Wing-slapping: A defensive behavior by honey bees against ants

Eusocial organisms are group living, and their nests, which are inhabited by numerous eggs, larvae, and adults, are attractive feeding grounds for predators and parasites. To effectively defend the nest against such natural enemies, eusocial organisms have evolved specific defensive behaviors. For example, honey bees (Apis spp.) counter attacks from hornets, an important predator, by the use of shimmering waves (Tan et al., 2012) and defensive balling (Ono et al., 1995); they also use grooming and other hygienic behaviors to suppress the spread of parasitic varroa mites (Harbo & Harris, 2005). Ants are also important consumers of honey, broods, and worker bees in beehives (Kern, 2017), and it has been reported that some honey bee species inhibit the entry of ants into the hive through fan-blowing, a defensive behavior in which bees use wind pressure from fanning their wings to keep ants away from the hive entrance (Spangler & Taer, 1970). Because this behavior inhibits ant ingress, while avoiding direct contact between ants and the bees, it is thought to reduce the risk of injury or death of the defending bees (Seeley et al., 1982). However, the Japanese honey bee (Apis cerana japonica), the Japanese subspecies of the eastern honey bee (A. cerana), repels ants by placing its wings directly in contact with intruding ants and flicking them away. Although this behavior has been briefly described in anecdotal accounts (Fujiwara et al., 2015; Yoshida, 1997), the detailed behavior sequence has remained unclear.

We filmed Japanese honey bee colonies being invaded by the Japanese pavement ant (Tetramorium tsushimae) with a high-speed camera (960 fps; DSC-RX10M4, Sony Group Corporation, Japan), and recorded honey bee defensive behaviors against the ants in detail. Our results show that the bees slap the ants through direct contact with their wings (Figure 1; Videos S1 and S2). Specifically, the worker bees executed the slap behavior by first tilting their bodies toward the ants, then flapping their wings while simultaneously turning their bodies. These observations demonstrate that the defensive behavior employed by Japanese honey bees against ants is clearly different from the noncontact fan-blowing behavior (Yang et al., 2010) reported in other honey bees. We named this new defensive behavior “wing-slapping.”

In addition to the above observations, to identify whether wing-slapping behavior is only carried out against specific ant species and whether there are differences in effectiveness among ant species, we introduced ants to the vicinity of the hive entrance and determined the frequency of wing-slapping by Japanese honey bees. The observations were carried out in June and July 2023 on two colonies (A and B) of Japanese honey bees in Tsukuba City, Ibaraki Prefecture (36°02′54.1″ N, 140°06′54.8″ E). The Japanese honey bee colonies were raised in movable-frame hives with a platform of dimensions 50 × 260 mm at the hive entrance. Worker bees that remained on the entrance platform for more than 5 s were regarded as guard bees, and were marked with paint for identification. For the ant intruders, we selected three species from among the common Japanese ant species that are known to invade Japanese honey bee hives: the Japanese queenless ant Pristomyrmex punctatus, the Japanese pavement ant, and the Japanese wood ant Formica japonica (Miyano et al., 2001; Okada, 1997). These species were collected at the study site. Specimens were placed in a plastic cup with free access to the hive entrance. Then, the platform was filmed from directly above with a digital camera for 30 min. From the video, the behavior of marked guard bees was observed and recorded during periods when the distance between an ant and the center of the bee's head was 10 mm or less. In addition, the success rate of the wing-slaps was calculated. A wing-slap was deemed to be successful if it resulted in the targeted ant being flicked away.

Wing-slapping was the most commonly observed behavior performed by guard bees toward nearby ants (Figure 2a). However, western honey bees show distinct responses to different beetle species (Atkinson & Ellis, 2011), and more observations of Japanese honey bees responding to a variety of ant species are needed to understand whether their responses might also be species specific. In our study, whereas wing-slapping had a success rate of roughly one in two or three attempts in repelling intruding Japanese queenless ants and Japanese pavement ants, the success rate against Japanese wood ants was relatively low (Figure 2b), suggesting the behavior might not be effective against all ant species. Given that Japanese wood ants have about twice the body length and move more quickly than the other species examined in the present study, it may have been difficult to successfully flick them away.

Wing-slapping behavior might have evolved among Japanese honey bees as a relatively low-energy, generic defense against the threat of ant incursion. Fan-blowing behavior is thought to function as a way for bees to avoid direct contact with predator ants (e.g., Asian weaver ants Oecophylla smaragdina; Seeley et al., 1982). However, although Japanese ants, especially Japanese queenless ants and Japanese pavement ants, will invade Japanese honey bee hives, they rarely bite and consume healthy workers in the hive (Okada, 1997; Yoshida, 1997; see Miyano et al., 2001 for an exception). Therefore, in Japan, where predator ants such as the Asian weaver ant are absent and the threat of direct predation by ants is low, noncontact defensive behaviors might not always be optimal. In other words, ants with low direct predation threat can be eliminated more efficiently (i.e., with lower energy cost) through direct contact rather than through generating wind pressure. To test this hypothesis, it will be necessary to investigate the aggression (predation ability) of ants and compare it against the defensive behaviors of bees in various areas. The behaviors employed to defend against ant incursion are yet to be identified for most honey bee species and subspecies. Such information, if collected in the future, would provide an interesting opportunity to examine the evolution of defensive behaviors among social insects through inter- and intraspecific comparison.

Anthropogenic range expansions can place honey bees in contact with predator species they have not evolved alongside, increasing the risk of predation. This risk has become a reality for some species, including the northern giant hornet (Vespa mandarinia) (Matsuura & Sakagami, 1973). The same logic would suggest that populations of the Japanese honey bee, which did not evolve alongside predatory ants, would suffer serious harm if exposed to nonnative ants that are capable of preying on bees. For example, big-headed ants prey upon healthy workers of western honey bees (Kern, 2017; Yang et al., 2010). Considering that big-headed ants are currently expanding their distribution in southern Japan (Yamane et al., 2019), Japanese honey bees could come into frequent contact with them in the future. If Japanese honey bees respond to these nonnative predators with wing-slapping, which involves direct contact, they may be subjected to a higher risk of mortality than other honey bees that exhibit noncontact defensive responses. Invasive ants are a widespread problem around the world, and there is no doubt that new invasions will occur in the future. Therefore, the resulting harm to honey bees may become more severe over time.

Additionally, the defensive behaviors of honey bees against ants continue to raise numerous research questions, such as how bees detect ants, whether there are variations in the innate aptitude of individual honey bees for a given behavior, and whether the success rate of behavior increases with experience. Although ants are important consumers of bees and bee products (Kern, 2017), defensive behaviors of bees against ants have received little attention, perhaps because these interactions are inconspicuous to human observers. Our study identifies a new defensive behavior, which we have named wing-slapping, and opens the door to further studies examining the evolution of defensive behaviors in social insects.

Yugo Seko and Kiyohito Morii collected and analyzed the data, took photographs and videos, and wrote the first draft. Yugo Seko, Kiyohito Morii, and Yoshiko Sakamoto designed the study and revised the manuscript.

The authors declare no conflicts of interest.