Vocal variability of chaffinch song ( Fringilla coelebs L.) Astakhov a O.A. (dissertation) 2006. A general characteristic • In early works geographical variability of song at many species of birds has been marked as a phenomenon of differentiation of song structures or song patterns (as the image on sonograms ) in different local populations (Marler, 1952 ; Marler and Tamura, 1962 ; Sick, 1939 ; Poulsen, 1951 ) . • Local variants of vocalizations of birds can be considered as analogy of human speech and to name as dialects . The word "dialect" has taken place from linguists and means a local speech variant, characteristic for region . Borders of a dialect define lexical ( word structure), morphological (an accent, structure ) and phonetic (pronunciation) characteristic s of the dictionary (Kurath, 1972 ) . • Song dialect of birds is a variant of the traditional form of song, shared by members of a local population of birds and forming borders of a dialect (separating from other variant of song pattern), within the limits of which there is a traditional training (song learning) of characteristic song components of the given population . But thus big song repertoir e s of some species of sparrow birds interfere (have problem) with objective definition of dialect borders (Thielcke, 1969 ; Kreutzer, 1974 ; Kroodsma, 1974 ; Baker, 1975 ; Baptista, 1975 ; Lemon, 1975 ; Payne, 1981 ; Mundinger, 1980 , 1982 ) . • Chaffinch ( Fringilla coelebs L . ) is classical object of study of becoming of vocal repertoir e (Thorpe, 1958 ; Marler, 1956 ; Nottebohm, 1967 ) and of geographical variability of species specific song in a population ( Promptov , 1930 ; Sick, 1939 ; Thielcke, 1961 ; Simkin, 1983 ; Slater e t al . , 1984 ) . One of problems of our researches has consisted in revealing macrogeographical distinctions of chaffinch song ( Fringilla coelebs L . ) in different local populations , which are removed approximately on 1000 km from each other . In northwest ( Curonian spit, the Kaliningrad region ) and in central parts ( Zvenigorod , Moscow, Michurinsk) of the European Russia (fig . 1 ) have been made tape records of singing males (N= 218 )during the spring - and - summer period of 2005 - 2006 . Fig. 1 . Map of Russia (points mark places of sampling of songs) Methods of research • Regional variability of song can be determined in qualitative aspect (the form of a syllable, syntax) and quantitative (time - and - frequency) parameters. The concept of a dialect should be closely (cautiously) and is carefully argued in an establishment of regional variability call and song patterns. • At the analysis of song sonogramms basically applied two qualitative methods (Mundinger, 1982): 1. Revealing of phonetic distinctions (frequency of a sound, its form on sonogram ) or of ways of a pronunciation of syllables of the phrases making songs; 2. Revealing of lexical distinctions (changes of phrases of songs as a whole). • A lso have been spent the quantitative analysis of dialect forms of songs of one type and comparison of their basic time - and - frequency parameters (duration of songs - sec , number of elements in type of song, duration of syllables in a trill, the maximal, minimal and average frequency of song - KHz , intervals between songs - sec ). • Sonograms of types of songs have been analyzed with the help of computer program Avisoft SASLab Light . In total have been analysed about five thousand songs. Types of songs have been marke d by L atin letters . At record, songs of one type have been met in different points of territory (considered that belong to repertoires of different males), therefore alongside with the letter have been designated by numbers in ascending order (for example, А 1, А 2, А 3, etc.). Results and discussion • In populations of the central part of the European Russia (N=65 of males ) we allocate ( distinguish) 15 types of song of chaffinch ( Fringilla coelebs L.) , which have been completely similar in structure or have been in part modified in the syllabic form (on sonograms ) in comparison with songs of same types in samples of a northwest part of the European Russia (N=153 of males ). • For seven types of songs from twenty two (tests on Curonian spit ) analogies (similarities) ha ve not come to light – probably, owing to their rarity . Many song types in the central part of the European Russia have been considered as combined phrases ( parts) of syllabic patterns known to us, but frequently with the changed figure (forms) on sonogram . • In result , have been found 12 dialect forms of songs of one type , which appeared similar in base structure of elements, of phrases, but in different regions of Russia frequently had distinguished manners, ways of their performance at singing (phonetic aspect). • The average size of chaffinch repertoir e ( Fringilla coelebs L . ) in populations of the central part of the European Russia from statistical calculations has been submitted 1,93 ± 0,22 type of songs (max – 4 types of song, min – 1 song type) at a capture (volume) on the average 24,7 ± 11,3 songs from one of male . • The basic time - and - frequency parameters of local variants of types of songs of chaffinch (Fringilla coelebs L.) at the quantitative analysis had insignificant macrogeographical variability that can speak about their importance in transfer and maintenance species specific song traditions . The most stable parameters of local variants of songs of one type appeared minimal and average (median) frequency , values of length and intervals of elements of a trill . The table. Character of vocal variability of chaffinch songs (Fringilla coelebs L .) in populations of northwest and the center of the European part of Russia (N = 218) * Vocal variability Song type ( token ) A B C D E F G I H J K L M N O P Q R S T V W U Total Dialects 1 + + + + + 5 Subdialects 2 + + + + + + + 7 European part of Russia Sample size of song type (n) in population Northwest 18 5 38 11 7 16 10 22 6 13 3 1 20 8 18 1 1 24 17 14 2 8 183 Center 8 8 24 3 5 1 11 12 2 5 2 2 3 3 93 Note: 1 dialects – different phonetic norms of song types; 2 subdialects – small phonetic distinctions of song types; * - sample size of chaffinch males (Fringilla coelebs L .) at record; the most widespread (frequently met) in populations types of songs (with sample n> 15) are allocated (distinguish) by a font. Probably, at song types are found the dialects (vocal variability), because of their high frequency of occurrence in populations at record. Perhaps, other types of song also have dialects in this territory, but have not come in our sample. Thus, on twelve types of chaffinch songs (Fringilla coelebs L .) the dialect forms (B C D F G I J M N V S U) have been found. It is interesting to note, that dialect forms of songs or separate phrases (in the "combined" songs) of one type can meet within the limits of one local population. Proceeding from samples, made by us in different regions of the European Russia ( northwest and the central part ), it is possible to draw a conclusion, that on some song types of chaffinch (Fringilla coelebs L .) exist subdialects (insignificant phonetic distinctions of syllables) - for example, on song types B, M, D, G, V, S, U, and dialects (different phonetic norms of song types) - C, C * and C # , F, I, J, N, and are also stable song types (in samples of different regions unchanged) – song type A. As a bright example of vocal variability of chaffinch (Fringilla coelebs L . ) it is possible to result (to show) some local variants of songs , which we have attributed to one type (fig . 2 , 3 , 4 ) . Samples of types of song , which will be resulted (be shown) , were not single in a population and have been recorded in repertoir e s more, than 1 - 2 individuals . The song t ype V 1 2 Fig. 2 . Dialect forms of song type V: 1 – the song type V 1 (record on Curonian spit , the Kaliningrad region); 2 – the song type V (record in Zvenigorod , the Moscow region) Fig . 3 . Dialect forms of song type S: 1 – the song type S5 (record on Curonian spit , the Kaliningrad region); 2 – the song type S (record in Zvenigorod , the Moscow region). The song t ype S The song t ype I Fig. 4 . Dialect forms of song type I: 1 - the song type I 6 (record on Curonian spit , the Kaliningrad region); 2 - the song type I 1 2 (record in Moscow); 3 - the song type I 6 (record in Zvenigorod , the Moscow region). 1 2 1 2 3 Despite of stability in a manner of performance of song type B, in different local populations , on sonogramms it is possible to note some variability of the final stroke (in the latter case (fig. 5.4) the number of the first elements of a stroke is increased, and the phrase is heard as a trill). The s ong t ype B Fig. 5 . Local variants of song type B (from different areas of Russia): 1 – the song type В5 (record on Curonian spit , the Kaliningrad region); 2 – the song type B (record in Zvenigorod , the Moscow region); 3 – the song type В8 and 4 – the song type В7 (record in Michurinsk, the Tambov region). 1 2 3 4 At the quantitative analysis of local variants of song type B (with the help of computer program Avisoft - SASLab Light ) come to light some distinctions (differences) in values of the basic time - and - frequency parameters, but in the majority they are similar (tab. 1). Table 1. The basic time - and - frequency parameters of local variants of song type B The note: average value and a standard deviation of parameters of song types from statistical calculations for all songs of one type which were reproduced chaffinch males in the given points of record are specified; the strongest differences counted a difference of parameters >0,5 KHz in frequency and >0,02 sec in length (are allocated ( distinguished) by a font); * - the trill of song type B will consist of two phrases (fig. 5). The length of songs, number of their syllables (elements) , the maximal frequency, intervals between songs and length of elements of the second phrase of a trill have the values varying in wide enough range. The minimal frequency and length of syllables of the first phrase of a trill of song type B are appeared completely similar as in northwest ( Curonian spit ), and in the center of the European Russia (Zvenigorod , Michurinsk). Type of song (token ) The name of record place The length of song , sec Min frequency , К Hz Max frequency , К Hz Median (average) frequency, К Hz Number of syllables in song type Length of syllables (trill) , sec* Intervals between songs , sec 1 phrase 2 phrase В5 Curonian spit 2,47 ± 0,104 1,57 ± 0,14 7,19 ± 0,32 3,77 ± 0,14 24,5 ± 0,93 0,043 ± 0,003 0,14 ± 0,01 5,32 ± 1,9 В Zvenigorod ( Moscow region ) 2,48 ± 0,36 1,68 ± 0,16 7,92 ± 0,31 3,83 ± 0,36 22,25 ± 2,99 0,04 ± 0,003 0,17 ± 0,008 В8 Michurinsk ( Tambov regio n) 2,34 ± 0,15 1,62 ± 0,09 8,044 ± 0,53 4,12 ± 0,29 22,3 ± 2,6 0,034 ± 0,006 0,127 ± 0,013 4,3 ± 0,59 В7 Michurinsk ( Tambov regio n) 2,99 ± 0,11 1,67 ± 0,08 7,63 ± 0,08 3,79 ± 0,18 27,4 ± 1,17 0,033 ± 0,003 0,15 ± 0,007 7,06 ± 0,19 In samples types of songs have been found , which are difficult for differentiating among themselves because of their similarity (fig. 6). Fig. 6 . D ifficultly differing chaffinch song types (record on Curonian spit , the Kaliningrad region): 1 – the song type D3; 2 – the song type F4; 3 - the song type G3. We have attributed the given samples of songs to different song types (D, F, G), but nevertheless they are similar among themselves in base structure of elements (especially in the started singing and trill) , which nevertheless are performance at singing by different manners, ways that makes their distinct or different . Probably, it – the initial forms of the further modification , being in themselves insignificantly transformed forms for precise revealing of type to which are attributed. 1 2 3 Many types of chaffinch songs (Fringilla coelebs L .) in samples of the central part of the European Russia were considered by us as combined phrases (parts) of already known types of the songs which have been recoded on Curonian spit ( the Kaliningad region), but their elements have been modified on sonograms . It is possible to result some examples. Fig. 7 . Dialect forms of phrases from song types B and F: 1 – the song type B (record on Curonian spit , the Kaliningrad region); 2 – the combined song type ВF 5 (record in Michurinsk, the Tambov region); 3 – the song type F (record on Curonian spit , the Kaliningrad region) The song type ВF 5 (fig. 7.2) is recoded in the central part of the European Russia (Michurinsk) and will consist of the modified combined phrases: the trill is similar with elements of started singing of song type F (fig. 7.3); the final stroke also is similar to song type F which has been recoded on Curonian spit ; and first two phrases of song type ВF5 - from song type B (fig. 7.1). Modification of forms of elements of one type can speak about their dialects ( special to a manner of a pronunciation) in a local population . 1 2 3 Table 2. The quantitative analysis of dialect phrases of song types B and F Type of song ( place of record ) Parameters of phrases В ( Curonian spit ) В F 5 ( Michurinsk) F ( Curonian spit ) 1 phrase ( started singing ) 2 phrase ( trill ) В - similar elements ( started singing ) F - similar elements ( trill ) 1 phrase ( started singing ) 1 phrase 2 phrase The length of phrase , sec 1,13 ± 0,09 0,32 ± 0,05 0,62 ± 0,13 0,24 ± 0,02 1,17 ± 0,25 1,27 ± 0,11 Max frequency , К Hz 6,57 ± 0,46 4,626 ± 0,32 7,3 ± 0,26 5,9 ± 0,52 7,44 ± 0,22 6,25 ± 0,2 Min frequency , К Hz 2,017 ± 0,13 2,067 ± 0,1 3,1 ± 0,3 2,4 ± 0,12 1,9 ± 0,17 1,722 ± 0,14 Median (average) frequency , К Hz 4,306 ± 0,24 3,17 ± 0,09 5,44 ± 0,2 3,8 ± 0,17 4,31 ± 0,172 4,2 ± 0,09 Number of syllables in phrase type 11,43 ± 0,79 5,29 ± 0,76 5,2 ± 0,8 5 ± 0 6,4 ± 1,34 7,6 ± 0,7 The length of syllables , sec 0,058 ± 0,01 0,037 ± 0,005 0,08 ± 0,014 0,04 ± 0,008 0,15 ± 0,01 0,12 ± 0,028 Intervals between syllables , sec 0,048 ± 0,01 0,037 ± 0,004 0,05 ± 0,01 0,02 ± 0,005 0,05 ± 0,008 0,06 ± 0,012 The note: average value and a standard deviation of parameters of dialect phrases from statistical calculations for all songs of one type which were reproduced chaffinch males in the given points of record are specified; the strongest differences of phrases of one type counted a difference of parameters >0,5 KHz in frequency and >0,02 sec in length (are allocated ( distinguished) by a font). Proceeding from the analysis (tab. 2) carried out by us, it is possible to draw a conclusion, that the length of dialect phrases and their maximal frequency basically are various , and the minimal frequency and intervals between syllables appeared are similar in the majority. Values of others frequency - time parameters can to vary widely , but within the limits of the certain range. It is interesting to note, that at the greater distinction of the form of syllables (elements) of dialect phrases on sonograms ( up to occurrence of new syllabic patterns) are observed the big distinctions of their basic frequency - time parameters (for example, the first phrases of the started singing of song types B and BF). Fig. 8. Dialect forms of phrases of song types Т and S: 1 – the song type Т (record on Curonian spit , the Kaliningrad region); 2 – the combined song type ТS (record in Michurinsk, the Tambov region); 3 - the song type S (record on Curonian spit , the Kaliningrad region) + Thus, in samples of different areas of Russia there can be the song types combined from phrases of song patterns of other populations , but with the modified elements (syllables) – it is possible to tell, their dialect forms. 1 2 3 Conclusions • Many characteristics of singing of birds are precisely mentioned by social training traditions , receiving their vocal patterns by species specific imitation . Therefore the tradition, custom should be taken into account at the analysis of geographical variability of vocal behaviour . • In local populations are formed certain song cultures which are capable to change du ring time and can to make dialect forms on all area of distribution of a species . Steady song dialects during time – the phenomenon of conservatism of the vocal traditions, transmitted to the subsequent generations by means of vocal imitation. • For many species the patterns of variability as purchases (changes) of vocal traditions are a product (result) of cultural evolution (Mundinger, 1980). Cultural evolution is a parameter for species with a historical variety of patterns of microgeographical variability, and also for species with differentiated syllables . But it is not a parameter in morphological variability (Lemon, 1975; Slater and Ince, 1979). • Vocal norms (installation s ), likely, can be judged by quantity of individuals in a population, which adher e to those or others song cultures (ways of singing) of different song types. If the majority of males sing any song type such certain style ( special manner of performance), then it is will be vocal norm (installation) i n the given song type (or in type of a phrase) in the given population . But it is difficult to define vocal norms of a population at repertoir e s of great volume (when song types much) (Kroodsma, 1974). • For chaffinch (Fringilla coelebs L .) typically selective training (song learning) of species specific song in sensitive period (Marler, 1956; Thorpe, 1958; Nottebohm, 19 6 7). Therefore young birds do not learn songs of other species , and are guided by a time - and - frequency range of songs of their species . Probably, as a result of such rigid genetic determination of species species song (Simkin, 1983) is observed small variability of the basic time - and - frequency parameters of its different dialect forms (even at the account of a wide variation of values). • Proceeding from samples , made by us in different regions of the European Russia (northwest and the central part), it is possible to draw a conclusion, that i n some song types of chaffinch (Fringilla coelebs L .) exist subdialects (insignificant phonetic distinctions of syllables) – for example, i n song type s B , M, D, G, V, S, U, and dialects (different phonetic norms of types of song) – C , C # and C *, F, I, J, N, and there are also stable types of song (in samples of different regions are invariable ) – song type A. • By f ull dialect of chaffinch song (Fringilla coelebs L.) in the certain territory it would be more correct to ac count the presence of dialect forms on all song types known to us in a population (originally 22 types have been found), that, likely, it is almost impossible because of limitation of a sample and constant mixing of vocal traditions of birds in different populations as a result of migrations (Slater and Ince, 1979, 1980; Espmark et al . , 1989). The literature • Promptov A . N . 1930 . Geographical variability of chaffinch song in connection with the common questions of seasonal flights of birds // Zool . journ . № 10 ( 3 ) : 17 - 40 . • Simkin G . N . 1972 . About biological value of bird singing // The Bulletin of the Moscow university ю № 1 . Moscow : 34 - 43 . • Simkin G . N . 1983 . The typological organization and population phylogeny of bird’s songs // Bulletin of Moscow community investigators of nature . Section of biology . V . 88 . № 1 . Moscow : 15 - 27 . • Baker M . C . 1975 . Song dialects and genetic differences in White - crowned Sparrows ( Zonotrichia leucophrys ) // Evolution, 29 : 226 - 241 . • Baptista L . F . 1975 . Song dialects and demes in sedentary populations of the White - crowned Sparrow ( Zonotrihcia leucophrys nuttalli ) . - Univ . Calif . , Berkeley, Publ . Zool . , 105 : 1 - 52 . • Espmark Y . O . , Lampe H . M . , Bjerke T . K . 1989 . Song conformity and continuity in song dialects of redwings Turdus iliacus and some ecological correlates . - Ornis Scand . , 20 : 1 - 12 . • Ince S . A . , Slater P . J . B . , Weismann C . 1980 . Changes with time in the song of a populations Chaffinches . - Condor, 82 : 285 - 290 . • Jellis R . 1977 . Bird sounds and their meaning . Published by the British Broadcastig corporation . Cambridge, 256 p . • Kreutzer M . 1974 . Stereotypie et varianttions dans les chants de proclamation territoriale chez le Troglodyte ( Troglodytes troglodytes ) . - Rev . Comp . Anim . , 8 : 270 - 286 . • Kroodsma D . E . 1974 . Song learning, dialects, and dispersal in the Bewick ` s Wren . - Tierpsychol . , 35 : 352 - 380 . • Kroodsma D . E . , Miller E . H . , Quellet H . 1982 . Communication and behavior an interdisciplinary series . Vol . 2 : Song leaning and its consequence / D . E . Kroodsma, E . H . Miller . – London : Academic press, 347 p . • Kurath H . 1972 . Studies in Areal Linguistics . - Indiana Univ . Press, Bloomington . • Lemon L . E . 1975 . How birds develop song dialects . - Condor, 77 : 385 - 406 . • Marler P . 1952 . Variation in the song of the Chaffinch Fringilla coelebs . - Ibis, 98 : 458 - 472 . • Marler P . 1956 a . The voice of the chaffinch and its function as a language . - Ibis, 98 : 231 - 261 . • Marler P . , Tamura M . 1962 . Song dialects in three populations of White - crowned Sparrows . - Condor, 64 : 368 - 377 . • McGregor P . K . , Krebs J . R . 1982 . Song types in a population of great tits ( Parus major ) : their distribution, abundance and acquisition by individuals . - Behaviour, 79 : 126 - 152 . • Mundinger P . C . 1979 . Call learning in the Carduelinae : Ethological and systematic considerations . - Syst . Zool . , 28 : 270 - 283 . • Mundinger P . C . 1980 . Animal cultures and a general theory of cultural evolution . - Ethol . Sociobiol . , 1 : 183 - 223 . • Mundinger P . C . 1982 . Microgeographic and macrogeographic variation in acquired vocalizations of birds // Acoustic communication in birds / Eds . D . E . Kroodsma, E . H . Miller . New York . P . 147 - 208 . • Kroodsma D . E . Miller E . H . (eds) Acoustic communication in birds, vol . 2 , New York : Academic Press, P . 147 - 208 . • Nottebohm F . 1969 . The song of the Chingolo, Zonotrichia capensis , in Argentina : Description and evaluation of a system of dialects . - Condor, 71 : 299 - 315 . • Nottebohm F . 1967 . The role of sensory feedback in development of avian vocalizations // Proc . Int . Ornithol . Cong . 14 . Oxford - Е dinburg : Blackwell scient . Publs . – P . 265 - 280 . • Payne R . B . 1981 . Population structure and social behavior : models for testing ecological significance of song dialects in birds . In «Natural Selection and Social behavior : Recent Research and New Theory» (R . D . Alexander and D . W . Tinkle, eds . ), Chiron, New York : 108 - 119 . • Poulsen H. 1951. Inheritance and leaning in the song of the Chaffinch ( Fringilla coelebs ). - Behaviour, 3 : 216 - 228. • Slater P. J. B., Ince S. A. 1979. Cultural evolution in chaffinch song. - Behaviour, 71 : 146 - 166. • Slater P. J., Ince S. A., Colgan P. W. 1980. Chaffinch song types: their frequencies in the population and distribution betwe en repertoires of different individuals. - Behaviour, 75 : 207 - 218. • Slater P. J. B. 1981. Chaffinch song repertoires: observations, experiments and a discussion of their significance. - Z. Tiërpsychol., 72 : 177 - 184. • Slater P. J. B., Clement F. A., Goodfellow D. J. 1984. Local and regional variations in chaffinch song and the question of dialects. - Behaviour, 88 : 76 - 97. • Sick H. 1939. Ueber die Dialektbildung beirn Regenruf des Buchfinken. - J. Orn., 87 : 568 - 592. • Thielcke G. 1961. Stammesgeschichte und geographische Variation des Gesanges unserer Baumläufer. - Verh Ornithol. Ges Bayern, 14 : 39 - 74. • Thielcke G. 1965. Gesangsgeographische Variation des Gartenbaumläufers ( Certhia brachydactyla ) im Hinblick auf das Artbildungsproblem. - Z. Tierpsychol., 22 : 542 - 566. • Thielcke G. 1969. Geographic variation in bird vocalizations. In «Bird Vocalizations» (R. A. Hinde, ed.), Cambridge Univ. Press., London and New York: 311 - 340. • Thorpe W. H. 1958. The leaning of song patterns by birds, with especial reference to the song chaffinch Fringilla coelebs . - Ibis, 100 : 535 - 570.