Saltmarsh Sparrow

Ammospiza caudacuta



Welcome to the Birds of North America Online!

You are currently viewing one of the free species accounts available in our complimentary tour of BNA. 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. Subscriptions are available for as little as $5 for 30 days of complete access! If you would like to subscribe to BNA, please visit the Cornell Lab of Ornithology E-Store or call us at 877-873-2626 (M-F, 8:00-4:00 ET).

Figure 2. Annual cycle of Saltmarsh Sparrow breeding, migration, and molt.

Thick lines show peak activity, thin lines off-peak activity. The lines for breeding and migration reflect average timing across range, but local peak and off-peak timing varies between northern and southern parts breeding range. Earliest migrants may be individual outliers. Molt initiation can vary depending on end of most breeding activity (primary sources: Hill 1965, Winder 2012; JSG, New York; WP, South Carolina; P. Pyle, molt, based on specimens. In New York, we found no evidence of body molt in early arrivals of local individuals on breeding grounds, latest April-early May; JSG banding notes). See text on migration chronology.

Figure 11. Saltmarsh Sparrow at the nest.

Nests of this species are usually built near the ground, in clumps of grass or low shrubs. In tidal salt-marsh habitats, such locations are often near the high tide line, so many nests get flooded during storms and new-/full-moon tides. Drawing by L. Zemaitis.

Figure 12. Growth curve of body mass of Saltmarsh Sparrow nestlings aligned against known age-zero chicks that were freshly hatched (mean, 95% confidence intervals).

Data from Greenlaw and Rising (1994).

© Brooke Keeney
Saltmarsh Sparrow nest.

Nests in tidal marshes, primarily in saltmeadow (supratidal high marsh above mean high water) zone dominated by Saltmeadow Cordgrass (Spartina patens). Nests are elevated above the substrate in the grass column, where some are sufficiently high to escape flooding.

© Jonathan Eckerson, Massachusetts, United States, 11 June 2016
Saltmarsh Sparrow nest with five eggs.

Nest ordinarily a simple cup. All nests elevated above ground substrate, and supported by graminoid stems along sides of nest, and sometimes by underlying semi-erect thatch. Structure variable, sometimes thin and flimsy but usually moderately bulky.

© Deirdre Robinson, Rhode Island, United States, 17 June 2016
Saltmarsh Sparrow nest with four eggs.

Egg ground color typically greenish white to greenish blue with some variation in intensity; occasionally grayish-white or bluish. Surface profusely marked by fine, variable reddish brown speckling.

© Deirdre Robinson, Rhode Island, United States, 9 July 2016
Saltmarsh Sparrow nest with four eggs.

Note possible brood parasitism; egg in upper right may be a cowbird egg, or perhaps more likely, a different female Saltmarsh Sparrow.

© Jonathan Eckerson, Massachusetts, United States, 11 June 2016
Saltmarsh Sparrow chick that has climbed out of its nest during tidal flooding.

~5 day old Saltmarsh Sparrow chick waiting out high tide above its flooded nest.

© Alyssa Borowske, Connecticut, United States, 24 August 2013
Juvenile Saltmarsh Sparrow.

Juveniles generally remain on nesting ground until molt has completed, although local movements may occur.

© Fleur Hopper, Maine, United States, 27 July 2015


Pair Formation

Not applicable; no pair bonds formed.

First Brood

Saltmarsh Sparrow begin nesting 1–3 wk after arrival on breeding grounds (176, JSG, CSE). First egg dates in sampled first nests of the season vary 9 May–9 June (Table 3) across breeding range. As expected, first nests are initiated earlier in southern populations (Maryland to New York, early to mid May) than farther north (Connecticut to Massachusetts, late May to early June) (Table 3; 185). Timing of first egg laying in spring may vary up to 10 d or so from one year to another in same area, evidently depending on prevailing weather or phenological factors related to habitat. Initial hatching peaks late May to early June in New York (JSG). Failed early and midseason nests are rapidly replaced on average; in New York, 4.9 d ± 0.5 SE (range 3–8, n = 13) (JSG), and in Maine, 2.9 d ± 0.6 SE (117). New England nests in riverine estuaries are secondarily entrained by early flooding failures of nests to the tidal cycle followed by rapid re-nesting (176, 116, 117, 186), but those exposed to lower tidal ranges in barrier beach populations farther south are not (JSG). Entrainment to a tidal cycle depends on one spring tide being higher than the other within a given lunar cycle; whether the highest tides, and most serious nest flooding, are associated with new or full moon tides varies among years (CSE).

Second Brood

Second broods are known rangewide in the species, although the true incidence of such nesting in populations of the species is unknown (K. Ruskin, personal communication; CSE). In New York and Rhode Island, where populations exhibited 2 peaks of nest initiations (176; JSG), second broods may contribute to this pattern. Elsewhere where tidal flooding is prevalent enough to cause synchronization of nests with tidal cycles, the pattern may be weak or may not occur (CSE). Timing of a mid-summer peak occurs in New York in second week of July, while in Rhode Island, it appears from mid-July to early August. New York females averaged 18.4 d ± 1.72 SD (range 15–21, n = 7) (JSG) from last to fledge in first brood to laying of first egg in second brood. DeRagon (176) estimated interval between first egg dates in 2 successive broods (first brood fledging young successfully) as 44 d and 47 d, and in 3 cases, 49 d. In New York, this interval averaged 43.0 d (range 42–44, n = 6) (JSG).

Nesting Synchrony

Where most nest failures result from spring tide flooding, nests become secondarily synchronized after such an event; correspondingly, frequency of sexual chases is greatest right after high spring tides, when nest failures and renestings surged, and decrease afterwards (117; CSE). These relationships were not detected in study populations in New York, where Saltmarsh Sparrows nest behind barrier islands, and tidal ranges and incidence of nest flooding are lower than in New England, where sparrows nest in mainland estuaries (JSG, manuscript in preparation).

Length of Breeding Season

Extreme first egg dates by region are provided in Table 3, generally mid May or early June to August; these apply to all reported regional nestings. Total length of breeding season, based on an information theory equation (187), varied from 83.4 to 99.3 d (mean = 94.2 d ± 6.23 SD) in a 4-yr sample in New York (150). Farther north in Rhode Island, with later early nests, season length varied in 4 study areas as 74 d, 85 d, 71 d, and 77 d (mean = 76.8 d ± 6.02 SD) (Table 3). These values are crude estimates from extreme seasonal nest starts (first egg dates, earliest in May, latest in August) that assume equally intense breeding in all months (176: Appendix B-2). Estimates based on data from across the species range also suggest a typical breeding season lasts about 85 d (188).

Nest Site

Selection Process

Nest sites selected by female only. Process poorly understood, but several studies show that nests are placed non-randomly with respect to vegetation and differ from sites selected by other Ammospiza sparrows nesting in the same marshes (116, 189, 190; JSG, unpublished data). Experimental broadcast of vocalizations has no discernible effect on nest placement patterns (Bayard and Elphick 2012), adding to evidence that behavior of conspecifics does not affect nest site selection (40).

Ruskin et al. (190) found nonrandom and somewhat different nest site selection in Saltmarsh Sparrow and its sister species (Nelson’s Sparrow) at a Maine site, but nest site characteristics did not predict nest fate in these cases. Although, they concluded that nest site selection is non-adaptive, additional work will be required to fully account for factors such as adult age (experience [191]), and for potential effects of predators and hybridization with Nelson’s Sparrow (rare and common, respectively, at this study site). Gjerdrum et al. (116) also found that nest site characteristics in Connecticut did not predict nest fate.


In tidal marshes, primarily in saltmeadow (supratidal high marsh above mean high water) zone dominated by S. patens (119). Relationship of species to high marsh long known by knowledgeable early naturalists (e.g., 192, 91, 140, 112, 113), and confirmed by modern, quantitative studies (see below). However, breeding birds in a young, natural (unmodified) section of Oak Beach marsh on Long Island, New York, primarily built nests, together with Seaside Sparrows, in a broad zone of upper low marsh dominated by an intermediate ecophene of S. alterniflora 40–80 cm tall (JSG, unpublished data). In New England, nests in most areas primarily fail from tidal flooding, especially during spring tides (176, 116, 117, 193). Relatively high tidal ranges and occurrence of most marshes in river or stream mouths (which magnifies the effect of tidal height) on the mainland influence this pattern. Most nests are elevated above the substrate in the grass column, where some are sufficiently high to escape flooding. From south shore of Long Island southward, most surviving marshes are sheltered behind barrier islands facing large bays. Here and southward tidal ranges are relatively low, and predation, varying inversely with increasing latitude, is the main cause of nest failure followed by flooding during spring tides or storm-driven high water (185; JSG, unpublished data).

Site Characteristics

Nests thought to be placed as high as possible in taller grasses at higher elevations in the marsh to minimize flood risk, but limited by availability of stable support, and low enough to provide cover overhead to protect against aerial predators (79). Given available information across species’ range, the general pattern is that Saltmarsh Sparrows seek sites for nests in high marsh graminoid vegetation. Most marshes—high marsh zones in the supratidal (119)—for which we have reports of patch-type use are human-modified by grid ditching, or by structures that influence tidal flow. But, siting details also vary according to marsh elevation and associated, local graminoid composition, and potentially by difference between experienced, returning and inexperienced, first-time females (no studies of latter). Deviations from widespread S. patens coverage in high marsh situations, here representing the most frequent “general pattern” mentioned above in many marshes in the Northeast, are reviewed below by state, along with special siting characteristics.

Virginia: A single outlier nest at southern margin of breeding range was studied in Four Points Marsh (unaltered) in Gloucester County. The nest was built in a dense stand of S. patens positioned against patch of J. roemerianus at edge of tide pool (84). No population information.

New Jersey: Two populations were examined near Lavallette, Ocean County, in altered (ditched) high marsh settings sheltered by barrier island on inside bay. Nests in higher, drier sections of marshes, located in patches dominated by J. gerardii and elevated in vegetation column 12–15 cm above substrate; no mention of nest canopy (10). A recent more detailed study in New Jersey affirmed the widespread pattern (above), but only relative to Seaside Sparrows, and no patch use mentioned. No information provided in this study on relationships to random availability of patch types in marsh plots (189).

New York (Long Island): Three populations were studied behind barrier island facing large inside bays: in altered (ditched) marshes (Tobay, West Gilgo), nest sites were variable, but all were in saltmeadow zone, chiefly in tufted thatch of S. patens, in D. spicata, and in open stands of shrub Iva frutescens in saltmeadow matrix, often near or beside ditches. Some nests in S. patensS. alterniflora ecotone. One nest in S. patens clump at base of Iva shrub. In unaltered, young marsh (Oak Beach), most nests in upper intertidal in medium height S. alterniflora and in patensS. alterniflora ecotone, fewer in small patches of pure S. patens hidden in tufted thatch. All nests variably elevated above substrate (127; JSG, unpublished data). During spring arrival and early post-arrival periods when first nests are initiated, marshes are predominantly brown from persistent, often flattened dead stems (thatch); early nests often placed in semi-erect tufts of thatch stems (S. patens, S. alterniflora) and under edges of wrack mats (upper intertidal) laying on tops of S. alterniflora thatch. Later in season, when growing, green stems change marsh aspect, nests placed in taller, green grass on unaltered marsh (S. alterniflora), often incorporating tops of persistent dead stems. Only at this time do sparrows weave a stem canopy over many nests, as they also do in other regions (194; CSE). As a young marsh historically, the unaltered marsh (in study area) in New York was unusual relative to other marshes in the region (Long Island) in that high marsh was restricted, so patch uses were primarily in upper intertidal in S. alterniflora. Relative to random sites on the marsh, nests tended to be placed in (early in summer) or just above thatch (later in summer) in taller, more clumped stands of S. alterniflora or under patches of drifted seaweed (JSG, unpublished data).

Connecticut: All marshes used in formal studies were altered (ditched) to one extent or another (Google Earth aerial maps), on mainland in estuarine sites: nests placed in taller, denser areas of saltmeadow vegetation with well-developed thatch layer. J. gerardii and S. patens patches commonly used, mid-height S. alterniflora in intertidal less frequently (116, 39). Nest site selection correlated with structural (thatch, graminoid stem density and height) and compositional (saltmeadow herbage) features, which collectively reflected slightly higher elevations on marsh offering some protection from flooding (39, 131, 125). Overall perspective stresses variability of vegetation use for nest sites within marshes in region (40). Models that used vegetative metrics to predict the occurrence of nesting birds performed better than those using other criteria (125) and performed well when applied to independent test plots not used for model development (131; CSE). No evidence of aggregation of nests on marshes by females (sometimes reported anecdotally in older literature); in the only formal study of this question to date, nest placement was random with respect to other nests (40). Both sexes commonly occur in portions of marsh not used for nesting (39); presence of sparrows in non-nesting microhabitats in marshes is chiefly associated with foraging and social behaviors (134; JSG).

Rhode Island: Mostly modified marshes were studied in Rhode Island, where all nests (n = 199) were in the landward saltmeadow zone, and 80% of those were in patches of S. patens, J. gerardii, and D. spicata (“high-marsh graminoids”). The rest were in patches of S. alterniflora found in depressions in the high marsh, in mixed graminoids in which no species dominated, and in salt marsh shrub (Iva)/graminoid areas (176). In a prerestoration (tidal restriction) marsh, nests were placed in S. patens, short [not dwarf] S. alterniflora, and in open edges of Phragmites australis stands. After tidal restoration, nests located in taller graminoids that permitted greater elevation on the marsh (115).

Massachusetts: Little information on patch use or other siting characteristics. On Cape Cod, nests conformed to the general pattern of S. patens use (93).

Maine: One population studied at Scarborough Marsh, where nests also located in areas dominated by S. patens, but deeper thatch immediately around the nest. At larger nest patch scale (10 m2), S. patens also dominated, although at both scales, graminoid vegetation was notably varied around nests (117).

Most recently, modeling has provided objective information on variables that are important to the species in nest site selection at local, regional, and range-wide scales (118: 59). The rangewide perspective is unusual in ecological studies. It stresses that habitat selection relative to available space is selective throughout species’ geographic range. Results are generally consistent with small-scale studies of nest-site selection reviewed above, notably that deeper thatch layers and greater cover of S. patens in comparison to random sites are widespread attributes. Across its range, most factors used to characterize the Saltmarsh Sparrow “nesting niche” vary locally.


Construction Process

Female builds cup nest, sometimes interweaving taller grasses that overhang the nest to form an overhead canopy, especially later in the season when the growing grass is taller. “Canopies” are not part of the cup nest, and thus strictly-speaking not “domes” (see Reinert [195: 91], who introduced term “canopy;” 194, who chose term “dome”). Rarely, a female may extend the back wall of the cup upwards higher than front wall, perhaps in some cases producing a partial dome (WP, CSE). Mean height of nest rim above substrate: New York, 105 mm ± 30 SD (range 35–185, n = 55) (JSG); Rhode Island, 133 mm ± 34 SD (range 70–250, n = 173) (176). In Connecticut, mean nest height was 129 mm (95% confidence interval [CI] = 18 mm, n = 27) (194). Mean height of eggs above substrate (rim–cup depth) is quite low: New York, 63 mm ± 27 SD (range 5–145, n = 57) (JSG). In Rhode Island, mean clearance between bottom of nest and substrate was 84 mm, and minimum clearance was 20 mm.

Females generally use the same approaches to nests, often via a narrow tunnel-like passage through the grass (Hill 1968; JSG). In New York, orientation of nest entrance varied from east to south in 71% of nests measured (median = SE, n = 62 nests) (JSG). In Connecticut, most nest entrances oriented to southeast (mean for nest groups was 139°) (194). Males that accidentally discover a nest during construction (or later in the nest cycle) are driven off by the female and evidently seldom return (JSG, unpublished data).

Structure and Composition Matter

Nest ordinarily a simple cup, rarely (in New York) a modified cup with one side built up and partly over cup (partially domed). All nests elevated above ground substrate, and supported by graminoid stems along sides of nest, and sometimes by underlying semi-erect thatch. Presence of a canopy above the nest (mean 11.2 cm ± 1.3 CI) in Connecticut functions to retain eggs in the nest during a flooding episode (194). Early nests, which are chiefly built into thatch before new growth is tall (New York at least), do not have a canopy other than overtopping thatch, and even later nests may not be canopied (JSG).

Structure variable, sometimes thin and flimsy but usually moderately bulky (JSG). Mean dry mass of nests collected in Connecticut was 10 g ± 3 SD (range 3.7–17.4, n = 23) (196). Nest consists of an external superstructure and a lining in cup. Superstructure built of relatively coarse, dry (dead) graminoid stems and leaf blades, and seaweed, all crudely interwoven with loose ends protruding tangentially in all directions. Finer grasses are used in lining (Hill 1968; JSG). Material appears to depend on local availability immediately around the nest where materials are collected (JSG). Overhead cover varies from very open to 100% closure, provided by naturally overarching grasses or wrack edges, and often is augmented by a constructed canopy. Data from Humphreys et al. (194) do not support the view that canopies reduce depredation risk, while the matter of nest-shading to influence nest microclimate is untested.


Cupped nest (rarely semi-domed) is hemispherical to somewhat oblong in shape. Outside diameter, mean 8.6 cm (range 7.6–10.8), inside diameter 5.3 cm (range 5.1–6.4), outside depth (rim to outside nest base) 7.1 cm (range 5.1–8.9), inside depth 3.8 cm (range 5.1–6.4) (10).

Maintenance or Reuse of Nests, Alternate Nests

No nest reuse is known. We have no reason to regard occasional nests that are completed but do not receive eggs (CSE, JSG) as anything more than those abandoned when the female was disturbed during nest discovery by a human or predator or flooded during the interval between nest checks. No evidence that female maintains nests apart from removal of fecal material, a dead nestling, and possibly an unhatched egg (which sometimes selectively disappears) during nestling development.

Nonbreeding Nests

No nonbreeding nests or use of nests after nest cycle is over.



Short ovate to ovate, elongation (length/breadth) 1.33; two dwarf eggs in same nest prolate spheroidal 1.17, 1.21 (elongation).


Egg size summarized by region: New Jersey–Massachusetts, length 19.5 mm (range 17.3–20.4), breadth 14.6 mm (range 13.7–15.5, n = 113 eggs in 25 clutches); New Jersey–Virginia, length 19.7 mm (range 18.5–20.8), breadth 14.8 mm (range 14.0–15.8, n = 55 eggs in 14 clutches) (WFVZ). Egg size within a single population in New York (JSG, unpublished data), length 19.5 mm ± 0.7 SD (range 18.4–20.9, n = 36), breadth 14.4 mm ± 0.4 SD (range 13.4–15.1, n = 36) (JSG, unpublished data). Two runt eggs found in a single nest in New York (197): 12.4 × 10.6 mm, 15.2 × 12.6 mm. Data on intraclutch variation in egg size not available; all eggs in a nest typically appear to be similar in size and shape (visual impression based on several hundred nests examined in New York, JSG).

Wet Mass

Mean mass of freshly laid eggs in New York: 2.12 g ± 0.16 SE (range 1.8–2.4, n = 32); 11% of female body mass (JSG, unpublished data).

Dry Mass

No data.

Shell Mass

Mean empty shell mass varied from 0.119 to 0.160 g (WFVZ).


Ground color typically greenish white to greenish blue with some variation in intensity; occasionally grayish-white or bluish. Surface profusely marked, often more or less evenly, by fine reddish brown (various shades) speckling. Some variation in distribution and amount of speckling may be evident within clutches. In some cases, eggs may be lightly speckled; in others, fine markings may be confluent (small blotches), especially around larger end of egg, where occasionally an egg may display a reddish “cap” (192, Hill 1968).

Surface Texture

Smooth but not glossy.

Clutch Size

Typical clutch ranges from 3–6 eggs, mode 4 eggs (JSG). Clutch size of 6 (perhaps 7) is rare (192, 198). Clutch size of 2 eggs has been reported, but may be adventitious, resulting from flooding egg loss. See Demography: Clutch for further details on clutch.

Egg Laying

One egg laid/d during early morning, often before 0600 (127). See Table 3 for regional details on extreme reported ranges of egg-laying dates.


Performed only by female (10, 113, 127). Incubation likely minimal or irregular before laying of penultimate or ultimate egg (JSG). In New York, mean incubation period 11.8 d ± 0.5 d (estimated mean error; range 11–12, n = 11) (127). Incubation patch develops only in female (127) over a period of 1 week or more between female arrival in breeding habitat and start of nest building, and remains unfeathered throughout nesting season (127).

Incubation schedule and nest microclimate studied in Connecticut using data-loggers (128). Mean inattentive period (“off-bout”) 11.8 min ± 3.2 SD. Ambient temperature affected off-bout duration and frequency; cooler temperatures related to shorter and more frequent periods. Females remained on nests at night. Night nest temperatures 78% higher than ambient temperatures (33.6°C vs. 18.9°C), and day nest temperatures 34% higher (34.7°C vs. 25.9°C). Clutch size did not affect pattern of off-bouts. Flooding during incubation, caused nest temperatures to drop 12.6 ± 2.6°C on average to a mean of 21.8 ± 1.5°C as nest goes underwater and females depart. Flooding depressed nest temperatures for 98 ± 41 min on average. Nest flooding caused partial or complete nest failure during incubation. Older eggs float during nest inundations, and may wash out of nests (194). Eggs that remain in a flooded nest often continue to develop (128). Nest canopies built in taller vegetation later in summer (see Nest) had significant effect on retention of eggs during flooding, but hypothesis that canopies provided an anti-predator benefit was unsupported (194). Hypothesis that nest canopies also may moderate nest microclimate during strong, mid-summer insolation not examined.


Hatching synchronous or asynchronous. Eggs hatch within 24 h after pipping. First evidence of hatching is a crack in the side of shell along line of greatest circumference; initial crack is extended by the embryo until shell splits into two parts. All eggs commonly hatch over a 24 to 36-h interval, indicating that incubation typically begins in earnest with laying of ultimate or penultimate egg (10, 127). Females remove eggshells but leave unhatched eggs in nest (127). In New York, 2.2% of eggs observed during hatching phase (n = 632 eggs) failed to hatch. Of these eggs, 4 examined later were “infertile" (no evidence of an embryo), and 2 were “inviable” (contained dead embryos) (JSG, unpublished data).

During nestling stage, nest microclimate changes when nests partly or fully inundated. Such nests suffered from inundations that lasted 90–100 min on average, when nest temperatures dropped about 14°C to an average of 24.1 ± 1.7°C. Most nestlings in a sample of flooded nests survived (128), but survival depends on flooding height and, presumably, the age of the chicks.

Young Birds

Growth and Development

Growth in nestling Saltmarsh Sparrow, as in other songbirds (199), is best described by a logistic function (Figure 12; see also 10). Most rapid growth occurs between day 3 and day 5. Growth characteristics in New York population (150), using method of Ricklefs (199): growth rate constant (K), 0.564, nestling asymptotic weight (A), 16.7; time taken to grow from 10 to 90% of asymptotic weight, 7.8 d. Tarsus development (10) most rapid between days 1 and 6, but not yet asymptotic at day 11. Eighth primary exhibited little growth until day 2, but afterward (days 4–11) growth nearly linear. Culmen length increased more or less steadily from about 2 mm at day 1 to about 5.5 mm at day 10. Rectrix growth minimal until day 6 when feathers about 1.0 mm, but afterward feathers elongate more rapidly to about 9.5 mm on day 11 (10). Limited published information for early ages in nest from Woolfenden (10); affirmed or extended by JSG and WP in New York.

Day 0. Nestlings of this age (hatch day) nearly naked, eyes closed, mostly helpless, but soon capable of gaping, and utter a very faint, single peep. Abdomen distended noticeably after hatch, perhaps from remnant yolk in yolk sac. Body sparsely covered with thin grayish wood-brown (5: 190) natal down on capital, alar, humeral, crural, spinal (midback only, absent near uropygium), and extreme posterior ventral (where down is whitish, very short) tracts. Down matted and wet at time of hatching, but dries within ~1 h. Skin yellowish orange; some feather papillae faintly visible. Mouth reddish orange with distinct yellow margin all around inside (outer) edge, but visible on gape when mouth closed. White egg tooth present. Mean body mass of freshly hatched nestlings (down-matted stage) 1.63 g ± 0.259 SE (range 1.0–2.1, n = 28) (127).

Day 1. Young still blind, somewhat stronger compared to post-hatch condition and can move short distances by using wings and feet. Feather papillae more visible on spinal and capital tracts, and thickened ridge of skin develops over eye where lids will form.

Day 2. Eyelids begin to open slightly as thin slits in some individuals. Visible feather papillae continue to develop and new papillae appear.

Days 3–4. Eyelids partly open in most individuals. Papillae prominent on all feather tracts. Down distribution similar to day 0. Initial appearance of teleoptiles by day 4. As judged by female brooding behavior, body temperature regulation in nestlings nearing maturation.

Day 5. Eyes fully open in most individuals. Egg tooth gone. Short teleoptiles all in sheaths. Movements stronger and better coordinated. Young still beg when disturbed.

Day 6. Eyes fully open in all individuals. Developing remiges still sheathed, but tips of body feather tracts unsheathed; still sheathed on head. Young begin to utter quiet, squealing distress calls when handled. Peep call disappeared earlier.

Day 7. Remigial sheaths become gray (previously dark blue) begin to slough off. Begging less frequent, and cowering prevalent. Breast streaking becomes apparent. If handled, first escape movements evident, but remain in nest when replaced. Distress calls, which stimulate mobbing in female, louder at this age.

Day 8. Young well-feathered at this age except on coronal region of capital tract. Breast streaks distinct. Escape behavior now well developed in most individuals, which are capable of leaving the nest; may fledge prematurely when handled (some individuals remain in nest if replaced, but others will persist in leaving). Two young that were closely observed climbed out of nest at this age and hid in a grass clump several cm from the nest (S. W. Young in Woolfenden [10]).

Days 9–10. Young normally leave nest on day 9 or day 10, and none vacated nest later than day 10. Some nestlings fledge from nests after sunset on flooding tide (128). Legs well developed, but wings still lag behind. On nest departure, all young disperse in vicinity of nest and hide in grass. Nestlings that are near or at fledging age may clamber out of nest into grass spontaneously to retreat as tidal water rises in a flooding event (128; photograph)—see Fledgling Stage below.

Parental Care


Performed only by female. Young brooded persistently during first few days post-hatch as early nestling capacity to regulate body temperature develops, but less frequently by days 4–5. Brooding periods may last at least 14 min, usually 2–6 min. On cool, windy days, female may continue to brood young that are at least 6 d old (150).


Only females feed young. In New York, female carried one larger to several smaller items at a time; several items delivered in a food bolus held together by saliva. Mean volume of food delivered/young/h (based on all food collected from young and from nest cup) was 0.18 ml ± 0.13 SE (n = 86) (150). Mean food deliveries/young/h overall in New York (1977–1978): 4.8 ml ± 1.5 SE (150). Pooled data for 4 yr in New York documented an increase in mean rate of food deliveries/young/h as chicks age: 0–2 d (age of nestlings), 3.4 ml ± 1.80 SE (n = 5); 3–5 d, 3.8 ml ± 1.98 SE (n = 17); 6–8 d, 6.6 ml ± 2.95 SE (n = 22); 9–12 d, 6.7 ml ± 3.5 SE) (n = 9) (JSG). Change in delivery rate is evidently not linear with advancing nestling age, but reflects a step-increase corresponding to timing of reduced brooding effort in days following hatching. Female may search for food > 250 m from nest (mean = 60.7 ± 3.3 [SE], n = 201 (150). She continues to feed young out of the nest for another 15–20 d (176). Female Saltmarsh Sparrow delivers food at same rate as male and female Seaside Sparrow together in same marsh over same time period (150).

Nestling dietary composition taken from Post and Greenlaw (141), but most important taxa in nestling diets studied in Maine and New Hampshire (insects, spiders, amphipods) also similar to those taxa found in New York chicks (below), and to overall adult diet (see Diet and Foraging: Main Foods Taken) (K. O’Brien and T. P. Hodgman, unpublished data). Young fed invertebrate prey exclusively. Based on comparisons to sampled food resources (availability), female sparrows preferred amphipods (Orchestia grillus) and larval Soldier Fly (Odontomyia microstoma) when foraging for nestlings. Among invertebrates inhabiting the standing vegetation, moths (adults and caterpillars, especially Noctuidae, ~ 7%) and salt marsh grasshoppers (Tettigoniidae, ~7%) were chosen preferentially. Other prey taken from the grass column was diverse, and included Diptera (chiefly Stratiomyidae, Tabanidae, ~21%), Coleoptera, and spiders (notably wolf spiders, Lycosidae). Saltmarsh Sparrow underused some small but abundant prey relative to availability, such as plant hoppers (Cicadellidae) and plant bugs (Miridae) (Hemiptera). Overall dietary importance (not considering availability), was greatest for larger but less abundant arthropods, ranked numerically as follows: amphipods, adult soldier flies, larval soldier flies, and wolf spiders. Items taken from the ground and water composed 68% of the prey. Foraging sparrows spent most time in the zones covered by dense saltmeadow grasses and by dead, medium-height cordgrass. Seasonal changes in the composition of prey that occupied the vegetation were evident, but the relative importance of items taken at ground level did not vary over time, correlated with a decrease in number of adult flies, and an increase in moth populations. The most common foraging tactics and substrates used by foraging sparrows in New York (either foraging for themselves by both sexes or for dependent young by females) was probing or gleaning mud (35% of maneuvers), gleaning new growth vegetation (23%), gleaning residual vegetation (19%), and gleaning tidally-deposited wrack (14%).

Nest Sanitation

Nests usually remain clean of fecal material; rarely, a nest late in nestling stage may become smeared with feces as active young apparently miss directing some fecal material over the rim of the nest. Young nestlings produce feces in discrete sacs that are removed in bill by adult female. Generally, she drops the sacs 10–50 m from nest on an outgoing feeding flight. Adult not observed to swallow sacs (10; JSG, WP). Dead nestlings are removed by female.

Invertebrate associates are found in the nests of this species, especially mites and ticks (Acari), amphipods (Amphipoda), and isopods (Isopoda) (196). The best predictor of ectoparasite loads in sparrow nests was found to be flooding failure. This study suggested that female sparrows that selected flood-prone nest-sites benefit from a reduction in exposure to nest parasites if a nest survives flooding.

Nest Defense

When observer approaches nest during egg-stage, female leaves and remains quiet and hidden. She utters alarm calls (Tuc, Tic, see above) during nestling and fledgling stages. No evidence of distraction display in species (JSG).

Cooperative Breeding

Not applicable in this species.

Brood Parasitism

No confirmed records known.

Fledgling Stage

In this nidicolous, passerine species, a cost of promiscuity is the unshared, continuing reproductive effort to nurture a fledged family to independence before the female can initiate a new nest cycle. Documented intervals between a successful nest and a subsequent nest by the same marked female range from 14–18 d (150, 135). A nestling that fledged on 5 July (1979) was recaptured 16 d later as an “independent” juvenile near its nest site on 21 July; it fledged at 15.9 g and was 18.0 g when recaptured. Females in Rhode Island continued to feed fledglings for 15–20 d after they left the nest (176). Fledging occurs as early as 8 days old if the nest is disturbed, but natural fledging typically occurs on days 9–10 (see Young Birds). In Connecticut, all young ready to leave at 6 nests fledged at night soon after sunset on a flooding tide (128). Late-age nestlings may avoid rising flood tides before they are ready to fledge by holding their head above the water, or by clambering higher in the nest or into the grass above the nest, returning to the nest after the water subsides in the last case (128; S. Apgar, unpublished data). During the period of dependency out of the nest, the female continues responding by calling (“mobbing”) when an intruder approaches her family (JSG). Young fledglings remain in the general vicinity of the nest, where they climb into the grass column to escape rising water as necessary during daily high tides.

Movements of telemetered females suggest that fledglings do not stay in one group during period of dependency (135); day-old fledged individuals capable of moving several m within a few min and crossing ditches of ~1 m width, although powered flight ability is limited during the first week (135; CSE). During post-fledging period, telemetered females generally stay within 100 m of nest, suggesting chicks also stay within this area, at least until independent (135).

Immature Stage

After independence, immatures in New York remained on natal marsh, but moved from inner areas to wetter, outer sections where they fed on seeding, tall Smooth Cordgrass or on exposed substrate during low tide. They roosted on low shrubby islands of dredge spoil in the outer zone. In Connecticut, adults and immatures begin disappearing from local natal sites by early August (101); adults at least evidently move to molting sites on the same marsh, while some immature birds show up in August in fairly large numbers in marshes where no nesting was known to have occurred, despite repeated searches (CSE). In both areas, individuals remained on the breeding grounds until molt completed, although local movements may occur (101; JSG, unpublished data).

Recommended Citation

Greenlaw, J. S., C. S. Elphick, W. Post, and J. D. Rising (2018). Saltmarsh Sparrow (Ammospiza caudacuta), version 2.1. In The Birds of North America (P. G. Rodewald, Editor). Cornell Lab of Ornithology, Ithaca, NY, USA.