Sounds

Development

In late summer a few independent juvenile males give variable subsong. On arrival in the next spring the first-year males sometimes give variable subsong, as do a few older adult males. The first songs of spring are short and the notes morph from first to last within a series. In males banded as nestlings in the previous year, the early songs are not like the songs of their father, or of another male on a nearby territory during the natal year. The first-year males then change from the first song to another song in their first breeding season. This later song usually matches the song of a neighboring territorial male. Males usually keep their definitive (last-sung) song themes into later years, as long as they continue to return to the area. Song improvisation is not evident insofar as song of hand-reared isolates is abnormal in structure and in playback tests this song does not attract other male buntings (Rice and Thompson 1968, Shiovitz 1975, Payne 1981, 1996, Payne and Payne 1997).

In laboratory experiments, first-year males reared from nestlings in isolation, before they could hear a male other than their father, developed a song, but it was not normal (it did not sound like a bunting song, and wild buntings ignored a playback of the recorded song). These males did not have the song of their father (Rice and Thompson 1968, Payne 1981). Other males were reared from nestlings and maintained in a cage between two adult males, both in auditory contact but only one in visual contact into their first spring. These males copied song of the male with which they could see and interact through the screen, and not song of the male they could only hear, and not song of their father. To trace song transmission, the tutors were chosen to differ in song from each other and from the young male's father (Payne 1981). In other experiments, males were tutored with tape-recorded bunting song rather than with a live tutor. These males did not learn the complete song; even though they learned certain notes in their first year and as late as their second year. They did not sing songs of their father, and even nestmates from the same brood developed different songs. A sensitive period for song learning extends into the second year of life (Rice and Thompson 1968). And males learn songs from a social tutor better than from a tape recording. In another experiment, first-year males that were captured in migration in autumn (their father's song and their own song experience unknown) developed odd songs, but when they lived next to an adult male, some changed to match the adult's song (Margoliash et al. 1994).

In the field, about 80% of first-year males in first spring season copy the song of a neighboring territorial male. The other 20% may learn a song elsewhere before settling on territory; this would explain the source of their songs. In first year, certain males switch songs repeatedly through the season, changing songs when they move from one neighborhood to another, each time matching the new neighboring territorial male. About 5% of older males also switch from one local song theme to another, mainly in a year when no neighbors match their own song. A few males copy songs of two neighbors, and they remain bilingual in later years or they drop the song that is not shared by any neighbor. Less often, first-year males recombine phrases of two males into a mixed or hybrid song theme; and one male recombined the two song themes of a bilingual neighbor into a mixed song theme (Payne et al. 1988, Payne 1996, Payne and Payne 1997).

Song Transmission And Cultural Success

First-year males often settle on a territory that borders the territory of two or more other males. How do first-year males decide which male and which song to copy; and what makes some males more successful as tutors in transmitting their song to a ready learner? Older adult males that arriver earlier in spring are more likely to be copied, just because they are on the area longer and their song theme has more time to be copied. Older males are more likely to be copied than first-year males, and the first-year males in bluer breeding plumage are more likely to be copied than the browner first-year males. Male "song models" (the copied males) do not have more females or more fledglings than the average male in the population, and so the males that are the cultural successes do not tend to have unusually high biological success. First-year males do not seem to assess in advance which neighbors will have success in breeding in the season or in their lifetime, insofar as they do not style their song after the song of the biologically successful neighbors. First-year males on territories in neighborhoods of high population density were more likely to copy a neighbor's song than were first-year males in sparse neighborhoods, but first-year males did not copy the most common song by conforming out of proportion to number of males with the song theme. Their time of arrival and their breeding plumage account in part for the cultural success of certain males in recruiting copiers (the number of first-year males that copy their song themes); other sources of variation in cultural success are unknown (Payne and Payne 1993b, 1997).

Vocal Array

Songs complex, highly ordered sequences of several notes, usually invariant in order and number of notes, usually given in pairs, and persisting for > 1 s (mean, 2.20 s; Thompson 1968; Fig. 5). Given by territorial males during the breeding season; appear to advertise a male’s occupancy of a territory to other males and to females.

Only males sing. Each male has a single complex song theme, which it retains from first spring through later years (Thompson 1970, Payne et al. 1988, Payne 1996). Songs composed of elements from a species repertoire of about 100 notes or “song figures.” Individual males often combine the elements in different sequences. With 6 to 8 kinds of notes in a song and no strict rules to order their sequence (except a short note, one of eight such short notes in the repertoire of the bunting species, to introduce a a song Shiovitz 1975), the permutations of notes in songs are immense (for only the 3-note sequences, the value is about 3¹⁰⁰). Odds are low that two birds develop the same sequence of three or more different notes—a song theme— independently.

Song is comprised of a series of loud, ringing metallic chirp-notes of nearly equal stress, the notes are usually given in pairs. Song is more patterened than melodic, and it lacks trills and complex bubbles. "He always introduces his song with a pianissimo downward chirp, then proceeds loudly with two or three upward chirps, continues with a series which alternates up and down...it is common for him to sing at the rate of five songs a minute for an hour at a time...not less than two thousand times in a day!" (Mathews 1904). Allen (1933) described song as like "Sweet, sweet–where, where–here, here—see it, see it." Verbal descriptions of songs in Michigan give the pattern of notes and are used in the field, by our students, as mnemomics for the local song themes: "Quick quick, run run, jump in the lake", "Have a sweet sweet cherry bing", "Oh my my it's a nice nice morning, don't you think?", and "I do chem, why don't you?" Most notes are given in pairs, a few long notes given as singles, ad short notes in triplets; some notes have more than one simple part.

The sequences of notes are unique to local song neighborhoods. Males a few hundred meters apart generally have different song themes as identified by 3 or more note types in common and in the same sequence. Certain males are individualistic in their song. However, bunting males on neighboring territories often have the same or nearly identical songs (Thompson 1970, Payne et al. 1988; Fig. 5). Song neighborhoods where buntings share or match the same song theme average 3 or 4 males, and a neighborhood may have as many as 22 males all with the same song theme (Payne et al. 1988). Song neighborhoods may be discontinuous in space, when a bird learns a song in one area, then disperses it to another area and retains its song, where it may be copied in turn (Payne 1996, Payne and Payne 1997).

Buntings that return to their natal area usually settle several territories distant from their natal site, and there they learn their song (Payne 1991, Payne and Payne 1997). Because males do not sing the song of their father, the song neighborhoods or dialects do not indicate kinship, and songs do not track genetic lineages.

Few first-year males have the song of their father or the song of another male in their father's song neighborhood, and these first-year males account for only 1 in 80 males. One first-year male that had his father's song theme first gave another song theme when the young male first arrived and had a territory, then he moved next to father's territory and copied father's song. Males reared in different broods on the same territory return next year with different songs, not the song of father; and even males fledged in the same brood have different songs (Payne et al. 1987). Most breeding adults are born outside their natal song neighborhood and birds of known natal history breed in a different song neighborhood and not the one where they were born. Many first-year males not born there arrive with songs unlike any songs in the study area during the previous natal year, then match a local song in their first spring, so these songs could not have been learned there in their natal year (Payne and Payne 1997). There is no suggestion of genetic differences between different song neighborhoods (mean F_ST= 0.0038, NS; Payne and Westneat 1988).

Song notes and song structure are consistent thorugh the breeding range. The song elements described in a population in Michigan (Thompson 1970), are widespread through the range of the species, and only about 10% of elements in any population do not match the Thompson catalogue (Shiovitz and Thompson 1970, Emlen 1971a Emlen et al. 1975, Payne et al. 1981, Margoliash et. al 1991, Baker and Boylan 1995, Payne and Payne 1997). No sequences of three or more note types are found shared between populations 250 km apart (Payne et al. 1981, 1988, Payne 1982, 1996). Song structure is similar in regions where Indigo Buntings are sympatric or allopatric with Painted Buntings (Forsythe 1974), and it is similar where they are sympatric or allopatric with Lazuli Buntings (Emlen et al. 1975), with no behavioral divergence in sympatry with other bunting species.

Phenology

Males do not sing on wintering grounds; they begin to sing just before or during spring migration in April (Taber and Johnston 1968, Jones 2004). Males sing on breeding grounds from time of arrival in spring through summer. Certain males become quiet in late June, others sing through summer until mid-Aug. when they begin a postbreeding molt. They do not sing during molt or in autumn migration.

Daily Pattern

Males sing from dawn to dusk, with highest rate in early morning before sunrise (more than 200 songs/h) decreasing to about 60/h through most of the day (Taber and Johnston 1968, Thompson 1972). Rarely sing at night.

Places Of Singing

Males sing from several song perches within their territory, the song sites are both peripheral and central sites within their territory. Males may sing an hour in one tree, more often they change song sites several times per hour. Some males sing in a tree over the nest; others sing on perches away from nest site. Males also sing on roads, the surface reflecting their song. Occasionally sing in flight in territorial chases with other males.

Repertoire And Delivery Of Song

Each male has a single song type or song theme. The song is nearly invariant in its notes and in their pattern of sequence, but the terminal notes are not always given. Songs tend to become shorter near the end of a singing bout (Thompson 1972). In territorial chases, songs may be given continuously, end to end, recycling the pattern with no pause between songs, and in a state of excitement a male inserts unpaired high-pitched, buzzy cheet notes into intervals between pairs of notes.

Males sing throughout breeding season. Song rates vary with stage of nesting, highest in unmated males (680 songs/h), decreasing during nest building (24/h), and it increases after the female has completed laying. Song rate increases when a neighbor sings (Thompson 1972); countersinging infrequent, less than 10% of the singing time of a male (Payne 1983a).

Social Context

Song given spontaneously through breeding season, less frequently when a male has a nest-building or laying female (Thompson 1972). Males attracted to playback of song in their territory, and attack the playback as if an intrusion by another male. Males sometimes sing when chasing an intruder, and even sing when grappling on the ground with another male in early spring. Song is typically on a perch, occasional at dawn and dusk in flight display (Saunders 1929, Sutton, in Taber and Johnston 1968). Females do not respond to playback in the field, even though unmated females may be attracted to males by hearing their songs. In captivity, adult males give a very soft, plastic song when isolated during winter and spring, so soft that it can be heard only within 1 m. (Margoliash et al. 1991). In the field in Belize in early Feb., when buntings come to roost in evening, perch in marsh at night, or before flying from roost to feeding areas at dawn, we heard no song, nor did birds sing in feeding areas (RBP).

Species Recognition

Indigo Bunting songs differ from songs of other buntings in form of figures and in syntax, especially in the pairing of song elements (Thompson 1968). In song playback experiments, in regions where the buntings occur with no other bunting species in their area (allopatric), the males respond to songs of their own species and not to songs of other bunting species (Thompson 1969, Emlen 1972). In sympatry, where Indigo and Lazuli buntings sometimes hybridize (Emlen et al. 1975), the two species share song elements or songs and they may not discriminate in response to conspecific and allospecific songs. In these areas, copying songs of the other species is associated with interspecific territoriality and the mixed songs are used to recognize and repel rival males of either species. Frequency range and timing differences among bunting species are more critical to the differential responsiveness of males than are the individual notes or their delivery in pairs within a song. Males use more than one cue in identification of their species' songs, and the birds' cues differ from those used by ornithologists who listen for the paired note sequences (Emlen 1972, Konishi et al. 1989). Females in the lab solicit copulation in response to song and visual model of a male of their own species more than to those of a Lazuli Bunting (Baker and Baker 1988, 1990, Baker 1991, 1994, 1996 . Females of the two species in the field tend to mate assortatively with respect to male song and to male plumage. Mixed species pairs have the same reproductive success per nest as same-species pairs; hybrid buntings, especially females, tend to have lower reproductive success than non-hybrid buntings (Baker and Boylan 1999).

Individual Recognition

In field playback experiments, males respond more strongly to songs of unfamiliar strangers than to songs of neighbors or to their own songs, but the response varies with the presence of a male dummy to attack, the position of the speaker in the male's territory, and the particular male. In tests where a speaker was located within a male's territory, with a mount of another male bunting, his level of resonse to his own song was the same as response to a male of a neighboring territory, whereas his response tended to be the higher to song of a stranger—but the difference was not significant (Belcher and Thompson 1969). In another test where the speaker was located near the center of a male's territory, and no mount was present, his response to the song of a strange male was higher than to the song of a male on a neighboring territory (Emlen 1971b . In these tests both the song of neighboring male and the song of stranger differed from the song of the subject male. In a third test, the speaker was placed within the male's territory near the side furthest from the neighbor with the test song. He responded equally to song of a neighbor and song of a stranger (Payne 1983a). In addition, a male challenged by songs responded to no more strongly to the song of a neighbor with the same song as his own than to the song of a neighbor with a different song. Although individual recognition is not well established, males tend to respond more aggressively to song that differs from their own song. In observations (not in playback tests), males tend to countersing more with a neighboring male when the neighbor matches his own song themes than when their song themes differ (Payne 1983a). Song copying may be a form of social mimicry, the copying first-year male being "recognized" by other males as the older established adult neighbor. Song copying appears not to be sexual mimicry, insofar as a female does not switch her territory to mate with a male with the same song theme as her earlier mate (Payne 1983a).

Cultural Evolution

Song themes in a local population change over time. Changes within a population occur in two ways, by song modification in learning, and by a balance of cultural extinction and immigration.

First, some first-year males modify the song theme learned from a neighbor. The first-year males may add or delete a song element, or substitute or modify a song element. In one historical song theme, a particular note changed in form across three cultural generations. When older males returning for a second year are copied, the new first-year males copy these modified songs. A gradual change within a song theme over generations results in a changing song tradition. And sometimes two first-year males modify the song theme in different ways, and each of these is copied the next year, in a process of splitting into sub-traditions. Some song traditions persist for 20 yr or longer and are transmitted across many bunting generations (Payne et al. 1981, Payne and Payne 1993b, Payne 1996).

The second way that song themes change within a population is by local extinction and immigration (not by invention, because a male without a model does not develop songs that other buntings recognize as a valent song). Because only a few males match a particular song theme (often 3 or 4 males in a season), and annual survival is about 50%, often none of these males return to the area in the next year. Or a song theme simply is not copied by new first-year males; these songs die when their singers do not return next year. Song themes shared by more males are more likely to be copied by the next generation because of a greater probability that at least one male with the song theme returns to the area. These local extinctions of songs are balanced when immigrant males bring their own songs into the population. These new songs seed new local traditions, when first-year males copy them. In Michigan, about 50% of song themes in a year were present in the previous year, and the others are songs of new immigrants. Most new songs disappear within the year when their singers are first-year males that then drop their old song and match the song of an adult neighbor. At the George Reserve, all song themes recorded in 1978 from nearly 100 males had gone extinct by 1994 (Payne and Payne 1990, 1993a, Payne 1996).

Young have a series of begging calls that grade into one another as the birds develop. Fig. 6A: peep, a nestling call about 0.1 s in duration and rising rapidly in tone to a peak at about 7 kHz. The call is audible from the day of hatching and becomes louder as the nestlings develop.

Fig. 6B: cheep, a call given at 5–11 d, develops directly from the peep, longer (to 0.2 s) and changing from an initial rapid rise in pitch to a broad-band harmonically complex harsh sound. Peep and cheep are used by the young when begging for food from their parents.

Fig. 6C: cheet develops directly from cheep, and is similar in fledged young and adults, but is usually lower pitched and more prolonged in the young, with a peak in energy about 6 kHz in the juvenile and 7 kHz in the adult, and with a descending frequency in the younger birds. Fledged young give the call when begging food from parents.

Fig. 6D: churr appears at 9–10 dof age and is loud and longer (about 0.3 s), with dominant formants at 3.7 and 7.4 kHz and with marked amplitude modulation (80 cps); given when disturbed. In the field this attracts the parents, who often fly repeatedly just over the young, e.g., to distract a predator.

Adults give a series of calls in certain contexts; there are no playback tests to confirm their effect on behavior of another bunting. Males insert a high buzzy sound (perhaps cheet but often higher than 8 kHz) into their song when chasing other males or responding to playback of a song (Thompson 1972).

Fig. 6E: adult aaaa, similar to the churr of fledglings and given when threatening another bunting.

Fig. 6F: chip, a mild distress call given in the wild when a person approaches. Chip is a short (< 0.1 s) call rapidly rising in pitch and readily located by ear. Young chip near fledging age. Adult buntings give a rapid series of chips (>10/0.5) near the nest, when their young are about to leave (or have just left the nest) and are approached by a person. Also chip and cheet in flocks on wintering grounds.

Fig. 6G: tink, a call given by adult males in high-intensity distress situations, a note with a narrow frequency band (< 0.5 kHz) and high in pitch (about 8 kHz) with a mean duration of 0.2 s. This call given when other males intrude onto territory and the resident male has a mate building a nest or laying (sexually receptive), when a hawk flies over, or when a predator approaches. Males often switch from chip to tink as a person approaches the nest, and the call is often given while males crouch and fluff the plumage, in pausing to attack an intruder or to approach a playback of a song on their territory, and when captured in a net.

Fig. 6H: eeee, a distress call given on handling, differs from other bunting calls in having a variety of harmonics and a rise then a fall in pitch; the duration is about 1.2 s. Call heard when bander (and presumably a predator) is at nest with young near fledging.

Fig. 6I: ti-ti-ti-ti, a twittering call given by a female as she crouches and actively solicits copulation from a male. The call, a trill with each note having two peak amplitudes around 4 kHz, is repeated at a rate of about 10 notes/s, with the duration and repetition indicating motivation of the female to mate. Females occasionally give this call and assume a crouched posture, wings outspread and quivering and tail elevated, as they flutter away from a bander at the nest—a behavior similar in form to the “solicitation invitation” of mating, both indicating arousal.

Fig. 6J: tseep, a buzzing sound given by the male while he flies toward her as she solicits. The call is structurally similar to cheet and is an amplitude-modulated sound lasting about 0.11 s, rising in pitch, usually given in bouts of four calls, whereas cheet is given singly.

None observed.