Breeding

Pair Formation

Some pairs form before arrival on breeding grounds; courtship may be initiated in wintering area, along migration route, or at or near breeding grounds (Bergman 1953, FJC). Birds arrive at breeding areas late Mar–late May (see Migration: timing and routes of migration, above). Courtship activities (Fish Flight, High Flight, Fish Call, etc.) and copulations begin as soon as birds arrive in spring; territories occupied simultaneously, about 4 d after arrival of main group. Peak courtship activity occurs in first 2–3 wk after arrival (Bergman 1953, Kirven 1969, Cuthbert 1981).

Nest-Building

In Great Lakes, birds begin making scrapes as early as 4 d after first birds arrive. After choosing permanent nest site, 1 individual is always present at nest territory; female usually remains by or on scrape for several days before laying first egg (Cuthbert 1981).

First/Only Brood Per Season

Figure 4 . First eggs are laid 2–3 wk after arrival (Bergman 1953); egg-laying period usually lasts 4–5 wk (Vermeer 1972, Cuthbert 1981, Penland 1981, Ohlendorf et al. 1985, Mitchell and Custer 1986). Egg-laying begins earlier in southern latitudes. Near Laguna Vista, TX, lays eggs mainly in Apr (Mitchell and Custer 1986). In San Diego Bay, CA, area, laying begins first week of Apr and continues through at least 10 Jun, including renesting attempts. In San Francisco Bay, CA, area, 1-wk-old chicks observed as early as 21 May (Miller 1943); laying must have begun as early as 25 Apr. In Grays Harbor, WA, laying begins third week of Apr and continues through most of May (Penland 1976). In Great Lakes, laying begins as early as 5 May and continues up to 1 Aug, but later clutches probably are replacements; peak laying period typically is 7 May–1 Jun (Cuthbert 1981). In Manitoba, laying from 7 Jun to 7 Jul (Vermeer 1972). Chicks hatch about 25–27 d later, and young fledge at approximately 37 days of age (see Incubation, below, and Fledgling and immature stages, below).

Second/Later Brood(S) Per Season

Only 1 brood per season, but pairs that lose nests to washouts, predation, or human disturbance before hatching often relocate and renest. In ne. Lake Michigan, 46% of birds (n = 33 pairs) that lost initial nests to storms began renesting within 2–3 wk (Cuthbert 1988). In Great Lakes, renesting continues through Jul (Shugart et al. 1978, FJC). In Utah, broods observed as late as 18 Sep, with chicks 10 d–2.5 wk old (Cottam 1946). In Grays Harbor, WA, nests with 2-wk-old chicks found as late as 25 Aug, and active nests with eggs still observed in early Sep. If these late records represent replacement clutches, about 7% of breeding pairs at this colony renested successfully (Penland 1976, 1981).

Figure 5 . Usually nests in colonies and single pair sites near or adjoining those of other birds, especially gulls and terns—e.g., Herring, California (Larus californicus), Ring-billed, Western, Glaucous-winged, and Mew (L. canus) gulls; Gull-billed, Least (Sterna antillarum), Royal, Elegant, Common, Arctic (S. paradisaea), Sooty (S. fuscata), Forster’s (S. forsteri), and Sandwich terns; also Black Skimmer (Rhynchops niger), American Avocet (Recurvirostra americana), Black-necked Stilt (Himantopus alexandrus), Snowy Plover (Charadrius alexandrinus), skuas (Stercorariidae), pelicans (Pelicanidae), and cormorants (Phalacrocoracidae; Bent 1921, Kirven 1969, Penland 1976, Cramp 1985, Martin and Lester 1990, Schew et al. 1994, Sirois et al. 1995, W. Golder and D. Paul pers. comm.).

Selection Process

Trial nest scrapes are made at several locations as soon as pairs form, and are usually defended for short periods of time (several minutes to 1–2 h; Cuthbert 1981). Roles of sexes in selection process not known. In many areas, selection appears to be influenced by elevation; highest point above water of dike or levee or highest point of island is often selected to avoid flooding (DeGroot 1931, Shugart et al. 1978, Sirois et al. 1995). However, proximity to other terns may be equally or more important than elevation. At a colony in n. California, terns were found clustered near the high points (2 m above water) of a levee; this clustering resulted in some birds taking low (<1 m) positions, even though higher positions were available in other parts of the colony (Miller 1943).

Site Characteristics

Typically open, sparsely vegetated areas; uses variety of substrates. Great Lakes birds nest on islands with sand-gravel or limestone substrate (Ludwig 1965). In San Francisco Bay, CA, nested on soft, spongy, marshy soil and on hard soil (DeGroot 1931, Miller 1943). At Hamilton Harbour, Ontario, preference for sand over pea-gravel and crushed stone was determined experimentally; in addition, sand, pea-gravel, and crushed stone were preferred over existing substrate of hard-packed ground (Quinn and Sirdevan in press). In Grays Harbor, WA, preferred sites adjacent to driftwood or along lengths of partly buried logs in sparsely vegetated or bare, open sand, possibly because driftwood and logs provide shade (Penland 1976).

Nest is depression large enough to hold 2–3 eggs, usually in vegetation-free area, elevated >2–3 m above water level. Depressions may be lined with dried vegetation, small pebbles or bits of broken clam shells, and debris (Bent 1921, Penland 1976). Sometimes nest rim contains mollusk or oyster shells, or crayfish chelipeds, and is built up elaborately like a gull nest. Some nests are piled masses of wood and stick debris; others appear primitive, with eggs merely lying on shells or in slight hollows built by birds or already present (Bent 1921, Hayward 1935, Miller 1943, Quinn 1990). In n. Lake Huron, eggs are laid in depressions in bedrock (FJC). At Klamath Lake, OR, nests were depressions in dead and decaying vegetation (Bent 1921).

Construction Process

Both sexes build by turning and scraping (Cramp 1985). Forms depression by pressing breast to ground and, with alternating motions of legs, kicks out substrate behind it. If vegetation, small pebbles, or bits of clam shells are in immediate vicinity, incubating tern lines nest with such materials (Penland 1976). Also carries nest materials short distances; during substrate experiments at Hamilton Harbour, Lake Ontario, nests on sand with rims of stone were >1 m from stone sources (Quinn and Sirdevan in press). Nest material is not added to initial scrapes made during courtship and pair-bonding period (FJC). Scraping and building of nest are most frequent at about dawn and dusk (Bergman 1953).

Dimensions

In n. Lake Michigan, average external diameter 19.5 cm, internal diameter 16.1 cm, depth 4.5 cm (n = 131 nests; FJC).

Microclimate

No information.

Maintenance Or Reuse Of Nests, Alternate Nests

From FJC. Nests with eggs are maintained in same manner as constructed; adults prevent eggs from being covered with sand as result of wind, and keep depression deep enough to hold eggs. Incubating birds continue to add to and maintain rim until eggs are hatched. Maintenance activity increases during inclement weather. Birds in n. Lake Michigan increased rim heights by >3 cm in section of colony immediately vulnerable to flooding. If nest is disturbed when chicks are small, adults call chicks away from original nest and construct and defend alternate scrapes. Depending on level of disturbance and available habitat, chicks may be moved >100 m.

Shape

Described as ovate or elliptical ovate (Bent 1921), or as subelliptical (Cramp 1985).

Size

Average measurements from different samples similar: 48 eggs reported by Bent (1921) averaged 64.5 x 45 mm (range 58–73 x 42–48); 54 eggs collected on Rock I., UT, averaged 64.39 x 44.98 mm (Hayward 1935); 30 eggs from n. Lake Michigan averaged 66.4 x 46.2 mm (range 60–70 x 45–50; FJC).

Occasional runt eggs: Two of 1,000 eggs (0.2%) examined in Michigan were approximately 35 x 30 cm and contained no yolk (FJC).

Mass

For 51 California eggs, range 53.8–76.9 g (mean 63.44 ± 5.81 SD); some eggs in sample were incubated longer than others (C. Collins unpubl.). First-laid eggs in a clutch (A-eggs) typically are heavier than second-laid eggs (B-eggs): mean 70.0 g ± 0.87 SE and 66.6 g ± 0.65 SE (n = 46 clutches), respectively. A-eggs that produced fledglings were significantly heavier than A-eggs whose chicks did not fledge; no such relationship held for B-eggs. Main effect of B-egg weight was to prolong life (Quinn and Morris 1986).

Color

Varies from pinkish or pale warm buff to very light buff; sparingly marked with small dark brown or black spots and speckles, or sometimes with large spots or irregular blotches (Bent 1921, Cramp 1985). At Grays Harbor, WA, 51% medium brown in color, 35% light brown (n = 156). Eggs at dark brown extreme accounted for 8% of sample; eggs on light brown end of spectrum (greenish brown and light blue) accounted for 2% of sample. Eggs nearly white with few markings represented 3% of sample (Penland 1981).

Surface Texture

Shell described as “lusterless and sometimes rather rough” by Bent (1921), and as “slightly more rough to the touch” than gull eggs by Hayward (1935); however, Cramp (1985) described the eggs as smooth and slightly or not glossy.

Eggshell Thickness

Mean shell thickness (with membrane) of sets of eggs from Grays Harbor, WA, collected since 1961 was 0.35 mm (n = 12 sets of eggs); mean shell thickness of eggs collected before 1933 was 0.34 mm (n = 8 sets of eggs)—not a sig-nificant difference despite fairly high levels of DDT and other contaminants in these waters during that period (Penland 1976, 1981). In Great Lakes, shells of eggs collected before introduction of DDT in 1945 (n = 18 eggs) had a Ratcliffe index of eggshell thickness (RTI = shell weight (mg)/length (mm) x breadth) of 1.537 mg/mm²± 0.134 SD, while shells of eggs collected between 1945 and 1978 (n = 13 eggs), the “DDT era,” had an RTI of 1.438 mg/mm²± 0.143 SD—not a significant difference. In 1991, eggshells slightly thicker than before 1945, with an RTI of 1.557 (Ewins et al. 1994). See also Conservation and management: effects of human activity, below.

Clutch Size

Typically 1–3 eggs. See Demography and populations: measures of breeding activity, below.

Egg-Laying

Eggs are laid 2–3 d apart (Bergman 1953, Quinn 1980, Penland 1981). See Phenology, above.

Egg-laying observed in n. Michigan (FJC): At 14:00, female sitting on nest scrape with hindquarters elevated; male nearby. Female appeared to strain periodically to push egg out. Male picked up tiny stones and tossed them on nest rim. Within seconds after egg was visible, it came out (blunt end first) quickly with copious mucus, fecal material, and small amount of blood. Male walked to female’s posterior, bent down, looked at cloaca and egg, stepped back, and took flight. Laying process took approximately 5 min. After laying, female appeared to sleep, wings drooped at side, breathing heavily, but did not cover egg lying on nest rim. At 14:09, female removed waste material from cloaca with bill and dropped it outside nest, added small stones to nest, and covered egg with body.

Onset Of Broodiness And Incubation In Relation To Laying

Incubation begins immediately with first egg laid, but brooding is less constant when there is only 1 egg in nest than when clutch is complete (Bergman 1953, Kirven 1969, Fetterolf and Blokpoel 1983).

Incubation Patches

Present in male and female (Shugart 1977, FJC).

Incubation Period

In Grays Harbor, WA, mean incubation period 27 d (Penland 1981). In Great Lakes, about 26 d (Ludwig 1965); range for first eggs 25–28 d (mean 26.1 ± 0.71 SD; n = 32 first eggs; Cuthbert 1981). In s. and central California, mean incubation period was 27 d ± 2.7 SD (n = 110 nests) and 25 d ± 2.5 SD (n = 85 nests), respectively (Schew et al. 1994).

Parental Behavior

Eggs are incubated fairly steadily by both sexes (Penland 1976). In Lavaca Bay, TX, during incubation, females typically attend nests more than males do (Quinn 1990). In Baltic, changeovers occur fairly frequently; during observations of about 80 breeding birds during a 2-h period, 12 feeding-related changeovers took place, as well as at least 12 changeovers not related to feeding. Concurrently, about 46 birds went out on fishing flights. During changeovers early in incubation, relieving bird arrives with fish for mate and gives Fish Call; when relieving bird lands, incubating bird usually rushes and seizes fish, and relieving bird begins brooding. If bird arrives without fish, it approaches nest in mild Erect Posture (see Behavior: agonistic behavior, above; Bergman 1953). Incubating parent leaves nest to defecate; in hot weather, stands off nest (Fetterolf and Blokpoel 1983).

To locate nest in crowded colony, generally uses spatial orientation, probably based on landmarks and position of neighbors and neighbors’ nests; does not appear to rely on individual features of clutch to find specific nest location. However, simultaneous choice trials indicated that parents recognize individual clutch features such as background color and spot patterns (Shugart 1987).

Hardiness Of Eggs Against Temperature Stress; Effect Of Egg Neglect

Eggs appear quite tolerant to low temperatures. After a mid-Jun storm in n. Michigan, many eggs were washed from nests and deposited on a gravel ridge as waves receded. Water temperature was approximately 5°C, ambient temperature 10°C. Of 30 eggs collected and opened, 25% had living embryos 12–24 h after being exposed to these conditions (FJC).

Preliminary Events And Vocalizations

Pipping chicks utter weak Begging Calls in response to Fish and Alarm calls (see Sounds: vocalizations, above; Bergman 1956). Parents may respond to these calls; J. Quinn (pers. comm.) observed parent trying to feed a pipping egg on South Limestone I. in Georgian Bay, Ontario.

Shell-Breaking And Emergence

Pipped eggs generally take about 1 d to hatch; a small portion require 2 d (Quinn and Morris 1986). Eggs usually hatch in same order laid (Quinn 1980). Because incubation begins immediately when first egg is laid, hatchings are asynchronous and result in chicks of different size in nest, which facilitates brood reduction (Soikkeli 1973a, Quinn 1980). In a brood of 3, age difference between oldest and youngest chick averaged 5–6 d (Bergman 1953). Hatching interval similar to laying interval, 23 d (Kirven 1969, Quinn 1980), but longer and possibly ineffective early incubation of first-laid eggs may reduce hatching interval; in Great Lakes, first-laid eggs hatched an average of 1.8 d ± 0.15 SE earlier than second-laid eggs (n = 77 clutches; Quinn and Morris 1986).

Parental Assistance And Disposal Of Eggshells

Adult often removes shells from nest cup and places them on ground adjacent to nest, or flies from nest with shells in bill (FJC).

Condition At Hatching

Semiprecocial; eyes open; covered with down; capable of limited mobility (FJC). Newly hatched chicks are seminidifugous; after down dries, chicks are capable of leaving nest cup within 3–6 h (Bergman 1953), but typically remain in nest for at least 1–2 d (FJC). Mean mass at hatching (2 Pacific Coast California estuarine colonies): 48.8 g ± 4.2 SE (n = 13) and 48.2 g ± 5.5 SE (n = 30), respectively (Schew et al. 1994). In Texas, first-hatched chicks have significantly longer culmen lengths than siblings: mean culmen length at hatching 1.65 cm ± 0.07 SD for first-hatched chicks, 1.62 cm ± .07 SD for second-hatched chicks (n = 46 chick pairs; Quinn 1980).

Ability To Get Around, Feed, And Care For Self

Chicks generally spend first week in or close to nest; within 1 wk, highly ambulatory (Kirven 1969). Can move quickly through colony to nearby water (>200 m) if disturbed by humans or terrestrial predators (FJC). On first excursion around nest, chick reacts to adult’s bill, snapping and pecking at it, and makes hesitant bending movements, similar to aggressive bending movements of adults (see Behavior: agonistic behavior, above). Also utters Begging Call to demand food; at about 15 d old, young begin to beg in crouched posture similar to Hunched Posture of adult, which they reinforce with gaping bill and loud juvenile Food Call (Bergman 1953). Swimming ability develops early, though downy young can not swim far (Miller 1943).

Growth Of Body Parts

Grows more slowly than other species of terns at temperate latitudes, probably because of physiological limitations imposed by large size. Chicks from 2 estuarine colonies in California reached asymptotic sizes of 570 g ± 4.03 SE on day 33 (n = 65 chicks), and 566 g ± 3.85 SE on day 32 (n = 84 chicks). Wing growth is linear between 10 and 35 d of age, increasing on average 8.3 mm/d (n = 149 chicks; Schew et al. 1994). In San Diego Bay, CA, chicks attained mean adult weight of 529 g over 29-d period (n = 6 chicks; Kirven 1969).

Behavior

From Bergman 1953, 1956 (Baltic colonies), except as noted. Chicks 3–4 d old leave nest when adults fly up at a disturbance; run few steps in response to parental Alarm Calls (see Sounds: vocalizations, above), then press themselves to ground; when warning cries weaken or abate, they return to nest and crouch. At about 10 d of age, chicks run for shelter to patch of vegetation or other cover, but they stop using such shelters at 25 d. Chicks form flocks (creches) and seek cover as a unit when disturbed. Chicks ≥4 wk old adopt Submissive Posture when unable to escape pursuing adults: Chick faces adult and lies prone with body from bill to tail pressed to ground; adults usually cease chasing and pecking chick (Penland 1976). Aggressive bows and calls are common during chick’s first 3 wk, especially when begging. Forward Posture (see Behavior: agonistic behavior, above) develops early. Aggressive Upright Posture (see Behavior: agonistic behavior, above), in which crest is raised as intimidation, develops when young are able to fly. When young become independent of territory, Begging and Contact calls combine with head-tossing as intimidation movement; begging and head-tossing are also used by young birds when adults return after separation from chicks. Young birds, even newly hatched, peck at and pick up small objects, especially if they resemble fish, and shake them repeatedly.

Brooding

Shared by both parents; in Lavaca Bay, TX, female brooded more than male (Quinn 1990). For first 5–7 d, depending on weather, chicks are brooded almost continuously. When 1 parent arrives with food, chick pokes head out of breast-feathers or scapulars to be fed (FJC). Parents recognize own chicks by at least 2 d of age, possibly earlier, and aggressively reject and sometimes kill foreign chicks (Bent 1921, Shugart 1977). Recognition appears facilitated by color of chick down, a cue present at hatching (Shugart 1990). Parents call their chicks by giving gentle krrr, krrr, krrr vocalizations, simultaneously ruffling breast- and belly-feathers and entire crest (Bergman 1953). Adults move chicks from nest site if in danger; to higher ground to avoid waves; and away from human dis-turbance (FJC).

Feeding

Parents begin feeding fish to chicks 0–1 d old (Kirven 1969, Quinn 1990). Typically, parent arrives with fish carried crosswise in bill, and chick seizes fish by head and swallows it headfirst (Bent 1921). In Michigan, adults observed holding 0.8-cm minnows crosswise in bill and manipulating fish lengthwise while pinching or crushing fish as if to soften it before feeding it to small chicks (FJC). At Lavaca Bay, TX, males foraged for and fed broods more frequently than females did, and feeding rates varied with time of day; highest delivery rates observed in early morning and in afternoon or evening. The following ranges of prey delivery rates to chicks <16 d old were observed on 2 islands: off times—0.46–0.48 prey/h ± 0.08–0.12 SE; peak times—1.22–2.35 prey/h ± 0.18–1.25 SE (n = 67 broods; Quinn 1990). At South Limestone I., Georgian Bay (Ontario), first-hatched chicks were fed more frequently and had higher fledging success than broodmates (Quinn 1980).

Adults typically adjust size of fish offered to age of chick (Quinn 1990), often bringing small (4–5 cm) minnows to newly hatched young (FJC). Chicks about 2 d old successfully are fed fish 5–7.5 cm long, some with heads removed (Bent 1921, Miller 1943). However, parents sometimes offer fish prey of inappropriate sizes. In Great Lakes, 2–3 small dead chicks are found annually with atypically large fish (>15 cm) protruding from their bills (FJC). At Rice I., OR, adults were observed offering chicks smelt about 20 cm long, with no success (D. Roby pers. comm.). In Texas, chicks that were fed fish too large to swallow remained stiff, unable to bend, with tail sticking out of mouths for some time, but gradually digested head and consumed entire fish (J. Quinn pers. comm.).

In Lake Michigan, 1977–1978, 57% of prey items (n = 1,219) brought to young were alewives, 34% smelt; fish captured are usually 5–15 cm long (Shugart et al. 1978). Roughly 30 yr earlier, the common food of chicks reported in Lake Michigan was lake herring (Leucichthys artedi), which has since decreased because of overfishing by humans (Ludwig 1965).

Nest Sanitation

Young back up to rim of nest and defecate on or outside rim (FJC).

Carrying Of Young

No reports of carrying.

No evidence that cooperative breeding occurs in this species.

No information on intraspecific brood parasitism; probably uncommon (J. Quinn pers. comm.). In mixed-species colonies, Caspian Tern nests very occasionally contain eggs of other larids (e.g., Ring-billed Gull) and vice versa. How and why nests acquire extra eggs is not known.

Caspian Terns have the longest period of parental care of any tern; most immatures remain partly dependent on parents and continue to be fed by them for several months (Jozefik 1969, Cramp 1985, L’Arrivee and Blokpoel 1988). In New Zealand, fledglings did not attain independence until about 9 mo of age (M. L. Barlow pers. comm. in Powlesland and Powlesland 1994). In Florida, an immature was observed in Feb following an adult and begging (M. Van der Voort pers. comm.).

After young fledge, families remain around nest site for at least 1 wk if colony is undisturbed; in disturbed colonies, families may abandon territories 2–30 d after chicks hatch (Bergman 1953, Shugart 1977). About 1 wk after fledging, immatures begin to accompany parents to feeding sites and on distant fishing trips (Soikkeli 1973a). Often are fed at sea, although some return to shore to be fed.

In late summer before migrating south, young birds disperse from natal colonies and sometimes wander; 2 Pacific Coast hatch-year birds were recovered 800 and 1,500 km north of natal colonies 2 mo after banding (Gill and Mewaldt 1983; see Demography and populations: range, below). On wintering grounds, immatures are sedentary and typically remain there for first full year (Ludwig 1965). Most birds do not return north to breeding colonies until their third summer (Gill and Mewaldt 1983). On South Limestone I., Ontario, however, J. Quinn (pers. comm.) observed an immature in May soliciting food with high-pitched Begging Call from an incubating adult. In addition, subadults (1- to 2-yr-olds) have been recovered during Jun–Jul in Great Lakes and on winter range, and of all age groups appear to have greatest tendency to wander (L’Arrivee and Blokpoel 1988).

Though diving and fishing abilities develop shortly after fledging, a long period is required to perfect fishing skills. Juveniles may fish independently of parents at about 60 d, but are rarely successful (FJC). In Texas, 2-yr-olds still missed prey more often and foraged longer each day than adults (D. Willard pers. comm. in Recher and Recher 1969).

Process of learning to fish is not well documented, but the following are observations on juveniles in n. Michigan (FJC): Adult enticed fledgling to leave colony by refusing to feed it on the ground. Adult landed next to the begging young and took flight with fish as juvenile attempted to grab it. Fledgling followed adult, giving Food Call; adult landed on water; juvenile landed next to it; adult passed fish underwater. An alternative strategy was taken when adult dropped fish in water and juvenile landed and picked fish out of water.