Diet and Foraging
Welcome to the Birds of North America Online!
You are currently viewing one of the free species accounts available in our complimentary tour of Birds of North America. In this courtesy review, you can access all the life history articles and the multimedia galleries associated with this species.
For complete access to all species accounts, a subscription is required.
Already a subscriber? Sign In
Main Foods Taken
As generalized summary (14 studies: 1 from 1893, remainder published 1933–1974 from Sherrod 1978), mostly birds, estimated at about 77–99% (frequency not biomass), passerines to small geese; occasionally mammals and rarely amphibians, fish, and insects. Most frequent mammals: bats (e.g., Tadarida, Eptesicus, Myotis, Pipistrellus, etc.), microtines (e.g., Microtus, Dicrostonyx, etc.), squirrels (Spermophilus), and rats (Rattus). In California, fish and mice pirated from Ospreys (Pandion haliaetus) and Red-tailed Hawks (Buteo jamaicensis; B. Walton pers. comm.). Depending on habitat, predominant foods differ: tundra, ptarmigan (Lagopus), shorebirds (e.g., Pluvialis, Gallinago, Phalaropus, Calidris), small passerines, such as longspurs (Calcarius) and Snow Buntings (Plectrophenax nivalis), and ducks; taiga, shorebirds (e.g., Gallinago, Tringa, Actitis), woodpeckers (Colaptes), passerines (jays, thrushes); coastal marine, podicipe-dids, alcids, procellarids, waterfowl; interior continental, columbids (e.g., Zenaida), swifts, passerines; urban, columbids (e.g., Columba), cuckoos, rails, passerines. Overall, in temperate continental latitudes Columbidae (pigeons, doves) may be most frequently taken and perhaps most important by biomass.
Microhabitat For Foraging
Most prey captured in air, while Peregrine is in flight; also from surface of water or ground; may walk on ground is search of nestling birds and rodents (Harris and Clement 1975; Dekker Dekker 1980, Dekker 1995, Dekker 1999; Rosenfield et al. 1995b). Mexican free-tailed bats (Tadarida brasiliensis) regularly hunted at dawn and dusk at mouths of roosting caves on Edwards Plateau, TX (Stager 1941, Skutch 1951b, Lee and Kuo 2001).
Food Capture And Consumption
Best described as sequence of actions consisting of search, pursuit (attack), capture, killing, and eating.
Search. Peregrine searches either from perched position (most commonly) or while flying, sometimes on foot on ground. Especially in breeding season, adult perches on some high vantage point on cliff, usually near eyrie, overlooking vast air space in which other birds fly (see Treleaven Treleaven 1977, Treleaven 1998 for many examples along Cornish coast). Position allows falcon to stoop down easily on low-flying prey or to ring up after high-flying ones. Initial height above prey prior to attack was positively correlated with success in capture in Africa (Jenkins 2000). Aerial search performed either by flapping flight or when soaring at height; some parents search up to 15–43 km from eyrie (White and Nelson 1991, Enderson and Craig 1997). On Cornish coast, hunts from perches more successful (60%) than hunts initiated in air (40%); latter occurred mainly in bad weather when soaring not possible (Treleaven 1980).
In fall and winter, often hunts from lower perches: trees, utility poles, fence posts, banks, mounds, driftwood—depending on habitat (Dekker Dekker 1980, Dekker 1999). Migrating Peregrine hunted from perches or in low flight in early morning (0.5 h before to 3.0 h after sunrise) and again following migratory movement after about 15:00; exceptionally launched attack during migration-soaring (Cochran 1975). Sometimes follows and uses human or animal to flush prey: gunners, harriers (Circus sp.), dogs (Canis familiaris), some-times machines (motorboats, harvesting equipment, vehicles; Palmer 1988c, TJC). Of 318 attacks on ducks, 55% initiated while prey was not in flight or just after taking off (Dekker 1987).
Sometimes hunts on foot after insects, other invertebrates, small mammals, and especially nestling and fledgling birds and precocial downy young (Sherrod 1983, Rosenfield et al. 1995b, Dekker 1999). Particularly noted in Greenland, where passerines provide bulk of food.
Pursuit (Attack). Several modes recognizable: stoop, ringing up, direct pursuit, contour-hugging, shepherding, running or hopping, and flapping on ground (see also Behavior: locomotion, below). Stoop is well-known mode: dive from above quarry, varying from <100 to >1,000 m long and from 90° angle to horizontal to <20°, in which gravity and body mass produce velocities in range of 25 to 100 m s-1; used to overtake and catch fleeing prey usually flying but sometimes running on ground or swimming in water. In shallow stoop or initial stage of steep one, falcon sometimes flaps wings, but in long, steep dives, wings are folded against body (Franklin 1999); rate of fall can be slowed by slight adjustments of wings, legs, and head (Tucker et al. 1998b). Occurs over land and water. Most often stooping falcon pulls out of dive some meters behind escaping prey and shoots forward at great speed to grab or strike prey, or repeatedly stoops to force it down into water or onto open ground where it can be grabbed. Sometimes stoops directly down on prey, striking it in head, wing, or back, killing it or breaking wing (Cade 1982b, TJC, CMW, WGH).
Ringing up: Attack used to pursue birds initially flying higher than falcon. Two variations: (1) falcon spots bird flying high above cover (e.g., dense forest) and takes off to intercept it, circling up until it is slightly above quarry; then executes series of shallow stoops until quarry becomes exhausted and easily caught in air, or quarry attempts to escape by diving earthward to cover, in which case falcon dives be-hind it and catches up to grab or strike. (2) falcon starts with direct pursuit of low-flying quarry, usually where there is little ground cover. Prey attempts to escape by rising up into air faster than falcon (the classic haut vol of French falconry, in which slower but more lightly wing-loaded quarry [heron, kite, gull, owl] gains altitude by ringing up in tight spirals, while heavily wing-loaded but faster-flying falcon mounts in wide rings around the quarry). Falcon also maneuvers to keep quarry flying upward and from diving for cover, until it becomes exhausted and can be grabbed (bound to) in air (Rudebeck 1951, Cade 1982b).
Direct pursuit: Often when stoop is unsuccessful and prey flees by flying straight away, falcon follows directly behind (tail-chasing) in powered, flapping flight in attempt to overtake and grab prey. Perched or flying, falcon may see prey get up and fly away unhurriedly; then launches accelerated flapping flight (high-intensity hunting; Treleaven 1980) to intercept prey. Alcids and other waterbirds caught during high-speed, low-level flights over water, surprising prey on surface or as it rises from water. Fish also caught as they broke surface of water (Cade 1982b).
Contour-hugging: Special form of direct pursuit in which falcon flies low, using concealing features of terrain or water (banks, fencerows, dunes, ridges, waves) to remain hidden from quarry until very close; in such surprise attacks, birds sometimes fly up directly in front of falcon and are caught before they can accelerate, while others freeze on ground or refuse to fly from water and are grabbed before they fly. Per-egrine uses surprise attacks more frequently than commonly thought (White and Nelson 1991, Dekker 1999).
Shepherding (: When hunting, Peregrine attacks flocking birds, many species (e.g., pigeons, shorebirds, starlings, waxwings, Snow Buntings, Lapland Long-spurs, grosbeaks) avoid capture by massing in tight formation-flights that zigzag about the sky (Tinbergen 1951, Buchanan 1996, Dekker 1998). Falcon appears hesitant to strike into mass of close-flying bodies; in-stead repeatedly harries the group by diving at its periphery; occasionally a panicked bird breaks formation and is then vulnerable.
Attack on ground: Sometimes lands and runs/hops and flaps after invertebrates, rarely reptiles, small mammals, and especially newly fledged birds and downy young of precocial species (waterfowl, shorebirds, gamebirds; TJC, CMW). Also runs down fallen birds injured in stoop.
Capture and Killing. Captures mainly by grabbing prey with feet (binding) but rarely kills small prey by forcing talons into body as accipiters do. Instead, falcon bites into neck, disarticulating cervical vertebrae and severing nerve cord; even with prey killed in stoop, falcon bites into neck before feeding begins. Prey frequently killed this way while falcon is flying. Young bird performs this action first time it has in-tact prey, dead or alive, in its feet. Stooping or chasing falcon often flies in under prey and rolls over or flips up to grab from below or side, but also grabs from back, holding onto wings or neck of large prey (Cade 1982b).
More spectacular strike or blow delivered in stoop is less common method of stunning or killing prey (Treleaven 1998, Dekker 1999). Confusion exists as to how falcon performs this action. Older and aboriginal ideas that falcon strikes with its breast or wing butts refuted. Strike is made with feet, but how? Long be-lieved that falcon strikes with loose fist, back talon protruding as gashing device (see Fuertes 1920, Cade 1982b), but High-speed cinematography revealed that all 4 toes are widely splayed at moment of contact, then immediately formed into fist after blow (Goslow 1971, M. W. Nelson film record). Many strikes may simply be failed attempts to bind at high speed, but some are delivered with enough force to displace or roll quarry several meters through air or cause it to bounce off ground on impact. Speed at which falcon reaches prey probably determines whether capture is by binding or striking. To grab, falcon must be near speed of prey to prevent momentum from pulling prey loose from talons. Some trained falcons learn to strike prey in head or gash pectoral area sufficiently to disarticulate from shoulder. Although some prey killed outright by striking, large prey (> size of falcon) usually only stunned and rendered easier to grab in air or on ground.
Eating. After biting into neck, falcon carries small prey to habitual plucking perch (tree snag, cliff side, driftwood, building) for consumption, or to cache site. Prey too heavy to carry in flight are partially con-sumed on ground; remains may later be carried to eyrie or plucking perch, or left in place for later re-turn. Sometimes eats small prey, especially bats, while flying (Skutch 1951b); male hunting over Colorado River in Grand Canyon caught and ate 7 bats in 20¿min of continual flight (TJC).
Much variation depending on size of prey. Small bodies <100 g usually totally consumed after wings, tail, and some body feathers deplumed; prey >250 g are picked clean of flesh and viscera, but skeletal elements left more or less intact. Wings with primaries still articulated with pectoral elements and sternum, sometimes neck also on larger birds; synsacrum and legs often remain articulated but usually separated from anterior skeleton; tail feathers usually plucked (Hagen 1952, TJC, WGH).
Begins eating by tearing off head; usually consumed if small, picked apart and eaten or discarded if large. Continues by pulling apart and eating skin and flesh of neck (also bones of small prey), working down to breast. Depluming of breast precedes tearing into pectoral muscles, which are usually totally consumed. Viscera may or may not be eaten; often gut is pulled out and discarded, but remaining organs, especially heart and liver, usually eaten. Legs of large prey may or may not be picked clean (TJC, CMW, WGH).
Appears to use tomial teeth to break long bones of wings and legs of smaller prey before swallowing. Large prey too heavy to carry back to eyrie are well plucked wing and tail feathers removed, head removed, eviscerated, and sometimes posterior half of carcass detached from breast, before latter carried back to eyrie to feed young.
Hunting Success. Varies with age of falcon, hunger level, time of day, season, species of prey, and behavior of prey (Dekker Dekker 1980, Dekker 1995, Dekker 1999; Treleaven 1980; Bird and Aubry 1982; Roalkvam 1985a; White and Nelson 1991). During winter in w. Washington, Peregrines hunting Dunlins (Calidris alpina) stooped directly at compact flocks with success rate of 47% (Buchanan et al. 1986) in estuaries, but rate dropped to 12.5% on coastal beaches (Buchanan 1996). Highest rate may be 93% (102 hunts) and 100% (68 consecutive hunts) by resident male still-searching and then ringing after high-flying migrants (mostly Blue Jays [Cyanocitta cristata]) over New Jersey salt marsh (Cade 1982b). For 23 different studies, average success rate was 23.7% (range 7–83%); outside breeding season, adults significantly more successful than immatures (12.7 vs. 7.3%; p < 0.01), and breeding-season adults were si-gnificantly more successful than adults outside breeding period (34.9 vs. 12.7%, p < 0.001; Roalkvam 1985a). Motivation of falcon, indicated by high- and low-intensity hunting (Treleaven 1980), appears to influence success rate greatly and sometimes leads to false impression that Peregrine is inefficient hunter.
Main Foods Taken
For North America, minimum of 429 species of birds, 10 bat species, and 13 other mammal species recorded (190 for California alone; B. Walton pers. comm.). Exceptionally, fish (4 species) and insects (mainly Orthoptera [grasshoppers, crickets] and Odonata [dragonflies, damselflies]) also. Rarely carrion (Holland 1989). Estimate of “well over 250 species” of birds captured worldwide often given (Palmer 1988c), but that number almost certainly exceeds 1,500 and probably 2,000 species. Primary literature on foods in North America too vast to be cited. Birds as large as Sandhill Crane (Grus canadensis; about 3,100 g) hit in the head and killed in midair in Alaska (CMW), also Bald Eagle (about 4,700 g) hit in head in Arizona, although probably not for food, and found dead within a month apparently from injuries exhibited during month (Hunt et al. 1992b). Smallest items, hummingbirds (Selasphorus, Archilochus spp.; 2.5–3.5 g), largest items in eyries, small geese (Brant [Branta bernicla]; 1,400 g; Cade et al. 1968). But known to have killed a small Canada Goose (Branta canadensis leucopareia), 1,700–2,200 g, on the ground (Stabins 1995). Frequently captures extremely aerial birds such as White-throated Swift (Aeronautes saxatalis), especially on Colorado Plateau, and Black Swift (Cypseloides niger) on San Juan Is., WA (C. M. Anderson pers. comm.). Of 20 prey captures seen in bottom of Grand Canyon, 45% were White-throated Swifts, remainder bats (Brown 1991a).
Dietary composition varies greatly among regions, habitats, seasons, age, and even individuals. Diet mainly quantified from prey remains at nest and plucking perches, less often from observations at nests (e.g., Greenland; Harris and Clement 1975, Rosenfield et al. 1995b) and least from pellets (difficult to determine prey items), and stomach analysis (Henderson 1927, Snyder and Wiley 1976). Each method has drawbacks, usually underestimating some category of prey.
Following are typical prey lists, frequently not quantified. Informative because of the nature or diversity of species, reported for several habitats:
(1) Across tundra, prey diversity decreases west to east, probably reflecting bird diversity. Alaska: 47 species minimum; most frequent prey may change annually (e.g., jaegers [Stercorarius spp.], 2% in 1 yr [n = 98], to 16% [n = 204] following year), but Common Snipe (Gallinago gallinago), golden-plovers (Pluvialis sp.), Gray-cheeked Thrush (Catharus minima), Yellow Wagtail (Motacilla flava), Lapland Longspur (Calcarius lapponicus), and American Tree Sparrow (Spizella arborea) consistently high. Nunavat: 28 minimum prey species with waterfowl, Rock Ptarmigan (Lagopus muta), Horned Lark (Eremophila alpestris), American Pipit (Anthus rubescens), Lapland Longspur, and Snow Bunting most frequent; among all Nearctic Peregrine populations yet reported, may have as high as 29% (about 15% biomass) mammalian food, lemmings (Dicrostonyx and Lemmus), and ground squirrels (Spermophilus), making difference between marginal and good breeding success in some years (Court et al. 1988b). Greenland: 11 species minimum, with 87% (frequency) made up of just 4 passerines: Lapland Longspur, Snow Bunting, Northern Wheatear (Oenanthe oenanthe), and redpoll (Carduelis spp.); also Rock Ptarmigan, 6% (Cade 1960, White and Cade 1971, Falk et al. 1986, Court et al. 1988b, Bradley and Oliphant 1991, Rosenfield et al. 1995b).
(2) In taiga (Alaska), prey taken more diverse than in tundra; minimum 60 species recorded. Depending on year, most important species varied: Lesser Yellowlegs (Tringa flavipes), 1.5–22.8% frequency (0.8 to 15.9% biomass); Gray Jay (Perisoreus canadensis), 5.1–19.2% (2.7–11.1%); Spotted Sandpiper (Actitis macularia), 3.0–10.2% (1.0–2.6%); Common Snipe, 3.8–8.8% (3.5–7.7%). Other frequently taken species: Catharus thrushes, Varied Thrush (Ixoreus naevius), Bohemian Waxwing (Bombycilla garrulus), and Northern Flicker (Colaptes auratus). By frequency in some years, passerines were 52% and ducks and grebes only 14%. By biomass, however, waterfowl and grebes made up to 63%, piciform and passeriform 22%, shorebirds 12%, and passerines as little as 11% in 1 study (Cade 1951; Cade et al. 1968; Hunter et al. 1988a; A. G. Palmer, D. Nordmeyer, and D. D. Roby unpubl.).
(3) Region of entire Pacific Coast shows some uniform characteristics in 4 sample regions, Aleutian Is., British Columbia, Channel Is., Baja California. At minimum, 78% (frequency) of food composed of alcids (auklets, murrelets) and procellariids (storm-petrels, shearwaters). In northern regions, alcids >75% of diet (Beebe 1960, White et al. 1973a, Nelson and Myres 1976). Departure from marine-based prey in some eyries on islands in the Gulf of Georgia, British Columbia, where 85% of prey (frequency) is European Starling (Sturnus vulgaris; R. W. Campbell pers. comm.). In Aleutians, minimum of 31 bird species and 1 mammal species taken. In Baja California, 99 bird species and 2 mammal species taken; numerical importance in Gulf and Sea of Cortez are Eared Grebe (Podiceps nigricollis), Black Storm-Petrel (Oceanodroma melania), and Red Phalarope (Phalaropus fulicaria; equal), Craveri's Murrelet (Synthliboramphus craveri) and Mourning Dove (Zenaida macroura; equal), Bona-parte's Gull (Larus philadelphia), Northern Phalarope (Phalaropus lobatus), Least Storm-Petrel (Oceanodroma microsoma), and Heermann's Gull (Larus heermanni), but order by biomass is Eared Grebe, Craveri's Murrelet, Black Storm-Petrel, Bonaparte's Gull, Red Phal-arope, and Mourning Dove. Within gulf, fishing bat (Myotis vivesi) was caught more frequently than any passerine and as or more frequently than anatids, ranking about equal to Heermann's Gulls (Porter et al. 1988, R. D. Porter and M. A. Jenkins unpubl.). In California Channel Is., gulls and alcids make up 66% of prey biomass in winter, 38% in spring (WGH).
(4) On neotropical nonbreeding grounds, many observations but no extensive published synthesis. Food varies regionally, somewhat by sex and habitat. Because many migrant falcons concentrate coastally, in urban centers, or in regions also occupied by other neotropical migrants, the latter, especially Charadriiformes, figure heavily in diet; e.g., North American waders (Risebrough et al. 1990). In n. Chile, 1 falcon occupying coastal city rarely took urban feral pigeons but went to sea for marine birds, especially phalaropes (Phalaropus spp). In one study along Chilean coast, with a few hundred prey remains of 35 species, most important were Nearctic Red Phalarope, at 18%; Franklin Gull (Leucophaeus pipixcan), 10%; Ruddy Turnstone (Arenaria interpres), 7%; and resident feral pigeons, 12%, and White-winged Dove (Zenaida asiatica), 6% (C. Gonzales unpubl.). Common Tern (Sterna hirundo) and Sanderling (Calidris alba) frequently mentioned prey in Peru (Bertochi et al. 1984, Blokpoel et al. 1989).
In or near urban regions, especially e. South America, feral pigeons (80+%) and bats (Molossus spp. and Tadaris spp.) important (11+%; Pierson and Donahue 1983, Risebrough et al. 1990, Silva E Silva 1997, J. L. B. Albuquerque pers. comm.). In Panama during winter over several years on same skyscraper, 1 female ate only neotropical migrants, no resident species (B. Walton pers. comm.). Bats frequently killed over water and retrieved from surface.
(5) Urban areas (breeding) refer to metropolitan centers, bridges connecting those centers, and associated power plants or near-urban areas (Bell et al., Cade et al., Septon et al., in Bird et al. 1996, J. B. Marks pers. comm.). Minimally, prey items exceed 117 species of birds, 6 mammals, and 1 fish: as expected, by percent, domestic or feral Rock Dove, Mourning Dove, Northern Flicker, European Starling, Blue Jay, and American Robin (Turdus migratorius) are major foods. Because power plants and bridges are near aquatic areas, grebes (Podiceps, Podilymbus spp.) and rails (Rallus, Porzana spp.), American Woodcock (Scolopax minor), and other shorebirds commonly taken.
Food Selection and Storage
Individual falcons and pairs often prey selectively on particular species and classes of prey. Selection appears related to factors that increase vulnerability of prey rather than to abundance per se, although latter can be factor influencing vulnerability too. Factors related to vulnerability and selection can be species-specific as well as specific to characteristics of individual prey. They include mass, plumage patterns, flight characteristics, oddity of appearance or behavior, molt or accidental loss of feathers, sickness or injury, and occurrence in unfamiliar or atypical habitats.
Average mass of prey difficult to determine as it varies regionally and may vary annually and season-ally; approximate mode of prey mass in Alaskan taiga 81 g, in Aleutians and Pacific Northwest about 185 g, in many urban sites about 189 g in summer and 350 g in winter.
Prey caught mostly in range of 50–500 g, perhaps optimal size in relation to aerodynamic characteristics and striking/grasping capabilities of falcon. May be selective tradeoff between advantage for aerial pursuer to be near to, or smaller than, size of escaping prey and capability of holding and killing large prey; latter may explain large size of feet and use of stoop to stun or kill large prey (see Cade 1982b).
Often selects birds with conspicuous flash patterns in flight (e.g., flickers, Blue Jay, Clark's Nutcracker [Nucifraga columbiana], Red-winged Blackbird [Agelaius phoeniceus], meadowlarks [Sturnella spp.], shrikes [Lanius spp.; Craighead et al. 1969); also, male ducks taken more frequently than females in breeding season (Cade 1960). Male birds with con-spicuous aerial courtship or territorial displays often taken during performance (e.g., Common Snipe, Lesser Yellowlegs, jaegers, Short-eared Owl [Asio flammeus], swifts [probably during aerial copulation when mates are locked together in falling flight; J. H. Enderson pers. comm., TJC], Yellow Wagtail, Horned Lark).
In interior Alaska taiga, 2-yr study tallied 83 avian taxa taken by Peregrines: 12 (14%) in 1985 and 11 (13%) in 1986 taken more frequently than expected by relative abundance within 3 km of eyries. Most selected species in descending order: Lesser Yellowlegs, Gray Jay, Spotted Sandpiper, Common Snipe, Solitary Sandpiper (Tringa solitaria), Bohemian Waxwing; 42 (51%) and 47 (57%) of taxa preyed on in proportion to their availability; 29 (35%) and 25 (30%) taken less frequently than expected from availability (Hunter et al. 1988a).
In w. Greenland, Lapland Longspurs, most abun-dant prey species, were most often delivered to nest, but Rock Ptarmigan and Snow Buntings were taken more frequently than expected from their relative abundance on prey transects; Northern Wheatears and Common Redpolls (Carduelis flammea) taken less frequently than expected; most prey were fledglings (Rosenfield et al. 1995b). Also, passerine abundance increased with increasing distance from Peregrine eyries (Meese and Fuller 1989).
In North America, other species apparently taken out of proportion (more frequently) to relative abundance include: rails and coots; small to medium-sized ducks (Dekker 1999); auklets and murrelets among seabirds (Beebe 1960, Nelson 1977b); golden-plover, Dunlin, Pectoral Sandpiper (Calidris melanotos), American Avocet (Recurvirostra americana), Black-necked Stilt (Himantopus mexicanus), Willet (Tringa semipalmata), oystercatchers (Haematopus sp.), and Black Skimmer (Rhynchops niger) among shorebirds; small gulls among Laridae; cuckoos; thrushes and jays among passerines (TJC, CMW); and especially doves and pigeons. Unusual items are Budgerigar (Melopsittacus undulatus), Monk Parakeet (Myiopsitta monachus), Cockatiel (Nymphicus hollandicus), and an array of other escaped caged parrots that are quickly selected because they are oddly conspicuous or naive.
Wherever they occur throughout range of Pere-grine, pigeons and doves—in particular Columba livia, wild, feral, and domestic—usually make up bulk of this falcon's diet (Ratcliffe 1993); Mourning Dove frequently taken in North America, as are White-winged Dove and Band-tailed Pigeon locally. These fast-flying, maneuverable birds with loose body feathers are well adapted to elude capture; yet for reasons not understood—perhaps having to do with taste, nutrition, or regularity with which they must travel to and from water sources, particularly during breeding—falcons pursue them relentlessly and catch many. Migrating Peregrines often prey on other small migrating raptors: American Kestrels (Falco sparverius), Merlin (F. columbarius), especially Sharp-shinned Hawk (Accipiter striatus; Dry Tortugas Is.; T. Smylie and J. Weaver pers. comm.).
Frequently hunts by sitting and watching activities of birds within its field of vision about 2-km radius allowing many birds to pass unchallenged, apparently waiting for disadvantaged individual or an especially stimulating one. Examples: odd black or white pigeon in a flock, homing pigeons released over unfamiliar landscape in Scotland (Ratcliffe 1993); stray Black Brant far inland along Yukon River (Cade 1955b); land bird crossing an expanse of water (e.g., Gray Jay crossing the Yukon, Cade 1960); sometimes exhausted migrant songbird or sandpiper far from escape cover (e.g., Blue Jays and American Woodcock along Atlantic coast, Cade 1982b); seabird flying over land (e.g., Dovekie [Alle alle], in Greenland, auklets on St. Lawrence I., AK, Marbled Murrelet [Brachyramphus marmoratus], in California; B. Walton pers. comm.).
Some individuals and pairs fixate on 1–2 prey species to virtual exclusion of everything else: migrant Sharp-shinned Hawks, nestling boobies (Sula sp.), doves and pigeons, avocets and stilts, murrelets and auklets, terns (Sterna sp.), corvids, and arctic passer-ines. Selective predation by Peregrine may well have both sanitary effect by culling weak, injured, unfit individuals from prey populations, as well as evo-lutionary influence on form and function of some species, particularly columbiforms. Fixation by falcons on 1 prey kind or class probably leads to increased skill in capture.
Peregrines cache and store surplus prey, especially during breeding season. Cache sites vary greatly: often in crevice or hole on cliff face near eyrie; under dense bush or in clump of grass on ledge or top of cliff; at base of fence post or dead tree; under drift wood or log, in marshes; sometimes in cavity or crotch of tree; on building ledges, recessed features, drain holes, under bridges, behind billboards and signs in urban areas. Male sometimes begins to cache early in breeding season before female arrives (unmated male on building in Boise, ID, accumulated cache of >20 Mourning Doves laid out in row on window ledge next to nest box; TJC). Cached prey often used in courtship feeding; sometimes male entices female to take cached prey before bill-to-bill transfers begin (Nelson 1977b). Prey often cached with head attached (Beebe 1960, TJC, CMW) but also headless; partially eaten carcasses also cached after feeding young. Nesting female observes where male caches and may take cached prey when male is slow to deliver food directly to her. Caching appears to also be response to both daily and multiple-day periods when prey is unavailable (e.g., prey available only at crepuscular times or nocturnal, such as Ancient Murrelets [Synthliboramphus antiquus] on Langara I.), and when general unavailability of prey during stormy conditions (Nelson 1977b, Fox 1979b, Cade 1982b, Ratcliffe 1993). At site in downtown Milwaukee, food including cuckoos (Coccyzus spp.) was cleaned from cache several hours after civil sunset; then cache again contained cuckoos 2–3 h before civil dawn, indicating nocturnal hunting, perhaps aided by city lights (Wendt et al. 1991).
Relatively few data, mostly from captive birds. Energy determined by body mass (Mb; larger female uses less mass-specific energy than male; thus total energy use is roughly equivalent between sexes), by ambient temperature, and by physiological work performed; total food consumption further influenced by nutritional quality of diet and efficiency of assimilation. In fall–winter conditions (average Ta= about 0°C), 3 males with mean Mbof 683 g (607–727 g) main-tained Mbby consuming mean of 104 g/d (101–107 g) of mainly lean beef with supplements (15.1% Mb/d); one 721-g male in spring–summer (average Ta= about 20°C) maintained Mbby consuming 83 g/d of same diet (11.5% Mb/d; Craighead et al. 1969). Amount of food for maintenance in laboratory con-ditions measured 89.2 g/kg Mb/d (Duke 1986), but sex and type of food not specified. In experiments on digestive efficiency (Barton and Houston 1993), 3 males with mean mass of 550 g (510–565 g) at 0°C consumed average of 142.3 g/d (141–143 g) of 1-d-old cockerels minus yolk sac and intestines (25.8% Mb/d) with digestive efficiency of about 75.3%; at 20°C, 5 males with mean mass 570 g (535–600 g) maintained Mbby consuming average of 94.5 g (78–117 g) cockerel/d (16.6% Mb/d) with digestive efficiency of 76.3%; 4 females with mean mass 789 g (740–896 g) consumed mean of 113.7 g/d (91–134 g) same diet (14.4% Mb/d) with efficiency of 74.0%.
Circulating levels of vitamins and minerals in-fluenced by diet. In captive birds, levels lower than in wild and poor nutrition affects appearance of soft parts (cere, feet), hatchability of eggs, and vigor of hatchlings. No age or sex difference noted in vitamin A/E or mineral levels of wild birds (Dierenfeld et al. 1989, NJC). Captive females subject to multiple-clutch-ing exhibited depressed levels of vitamins A/E relative to captive males (NJC).
Oxygen consumption or CO2production not mea-sured; metabolic rate higher than other congeners in North America based on frequency and amount of food consumption, especially in relation to Ta(D. Bird pers. comm.). Thermoregulation accomplished primarily by behavioral mechanisms that modify radiative and convective heat transfer, by adjusting body insulation to decrease conductive heat loss, and by evaporative cooling. Behaviors include seeking favor-able microclimates, changing orientation of body relative to sun, changing position of extremities (e.g., drooping legs or wings, tucking in feet), erecting or depressing feathers to change levels of insulation, and panting. In American Kestrel (Falco sparverius), and presumably Peregrine, primary route of radiative/convective heat transfer is bare tarsus (Bartholomew and Cade 1957, Mosher and White 1978). As expected, tarsal index (tarsal surface area divided by body weight) of falcons is related to climate, with species from cooler climates having less tarsal surface area. Males also tend to have higher tarsal surfaces than females. Most Peregrines able to maintain relatively constant cloacal temperatures over a span of 20–50°C. Indications that tundra Peregrines (F. p. tundrius) may be able to maintain more constant Tbover wider range of Tathan Peregrines at lower latitudes (F. p. pealei) or other species (Mosher and White 1978).
Drinking, Pellet-Casting, and Defecation
Drinks frequently, often when bathing. Pellets usually egested once/d early in morning prior to feeding. They are long and oval; about 16 × 41 mm (1.44 g dry) for female and 13 × 29 (0.80 g dry) for male; consist mainly of feathers (or fur), with some bone (usually broken), and hard indigestible parts such as gizzard linings, bills, and toenails. Sand, gravel, and small, usually rounded, stones (gastroliths “rangle”), usually between 6 and 12 mm and as large as 20 × 20 mm, eaten; may be used to clean stomach, promote gastric stimulation, or slough koilin lining; also associated with falcons that have fasted before eating the rangle (Albuquerque 1982, Cade 1982b). May do so up to once a week. Nestlings eject fecal material away from nesting scrape, often causing buildup of fecal material on back walls of nesting ledges or at front edges of ledges.