Social insects forage food resources (proteins, carbohydrates), material for nest construction (plant fibers, clay), and water (Beggs 2001). Most social wasps (Vespidae) are generalist predators of agricultural pests; therefore, they are frequently found in agroecosystems (Auad et al. 2010; Brügger et al. 2011, 2017) controlling lepidopteran caterpillar populations (Saraiva et al. 2017).

The prey capture by social wasps consists of approaching the prey (generally by walking), antennation, attack, killing it by slicing pieces of flesh (Dejean et al. 2017), and transport of prey pieces to the nest (Brown et al. 2012).

Polybia fastidiosuscula Saussure (Hymenoptera: Vespidae) is a social wasp that is widely distributed in South America, and which constructs large nests (“fragmocitaro”) that are about 30 cm long (Saraiva et al. 2017). The foraging activity of these worker wasps covers a radius of approximately 81 m from their nest (Souza et al. 2010).

Behavioral aspects of social wasps, such as daily activity, diet, ecology, flight, and transport capacity (Moreyra et al. 2012; Brügger et al. 2017) should be studied to assess their potential for integrated pest management. The objective was to identify the prey captured and determine the foraging activity patterns of P. fastidiosuscula in an urban area.

A colony of P. fastidiosuscula (nest of 23 cm diam) in an experimental field of the Universidade Federal de Viçosa (UFV) in Viçosa, Minas Gerais, Brazil (20.764417°S, 42.868278°W) was evaluated. Polybia fastidiosuscula workers that were returning to the colony were intercepted in Jul 2015. Seven sampling sessions of 13 h each were conducted from 6:00 AM to 7:00 PM h, totaling 91 h of sampling effort. Wasps carrying prey, which were identified by their slow flight, were captured with an entomological net, and killed in a lethal chamber containing cotton moistened with ether. Dead wasps and preys were saved in 2 mL Eppendorf tubes with 70% alcohol. The gross sample weight was determined using a digital analytical scale (± 0.1 mg). The alcohol was dried by placing the samples on filter paper for 10 min at room temperature before weighing. Prey was identified under a stereomicroscope (DIGILAB, Piracicaba, São Paulo, Brazil).

Data on foraging activity and prey capture by P. fastidiosuscula throughout the d were subjected to ANOVA using Statistix, Version 9.0 (Analytical Software, Tallahassee, Florida, USA) with each 1 h sampling representing an experimental unit. Differences with P < 0.05 were significant, and the means compared with the Tukey test at 95% probability level. The Spearman correlation coefficient (r) of wasp weight vs. prey/flesh-piece transported was obtained using the PAST program, Version 3.08 (Analytical Software, Oslo, Norway).

Foraging by P. fastidiosuscula typically varied throughout the d. The peak foraging period of this wasp occurred from 12:00 Noon to 2:00 PM (Fig. 1). A total of 322 departures of P. fastidiosuscula workers were counted during the observation period, with a return rate of 89.4% (288 arrivals). Of the 288 arrivals, 182 P. fastidiosuscula workers brought prey of Lepidoptera (caterpillars) (92.6%), Araneae (3.1%), Diptera (2.5%), or Orthoptera (1.8%) species. The other arrivals were wasps with material for nest building.

Mean (± SD) wasp weight and prey/flesh-pieces transported were 17.4 ± 3.3 mg (n = 288; ranging from 11.6 to 28.4 mg) and 5.2 ± 2.5 mg (n = 182; ranging from 1.1 to 13.1 mg), respectively. Polybia fastidiosuscula workers can carry a prey load up to 30% of their own body weight; however, the weights of wasp vs. prey/flesh-piece transported were not correlated (r = 0.0424; P = 0.5692). The wasps had different behaviors at certain periods of the d. Peak activity occurred during the hottest periods of the d (Hernández et al. 2009; Santos et al. 2009).

Fig. 1.

Typical Polybia fastidiosuscula foraging patterns. Means with the same letter at each evaluation time do not differ at 95% probability by the Tukey's test. Vertical bars represent the standard error of the mean (n = 7).

The high return rate of P. fastidiosuscula to the nest is due to their ability to locate themselves through visual signals during foraging up to a distance of 500 m from their colony (Souza et al. 2010). The Lepidoptera caterpillars were the principal prey of P. fastidiosuscula workers; this is similar to the report by Pereira et al. (2013) of caterpillars being a major protein source for social wasps.

The weight of the wasp and its prey was found to vary considerably for P. fastidiosuscula workers returning with a prey load. The prey weight was about 30% of their own body weight, which is lower than that of Polybia occidentalis (Olivier) (39%) (Hernández et al. 2009), but similar to Polybia rejecta (F.) (30.7%) (Dejean et al. 2017), and higher than that for Polybia platycephala Richards (all Hymenoptera: Eumenidae) (13.4%) (Prezoto et al. 2005). This weight variation of the prey captured and transported by P. fastidiosuscula may be related to the opportunistic and generalist behavior of social wasps, which capture the first prey regardless of size (Raveret-Richter 2000). In addition, wasps divide the large prey into small pieces of different weights (Raveret-Richter 2000; Dejean et al. 2017).

The P. fastidiosuscula preference for foraging caterpillars demonstrates the importance of this wasp to reduce Lepidopteran populations in agroecosystems. Chemical pest control, when necessary, should be conducted before 9:00 AM or after 4:00 PM, when the foraging activity of P. fastidiosuscula is reduced.