Natural History
This material is excerpted from Brown, C. R. et al., 2017, Cliff Swallow (Petrochelidon pyrrhonota) in The Birds of North America. Citations are provided on the References page, which also serves as a comprehensive bibliography on the species.
The Cliff Swallow is one of the most social land birds of North America. These birds typically nest in large colonies, and a single site may contain up to 6,000 active nests. Cliff Swallows originally were birds of the western mountains, where they still nest underneath horizontal rock ledges on the sides of steep canyons in the foothills and lower elevations of the Sierra Nevada and Rocky and Cascade mountains. In the past 100 to 150 years, these birds have expanded their range across the Great Plains and into eastern North America, a range expansion coincident with the widespread construction of highway culverts, bridges, and buildings that provide abundant alternative nesting sites. New colonies continue to appear each year in areas where Cliff Swallows were previously unrecorded as breeders.
The Cliff Swallow was one of the first North American birds to be described. Although its discovery in Colorado is usually credited to Thomas Say on Stephen Long’s expedition to the Rocky Mountains in 1820 (James 1823), the bird and its colonial breeding habits were first mentioned by the Spaniard Silvestre Velez de Escalante in September 1776 when he encountered large numbers in the Wasatch Range of Utah (Coues 1899). This species is known to the public for its mythical return to Mission San Juan Capistrano on 19 March each year (Bruton 1975), with this legend serving as a metaphor for the regular occurrence of events.
The Cliff Swallow’s highly colonial life style has led to the evolution of some complex behavioral traits. For instance, Cliff Swallows brood-parasitize neighboring nests both by laying parasitic eggs and by moving eggs from their own nest into others (Brown 1984, Brown and Brown 1988c); they have a sophisticated vocal system for distinguishing their own young from the offspring of many other individuals within a colony (Beecher et al. 1985, Medvin et al. 1992, 1993); and they observe each other’s foraging success and learn from other colony residents the locations of food (Brown 1986). The Cliff Swallow’s social behavior during the breeding season has been studied extensively, and this species has figured prominently in our understanding of the evolution of coloniality in birds (Brown and Brown 1996, 2000a, Brown et al. 2016). Cliff Swallows have served as a model organism for studying rapid evolution in response to natural and anthropogenic environmental change (Brown and Brown 1998a, 2011, 2013, 2014) and the dynamics of fluctuating selection in the wild (Brown et al. 2013, 2016). In addition, the species is closely associated with an endemic vector-borne virus that has led to insights into how changes in hosts drive the evolution of different pathogen strains (Brown et al. 2009, 2012, O’Brien et al. 2011).
Breeding Range
The Cliff Swallow’s breeding range extends from w. and central Alaska, n. Yukon, n. Mackenzie, central Keewatin, n. Manitoba, n. Ontario, s. Quebec (including Anticosti I.), New Brunswick, Prince Edward I., and Novembera Scotia south to s. Alaska, n. Baja California (Godfrey 1986, Phillips 1986, Am. Ornithol. Union 1998), the Pacific slope of Mexico south to Sinaloa and along the Mexican plateau south to central Oaxaca (Howell and Webb 1995) and s. Texas, and east to the Atlantic and Gulf coasts of the U.S. Colonies have been found on the north side of the Brooks Range, with some within ~30 km of the Arctic Ocean, in ne. Alaska (Sage 1973, P. Goldman pers. comm.). Less common breeder in the Gulf states and the e.-central and ne. U.S., but the species is increasing in most of these areas, especially the Gulf states and the Ohio Valley.
Since at least 1980, there have been scattered reports of nesting activity on the winter range, primarily in Buenos Aires Province, Argentina (Petracci and Delhey 2004). Nests have been built, but no known cases of egg laying have been documented to date.
Winter Range
The winter range extends apparently from s. Brazil (São Paulo province) and possibly se. Paraguay south to s.-central Argentina. Distribution within this region is poorly known, although most birds apparently winter in lowlands along the Rio Paraná and Rio Uruguay north and northwest of Buenos Aires (Buenos Aires, Entre Rios, and Corrientes provinces of Argentina, and w. Uruguay; Olrog 1967, P. Burke pers. comm., A. Jaramillo pers. comm., V. Bowers pers. comm.). Wintering birds have also been recorded west to Tucumán province, Argentina. South of Buenos Aires, birds occasionally occur in large numbers but are more irregular than farther north (A. Jaramillo pers. comm.). Recorded several times as far south as Tierra del Fuego and occasionally in the Falkland I. Rare in Chile and the high Andes. The subspecies P. p. pyrrhonota winters in ne. Argentina, P. p. melanogaster in nw. Argentina; the wintering ranges for P. p. ganieri, P. p. tachina, and P. p. hypopolia are unknown (Fjeldså and Krabbe 1990). Small numbers are reported to winter occasionally with flocks of Barn Swallows in the Pacific slope lowlands of Panama (Ridgely 1976). Casual in Barbados, St. Kitts, Guadeloupe, St. Lucia, and Dominica, mostly in winter but a few records in spring and fall (Bond 1971, Am. Ornithol. Union 1998, Feldmann et al. 1999). Stragglers have been reported in December in the Imperial Valley of California (Grinnell and Miller 1944), along the Lower Colorado River in Arizona (Phillips et al. 1964), and on the Pacific Coast (north to Vancouver) and the Gulf Coast on Christmas Bird Counts. Report of Cliff Swallows being “common” in the Salton Sea region in winter (van Rossem 1911) is likely erroneous.
Outside America
Accidental on Wrangel I., Siberia, and s. Greenland (Am. Ornithol. Union 1998). At least 6 records for the British Isles (Rogers 1997, 1998).
Historical Changes
This species’ breeding range has been influenced heavily during the last 100–150 years by widespread construction of bridges, buildings, and culverts, which provide nesting sites in areas formerly uninhabited, and by the introduction of House Sparrows (Passer domesticus) which usurp nests. In the ne. U.S., where Cliff Swallows were probably never common, this swallow began to increase in the early to mid-1800s as land was cleared and more buildings constructed (Bent 1942). With the introduction of the House Sparrow in the late 1800s, and its usurpation of swallow nests, Cliff Swallow numbers began to decline in the ne. U.S. and remain low today (Forbush 1929, Silver 1993, 1995).
In the se. U.S., the breeding range has expanded south and east during the last 50-75 years with new colonies found each year. A major eastward range expansion has occurred in Tennessee since the 1930s (Alsop 1981). Breeding began in Alabama in 1951 (Imhof 1976), Georgia in 1965 (Dopson and Peake 1967), Florida in 1975 (Sykes 1979, Lewis and McNair 1998), the Carolina piedmont in 1965 (McConnell 1981) with a major expansion in the 1980’s (McNair 2013), and Louisiana in 1980 (Viers 1991). Breeders reached the Mississippi coast by 1986 (Spence and Toups 1986).
The breeding range has also expanded elsewhere east of the Great Plains. Breeders have increased since the mid-1970s in the coastal plain of Maryland (Patterson 1981) and Virginia (Watts et al. 1996), since the early 1980s in W. Virginia (Igou 1986), and since the late 1980s in s. Illinois (Robinson 1989). Breeding was detected in Delaware in 1993 (Ednie 1994). The species is now breeding in larger numbers in the eastern Great Plains (e.g., in e. Oklahoma, e. Kansas, e. Nebraska) than 50 years ago (CRB, MBB) and has expanded into s. Arkansas (Tumlison 2009). Only in the ne. U.S. does nest usurpation by House Sparrows (or forest cover) appear to be limiting breeding-range expansion.
Geographic Variation
Body size varies clinally, with birds in the north (Alaska and w. Canada) larger than those in the south (s. Arizona and n. Mexico). Wing and tail length vary the most among populations, whereas bill size and tarsus length vary little (Behle 1976). Southern birds have the forehead patch darker (most chestnut, less buff or white). Throat color varies from dark chestnut in the north to pale tawny in the south. Rump patches and underparts tend to be whiter in the north, whereas to the south birds often have the flanks tinged with rust or rufous (Behle 1976).
Subspecies
Four subspecies, following Phillips (1986), diagnosed chiefly on the coloration and pattern of the plumage on the head and on body size. Distinguishing characters listed below are for an adult with unworn plumage.
P. p. pyrrhonota (Vieillot, 1807). Includes Hirundo albifrons Rafinesque, 1822; P. lunifrons (Say, 1823); Hirundo opifex De Witt Clinton, 1824; H. republicana Audubon, 1824; P. p. hypopolia Oberholser, 1920; and P. p. aprilophata Oberholser, 1932 (see Browning 1992). Breeds across n. North America, from n. Alaska east through central Canada to the Maritimes and ne. United States and south; in the West, to nw. Baja California, n. Nevada, and central Colorado; and in the East, from the Ohio Valley, Chesapeake Bay, and (less commonly) to se. United States and the Gulf coast; winters south to central Argentina [type locality = Paraguay]. Throat and cheeks dark chestnut; forehead pale buff or whitish; rump chestnut; large (wing chord > 105 mm).
P. p. ganieri Phillips, 1986 (see Browning 1990). Breeds west of the Appalachians from (probably) s. Oklahoma south through central Texas to s. Texas and east to w.-central Tennessee [type locality = Kerr Co., Texas]; winter range unknown. Like P. p. pyrrhonota, but forehead darker buff; averages smaller (wing chord 102–108 mm; Phillips 1986).
P. p. tachina Oberholser, 1903. Breeds from central California east through s. Nevada to sw. Utah and south through the Lower Colorado River valley to ne. Baja California and east through central Arizona and New Mexico to sw. Texas [type locality = Langtry, Val Verde Co., Texas]; apparently winters south to n. Argentina (Phillips 1986). Similar to P. p. pyrrhonota, but throat and cheeks paler, forehead brownish buff, and chestnut rump darker; averages markedly smaller overall (wing chord < 108 mm).
P. p. melanogaster (Swainson, 1827). Includes Hirundo coronata Lichtenstein, 1831; P. p. swainsoni Sclater, 1858; and P. p. minima van Rossem and Hachisuka, 1938. Breeds from extreme se. Arizona and sw. New Mexico south over the Mexican plateau to Oaxaca and the Pacific plains to Nayarit [type locality = central Mexican Plateau]. Similar to P. p. tachina, but forehead dark chestnut. The ranges of tachina and melanogasteroverlap in parts of se. Arizona, and both were found in the same colony near Fairbank, AZ (Jeter 1959).
Related Species
Monophyly of the Hirundinidae, the swallows and martins, has never been questioned, but relationships within the subfamily Hirundininae—a subfamily that includes all but the two species of river martins (Pseudochelidoninae)—have been difficult to tease apart. On the basis of behavior and basic morphology (i.e., ignoring sexually selected traits such as elongated rectrices), the genus Hirundo formerly included a large group of swallows that generally are blue dorsally, white and rufescent ventrally, and build their nests out of mud (Turner and Rose 1989). This treatment changed with the advent of molecular phylogenetics, such that Hirundo was split into three core genera: Hirundo sensu stricto (which includes the familiar Barn Swallow, H. rustica), Cecropis (the striated or red-rumped swallows), and Petrochelidon (the cliff swallows), although this taxonomic split had been recognized earlier by some authorities (e.g., Ridgway 1904, Peters 1960). The genera Delichon (the house martins) and Ptyonoprogne (the crag martins) round out this mud-nester clade (Sheldon et al. 2005). Despite Petrochelidon and Cecropis being lumped with Hirundo in various earlier works, it is actually Ptyonoprogne that is sister to Hirundo, whereas Petrochelidon and Cecropis are sister to each other and together are closest to Delichon (Sheldon et al. 2005).
The nearest living relative in North America of P. pyrrhonota is P. fulva, the Cave Swallow (Sheldon et al. 2005). Formerly little breeding sympatry, but these species have come into contact in Texas as bridges and highway culverts have been constructed. Although there are no known cases of Cliff x Cave hybridization, an extralimital Cave Swallow paired and attended a nest with a Cliff Swallow in Tucson, AZ, in 1984 (Huels 1985); it is unknown whether the offspring produced were hybrids. Hybrid Cliff x Barn swallows have been reported from Pennsylvania (Trotter 1878), New York (Wood et al. 2011), Louisiana (Dittmann and Cardiff 2002), sw. Texas (Mearns 1902), Nebraska (Brown and Page 2015), California (Rogers and Jaramillo 2002) and e. Washington (P. Stoddard pers. comm.), along with internet photos of apparent hybrids from Colorado and Vermont. There is 1 specimen of a hybrid Cliff x Tree (Tachycineta bicolor) swallow (Chapman 1902) from Massachusetts.
Repeated hybridization between Cave and Barn swallows in s. Texas (Martin 1980) was the basis for merging Petrochelidon into Hirundo in 1982 (Am. Ornithol. Union 1982). However, Petrochelidon, a more phylogenetically derived genus than Hirundo (Sheldon and Winkler 1993, Sheldon et al. 2005), represents a distinct group of 7 species of red-rumped, retort-nesting, colonial species distributed worldwide (pyrrhonota, fulva, preussi, rufigula, spilodera, fluvicola, and ariel). Petrochelidon was re-established for the Cliff Swallow in 1997 (Am. Ornithol. Union 1997).
Cliff Swallows migrate from the breeding range to the winter range via Mexico, the Central American isthmus, and n. South America, staying east of the Andes. Apparent migrants are also recorded rarely in Bahama I., Cuba, the Lesser Antilles, and Virgin I. (Am. Ornithol. Union 1998, Feldmann et al. 1999). It is not known whether any intraseasonal movement occurs on the wintering range, although the species probably is nomadic at that time. Most migrants presumably follow the Central American isthmus between North and South America. Migration in both directions seems leisurely and spans several months; there are fewer observations of spring than of fall migrants.
Spring Migration
Birds begin leaving the wintering range in early February, although some individuals are still present in April (Hudson 1920). Thousands were observed migrating through Panama on 24 February 1994, passing continuously for at least 1 h in a narrow front about 0.4 km wide at a rate of about 150 birds/min (E. Morton pers. comm.). Some birds pass through Colombia and Panama as late as early May (Ridgely 1976, Hilty and Brown 1986, Paynter 1995). Migrants are seen in Sinaloa, Mexico, as early as 14 February and commonly in Oaxaca by mid-March, with migration in Mexico lasting until at least 30 May (Phillips 1986).
Birds first arrive in s. California in early February (rarely in late June; Small 1994), Arizona usually in early March (rarely as early as 9 February; Phillips 1986), Texas in early March (rarely 24 February; Oberholser 1974), Arkansas in late March (Tumlison 2009), Kansas in mid-April (rarely 26 March; Thompson et al. 2011), Nebraska in mid-April (usually 16–18 April but an extreme date of 22 March; CRB, MBB, Sharpe et al. 2001), Illinois in early April (rarely 29 March; Graber et al. 1972), Minnesota in late April (rarely 13 April; Roberts 1936), Idaho in early April (Burleigh 1972), Massachusetts in mid-April (rarely 9 March; Veit and Petersen 1993), the Yukon in late April (earliest 20 April; Sinclair et al. 2003), and Alaska in mid-May (rarely 7 May; Gabrielson and Lincoln 1959). For other first arrival dates, see Bent (1942). The peak period of migration in the Texas panhandle is 28 April to 18 May, with the earliest record being 9 April and the latest 2 June (C. D. Littlefield pers. comm.). More northerly populations generally arrive later than more southerly ones; however, P. p. melanogaster typically arrives on its se. Arizona breeding range (at the time of the summer monsoon) 6–8 wk later than P. p.tachina and P. p. pyrrhonota in n. Arizona (Phillips et al. 1964). The first birds to arrive in a breeding area usually do so in groups (Shaw 1991, CRB, MBB), possibly reflecting associations of birds that have been together since the previous summer. Arrival date is subject to selection: unusual cold weather in spring in Nebraska that caused the deaths of large numbers of birds (Brown and Brown 1998a) resulted in later arrivals and colony initiation dates in years following the mortality (Brown and Brown 2000b).
Fall Migration
Begins when nestlings fledge and as colony sites are vacated, so departure can be staggered within a locale and quite variable between years. In sw. Nebraska, birds begin departing in early July in some years; most are gone by early August, although some late nesters may not leave until late August (CRB, MBB). Peak of migration in U.S. apparently is in August and early September when flocks of thousands may be seen moving south (e.g., Bent 1942). In the Texas panhandle, the peak of migrant passage was 24 July to 30 August, with extreme dates of 18 June and 3 October (C. D. Littlefield pers. comm.). Migrants recorded in Mexico from 24 July (Veracruz) to 5 November (also Veracruz); in Costa Rica from 29 July to 30 November (Phillips 1986); in Panama from 29 July to late October (Ridgley 1976); in Colombia from early September to mid-October, with the largest numbers in mid-September (Hilty and Brown 1986; Bayly et al. 2014); in Venezuela from early August to October (Meyer de Schauensee and Phelps 1978); in Bolivia from early October to early December, with the bulk of birds 23 November to 5 December (Parker and Rowlett 1984); and in Paraguay 15 October to 5 December (Lowen et al. 1997, J. Unger pers. comm.), with an extreme date of 2 September (Hayes 1995). Hundreds of thousands of Cliff Swallows pass through Mucubaji Pass in the Merida Andes, nw. Venezuela, from late August to early November, with peak passage in mid October (estimated 150 birds/min; C. Rengifo pers. comm.). Some birds are in Argentina, presumably on the winter range, by October (Pereyra 1938) and continue to arrive through December. Migration dates divided by subspecies are given in Phillips (1986).
Migratory Behavior
Usually seen in groups of up to several hundred, occasionally several thousand, birds. Probably exclusively diurnal migrants, foraging as they move. Cool and rainy weather forces spring migrants in Nebraska to concentrate over lakes, ponds, and rivers where they may spend several days foraging low over the water surface. These concentrations can sometimes exceed 5,000 birds in a 1- to 2-km stretch of lake or river (CRB, MBB). Often seen in flocks with other swallows. Sleeps in marsh vegetation during migration (Kirby 1978).
Breeding Range
Historically inhabited open canyons, foothills, escarpments, and river valleys that offered a vertical cliff face with a horizontal overhang for nest attachment. With the present use of artificial nesting structures such as bridges and buildings, the species is now found in a wide variety of habitats: grasslands, towns, broken forest, riparian edge. Avoids heavy forest, desert, and alpine areas. Most colony sites are located near open fields or pastures where the birds forage, and a water source is often nearby. Proximity to a mud source (for nest-building) is often cited as a breeding-habitat requirement (Emlen 1941, 1952), although some colonies are located several kilometers from the nearest mud supply (Coffey 1980, CRB, MBB). The species probably has more specific habitat requirements that are presently unknown, as Cliff Swallows are strangely absent from certain localities within their breeding range that would seem to offer appropriate nesting sites (Phillips et al. 1964, CRB, MBB).
Altitudinal range extends from the Lower Sonoran through the Transition zones, from sea level to about 2,770 m. Colonies occur rarely to 3,000 m; highest known to us is one of 50–100 nests at 3,200 m on Rendezvous Mtn., Teton Range, WY (CRB, MBB). There are no clear differences among subspecies in preferred breeding habitat.
Spring and Fall Migration
Often seen in savannahs and near bodies of water, but probably migrates through (over) a wide variety of habitats. Thousands of migrating birds assemble and roost in cornfields near playa lakes in s.-central Nebraska and in the Oklahoma and Texas panhandles during fall (C. D. Littlefield pers. comm., CRB). Commonly seen in coastal lowlands in Panama (Ridgley 1976). Probably migrates mostly at elevations below 1,000 m, but transients recorded to 3,800 m in South America (Meyer de Schauensee and Phelps 1978, Ridgley and Tudor 1989). Migrants concentrate over water surfaces and marshes when poor weather reduces abundance of flying insects (see Migration: migratory behavior).
Winter Range
Occurs in grasslands, agricultural areas, and in marshes. A roost of up to 50,000 birds was found in wetlands along the Rio Paraná in Argentina, about 50 km north of Buenos Aires (P. Burke pers. comm.). Another very large roost was in the Iberá Wetlands, Corrientes province, Argentina, about 700 km north of Buenos Aires (V. Bowers pers. comm.). Birds were sleeping in marsh vegetation, fanning out to forage over surrounding areas during the day, perhaps up to 20 km distant from the roost sites. Smaller roosts have been seen in marshes in Entre Rios province, Argentina (A. Jaramillo pers. comm.).
Main Foods Taken
Flying insects at all times of the year. Occasional pieces of seeds are found in stomachs (Beal 1918), but these represent either accidental ingestion or use as grit. Birds sometimes pick up small bits of gravel, probably to aid digestion of insect exoskeletons. A report of 2 birds with stomachs full of juniper (Juniperus) berries (Beal 1918) was likely based on misidentifications of Tree Swallows.
Microhabitat for Foraging
Feeds above the ground at altitudes of 50 m or more. Seems to prefer to feed over grassy pastures, plowed fields, and other open areas, but also feeds over floodplain forest, above canyons, and near towns. Forages over water (lakes, ponds, rivers) primarily when cool or rainy weather reduces insect availability and prevents formation of thermals that concentrate insects (Brown 1988, Brown and Brown 1996). Birds have been seen walking on the ground and picking ants off bare dirt in e. Washington (P. Stoddard pers. comm.) and brine flies off the shoreline at the Great Salt Lake, UT (Paton and Fellows 1994).
Food Capture and Consumption
Exclusively a diurnal forager, usually feeding in groups on aggregations of insects. In Nebraska, foraging groups during the breeding season vary from 2 to >1,000 birds, and some individuals feed solitarily. Birds often rely on local enhancement to discover insect swarms, watching nearby foragers and converging on a spot where the prey-capture behavior of other birds indicates a food source (Brown 1988, Brown and Brown 1996). While foraging, Nebraska birds use the Squeak Call (see Vocalizations) to signal when a food patch has been discovered. This call attracts other foragers and may serve to ensure that the insect swarm will be effectively tracked and that the discoverer can remain knowledgeable of its whereabouts (Brown et al. 1991). However, this call is used only in bad weather (poor foraging conditions) and relatively early in the season, and thus the contexts promoting calling are not well understood. The Squeak Call is used exclusively by birds on the foraging grounds and not at the colonies (Brown et al. 1991).
Foraging groups often feed on the lee side of bluffs or road cuts where insects concentrate (Brown 1988). Birds cue on thermals that passively transport insects aloft and on insect mating swarms and other types of aggregations (Brown and Brown 1996). Thermals and convection currents lead to a patchy distribution of insects, with the birds’ prey abundantly but unpredictably concentrated in several spots near a colony. When the air temperature is not warm enough for convection, birds feed lower over grass tops or water surfaces and in a more dispersed fashion (smaller groups). In cold weather, birds in Nebraska forage a few centimeters above the water and pick aquatic insects off the surface. Before nest-building starts, birds feed throughout the day in long bursts and may spend the entire afternoon away from the colony sites. After egg-laying begins, birds feed in more frequent and shorter bursts and are not absent from the colony for prolonged periods at any time of the day. After nestlings fledge, birds resort to longer foraging periods, like those early in the season.
While parents are feeding nestlings, Cliff Swallow colonies serve as information centers (Brown 1986). When a bird unsuccessful at finding food returns to its nest, it may watch its close neighbors; after a neighbor returns with food, the unsuccessful bird may follow that neighbor to a current food source when the neighbor next leaves the colony. Information transfer is unintentional; birds simply observe each other, with no evidence of active signaling at the colony to alert others that food has been found. However, Stoddard (1988) reported a tseer call used in rare circumstances by birds at colonies in Washington; the call seemed to signal that food was available. There is no evidence that birds try to disguise their foraging success to prevent others from following them (Brown 1986, Brown and Brown 1996). All birds alternate being followers and leaders, although how they discover insect swarms initially is unclear. Birds in small colonies (with few neighbors) do not wait at nests to monitor neighbors and instead spend that time searching for prey themselves. In huge colonies (≥1,000 nests), birds are less likely to monitor specific neighbors, and they often join the large groups that continually stream between the colony and food patches (Brown and Brown 1996).
Foraging in groups and using others to find food results in higher mean food intake rates for Cliff Swallows in groups than for birds feeding solitarily. Variance in prey-encounter rates is lowest for birds foraging in large groups (Brown 1988, Brown and Brown 1996). Consequently, birds nesting in larger colonies feed more efficiently and deliver more food to their offspring than do birds in small colonies. See Brown and Brown (1996) for a full discussion of how foraging efficiency is affected by colony size.
Nothing is known about feeding behavior on the winter range, but the large numbers of birds typically seen together (Hudson 1920, P. Burke pers. comm., V. Bowers pers. comm.) suggest that social foraging continues in winter and during migration.
Diet
Insects taken reflect local availability and may vary considerably among colonies located only a few kilometers apart (Brown and Brown 1996). The only generalization possible is that the birds prefer swarming taxa; 10 of the 15 most common families taken in Nebraska were ones known to swarm or otherwise aggregate (Brown and Brown 1996). A total of 84 insect families were represented in the diet of Nebraska birds, including homopterans, dipterans, hymenopterans, coleopterans, neuropterans, ephemeropterans, hemipterans, lepidopterans, orthopterans, and odonates. Grasshoppers are commonly taken during mid- to late summer when hot temperatures apparently reduce populations or activity levels of other taxa (Brown and Brown 1996). Insects not normally considered aerial are sometimes taken when they are transported aloft by thermals and convection currents. A Cliff Swallow colony can deplete the insect-pollinator population to the extent that it negatively affects seed-set of some plants near the colony site (Meehan et al. 2005).
In Nebraska, the most common family taken was Cicadellidae, followed in order by Dolichopodidae, Simuliidae, Formicidae, Empididae, Chironomidae, Muscidae, Culicidae, and Argidae (Brown and Brown 1996). In California (Beal 1907), families were not given, but the most common order was Hymenoptera (39% of total food), followed by Hemiptera (including Homoptera; 27%), Coleoptera (19%), and Diptera (12%). In a larger study of birds from unspecified parts of North America (Beal 1918), Hymenoptera again was the most frequent (28.7%), followed by Coleoptera (26.8%), Hemiptera–Homoptera (26.3%), and Diptera (13.9%).
Metabolism
Food-harvest rates of adults in California are estimated to be at least 3.40, 3.80, and 3.50 kcal/h during nest-building, incubation, and nestling periods, respectively (Withers 1977). Birds extend legs in flight to dissipate heat when ambient air temperature reaches 21–28°C (Butler 1982a) and gape and pant when hot.
Drinking
Birds drink exclusively on the wing by skimming water surface and lapping up water with lower mandible. Drinking is often done in groups, with many individuals suddenly starting and stopping simultaneously.
Development
Young begin vocalizing at least by 5–6 days of age. The call of each chick becomes uniquely recognizable by day 15 and is pure and consistent in structure by day 18–21 (Stoddard and Beecher 1983). Calls of siblings are structurally similar. Sib-sib similarity is genetically based, rather than reflecting vocal imitation among nest mates (Medvin et al. 1992). There is no evidence for vocal learning, sensitive periods, or vocal mimicry. The juvenile’s Begging Call (see below) develops into the Chur Call of adult. A call resembling the Purr Call (alarm) was given by juveniles that were about 6 weeks old (CRB, MBB).
Vocal Array
Limited vocal repertoire. Five vocalizations: the Begging Call (Stoddard and Beecher 1983), used by juveniles when soliciting food from adults; the Purr Call, used as alarm call when predators approach; the Chur Call, commonly used in many contexts (Brown 1985); the Twitter-squeak Song, up to 6 s in typical duration, composed of many guttural gratings, and used primarily during courtship and nest establishment; and the Squeak Call, used as a food-finding signal (Brown et al. 1991). The Squeak Call is structurally similar to the much longer Twitter-squeak Song and may be derived from that song. Both sexes give the Begging Call, the Purr Call, the Chur Call, and the Squeak Call; presumably only males give the Twitter-squeak Song, but no study of singing with marked birds has been done. Apparently there is little geographic variation. Descriptions of vocalizations were similar for birds from W. Virginia, Texas, and Nebraska (Samuel 1971a, Brown 1985), although the rarely used tseer call of birds in the Pacific Northwest does not apparently occur in the Great Plains.
Cliff Swallow Vocalization - Twitter-Squeak SongCliff Swallow Project00:00 / 00:08Cliff Swallow Vocalization - Chur- Purr callCliff Swallow Project00:00 / 10:37Phenology
The Begging Call is used only until juveniles become independent of their parents in midsummer, then changes to the Chur Call. Purr and Chur calls are used at all times of the breeding season; birds seem rarely to vocalize in winter (A. Jaramillo pers. comm.). The Twitter-squeak Song is used primarily in spring while birds are establishing pairs, declines in usage when parents are feeding nestlings, and then is used again during a brief period in late summer while postbreeding birds are defending nests and assessing colony sites (Brown and Brown 1996). Songs presumably are not used on migration or during the winter. The Squeak Call is used from the time of birds’ arrival on the breeding grounds until midsummer; it is unknown if this call is used at other times of the year (Brown et al. 1991).
Time and Places of Vocalizing
Time of day has little influence on the pattern of vocalizing. The Begging Call, the Chur Call, and the Twitter-squeak Song are given by birds at night (2200–0400 h; MDT) while inside their nests (CRB, MBB).
The Begging Call is given by juveniles inside nests and when assembled away from a colony in creches. The Purr Call is usually given in flight, although occasionally a bird in a nest uses a Purr Call upon approach of a terrestrial predator. The Chur Call is used at nests and in flight. The Twitter-squeak Song is given by birds at nests and occasionally in flight as 1 or more birds (males?) chase another (female?). The Squeak Call is given only in flight.
Repertoire and Delivery of Songs
The Begging Call of the juvenile is individually distinctive enough to represent a “signature” that parents use to identify their own chicks (Stoddard and Beecher 1983, Medvin et al. 1993). The Chur Call of the adult may also be distinctive enough to allow juveniles to recognize their own parent (Beecher et al. 1985).
Presumed Functions of Vocalizations
The Begging Call probably reflects food needs of chicks, as hungrier nestlings call more loudly and more readily when adults pass by. Primary function of the Begging Call may be to allow parents to recognize their offspring. Parents learn their chicks’ calls as they develop, so that parental recognition of offspring is well developed by the time nestlings fledge (Stoddard and Beecher 1983, Medvin et al. 1993). Parents in small colonies in Washington discriminate among chicks of similar age and evict intruding ones from their nest (P. Stoddard pers. comm.); eviction of intruders has not been observed in Nebraska colonies which tend to be larger (Brown and Brown 1996). Parents locate their own offspring in a creche (see Breeding: fledgling stage) by the juveniles’ Begging Calls, although the parents’ efficiency at discriminating their own chicks’ voices from others declines in large groups (Medvin et al. 1993, Brown and Brown 1996).
The Purr Call is the Cliff Swallow’s alarm call, used whenever an aerial or terrestrial predator approaches a colony. The call often seems directed at the predator, with birds typically swirling directly above a predator and emitting barrages of Purr Calls. Other colony members respond by exiting their nests. The Purr Call is occasionally given when no apparent predator is present, causing some birds to flush from their nests, whereupon others in the colony use the opportunity to intrude into neighboring nests (Brown and Brown 1996). Such circumstances might represent a deceptive use of alarm calling (Brown and Brown 1989). The Chur Call is a multipurpose vocalization used widely, most often when birds are undisturbed, and may be used for individual recognition between members of a pair, parents and offspring (Beecher et al. 1985), and perhaps neighboring residents within a colony. The Twitter-squeak Song may serve to attract a female to a male’s partially or completely constructed nest and allow her to assess male quality. The Squeak Call and tseer call (Brown et al. 1991, Stoddard 1988) are used by foragers to signal the location of a food patch (see Food Habits).
Walking
Goes to the ground only to collect mud or grass, to attempt forced copulations, to pick up bits of gravel, to sunbathe, or occasionally to eat insects. Sidles along a wire, tree branch, or cliff face using a sideways walk, usually to fight with another Cliff Swallow for unknown reasons.
Flight
Flies at various heights, from just above the ground to 60 m or more. Typical flight speed is estimated at 8.7 m/second (Withers 1977), although some birds commuting from mud holes to colony sites were clocked at 15.5 m/second (CRB, MBB), close to the apparent maximum speed (Shelton et al. 2014). Maximum acceleration estimated at 78.1 m/second/second (Shelton et al. 2014). Changes altitude more frequently than other swallows (Blake 1948). Glides are short and frequent, usually from 2–3 seconds in length but sometimes up to 10 seconds. This is the only North American swallow that customarily slants its wings downward when gliding (Blake 1948). Flapping rates range from 2.9 to 4.5 flaps/second, with a mean of 3.9. Faster flapping rates are employed for climbing and turning, averaging 4.6 flaps/second (Blake 1948). When turning, tail is overspread, showing a convex terminal margin. When pursuing insect prey, the birds make sudden turns to either side or upward, accelerate, and then flare the tail as the insect is caught, whereupon there is a return to close to the original altitude (Brown 1988). Flight dynamics of birds during intraspecific chases were studied in detail by Shelton et al. (2014). Fighting birds sometimes fall out of nests over water (Brown and Brown 1996); some become waterlogged and “swim” to shore by propelling themselves with backward strokes of the wings.
Self-Maintenance
Assembles in groups to preen, often on wires or a rock face near the colony site. Birds spend more time preening (and less time watching for predators) in larger flocks, and birds on the edge of a group preen less than those closer to the center (Brown and Brown 1987, 1996). Preening occurs most often in mid- to late summer after the young fledge, when adults and independent juveniles gather in huge premigratory flocks. During the breeding season, preening occurs mostly in early to mid-morning and for an hour or so before sunset. Head-scratches over the wing. Stretches by extending 1 wing at a time below the feet, then extends both in a “V” over the back. This stretching sequence often immediately precedes taking flight. Yawns sometimes accompany stretches. Bathes by skimming a water surface and “hitting” the surface briefly in a violent collision, sometimes several times in succession. Bathing is communal, and many birds often simultaneously start and stop bathing. Anting is not known to occur. Did not respond when solicited by a Brown-headed Cowbird (Molothrus ater) in a preening-invitation display (CRB, MBB).
Sleeping, Roosting, Sunbathing
Sleeps in the nest once ownership is established and a nest becomes 50–75% complete. Before the nest is large enough to sit in or before a colony site is selected, sleeps in trees. Early in the breeding season, one radio-tagged female returned to the same tree on 4 consecutive nights to sleep (Brown and Brown 1996). Once the young fledge and become independent, some birds (including independent juveniles) continue sleeping in nests, but others start using trees. Presumably sleeps in trees and marshes (Kirby 1978, P. Burke pers. comm.) during migration and in the winter. Sunbathes by rolling over to one side, ruffling feathers, drooping the wings, fanning the tail upward, opening the bill, and pointing 1 eye toward the sun (Barlow et al. 1963). Sunbathing often occurs in preening flocks, especially among birds gathered on cliff faces or bare ground exposed to hot afternoon sunlight.
Daily Time Budget
During nest construction, birds in California spent 9.5 hours each day foraging, 3.0 hours building the nest, and 11.5 hours in the nest (including sleeping; Withers 1977). During incubation, 6.8 hours foraging, 0.4 hours refurbishing the nest, and 16.8 hours in the nest. During nestling period, 7.5 hours foraging, 0.2 hours refurbishing the nest, and 16.3 hours in the nest. Time spent nest-building and foraging varies with colony size in Nebraska (Brown and Brown 1996). Generally, birds upon arrival in the spring spend much of the day foraging, gradually spending more time at the colony site each day as the season advances. After the young fledge, birds begin gradually to spend more time foraging away from the colony site each day until migrating.
Agonistic Behavior
Birds fight for nest sites by grappling and falling out of a partially built nest or off the substrate wall. Physical contact is common among birds fighting for nests. In fights, they peck with their bills and strike with their wings, and they often pull out feathers. Some birds fight repeatedly with each other for 15 min or more. When fighting birds separate after falling out of a nest, one often chases the other for several meters. Birds have been known to fall into the water below nests while fighting and drown (Brown and Brown 1996). Once nests are built, owners defend the nest by sitting in the tubular entrance and lunging at intruders. Intruders usually retreat without a fight, but sometimes an intruder forces its way into a nest, leading to a fight in the nest. The owner ousts the intruder from the nest by using its bill to hold the intruder’s back and shoves it out the entrance. Later in the summer, conspecific intruders enter unattended nests with young and peck nestlings on the head, visibly wounding and occasionally killing them for unknown reasons (CRB, MBB). In preening flocks on wires, a bird often approaches another from the back and tries to knock it off the wire for unknown reasons; others sidle toward the bird next to it and try to peck it and force it to fly. In early spring, several birds sometimes chase another in flight; this may be a form of courtship, as this behavior is often accompanied by Twitter-squeak Songs. When attempting extra-pair copulations at mud holes, males sometimes seem to mistake other males for females, and a copulation attempt turns into a fight in the mud (Brown and Brown 1996). Possibly because of increased competition for nesting sites in larger colonies, the levels of circulating testosterone in both males and females increase with colony size (Smith et al. 2005).
No threat or appeasement displays are known. When defending the nest, both sexes often slightly raise the feathers of the head and neck, making them look larger (“puffed up”). The white forehead patch, which shows easily in the darkness of the nest entrance, probably serves as a signal to potential intruders that a nest owner is home, as birds constantly face out of the entrance when present at the nest.
Territoriality
The only defended area is the nest or (early in the breeding season) a region on a vertical wall where a nest is to be built. Space defended is the interior of the nest and that area within a bird’s reach when sitting in a partial nest or clinging to a substrate. Once a nest is complete, the outside of the nest is not defended; other birds may sit atop a nest while an owner is inside peering out. Nest owners attack other birds that try to build a nest within 8–12 cm directly below a nest’s entrance; this usually prevents later-nesting birds from blocking the entrance of existing nests (Brown and Brown 1996), leading to a honeycombed pattern of nest placement in most colonies.
Defense of space is suspended during unusual cold-weather events that lead to mortality (Brown and Brown 1998a; see Demography and Populations). At these times, adults crowd together inside nests to conserve heat. Up to 12 adults have been found packed into a single nest; when the bird nearest the entrance dies, it may trap the others that are unable to exit the nest (CRB, MBB).
Individual Distance
Cliff Swallows are extremely social at all times, seeking out other individuals whenever away from their nests. Preening birds on wires are often spaced as closely as 10 cm (Emlen 1952), and sometimes to 3–4 cm or with shoulders touching (CRB, MBB; see Degree of sociality, below).
Mating System, Sex Ratio, and Pair Bond
Socially monogamous; only 1 male and 1 female tend a nest; neither sex is known to establish ownership of >1 nest. Genetically polygamous, as both sexes routinely mate with multiple members of the other sex (see below). Based on mist-net captures, the sex ratio in Nebraska appeared male-biased at about 1.3 males:1 female (Brown and Brown 1996), but for a sample of 1856 birds dying during cold weather and dissected (Brown and Brown 1998a), the sex ratio was exactly 1:1.
Pair bond is more accurately a form of “mutual tolerance” of the other sex at the nest (Emlen 1954); sexes do not associate together away from the nest. Male sings to female while nest ownership is being decided, but there is little formal courtship, and singing declines once egg-laying and incubation begin. There is no mate-guarding (Brown and Brown 1996).
Copulation between nest owners occurs within the nest after the nest has been built to at least a shallow cup (Emlen 1954). Some copulations are preceded by the male leaving his mate at the nest entrance, retiring to the back of nest, and uttering a soft Chur Call. The female follows the male to the back of the nest and crouches, whereupon he mounts her. Copulating birds often tumble out of the nest if it is still incomplete (Emlen 1954), but in a complete nest, copulation ends with both birds returning to the nest entrance. The male often repeats copulatory invitations by going to the back of the nest several times in succession; female may ignore him and remain at the entrance. The male also frequently attacks his mate just after her return from a mud hole and copulates in a forced way. This may reflect sperm competition; a male’s probable defense against extra-pair copulations experienced by his mate at mud holes is frequent intrapair copulation (Brown and Brown 1996). Copulation begins 4–6 d before the first egg is laid and continues frequently until the afternoon preceding the laying of the last egg (Emlen 1954). The pair bond dissolves after the young fledge, and any re-pairing in a subsequent year is merely coincidental when both birds return to the same part of a colony (Mayhew 1958, Brown and Brown 1996, Meek and Barclay 1996). Mutual tolerance by 2 birds at the same nest in late summer during postbreeding colony visitation may reflect former nest owners reuniting briefly, but no studies of marked birds have been done to confirm this.
Extra-Pair Copulations
Common at mud holes where birds collect mud for nest-building (Emlen 1952, Butler 1982b, Brown and Brown 1996). Also occurs when birds go to the ground to gather grass for nest lining. Both males resident at a colony and nonresident males engage in extra-pair copulations (EPCs; Brown and Brown 1996). Females sometimes resist, other times accept EPCs. The number of EPCs/female increases with the size of the mud-gathering group and in larger colonies (Brown and Brown 1996). When gathering mud, both sexes flutter wings above their back, possibly to prevent being attacked by males seeking EPCs (Butler 1982b, Brown and Brown 1996; see Agonistic Behavior). Other EPCs occur at the colony when a male intrudes into a neighboring nest and forcibly copulates with the female nest-owner. The success of EPCs in leading to fertilizations probably varies: allozyme exclusion analyses (Brown and Brown 1988a) showed that collectively up to 43% of nests in Nebraska contained 1 or more nestlings not related to either the father (EPC) or the mother or both (intraspecific brood parasitism; see Breeding: brood parasitism). A study in Pennsylvania suggested that 13% of nests had extra-pair young and that 5.6% of the total offspring resulted from EPCs (L. Reichart pers. comm.). Perhaps as a result of increased rates of EPCs that lead to more intrapair copulations and sperm competition in larger colonies, testis volume increases with colony size (Brown and Brown 2003). EPCs may be a strategy perpetrated especially by inferior males, because those males had a probability of annual survival about 33% lower than that of males not engaging in EPCs (Brown and Brown 1998b).
Degree of Sociality
The Cliff Swallow shows the highest degree of coloniality of any swallow in the world. Colonies often number 200–400 nests and routinely range up to 1,000 nests, with ones as large as 6,000 nests in Nebraska (Brown et al. 2013a) and 5,000 nests in Kansas (Thompson et al. 2011). Solitary nesting does occur, however, sometimes only a few kilometers from the largest colonies. Colonies are smallest in e. North America, especially in areas where the species has been breeding only a short time, and in parts of the sw. U.S. There is great diversity in colony size within a population, although the basis for colony size variation is still poorly understood (Brown et al. 2013a, 2016). In Nebraska, colonies on bridges and highway culverts average larger than those on cliffs, but in the Rocky Mountains and other parts of w. North America, substrate type probably has no effect on colony size. Some colony sites are used perennially, others more erratically. Often 1 year (and occasionally up to 5 years or more) may elapse between use of a given site in California (Grinnell et al. 1930), Texas (Sikes and Arnold 1984), Oklahoma (Loye and Carroll 1991), Arizona (S. Speich pers. comm.), and Nebraska, but the reason(s) for alternate-year usage patterns are not clear (Brown et al. 2013a). See Brown et al. (2013a) for a detailed, long-term study of colony-size dynamics and site usage in Nebraska. Coloniality probably evolved initially to facilitate efficient social foraging during the breeding season, and birds may have subsequently clustered their nests in high densities to exploit secondary benefits of group living (see Brown and Brown 1996 for details). Species remains in large groups during the nonbreeding season; flocks of thousands are often seen together on the Argentine winter range (Hudson 1920, P. Burke pers. comm., V. Bowers pers. comm.). Birds may be nomadic during the winter, traveling over large areas in search of insect emergences (A. Jaramillo pers. comm.).
In Nebraska, play appears to occur when groups of adults (and, later in the summer, independent juveniles) all try to crowd onto the same space of ≤1 m along a wire. Birds on a wire pack themselves together tightly (bodies touching) and try to maintain their position as dozens of others hover behind them and try to knock them off and usurp their places. Incumbents often hang off a wire upside down in an attempt to keep their places. Sometimes 75–100 birds engage in these jousting events; that there is always ample perching space and that birds cease this activity after 10–15 min and resume normal spacing on nearby parts of the wire suggest that it is a form of play (CRB, MBB).
Nonpredatory Interspecific Interactions
Usurps inactive and active Barn Swallow (Hirundo rustica) nests, expelling the owners. Domes over Barn Swallow nests, turning them into typically shaped Cliff Swallow nests. Extra-pair copulation with female Barn Swallows may lead to the reported cases of Cliff x Barn swallow hybrids (Brown and Page 2015). Usurped a Say’s Phoebe (Sayornis saya) nest that had been constructed in an old Cliff Swallow nest fragment; invading swallows killed the nestling phoebes and threw them out of the nest (Brown and Brown 1996). Sometimes nests in Bank Swallow (Riparia riparia) colonies (Carpenter 1918, Monroe and Mengel 1942, Emlen 1954, CRB, MBB), although it is unknown if active Bank Swallow burrows are usurped. Comes into frequent contact with Cave Swallows in mixed-species culvert colonies in s.-central Texas (Thayer 1915, Martin 1980, Weaver and Brown 2004), but behavioral interactions between Cliff and Cave swallows have not been studied.
Cliff Swallow nests have been used for breeding by Say’s Phoebes, Chestnut-backed Chickadees (Parus rufescens), Plain Titmice (P. inornatus), House Wrens (Troglodytes aedon), Eastern Bluebirds (Sialia sialis), House Sparrows, and House Finches (Carpodacus mexicanus) (Mayhew 1958, Weeks 1995, CRB, MBB); all but the phoebe may usurp active nests (see Predation, below, for discussion of House Sparrows). White-throated Swifts (Aeronautes saxatalis) occasionally nest and forage among Cliff Swallows (Mayhew 1958). A House Sparrow repeatedly fed nestling Cliff Swallows in an Alberta colony (Hofman 1980). Cliff Swallows routinely flock with other swallow species during migration and foraging, but there is no evidence of any cooperative or commensal feeding with these species. In mixed-species perching flocks, Cliff Swallows attack Bank and Barn swallows and drive them off wires. In Nebraska, both Eastern (Tyrannus tyrannus) and Western (T. verticalis) kingbirds often chase Cliff Swallows for no apparent reason, sometimes driving a swallow to the ground (CRB, MBB). Various species of bats roost in abandoned Cliff Swallow nests, and during the winter Canyon Wrens (Catherpes mexicanus) and Black (Leucosticte arctoa) and Gray-crowned (L. tephrocotis) rosy finches use Cliff Swallow nests as dormitories (Sooter et al. 1954, MBB).
Ants (Crematogaster lineolata and Formica spp.) prey on swallow bugs (see Demography: body parasites) on the outsides of active Cliff Swallow nests in Nebraska, but these ants confine themselves to the outsides of the nests and have not been observed going inside nests or affecting birds (Brown et al. 2015c). Ants very effectively control bugs on the outsides of nests and benefit Cliff Swallows in this way. Various species of spiders also prey on swallow bugs and probably reduce parasite numbers within Cliff Swallow colonies.
Kinds of Predators
Primarily birds and snakes. In Nebraska, Great Blue Herons (Ardea herodias), Sharp-shinned Hawks (Accipiter striatus), Cooper’s Hawks (A. cooperii), Peregrine Falcons (Falco peregrinus), American Kestrels (F. sparverius), Barn Owls (Tyto alba), Great Horned Owls (Bubo virginianus), Blue Jays (Cyanocitta cristata), Black-billed Magpies (Pica hudsonia), Loggerhead Shrikes (Lanius ludovicianus), Common Grackles (Quiscalus quiscula), and bull snakes (Pituophis catenifer) attack colonies (Brown and Brown 1996). In other areas, predators include American Kestrels, Acorn Woodpeckers (Melanerpes formicivorus), Loggerhead Shrikes, and unspecified ants in California (Bent 1942, Wilkinson and English-Loeb 1982, Fajer et al. 1987); Black-billed Magpies in the Yukon (Sinclair et al. 2003); Peregrine and Prairie (Falco mexicanus) falcons and Mississippi Kites (Ictinia mississippiensis) in Oklahoma (Byard et al. 1979, C. Hopla and J. Loye pers. comm.); Red-headed Woodpeckers (Melanerpes erythrocephalus) in Ohio (Jones 1883); bull snakes in Washington (Thompson and Turner 1980), Oklahoma (C. Hopla pers. comm.), and Utah (Czaplewski et al. 2012); rat snakes (Elaphe obsoleta) in Oklahoma (Oliver 1970), Texas (W. Pulich pers. comm.), and Tennessee (Bullard 1963); coachwhip snakes (Masticophis flagellum) in Oklahoma (C. Hopla pers. comm.); rattlesnakes (Crotalus sp.) in Montana (Bent 1942); minks (Mustela vison) in Washington (P. Stoddard pers. comm.); and fire ants (Solenopsis invicta) in Texas (Sikes and Arnold 1986). Domestic cats prey on mud-gathering birds in Massachusetts (M. Silver pers. comm.). House Sparrows and deer mice (Peromyscus maniculatus) usurp nests and in the process destroy large numbers of eggs and nestlings (CRB, MBB). Box turtles (Terrapene ornata) prey on birds (primarily nestlings) on the ground (Brown and Brown 2008).
Manner of Predation
Great Blue Herons alight on bridges at night and reach underneath the overhangs, using their bill to grasp and pull swallows out of the nest. Sharp-shinned Hawks attack colonies at dusk by catching adults in flight as they come into roost. American Kestrels hunt adults and fledged juveniles primarily by diving from above the colonies and striking birds flying below them. Occasionally kestrels fly up to a nest and try to pull nestlings out of the entrance hole. Peregrine Falcons, kestrels, and Cooper’s Hawks in Nebraska fly through culverts and catch Cliff Swallows flushing from their nests. Owls visit colonies at dusk and use their talons to pull birds from nests. Blue Jays sit on adjacent nests and extract small nestlings from nests. Magpies perch on top of a cliff or a bridge containing nests and fly out toward incoming adults, trying to collide with them, and also scavenge birds of all ages found on the ground. Loggerhead Shrikes fly into large colonies and try to collide with incoming or outgoing birds. Grackles attack mud-gathering and grass-gathering adults by walking toward them and pouncing on a bird from the side or above. Grackles also cling to nest exteriors and try to pull nestlings out, attack birds perching on wires, chase down and catch recently fledged juveniles near colonies, and scavenge nestlings that fall out of nests. In Nebraska certain grackles learn to specialize on Cliff Swallows; one grackle killed 70 birds (mostly yearlings) over a 12-d period, often eating only the brains (Brown and Brown 1996). Woodpeckers alight at nest entrances and pull eggs and nestlings out; Red-headed Woodpeckers have been seen to drill holes in the mud nest to reach inside (Jones 1883). Avian predators recruit to larger Cliff Swallow colonies, and the per-capita risk of predation increases for birds breeding in large colonies (Brown and Brown 1996).
Snakes climb to nests and can reach colonies located on cliffs, buildings, concrete culverts, and metal bridges. Bull snakes may spend up to 3 d in a colony, coiling inside a nest, out of sight, and grabbing nest owners when they enter the nest. One bull snake in Nebraska consumed about 150 eggs in a single colony over a 3-d period (Brown and Brown 1996). Snakes, probably the most important predators, are also attracted to larger colonies. Stacking of nests close together in large colonies enhances snakes’ access and may represent a cost of coloniality (Brown and Brown 1996). Fire ants crawl up the substrate to reach nests and feed on eggs and nestlings (Sikes and Arnold 1986).
House Sparrows destroy eggs in attempts to usurp nests; a single House Sparrow may clean out 12–15 adjacent nests before selecting one as its own. Cliff Swallows seem intimidated by House Sparrows and do not attempt nest defense against them. In W. Virginia, 48% of Cliff Swallow nests were lost to House Sparrows in 1 year (Samuel 1969a), and in Nebraska House Sparrows destroyed contents of all nests in a 100-nest colony (Brown and Brown 1996). Deer mice scale vertical walls on which nests are located and chew through the walls of adjacent nests. In addition to destroying eggs and small nestlings, mice chew on wing and tail feathers of older nestlings, often impairing flight when these birds fledge (CRB, MBB). An unknown predator (fish or turtle) captured a bathing Cliff Swallow by grabbing it and pulling it below the water surface (Brown and Brown 1996). Snapping turtles (Chelydra serpentina), bull snakes, and ornate box turtles scavenge doomed nestlings that fall out of nests.
Response to Predators
Typical response to most predators consists of colony residents milling above the predator and alarm-calling (Purr Call) heavily. When a falcon or hawk approaches, colony residents exit the colony, fly in a very coordinated, tight flock to the altitude of the predator, then spread out above the predator in a loose group and follow it as it moves, alarm-calling continuously. This seems to signal to the predator that it has been detected (Brown and Brown 1996). Birds mill above predators that approach from ground level (snakes, woodpeckers, grackles) and give barrages of alarm calls. They typically do not dive at predators (Brown and Hoogland 1986). Less pronounced responses are given to predators such as shrikes, grackles, and snakes, with some colony residents not exiting the colony during an alarm. Alarm calls are not given to House Sparrows.
The distance at which an approaching predator is detected increases with colony size, and thus large colonies may confer a benefit by enabling mates and nearly fledged juveniles to escape predation more often (Brown and Brown 1987, 1996). Vigilance is enhanced in preening flocks away from colonies, in creches, and in mud-gathering groups; an important advantage of flocking in both the breeding and nonbreeding seasons is that per-capita time spent in vigilance can be reduced (Brown and Brown 1987, 1996). Birds on the edges of preening flocks (closest to a predator’s approach) are more vigilant than the birds closer to the center. Birds also exhibit vigilance at their nests, and individual differences in the extent of vigilance may reflect different personality types among birds in different colonies (Roche and Brown 2013).
A major objective during the period from about 2006-2010 was to study a viral pathogen, Buggy Creek virus, and how it affected cliff swallows in colonies of different sizes. The virus is transmitted to birds by blood-feeding swallow bugs that live in the birds’ nests year-round. Buggy Creek virus has little effect on cliff swallows, but it severely affects invasive house sparrows that occupy cliff swallow nests at some sites. The invasive sparrows appear not adapted to this virus as a result of their relatively recent exposure. We have used the cliff swallow/swallow bug/Buggy Creek virus system to explore general questions about the ecology of bird-associated virus transmission. Although not a human pathogen, Buggy Creek virus is similar in some ways to viruses such as West Nile and western equine encephalitis viruses, which do affect human health.
One advantage of a long-term study is the ability to document rapid evolutionary changes in response to variability in the environment, including climate change. We have found that body size of cliff swallows has undergone a change since the early 1980’s, with birds now skeletally larger but with shorter wings and tails than when our study began. These changes reflect natural selection brought about in part by severe weather events (late spring cold snaps that reduce the availability of the birds’ food), causing the population to shift from smaller to larger birds almost overnight. Vehicles also exert a selection pressure on cliff swallows, because the birds often nest around roads. Swallows with shorter wings are more likely to escape an oncoming car, and consequently selection has favored birds with shorter wings over time and has resulted in fewer birds being killed on roads now than in the 1980’s. Finally, we are finding evidence that cliff swallows’ ability to tolerate the parasites that live in their nests has improved over time, a response to the high infestations of swallow bugs these birds have been exposed to over the last 35-40 years in our study area.
We recently initiated studies of personality in cliff swallows, building on research with other species illustrating that the behavioral composition of animal groups can often vary. Some groups consist of a higher percentage of bold or risk-taking individuals, while other groups contain more shy animals. We are investigating whether cliff swallows sort among colony sizes based on their personalities and whether the reproductive performance of an individual depends on its personality and/or those of others in the colony.
