Notas Científicas Pachycondyla obscuricornis as natural enemy of the spittlebug Deois flavopicta

The objective of this work was to evaluate the potential control of the ant Pachycondyla obscuricornis Emery (Hymenoptera: Formicidae) on populations of nymphs of the spittlebug, Deois flavopicta Stal (Hemiptera: Cercopidae). Foraging behavior and prey consumption rate of P. obscuricornis were evaluated. Field data revealed that P. obscuricornis does not show aggressive behavior against individuals of the same species, when they are not carrying a prey; they can patrol distances larger than 10 m searching for prey, and they can build their nest as close as 1 m from each other. The ant has a solitary patrolling habit, there is no recruitment behavior, and individuals dislocate fast, browsing on soil and vegetation for prey. Predation rate on spittlebug nymphs increased relative to the spittlebug abundance, reaching 93.8% of captured prey. Pachycondyla obscuricornis is a voracious predator and may control the population of spittlebugs in cultivated pastures.

Spittlebugs (Hemiptera: Cercopidae) are the main pest in cultivated pastures of Central Brazil due to their frequent outbreaks and wide distribution (Valério & Koller, 1992).Production of spittle protects the nymphs against natural enemies and desiccation and, besides the high capacity of colonizing introduced grasses, represents another trait that increases spittlebug's survivor and fitness, facilitating its population growth and damage to pastures (Valério & Koller, 1992;Pires et al., 2000).
Surveys in pastures have shown that ants are the most abundant predators and a potentially important biological control agent of spittlebug populations.Individuals of Pachycondyla, Camponotus, Ectatoma, Pheidole, Labidus and Mycocepurus were observed in pastures carrying spittlebug nymphs, and Pachycondyla obscuricornis outstands as the most frequent (Sujii et al., 2002).
The objective of this work was to determine if the ant Pachycondyla obscuricornis has potential to control spittlebug nymph populations.
Methods used focused on P. obscuricornis foraging behavior and prey consumption rate.The foraging behavior of P. obscuricornis was observed by following individuals leaving the nests, attacking prey and carrying them back to the nests, at different periods of the day.At least 40 individuals were observed, 20 in the morning and 20 in the afternoon, to confirm the behavior patterns.To further explore the potential of this species for the spittlebug's population control, this study focused on its patrolling behavior, how it locates and captures the spittlebug nymphs, and its response when encountering other ants.
To evaluate the distribution of P. obscuricornis in a pasture of Brachiaria ruziziensis, a 30x30 m grid was set, with sardine baits placed in Petri dishes 5x5 m apart.Ants foraging at the baits were followed to locate their nests.Some individuals were collected to confirm identification.
A study to evaluate the importance of D. flavopicta in the diet of P. obscuricornis was carried out by observing ants arriving at two nests during a period of intense activity (9-12 AM) on January 16 and 30.Each prey carried to the nest was taken from ants and identified.The densities and instars distribution of D. flavopicta in the field for the same day was estimated by sampling 10 randomly selected quadrats of 0.25 m 2 in the pasture.To determine occurrence of preference for prey instars, all nymphs were counted, had their developmental stage determined, and had their frequency distribution compared to the frequency distribution of nymphs observed being carried by the ants to the nestsé ntrances.The frequency distribution of spittlebug nymphs present in the area and captured by the ants was compared by a Kolmogorov-Smirnov test for the two samples (Wilkinson, 1990).
During the study, D. flavopicta was the main item found in P. obscuricornis diet.The ant was particularly successful in finding nymphs and attacking them inside the spittle.This ant, 12 mm long and displaying a black coloring, has a patrolling habit; individuals hunt and quickly catch all individuals observed.This species shows a solitary foraging behavior; there is no recruitment of other ants for catching or carrying the prey.The Pachycondyla genus has several species specialized in preying on termites (Holldobler & Wilson, 1990).Although there is no reference of P. obscuricornis predating termites, these insects could be the preferred prey attacked in the absence of spittlebugs.Longino (2003) reports P. obscuricornis hunting and carrying caterpillars, in Costa Rica.
P. obscuricornis does not show territorial behavior, but may aggressively dispute prey.Antagonistic or fighting behavior in defense of territories was not observed between individuals from different nests encountering each other (12 observations).The proximity of nests, which can be separated by distances shorter than 1 m (Figure 1), reinforces the absence of territorial aggressive behavior in P. obscuricornis, a trait that can be useful for a biological control agent.Aggressive behavior may occur however, if one of the ants is carrying a prey.In this case, the encounter usually resulted in fight for the prey (three observations).Therefore, even though individuals seem not to frequently compete by interference with each other, competition for food may occur in this system, and is more related to food exploitation, with individuals trying to attack and carry more of the same prey item to their nests.
Observation at the entrance of P. obscuricornis nests showed that the amount of D. flavopicta in ant's diet rose from 63.6% to 93.7% when prey density increased from 40 nynphs/m 2 , at the beginning of the infestation, to 55 ninphs/m 2 (2 nd -5 th ) (Figure 2).The number of nymphs removed by the ants varied from 14 to 61 nymphs, in a three-hour-period (9-12 AM).The frequency of capture was positively correlated with the relative abundance of D. flavopicta and the distribution of prey instars captured was positively correlated with the distribution of instars available in the field.Kolmogorov-Smirnov test on January 16 (MD = 0.75, p=0.13) and January 30 (MD = 0.50, p =0.5) showed that there is no preference for a specific stage or size of spittlebug nymphs, although relative contributions to the diet changed as spittlebug nymphs´ size and abundance increased.
Even though P. obscuricornis patrols distances greater than 10 m from the nest (Figure 1), field observations of presence of nymphs next to the nests (<1 m) suggest that the stimulus for attack is generated by visual perception of nymph movement inside the spittle, and not by the encounter with the spittle itself.This may allow the ant to save time by not searching empty spittle.
From the observations in this study, it can be inferred that, although D. flavopicta is the main component of the P. obscuricornis diet during some periods of the year, and although ants actively search for nymphs in the field, the presence of remaining nymphs of spittlebugs close to the nests indicates the need of further studies to evaluate the control level imposed by P. obscuricornis ants on the spittlebug population.Received on November 11, 2003 andaccepted on March 19, 2004

Figure 1 .
Figure 1.Distribution of sardine baits ( ) grid (5x5 m) in the field, baits visited by ants and the relative location of the ant Pachycondyla obscuricornis's nests ( ).