Are Different Song Types used in Different Contexts in

Are Different Song Types used in Different Contexts in Brownish-flanked Bush Warblers? Author(s): Canwei Xia , Jiayu Liu , and Yanyun Zhang Source: Zoological Science, 30(9):699-703. 2013. Published By: Zoological Society of Japan DOI: http://dx.doi.org/10.2108/zsj.30.699 URL: http://www.bioone.org/doi/full/10.2108/zsj.30.699

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research.

¤ 2013 Zoological Society of Japan

ZOOLOGICAL SCIENCE 30: 699–703 (2013)

Are Different Song Types Used in Different Contexts in Brownish-flanked Bush Warblers? Canwei Xia, Jiayu Liu, and Yanyun Zhang* Ministry of Education Key Laboratory for Biodiversity and Ecological Engineering, College of Life Sciences, Beijing Normal University, Beijing 100875, China

The repertoires of most songbirds consist of two or more different song types. Studies on a growing number of species have indicated that song types are used selectively in different contexts. For example, two bush warblers (Cettia diphone and C. cettia) use song types selectively in spontaneous singing and playback-evoked singing. Most male brownish-flanked bush warblers (C. fortipes) have two distinct song types. In this study, we tested whether these two song types are used selectively in different contexts. Most frequency and temporal variables are significantly different between these two song types, but neither song type is used more frequently during or after playback compared to spontaneous singing. Additionally, males did not respond differently to the two song types during playback experiments. In conclusion, these two song types differ significantly in both frequency and temporal variables, but we did not find any evidence that they are used selectively in different contexts. Comparing two related bush warblers (C. diphone and C. cettia) that selectively use different song types in different contexts, our study offers further evidence that avian vocalization is a plastic characteristic and can vary considerably, even among closely related species. Key words: brownish-flanked bush warbler, song types, spontaneous singing, playback-evoked singing, aggressive motivation

INTRODUCTION Approximately 70% of songbirds sing two or more different song types (Beecher and Brenowitz, 2005). Some studies have suggested that different song types have the same function (both territorial defense and mate attraction), and the function of having different song types is to increase song diversity (e.g., Catchpole and Slater, 2008; Berglund et al., 1996; Searcy, 1992). However, some studies have indicated that song types are used selectively in different contexts (Brunner and Pasinelli, 2010; Byers et al., 2010), such as spontaneous or territorial dispute singing, or for longrange or short-range communication. For example, several North American wood warblers (Parulidae) have two distinct singing modes, one that is mainly used in spontaneous singing and another that is mainly used during territorial disputes (Spector, 1992). Some New World passerines use one song category with low amplitudes for short-range communication, but another song category with high amplitudes for long-range communication (Morton, 2000). Apart from being used in different contexts, different song types can reflect different levels of aggression. For example, song types with a long duration tend to induce more aggressive responses in the white-crowned sparrow Zonotrichia leucophrys (Nelson and Poesel, 2012), and yellowhammer Emberiza citrinella responds strongly to song types with low frequency ele* Corresponding author. Tel. : +86-10-58805399; Fax : +86-10-58807720; E-mail: [email protected] doi:10.2108/zsj.30.699

ments (Gruber and Nagle, 2010). Cettia warblers are a group of old world warblers that includes 17 species (Kennerley and Pearson, 2010). Most Cettia warblers sing loud songs during the breeding season. To date, two Cettia warblers have been found to use different song types in different contexts. For example, male Japanese bush warblers Cettia diphone mix two song types during spontaneous singing, but there is a significant increase in the use of one song type during playback (Park and Park, 2000). Male Cetti’s warblers, Cettia cettia, use one song type during spontaneous singing and another song type when facing a territorial dispute (Luschi and Seppia, 1996). Whether it is a general rule that Cettia warblers use different song types in different contexts needs further research. As is the case for Japanese bush and Cetti’s warblers, the majority of brownish-flanked bush warblers Cettia fortipes have two distinct song types (Fig. 1). One type (the alpha song) consists of two notes within the terminal part whereas the other (the beta song) comprises three notes within the terminal part (Xia et al., 2010). Males sing these two song types both during spontaneous singing and playbackevoked singing (Xia et al., 2010), but the proportion of use of each type in these different contexts has never been studied. Song structure is quite similar between Japanese bush warblers and brownish-flanked bush warblers; both species have two song types, and song consists of an introductory whistled part and a more complex terminal part. We hypothesized that these similarities in song structure imply that the two song types have similar functions in both species. To determine if brownish-flanked bush warblers uses

700

C. Xia et al. and a time resolution of 5.8 ms. We defined a note as a continuous signal in the spectrogram. For the introductory whistled part, we measured duration (D_w), number of notes (N_w), and frequency (F_w); for the terminal part, we measured duration (D_t), number of notes (N_t), minimum frequency (Fmin_t), and maximum frequency (Fmax_t); for the whole song, we measured duration (D_s) and relative sound pressure (P_s). When measuring maximum frequency, we ignored harmonic components, as they were quite weak. We used root mean square values to measure the relative sound pressure of the whole song, which set the loudest sound at 0 dB; these values can only be compared within one recording. We also calculated frequency bandwidth (maximum frequency minus minimum frequency) for the terminal part (Fr_t), note rate (number of notes/ duration) for both the introductory whistled part and the terminal part (Nr_w, Nr_t), and the interval between the introductory whistled part and the terminal part (duration of the whole song minus the durations of the introductory whistled part and the terminal part) (I_s). In total, we obtained values for 13 variables for each song.

Fig. 1.

Spectrograms of alpha (A) and beta (B) song types.

different song types in different contexts, we compared the acoustic features of the two distinct song types and tested whether the proportion of each song type differed between spontaneous singing and playback-evoked singing. We also used playback experiments to compare the induced behaviors of these two song types to test whether they reflect different levels of aggressive motivation. MATERIALS AND METHODS Study area and target species This study was conducted on a population of brownish-flanked bush warblers at Dongzhai National Nature Reserve, Henan Province, southern China (31.9°N, 114.3°E) in 2012. The reserve consists of mature montane forests at 100–466 m elevation, interspersed with small tea estates, villages, and several pedestrian trails. This warbler inhabits scrubland habitats dominated by tea plants Camellia sinensis, and defends territories from March to July. Most males were trapped by mist net and banded with a unique combination of color-rings at the beginning of the breeding season. For playback stimulus, we used songs that were recorded during the 2011 breeding season from a population at Guangshui (31°50′N, 113°55′E), approximately 35 km from Dongzhai National Nature Reserve, where the landscape was dominated by similar habitats and topographical characteristics. Comparing acoustic features of alpha and beta song types Songs were recorded at a sampling rate of 44.1 kHz and a sample size of 16 bits with a TASCAM HD-P2 portable digital recorder (Tascam Co., Japan) and a Sennheiser MKH416 P48 external directional microphone (Sennheiser Co., Germany). Songs were recorded during the dawn chorus from 30 males at Dongzhai and 20 males at Guangshui, and both song types were included in their songs. Because acoustic features are strictly stereotyped within individuals throughout the entire breeding season (Xia et al., 2010; Xia et al., 2012), we randomly selected one recording for each male, and measured one song, which was also randomly selected, of each song type within this recording. We used AvisoftSASLab Pro 4.3 (Avisoft Bioacoustics, Germany; Specht 2005) to analyze the recordings. First, we used a high-pass filter at 1 kHz, which is lower than the minimum frequency of the songs, to remove background noise (mostly caused by wind). We resampled the recordings at 22.05 kHz and created spectrograms with fast Fourier transform lengths of 256 points, a hamming window with a frame size of 100% and an overlap of 50%, a frequency resolution of 86 Hz,

Generating playback songs To avoid dear-enemy effects (Fisher, 1954), we used playback songs that were recorded from Guangshui. We used songs from 20 males as stimulus songs. To imitate the natural sound pressure level and song rate of this species, we used a CEL-240 sound level meter (Casella Co., UK) to adjust the amplitude to 80 dB at a location 1 m above the speaker, and used Goldwave 5.25 (GoldWave Inc., Canada) to adjust the song rate to one song per 10 s for stimulus songs. This type of stimulus is termed a normal song stimulus in this study. The proportion of the alpha song type in a normal song stimulus ranged from 33.3% to 66.6%, with a mean of 50%. For alpha (beta) song stimulus, we replaced the beta (alpha) song with the alpha (beta) song in the normal song stimulus. In total, we used 60 bouts of stimulus songs: 20 normal songs, 20 alpha songs, and 20 beta songs. Playback was conducted using a Teclast X18 Mp3 player (Teclast Co., China) connected to a Senway loudspeaker (Shenzhen Senway Amplifier Co., China). Comparing spontaneous singing and playback-evoked singing We compared spontaneous singing and playback-evoked singing among seven target males. First, we recorded 3 min of spontaneous singing at dawn. We then broadcast a normal song stimulus for 3 min within the territory of the target male. To avoid pseudoreplication, each of seven males was presented with one stimulus song that was randomly selected from the 20 normal song stimuli. The target male responded to the playback by moving toward the loudspeaker, singing and calling. We recorded songs from target males during the 3 min playback and for 3 min immediately after playback. Comparing behaviors induced by alpha and beta song types Following Gruber and Nagle (2010) and Nelson and Poesel (2012), we used playback experiments to test whether alpha and beta song types reflect different levels of aggressive motivation. As this species of warbler is cryptic, it is generally difficult to observe individuals in their bush habitats. Therefore, two-speaker playback was used here because it can force a response choice between different stimulus songs during playback and can enlarge the response difference (Stoddard et al., 1990). We compared the birds’ responses to alpha vs. beta songs, alpha vs. normal songs, and beta vs. normal songs. Three speakers were placed at the center of a target male’s territory. Two experimental speakers were placed 15 m apart, without exceeding the length of a typical territorial boundary or the audible range of songs. A lure speaker was placed in the center between the two experimental speakers. Playback was divided into two steps. During step 1, we played a normal song for 30 s using the lure speaker to attract the target male to the center of the two experimental speakers. Then we played 3 min of alpha vs. beta

Song Types in Different Contexts

701

Table 1. Values of measured variables for alpha and beta song types. songs (or alpha vs. normal, or beta vs. normal) using the two Dongzhai population Guangshui population experimental speakers. After 95 s Variables* Alpha song type Beta song type of silence, step 2 was carried out: Alpha song type Beta song type P P we repeated the 30 s of normal (Mean ± SD) (Mean ± SD) (Mean ± SD) (Mean ± SD) song stimulus and then played D_w (s) 1.451 ± 0.150 1.462 ± 0.126 0.635 1.379 ± 0.133 1.398 ± 0.115 0.509 3 min of songs with the experiN_w 2.467 ± 0.629 2.133 ± 0.507 0.004 2.050 ± 0.394 1.900 ± 0.308 0.083 mental speakers, interchanging Nr_w 1.698 ± 0.392 1.455 ± 0.306 < 0.001 1.499 ± 0.278 1.373 ± 0.262 0.032 the songs between the speakers. F_w (kHz) 2.836 ± 0.200 2.127 ± 0.179 < 0.001 2.907 ± 0.184 2.046 ± 0.233 < 0.001 These two steps insured that our D_t (s) 0.464 ± 0.046 0.679 ± 0.071 < 0.001 0.487 ± 0.063 0.691 ± 0.049 < 0.001 results were mainly the effect of N_t 2±0 3±0 2±0 3±0 the contents of the broadcasts, Nr_t (s) 4.364 ± 0.440 4.466 ± 0.522 0.274 4.179 ± 0.454 4.372 ± 0.308 0.043 not the positions of the speakers Fmin_t (kHz) 1.939 ± 0.136 2.102 ± 0.235 0.001 1.661 ± 0.122 1.976 ± 0.199 < 0.001 (Stoddard et al., 1990). We Fmax_t (kHz) 5.405 ± 0.362 5.927 ± 0.397 < 0.001 5.495 ± 0.376 5.993 ± 0.272 < 0.001 recorded males that appeared Fr_t (kHz) 3.466 ± 0.419 3.825 ± 0.471 0.001 3.833 ± 0.446 4.018 ± 0.331 0.147 within 3 m of either experimental D_s (s) 1.989 ± 0.140 2.213 ± 0.143 < 0.001 1.932 ± 0.153 2.160 ± 0.098 < 0.001 speaker during both experimental I_s (s) 0.075 ± 0.023 0.072 ± 0.021 0.624 0.069 ± 0.019 0.072 ± 0.020 0.358 steps. If a target male did not 0.104 −22.64 ± 6.197 −21.965 ± 5.507 0.326 P_s (dB)+ −25.101 ± 4.572 −25.979 ± 4.121 appear within 3 m of any experi* See main text for descriptions of variables. mental speaker during the 3 min, + Root mean square values were used in Avisoft-SASLab Pro 4.3, which set the loudest sound at 0 dB; we considered that playback these values can only be compared within one recording. experiment to have failed and the target male was not used in this study. In total, we conducted playback experiments using alpha vs. beta songs on 10 males, alpha vs. normal on 12 males, and beta vs. normal on 10 males. Statistical analyses To compare acoustic features between alpha and beta song types, we used Wilcoxon signed ranks tests with two related samples to compare discrete variables (number of notes), and paired-sample T-tests to compare continuous variables. Because we conducted multiple comparisons for 12 variables, we used a Bonferroni correction to an adjusted α = 0.004. To determine whether the proportion of the alpha song type differed between spontaneous singing and playback-evoked singing, we used repeated measures ANOVA with singing stage (spontaneous singing, during playback, after playback) as a factor. We also used Fisher exact tests to determine whether one song type was used more frequently during or after playback compared to spontaneous singing for each male. Similar to Stoddard et al. (1990), time spent near a speaker (within 3 m) was summed for the two steps and paired-sample T-tests were used to assess differences in the response between song types. All data are given as means ± standard deviation. We ensured that the data were sufficiently normal (Kolmogorov-Smirnov test, P > 0.05) for parametric tests. Data were analyzed using R 2.15.0 (R Development Core Team, 2012). Ethics statement This research protocol was approved by the Animal Management Committee at the College of Life Sciences, Beijing Normal University, the National Bird-banding Center of China (NBCC), under license number 201000042 and the Dongzhai National Nature Reserve under license number 2011002.

RESULTS Acoustic features of alpha and beta song types In addition to the note number of the terminal part, which we used to define the song types, there were significant differences in some variables between alpha and beta song types at the significance level α = 0.004 (Table 1). In both the Dongzhai and Guangshui populations, the alpha song type had a higher frequency in the introductory whistled part (F_w), a lower frequency (both minimum and maximum frequency) in the terminal part (Fmin_t, Fmax_t), and shorter

Fig. 2. Proportion of alpha song type during spontaneous singing and playback-evoked singing.

durations in both the terminal part (D_t) and the whole song (D_s). The alpha song type had a faster note rate in the introductory whistled part (Nr_w) and a narrower frequency bandwidth in the terminal part (Fr_t) in the Dongzhai population, but it was not significantly different from the beta song type in the Guangshui population. There were no significant differences between the alpha and beta song types in the duration of the introductory whistled part (D_w), note number in the introductory whistled part (N_w), note rate in the terminal part (Nr_t), interval between the introductory whistled part and the terminal part (I_s), or relative sound pressure (P_s) in either population. Spontaneous singing and playback-evoked singing All seven target birds sang during the 3 min playback, and six of them sang after playback stopped. Similar to their spontaneous songs, all of the target birds sang a mixture of two song types during playback and after playback (Fig. 2, Table 2). The proportion of the alpha song type was not significantly affected by the singing stage (repeated measures ANOVA: F2,10 = 0.180, P = 0.838). Neither song type was used more frequently during or after playback compared to spontaneous

702

C. Xia et al.

Table 2. Number of songs sung during spontaneous singing and playback-evoked singing. Individual 1 2 3 4 5 6 7

Spontaneous singing During playback After playback alpha

beta

alpha

beta

alpha

beta

8 6 8 6 5 2 8

9 6 9 3 5 5 6

3 7 4 8 4 1 6

4 6 5 6 3 2 4

8 12 3 0 1 0 5

7 11 3 0 0 2 4

Fig. 3. Time (mean ± SE, seconds) spent within 3 m of the speakers during two-speaker playback.

singing for each male (Fisher exact test, P > 0.05). Behaviors induced by different song types Target males approached the lure speaker when it first broadcast and then approached the experimental speakers when alpha and beta (or alpha and normal, or beta and normal) song types were played back. Males flew back and forth and showed no significant difference in response between the song types (paired-sample T-tests: alpha vs. beta, t9 = –0.434, P = 0.675; alpha vs. normal, t11 = 0.259, P = 0.801; beta vs. normal, t9 = –0.380, P = 0.713) (Fig. 3). DISCUSSION We compared acoustic features between two song types in brownish-flanked bush warblers. The number of notes within the terminal part of alpha songs was obviously less than in beta songs, which was used to define the song types (Xia et al., 2010). In addition, compared to the beta song type, the alpha song was shorter, had a higher frequency in the introductory whistled part, and had a lower frequency in the terminal part in both the Dongzhai and Guangshui populations (Table 1). Alpha song type also had a faster rate in the introductory whistled part and narrower frequency bandwidth in the terminal part than beta song type in Dongzhai population (Table 1). In Guangshui population, there was similar trend in these two variables, but the differences were not significant (Table 1). The songs of Japanese bush warblers and Cetti’s warblers have also been analyzed in detail. In the former, a song consists of an introductory whistled part and a more complex terminal part (Park and Park, 2000). In the latter, a song consists of a beginning note, instead of introductory whistled part, and a series of terminal notes (Luschi and

Seppia, 1996). Similar to brownish-flanked bush warblers, there are generally two song types in these other two warblers. In the former, the number of notes in the introductory whistled part is the most distinctive feature, and is used to define the song type (Yoon et al., 1995); in the latter, the number of notes in the terminal part is used to define the song types (Luschi and Seppia, 1996). In brownish-flanked bush warblers, the difference in the number of notes in the terminal part was more apparent. In general, bird songs are produced at high amplitudes and can extend well beyond territorial boundaries (Naguib and Wiley, 2001). However, the repertoires of some birds contain lower-amplitude song types. These songs have a restricted transmission range, and are not easily subject to eavesdropping by predators or other individuals (Dabelsteen et al., 1998; Morton, 2000). In brownish-flanked bush warblers, the amplitudes of the alpha and beta song types were nearly the same. Therefore, our results do not support the hypothesis that the different song types are used to communicate at different ranges in this species. Brunner and Pasinelli (2010) inferred that different song types, which typically consist of songs that differ in quality and sometimes in output, are delivered in different temporal and social contexts. Lampe and Espmark (1987) suggested that shorter songs may be used during male–male interactions. Consistent with these ideas, Cetti’s warblers use a long song type during spontaneous singing and a short song type when involved in territorial disputes (Luschi and Seppia, 1996). Japanese bush warblers show a similar pattern. During spontaneous singing, the two song types are mixed, but there is a significant increase in the use of the short song type during playback and a quick return to a mixture of the two song types after playback (Park and Park, 2000). In our study, we examined 205 songs from seven brownish-flanked bush warblers, and no one song type was used more frequently during or after playback compared to spontaneous singing (Fig. 2, Table 2). Although the sample size in our study was relatively low compared to the study by Park and Park (2000) that was based on 992 songs from 13 Japanese bush warblers, the failure to detect an increase in the use of one song type during playback in brownish-flanked bush warblers was not merely due to statistical power. During playback, four of 13 Japanese bush warblers sang only the short song type, and most of the other individuals increased the proportion of this song type (Park and Park, 2000). In contrast, none of the males increased the use of one song type during playback in our study. In some bird species, different song types reflect different levels of aggressive motivation and can induce different responses in the receiver. In at least 12 species that have been subjected to playback tests, males approached a sound box that was producing long songs more closely than one that was emitting short songs, and no species showed the opposite pattern (Nelson and Poesel, 2012). In brownish-flanked bush warblers, there were significant differences in some variables between alpha and beta song types, but the warblers treated the different song types equally and did not respond differently to normal song playback or playback of a single song type (Fig. 3). Therefore, our results do not support the hypothesis that alpha and beta song types reflect different levels of aggressive motivation. Similar results were

Song Types in Different Contexts

found in Cetti’s warblers. In Cetti’s warblers, two song types were used selectively in spontaneous and playback-evoked singing, but the reactions to the different song types were not significantly different (Luschi and Seppia, 1996). Based on bird song studies in oropendolas (genera Psarocolius, Gymnostinops, and Ocyalus), Price and Lanyon (2002) pointed out that avian vocalization is a plastic characteristic that can rapidly diverge within closely related taxa. Our study offers another example of this pattern. Unlike two closely related species, which both use different song types selectively during spontaneous singing and playbackevoked singing, we did not find evidence that song types were used selectively in brownish-flanked bush warblers, although there were significant differences in both frequency and temporal variables between the two song types. Given that the evidence did not support the hypothesis that song types are used selectively in different contexts in brownishflanked bush warblers, what alternate hypotheses might account for the presence of multiple song types in this species? The anti-exhaustion hypothesis can account for versatility in bird songs. According to this hypothesis, song diversity is beneficial because singing different song types, which involve different patterns of muscle contraction, leads to less exhaustion in the singer (Lambrechts and Dhondt, 1988). The number of notes in the whistle and syllable portions of short songs in Japanese bush warblers is 4.46 ± 0.61 and 2.36 ± 0.89 (Park and Park, 2000), respectively, and the short song of Cetti’s warblers also consists of more than 10 notes. Compared to these two Cettia specis, both song types in brownish-flanked bush warblers are relative simply, especially the short song (Table 1). A long series of the same song can lead to fatigue in the muscles that are involved in its production, which would cause the song to become more difficult to generate (Catchpole and Slater, 2008), especially for a short and dull song. Brownish-flanked bush warblers sing clear, high-pitched songs (del Hoyo et al., 2006; Kennerley and Pearson, 2010). Perhaps by mixing the two song types individuals can avoid muscle exhaustion and benefit from a longer period of singing in this species. Bird song has long been treated as a sexual signal, and song diversity is hypothetically raised and maintained by female mate choice (Catchpole and Slater, 2008). Many species use ‘dual-function’ signals to interact with males and to attract females (Searcy and Andersson, 1986). It may be that the two song types have equal functions and are maintained through inter-sexual selection in brownish-flanked bush warblers. We will study female responses to the playback of normal songs and individual song types in the future. ACKNOWLEDGMENTS We thank Mr. Chentao Wei and Ms Qiao Wu for their assistance in the field, Prof Yong Wang for his helping of data analysis. Many thanks to two anonymous reviewers for their constructive suggestions and comments on the manuscript. This work was supported by the National Natural Science Foundation of China (No.31172098), the Foundation for Excellent Doctoral Dissertation of Beijing Normal University, the scholarship award for excellent doctoral student granted by Ministry of Education, China.

REFERENCES Beecher MD, Brenowitz EA (2005) Functional aspects of song

703

learning in songbirds. Trends Ecol Evol 20: 143–149 Berglund A, Bisazza A, Pilastro A (1996) Armaments and ornaments: an evolutionary explanation of traits of dual utility. Biol J Linn Soc 58: 385–399 Brunner P, Pasinelli G (2010) Variation in singing style use in the reed bunting Emberiza schoeniclus: influencing factors and possible functions. J Avian Biol 41: 388–397 Byers BE, Belinsky KL, Bentley RA (2010) Independent cultural evolution of two song traditions in the chestnut-sided warbler. Am Nat 176: 476–489 Catchpole CK, Slater PJB (2008) Bird Song: Biological Themes and Variations. 2nd ed, Cambridge University Press, London Dabelsteen T, McGregor PK, Lampe H, Langmore N, Holland J (1998) Quiet song in song birds: an overlooked phenomenon. Bioacoustics 9: 89–105 del Hoyo A, Elliott A, Sargatal J (2006) Handbook of the Birds of the World, Vol. 11: Old World Flycatchers to Old World Warblers. Lynx edicions, Barcelona Fisher RA (1954) Evolution and bird sociality. In: “Evolution As a Process” Ed by J Huxley, AC Hardy, EB Ford, Allen & Unwin, London, pp 71–83 Gruber T, Nagle L (2010) Territorial reactions of male yellowhammers toward a specific song structure. J Ornithol 151: 645–654 Kennerley P, Pearson D (2010) Reed and Bush Warblers. Christopher Helm, London Lambrechts M, Dhondt AA (1988) The anti-exhaustion hypothesis: a new hypothesis to explain song performance and song switching in the great tit. Anim Behav 36: 327–334 Lampe HM, Espmark YO (1987) Singing activity and song pattern of the redwing Turdus iliacus during the breeding season. Ornis Scand 18: 179–85 Luschi P, Seppia CD (1996) Song-type function during territorial encounters in male Cetti’s warblers Cettia cetti. Ibis 138: 479– 484 Morton ES (2000) An evolutionary view of the origins and functions of avian vocal communication. Jpn J Ornithol 49: 69–78 Naguib M, Wiley RH (2001) Estimating the distance to a source of sound: mechanisms and adaptations for long-range communication. Anim Behav 62: 825–837 Nelson DA, Poesel A (2012) Responses to variation in song length by male white-crowned sparrows. Ethology 118: 24–32 Park SR, Park DS (2000) Song type for intrasexual interaction in the bush warbler. Auk 117: 228–232 Price JJ, Lanyon SM (2002) Reconstructing the evolution of complex bird song in the oropendolas. Evolution 56: 1514–1529 Searcy WA (1992) Song repertoire and mate choice in birds. Am Zool 32: 71–80 Searcy WA, Andersson M (1986) Sexual selection and the evolution of song. A Rev Ecol Syst 17: 507–533 Spector DA (1992) Wood-warbler song systems: a review of Paruline singing behaviors. Curr Ornithol 9: 199–239 Stoddard PK, Beecher MD, Horning CL, Willis MS (1990) Strong neighbor-stranger discrimination in song sparrows. Condor 92: 1051–1056 Xia C, Xiao H, Zhang Y (2010) Individual variation in brownishflanked bush-warbler songs. Condor 112: 591–595 Xia C, Lin X, Liu W, Lloyd H, Zhang Y (2012) Acoustic identification of individuals within large avian populations: a case study of the brownish-flanked bush warbler, south-central China. PLoS ONE 7: e42528 Yoon MB, Park DS, Park SR (1995) The characteristics of the alpha and beta song modes and comparison among alpha, beta mode, and inland geographic variation groups’ songs of the bush warbler (Cettia diphone). Korean J Beh Biol 4: 31–42 (Received November 5, 2012 / Accepted May 8, 2013)