Species names in all available languages
|English (United States)||Purple Martin|
|Haitian (Haiti)||Irondèl vyolèt|
|Lithuanian||Purpurinė miškinė kregždė|
|Spanish (Argentina)||Golondrina Purpúrea|
|Spanish (Costa Rica)||Martín Purpúrea|
|Spanish (Cuba)||Golondrina azul americana|
|Spanish (Dominican Republic)||Golondrina Migratoria|
|Spanish (Ecuador)||Martín Purpúreo|
|Spanish (Mexico)||Golondrina Azulnegra|
|Spanish (Panama)||Martín Purpúreo|
|Spanish (Paraguay)||Golondrina purpúrea|
|Spanish (Peru)||Martín Purpúreo|
|Spanish (Puerto Rico)||Golondrina Púrpura|
|Spanish (Spain)||Golondrina purpúrea|
|Spanish (Uruguay)||Golondrina Purpúrea|
|Spanish (Venezuela)||Golondrina de Iglesias|
|Turkish||Büyük Mor Kırlangıç|
In this revision, Charles R. Brown and Daniel A. Airola revised all content. Peter Pyle contributed to the Appearance page. Arnau Bonan Barfull and Peter Pyle curated the media.
Progne subis (Linnaeus, 1758)
The Key to Scientific Names
Purple Martin Progne subis Scientific name definitions
Version: 2.0 — Published September 10, 2021
Account navigation Account navigation
Welcome to Birds of the World!
You are currently viewing one of the free accounts available in our complimentary tour of Birds of the World. In this courtesy review, you can access all the life history articles and the multimedia galleries associated with this account.
For complete access to all accounts, a subscription is required.
Walking, Hopping, Climbing, etc.
Flies to the ground only to collect nesting material, to attempt forced copulations, or to pick up grit. When on the ground, martins walk exclusively. Commonly walks along porches below the entrance holes of birdhouses. Occasionally sidles along a wire or tree branch using a sideways walk.
From Blake (198), except where noted. Flies at various heights, from just above the ground to 1,889 m (167). There are no estimates of the typical flight speed, although a martin in a homing experiment averaged 43.5 km/h during a nocturnal homing flight (199). The pattern of flight is very much like that of the Tree Swallow (Tachycineta bicolor)—so much so that the two species can be confused at a distance. Sails in circles estimated to be 10–20 m in diameter, with regular alteration between quick flapping and gliding. Glides are mostly short, < 7 s and usually 3–4 s, though occasionally up to 26 s. Holds wings horizontal during glides. Changes altitude usually slowly, although it can descend rapidly from high altitudes by closing the wings completely and hurtling downward in an almost vertical dive or fall, pulling out just before reaching the nest site. Has more speed and power, but less maneuverability than smaller swallows. When pursuing insect prey, individuals make sudden turns to either side or upward, accelerate, and then flare the tail while catching an insect, whereupon they return to near the original altitude (CRB). Young birds are very competent flyers at fledging (128).
Swimming and Diving
Not known to swim or dive.
Preening, Head-Scratching, Stretching, Bathing, Anting, etc.
Often preens, probably more so than the smaller swallows, while sitting on wires, in trees, on buildings, on birdhouses, and inside the nest cavity (200). Preens relatively constantly throughout the nesting season, except for a decline during feeding of nestlings; preening increases markedly once birds finish breeding and join late-summer roosts. Preening at a birdhouse seems to indicate residency; visitors and prospectors at a site seem too agitated to preen until they become established. Preens at all times of the day. Preening often exposes a patch of white feathers on the back. White patch is believed to be a visual signal associated with preening (171), but its function is unclear.
Scratches its head over the wing; birds in flight bring the leg forward to scratch the throat and upper breast (201, CRB). Stretches by extending one wing at a time below the feet, then extends both wings simultaneously in a "V" over the back. This stretching sequence often immediately precedes flight takeoff. Birds sometimes yawn while stretching. Bathes by skimming the water surface in flight and hitting surface briefly in a violent collision, sometimes several times in succession (201). Anting not known to occur.
Sleeping, Roosting, Sunbathing
In a Texas birdhouse population, birds usually sleep inside nest compartments until midsummer (15 June), after which they begin sleeping in trees (202). Residents sleep in a compartment on their territory; members of a pair sometimes sleep in the same compartment, sometimes in separate compartments. Male often seems to choose whether to sleep with the female, especially early in the season, and pairs that have been together for a week or more often sleep together (202). As the breeding season progresses, sleeping is gradually confined to a compartment containing the nest, and females always sleep in the nest during egg-laying and incubation. Some males continue sleeping in the nest after eggs are laid; others move to other compartments on the territory (202). Some females (and males) stop sleeping in the nest when nestlings exceed 13–15 d old, and these adults either use another compartment or begin sleeping in trees. Sleeping behavior is described in detail by Brown (202).
Typically sleeps in trees in late summer, when pre-migratory flocking begins. Sleeps in late-summer roosts only at night (128); roost sites are usually deserted during the day. Birds from multiple colonies in central California gather during the breeding season at evening roost sites before dispersing back to colonies for the night (160). Doppler radar data from birds in a pre-migratory roost in South Carolina show morning roost departure occurred an average of 41 min before sunrise (peak 10 min), with birds departing in all directions for up to 100 km (203, 204). Flight speeds ranged from 10.8–13.4 m/sec during roost departure. The number of individuals in radar image plumes was estimated to range from 3,988 (05:50 h) to 32,662 (06:10 h;203). Evening arrival at a roost is more variable spatiotemporally, but most birds approach the roost near sunset. Cloud cover shortens the time away from roosts (204). One roost in South Carolina was occupied by approximately 700,000 birds (204). Western martins (subspecies arboricola) formerly used large premigratory roosts (e.g., Seattle, Washington; 161), but none are known today, likely because of lower population sizes (50). In British Columbia, they continue to use small local communal roosts after breeding and until departure on migration (B. Cousens, personal communication).
In northwestern Pennsylvania, radio-tagged fledglings (n = 15) often returned to their natal colony to sleep for several consecutive nights in either their own nest compartment or another vacant compartment (ST). Fledglings are first detected at pre-migratory roosts 11–16 d after fledging. Young birds occupied roosts an average of 14.7 nights each (range 2–24) prior to migration. Some fledglings appear to begin fall migration approximately 14 d after fledging (128).
Suns on perches such as wires, tops or porches of birdhouses, or roofs of buildings. Ruffles feathers of rump and head, visibly pants with mouth agape and tongue extended, droops wings, and rolls body partly over to one side (171). Eyes remain open, and one is exposed to direct sunlight; there is no conspicuous action of the eyelids or nictitating membranes. Can hold the sunning posture for 3–5 min in the absence of disturbance (171).
Daily Time Budget
Few data. During a 15-h period of daylight, 8 females that were feeding nestlings spent on average 9.2 h/d flying (away from colony, presumably foraging) and 5.8 h/d sitting at the colony (presumably mostly brooding or tending nestlings;190). For an unspecified number of males that were not feeding nestlings, time spent flying was 7.1 h/d and sitting 7.9 h/d (190).
Usually limited to instances when an intruder enters the nest compartment that is claimed by a bird already resident. Fights inside the compartment can be violent, with combatants pecking each other with their beaks, beating each other with their wings, and holding on to each other with their feet. Intruders are usually ousted from a compartment quickly; the defeated bird struggles to exit while territorial resident holds on to its back or tail with its beak and continues to peck it. If a particular intruder is persistent at trying to claim part of a resident's territory, the resident soon seems to recognize that bird, and fights may develop even before the intruder enters a compartment. Birds are occasionally injured seriously in these repeated fights (205).
Generally, males defend the territory from other males, and females from other females , and both sexes exhibit the same behavior when fighting. Paired male becomes excited and begins profuse vocalizing with Zweet calls (see Sounds and Vocal Behavior: Vocalizations) when a foreign female enters his territory, and he does not try to drive her away. Females seem to ignore foreign males, although if her mate is absent, a female will oust an intruding male who enters a compartment on the territory. Most territorial intrusions occur early in the breeding season and usually cease by the time eggs hatch.
Little information on agonistic behavior during the nonbreeding season. At roosts away from nest sites, incoming birds frequently try to displace others already perched (14). Individuals prefer upper perches, and contest these favored positions through attempted displacements. In a roost of subspecies hesperia in Arizona, light-breasted birds (females and yearling males) were often successful at displacing dark-breasted adult males who were submissive in displacement interactions (14). Greater mortality of yearling females during a thunderstorm at an Oklahoma roost may have reflected differences in perch access, with yearling females relegated to more exposed sites within the roost through behavioral interactions (206).
Frequently tolerant of intruders on territories; martins advertise ownership by postures and behavior. Birds engage in physical fighting only if an intruder goes inside a compartment or is persistent. Territory owners of both sexes enter a compartment as soon as a same-sex intruder arrives nearby, an obvious signal of ownership that is displayed both by birds using birdhouses and by birds nesting in tree cavities (CRB). Residents sit inside the compartment, looking out and vocalizing, as long as an intruder is present. Both sexes repeatedly flip the wings and tail upward, a sign of general anxiety often given by residents when an intruder arrives, by males when courting females, and by both sexes when terrestrial predators (e.g., humans) approach a nest site. Birds gape at interspecific intruders such as House Sparrow (Passer domesticus) and European Starling (Sturnus vulgaris), but do not gape during intraspecific agonistic interactions. Males snap their bill in a gaping motion at their mates, resembling juveniles begging for food. Johnston and Hardy (171) believed bill-snapping was an appeasement display, although little is known about it.
Function of yearling male's female-like plumage is unclear (207, 25). It does not seem to deceive adult males into identifying yearling males as females (thereby improving yearlings' chances of taking over the adults' territories), because adult male territory owners are equally aggressive toward yearling-male and adult-male intruders (207). Female-like yearling males did not have an advantage in territory acquisition over yearling males dyed black; yearlings made to look like adults gained territories faster than female-like males. The female-like plumage of the yearling male does not seem to reflect a molt constraint that prevents yearlings from growing in their blue feathers before the breeding season (207).
In eastern populations, individuals defend territories that often consist of multiple compartments within a birdhouse, and occasionally multiple birdhouses or gourds (208). A single male may defend up to 36 compartments early in the season; typically, he defends the compartments themselves and adjoining porches, but not the tops of birdhouses. Males tend to defend more rooms early in the season; for males (n = 19) arriving in February in northern Texas, mean territory size early in season was 12.5 compartments (208). Later-arriving males defend fewer compartments, and almost all birds gradually relinquish portions of their territories as the season progresses, although many birds continue to defend several compartments throughout the breeding season. In Texas, mean minimal territory size among all males (n = 158) was 2.6 compartments (208). The consequence is that not all rooms in a martin house are occupied by nesting pairs; most birds have adjacent spare rooms where one or both members of pair may sleep (202). One male defended 18 compartments throughout one season in a Texas colony (208).
Females also defend territories containing multiple rooms, although they often have smaller territories (fewer compartments) and usually restrict their activities almost entirely to the nest compartment by the time eggs are laid. Occasionally a single female will defend a territory containing two resident males. Until she relinquishes part of that territory, the second male stays unpaired. A female in a northern Texas colony controlled a territory with two males for 26 d (CRB). See Brown (208) for a full description of territorial behavior.
Nonresident males secure nest sites by repeatedly challenging territory holders (209). Territory owners win all fights with intruders, but eventually some owners appear to tire of fighting a particular male and relinquish one or more compartments to him. Nonresidents seem to win through a “war of attrition” (209). Females exhibit the same pattern, although the smaller territories defended by females, especially later in season, reduce conflict among females over space. Males may remain without a territory, as floaters, for up to 29 d (mean 8 d; 209). Nonresident floaters contest territories unequally; some resident males receive many intrusions and others almost none, but intrusion rate does not correlate with any known attribute of the owners (209).
Relatively little is known about territoriality in western populations that nest in natural cavities. In subspecies hesperia in Arizona, males were reported to defend much larger areas, in some cases controlling extra nest cavities up to 115 m away; territories have an average radius of 20–30 m from the chosen nest cavity (13). Recent studies, however, have documented multiple cases of two pairs occupying two cavities in a single saguaro (DAA; J. MacFarland, personal communication) indicating greater territorial flexibility . The large territory size ensures that additional cavities will be available to a pair should the chosen saguaro blow over, as sometimes happens. Defense of extra cavities by saguaro-nesting birds suggests that territorial defense in this species is not entirely an artifact caused by the high density of nest compartments found in birdhouses. However, eastern populations still defend substantially more nest cavities than those using natural nest sites in Arizona (13, CRB) or clustered single nest boxes in British Columbia (B. Cousens, personal communication); birdhouse design probably contributes to this difference. Houses built with rooms that open onto a common porch are much easier for Purple Martins to defend by simply walking from room to room. When the porch is physically divided by a barrier, and thus adjacent rooms are not visible from the same porch (210, 211), territorial defense is reduced, possibly because of the extra cost to males of flying from room to room. For this reason, dividing porches can increase the total number of martins occupying a house (see Conservation and Management: Management).
Subspecies arboricola nesting in clustered single nest boxes < 0.5 m apart in British Columbia may defend one to several (2–3) nest boxes at the start of the season, but only rarely more than one by the time they are feeding young (and when many yearling pairs are incubating eggs); most defended nest boxes early in the season are relinquished to later-arriving adult and subadult pairs (B. Cousens, personal communication). Western birds nesting in box-girder bridges in central California readily nest in adjacent holes located 5 m apart (101), and show territorial defense only for the occupied nest hole (DAA).
Mating System and Operational Sex Ratio
Usually socially monogamous—one male and one female tending a nest. Social polygyny occurs at low frequency (< 5% of nestings), both in eastern populations (212, 213, 208, 214) and in P. s. hesperia in Arizona (13). Polygyny happens when one female allows a second to settle on a male's multi-compartment territory before he has relinquished part of it. In some cases, males defend a second female for several days, then abandon her to the intrusions of another male. When males maintain both mates, they provide parental assistance to both females, although generally more to the first female (213). Male may defend multiple nest cavities to enhance his chances of securing a second mate (208, 13), although the low observed incidence of polygyny suggests that other factors may be more important in causing territorial defense.
Adult sex ratio at two breeding colonies (202–240 pairs per year) in northwestern Pennsylvania was slightly male-biased (0.58; ST).
Courtship, Copulation, and Pair Bond
Unpaired males advertise territory ownership to prospective females through the Claiming-Reclaiming display (171): male flies from a birdhouse or tree/cactus containing a nest cavity, sails in a wide arc as much as 0.8 km in radius, and abruptly returns to the nest site, terminating the flight in a steep dive. As he lands, the bird enters the nest compartment, turns around, projects his head from the hole, and sings (171). Often, this behavior attracts the notice of a passing female, who follows the male to the nest site. This Claiming-Reclaiming display is used only in mate attraction, and males engage in it only if a female is present.
Pair bond forms relatively quickly once the female chooses a male and his associated nest cavity. Females seem to evaluate nest sites, rather than males per se (171), and occasionally settle at a site that has no attendant male (CRB). Once the same female returns to a male's nest site repeatedly over a period of several hours, he begins to treat her as a mate and may begin escorting her, at least briefly, when she leaves the nest site. After several days, male and female often sleep in the same compartment (202) and greet each other with vocalizations when reuniting. Both sexes clearly treat other members of the opposite sex differently from how they treat their mate (see Behavior: Agonistic Behavior). In the eastern Purple Martin, a male and female only occasionally arrive at a nest site already paired, with the male escorting the female and keeping her away from occupied territories. However, pairs' arriving together occurs relatively commonly in British Columbia (B. Cousens, personal communication). There is no evidence that pairs remain together after fall migration or spend the winter together, and members of a pair from the previous year do not arrive synchronously in the spring (215). See Johnston and Hardy (171) for further details on pair formation.
A poorly understood display that occurs during pair formation is the Stooped-Submissive Posture (171, 216). In this display, a male flies with its upper back humped, the head lowered, and the tail held low; the rectrices are abnormally constricted so that the tail resembles a tapered spine. Flight is labored and seems to lack a martin's typical coordination. Some birds stay in this posture for up to 30 min; only males engage in it. This posture is related to sexual behavior and seems to occur when martins are attempting copulations (216), possibly extra-pair copulations.
Extent of mate-guarding varies in the Purple Martin. In Maryland, some males escorted their mate on 100% of their forays from the nest site, while others never accompanied their mate (217). In Arizona, male P. s. hesperia rarely guarded their mate (13). When escorting a female, the male stays close to her, often attacking her if she visits another male's territory or lunging at and trying to displace a male that comes close on a perching site. Pairs frequently feed together early in the breeding season (CRB), probably a consequence of mate-guarding. While foraging, a male occasionally chases his mate for 15–40 s for unknown reasons, while the female vigorously tries to elude him; when the chase ends, both resume normal flight (218). This behavior may allow the male to assess how close his mate is to egg-laying, perhaps by observing her flight characteristics, which are influenced by the heavy body mass associated with laying (219).
Mate-guarding ceases after females begin incubation; but in rare instances when second broods are reared (220, 221), males resume mate-guarding in early stages of a second attempt. Yearling males tend to guard mates more than adult males do, possibly because yearlings are more likely to be cuckolded (197, 222). Purple Martin is unusual among swallows in showing such high variance in mate-guarding; in other colonial species, almost all males tend to guard mates (223) or none do (127).
Pairs are rarely observed copulating; almost none were seen copulating in ≥ 14 yr of observation at two northern Texas colonies (CRB; T. Dellinger, personal communication). Within-pair copulation probably occurs inside the nest cavities (217), as in other swallows (127), and often may happen at night (224, 33). P. s. hesperia pairs in Arizona, however, copulated in the open (13).
The pair bond dissolves after young fledge, although the same pair may visit their former nest site and exhibit post-breeding nest defense after their young become independent (195, CRB). They tolerate each other, even as they continue to defend the territory from other Purple Martins. However, any re-pairings of the same birds the next year appear to be coincidental (124, 129).
Extra-Pair Mating Behavior/Paternity
Occurs among birds gathering nesting material. Has been observed in Texas, Maryland, and Pennsylvania (218, 217, 225), but not in P. s. hesperia of Arizona (13). Females are attacked by other males while on the ground; a male mounts a female, spreads his wings laterally, and grasps her nape with his beak. Females peck at attacker or try to flee. If a female's mate observes the extra-pair copulation (EPC), he will try to disrupt it. Sometimes several males attack a female simultaneously, and they may chase her for up to 60 s while she tries to elude them by plunging into tree branches or landing on a martin house (218). However, females are thought to sometimes be willing participants in EPCs (222), seeking to achieve matings with males of higher genetic quality than their social mate. Most EPCs seem to be obtained by neighboring males, and adult males frequently attempt EPCs with mates of yearling males living nearby (197, 222). Earlier-arriving older males generally pair socially with older females that similarly arrive early in the breeding season, and then use a dawn song possibly to attract yearling birds to the colony and cuckold the young males (197, 222)
DNA fingerprinting analysis of paternity at a single colony in Maryland showed that EPCs often led to extra-pair fertilizations (EPFs), and that male age class was a powerful predictor of male genetic mating success (197). Forty-seven percent of yearling males (n = 17) had some offspring unrelated to them (43% of offspring produced in the yearling males' nests resulted from EPCs, n = 53), versus only 8% of adult males (n = 24; 4% of adult males' offspring resulted from EPCs [n = 85]; 222). Opportunities to cuckold neighboring males increase realized reproductive success for older males; in Maryland, an average 3.6 additional fertilized eggs/male were accrued through EPCs (197). In Pennsylvania, males who gained EPFs sired 1.07 and 1.35 additional offspring per bird for yearling and adult males, respectively (225), with 46% of nests containing extra-pair young (EPY) and 22% of all offspring sired by extra-pair males (n = 297 nests; 225). For nests containing at least one EPY, the proportion of EPY per nest averaged 0.53 ± 0.03 (n = 137 nests). Ten polymorphic microsatellite loci, each with 2–19 alleles, have been developed for improved resolution in genetic studies involving Purple Martin (226). The frequency of EPY based on the HrU10 microsatellite locus was 26% (n = 22 nests; 227). Multiple extra-pair sires in nests are common (225).
Variance among males in achieving successful EPCs is high. In Maryland, a single male was responsible for 87.5% of all assignable EPFs in one year, and another male for 83.3% in another year (222, see also 225). Whether a female engages in EPCs seems to be determined by either how well her mate guards her, thus influencing her encounters with other males, or by a female's willingness to participate in EPC attempts by extra-pair males. Bigger males seem better able to prevent their mates from engaging in EPCs than smaller males (197, 222). Male genetic mating success increases linearly up to 3 years of age and then levels off, with no clear evidence of senescence (225).
Brood Parasitism of Conspecifics
No evidence of intraspecific brood parasitism in hundreds of nest checks in a Texas study (no cases of ≥ 2 eggs laid in a single day; CRB). Based on the appearance of more than one egg laid per 24-h period, 3.6% of nests (n = 394) in Oklahoma had evidence of parasitism by a conspecific; 2 of the 7 nests had evidence of multiple parasitism (228). We do not know whether intraspecific brood parasitism is perpetrated by colony residents or non-resident, transient individuals.
Brood Parasitism of Other Species
Purple Martin is not known to parasitize other species.
Social and Interspecific Behavior
Degree of Sociality
Highly social during the nonbreeding season, concentrating in enormous pre-migratory roosts and assembling in huge groups to sleep at night in town squares of Brazilian cities (see Habitat in Nonbreeding Range). Late-summer roosts commonly contain several thousand birds; one in South Carolina was estimated at > 700,000 birds (159; see Movements and Migration: Migratory Behavior). Birds probably migrate in groups. Communal roosts persist through the entire nesting period in Arizona (155), suggesting that nonbreeders maintain their gregariousness throughout the summer. Also reported roosting throughout the summer in Louisiana (154).. Purple Martins from multiple colonies in California moved 3.1–7.4 km to aggregate in the late afternoon throughout the nesting season, including adults with dependent juveniles (160). Little else is known about spring and early-summer roosting for either subspecies.
For nesting, depends on cavities that they do not construct; thus the martin's spatial distribution during breeding is determined largely by the distribution of its nest sites. Before eastern forests were cut and these birds shifted to birdhouses and gourds, eastern subspecies probably at times occupied huge dead snags riddled with woodpecker holes, a situation resembling present birdhouse-nesting (196). Solitary nesting also was common among ancestral populations (13). In rare instances of eastern birds nesting in natural sites recently, records have been of single pairs in Florida (229), Alabama (R. K. Crawford, personal communication), Texas (152), and of 2 pairs nesting together in Mississippi (J. Jackson, personal communication). Overall, eastern birds have probably shifted to higher-density breeding as consequence of nesting in birdhouses and gourds; some martin houses contain as many as 620 compartments (230), and an enormous “martin tower” at Lake Charles, Louisiana, contained 46 separate houses and more than 1,100 total rooms (photo in 136). Most martin colonies in eastern North America, however, contain < 12 pairs, because 12-compartment houses are most frequently installed and most people erect only one house (CRB).
Colony size can increase with the installation of more houses, suggesting that the birds are limited by nesting opportunities, but colonies > 35 pairs are unusual even with multiple houses (CRB). Colonies > 100 nests are noteworthy, and the largest reported colony had about 300 pairs nesting under boulders and in rock crevices on islands in Minnesota up to 1930 (231), although the accuracy of this colony size estimate is unknown.
In western populations, many birds nest largely solitarily or in small groups in natural or woodpecker-made cavities in trees or cacti (13, 108, CRB), although they assemble in large late-summer roosts even in these areas (232). In Arizona, no more than a single pair of Progne subis hesperia was found per saguaro, and occupied nest sites were at least 100 m apart; median nearest neighbor distances in two years were 325 m and 238 m, respectively (13), although recent studies have in areas with sparser concentrations of saguaro cacti, nine of 110 saguaro cacti supported two pairs nesting in them (J. MacFarland, DAA). Largest natural colony of western Progne subis arboricola had > 100 pairs nesting in crevices of a sandstone cliff in 1996 near Flagstaff, Arizona (E. Morton, personal communication). In California where nest cavities in bridges are abundant, nesting individuals appear to aggregate to nest in adjacent holes despite colony populations dramatically declining (101, 160, 93, 94).
The extent to which the Purple Martin is "colonial" and the benefits it derives from nesting together, if any, is controversial. The species has been described as engaging in coordinated group behavior (171, 13), implying potential advantages to forming and maintaining colonies or flocks. Although these birds are clearly attracted to conspecifics during the nonbreeding season and at times while nesting, their level of social interaction during the breeding season is less than that of the more colonial North American swallows such as the Cliff Swallow (Petrochelidon pyrrhonota) (127). Except for the tendency of neighboring pairs in western subspecies (hesperia and aboricola) to join together in mobbing predators (13, 52, 94), little coordinated social behavior seems to occur during breeding, and group defense against predators can simply represent a summation of individual-level responses (233, 127). No clear potential advantages of group living like those seen in more social species (e.g.,127) have been identified (234): birds in larger colonies have not been shown to be more successful than those in smaller colonies or those nesting solitarily. For example, a range-wide species distribution model showed no relationship between fecundity and habitat suitability as partly defined by the number of breeding pairs (95).
Clutch size or the number of nestlings surviving to day 15 is not known to vary with the size of a colony in the single study that has measured reproductive success in relation to colony size (234). Disadvantages of coloniality in the eastern subspecies include increased agonistic interactions, increased need for males to mate-guard, and increased exposure to mites, all that vary directly with colony size (234). Ectoparasitic mites, however, only affected nestling body mass adversely for large broods in an Oklahoma study (n = 72 nests; 234). Increased risk of transmission of other parasites and resource competition should apply to birds forced into high-density nesting in birdhouses if colonies reach a large enough size for these effects to be expressed (235). Martin behavior largely conforms to that expected for solitary species—e.g., high-risk diving at predators (194, 13) and parental inability to recognize fledglings (193, 236).
Colonies most likely form simply in response to localized abundance of limited nest sites (234, 95). Morton et al. (197) proposed that colonies form largely to benefit adult males, who recruit yearling males whose mates provide the older males with potential extra-pair copulation (EPC) partners. But this cannot explain the evolution of coloniality, because it does not explain why yearling males would join a colony in the first place (127,235). Another hypothesis that coloniality evolved because females seek to cluster around males with whom they can seek EPCs (222) requires identifying the advantages of EPCs to female martins, but there seem to be none. Individuals sired by extra-pair males had similar first-year survival to that of birds sired within a pair, with no other obvious benefits of EPCs to females (237). Evolution of EPC strategies seems more likely a consequence of the birds' having already formed colonies in response to a clumped distribution of nest sites.
Rarely reported. Only case well described is the observation of a male in South Carolina obviously playing with a piece of straw, dropping it from a height of about 30 m, diving underneath it close to the ground to catch it while flying upward, regaining altitude, then repeating the performance several times (238). The bird went through remarkable flight movements, repeatedly soaring upside down, nose-diving, and once climbing upward to stall out, hanging motionless, and then gliding backward, tail first. The display was terminated finally when the bird missed catching the straw, which fell to the ground (238).
Nonpredatory Interspecific Interactions
Principally with nest-site competitors, most frequently House Sparrow and European Starling (239, 240, 241, 242). These species are serious competitors, usurping suitable nest cavities and rendering them unsuitable for martin use by stuffing them full of grass. The more aggressive starlings generally dominate martins in direct behavioral interactions, sometimes trapping martins in birdhouse compartments and killing them (240, 242). Purple Martins threatens starlings by gaping at them, but only occasionally chase or dive at them like predators.
Purple Martins frequently win fights with House Sparrows, but the sparrows' persistence usually allows them to take over compartments. Once established at a site, and if there before any martins arrive, House Sparrows can deter martin settlement by aggressively chasing martins and filling nest compartments with so much grass that martins cannot enter. Purple Martins can occasionally get trapped in the densely packed nesting material within a House Sparrow nest and die (243).
Both European Starlings and House Sparrows usurp active martin nests, throwing out eggs or small young. Starlings often drag part of a martin's nest out of a nest cavity. Sparrows seeking nest sites may intrude into martin nests that are left temporarily unattended, and although they do not usurp the nests, they throw out or puncture martin eggs. A single House Sparrow can destroy ≥ 15 martin eggs in a few minutes, especially during cool weather when martins leave their nests unguarded to forage (CRB). Martins show little ability to compete effectively with starlings and sparrows; see Jackson and Tate (239) and Brown (240) for further details on nest-site preferences of these species and their behavioral interactions.
In the West, Purple Martin populations were reduced in numbers and eliminated from large areas of their range after the European Starling's arrival and population increase, including in southern California and the Central Valley of California (90, 101, 51), the Williamette Valley of Oregon (102), much of Washington (50), and British Columbia (52). In the last Central Valley colonies to persist, starlings are not significant competitors because nest sites are superabundant, and starlings begin nesting earlier than martins and usually select holes on the outer edge of bridges, allowing martins to occupy other holes without contesting with starlings.
Native North American species that occasionally nest in Purple Martin houses in the East include Great Crested Flycatcher (Myiarchus crinitus), Northern Rough-winged Swallow (Stelgidopteryx serripennis), Tree Swallow, House Wren (Troglodytes aedon), Eastern Bluebird (Sialia sialis), and House Finch (Haemorhous mexicanus); 244, 20, 245, 246, CRB). House Wren may destroy martin eggs, but the other species are usually dominated by martins and have little effect on martin use of a nest site. Eastern Screech-Owl (Megascops asio) occasionally roost in martin houses (247). In rare instances, late-arriving yearling martins may kill small nestlings of conspecifics or of other species (such as the Great Crested Flycatcher, for example) in apparent attempts to take over compartments (248, 249). In the West, Purple Martins may displace Tree Swallow, Violet-green Swallow (Tachycineta thalassina), Northern Rough-winged Swallow, White-throated Swift (Aeronautes saxatalis), and occasionally European Starling from nest sites (250, 52, 94; S. Kostka, personal communication). In montane and desert regions of western North America, Purple Martins interact mostly with woodpeckers over access to cavities. In the Chiricahua Mountains, Arizona, martins competed with Northern Flicker (Colaptes auratus) and Acorn Woodpecker (Melanerpes formicivorus); martins dived at the woodpeckers, and the woodpeckers chased the martins from dead snags (CRB). P. s. hesperia sometimes nests in same saguaro as Northern Flicker, Gila Woodpecker (Melanerpes uropygialis), and Brown-crested Flycatcher (Myiarchus tyrannulus; 13).
Perhaps as a consequence of the close nesting of different species made possible by birdhouses or bridge nesting, several instances of misdirected foster feeding involving Purple Martins have been observed: a European Starling feeding young Purple Martins (251), Purple Martins feeding young European Starlings (252), Purple Martins feeding young House Sparrows (253, 254), and House Sparrows feeding young Purple Martins (CRB). Purple Martin and White-throated Swift occasionally occupy the same bridge nesting chambers (5 of 66 nest sites checked), and young swifts have occasionally been observed in martin nests (255). Such cases probably result from the wandering of swift young from swift nests into martin nests.
Purple Martin flocks during migration and overwintering periods, especially to roost at night, with Gray-breasted Martin (Progne chalybea), Southern Martin (Progne elegans), Brown-chested Martin (Progne tapera), and Caribbean Martin (Progne dominicensis) (75). Some South American roosts contain higher percentages of these species than of Purple Martin; behavioral interactions among the various Progne have not been studied.
Kinds of Predators
Primarily birds and snakes, occasionally domestic cat (Felis domesticus), raccoon (Procyon lotor), and squirrels (149, 20, 256, 257, 258, 259, 260, 261, 262). Owls (128) and possibly snakes are likely the most common and significant predators, taking both adult and nestling Purple Martins. Documented predators include Great Blue Heron (Ardea herodias), Mississippi Kite (Ictinia mississippiensis), Sharp-shinned Hawk (Accipter striatus), Cooper's Hawk (Accipiter cooperi), Swainson's Hawk (Buteo swainsoni), Peregrine Falcon (Falco peregrinus), Merlin (Falco columbarius), American Kestrel (Falco sparverius), Greater Roadrunner (Geococcyx californianus), Eastern Screech-Owl (Megascopus asio), Barred Owl (Strix varia), Great Horned Owl (Bubo virginianus), Blue Jay (Cyanocitta cristata), Fish Crow (Corvus ossifragus), Black-billed Magpie (Pica hudsonia), and Great-tailed Grackle (Quiscalus mexicanus) (263, 52, 264, 94). Little is known about predators on the overwintering range.
Manner of Predation
Owls, roadrunners, jays, crows, and magpies alight on birdhouses and pull adults or young out of nest cavities—owls usually at night. An owl can wipe out entire colonies, as can accipiters. Snakes, squirrels, and raccoons climb birdhouse poles to reach nests, then pull out adults and young or enter cavities to eat eggs. House cats catch adults that fall to the ground during fights. Cats and hawks capture martins on the ground and on perches when they come to the ground to gather nesting material (265, 94). American Crow (Corvus brachyrhychus) scavenges young that fall from nest holes. Kites, accipiters, falcons, and Great Blue Heron are known to catch birds in flight.
Response to Predators
Typical response to most predators is alarm-calling with Zweet or Zwrack calls (see Sounds and Vocal Behavior: Vocalizations), while usually milling or perching above predator. If predators (including people) approach a nest closely, owners of a threatened nest dive at the predator, sometimes swooping within a few centimeters. These close dives are risky (194). No coordinated, cooperative defense by colony members is reported in eastern martins; each bird responds individually (CRB). In California and British Columbia, groups of martins, often including most of the nesting individuals at a colony site, routinely harass hawks and falcons that approach nesting colonies (263, 94). In subspecies hesperia, neighboring pairs assembled in a “mob” when models of predators were presented at a nest, but only the owners of the threatened nest actively dived at the predator (13). When raptors fly over a nest site, martins there flush and fly up to the altitude of the predator, as do other swallows (127).
Although Purple Martins occasionally have been seen attacking soaring raptors in Mississippi (K. Kimmerle, personal communication), martins generally do not chase predators or even approach them closely unless predators are within a few meters of a nest containing young (CRB).