Tuatara: Volume 26, Issue 1, September 1982
On the Evolution of Song Repertoires; A Discussion of the Evidence
On the Evolution of Song Repertoires; A Discussion of the Evidence
The general question of the evolution of song repertoires is discussed. The major hypotheses advanced on the issue are introduced and the evidence from the literature is briefly reviewed. It is concluded that no one hypothesis adequately explains the evolution of song repertoires. It is suggested that the guidelines provided by Gould and Lewontin (1979) could be profitably utilised in song research.
Keywords: bird song repertoires; evolution; review; behaviour.
One of the most interesting and puzzling features of bird song is the fact that in many species it is so complex. It has been shown that, broadly, song functions in territory defence (Peek, 1972; Krebs, 1977; Krebs et al. 1978), in mate attraction (Catchpole, 1980; Krebs et al. 1981) and in some species in the physiological preparation of the female for reproduction (e.g. budgerigars, Brockway, 1965; canaries, Kroodsma, 1976). It appears that in some species quite simple songs are capable of fulfilling these functions (e.g. fernbird, Jenkins, pers. comm.), yet many birds produce songs of great diversity and complexity (e.g. brown thrasher, Kroodsma and Parker, 1977; New Zealand starling, Jenkins in prep.). A large number of species attain a complex vocal output through the possession of a repertoire containing several song types e.g. chaffinch, Marler, 1952; great tit, Krebs 1976b).
Hartshorne (1956) estimated that roughly two-thirds of extant bird species possess repertoires. Research into the significance and possible evolutionary history of repertoire-singing seems important and of special relevance to New Zealand's vertebrate fauna, as it is dominated by birds. As so many of our bird species seem to have evolved in a dense forest habitat, where acoustic communication is of great importance, song repertoire studies may be of great interest. Certainly several species possess complex song repertoires. Tui song, for example, has so far defied detailed analysis and Hughes (1981) found considerable complexity in the vocalisations of the North Island kokako.
The origin of such vocal complexity is a question to which many authors have responded with hypotheses, but unfortunately few of these hypotheses are supported by convincing data. Given the importance of these ideas to present and future New Zealand bird studies, it seems appropriate here to briefly review the hypotheses that have been advanced on the issue.
1. Repertoires Enhance Recognition
Emlen (1971) suggested that factors promoting the distinctiveness of an individual's song(s) may allow others to more easily identify that individual. He argued that the possession of more than one song type (i.e. a repertoire) allows more opportunity for variability between individuals, and so may enchance individual recognition. Other authors (Krebs, 1977; Kroodsma, 1976b; Smith and Reid, 1979) have deemed this hypothesis unlikely, and pointed out that there is no evidence supporting this view. Krebs (1977) and Kroodsma (1976b) have in fact argued the opposite, that repertoires, if anything, probably make individual recognition more page 28 difficult. Their conviction is based on a study which concluded that individual recognition was facilitated, as one might expect, by features within the song which are constant within an individual but vary between individuals (Brooks and Falls, 1975; see also Emlen, 1971; Shiovitz, 1975; Miller, 1979). Their interpretations differ however; Krebs argues, “A varied repertoire is very likely to decrease the chance that any feature of the song is constant within an individual” (Krebs, 1977, p. 475). Kroodsma centres his objections around the problem of repetition and the familiarity required for the recognition of each song type. “With larger repertoires, the amount of exposure to and probably the familiarity with each song type of neighbouring males will decrease proportionately; regardless of the neural mechanisms involved, distinguishing songs of neighbours from those of strangers may then become more difficult.” (Kroodsma, 1976b, p. 97).
Treisman (1978, 1980) proposed a slightly different scheme, that repertoires aid the recognition of KIN. By allowing for greater variability, repertoires might aid the formation of dialects. He interpreted dialects as being advantageous through restricting gene flow and thereby preserving local adaptation (see also Nottebohm, 1970, 1972, for similar adaptive explanation of dialects). Treisman also argued that animals settling near kin may obtain a selective advantage “… provided both sides can recognise their relationship in the sense that they apply appropriately modified aggressive strategies” (Treisman, 1978, p. 814). He suggests that the similarity of an individual's song with the dialect of an area may be a cue that allows that individual to find an area where many of its neighbours are likely to be relatives. Treisman presents no data in support of his view and Trainer (1980) (see also Slater and Ince, 1979) suggests that the literature lends itself better to alternative explanations (e.g. Nottebohm, 1969). Treisman's idea that repertoires aid the recognition of kin has found little support.
The importance of song repertoires to individual recognition is unclear. It must be concluded that the hypothesis is not supported by available data and seems unlikely to have been important in the evolution of repertoires.
2. Repertoires Evolved in Response to Intersexual Selection
Several authors have suggested that repertoires may be the product of intersexual selection (Nottebohm, 1972: Catchpole, 1973, 1976, 1979, 1980, 1981; Howard, 1974; Kroodsma, 1976a, 1977; Krebs and Kroodsma, 1980). “Once good quality males start being more successful by using more complicated songs, then the runaway process of intersexual selection leading to extreme elaboration could be set in motion” (Catchpole, 1981, p. 31). For this system to evolve, high genetic quality in the males must have been positively correlated with repertoire size. Natural selection would then favour females mating with males possessing large repertoires, as those males would tend to produce more offspring.
To demonstrate convincingly that repertoires are a product of intersexual selection alone, a study must show the following: (Note; the word “choice” is used in evolutionary sense).
That females preferentially mate with males possessing large repertoires.
That females are “choosing” males on the basis of repertoire size alone (as an indicator of genetic quality), and Not on the quality of the page 29 resources controlled by those males (e.g. territory) or any other factor (e.g. reproductive experience).
That males with larger repertoires leave more offspring over their entire lifetime (i.e. that they have greater whole-lifetime fitness).
No study has yet been able to satisfy these conditions.
Existing studies (Nottebohm, 1972; Catchpole, 1973, 1976, 1979, 1980, 1981; Howard, 1974; Kroodsma, 1976a, 1977; Payne, 1979; Krebs and Kroodsma, 1980) suggest but do not unequivocally establish that intersexual selection has been important in the evolution of vocal complexity in some species. For example, Catchpole (1973) has argued that the principal function of the sedge warbler's extremely complex song is one of mate attraction, because the males cease singing immediately after pairing. He suggests that intersexual selection “might well be the main evolutionary force responsible for this extraordinary elaboration” (Catchpole, 1979, p. 55).
The importance of intersexual selection in the evolution of behaviour is an aspect of evolutionary theory which is complex, hotly debated, and in general poorly supported by convincing studies (Lambert et al. in press). Catchpole, one of the chief proponents of this hypothesis (Catchpole, 1973, 1976, 1979, 1980, 1981), in reviewing his own work and the work of others, states, “Sexual Selection may be important, but there is little evidence for it” (Catchpole, 1979, p. 58). At present this hypothesis remains an undemonstrated possibility.
3. Repertoires Increase Success in Territorial Competition
Although Slater (1981) found no obvious relationship between repertoire size and territory size or quality, the studies of Howard (1974), Krebs (1976b), Krebs et al. (1978), Yasukawa et al. (1980) and McGregor et al. (1981) strongly suggest that males with large repertoires may be more successful in territorial competition. “Speaker replacement” experiments have shown that territories are more successfully defended by larger repertoires (Krebs, 1976b; Krebs et al. 1978; Yasukawa, 1981).
Two recent studies (Yasukawa et al. 1980; McGregor et al. 1981) have determined that large repertoires may convey a reproductive advantage. It is known that survival of young is closely correlated with fledgling weight (Perrins, 1965). McGregor et al. (1981) found that great tits fledge heavier on territories defended by a male with a large repertoire; thus large repertoire males may have a reproductive advantage. Yasukawa et al. (1980) showed a correlation between repertoire size and pairing date in red-winged blackbirds (Males pairing earlier leave more offspring). Both studies interpreted this reproductive advantage as being the result of large-repertoire males being more successful in the male-male (intrasexual) competition for high quality territories; the reproductive difference being merely a reflection of the variation in territory quality. Both studies argued that the large-repertoire advantage was not the result of “female choice” of large-repertoire males because of their intrinsic fitness (see 2). Yasukawa et al. (1980) state in their final sentence, “Red-winged Blackbird song repertoires probably evolved in response to intrasexual selection” (p.237).
As yet there have been no experimental studies which clearly discriminate between the forces of intra and intersexual selection in the evolution of vocal complexity. Such a study would seem extremely difficult page 30 to design due to the confounding factors of age, reproductive experience and all the manifestations of male-male competition. Until such studies are done this area will remain grey, but on the basis of evidence currently available, it seems likely that intrasexual selection has been important in the evolution of some repertoires.
4. Repertoires Allow the Use of Different Songs in Different Contexts
Lein (1978) and Smith et al. (1978) suggested that different song types may represent different display intensities, so that a repertoire allows the use of graded vocal signals. They noted that some song types were used primarily near the territory centre, while other types were used closer to the periphery. This differential usage was interpreted as reflecting the probability of attack by the resident on a trespasser. However, studies by Krebs et al. (1978) and Smith and Reid (1979) have shown that different song types are used equally in all parts of the territory, and that (with some suggested exceptions, Morse, 1970; Lein, 1972) different song types (at least in great tits, Krebs et al. 1981) do not appear to have different overall functions (this study showed that all songs were used for both territorial competition and mate attraction, and that no one song was used as if it was “sexier” or more aggressive than the others). Slater (1981) found that the sequence of chaffinch song types changes little between normal song and response to playback, suggesting that chaffinches do not use different songs in different contexts.
Hinde (1958) and Bremond (in Armstrong, 1973) suggested that if it is important in territory defence to match the songs of a neighbour (matched countersinging), then the possession of a repertoire may allow an individual to match songs with more than one neighbour. “Song matching occurs when a bird responds to playback or a rival with a song out of its repertoire that closely resembles the stimulus song” (Krebs and Kroodsma, 1980, p. 151). Song matching appears common and has usually been interpreted as an aggressive response. Krebs et al. (in press) suggest that the degree of matching acts to signal different levels of response to the intruder. Alternative explanations exist, however, suggesting that matching may not be an aggressive response. Kroodsma (1979) hand-raised two marsh wrens and found that the subordinate bird tended to match the dominant bird's song, but not vice versa. Also the study by Baker et al. (1981) (and others, see 7) showed that the response of residents was stronger to the playback of a neighbouring dialect than it was to playback of songs from within their own dialect area. These observations are consistent with the Acoustic Camouflage hypothesis (Craig and Jenkins, in press; see 7) and suggest that song matching may not be primarily aggressive in function.
It seems possible that “different songs in different contexts” may have been a real influence in the evolution of repertoires. Certainly song matching is likely to be far more important in certain species (for example, chaffinches and great tits show a tendency to match playback; Hinde, 1958; Slater, 1981; Krebs et al, in press; whereas no such tendency is found in western meadowlarks; Falls and Krebs, 1975). If this is so, song matching (per se) may have many functions. The importance of song matching to the evolution of repertoires is, at present, unclear.
5. Repertoires Prevent Habituation
Hartshorne (1956, 1973) advanced the “anti-monotony principle” which proposes that diversity in a song sequence counters the monotony of a continuously repeated signal. Possession of a repertoire may then allow an individual to sing continuously without becoming monotonous. If the song conveys a territorial “Keep Out” signal, a varied performance may ensure that the songs remain effective (i.e. neighbours and intruders do not habituate to them). In support of this, Hartshorne presented evidence of a general interspecific relationship between “continuity” and “versatility” (see Slater, 1981; for alternative explanation of this relationship). Whether this relationship existed was questioned by Dobson and Lemon (1975) but confirmed by Kroodsma (1978).
Also in general support of this hypothesis is the work showing that individuals habituate more rapidly to the repeated playback of a single song type than to a repertoire (Krebs, 1976b). Although some features of song and behaviour are not consistent with this hypothesis (e.g. the enormous repetition of song types before switching; Krebs, 1978; Krebs and Kroodsma, 1980; Slater, 1981; all song types are not necessarily repeated with equal frequency; Slater, 1980, 1981; Dawson, in prep.), it is supported by considerable observational data (see Krebs and Kroodsma, 1980, for review).
Krebs (1976a, 1976b, 1977; with Kroodsma, 1980), Falls (1969) and Slater (1981) have considered why habituation occurs. Basically three ideas have been proposed.
That habituation is a fundamental property of neural systems, and has arisen because it is generally advantageous to ignore repeated stimuli (see Krebs and Kroodsma, 1980).
That habituation in the neighbour-neighbour context is adaptive in that it saves time and energy; then established neighbours do not waste time responding vigorously to each other's songs (Falls, 1969).
That habituation plays a major role in the settlement of immigrants into an area. Krebs (1976a, 1976b. 1977) suggested that habituation is part of a mechanism by which intruders assess the density of residents in an area. He suggests that perhaps the bird will not settle if it cannot habituate to the songs of the area, and will move on to a less densely populated area.
There seems no doubt that habituation in this context does occur, and that the possession of repertoires by residents would make habituation (in an intruder) slower or less likely. No study has yet shown why habituation (per se) occurs to song performances. Although the issue is cloudy, it seems likely that habituation may have been important in the evolution of repertoires.
6. The Beau Geste Hypothesis
One of the most recent hypotheses, the Beau Geste hypothesis (Krebs, 1977), suggests that “… song repertoires in some species have evolved in the context of density assessment… I suggest that repertoires are used by resident birds to increase the apparent density of singing residents, and hence decrease the apparent suitability of the area to new birds” (Krebs, 1977, p. 475). The idea is extremely simple, if, for example, an individual sings five different song types, then that individual may appear to an intruder to be five different birds. As previously mentioned, Krebs (1976a, 1976b, 1977) has argued that there needs to be a “functional explanation page 32 for intruder habituation” (Krebs, 1977, p. 477), and that is that “habituation is a plausible proximate mechanism by which a bird estimates the number of song types in an area” (Krebs, 1977, p. 477). He incorporates the habituation hypothesis into the Beau Geste hypothesis by suggesting that a bird may not settle in a new area until it has habituated to the songs of that area; therefore if each resident sings a number of different song types habituation will be slower and settlement more difficult. Clearly, if the Beau Geste hypothesis is applicable to a species, certain features of the song and singing behaviour of that species must be consistent with the “desired” deception. For example, the singer would be expected to change song post between renditions of different song types, so as to give the impression that he was more than one bird. These aspects have been studied for red-winged blackbirds (Smith and Reid, 1979; Yasukawa, 1981), great tits (Krebs et al. 1978) and chaffinches (Dawson, in prep.). Jenkins and Dawson (submitted for publication) have argued that the data from red-winged blackbirds (Smith and Reid, 1979; Yasukawa, 1981) are not relevant to the testing of the Beau Geste hypothesis, as song post changes in this species are easily visible to an intruder.
Rechten (1978) suggested that the hypothesis may be important in explaining the evolution of interspecific mimicry, but unfortunately there are no hard data pertaining to her suggestion.
The Beau Geste hypothesis has been criticised on theoretical grounds (Slater, 1978) chiefly due to problems of song type repetition. The question of how often a song type should be repeated before the singer changes song type is an area of disagreement for the two major workers in this field (Krebs, 1977, 1978; Slater, 1978, 1981). Slater has also suggested that according to the Beau Geste hypothesis, a bird should use all its song types equally (Slater, 1981). The studies of Slater (1981) and Dawson (in prep.) show that some chaffinches repeat one song type almost to the exclusion of the others. By Krebs’ (1978) own admission, the Beau Geste hypothesis has problems, but with relevant data from only two species (great tits and chaffinches) it is impossible to justify a generalisation regarding its overall importance.
7. The Acoustic Camouflage Hypothesis
A number of studies have shown that the response of resident birds is greater to the playback of a stranger's song than it is to the playback of a neighbour's song (Weeden and Falls, 1959; Brooks and Falls, 1975; Kroodsma, 1976b; Wunderle, 1978; Baker et al, 1981). Craig and Jenkins (in press) argue that it would be advantageous for intruders to be able to sound like residents, as presumably by doing so they would evoke a less aggressive response. They argue that the pressure on intruders to match may be countered by a pressure on residents to make the matching of their songs more difficult. They suggest that repertoires evolved as a means of achieving a complex vocal output so that intruders find it more difficult to match songs (sound like a resident), and so are discouraged from settling. Supporting this hypothesis is Jenkins’ (1978) observation of a saddleback that modified its song as it dispersed to a new area, the modified song more closely matching the “dialect” of the new area. As discussed previously, some authors are in doubt as to whether song matching is aggressive or not (see 4). If song matching is highly aggressive, then an individual that matched songs from an area would be expelled from that area faster than an individual that did not match songs. Also possibly at variance with the page 33 Acoustic Camouflage hypothesis is the evidence that nightingales share less of their song types with close neighbours than they do with more distant birds (Hultsch and Todt, 1981).
A strong point of this hypothesis is that the authors have (as Krebs has for the Beau Geste hypothesis) generated a number of predictions which will be useful in evaluating its importance.
One obvious characteristic of the literature concerning the significance of vocal repertoires is that there are many hypotheses but few hard data. Many of the hypotheses have appealing features and are supported by some field evidence. As yet no one hypothesis adequately explains the evolution of repertoires. On the basis of the variety of species possessing repertoires and the variety of social systems involved, it seems doubtful whether such a general explanation exists. Also, only adaptationist (sensu Gould and Lewontin, 1979) explanations have been stressed in the literature, and the existence of constraints in the evolution of song repertoires has not been widely considered. Gould and Lewontin (1979) suggest a more holistic approach to evolutionary questions. They advocate the consideration of apparently adaptive aspects of a behavioural or morphological feature in the light of the constraints imposed by epigenesis, the environment, conflicting selective forces, and random effects in allele fixation. In short, they argue that organisms must be analysed as integrated wholes.
Although the merit of this approach seems obvious, its influence is absent in most song studies — perhaps that is why we have so few answers to the basic questions.
Peter Jenkins, Elisabeth Slooten, Chris White, John Craig, Ian McLean and Tony Hughes discussed the material presented here on a number of occasions, and straightened my thinking where necessary. I thank Dave Lambert for general discussions on evolution and Professor E. C. Young for urging me to publish this work.
Armstrong, E. A., 1973. A Study of Bird Song. Dover, London.
Baker, M. C., Thompson, D. B., Sherman, G. L. and Cunningham, M. A., 1981. The role of male vs. male interactions in maintaining population dialect structure. Behav. Ecol. Sociobiol. 8: 65-69.
Brockway, B. F., 1965. Stimulation of ovarian development and egg laying by male courtship vocalisation in budgerigars (Melopsittacus undulatus). Anim. Behav. 13:575-578.
Brooks, R. J. and Falls, J. J., 1975. Individual recognition by song in white-throated sparrows. III. Song features used in individual recognition. Can. J. Zool. 53: 1749-1761.
Catchpole, C. K., 1973. The functions of advertising song in the sedge warbler and reed warbler. Behaviour 46: 300-320.
——, 1976. Temporal and sequential organisation of song in the sedge warbler (Acrocephalus schoeobaenus). Behaviour 59: 226-246.
——, 1979. Vocal communication in Birds. Edward Arnold. London.
——, 1980. Sexual selection and the evolution of complex songs among European warblers of the genus Acrocephalus. Behaviour 74: 149-166.
——, 1981. Why do birds sing? New Scientist, 2nd April: 29-31.
Craig, J. L. and Jenkins, P. F. The evolution of complexity in broadcast songs of passerines. J. Theoret. Biol. (In press).
Dobson, D. W. and Lemon, R. E., 1975. Re-examination of monotony threshold hypothesis in bird song. Nature (London) 251: 126-128.
Emlen, S. T., 1971. The role of song in individual recognition in the indigo bunting. Z. Tiepsychol. 28:241-246.
Falls, J. B., 1969. Functions of territorial song in the white-throated sparrow. In Bird Vocalisation (R. A. Hinde, ed.): 207-232. Cambridge Univ. Press, London and New York.
Falls, J. B. and Krebs, J. R., 1975. Sequences of songs in the repertoires of western meadowlarks. Can. J. Zool. 53: 1165-1178.
Gould, S. J. and Lewontin, R. C., 1979. The spandrels of San Marco and the panglossian paradigm: a critique of the adaptationist programme. Proc. Roy. Soc. London B205:581-598.
Hartshorne, C., 1956. The monotony threshold in singing birds. Auk, 73: 176-192.
—— 1973. Born to Sing. Indiana University Press, Bloomington.
Hinde, R. A. 1958. Alternative motor patterns in chaffinch song. Anim, Behav. 6: 211-281.
Howard, R. D., 1974. The influence of sexual selection and interspecific competition on mocking bird song (Mimus polglottus). Evolution 28: 428-438.
Hughes, A. J. 1981. The vocal dynamics of the North Island kokako. Unpub. M.Sc. thesis, University of Auckland.
Hultsch, H. and Todt, D. 1981. Repertoire sharing and song-post distance in nightingales (Luscinia megarhynchosB.). Behav. Ecol. Sociobiol. 8: 183-188.
Jenkins, P. F., 1978. Cultural transmission of song patterns and dialect development in a free-living bird population. Anim. Behav. 26: 50-78.
Krebs, J. R. 1976a. Bird song and territorial defence. New Scientist. 3rd June: 534-536.
—— 1976b. Habituation and song repertoires in the great tit. Behav. Ecol. Sociobiol. i: 215-227.
——, 1977. The significance of song repertoires: the Beau Geste hypothesis. Anim. Behav. 25: 475-478.
—— 1978. Beau Geste and song repetition: a reply to Slater. Anim. Behav. 26: 304-305.
Krebs, J. R., Ashcroft, R. and Webber, M. I., 1978. Song repertoires and territory defence, Nature 271: 539-542.
Krebs, J. R., Ashcroft, R. and van Orsdol, K. Experiments on song matching in the great tit. Parus major L. Anim. Behav. (In press).
Krebs, J. R., Avery, M. and Cowie, R. J., 1981. Effect of removal of mate on the singing behaviour of great tits. Anim. Behav. 29: 635-637.
Krebs, J. R. and Kroodsma, D. E., 1980. Repertoires and geographical variation in bird song. Advances in the Study of Behaviour, vol. 11: 143-177.
Kroodsma, D. E., 1976a. Reproductive development in a female songbird: differential stimulation by quality of male song. Science 192: 574-575.
——, 1976b. The effect by large song repertoires on neighbour “recognition” in male song sparrows. Condor 78: 97-99.
——, 1977. Correlates of song organisation among North American Wrens. Am. Nat. 111:995-1008.
——, 1978. Continuity and versatility in bird song: support for the monotony threshold hypothesis, Nature 74: 681-683.
——, 1979. Vocal duetting among male marsh wrens: evidence for ritualised expressions of dominance/subordinance. Auk 96: 506-515.
Kroodsma, D. E. and Parker, L. D., 1977. Vocal virtuosity in the brown thrasher. Auk. 94: 783-785.
Lambert, D. M., Kingett, P. D. and Slooten, E. Intersexual selection: the problem and a discussion of the evidence. Evolutionary Theory. In press.
Lein, M. R., 1972. Territorial and courtship songs of birds. Nature 237: 48-49.
——, 1978. Song variation in a population of chestnut-sided warblers (Dendroica pensylvanica: Its nature and suggested significance. Can. J. Zool. 56: 1266-1283.
McGregor, P. K., Krebs, J. R. and Perrins, C. M., 1981. Song repertoires and reproductive success in the great tit. Am. Nat. 118: 149-159.
Marler, P., 1952. Variation in the song of the chaffinch, Fringilla coelebs. Ibis 94: 458-472.
Miller, O. B., 1979. Long term recognition of father's songs by female zebra finches. Nature 280: 389-391.
Morse, D. H., 1970. Territorial and courtship song of birds. Nature 226: 659-661.
Nottebohm, F., 1969. The song of the chingolo, Zonotrichia capensis, in Argentina: description and evaluation of a system of dialects. Condor 71: 299-315.
——, 1970. Ontogeny of bird song. Science 167: 950-956.
——, 1972. The origins of vocal learning. Am. Nat. 106: 116-140.
Payne, R. B., 1979. Sexual selection and intersexual difference in variance of breeding success. Am. Nat. 114: 447-452.
Peek, F. W., 1972. An experimental study of the territorial function of vocal and visual display in the male red-winged blackbird (Agelaius phoenicius). Anim. Behav. 20: 112-118.
Perrins, C. M., 1965. Population fluctuations and clutch size in the great tit (Parus major L.). J. Anim. Ecol. 34: 601-647.
Rechten, C., 1978. Interspecific mimicry in birdsong: Does the Beau Geste hypothesis apply? Anim. Behav. 26: 305-306.
Shiovitz, K. A., 1975. The process of species-specific song recognition by the indigo bunting Passerina cyanea and its relationship to the organisation of avian acoustical behaviour. Behaviour55: 128-179.
Slater, P. J. B., 1978. Beau Geste has problems. Anim. Behav.26: 304.
——, 1980. Individual differences in animal behaviour. In Perspectives in Ethology IV. (P.P.G. Bateson and P. H. Klopfer, eds.). Plenum Press, New York.
——, 1981. Chaffinch song repertoires: Observations, experiments and a discussion of their significance. Z. Tierpsychol.56: 1-24.
Slater, P. J. B. and Ince, S. A., 1979. Cultural evolution in chaffinch song. Behaviour71: 147-166.
Smith, D. G. and Reid, F. A., 1979. Roles of song repertoire in red-winged blackbirds. Behav. Ecol. Sociobiol.5: 279-290.
Smith, W. J. Pawlukiewicz, J., and Smith, S. T., 1978. Kinds of activities correlated with singing patterns of the yellow-throated vireo. Anim. Behav. 26: 862-884.
Trainer, J. M., 1980. Comments on a kin association model of bird song dialects. Anim. Behav. 28: 310-311.
Treisman, M., 1978. Bird song dialects, repertoire size and kin association. Anim. Behav. 26: 814-817.
——, 1980. Some difficulties in testing explanations for the occurence of bird song dialects. Anim. Behav. 28: 311-312.
Weeden, S. J. and Falls, J. B., 1959. Differential responses of male ovenbirds to recorded songs of neighbouring and more distant individuals. Auk 76: 343-351.
Wunderle, J. M., 1978. Differential responses of territorial yellowthroats to the songs of neighbours and non-neighbours. Auk 95: 389-395.
Yasukawa, K., 1981. Song repertoires in the red-winged blackbird (Agelaius phoenicius): A test of the Beau Geste hypothesis. Anim. Behav. 29: 114-125.
Yasukawa, K., Blank, J. L. and Patterson, C. B., 1980. Song repertoires and sexual selection in the red-winged blackbird. Behav. Ecol. Sociobiol.7: 233-238.