return, the ants receive large amounts of honeydew, which is essential to their colony growth and survival because it contains sugars mixed with various amino acids (Yao and Akimoto 2001; Fischer and Shingleton 2001; Helms and Vinson 2002, 2003; Brightwell and Silverman 2010). Previous studies have shown that the mutualism between Solenopsis invicta and Phenacoccus solenopsis facilitates population increase and fitness of each other (Zhou et al. 2012a). The invasive ants S. invicta are able to protect the exotic honeydew-producing mealybug P. solenopsis with their ability to deter natural enemies (Zhou et al. 2013). Honeydew produced by P. solenopsis significantly enhanced the colony growth and worker survival of S. invicta (Zhou et al. 2012b). Similar results indicated that carbohydrates could increase worker survival and colony growth rates of fire ants (Wilder et al. 2011a, b). Adult workers can not digest solid food sources directly because they have a sieve-like structure in their throats that prevents them from swallowing solids. Honeydew from mutualist hemipterans is considered as an important liquid source of energy for adults S. invicta workers (Vinson 1983; Tschinkel 2006).

Besides invasive ants, some co-occurring ant species can also utilize the honeydew excreted by P. solenopsis. Honeydew is a critical resource for both introduced and native ants (Wilder et al. 2013). Colony growth of native ants decreased if they lost mutualist-provided resources even when other sources of energy and nutrients were available (Wilder et al. 2013). The ghost ant, Tapinoma melanocephalum, is a worldwide invader, whose native range is unknown but is believed to come from Africa or Asia (Wheeler 1910). It has been a well established species in southern China. T. melanocephalum is a highly competitive sugar-feeding species. They can also establish a close relationship with hemipterans. We found that the density of mealybugs had a significantly linear correlation with the number of foraging T. melanocephalum workers on a given plant (unpublished data, Zhou et al.). They compete with S. invicta frequently for honeydew resources produced by P. solenopsis on plants. Previous study has demonstrated that tending by both S. invicta and T. melanocephalum has positive impacts on the fitness of individuals of P. solenopsis. However, compared with T. melanocephalum, S. invicta acquired more honeydew and protected P. solenopsis more effectively, which may facilitate the invasion of these two alien species in south China (Zhou et al. 2012a).

Resource competition by S. invicta has been considered as one of the key factors shaping the reduction of native ant diversity (Epperson and Allen 2010). Recent studies indicate that introduced ants compete for honeydew produced by hemipterans with native ants, and exclude native ant species from this critical resource (Wilder et al. 2013). A lot of literature shows that invasive ants control the food resources based on their forceful competitiveness, and inhibit native species (Holway et al. 2002; Zhou et al. 2012a; Wilder et al. 2013). Previous studies also indicate that native ant species could coexist and neighbor each other (Calixto et al. 2007a, b). However, few efforts have focused on the competitive mechanisms between exotic ants S. invicta and its co-occurrence with T. melanocephalum.

Here, we investigated possible behavioral mechanisms underlying the coexistence of S. invicta and T. melanocephalum using short geographical range competition experiments in the laboratory to compare their competitive ability for honeydew resource.

Materials

Plants

Hibiscus rosa-sinensis, a Chinese native species cultivated worldwide, was purchased from a commercial horticultural farm. Each plant was approximately 25–30 cm in height and had 25–30 true leaves. All plants were cultivated in plastic flowerpots (the diameters of the upper and lower edges were 18 and 14 cm, respectively, with a height of 17 cm) in greenhouses.

Insects

Colonies of P. solenopsis were collected from the campus of South China Agricultural University and fed on H. rosa-sinensis. First instar mealybug nymphs were inoculated on each plant and raised for several generations. H. rosa-sinensis plants with established mealybug colonies were used for subsequent experiments. All colonies were reared in the laboratory with the temperature maintained at 27 ± 2 °C and a relative humidity of 60–70 %. Colonies of S. invicta and T. melanocephalum were collected from a suburb of Guangzhou and reared in plastic boxes (116 L 72 cm × 56 cm × 29 cm). Ant colonies were separated from the soil by dripping water into plastic boxes until the colonies floated (Jouvenaz et al. 1977). Each colony was divided into sub-colonies (approximately 1.0 g) using a microbalance (Sartorius, BS, 224S). Each sub-colony included one queen and adult workers, pupae, larvae, and eggs. The ants were placed in a 9-cm plastic Petri dish, which served as an artificial nest. All ant colonies were reared in plastic cases (26 cm × 18 cm × 8 cm) at 28 °C. Solenopsis invicta colonies were given fresh live Tenebrio molitor larvae, frozen crickets, and a 10 % solution of honey mixed with water (50 ml) weekly. T. melanocephalum sub-colonies were maintained with tubes filled with distilled water plus 10 % honey solution.

Experimental procedure

Before the beginning of the experiments, all experimental colonies were starved for 48 h to produce a uniform state of hunger. Each experimental colony consisted of workers (0.8 g), brood (0.2 g), and a single queen. A microbalance (Sartorius, BS, 224S) was used to weigh workers and brood. Five adjacent boxes (26 cm × 18 cm × 8 cm) were arranged in a row and connected by tubes. The cases containing the ant colonies on each end are considered as colony cases; the empty cases adjacent to each colony case are considered as adjacent cases; and the connecting case in the middle is considered as the center case (Morrison 2000). The whole equipment included S. invicta colony case (SCC), S. invicta adjacent case (SAC), center case (CC), T. melanocephalum adjacent case (TAC), and T. melanocephalum colony case (TCC) (Fig. 1). At the beginning of the experiments, 0.4 g honeydew produced by P. solenopsis was placed in CC, 0.2 g in SAC and TAC, respectively. Each plastic case was connected to (1 cm inside diameter) to an adjacent case via a 12-cm long silicone tube (Fig. 1). The number of foraging workers of the two ant species in SAC, CC, and TAC after 0.5 h, 24 h, 48 h, and 72 h was counted, respectively. The mortality of each ant species in SCC, SAC, CC, TAC, and TCC was recorded after 72 h. We also recorded the number of S. invicta and T. melanocephalum colonies, which intruded the opposite colony case after 72 h. All treatments were replicated 10 times.

Fig. 1
figure 1

Design of the short-distance, colony-level interference competition experiments between S. invicta and T. melanocephalum (Morrison 2000)

Full size image

Statistical analysis

We used three-way ANOVA to test the number of foraging workers versus the effects of the box (SCC, SAC, CC, TAC, and TCC), the time interval and the species. One-way ANOVA using type III sum of squares was performed to compare the means among different resource cases. If ANOVA results were significant, multiple comparisons of means were performed using Tukey HSD post hoc analysis. Paired t test was used to analyze the foraging activity of the two ant species in same resource case. We used chi-square test to compare the invasiveness of S. invicta and T. melanocephalum. All statistical analyses were conducted using SPSS version 14.0 (SPSS Inc., Chicago, IL, USA).

Results

Foraging activity of S. invicta and T. melanocephalum

Foraging intensity showed significant difference between the two ant species in SAC (F = 63.076, P < 0.001). Number of foraging S. invicta was significantly more than that of T. melanocephalum in SAC (Table 1, Ants; Fig. 2a). In CC, number of foraging workers was not significantly different between the two ant species (Table 1, Ants; Fig. 2b). Foraging intensity was significantly different between the two ant species in TAC (Table 1, Ants; Fig. 2c). However, this difference declined with our investigative time sequence (Table 1, Ants*Time: P = 0.043).

Table 1 Analysis of foraging intensity of the two ant species based on two-way analysis of variance
Full size table
Fig. 2
figure 2

Foraging intensity of S. invicta and T. melanocephalum (a in S. invicta adjacent case; b in central case; c in T. melanocephalum adjacent case)

Full size image

Mortality of S. invicta and T. melanocephalum

Our results show that in both ant species mortality differed between cases (S. invicta: F 4,45 = 18.271, P < 0.0001; T. melanocephalum: F 4,45 = 23.134, P < 0.0001; Table 2). Mortality of S. invicta between SCC, SAC, and CC was no significantly different after 72 h (P = 0.960). It was obviously different between TAC and TCC (P = 0.009). Mortality of T. melanocephalum between SCC, SAC, and CC was also not significantly different after 72 h (P = 0.257). Mortality of T. melanocephalum in TCC was markedly greater than that in other case.

Table 2 Mortality of S. invicta and T. melanocephalum in the cases after 72 h
Full size table

Mortality of S. invicta was observably more than T. melanocephalum in SCC (t = 3.334, df = 12.200, P = 0.006; Table 2). There was no significant difference in mortality between the two ants species in SAC, CC, TAC, and TCC, respectively (t = 0.798, df = 18, P = 0.435; t = 0.796, df = 18, P = 0.436; t = 1.259, df = 18, P = 0.224; t = 1.290, df = 18, P = 0.213; Table 2).

Invasiveness of S. invicta and T. melanocephalum

Our results indicated that the number of colonies in home-side case between S. invicta and T. melanocephalum was significantly different after 72 h (χ 2 = 5.333,df = 1, P = 0.021). There were 8 colonies of S. invicta which had intruded into TCC. However, no colony of T. melanocephalum had intruded into SCC (Table 3).

Table 3 Comparison of invasiveness of two ant species after 72 h
Full size table

Discussion

Interaction between ants and hemipterans facilitates colony growth and fitness of each other (Kaplan and Eubanks 2002; Flatt and Weisser 2000; Kaplan and Eubanks 2005; Daane et al. 2007). Honeydew produced by hemipterans is a critical resource for multiple species of ants. They can not be substituted by nutrients in insect prey (Wilder et al. 2013). Many studies show that honeydew is important as a fuel for activity and aggression of workers. Colony growth and reproduction of brood are facilitated by the mutualist-provided carbohydrates (Kay et al. 2010; Wilder et al. 2011a, b; Grover et al. 2007). Exotic ant species such as S. invicta compete with native species extensively for mutualist-provided carbohydrates and suppress the exploitation of mutualist carbohydrates by native ants. To honeydew-dependent native ant species, the effects of honeydew monopoly by S. invicta could negatively affect colony growth (Wilder et al. 2013).

Strength in numbers of foraging ants is key factor in interspecific competition for food resource (Holway et al. 2002). Our study determined the short-distance competitiveness of S. invicta and T. melanocephalum at colony level. The results demonstrate that the competitive ability of S. invicta is stronger than that of T. melanocephalum. Forager numbers of S. invicta was continually increased in determined resource case. On the contrary, foraging intensity of T. melanocephalum was restrained by interference from S. invicta (Fig. 1; Table 1). T. melanocephalum found the honeydew resource much earlier and arrived at honeydew more quickly than S. invicta (Zhou et al. 2012a). However, we found that most of honeydew was occupied by S. invicta after 72 h. The advantage of greater foraging intensity of T. melanocephalum gradually disappeared (Fig. 1; Table 1). Due to forceful aggressiveness by S. invicta, most of food resource was dominated by this ant. Previous study also indicated that domination of food resources by S. invicta depended on their forceful competitiveness and inhibition of native species. S. invicta occupied more of the foraging arenas against both native ants when colonies were equivalent by worker biomass (Morrison 2000).

Mortalities of S. invicta and T. melanocephalum in TCC were greater than that in other cases. Food competition was more intense in TCC than in other cases. Because of higher aggressiveness, S. invicta often intruded into colony of T. melanocephalum. However, T. melanocephalum barely intruded into colonies of S. invicta.

Although most of honeydew resource was forcibly occupied by S. invicta, our results indicate that there were still small numbers of T. melanocephalum workers in resource cases. S. invicta could not completely drive T. melanocephalum out of honeydew competition. T. melanocephalum could coexist and share the honeydew with S. invicta. Activity of native ants can play a certain restraining role to invasive ants (Vogt et al. 2005). Previous studies also show that some native ant species such as Dorymyrmex spp. and Forelius spp. are usually located adjacent to S. invicta colonies, and they are rarely attacked by S. invicta. Native ant species could regulate their abundance to suit the density of fire ants (Calixto et al. 2007a, b). T. melanocephalum can scare off attackers through their chemical defense. We found that secretions from pygidial gland could be sprayed over the S. invicta workers by T. melanocephalum when they encounter each other. Workers of S. invicta usually act slowly after they are sprayed by the secretions of T. melanocephalum. Similar results also indicate that pygidial gland secretions of T. melanocephalum play an important role in fighting against each other between S. invicta and T. melanocephalum (Tomalski et al. 1987).

Compared with co-occurring ant species, such as T. melanocephalum and Pheidole fervida, S. invicta not only have significant advantage of individual body size, but also can attack native species using their well developed mandibles and venom (Gao et al. 2011). Our results suggest that although S. invicta are more aggressive than T. melanocephalum, mutual tussle was not intense at individual level between the two ant species. Chemical defense of T. melanocephalum may restrain the aggressiveness of S. invicta. The aggressiveness of S. invicta is stronger than that of the two ants T. melanocephalum and P. fervida. Aggressiveness between S. invicta and P. fervida was stronger than that between S. invicta and T. melanocephalum (Gao et al. 2011). This result also indicates that chemical defensive ability was important for T. melanocephalum against fire ants.

Conclusion

This study determined the competitive ability for food resource of the fire ant and the ghost ant at short range in colony level. We found that S. invicta suppressed the exploitation of honeydew-producing mealybug by ghost ants, and occupied most of honeydew resource due to forceful aggressiveness. However, ghost ants have not been completely driven out of honeydew competition by fire ants, because chemical defense of ghost ants may restrain the aggressiveness of fire ants. The exact competitive mechanisms between invasive and native ants in a recently invaded area may vary among species, because ants are a very diverse group. Effects of invasive ants on subdominant and subordinate ant species in the recipient biota may be mediated through a diversity of competitive mechanisms (Morrison 2000). Therefore, the study which elucidate the underlying mechanisms of species replacement and interspecific competition are important to the understanding of why invasions fail or succeed.