Intimacy across species boundaries: Interspecific allopreening between Spot‐necked (Stachyris strialata) and Nonggang Babblers (S. nonggangensis)

Abstract Allopreening occurs in many species of birds and is known for providing hygienic and social benefits. While this behavior has been extensively studied among conspecifics, its occurrence among different species remains little known. Outside of captive environments, only a few records of interspecific allopreening exist. In this study, we describe our observations of the interspecific allopreening behavior between Spot‐necked (Stachyris strialata) and Nonggang Babblers ( S. nonggangensis) in a non‐captive environment in southern China. To our knowledge, these observations represent the first record of interspecific allopreening in the family Timaliidae. We suggest that this understudied behavior is most likely related to the dominant–subordinate relationship between these two species: either the dominant species preening the subordinate species to assert dominance or the subordinate species preening the dominant species to reduce tensions by appeasement. We also suggest interspecific allopreening may not be as rare as we thought if we study this behavior under circumstances where different species are close to each other. This study contributes to our understanding of not only the potential mechanism(s) behind interspecific allopreening but also the behavioral ecology of the vulnerable Nonggang Babbler.

environment in southern China. To our knowledge, these observations represent the first record of interspecific allopreening in the family Timaliidae. We suggest that this understudied behavior is most likely related to the dominant-subordinate relationship between these two species: either the dominant species preening the subordinate species to assert dominance or the subordinate species preening the dominant species to reduce tensions by appeasement. We also suggest interspecific allopreening may not be as rare as we thought if we study this behavior under circumstances where different species are close to each other. This study contributes to our understanding of not only the potential mechanism(s) behind interspecific allopreening but also the behavioral ecology of the vulnerable Nonggang Babbler.
Despite these records, we still lack an understanding of the mechanism(s) underlying interspecific allopreening and the associated allopreening invitation display. Here, we report the first documented record of interspecific allopreening between two gregarious babblers in the family Timaliidae in a non-captive environment.
While SNB occurs from southern China to southern Sumatra (Collar & Robson, 2020), NB is only found in the limestone forest in southern China and northern Vietnam (Jiang et al., 2020). In regions where these two species co-occur, they often form mixed-species aggregations in bird blinds (W. Zhou, Z. Zhang, pers. obs.), which are birdfeeding stations built by local people as a form of bird photography tourism.

106°57′24.6″E) near Nonggang National Nature Reserve in Guangxi,
China (see Table 1 for a summary). Following Gill (2012), we defined an allopreening bout as continuous preening motions involving two birds, with motions separated by 10s considered different bouts.
During all three allopreening bouts, allopreening only occurred from SNB to NB, not vice versa. Because no birds were marked, we could not determine if the participants in each allopreening bout were the same individuals. We only observed the third allopreening bout from the beginning to the end. We did not observe how the first and second bouts began. Apart from these three interspecific allopreening bouts, we also observed frequent intraspecific allopreening among NBs.
Our first observation lasted for 15 s at 10:10 a.m. on February 3, 2022. One SNB and one NB perched side by side on a vine branch 0.8 m above ground (Figure 1). During this observation, the SNB preened the nape region of the NB several times. The NB did not react to the preening. Instead, it only turned its head slightly away from the SNB as it was being preened. When the bird blind owner began tossing mealworms to feed the birds, the allopreening stopped and both birds flew off to eat the mealworms. Throughout the SNB, with its throat and breast regions exposed and feathers of those regions erected (Figure 2). This posture was similar to the head-up allopreening invitation display described in Scaly-breasted Munias (Lonchura punctulata) (Moynihan & Hall, 1954), Brownheaded Cowbirds (Molothrus ater) (Selander & La Rue Jr, 1961), and Jungle Babblers (Argya striata) (Gaston, 1977). We also observed this display in NBs when they allopreened intraspecifically. Upon no-

| DISCUSS ION
During our observations, the allopreening from SNB to NB included many self-accessible body parts such as the flank and breast regions. While allopreening on self-inaccessible regions such as the head and nape tends to serve a hygienic function, allopreening on self-accessible regions tends to serve a social function (Radford & Du Plessis, 2006). We believe the observed interspecific allopreening instances between SNB and NB is most likely related to the dominant-subordinate relationship between the two species. Such a relationship can lead to allopreening between conspecifics of different social dominance rankings (Miyazawa et al., 2020;Radford & Du Plessis, 2006). Therefore, it may also lead to interspecific allopreening if different interacting species have different social dominance rankings (Ng & Jasperson, 1984). This possible link between interspecific allopreening and dominant-subordinate relationship has been studied in Icterids (Post & Wiley, 1992;Rothstein, 1980;Scott & Grumstrup-Scott, 1983). However, the mechanism behind interspecific allopreening in these species is likely different because they actively seek preening from other species through chasing followed by allopreening invitation displays (Selander & La Rue Jr, 1961). We did not observe such behavior from NB toward SNB.
In social animals, intraspecific allopreening can occur both from subordinate to dominant individuals and vice versa. On the one hand, subordinate individuals are known to preen dominant ones to reduce social tensions (Kutsukake & Clutton-Brock, 2006;Radford & Du Plessis, 2006). The allopreening behavior in this case serves a conciliatory function. If SNB were subordinate to NB, the purpose of its allopreening could be reducing potential conflict by pleasing the dominant NB. On the other hand, dominant individuals are known to preen subordinate ones to assert dominance (Harrison, 1965;Miyazawa et al., 2020). The allopreening behavior in this case perhaps originates from an agonistic intention of the dominant preener (individual offering F I G U R E 1 A Spot-backed Babbler preening the nape region of a Nonggang Babbler.
F I G U R E 2 Head-up allopreening invitation display from a Nonggang Babbler to a Spot-necked Babbler.
preening) to attack the subordinate preenee (individual receiving preening). However, instead of fleeing from the preener, the preenee performs the allopreening invitation display by exposing its vulnerable area to the preener, perhaps to express submission. The preener then replaces its original intention to attack with allopreening as an outlet behavior (Harrison, 1965). If SNB were the dominant species, the purpose of allopreening could be asserting dominance over NB.
Whether the observed allopreening behavior was the subordinate species pleasing the dominant species or the dominant species asserting dominance over the subordinate species, it might have occurred because of the need to establish a flock hierarchy (Scott & Grumstrup-Scott, 1983). With both species seeking access to the concentrated food resources at the bird blind, an established dominant-subordinate hierarchy would clarify the pecking order in this mixed-species aggregation. While we could not determine the exact dominance rankings of these two species, we observed them chasing each other off the feeding platform. Measuring aggressive behaviors such as chasing/supplanting will allow future research to determine the dominant-subordinate relationship between these two species, therefore improving our understanding of the interspecific allopreening behavior between them.
Besides the dominant-subordinate relationship, we also considered the possibility of a cleaning association, in which the preener eats the ectoparasites from the preenee through allopreening, and hybridization, in which two allopreening participants form a mated pair. However, we found insufficient support for both hypotheses: cleaning associations have rarely been recorded between two bird species (Sazima et al., 2012) and hybridization has never been recorded between SNB and NB (McCarthy, 2006). However, because our study does not eliminate the possibility of these hypotheses, future research on interspecific allopreening should consider them as possible alternative hypotheses that need to be tested.
With an increasing number of instances being discovered, interspecific allopreening may not be as rare as we thought. Perhaps, we only perceive it being rare because we have not studied this behavior in circumstances where it most frequently occurs. Future research should study this behavior in locations where birds are near each other, as allopreening is more likely to happen under such circumstances (Cullen & Ashmole, 1963;Harrison, 1965;Morales Picard et al., 2020). These locations include resting and bathing areas, captive environments, roosting sites, and feeding sites such as bird blinds and feeders, carcasses, and clay licks. Also, future research should focus on social species, particularly species that display clumping behavior and ones that form mixed-species associations and aggregations with others. Lastly, it is essential to be attentive because interspecific allopreening may be easily overlooked without being taken into consideration.

ACK N OWLED G M ENTS
We are grateful to people from Nonggang Aviary for access to their bird blinds. Ei-Ichi Izawa and Loma Pendergraft helped us interpret various behaviors in the video footage. Nancy Coomer, Shruti Parikh, and Scott Robison provided insightful feedback on the first and final drafts of the manuscript. Dan Liang improved the Mandarin abstract. Finally, we thank two anonymous reviewers for their invaluable comments on the manuscript.

CO N FLI C T O F I NTE R E S T
The authors declare no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
This study does not include any data (video footage of the allopreening instances is publicly available at https://doi.org/10.5061/dryad. mw6m9 0609).