Purple Gallinule Porphyrio martinica

Richard L. West and Gene K. Hess
Version: 1.0 — Published March 4, 2020
Text last updated January 1, 2002


Welcome to Birds of the World!

You are currently viewing one of the free accounts available in our complimentary tour of Birds of the World. In this courtesy review, you can access all the life history articles and the multimedia galleries associated with this account.

For complete access to all accounts, a subscription is required.

Subscribe Now

Already a subscriber? Sign in


Phenology given for Louisiana and Texas is generally applicable for the Southeast, but not for an unstudied resident population in central and s. Florida.

Pair Formation

Arrives unpaired in Texas; first seen 18 Apr, and 28 adults had arrived by 12 May 1974. Courtship displays first noticed 10 May; by 21 May, all individuals were paired and courtship behavior ended. First arrival 21 and 12 Apr in 2 yr in Louisiana (Helm 1982), but arrival reported as early as last week in Mar (Lowery 1960).


In S. Carolina, nest-building in early May; full clutches usual by 21 May (Sprunt 1970). These dates consistent with arrival and first brood dates in other southeastern states. Nest-building earlier in central Florida; see below.

First Brood Per Season

Earliest egg record in Louisiana 15 Apr (Oberholser 1938). In a coastal Louisiana study, clutch initiation from 1 May 1977 with peak in mid-Jun in a marsh and impoundment. In a 24-ha rice field, arrival delayed until mid-May; clutch initiation delayed until Jun and peaked in Jul, as suggested by the 11 nests located during a 7–9 Jun search and 26 nests found 18–20 Jul; these latter probably represented renesting or second nesting. This delay associated with inadequate nest cover until late May. Gallinules began using fields about 7 wk after planting, when rice attained height of 80–90 cm; continued use until harvest, about 1 Aug. These rice fields provided 70 d of favorable nesting conditions; period shorter in areas with faster-maturing rice varieties (Helm 1982).

Drought-induced low water levels concurrent with delayed nesting in marshes; e.g., first clutch initiation 23 May 1978 (a drought year), but nesting still peaked mid-Jun; total nests found depressed 55% in drought year. Rice fields relatively more attractive in drought year with 145% increase in nests found in Jun during drought, but drought conditions in Jul affected water level in rice field, leading to predation, abandonment, and end of nest activity (Helm 1982).

In Texas, egg dates 9 Apr–12 Aug, with some young barely able to fly as late as Sep (Oberholser 1974). Earliest clutch collected in Florida 12 Apr, but 8 downy chicks seen 12 Apr indicate a clutch completed by 20 March in Polk Co. (Stevenson and Anderson 1994b). Northernmost nest known, 7 eggs in Ohio, completed 18 Jun (Trautman and Glines 1964).

In Panama, eggs or newly hatched young recorded from Mar to early Nov (Wetmore 1965b). Breeding appears to occur year-round in Puerto Rico, with peaks Apr–May and late Aug–Oct (Biaggi 1970, Raffaele 1989).

Sometimes renests following nest failure in Louisiana (Matthews 1983b), but Helm (Helm 1994: 160) circumspectly states, “no definitive information on double brooding in North America,” presumably referring to a lack of studies using marked birds. Peninsular Florida population, however, like those in Central America, is nonmigratory with a long potential nesting season. Polk Co., FL, residents reported multiple broods coming from same territories, on which juveniles routinely fed downy young. Landowners could identify adult birds by shield characteristics, the way they walked, and especially by the way they interacted with people who routinely offered them scraps of bread. These were birds that were under daily observation for periods of several years from back yards that open onto lakes (RLW).

Even in migratory populations, some double-brooding occurs. In Duval Co., FL, 2 broods reared when conditions favorable (Grimes 1944, RLW). Second broods in Louisiana are implied by Helm (Helm 1982) who observed juveniles (7–10 wk old) and older accompanying adults with newly hatched chicks. A juvenile was observed feeding a downy young from a second brood produced from same territory in Leon Co., FL (RLW).

Recognized as having an extended breeding season from eggs collected in Louisiana 15 Apr 1894 (U.S. National Museum's egg collection) to downy young taken in S. Carolina 17 Sep 1887 (A. Wayne in Cooke 1914a).

Nest Site

Selection Process

In Costa Rica, usually 2–4 nests constructed prior to egg-laying; only 1 lined with leaves or grasses and becomes egg nest. Not known which bird selects nest and what criteria used in selection process. Before egg-laying begins, both pair members bring additional pieces of vegetation to selected nest and carefully weave them into nest bowl (Hunter 1986). This is the only study done with marked birds where the sex could be determined by behavior.


In a Texas impoundment, nests located in mixed maidencane and bulltongue (Sagittaria lancifolia); averaged 31.8 cm ± 8.2 SD (range 18–48, n = 12) above water (Bell and Cordes 1977).

Similarly, in a Colombian rice field (Mckay 1981), nests placed 29.8 cm ± 1.0 SE (range 21–42, n = 30) above water. Nest-site means for water depth (14.7 cm ± 0.8 SE; range 6–26), rice height (65 cm ± 1.3 SE; range 55–85), and distance to nearest border (108 m ± 16.7 SE; range 29–200) did not differ significantly from those of 30 random points, indicating that pairs nested randomly with respect to those factors. These nests often shaded on top by partial enclosures, with 44% shaded >25%, 13% not shaded at all.

Site Characteristics

Nesting in Louisiana rice fields delayed until crop grown to a height of 80–90 cm, generally about mid-May (Helm et al. 1987).

Egg nests in s. Louisiana marshes constructed in one of 3 basic vegetational strata:

(1) On floating vegetative mats such as water hyacinth, alligator weed (Alternanthera philoxeroides), or pennywort. Base of such floating nests usually constructed of buoyant plant material, especially inflated water-hyacinth petioles. Floating nests often found near center of canals and exposed to wind drift; lack overhead concealment but not affected by fluctuating water levels (Helm 1982).

(2) Anchored in base of emergents such as southern bulrush (Scirpus californicus) or bulltongue. These midstrata nests located in crotch of emergents that would not support a nest in their upper extremities. Such nests usually adjacent to open vegetative mats, but also typical of those in rice fields early in growing season. Midstrata nests anchored to emergents often flooded when water levels increase (Helm 1982).

(3) Fixed higher in dense stands of emergents, such as giant cutgrass, southern bulrush, or rice (Helm 1982). Nests located in dense, high grasses routinely placed higher in Louisiana than in the 2 situations described above: In an impoundment with stable water levels and giant cutgrass used for nest support, nest heights averaged 80 cm; in a disturbed marsh with nests supported either by giant cutgrass or southern bulrush, 24 nests averaged 32 cm but were higher above water level in a drought year with receding water levels than in prior year; in rice fields, 67 nests averaged 35 cm but were lower in drought year (Helm 1982).

Nests and nest sites vary considerably in Duval Co., FL. One found 60 cm above waist-deep water in a clump of dog fennel (Eupatorium capillifolium) in a buttonbush pond that contained several hundred nests of herons and egrets; another in a populous White Ibis (Eudocimus albus) colony was merely a shallow, saucer-shaped depression in hardened mud, lined with dead leaves. Other nests found in cattails, sawgrass, willow bushes (Salix sp.), and mixed marsh grasses (Grimes 1944). One Ohio nest in a small stand of cattails 6 m from shore growing in 40–50 cm of water, surrounded by a dense stand of yellow pond lilies (Trautman and Glines 1964).

Seven nests found near Colombian rice fields in dense growths of platanillo (Thalia geniculata) in narrow strips of marsh habitat along streams and canals were constructed of platanillo leaves and differed from rice-field nests in being higher above water (mean 56.7 cm ± 5.0 SE) and having greater outside depths (mean 15.6 cm versus 10 cm in rice; Mckay 1981). In Louisiana, Purple Gallinule nests resemble slightly larger nests of Common Moorhens but were placed about twice as high on average (Helm 1982).


Construction Process

Base/outer layer a thin, cupped layer of grass, much of it merely pulled over and pressed into position without being severed from clump on which it grew. Other blades drawn together above nest to form an arch of camouflage. Adds nest material throughout incubation, primarily at nest changeovers after relieving bird has settled onto nest (Gross and Van Tyne 1929, Helm 1982). As incubation progresses, a lining of fine grasses is added (Helm 1982).

Ramps (typical of rallids) are constructed in emergent vegetation when growth form of vegetation in which nest is located necessitates a ramp for access (Bell 1976; Helm 1982). Construction of a ramp (after clutch completion) is accomplished by pulling down an overhanging blade of grass with bill and then holding it down with one foot, dexterously weaving it into structure. Bird selects another long grass stem and repeats process until many such additions are made (Gross and Van Tyne 1929).

In a Colombian rice field, ramps of bent leaves were approximately 15 cm wide, 20 to >100 cm long, and either led directly to nest entrance or were built tangentially to it, apparently allowing bird to enter nest from either of 2 directions. Several ramps were poorly defined, and 3 nests lacked evidence of ramps or entrances (n = 30; Mckay 1981).

Structure And Composition Matter

Plant community in which nest is constructed normally dictates primary nest material, but in a Louisiana marsh, southern bulrush selected over other nearby plants. In tall grass and similar substrates, nest is firmly attached to stems; leaves from other stems are pulled down and woven into structure (Gross and Van Tyne 1929, Grimes 1944, Trautman and Glines 1964, Mckay 1981).


Mean dimensions (cm) of 10 random nests in Louisiana: outside diameter, 21.7; height, 10.9; inside diameter, 10.8; depth, 3.6 (Helm 1982). In a Colombian rice field (cm; n = 30), mean diameter 21.8 (range 16–28), outside depth, 10.0 (range 6–16); bowl depth, 5.2 (range 2–9; Mckay 1981).


No data. On hot days in Panama, shades eggs by standing over them (Gross and Van Tyne 1929).

Maintenance Or Reuse Of Nests, Alternate Or Nonbreeding Nests

In Ohio, 1 wk after all young hatched, a completed “brood” nest with runway was found about 4.6 m from first nest. Adult and young occupied that nest with decreasing frequency until the young were 3 wk old. Larger than egg nest: approximately 84 cm in diameter, 23 cm deep; depression 8 cm deep; rim 80 cm above water's surface. Adult and young used this nest with decreasing frequency until young were 3 wk old. Three additional partial nests found in area and believed to have been constructed by gallinules were never completed (Trautman and Glines 1964).



Regular ovals, sometimes biconical (Herklots 1961).


All measurements mean length × mean breadth (mm).

Louisiana. 37.3 (range 33.2–41.5) × 25.6 (range 22.133.2, n = 100 eggs; Helm 1982).

Florida. 39.8 ± 1.7 SD (range 35.6–43.2) × 28.4 ± 1.3 SD (range 24.5–29.0, n = 168 eggs; Delaware Museum of Natural History [DMNH] egg collection data).

Colombia. 41.0 ± 0.3 SE × 29.3 ± 0.2 SE (n = 53). Eggs within clutches varied by as much as 5.5 mm in length and 2.5 mm in width (Mckay 1981).

Panama. 40.1 (range 38.9–42.8) × 28.4 (range 27.1–29.8, n = 7 eggs from 2 clutches; Gross and Van Tyne 1929).


Mean 15.7 g (range 15.0–16.4, n = 7; Gross and Van Tyne 1929); 6.8% of mass of a 230-g female.


Pale creamy white to deep pinkish buff with small, irregular spots of rich brown and pale purple most abundant at larger end (Herklots 1961).

Surface Texture

Smooth with little or no gloss (Bent 1926).

Eggshell Thickness

Mean eggshell thickness in 89 clutches collected in Florida 1885–1940 averaged 0.218 mm; 15 clutches collected there 1948–1973 averaged 0.213 mm, a nonsignificant difference. Similarly eggshell thickness of 45 pre-1947 clutches from S. Carolina averaged 0.210 mm and 13 clutches from Louisiana averaged 0.214 mm (DMNH). Eggshell thickness of eggs collected in a Louisiana rice field in 1968–1969: 0.203 mm ± 0.013 SD (n = 107; Fowler et al. 1971).

Clutch Size

Varies greatly; average usually >6 in North America, about 4 in the Tropics (Table 2). In Louisiana, clutches in rice fields averaged 3 eggs more than those found in either an impounded marsh or a disturbed natural marsh in 1977 and 1978 (a normal and dry year, respectively; see Table 2 and Helm 1982). Greater availability of animal matter in rice fields may have increased egg production or egg-dumping could account for larger clutches (Helm et al. 1987).


One egg/d (Trautman and Glines 1964, Matthews 1983b).


Onset Of Broodiness And Incubation In Relation To Laying

In many nests, incubation begins before last egg is laid, usually with penultimate egg. Hatching asynchronous, extending ≥2 d (Helm 1994).

Incubation Patch

No information.

Incubation Period

Eighteen to 20 d in Louisiana (Matthews 1983b). In a Florida nest, first egg laid hatched in 21 d; other egg that hatched the same day had not been incubated >20 d (Grimes 1944). Extremes of incubation period in an Ohio nest were 20.5 and 23 d if eggs hatched in same order as they were laid (Trautman and Glines 1964). Incubation about 22 d in Panama, probably started after second egg laid; 2 eggs hatched 20 d after fourth (last) egg was laid, 2 hatched 21 d after fourth egg was laid (Gross and Van Tyne 1929).

Parental Behavior

Parents take turns incubating regularly every 34 h. Relieving bird often picks a grass stem or other nest material, then presents this to incubating bird to add to nest. This might be repeated with other nest material until incubating bird abruptly walks past relieving bird on runway and departs. Relieving bird examines eggs, turns them, and settles facing ramp (Gross and Van Tyne 1929).

In Panama, when weather warm and cloudy at midday, gallinules would leave nest unattended for short periods. With this exception, adults kept nest covered continually. As incubation continued, sitting bird would often arise, turn eggs, and settle on them again. Sometimes this was done as often as 3 times in a half-hour. Although sitting bird appeared to doze, an unusual noise would instantly arouse it and cause it to slightly raise feathers of its crown (Gross and Van Tyne 1929).

Hardiness Of Eggs Against Temperature Stress; Effect Of Egg Neglect

No information.


Preliminary Events And Vocalization

In an Ohio nest, occasional high notes heard from some of the 5 eggs pipped the same day the first 2 hatched (Trautman and Glines 1964). In Panama, by 13:30 of the day when first 2 eggs hatched, chick in third egg could be heard peeping but had not pipped shell; it had made no further progress by dark, but early the next morning, it was out of shell (Gross and Van Tyne 1929).

Shell-Breaking And Emergence

In Florida, a chick emerging from shell had kicked itself free within 2 h from time of first pipping; fourth and last egg was pipped by noon of second day and hatched by night (Grimes 1944). Hatching of 4 eggs—from pipping of first egg to emergence of last chick—required about 48 h in Panama (Gross and Van Tyne 1929). In Ohio, last chick (of 7) hatched 4 d after first 2 hatched, penultimate one only 4 h earlier; last 2 took about 4 d from first pipping to emergence (Trautman and Glines 1964).

Parental Assistance And Disposal Of Eggshells

Eggshells usually remain in nest a few hours and then are carried away by adults (Gross and Van Tyne 1929).

Young Birds

Condition At Hatching

Mean mass, after drying, of 11 incubator-hatched chicks: 13.4 g (Helm 1982); of 1- to 2-d-old chicks in Panama, 11.0 g (Gross and Van Tyne 1929); and of a dry, first-day male chick in Ohio, 10.1g (Trautman and Glines 1964).

Pollex tipped with claw 3 mm long; toes large, middle-toe length 2.60 cm. Chick completely covered with glossy black down; silver-tipped down encircling neck and also randomly dispersed over entire body; down less dense on alar and capital tracts, but pate not bald. Anterior of bill primarily black; stripe of pink just anterior to nares; irregular black band encircles both mandibles at nares; posterior portion of both mandibles bright red; area of culmen shield pinkish. Terminal egg tooth pearl white, diminishes by 12 d, and lost by 4 of 6 chicks by 21 d (Helm 1982).

Chicks are subprecocial (Skutch 1976), belonging to a group fed from adult's bill, at least until they are some weeks old. Typical of this category are rails and coots (Fulica spp.). Brightly colored bill of young gallinule may makes it a more conspicuous target for feeding. Chicks are open-eyed and slip from nest within a day or 2 after hatching, sooner if disturbed. They do not feed themselves but are capable of following parent to food.

Eggs hatch asynchronously over 3–4 d, and young normally remain in egg nest until a majority of eggs hatch, but when disturbed, chicks <6 h old scramble out of nest and swim to cover. Large feet enable rapid mobility in an aquatic ecosystem so young can swim, dive, and run rapidly across emergent vegetation. Hours-old young return to nest via runway, using both feet and wings. When wings are employed, they are partially folded in such a manner that the well-developed claw on the pollex can be hooked most advantageously into the vegetation (Gross and Van Tyne 1929, Trautman and Glines 1964, Helm 1982). Departure may be hastened by an alarmed adult giving a peculiar note, whereupon young, even just hatched, scramble from nest (Gross and Van Tyne 1929).

Growth And Development

Little plumage change occurs during first 2 wk, but tarsi and toes become light brown by 7 d. By 12 d, down loses its sheen and pate becomes covered with black silver-tipped down; red on bill lighter in intensity; area of culmen shield darkens. Neck lengthening by 12 d and prominent by 21 d. Juvenile feathers appear on ventral tract by 18 d, and on alar and crural tracts by 21 d; emerging feathers white, but appear brown as black down adheres to tips; black down retained on capital tract (Helm 1982). Body mass of Costa Rican chicks 3–4 d old averaged 15.3 g (range 13.0–18.5, n = 9); at 4 wk, averaged 100.1 g (range 84–118, n = 7); at 12 wk, averaged 177 g (ranged 163–198, n = 13; Hunter 1986).

Small young are active; great ability in climbing over tangled reeds and grass stems in response to adult's call. Outspread wings aid young in balancing their tiny bodies; little clawlike appendages at tip of manus (pollex) of wing frequently employed in holding on to a stem of climbing over an obstacle. Adept at swimming; rapidly escape toward grass and thick vegetation, concealing skillfully; extremely difficult to find (Gross and Van Tyne 1929, Stoddard 1978).

Parental Care


Adults share concurrent brooding and incubating duties (Helm 1982). Brooding period not firmly established but primarily at night after first week; chicks brooded for short periods during day and rain showers (Helm 1982). Two chicks left huddled together first day for about 20 min before adult returned. Brooding adults trample peeping chicks with indifference. When covered, chicks continue to peep vigorously. One such chick shortly wiggled forward so its head protruded through feathers of adult, which immediately pushed the chick back out of sight. Later, chicks often allowed to remain with heads out. Brooding adult sometimes arises to examine chicks, but more often it puts its head under its breast feathers without rising (Gross and Van Tyne 1929). Young may be brooded by male in a nursery nest 1 m from egg nest, but in one instance, a newly hatched first chick was pecked to death by male, who pulled off pieces of its body and ate them in an aberrant act of kin cannibalism (Hunter 1985a).


For 1–4 d after hatch, precocial young become excited when an adult peers over edge of nest, peeping and waving their heads, jumping up and grasping adult's bill. Adult offers small insects and spiders, reaching into nest with neck outstretched; food held before eldest chick, which snatches at it. First day after hatch, adults may present some inappropriate food: nest material, a too-large spider, or a water beetle too tough for small chicks. Food items delivered to Louisiana nests included mollusks, crayfish (Procambarus clarkii), dragonflies, grasshoppers, garter snakes (Thamnophis sp.), water beetles, spiders, and insect larva (Helm 1982; see also Diet, above). Animal food continued to be primary diet during first 2 wk. Chicks begin to feed themselves at 7–10 d and are primarily self-feeding by 21 d. Coontail, duckweed, pondweed, and flowers of water hyacinth were foods consumed directly by downy young (Gross and Van Tyne 1929, Trautman and Glines 1964, Bell 1976, Helm 1982).

Three phases of feeding recognized in Costa Rica: phase 1, hatching to 3–5 wk old, chicks typically follow ≥1 adults or juveniles and wait while family members gather food nearby. Once family member has food, it either walks or flies to chick and passes food from its bill to chick's bill. Chick pecks at bill of adult or juvenile and frequently shows a “begging posture” when family member approaches. In this posture, neck of chick is stretched forward and lowered below level of rump. At the same time, wings are elevated and often vigorously moved up and down synchronously. During this performance, chick frequently peeps.

During phase 2, approximately 5–7 wk old, chicks more actively solicit food from elders by walking or running to feeding family members. Lily fruit fed from bill of helper or parent as pieces are chipped off. Chicks began pecking at lily fruit during this phase, usually with little success, and rely on adults to remove at least a portion of outer, black, shell-like covering. Chicks also take some insects, spiders, and grass seeds.

In third phase, 7- to 8-wk-old chicks take lily fruit from adults and juveniles and quickly swallow it or take it to another place to eat. Adults and juveniles exhibit no aggression toward chicks and make no attempt to retrieve lily fruit that chicks carried off. Chicks become independent at about 8–9 wk old in Costa Rica (Krekorian 1978, Hunter 1986). A male successfully brought brood to independence from 4 wk old when female was killed (Hosford 1967).

Nest Sanitation

Adults known to eat feces of young (Gross and Van Tyne 1929).

Carrying Of Young

Adult reported carrying a downy chick by neck or back; adult, holding chick, flew about 75 m to a patch of water lilies and then walked another ≥50 m across pads to disappear into an area of emergent grass. This was interpreted as care of young, and observations were cited of a number of rallid species and 2 jacanas that carried young (Olson 1974b), but an act of predation or cannibalization might also explain this observation (Hunter 1985a).

Cooperative Breeding

Juveniles (9–24 wk old) commonly fed chicks in families of color-banded Purple Gallinules in which there were both juveniles and chicks in a Costa Rican study (Krekorian 1978). In this study, these families also included additional, possibly related birds in adult plumage that fed the chicks and participated in defense, but the study was not continued sufficiently long to establish the relationships of the individuals. In one group in which a second brood of chicks later hatched, juveniles of the first brood fed chicks of the second brood. This brood was presumed to be from the same pair, since the same territory was used continuously, and second brood was fed by same marked adults as well as juveniles of the previous brood. The size of this family group grew to 13 or 14, which probably included 3 successive broods of young (Krekorian 1978). In one instance, a food morsel was passed from adult to juvenile to new downy chicks (L. Reeves in Krekorian 1978). Krekorian (Krekorian 1978) lists 7 other Rallidae species for which similar behavior has been documented. A subsequent Costa Rican study (Hunter 1986) found that cooperative breeding benefits breeders and probably benefits helpers. Breeders raised more young and make fewer visits to feed chicks when they had helpers. Helpers may increase their chance of survival by remaining on their natal territories, increase production of genetically related young, and gain experience in rearing chicks, avoiding predators, finding food, and defending territory (Hunter 1986).

In North America, juveniles observed feeding chicks in Louisiana and n. Florida (Helm 1982, RLW), but familial relationships have not been established with color-banded individuals. In central Florida, where multiple broods on same territories appears to be the rule, juveniles were routinely observed feeding chicks. Although individuals were not color-banded, observers recognized territorial adults by peculiarities in plumage, soft parts, and behavior (RLW).

Brood Parasitism by Other Species

Clutch sizes of up to 13 in a rice-field study suggest possibility of laying by more than 1 adult (Helm et al. 1987). In related Common Moorhen, such intraspecific parasitism has been suggested by a number of authorities (Cottam and Glazener 1959; see Greij 1994).

Fledgling Stage

From Helm 1982; sw. Louisiana. Juvenile plumage complete, and young capable of short flights by 7 wk of age (Helm 1982). By 10 wk of age, juveniles three-fourths of adult size and capable of sustained flight. Family units maintained though 10 wk of age. Juveniles remained in parents' established territory indefinitely; agonistic territorial behavior by adults toward their young not observed. Juveniles commonly observed in small groups of 15–20 in late Aug and Sep.

Immature Stage

In an isolated Costa Rican pond (Hunter 1986), juveniles became helpers and remained on territory until they achieved adult plumage. Most young adults dispersed from study site, but some returned after several months. Returning birds became floaters. Floaters acquire territories when a breeding vacancy occurs in an established territory or by first forming a pair and then establishing a new territory. Some adults acted as helpers. Migratory populations in se. U.S. are not known to have a constraint on territories nor a recognized floater population. One bird banded as an immature returned to its n. Florida natal site the next spring (RLW).

Recommended Citation

West, R. L. and G. K. Hess (2020). Purple Gallinule (Porphyrio martinica), version 1.0. In Birds of the World (A. F. Poole and F. B. Gill, Editors). Cornell Lab of Ornithology, Ithaca, NY, USA. https://doi.org/10.2173/bow.purgal2.01