Abstract Knowledge of the mechanisms facilitating the coexistence of closely related competing species is crucial for understanding biodiversity patterns. The concept of convergent agonistic character displacement (ACD) suggests that interspecific interference competition may lead to convergence in territorial signals between species, which helps to establish interspecific territoriality and thus facilitate the species coexistence. Despite a strong theoretical background, however, empirical evidence for convergent ACD in nature is still scarce. Here we tested whether mixed singing (i.e. copying of elements from songs of a closely related sympatric species) in the Thrush Nightingale (Luscinia luscinia) in a secondary contact zone with the Common Nightingale (L. megarhynchos) could represent a case of convergent ACD. Using playback experiments, we measured the intensity of physical and vocal territorial responses of Common Nightingale males to 3 stimuli: conspecific song, pure heterospecific song, and mixed heterospecific song of the Thrush Nightingale. We found that Common Nightingale males showed a stronger physical territorial response to conspecific than both pure and mixed heterospecific songs. However, the intensity of vocal territorial response significantly increased with the presence of Common Nightingale elements in the stimuli, being lowest to pure heterospecific songs, intermediate to mixed heterospecific songs, and strongest to conspecific songs. These results indicate that mixed singing in the Thrush Nightingale may indeed be a case of convergent ACD. Our findings highlight the potential importance of mixed singing in facilitating species coexistence in the early stages of divergence. INTRODUCTION Interspecific interactions—including competition over resources, aggressive behavior, and hybridization—very often result in the phenotypic divergence of species in sympatry known as character displacement (Brown and Wilson 1956). Such divergence is assumed to be adaptive in terms of reducing the intensity of ecological competition (ecological character displacement; Pfennig and Pfennig 2009), avoiding misdirected aggression (divergent agonistic character displacement, hereafter ACD; Grant and Grant 2006; Grether et al. 2009), or decreasing the cost of hybridization (reproductive character displacement or reinforcement; Dobzhansky 1937; Butlin 1989). An apparent exception to this pattern concerns agonistic territorial signals involved in interference competition, which, compared to most other traits, relatively often converge between species in sympatry (Cody 1969; Brown 1977; Tobias et al. 2014). It has been suggested that convergence in these signals could be adaptive in terms of facilitating the communication between species and the establishment of interspecific territoriality. This would allow animals to resolve territorial disputes without direct physical conflict and avoid resource competition by forcing competitors from their territories. As a consequence, convergence in agonistic territorial signals can reduce both the costs of aggressive physical conflicts and the intensity of competition over resources between species, allowing species with close ecological requirements to coexist (Brown 1977; Sorjonen 1986; Tobias and Seddon 2009; Gil 2010; Reif et al. 2015). Such convergence in agonistic signals has been called convergent ACD (Grether et al. 2009, 2017). Bird song is an important agonistic territorial signal that helps birds defend their territories without resorting to harmful physical attacks. It has been repeatedly demonstrated that bird song may converge in sympatric species, both in suboscine birds whose song is genetically determined (Tobias and Seddon 2009; Tobias et al. 2014) and in oscine birds whose song is transmitted culturally (Slater 1989; Catchpole and Slater 2008). Convergence in song may be caused by shifts in various song characteristics (e.g. temporal or syntax parameters; Secondi et al. 2003). However, in oscine birds this convergence is often caused by heterospecific song learning. This typically leads to a phenomenon known as mixed singing when males of one species incorporate song elements of another closely related species in their repertoires (Helb et al. 1985; Haavie et al. 2004; Gorissen et al. 2006; Vokurková et al. 2013). Mixed singing has long been assumed to result from erroneous heterospecific learning that brings no advantage to birds (Helb et al. 1985) or could even be maladaptive as it increases the likelihood of hybridization (Qvarnström et al. 2006). However, some indications suggest that it might be adaptive and represent a case of convergent ACD (Gorissen et al. 2006; Reif et al. 2015), although direct evidence is still missing. In this context, mixed singing may be a good example of vocal mimicry as defined by Dalziell et al. (2015), where a change in vocalization causing vocal resemblance between species confers a selective advantage to the mimic. An important criterion for demonstrating convergent ACD (Grether et al. 2009; Tobias and Seddon 2009) is to show that convergence in a territorial signal increases the intensity of the territorial response towards heterospecifics. We tested this prediction in 2 closely related oscine species, the Common Nightingale (Luscinia megarhynchos) and the Thrush Nightingale (L. luscinia), which show song convergence in sympatric areas. These species diverged about 1.8 Mya (Storchová et al. 2010) and currently co-occur in a secondary contact zone spreading from north-eastern Germany to the Black Sea (Sorjonen 1986), where they occasionally hybridize (Becker 2007; Kverek et al. 2008; Reifová et al. 2011a; Mořkovský et al. 2018). The 2 species have distinct vocalizations but otherwise are very similar morphologically (with only minor differences in size and the intensity of plumage coloration) as well as ecologically (Cramp and Perrins 1994), which leads to aggressive territorial interactions (Reif et al. 2015). Competition between the species has apparently resulted in partial habitat segregation in sympatry (Reif et al. 2018; Sottas et al. 2018) and ecological character displacement in beak morphology (Reifová et al. 2011b). Several studies also suggest that the Thrush Nightingale, which is slightly larger than the Common Nightingale, might be competitively dominant (Sorjonen 1986; Stadie 1991; Reifová et al. 2011b), although the levels of heterospecific aggression do not differ between the species (Reif et al. 2015). Song convergence in this pair of nightingale species is asymmetric. Although the songs of the 2 species clearly differ in allopatry, Thrush Nightingale males from sympatric populations often incorporate songs of the Common Nightingale in their repertoire (Sorjonen 1986; Vokurková et al. 2013). The frequency of Thrush Nightingale mixed singers varies among different sympatric localities but can be over 70% in localities where both species breed close to each other (Vokurková et al. 2013). The proportion of heterospecific songs in the repertoires of mixed singers is highly variable among individuals; it often exceeds 50%, occasionally reaching up to 100% (Vokurková et al. 2013). To investigate the potential role of Thrush Nightingale mixed singing in interspecific communication and in mediating territorial interactions with the Common Nightingale, we measured the intensity of physical and vocal territorial signals of Common Nightingale males exposed to 3 playback stimuli: conspecific song, pure heterospecific song, and mixed heterospecific song of the Thrush Nightingale. We predicted that if mixed singing has an influence on interspecific male interactions, it should trigger a higher level of aggressive response than pure heterospecific song. Our results are discussed in the context of the theory of convergent ACD. METHODS Study area Common Nightingale males were tested at localities within the zone of sympatry in Poland, in an area of ca 20 × 30 km along the floodplains of rivers Prosna (sites: Chocz, Czołnochów, Lisewo, Ruda Komorska) and Warta (sites: Zagórów, Ląd, Pyzdry) in the Greater Poland Voivodeship (52.12°N, 17.78°E), where both species are common breeders. Individuals were typically found in dense vegetation near riverbanks, where males were commonly observed to defend territories in close range to either conspecific or heterospecific neighbors. Song stimuli Three types of playback stimuli were prepared for the experiment: conspecific stimuli from recordings of Common Nightingale males (CN), heterospecific pure stimuli from recordings of pure-singing Thrush Nightingale males (TN), and heterospecific mixed stimuli from songs of mixed-singing Thrush Nightingale males (Mix) (Figure 1). A previous study, from which our experimental design was derived (Reif et al. 2015), already demonstrated that neither Common nor Thrush Nightingale males show any response to playback of a noncompeting species’ song; we thus did not include such control stimulus in the present experiment. Figure 1 View largeDownload slide Spectrograms from samples of the 3 categories of stimuli: conspecific (Common Nightingale song; a), heterospecific pure (Thrush Nightingale song; b), and mixed (c, d). Mixed Thrush Nightingale songs may include either completely copied Common Nightingale song types (indicated by full horizontal bars), or song types in which syllables characteristic for song types of both species are mixed (striped horizontal bars). Songs are separated by pauses of silence, usually at least several seconds long, and consist of numerous syllables (which we define here as the invariant units within the vocalization, composed of one or multiple continuous traces in the spectrogram). Song types are then defined by a characteristic order of dominant syllables (Vokurková et al. 2013). Detailed spectrograms of selected song types are provided in Supplementary Figure 1. Figure 1 View largeDownload slide Spectrograms from samples of the 3 categories of stimuli: conspecific (Common Nightingale song; a), heterospecific pure (Thrush Nightingale song; b), and mixed (c, d). Mixed Thrush Nightingale songs may include either completely copied Common Nightingale song types (indicated by full horizontal bars), or song types in which syllables characteristic for song types of both species are mixed (striped horizontal bars). Songs are separated by pauses of silence, usually at least several seconds long, and consist of numerous syllables (which we define here as the invariant units within the vocalization, composed of one or multiple continuous traces in the spectrogram). Song types are then defined by a characteristic order of dominant syllables (Vokurková et al. 2013). Detailed spectrograms of selected song types are provided in Supplementary Figure 1. Both pure song categories (CN, TN) contained only species-specific song types (based on a prior exploration of local song repertoires and syntax, as well as on comparison with allopatric populations of either species). The mixed stimulus (Mix) was characterized by the presence of both Thrush Nightingale and Common Nightingale song types (identified by comparison with an extensive catalogue of Common Nightingale song types; Vokurková et al. 2013) within the song of one Thrush Nightingale individual, but could also include song types where typical Common Nightingale syllables occur within Thrush Nightingale song structures (Figure 1, Supplementary Figure 1). The playback stimuli were derived from recordings from different localities in the same study area obtained in 2008 and 2015 in order to minimize potential familiarity with an individual in the same location. All song manipulations and sound analyses were performed using Avisoft-SASlab Pro version 5.2 (www.avisoft.com). For each stimulus category, we used 90-s long sections from 15 to 19 recordings containing active singing without long pauses (pause duration below 6.6 s), selected according to their signal quality (song rate by category (mean ± SD): TN 7.56 ± 1.30, CN 10.89 ± 1.48 and Mix 7.92 ± 1.47 song types/min). The number of stimuli for testing at each locality was constrained by the need to use songs originating in the region (and thus unlikely to be perceived as foreign) but at places distant enough to reduce the risk of potential previous experience of the tested birds with the same singer or repertoire. Thus, while the majority of stimuli were only used once in the experiments, some had to be re-used (number of use (mean ± SD): 1.53 ± 0.65). All stimuli were edited to remove the background noise (including the songs of other species). Some recordings of Thrush Nightingale mixed songs were further edited by removing some songs in order to balance the ratio of each species’ song types (proportion of Thrush Nightingale song types in stimuli (mean ± SD): 63.5 ± 19.5%). The amplitude of the tracks was normalized to 50% of the dynamic range (with bias offset from the signal removed) in Avisoft-SASLab Pro. During playback, the volume was standardized according to local conditions to match the natural amplitude of a male singing from the same distance. Experimental procedure Playback experiments were conducted during the first 2 weeks of May 2016, soon after the return of the 2 species from their wintering grounds in Africa, as this is the most active period of territorial competition for males (Cramp and Perrins 1994; Reif et al. 2015). The tests were performed during daylight (from 0600 to 1900 h) in the absence of rain and strong wind, when the birds were spontaneously singing and most likely willing to react to intruders. The experimental design generally followed Reif et al. (2015) with some modifications needed for the purposes of this study. We used a speaker (MIPRO MA-101) mounted on a tripod, on top of which a taxidermic dummy of either species was attached: Common Nightingale for conspecific playbacks and Thrush Nightingale for heterospecific playbacks. These visually very similar dummies (reflecting the close phenotypic similarity of both species) were used to provide a target allowing the tested bird to express a full range of behavioral responses (see Petrusková et al. 2008 for further discussion of the pros and cons of dummy use in playback experiments). Vocal responses of the tested males were recorded using a digital recorder (Marantz PMD660) connected to a directional microphone (Sennheiser ME67), and direct behavioral observations were performed by a pair of observers (AS, HK) from a safe distance (ca. 15–20 m), to avoid potential disturbances by the presence of humans. The presence of a bird in its territory was assessed from both acoustic cues and visual observations. Each tested individual was presented with all 3 stimuli in a randomized order. For the experiment, we chose actively singing territorial males that did not directly interact with their neighbors. The experimental device (loudspeaker with the taxidermic dummy) was set ca 1 m from the vegetation patch where the bird had been singing. Several minutes after the equipment set-up, the first stimulus was played for 3 min. Song and other behavioral reactions were recorded during the playback and 3 min after the playback ceased (to also include late responses). The same procedure was repeated with the remaining stimuli, each after a break of at least 1 h to avoid overstimulation and to clearly differentiate among stimuli. All 3 stimuli were tested on the same day, to minimize the chance of predation or territorial shifts between days. Directly neighboring birds were not tested on the same or subsequent days. Shortly after the experiment, the tested bird was lured by playback and captured by mist-netting to confirm the species identity, and to avoid any risk of repeated testing of the same individual in the future. Evaluation of male territorial response to playback Overall, we analyzed the responses of 27 Common Nightingale males. The full data set consisted of 21 males for all 3 types of stimuli; for 6 individuals, 1 stimulus had to be excluded (in 3 cases heterospecific pure stimulus, twice heterospecific mixed stimulus, once conspecific stimulus) due to a disturbance in experimental conditions (weather, intrusion by another male, etc.) or the absence of the bird in its territory for a given playback trial. The territorial response of nightingale males to a simulated intrusion of the competitor into its territory (as already studied by Reif et al. 2015) typically begins with an expression of interest by approach, that is, an initial exploratory behavior within the vegetation cover, for example jumping to a closer perch soon after the playback starts. More intensive responses, ordered by increased aggressiveness (Reif et al. 2015), include the following: 1) flyover, that is, threatening the presumed intruder by a short flight from a nearby location (but more than 1 m away from the dummy); 2) jumping out of the vegetation cover to restlessly investigate the playback device from the ground (jumping, or “running” in Reif et al. 2015), 3) flight-attack, that is, flight directed towards the dummy with a very close approach (below 1 m), and 4) physical contact, that is, escalating aggression to direct physical attacks on the dummy. Furthermore, most males respond to playback by singing either during or soon after the playback stimulus. The physical territorial response was monitored during the playback itself (3 min) and during the subsequent 3 min of the post-playback phase (i.e. 6 min in total). The responses were described by a set of behavioral categories (see Table 1) ranging from low to high level of aggressiveness as follows: jumping on the ground (count), flyovers (count), and direct physical contact (% of time attacking); no flight attacks were observed in the present experiments. In addition, we evaluated the strongest physical response shown by the bird during the experiment (as defined above, ranging from 1 = interest in the dummy without close approach, to 5 = physical attack) as an ordinal variable. The distance to the dummy was assessed throughout the experiment, as it is also expected to reflect different degrees of risk taken by the bird and its motivation to get involved in agonistic behavior (Searcy et al. 1997; Hyman et al. 2004; Reif et al. 2015), and the minimal estimated distance category was noted. Time spent by a tested individual within 4 different distance ranges from the dummy (<1 m, 1–3 m, 3–5 m, and >5 m) was also quantified. Table 1 Factor loadings obtained by principal component analysis on the responses of Common Nightingale and Thrush Nightingale males during playback experiments Variable Description PC1 PC2 Song duration mean duration of songs −0.21 0.34 Pause duration mean duration of pauses (shorter than 6.6 s) −0.17 −0.84 Song rate number of songs per minute 0.25 0.58 Singing time % of time spent actively singing −0.07 0.82 Jumping excited jumping below the dummy 0.82 0.14 Flyovers aggressive flying over the dummy 0.27 −0.06 Physical contact time spent on the dummy, attacking 0.65 −0.19 Max aggressiveness highest level on aggressiveness scale 0.92 0.07 Closest distance minimal distance to dummy (m) −0.59 −0.51 Time <1 m time spent at distance 0 to 1 m 0.76 0.13 Time 1–3 m time spent at distance 1 to 3 m 0.60 0.40 Time 3–5 m time spent at distance 3 to 5 m 0.00 0.46 Time >5 m time spent at distance over 5 m −0.57 −0.49 Variable Description PC1 PC2 Song duration mean duration of songs −0.21 0.34 Pause duration mean duration of pauses (shorter than 6.6 s) −0.17 −0.84 Song rate number of songs per minute 0.25 0.58 Singing time % of time spent actively singing −0.07 0.82 Jumping excited jumping below the dummy 0.82 0.14 Flyovers aggressive flying over the dummy 0.27 −0.06 Physical contact time spent on the dummy, attacking 0.65 −0.19 Max aggressiveness highest level on aggressiveness scale 0.92 0.07 Closest distance minimal distance to dummy (m) −0.59 −0.51 Time <1 m time spent at distance 0 to 1 m 0.76 0.13 Time 1–3 m time spent at distance 1 to 3 m 0.60 0.40 Time 3–5 m time spent at distance 3 to 5 m 0.00 0.46 Time >5 m time spent at distance over 5 m −0.57 −0.49 View Large Table 1 Factor loadings obtained by principal component analysis on the responses of Common Nightingale and Thrush Nightingale males during playback experiments Variable Description PC1 PC2 Song duration mean duration of songs −0.21 0.34 Pause duration mean duration of pauses (shorter than 6.6 s) −0.17 −0.84 Song rate number of songs per minute 0.25 0.58 Singing time % of time spent actively singing −0.07 0.82 Jumping excited jumping below the dummy 0.82 0.14 Flyovers aggressive flying over the dummy 0.27 −0.06 Physical contact time spent on the dummy, attacking 0.65 −0.19 Max aggressiveness highest level on aggressiveness scale 0.92 0.07 Closest distance minimal distance to dummy (m) −0.59 −0.51 Time <1 m time spent at distance 0 to 1 m 0.76 0.13 Time 1–3 m time spent at distance 1 to 3 m 0.60 0.40 Time 3–5 m time spent at distance 3 to 5 m 0.00 0.46 Time >5 m time spent at distance over 5 m −0.57 −0.49 Variable Description PC1 PC2 Song duration mean duration of songs −0.21 0.34 Pause duration mean duration of pauses (shorter than 6.6 s) −0.17 −0.84 Song rate number of songs per minute 0.25 0.58 Singing time % of time spent actively singing −0.07 0.82 Jumping excited jumping below the dummy 0.82 0.14 Flyovers aggressive flying over the dummy 0.27 −0.06 Physical contact time spent on the dummy, attacking 0.65 −0.19 Max aggressiveness highest level on aggressiveness scale 0.92 0.07 Closest distance minimal distance to dummy (m) −0.59 −0.51 Time <1 m time spent at distance 0 to 1 m 0.76 0.13 Time 1–3 m time spent at distance 1 to 3 m 0.60 0.40 Time 3–5 m time spent at distance 3 to 5 m 0.00 0.46 Time >5 m time spent at distance over 5 m −0.57 −0.49 View Large Several song parameters (Table 1) were extracted from the recordings covering 6 min of both the playback and post-playback phases, based on variables already used in previous studies on Luscinia species (e.g. Kunc et al. 2005; Turčoková et al. 2011): mean song duration (in seconds), mean pause duration (the duration of silence between 2 consecutive songs, in seconds), song rate (mean number of songs per minute), and proportion of time spent by singing (% of time of bird presence). Statistical analysis Statistical analysis of the physical and vocal territorial responses of the tested males was performed with R software 3.2.2 (R Core Team 2014). We first used Principal Component Analysis (PCA) to summarize multiple song and behavioral variables measured during the experiments into independent factors suitable for further analyses. PCA was based on a correlation matrix of all 13 variables describing both physical and vocal responses (Table 1). It was run using the “principal” function in the “psych” package v. 1.5.8 (Revelle 2017) with a varimax rotation to maximize the variances of the squared normalized factor loadings across variables for each factor, thus facilitating interpretation of the resulting principal components. Then, we used the first 2 principal components (i.e. those accounting for most of the variation in the data) as response variables in 2 linear mixed-effects models (one for each principal component) to test whether the responses differed between different types of stimuli. Besides song stimulus, each model contained a fixed-effect variable expressing the order of stimuli (from 1 to 3). This allowed us to control for a possible influence of habituation or overstimulation (e.g. Petrinovich and Patterson 1979; Petrusková et al. 2008; Turčoková et al. 2011) caused by presenting several stimuli to the same individual during the experiment. Individual identity was treated as a random effect in the models. Linear mixed-effects models were run with the “lme” function in the “nlme” package v. 3.1-122 (Pinheiro et al. 2017), and we used type III ANOVA from the “car” package v. 2.1-0 (Fox and Weisberg 2011) to test for the significance of respective variables. Ethical note No individual was hurt during our noninvasive experiments. Upon completion of the experiment, individuals were captured by mist-netting with a conspecific playback lure and ringed for subsequent recognition. Manipulation with birds was approved by the Local Ethical Committee for Scientific Experiments on Animals in Poznań (permission no. 17/2015) and capturing the birds was performed under a license provided by the Gdansk Bird Ringing Centre (license no. 405/2051). RESULTS All tested males responded to the playback stimulation, irrespectively of the presented stimulus. In the vast majority of cases, reaction to playback started with a cautious approach to observe the “intruder,” usually followed a moment later by actively counter-singing towards the speaker, before escalating to stronger physical territorial reactions in many cases. The first 2 principal components explained altogether 50.4% of the overall variance. The first PC axis (PC1, explaining 34.6% of the variance) was positively correlated with the intensity of the physical territorial response and the time spent closer than 3 m from the loudspeaker with the dummy. The second axis (PC2, explaining 15.8% of the variance) was positively correlated with the intensity of the vocal territorial response and the time spent at an intermediate distance, from 3 to 5 m (Figure 2). Figure 2 View largeDownload slide Principal Component Analysis (after normalized varimax rotation) based on the categories of distance (purple), singing characteristics (green) and physical aggressive responses (orange). The first 2 principal components explain 34.6% (first axis) and 15.8% (second axis) of the overall variation. The dots represent Common Nightingale individuals tested for the respective stimuli: conspecific (red triangle), mixed (green square), and pure heterospecific (blue circle). Figure 2 View largeDownload slide Principal Component Analysis (after normalized varimax rotation) based on the categories of distance (purple), singing characteristics (green) and physical aggressive responses (orange). The first 2 principal components explain 34.6% (first axis) and 15.8% (second axis) of the overall variation. The dots represent Common Nightingale individuals tested for the respective stimuli: conspecific (red triangle), mixed (green square), and pure heterospecific (blue circle). According to the linear mixed-effects model, the physical territorial response of the tested males (reflected by PC1) significantly differed among playback stimuli (Χ2 = 34.4, P = 3.4 × 10−8, df = 2, 70), with a more aggressive response to conspecific stimulus than to both heterospecific ones, either pure or mixed (Figure 3a, Supplementary Table 1). We did not observe a significant difference between the mixed and pure heterospecific stimuli (Figure 3a). This outcome was in line with the general expectation of a more aggressive physical territorial response to conspecific than to heterospecific stimuli, being reflected by closer approaches to the device, more time spent on the ground investigating it, as well as the occurrence of physical attacks (which occurred in conspecific trials only). Figure 3 View largeDownload slide Mean scores for responses of Common Nightingale males to particular song stimuli (LL = Thrush Nightingale, LM = Common Nightingale, mix = mixed singing from Thrush Nightingale) during playback experiments estimated by a linear mixed-effects model run separately for PC1 corresponding to physical responses (a) and PC2 corresponding to vocal responses (b). Box plots depict the median, interquartile range, and range. (NS: P > 0.05, *P < 0.05, **P < 0.01, ***P < 0.001). Figure 3 View largeDownload slide Mean scores for responses of Common Nightingale males to particular song stimuli (LL = Thrush Nightingale, LM = Common Nightingale, mix = mixed singing from Thrush Nightingale) during playback experiments estimated by a linear mixed-effects model run separately for PC1 corresponding to physical responses (a) and PC2 corresponding to vocal responses (b). Box plots depict the median, interquartile range, and range. (NS: P > 0.05, *P < 0.05, **P < 0.01, ***P < 0.001). The vocal territorial response of the tested birds (reflected by PC2) also significantly differed among stimuli (Χ2 = 27.4, P = 1.1 × 10−6, df = 2, 70). Specifically, the males exhibited a stronger reaction to the conspecific stimulus than to pure and mixed heterospecific stimuli (Figure 3b, Supplementary Table 1), demonstrated by more time spent singing, a higher song rate, longer song duration and accordingly shorter pauses in song bouts. In contrast to the physical response, however, the vocal response to the mixed heterospecific stimulus was significantly more intense than the response to the pure heterospecific stimulus (Figure 3b, Supplementary Table 1). The intensity of the response to the mixed heterospecific stimulus was thus intermediate between the responses to pure heterospecific and conspecific stimuli. Although a habituation effect could be expected due to presentation of the respective stimuli to the same individual within a few hours, we found no significant effect of the order of the presented stimuli on either physical or vocal territorial responses (PC1: Χ2 = 3.84, P = 0.15, df = 2, 70; PC2: Χ2 = 1.92, P = 0.38, df = 2, 70). DISCUSSION Our playback experiments testing whether mixed singing of Thrush Nightingale males in the secondary contact zone with the Common Nightingale affects the level of interspecific territoriality between the species resulted in 2 main findings. 1) Common Nightingale males discriminated between intruders singing conspecific and heterospecific songs, regardless of whether the heterospecific stimulus was pure or mixed. This was manifested by a stronger physical territorial response of the tested males to the conspecific stimulus than to both pure and mixed heterospecific stimuli. 2) However, the vocal territorial response was gradually more intense with an increasing presence of Common Nightingale elements in the stimuli, from a low response to pure heterospecific, an intermediate response to mixed heterospecific, to a strong reaction to conspecific songs. This suggests that the presence of Common Nightingale song characteristics in Thrush Nightingale songs triggers a higher level of the vocal territorial response in Common Nightingale males but does not affect the physical territorial response. Our results support the view that mixed singing performed by Thrush Nightingale males has a role in between-species communication and can mediate interspecific territoriality. Below, we discuss our results in the context of convergent ACD (Grether et al. 2009). We argue that mixed singing in oscine birds might be adaptive in terms of facilitating the coexistence of closely related species, and thus fits the concept of vocal mimicry (Dalziell et al. 2015). However, we also mention possible alternative explanations of the presented results. Grether et al. (2009) defined 8 criteria for demonstrating ACD. The phenotypic shift 1) should be at least partially genetically determined, 2) should be unlikely to have arisen by chance, 3) should represent a true evolutionary change and not the result of biased colonization or extinction events, 4) should persist after controlling for environmental variables other than the presence of the competing species, 5) is not likely to be a pleiotropic effect of another evolutionary process, and 6) is not a product of hybridization. In addition, there should be 7) evidence that the observed phenotypic shift affects the intensity of interspecific interference competition and 8) independent evidence for interspecific interference competition. Meeting all these criteria is quite difficult. Grether et al. (2009) reported 18 possible examples of convergent ACD across fishes, amphibians, reptiles, birds, and insects, including as many as 12 cases of convergence in bird vocal signals and vocal recognition. However, only one of these examples, concerning heterospecific aggressive behavior in the brook stickleback Culaea inconstans (Peiman and Robinson 2007), satisfied all 8 criteria for demonstrating ACD, whereas the rest of examples satisfied only 3 to 5 criteria. Based on the results presented in this study and previous studies, we have assembled evidence for 6 criteria, making our system one of the best-supported examples of convergent ACD. Character displacement is typically demonstrated as a phenotypic shift between sympatric and allopatric populations (Brown and Wilson 1956). In the case of convergent ACD, this shift results in more similar phenotypes in sympatry than in allopatry. To show that this pattern is not caused by chance (criterion 2), it is desirable to show that the convergence occurs in multiple independent populations. In these 2 nightingale species, song convergence has been independently demonstrated in multiple parts of the secondary contact zone, including Poland (Sorjonen 1986; Vokurková et al. 2013), Germany (Becker 2007) and Hungary (Schmidt 1973). Moreover, the change in the Thrush Nightingale song cannot result from biased colonization or extinction (criterion 3), as no mixed singers have been observed in allopatry (Vokurková et al. 2013). It is also highly unlikely that the mixed song in the Thrush Nightingale represents an adaptation to the local acoustic environment (criterion 4), as both species occupy very similar habitats in sympatry and allopatry (Kuczyński and Chylarecki 2012), and if anything there is a tendency to habitat divergence rather than convergence in sympatry (Reif et al. 2018). The effect of hybridization (criterion 6) has also been ruled out as a possible cause of mixed singing in nightingales (Vokurková et al. 2013). Finally, in this study we provide evidence that the observed phenotypic shift affects the intensity of interspecific interference competition (criterion 7), which complements the previously provided evidence for interspecific interference competition between both nightingale species (criterion 8) (Reif et al. 2015). One strong argument against the possible adaptiveness of mixed singing concerns the fact that in oscine birds, the song is not genetically determined (criterion 1), but culturally learned (Catchpole and Slater 2008). In this situation, mixed singing does not need to be a heritable and thus evolvable trait. However, new studies on the genetic basis of the song learning process have shown that although song is learned in oscine birds, there is a genetically determined ability to discriminate the species-specific song, which is then preferentially learned (Scharff and Adam 2013; Wheatcroft and Qvarnström 2017). This genetically determined song discrimination can play an important role in maintaining reproductive isolation between closely related species by reducing the probability of heterospecific song learning (Wheatcroft and Qvarnström 2017). Nevertheless, mutations in genes for song discrimination could cause the bird to learn not only conspecific but also heterospecific songs. Mixed singing might thus in principle have a genetic basis. However, to prove this in our system, and demonstrate that mixed singing in the Thrush Nightingale evolved to reduce the level of interspecific competition, it would be necessary to show that the ability (or tendency) to learn heterospecific songs is higher in sympatric than in allopatric populations. According to criterion 5, a pleiotropic effect of another evolutionary process should be ruled out. Vokurková et al. (2013) speculated that mixed singing might possibly be the result of sexual selection rather than selected to reduce interspecific competition. Without the knowledge whether Thrush Nightingale females respond differently to pure and mixed song, it is difficult to distinguish between these alternatives. However, we consider the explanation based on sexual selection alone unlikely, given that a preference of Thrush Nightingale females for Common Nightingale song would lead to increased interspecific hybridization, as has been observed in Ficedula flycatchers (Qvarnström et al. 2006). It appears surprising that mixed singing is so common in the Thrush Nightingale but has not been observed in the Common Nightingale. Copying of song types from conspecific neighboring males, which facilitates territorial interactions, has been documented within populations of both species (Hultsch and Todt 1981; Naguib and Todt 1998), and experiments in captivity demonstrated that both are able to learn the heterospecific song (Stadie 1983). Nevertheless, the reaction to heterospecific songs in sympatry differs between species: in playback experiments, the Thrush Nightingale males reacted with the same intensity to conspecific songs and to heterospecific songs of the Common Nightingale, while Common Nightingale males reacted to heterospecific songs more weakly (Reif et al. 2015). This difference, however, may be a direct consequence of mixed singing. Some Thrush Nightingale males in sympatry seem to compose their song entirely from Common Nightingale song types (Vokurková et al. 2013), and their conspecifics should thus react to the Common Nightingale song as to their own. According to the physical territorial response in our experiments, Common Nightingale males were clearly able to distinguish between conspecific and heterospecific intruders, no matter whether the latter sing pure or mixed songs. By contrast, the vocal territorial response of Common Nightingale males was strongest for the conspecific stimuli, less strong for the mixed heterospecific stimuli and weakest for the pure heterospecific stimuli. The differences in responses to conspecific and heterospecific stimuli are unlikely to have been caused by accompanying taxidermic dummies. The tested males usually did not approach closer than 5 m from the dummy in heterospecific trials, and thus were unlikely to distinguish the minor species-specific differences between the Common and Thrush Nightingale dummies. Thus, we presume that the male responses were triggered by the acoustic stimuli alone, not only when comparing pure and mixed heterospecific song (accompanied by the same Thrush Nightingale dummy) but also when comparing the conspecific and heterospecific stimuli. This is also consistent with the species’ biology: singing interactions of nightingales usually happen from a large distance out of sight range, often at night (Hultsch and Todt 1981; Griessmann and Naguib 2002), and therefore play a key role in their territoriality. Our findings, together with other recent works on the nightingale system (Vokurková et al. 2013; Reif et al. 2015), support the view that mixed singing performed by Thrush Nightingale males in the secondary contact zone with the Common Nightingale may represent a case of convergent ACD, and, in a broader sense, a prominent case of vocal mimicry. However, further work on the genetic basis of song discrimination in sympatric and allopatric populations of the Thrush Nightingale as well as studies demonstrating that mixed singing increases the fitness of the sympatric Thrush Nightingale males would be needed to confirm the adaptive role of mixed singing in facilitating the coexistence of these closely related species before divergence in ecological niches evolves. FUNDING The study was supported by grants of the Czech Science Foundation (grant numbers 15-10884Y, 18-14325S) to R.R. We wish to thank T. Albrecht, B. Cramer, P. Dolata, P. Kverek, K. Opletalová, and C. Sottas for field assistance and 2 anonymous reviewers for their constructive comments on earlier versions of this paper. T. Osiejuk kindly helped us with obtaining permissions for research. Authors’ Contributions: R.R., T.P., A.P., and J.R. conceived the idea; A.S., J.R., T.P., A.P., and R.R. designed the research; A.S., H.K., J.V., and J.R. collected data; A.S., H.K., T.P., A.P., and J.R. analyzed data; A.S., R.R., and T.P. drafted the manuscript with inputs from all authors. All authors gave final approval for publication. Conflict of Interest: The authors declare no competing interests. Data accessibility: Analyses reported in this article can be reproduced using the data provided in Souriau et al. (2018). REFERENCES Becker J . 2007 . Nightingales Luscinia megarhynchos, Thrush Nightingales L. luscinia and their hybrids in the area of Frankfurt (Oder)–further results of long-term investigations using bird ringing . Vogelwarte . 45 : 15 . Brown RN . 1977 . Character convergence in bird song . Can J Zool . 55 : 1523 – 1529 . Google Scholar CrossRef Search ADS Brown WL , Wilson EO . 1956 . Character displacement . Syst Zool . 5 : 49 – 64 . Google Scholar CrossRef Search ADS Butlin RK . 1989 . Reinforcement of premating isolation . In: Otte D , Endler JA , editors. Speciation and its consequences . Sunderland (MA) : Sinauer Associates . p. 158 – 179 . Catchpole CK , Slater PJ . 2008 . Bird song: biological themes and variations . Cambridge : Cambridge University Press . Google Scholar CrossRef Search ADS Cody ML . 1969 . Convergent characteristics in sympatric species: a possible relation to interspecific competition and aggression . Condor . 71 : 223 – 239 . Google Scholar CrossRef Search ADS Cramp S , Perrins CM . 1994 . Birds of the western palearctic . Oxford : Oxford University Press . Dalziell AH , Welbergen JA , Igic B , Magrath RD . 2015 . Avian vocal mimicry: a unified conceptual framework . Biol Rev . 90 : 643 – 668 . Google Scholar CrossRef Search ADS PubMed Dobzhansky T . 1937 . Genetics and the origin of species . New York : Columbia University Press . Fox J , Weisberg S . 2011 . An R companion to applied regression . 2nd ed . Thousand Oaks : Sage Publishing . Available from: http://CRAN.R-project.org/package=car. Gil D . 2010 . Increased response of the Short-toed Treecreeper Certhia brachydactyla in sympatry to the playback of the song of the Common Treecreeper C. familiaris . Ethology . 103 : 632 – 641 . Google Scholar CrossRef Search ADS Gorissen L , Gorissen M , Eens M . 2006 . Heterospecific song matching in two closely related songbirds (Parus major and P. caeruleus): Great Tits match Blue Tits but not vice versa . Behav Ecol Sociobiol . 60 : 260 – 269 . Google Scholar CrossRef Search ADS Grant PR , Grant BR . 2006 . Evolution of character displacement in Darwin’s finches . Science . 313 : 224 – 226 . Google Scholar CrossRef Search ADS PubMed Grether GF , Losin N , Anderson CN , Okamoto K . 2009 . The role of interspecific interference competition in character displacement and the evolution of competitor recognition . Biol Rev . 84 : 617 – 635 . Google Scholar CrossRef Search ADS PubMed Grether GF , Peiman KS , Tobias JA , Robinson BW . 2017 . Causes and consequences of behavioral interference between species . Trends Ecol Evol . 32 : 760 – 772 . Google Scholar CrossRef Search ADS PubMed Griessmann B , Naguib M . 2002 . Song sharing in neighboring and non-neighboring Thrush Nightingales (Luscinia luscinia) and its implications for communication . Ethology . 108 : 377 – 387 . Google Scholar CrossRef Search ADS Haavie J , Borge T , Bures S , Garamszegi LZ , Lampe HM , Moreno J , Qvarnström A , Török J , Saetre GP . 2004 . Flycatcher song in allopatry and sympatry - convergence, divergence and reinforcement . J Evol Biol . 17 : 227 – 237 . Google Scholar CrossRef Search ADS PubMed Helb HW , Dowsett-Lemaire F , Bergmann HH , Conrads K . 1985 . Mixed singing in European songbirds - a review . Z Tierpsychol . 69 : 27 – 41 . Google Scholar CrossRef Search ADS Hultsch H , Todt D . 1981 . Repertoire sharing and song-post distance in Nightingales (Luscinia megarhynchos B.) . Behav Ecol Sociobiol . 8 : 183 – 188 . Google Scholar CrossRef Search ADS Hyman J , Searcy W , Nowicki S , Hughes M . 2004 . Individual variation in the strength of territory defense in male Song Sparrows: correlates of age, territory tenure, and neighbor aggressiveness . Behaviour . 141 : 15 – 27 . Google Scholar CrossRef Search ADS Kuczyński L , Chylarecki P . 2012 . Atlas of common breeding birds in Poland: distribution, habitat preferences and population trends . Warszawa : GIOŚ . Kunc H , Amrhein V , Naguib M . 2005 . Acoustic features of song categories and their possible implications for communication in the Common Nightingale (Luscinia megarhynchos) . Behaviour . 142 : 1077 – 1091 . Google Scholar CrossRef Search ADS Kverek P , Storchová R , Reif J , Nachman MW . 2008 . Occurrence of a hybrid between the Common Nightingale (Luscinia megarhynchos) and the Thrush Nightingale (Luscinia luscinia) in the Czech Republic confirmed by genetic analysis . Sylvia . 44 : 17 – 26 . Mořkovský L , Janoušek V , Reif J , Rídl J , Pačes J , Choleva L , Janko K , Nachman MW , Reifová R . 2018 . Genomic islands of differentiation in two songbird species reveal candidate genes for hybrid female sterility . Mol Ecol . 27 : 949 – 958 . Google Scholar CrossRef Search ADS PubMed Naguib M , Todt D . 1998 . Recognition of neighbors’ song in a species with large and complex song repertoires: the Thrush Nightingale . J Avian Biol . 29 : 155 – 160 . Google Scholar CrossRef Search ADS Peiman KS , Robinson BW . 2007 . Heterospecific aggression and adaptive divergence in brook stickleback (Culaea inconstans) . Evolution . 61 : 1327 – 1338 . Google Scholar CrossRef Search ADS PubMed Petrinovich L , Patterson TL . 1979 . Field studies of habituation: I. Effect of reproductive condition, number of trials, and different delay intervals on responses of the white-crowned sparrow . J Comp Physiol Psychol . 93 : 337 – 350 . Google Scholar CrossRef Search ADS Pfennig KS , Pfennig DW . 2009 . Character displacement: ecological and reproductive responses to a common evolutionary problem . Q Rev Biol . 84 : 253 – 276 . Google Scholar CrossRef Search ADS PubMed Pinheiro J , Bates D , DebRoy S , Sarkar D , R Core Team . 2017 . nlme: linear and nonlinear mixed effects models. R package version 3.1–122 . Available from: http://CRAN.R-project.org/package=nlme. Petrusková T , Petrusek A , Pavel V , Fuchs R . 2008 . When an alien sings at a rival’s post: a passerine excited by conspecific stimulus may show aggressive behaviour towards heterospecific individuals . Folia Zool . 57 : 201 – 211 . Qvarnström A , Haavie J , Saether SA , Eriksson D , Part T . 2006 . Song similarity predicts hybridization in flycatchers . J Evol Biol . 19 : 1202 – 1209 . Google Scholar CrossRef Search ADS PubMed R Core Team . 2014 . R: a language and environment for statistical computing . Vienna (Austria) : R Foundation for Statistical Computing . Available from: http://www.R-project.org. Reif J , Jiran M , Reifová R , Vokurková J , Dolata PT , Petrusek A , Petrusková T . 2015 . Interspecific territoriality in two songbird species: potential role of song convergence in male aggressive interactions . Anim Behav . 104 : 131 – 136 . Google Scholar CrossRef Search ADS Reif J , Reifová R , Skoracka A , Kuczyn′ski L . 2018 . Competition driven niche segregation on a landscape scale: evidence for escaping from syntopy toward allotopy in coexisting sibling passerine species . J Anim Ecol . https://doi.org/10.1111/1365-2656.12808 Reifová R , Kverek P , Reif J . 2011a . The first record of a female hybrid between the Common Nightingale (Luscinia megarhynchos) and the Thrush Nightingale (Luscinia luscinia) in nature . J Ornithol . 152 : 1063 – 1068 . Google Scholar CrossRef Search ADS Reifová R , Reif J , Antczak M , Nachman MW . 2011b . Ecological character displacement in the face of gene flow: evidence from two species of nightingales . BMC Evol Biol . 11 : 138 . Google Scholar CrossRef Search ADS Revelle W . 2017 . psych: procedures for personality and psychological research . Evanston (IL) : Northwestern University . Available from: http://CRAN.R-project.org/package=psych. Scharff C , Adam I . 2013 . Neurogenetics of birdsong . Curr Opin Neurobiol . 23 : 29 – 36 . Google Scholar CrossRef Search ADS PubMed Schmidt E . 1973 . Zur Ökologie des Sprossers (Luscinia luscinia L.) in der Theiss-auen bein Tiszatelek, nach seiner Ankunft im Frühling . Tiscia . 8 : 79 – 81 . Searcy WA , Nowicki S , Hughes M . 1997 . The response of male and female Song Sparrows to geographic variation in song . Condor . 99 : 651 – 657 . Google Scholar CrossRef Search ADS Secondi J , Bretagnolle V , Compagnon C , Faivre B . 2003 . Species-specific song convergence in a moving hybrid zone between two passerines . Biol J Linnean Soc . 80 : 507 – 517 . Google Scholar CrossRef Search ADS Slater PJB . 1989 . Bird song learning: causes and consequences . Ethol Ecol Evol . 1 : 19 – 46 . Google Scholar CrossRef Search ADS Sorjonen J . 1986 . Mixed singing and interspecific territoriality – consequences of secondary contact of two ecologically and morphologically similar nightingale species in Europe . Ornis Scand . 17 : 53 – 67 . Google Scholar CrossRef Search ADS Sottas C , Reif J , Kuczyn′ski L , Reifová R . 2018 . Interspecific competition promotes habitat and morphological divergence in a secondary contact zone between two hybridizing songbirds . J Evol Biol . https://doi.org/10.1111/jeb.13275 Souriau A , Kohoutová H , Reif J , Vokurková J , Petrusek A , Reifová R , Petrusková T . 2018 . Data from: Can mixed singing facilitate coexistence of closely related nightingale species ? Dryad Digital Repository . http://dx.doi.org/10.5061/dryad.9987q2k. Stadie C . 1983 . Erste ergebnisse zur verwandtschaftlichen beziehung zwischen Nachtigall (Luscinia megarhynchos) und Sprosser (Luscinia luscinia) . Voliere . 6 : 87 – 91 . Stadie C . 1991 . Erdsanger I; Nachtigall und Sprosser . Europäische Vogelwelt (Sonderheft) . 3 : 130 – 189 . Storchová R , Reif J , Nachman MW . 2010 . Female heterogamety and speciation: reduced introgression of the Z chromosome between two species of nightingales . Evolution . 64 : 456 – 471 . Google Scholar CrossRef Search ADS PubMed Tobias JA , Cornwallis CK , Derryberry EP , Claramunt S , Brumfield RT , Seddon N . 2014 . Species coexistence and the dynamics of phenotypic evolution in adaptive radiation . Nature . 506 : 359 – 363 . Google Scholar CrossRef Search ADS PubMed Tobias JA , Seddon N . 2009 . Signal design and perception in Hypocnemis antbirds: evidence for convergent evolution via social selection . Evolution . 63 : 3168 – 3189 . Google Scholar CrossRef Search ADS PubMed Turčoková L , Pavel V , Chutný B , Petrusek A , Petrusková T . 2011 . Differential response of males of a subarctic population of Bluethroat Luscinina svecica svecica to playbacks of their own and foreign subspecies . J Ornithol . 152 : 975 – 982 . Google Scholar CrossRef Search ADS Vokurková J , Petrusková T , Reifová R , Kozman A , Mořkovský L , Kipper S , Weiss M , Reif J , Dolata PT , Petrusek A . 2013 . The causes and evolutionary consequences of mixed singing in two hybridizing songbird species (Luscinia spp.) . PLoS One . 8 : e60172 . Google Scholar CrossRef Search ADS PubMed Wheatcroft D , Qvarnström A . 2017 . Genetic divergence of early song discrimination between two young songbird species . Nature Ecol Evol . 1 : 192 . Google Scholar CrossRef Search ADS © The Author(s) 2018. Published by Oxford University Press on behalf of the International Society for Behavioral Ecology. All rights reserved. For permissions, please e-mail: firstname.lastname@example.org This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
Behavioral Ecology – Oxford University Press
Published: Apr 18, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera