Song Sparrow Melospiza melodia Scientific name definitions

Peter Arcese, Mark K. Sogge, Amy B. Marr, and Michael A. Patten
Version: 1.0 — Published March 4, 2020
Text last updated January 1, 2002

Behavior

Locomotion

Walking, Hopping, Climbing, Etc

Primarily walks, but hops on rough or uneven substrates, or from branch to branch. Often runs through grass and weeds (Judd 1901b, Stone 1937). Hailman (Hailman 1973) described “skip” (asynchronous movement of legs). Commonly furtive in manner, working downward into bushes with bobbing tail (especially females returning to nest; PA), hopping twig to twig, or through underbrush, grass, and leaves. Chicks climb on edge of nest or branches by 9 d, earlier in warm conditions late in nesting season (Nice 1943b, PA). Hopping and walking develop 1 wk after fledging; runs rapidly with sharp turns when “frolicking” (17 d; Nice 1943b).

Flight

Flights usually short between perches or to cover; direct and low, often “twisting” with bursts of flapping; exception when territorial birds pursue intruders and each demonstrates purposeful flight (PA). Because strong flight required in migration, suggests jerky, tail-pumping flight serves as a display function.

Other display flights more obvious and similar to those of Savannah Sparrow (Wheelwright and Rising 1993); e.g., male and female of pair often use a fluttering flight with tail cocked and feet dangling. Male employs flutter flight to approach females early in season, often in combination with flight song, sometimes landing near or on female (Nice 1943b, PA). Nice (Nice 1943b) described related “pounce” wherein male flies directly at female, occasionally making contact, to land and then sing strongly. Male uses similar but more cautious approach in territorial disputes with new neighbors, and with established neighbors if females construct new nests near old border (Arcese 1989a).

Wing area (approx. 86.5 cm2) equals approximately 3.94 cm2/g total body weight (Poole 1938), similar to other ground-feeders and similarly sized sparrows. Maximum flight speed 26–48 km/h (Cottam et al. 1942b, Pearson 1961). Flight song noted (see above).

Self-Maintenance

Preening, Head-Scratching, Stretching, Bathing, Anting, Etc

Preening. Preening common, especially after bath or rain; does so habitually after copulation, often after territorial conflict (PA). Typically preens atop or within bush or low tree; interrupted frequently to scan surroundings.

Head-Scratching. Wing dropped, leg brought up and over wing to scratch head (Nice 1943b); most frequent during or after bath. Also scratches head on perch (Hailman 1959a).

Bill-Wiping. Wipes bill by lowering and rolling head side to side, alternately rubbing sides of tomium on perch or ground (Nice 1943b).

Stretching. Most common as simultaneous stretching down of one wing, corresponding leg and half of tail, often just prior to sunbath (PA); also stretches legs up with wings closed (Nice 1943b).

Bathing. Bathes frequently; during day as opportunity allows or after sunset (Bent 1968a; details in Nice 1943b). On Mandarte I., British Columbia, bathing most common 09:00–16:00; typically once/d in spring and summer, rare in winter (PA). Bathes in puddles, including salt water, and on shrub and grass leaves (“leaf-bathing”) by striking foliage with wings and body (Verbeek 1962b, PA).

Anting. Adults and juveniles ant. Groskin (Groskin 1950a) described adult's use of queen Lasius niger and worker Formica sanguinea subintegra; Nice (Nice 1943b) described anting by juvenile: ant crushed in bill, rubbed on feathers, wings, tail, and legs.

Sleeping, Roosting, Sunbathing

Sleep. Adult turns head back and tucks bill into wing coverts (Nice 1943b); young adopt this method upon leaving nest. In Alaskan winter, Sutton and Wilson (Sutton and Wilson 1946) saw bird asleep in a niche on rock face; fluffed up, bill between back and scapular plumage, chest against rock, and tail out, down, and spread slightly.

Waking. Wright (Wright 1913) estimated waking based on dawn song at 55–90 min before sunrise (n = 22), but Fisler (Fisler 1962) and PA noted “night song” common. Nice (Nice 1937) suggested females woke approximately 25 min before sunrise on clear mornings, 5 min later when cloudy; light 0.3–0.4 and 0.2–0.4 ft-candles, respectively. Male “waking song” 16–49 min (n = 360) before sunrise when clear, 5 min later (n = 238) when cloudy; earlier May–Jun than remainder of year (Nice 1943b). In s. Michigan, waking 18–46 min before sunrise; earliest before summer solstice (Fisler 1962).

Roosting. Roosts in dense vegetation, including weed or shrub thickets, usually close to ground. Sutton and Wilson (Sutton and Wilson 1946) described usual winter roost on Attu I., AK, as rock crevice or grass tussock. Roosted in Ohio from 10 min presunset to 22 min post (Nice 1943b); in New Hampshire (Wright 1913), average was 23 min after sunset (n = 22).

Sunbathing. Sunbathes on warm days, often after water bath in association with preening. Wing and tail spread, body feathers fluffed, mouth open, with body inclined to one side (Nice 1943b, MKS).

Daily Time Budget

From Arcese (Arcese 1989a) or as noted. Varies seasonally, especially in females whose time when breeding is limited by available energy and occupied mainly by activities of nesting. Six to 8 wk prebreeding, males and females more involved in territory defense, advertisement, and interactions with mate, especially with food plentiful. Females on Mandarte I. spent approximately 63% of time out of sight, 30% feeding, 5% alert, 1% defending or displaying, and 1% preening with natural food available (n = 14), but with supplemental food spent less time out of sight (48%) and feeding (8%) and more alert (22%), defending (6%), displaying (6%), and preening (10%; n = 13). Unfed males (n = 14) similar to fed females (54% out of sight, 23% feeding, 17% alert, 1.5% defense, 4.5% display and <1% preening), but fed males (n = 14) increased time alert (57%) and nearly ceased display (mainly song; <1%), spent less time out of sight (27%) and 5% feeding, and spent about same time in defense (3%) and preening (4%). Because alert is main behavior used to detect intruders, results show each sex allocated more time to defense with energy limits relaxed. Adults out of sight nearly continuously during molt. In Dec–Feb on Mandarte I., birds spent approximately 90% of daylight hours feeding (PA).

Agonistic Behavior

Physical Interactions

Uncommon; fights involving contact of wings, feet, or bills, and bodies of contestants at feeders made up <10% of 3,500 interactions between juveniles (Arcese and Smith 1985). Physical interactions between territorial neighbors rare and confined to same-sex interactions, usually during territory establishment (Nice 1937, Tompa 1964, Smith et al. 1982b). In these cases, contact occurs as culmination of escalated border disputes between recently established neighbors or when floaters of either sex attempt to settle and fail to yield to display or chase by owner (Nice 1943b, Arcese et al. 1988, Arcese 1989a). Highest level of escalation rare; involves birds locked in tumbling fights with bills used as spears and feet clasping opponent's body (Arcese Arcese 1987, Arcese 1989a).

Communicative Interactions

Threat Displays. Nice (Nice 1943b: 157) details 10 “methods of intimidation”; 4 noted here. “Ballooning” exclusive to territorial birds or persistent intruder; often by males within several centimeters, at common border on ground or low in shrub, with repeated quiet song. Birds with feathers erect, body hunched, wings out of pocket of flank-feathers. “Puff-sing-wave” mainly by males; with feathers puffed, wings vibrating, and in song, bird flies slowly toward challenger or mate. Mounted males in Ohio elicited this display from territorial males prior to attack, but on Mandarte I., males new on territory often perform this toward resident female, who responds with “caterwaul” (PA). “Menace” entails lowered crest, with body crouched forward and closed bill pointed at victim. “Pounce” performed exclusively by males and directed toward female mate (rarely a female neighbor); male flies to approach female, occasionally alighting as though initiating coition, then flies immediately to second perch, often with simultaneous song. Nice called “pounce” a “method of intimidation”; but this display typical of males in a high state of excitement during fertile period of females, who often respond with “caterwaul” and subsequently displace male (PA).

Appeasement Displays. Nonterritorial floaters when perched prominently, or when approached from a distance by residents, stand erect with contour feathers pressed close to body, wings folded onto back, but crest erect (“crest-up”; PA). Common in all nonterritorial birds of each sex; not observed in territory owners on Mandarte I., even when off territory; hence, a reliable clue to status. “Crest-up” often accompanied by pseet (PA).

Spacing

Territoriality

Nature and Extent. Males and females territorial year-round where resident. Characteristic behaviors (vocalization, chasing, and perching prominently) ubiquitous 1–3 mo prebreeding, subsiding somewhat toward end of breeding unless replacement mates, neighbors, or nonterritorial floaters attempt to usurp territory; reappears Sep–Oct (Johnston Johnston 1956c, Johnston 1956d; Tompa 1964; Arcese 1989c; Wingfield 1994; Wingfield and Hahn 1994), even in partially migratory populations (Nice 1937). Site attachment and zeal with which exclusive use of territories enforced varies in nonbreeding period; weaker where migratory (Nice 1937), stronger where resident (Johnston Johnston 1956c, Johnston 1956d; Tompa 1964; Arcese 1989c; Wingfield 1994). Attachment to breeding territory strong at low elevation on West Coast, with birds often remaining in territory even when not “actively defended” (Johnston 1956d), or active defense year-round but evidenced less by song and more by eviction of trespassers at close range (Arcese Arcese 1987, Arcese 1989a; Wingfield and Hahn 1994). Young may defend territory upon completion of Prebasic molt, as early as Aug in hatch year (Arcese 1989b, Wingfield and Hahn 1994).

Residents on Mandarte I. intensely territorial during fall settlement, Sep–Oct; mild conflicts in fair weather through Jan, intensifying again Feb–May (Tompa 1964; Arcese Arcese 1989c, Arcese 1989b, Arcese 1989a). Defense throughout year focused on birds of same sex (Arcese Arcese 1987, Arcese 1989a; Arcese et al. 1988). In cold periods (Nov–Feb), owners left territory daily to feed at concentrated food source (Arcese 1989b), where local owners exerted dominance in former/future breeding territories (PA). Territoriality of residents in winter not much studied elsewhere, but probably similar with reduced attachment to breeding area as environment more marginal. Nice (Nice 1937) noted male range in winter of 3.4 ha, 6–10 times larger than breeding territory; in cold spells, birds came 270–500 m to visit feeder. On tiny Mandarte I. (approx. 6 ha), birds moved up to 350 m to feed, depending on distance between territory and winter food (Arcese 1989b). Thus, with commutes included, winter range on Mandarte similar to winter range in Ohio.

Wingfield and Monk (Wingfield and Monk 1992) suggested function of territoriality in Washington varied seasonally, being unrelated to reproduction in fall–winter even for birds remaining in breeding area. But much work on residents shows maintenance of winter territory essential to retain ownership in spring, and that territory tenure a key predictor of lifetime reproductive success (Knapton and Krebs 1974, Arcese 1989c, Hochachka et al. 1989, Beecher et al. 2000b). Related experiments on seasonal variation in territorial behavior, response to simulated intrusion and role of testosterone in territorial behavior also demonstrated generally consistent nature of territorial response throughout year in males (Wingfield 1994) and females (Elekonich and Wingfield 2000). Overall, evidence suggests fall territoriality and year-round defense part of an integrated suite of life-history behaviors, with temporary absences from territory a result of balancing risks of starvation and usurpation of territory left undefended.

Territorial females aggressive to female floaters, deterring settlement (Smith et al. 1980, Arcese et al. 1988) and limiting polygyny (Arcese 1989a), but variation across populations probably occurs. Nice (Nice 1937: 88) noted females “do not fight . . . over mates” but “exhibit a defensive attitude towards neighbors of like sex, dogging each others' footsteps in a hunched up or puffed out attitude.” Nonbreeders absent in Ohio (Nice 1937), rare on Mandarte I. (PA). Food addition increased territoriality in females and reduced settlement by female floaters on occupied territories, instead of enhancing settlement as predicted by “polygyny-threshold” (Arcese 1989a). Hence, level of female aggression may reflect energy limitation or geographic variation in fitness premium of early nesting when latter trades off against effort allotted to territory defense (Arcese 1989a). Overall, evidence suggests female territorial behavior spaces individuals in populations and limits polygyny, but rarely prevents females from breeding.

Boundaries denoted by location of song perches and agonistic displays with neighbors; obvious in open habitat where territories as small as 65 m(un-mated male), less obvious in shrub-forest mix where territories to approximately 6,500 m(Nice 1937, PA). Borders in open areas harder to define, perhaps overlapping or undefended (Suthers 1960, Tompa 1964, Sogge and Van Riper III 1988). Song perch typically a bare branch of shrub or tree, often with overhead cover, on or near border with expansive view of territory or portion thereof (Nice 1937, Tompa 1964).

Territories often arranged linearly in marsh (Marshall 1948a; Johnston Johnston 1956c, Johnston 1956d) and agricultural habitat (Knapton 1976), along sloughs, streams, or hedgerows bounded by water, grassland, agricultural field; but ≥3 deep in floodplain (Nice 1937), clustered or disjunct in forest or suburban habitats (PA).

Territory size ranges widely within and between populations depending on annual variation in density, mated status and age of owner, and type of habitat. On Mandarte I., territories of pairs as small as 200 mat high density (approx. 25 pairs/ha; Arcese 1989b) to approximately 5,000 mat low density (approx. 2 pairs/ha). Also variable on San Miguel I., CA (700–4,900 m2); smaller in dense shrub (mean 0.15–0.25 ha; 2 yr) than open areas (mean 0.28–0.42; 2 yr), and in newly settled (1,800 m± 700 SD) versus established males (2,300 m± 1,100 SD; Sogge and Van Riper III 1988). Territories of 5 unmated males on Mandarte I. smaller (range 65–105 m2) than all those of breeders (range 110–400 m2; Tompa 1971); average (288 m2; n = 47) approximately 9% of minimum size in Ohio (approx. 3,300 m2; Nice 1943b). Territories in tidal marsh 380–500 m(Johnston 1956d), smaller than those on Minnesota lakeshore (approx. 1,100 m2; Suthers 1960) or in rural Massachusetts (approximately 900–5,700 m2; DeGraaf 1989). Size related positively to passerine species richness (Yeaton and Cody 1974). Overall, evidence suggests variation in local density of conspecifics the primary factor affecting territory size.

Territory size of returning summer-resident males implanted with testosterone about double that of controls (mean 1,466 m± 145 SE vs. 2,955 m± 325 SE; Wingfield 1984). But subsequent experiments on castrated males showed territories maintained in absence of testosterone (Wingfield 1994). Date of territory occupation by male migrants correlated positively with settlement date of females, suggesting each related positively to territory quality (Searcy 1984). Nearest-neighbor analysis of adult survival and reproduction suggested marked variation in territory quality (Hochachka et al. 1989), but fine-scale attributes of habitat in territories on Mandarte I. related only weakly to record of occupation over 30 yr (PA).

Territory boundaries remarkably stable year to year (Nice 1937, Johnston 1956d, Tompa 1964) despite changes in ownership (Arcese Arcese 1987, Arcese 1989c). Knapton and Krebs (Knapton and Krebs 1974) found that males removed experimentally from territories were replaced within 12–72 h by dominant floaters from juvenile flocks, and showed experimentally that pattern of territory settlement affects territory and population size.

Manner Of Establishing And Maintaining Territory

Descriptions for males; similar but less studied in females (Nice 1937, Tompa 1964, Smith et al. 1982b, Arcese et al. 1988, Arcese 1989a . Elekonich and Wingfield 2000). Establishment consists “of five parts: assuming the role; staking out the claim; the chase; the fight”; then “the proclamation of ownership” (see Nice 1937: 57 for details; similar in resident populations, PA). Most territories established by hatch-year young. On Mandarte I., young “float” after independence from parental care for an average of 7 mo (Arcese 1989b); maintain home range encompassing 3–25 territories and nonbreeding habitat. Floaters gain territories by deposing territorial birds, inserting themselves between adjacent territories, or fighting with other floaters for ownership of vacant territory (Arcese 1989c). Male floaters continually intrude on territories within home range and account for approximately 25% of territory losses by established owners (Arcese Arcese 1987, Arcese 1989c); further 25% of losses due to eviction by territorial neighbor; remainder by death of owner (Arcese 1989c). Studies of resident populations in Washington suggest mechanisms of acquisition and loss similar (Beecher et al. 1994, Beecher et al. 2000b; Wingfield 1994).

Young may settle after completing Prebasic molt in Aug, but some delay to second or third year of life (Nice 1937, Arcese 1989c, Smith and Arcese 1989, Wingfield and Hahn 1994). Territories also established by hatch-year young from fall to late Feb in San Francisco Bay, CA (Johnston Johnston 1956c, Johnston 1956d), and in Ohio, even though some fall settlers migrate thereafter in Ohio (Nice 1937). Young settlers sometimes evicted by deposed owners upon latter's completion of Definitive Prebasic molt (Nice 1937, PA). Among summer residents that return to find prior territory occupied, “there is little wing fluttering and puffing, merely the singing, chasing and fight” (Nice 1937: 58). In late stages of establishment, contestants sing frequently but softly in rapid succession in border disputes (Nice 1937, Arcese 1989c, see Agonistic behavior, above). Unmated males often stop singing and desert territory early to midseason, unless mated earlier (Nice 1937, Tompa 1964, PA).

Much evidence suggests territorial behavior is under hormonal control (e.g., Wingfield 1994, Wingfield and Hahn 1994, Soma et al. 2000), but which hormones involved and mechanisms unclear. Plasma testosterone increased as resident males began spontaneous song (Feb–Mar), peaked as females began to lay, again during laying of second brood, then declined to basal level prior to molt (Wingfield and Hahn 1994). But expression of territoriality and response to simulated intrusion largely independent of circulating testosterone (Wingfield 1994, Wingfield and Hahn 1994). Castrated and control males responded similarly to intrusion in fall and spring; suggests testosterone does not activate territorial aggression, but may increase intensity of aggressive response (Wingfield 1994). Recent work suggests estrogen regulates male aggression in nonbreeding period (Soma et al. 2000).

Individual Distance

Interindividual distance generally >45 cm for nonterritorial birds, greater for territorial and dominant birds (Nice 1937, PA; see below), but ≤10 cm among agonists. Nonbreeders in loose flocks of 2–20 birds (Nice 1937; Johnston Johnston 1956c, Johnston 1956d; Arcese and Smith 1985). Resident adults in winter flocks noted in Ohio (Nice 1937) and Washington State (Wingfield and Monk 1992).

Interspecific Territoriality

Chases most bird species, except those much larger; Nice (Nice 1937: 68) lists 21 species pursued by male/female in Ohio. Territorial males in particular chase similar-sized sparrows during pre-breeding (Savannah, Lincoln's, White-throated, White-crowned, and Golden-crowned sparrows; Johnston 1956d, Tompa 1964), but zeal wanes as breeding starts (PA). Aggression between Swamp and Song sparrows uncommon during breeding despite occupying juxtaposed territories; and males unresponsive to heterospecific song (Greenberg 1988b). Also tolerates Fox and White-crowned sparrows in territory once breeding under way (PA). In contrast, generally exerts dominance over all similar-sized species during nonbreeding, including White-throated Sparrows in lab (Wagner and Gauthreaux 1990) and field (Dennis 1950), despite larger size of latter; 99% of 251 encounters between White-throated and Song sparrows won by latter (Schneider 1979). Swamp Sparrow also subordinate to Song Sparrow (Greenberg 1988b), but Song Sparrow dominated by Fox Sparrow in 11 of 12 encounters at feeders (Smith et al. 1980).

Winter Territoriality

For residents, see above; function and exclusivity of winter territories in migrants and winter residents unclear and little studied.

Dominance Hierarchies

From Arcese and Smith 1985, or as noted. Well established at and away from artificial food sources (Knapton 1976). In resident populations, dominance status at feeders in summer/fall a predictor of survival and settlement on territory the following spring in males and females (Knapton 1976). Status independent of body mass, wing length, or tarsus length, but related negatively to hatch date; latter accounting for 59% of variation in status overall. Twice as many males as females won ≥50% of encounters; no detectable effect of parental age on dominance of young; smallest surviving young in broods as likely as largest to become dominant. Accumulated experience in agonistic encounters or physiological development probably a key determinant of dominance.

Males generally dominate females in nonbreeding period (Smith et al. 1980); may facilitate higher average survival of males versus females (Arcese et al. 1992). Territorial males spent more time at feeders and achieved twice as many displacements as they suffered, followed by territorial females, yearling males, and yearling females (Smith et al. 1980; see also Knapton 1976, Wagner and Gauthreaux 1990). Females often dominate their mates during breeding (Dixon 1986, Arcese 1989a).

Sexual Behavior

Social Mating System And Sex Ratio

Based on Nice 1937Johnston 1956c, Johnston 1956d; Tompa 1964; Smith et al. 1982b; and Arcese 1989a; except as noted. Primarily monogamous, but up to 21% of females and 13% of males breed in a polygamous group at least once in their lifetimes. Nice recorded 4 cases of bigamy in 6 yr (2.3% of 173 male-yr), similar to long-term average of 3% of mated males polygynous on Mandarte I. Males rarely with 3 females; 1 male mated to 4 females on a tiny island in British Columbia with only those 5 adults present (Smith et al. 1996c). Overall, available data suggest rate of polygyny varies little geographically.

Polygyny related positively to ratio of territorial females to males (r = 0.68; Arcese 1989a). Slight to moderate biases in number of territorial females compared to territorial males occurred in 14 of 15 yr (mean 0.98; range 0.64–1.07) on Mandarte I., similar to Ohio and California (Nice 1943b, Johnston 1956d). However, polygyny noted even when territorial males outnumbered females 1.4:1 (Arcese 1989a). Nice (Nice 1943b) claimed little evidence of a nonbreeding reserve of unmated birds of either sex, but reported several cases wherein floating males replaced missing territory owners or survived locally to settle in next year; unmated males also noted to abandon territory in absence of female mate. Male floaters on Mandarte I. composed up to 35% of adult males alive in summer, but were uncommon at low density (Smith and Arcese 1989, PA); suggests Nice's observations, made at densities similar to low range on Mandarte I., were consistent with density accounted for.

Polygynous groups form in 2 ways in about equal occurrence. First, territorial males may annex the adjacent territories of missing or ailing males (Arcese 1989c). When this occurs after a widow has already commenced breeding, bigamy may result (Nice 1937, Smith et al. 1982b, Arcese 1989a). Second, female floaters sometimes settle on the territory of an already mated male, usually during incubation period of resident female early in breeding period (Arcese 1989a). Polygyny also induced experimentally in 2 ways, both related to manipulation of local sex ratio. First, Smith et al. (Smith et al. 1982b) altered adult sex ratio directly by removing 9 males after their females' first broods had hatched, with the result that 4 of 9 widows became polygynous. Second, Wingfield (Wingfield 1984) implanted migrant males with testosterone upon their arrival in New York in spring, prior to peak arrival of females. Experiment affected local sex ratio indirectly because implanted males defended larger territories than control males and females, and they became polygynous more often. Chance that individual females participated in polygyny was also reduced by providing them with supplemental food prior to breeding; fed females increased time in defense and repelled settlement by floaters more successfully than controls (Arcese 1989a).

Overall, low average level of polygyny probably regulated by variation in adult sex ratio and related to costs of defense against floaters by territorial females. Reluctance of females to mate in polygamous groups (Arcese et al. 1988, Arcese 1989a) probably a result of the fact that males usually feed only first brood to hatch (Nice 1937, Smith et al. 1982b, Arcese 1989a). Consequently, unaided females raise about half as many young as females with male help (Smith et al. 1982b, Arcese 1989a). Males probably attempt polygyny because it increases their reproductive rate (Smith et al. 1982b, Arcese 1989a).

Close inbreeding a rare but regular occurrence, most often a result of natal site fidelity for each sex (Nice 1937, Arcese 1989b, Keller 1998a). No evidence of inbreeding avoidance based on Wrights' F statistics and a pedigree (Keller and Arcese 1998).

Pair Bond

Nice (Nice 1943b: 170–184) described in detail pair formation, courtship, and relation of each to nesting cycle. These descriptions probably suffice except where birds are resident year-round because in those populations, pair formation occurs at any time, more or less synchronously with territory establishment. In general, pairing followed by marked reduction in male song except as associated with “pounce.” Pair often feed and perch within ≤3 m for approximately 10 d–3 wk prior to female beginning incubation. Thereafter, male behavior more characteristic of unmated male, except to “pounce” on females leaving nest. When pairing outside breeding period, interactions between sexes subdued relative to descriptions of Nice for Feb–Mar; involves mainly quiet song by males, “caterwaul” by females, and feeding or loafing in close association (PA). At these times, sequence and elements of pair formation often obscure owing to circumstances over which pairing occurs; e.g., ≥50% of males (fewer females) acquire territory by evicting prior owner (Arcese 1989c, Arcese et al. 1988).

Nice (Nice 1943b) claimed no obvious male trait influenced female choice; similar to Mandarte I., where yearround residency makes it as likely that female or male is “first” bird settled (Arcese 1989b). Overall, “pair” on Mandarte I. appears to be a de-facto result of male and female territory overlap; establishment of social unit related to coordination of behavior preparatory to nesting. But degree to which these behaviors vary by region or life history an exceptional topic for further study.

Courtship Displays And Mate-Guarding

Nice (Nice 1937) describes males as “dominating” their female and neighboring females when male neighbor out of sight. Males “pounce” on female neighbors more roughly than their mate. Neighbors typically repulsed or chased by unaccompanied female, often issuing “caterwaul,” or chased vigorously by returning male mate. Males typically maintain proximity to mate during whole of nesting cycle, but particularly during fertile period prior to nest-building and incubation. At this time, males reduce song rate dramatically to approximately 5 songs/h or less, often staying within 3 m of mate while feeding, bathing, or loafing (Arcese 1989a, PA). As females incubate, males resume higher song rate, similar to unmated males; may then acquire additional territory and mates by annexation (Smith et al. 1982b, Arcese 1989c).

Copulation; Pre- And Postcopulatory Displays

Copulations observed most often just prior to nest-building and incubation, to 6 d into incubation; any time during day (Nice 1943b, PA). Typically, female emits nasal ee-ee-ee, flies to low perch in open or on ground, lowers head, raises and opens bill slightly, flutters wings, raises tail to expose cloaca. May flutter ≥4 s, occasionally flying closer to male and repeating above if male fails to respond. More often, male flies immediately to female, alighting on back to initiate cloacal contact. Coition lasts approximately 1 s; male often returns to mount 2–4 times (Nice 1943b, PA, MKS). Male quiet, preening immediately afterward; female may fly to feed, preen, or build nest.

Duration And Maintenance Of Pair Bond

Mate desertion somewhat (Nice 1937) to very common (Weatherhead and Boak 1986). On Mandarte I., approximately 0–25% of pairs remain intact from year to year, owing mainly to mortality (PA). Often no obvious attempt of female to maintain bond even between broods (Arcese 1989a; see also Nice 1943b: 183 for detail).

Extra-Pair Mating Behavior

Nice (Nice 1937: 85) noted “faithfulness during a whole season is the rule,” but Keller (Keller 1998a) used microsatellites in nDNA to show that ≥15% of 200 young were not sired by the social mate. Similar to resident population in Washington State, where all extra-pair sires identified as neighbors (C. Hill pers. comm.). Behavioral precursors to extra-pair matings identical to those with social mate, but occasionally initiated by female flight to territory of neighbor (PA).

Social and Interspecific Behavior

Degree Of Sociality

Social interactions limited mainly to those of pair and aggressive interactions of neighbors or territory owners and floaters. Juveniles form loose flocks and congregate at feeding sites, such as shrubs with ripening berries, or at bathing sites with fresh water, and may maintain contact with tseep notes when traveling between such sites. However, systematic plotting of juvenile home ranges suggests little continuity in group membership and no tendency for brood mates to disperse together (L. F. Keller and PA unpubl.).

Play

Nice (Nice 1943b) described “frolicking” as chief form of play in young birds; characterized as sudden runs or flights with sharp turns without particular context. Noted in wild from 40 d of age through first fall (approx. 120 d) and on occasion in a 3-yr-old captive.

Nonpredatory Interspecific Interactions

Aggressive interactions noted with at least 23 species, including Northern Cardinal (Cardinalis cardinalis); Eastern Towhee (Pipilo erythrophthalmus); White-throated, White-crowned, Golden-crowned, Savannah, Swamp, Lincoln's, and Field (Spizella pusilla) sparrows; and Purple Finch (Young 1990; PA); all but Field, Swamp, Lincoln's, and Savannah sparrows larger. Supplanted repeatedly by Fox Sparrow (Smith et al. 1980, PA; see also Spacing, above).

Predation

Kinds Of Predators

Adults killed by gray fox (Urocyon littoralis) and Loggerhead Shrike (Lanius ludovicianus) on San Miguel I., CA (Sogge and Van Riper III 1988). Northern Harrier (Circus cyaneus) and Short-eared Owl (Asio flammeus) important predators in 1950s in San Francisco Bay, CA (Johnston 1956c), but invading red fox (Vulpes vulpes) now also a regular predator (Y. L. Chan pers. comm.). Other predators of adults on Mandarte I. include Barn (Tyto alba) and Burrowing (Athene cunicularia) owls, Sharp-shinned (Accipiter striatus) and Cooper's (A. cooperi) hawks, and Glaucous-winged Gulls (Tompa 1964, Arcese 1989b, PA); one Cooper's Hawk removed 16+ birds (approx. 20% of population) in winter 1998–1999, as indicated by bands recovered in castings (PA; see also Arcese 1989b).

Garter snakes (Thamnophis spp.) thought to take many eggs and young (Nice 1937). Where snakes and most other predators absent, 93% of depredated nests on San Miguel I. attributed to endemic fox (Sogge and Van Riper III 1988). On Mandarte I., cowbirds a major cause of nest depredation until young ≥6 d of age (Arcese et al. 1996; see Breeding: brood parasitism, below). Other nest predators probably include most small to midsized mammals, many snakes, corvids, and cowbirds (Nice 1937, Arcese and Smith 1999). Nest predators confirmed by sight or circumstance include: deer mouse (Peromyscus maniculatus), raccoon (Procyon lotor), mink (Mustella vison), river otter (Lutra canadensis), rat (Rattus spp.), opossum (Didelphis sp.), mule deer (Odocoileus hemionis), red squirrel (Tamiasciurus hudsonicus), gray fox, various snakes, Brown-headed Cowbird (Molothrus ater), Northwestern Crow (Corvus caurinus), Glaucous-winged Gull, Barn Owl, and domestic cats (Y. Chan pers. comm., PA).

Manner Of Predation

Accipiters fly fast and low at height of shrub canopy to attack perching birds and those on ground in openings (PA). Open-country owls and harriers soar over grassy areas to drop on birds, and Barn Owls attack brooding females on nest (PA). A Glaucous-winged Gull snatched a singing male from perch in a windstorm (PA) and depredated eggs and nestlings incidentally. Crows observe fledglings and parents delivering food and dive into shrub cover to capture young; may also search for nests. Owls and accipiters consume whole bird, but crows often drop legs disarticulated at pelvis under perch. Deer mice unable to open intact eggs in trials on Mandarte I. (B. T. Martinez and PA unpubl.), but do appear to kill nestlings, sometimes one at a time over 2–3 d (PA). Cowbirds puncture and eject eggs and nestlings, presumably to initiate renesting by host (Arcese et al. 1996; see Breeding: brood parasitism, below). In British Columbia, signs of cowbird depredation rare on islands with mice, probably due to scavenging of ejected eggs and nestlings; but these signs common on islands with no resident mammals (dead nestlings with contusions, dented skulls, skin breaks, and missing feathers, on ground or in nest; PA, ABM). Revisits to >200 experimental nests with sparrow and plasticine eggs on Mandarte I. indicated approximately 50% were visited by mice within 48 h of placing nests out (B. T. Martinez and PA unpubl.). DuBois (Dubois 1956) described cowbird killing nestlings in Michigan, matching our observations in British Columbia (Arcese and Smith 1999). As expected by these observations, removing cowbirds reduced markedly rates of nest failure (Arcese and Smith 1999, J. N. M. Smith pers. comm.).

Response To Predators

Hand-raised birds responded with escape reactions to threatened capture by human; movement of immediate environment; fright in companions; fear note (tik) and sounds resembling it; and approach of large objects, small children, a half-grown chicken, and birds flying overhead; but hand-raised birds did not recognize cat or dog as enemy (Nice 1943b). In field, fledgling ceased begging instantly after male gave tik; garter snake coiled under nest with young elicited “curiosity” and a peck from adult male, whereas a female “attacked small snakes”; hand-reared bird shown a garter snake craned its neck but no other alarm (Nice 1943b: 257). Many probably benign mammals (e.g., rabbit, domestic dog) elicit no response or alarm if close to nest; whereas domestic cats, raccoons, squirrels, and foxes invariably elicit alarm response (tchunk, tik-tik-tik). Accipiters, Northern Harrier elicit tik and hiding response lasting ≥5 min (Nice 1943b, PA). Hand-raised birds showed strong response to cardboard model of Barred Owl (Strix varia: Nice 1943b), but roosting Saw-whet (Aegolius acadicus) and Great Horned (Bubo virginianus) owls seemed to go unnoticed by birds in winter on Mandarte I. (PA). Other potential predators capable of eliciting alarm include: Blue Jay (Cyanocitta cristata), grackle (Quiscalus spp.), Northwestern and American (Corvus brachyrhynchos) crows, and Brown-headed Cowbird. Mobbing response limited to cowbirds; crows elicit tchunk in the territory and tik within approximately 5 m of nest or fledgling.

Recommended Citation

Arcese, P., M. K. Sogge, A. B. Marr, and M. A. Patten (2020). Song Sparrow (Melospiza melodia), version 1.0. In Birds of the World (A. F. Poole and F. B. Gill, Editors). Cornell Lab of Ornithology, Ithaca, NY, USA. https://doi.org/10.2173/bow.sonspa.01
Birds of the World

Partnerships

A global alliance of nature organizations working to document the natural history of all bird species at an unprecedented scale.