of the
FLORIDA STATE MUSEUM
Biological Sciences
Volume 26
1981
Number 3
BIOLOGY OF THE YELLOW-SHOULDERED BLACKBIRD-
AGELAIUS ON A TROPICAL ISLAND
WILLIAM POST
UNIVERSITY OF FLORIDA
GAINESVILLE
Numbers of the BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL SCIENCES,
are published at irregular intervals. Volumes contain varying numbers of pages and are not
necessarily completed in any one calendar year.
OLIVER L. AUSTIN, JR., Editor
RHODA J. BRYANT, Managing Editor
Consultants for this issue:
FRANCES C. JAMES
EDWIN O. WILLIS
Communications concerning purchase or exchange of the publications and al manuscripts
should be addressed to: Managing Editor, Bulletin; Florida State Museum; University of
Florida; Gainesville, Florida 32611.
Copyright 0 1981 by the Florida State Museum of the University of Florida.
This public document was promulgated at an annual cost of
$4,074.00 or $4.07 per copy. It makes available to libraries,
scholars, and all interested persons the results of researches in the
natural sciences, emphasizing the circum-Caribbean region.
Publication date: November 23, 1981
Price: $4.10
BIOLOGY OF THE YELLOW-SHOULDERED
BLACKBIRD-AGELAIUS ON A TROPICAL ISLAND
WILLIAM POST'
SvNopSIS: The Yellow-shouldered Blackbird (Agelaius xanthomus) nests in mangroves, scat-
tered trees in pastures, and in suburbs. In the most frequently used habitat, mangroves, birds
use isolated cays or scattered trees in salt pans. Nests are either open cups placed on branches
or are built in cavities.
The breeding season starts with the spring rains in April and May. These rains stimulate
new vegetative growth, which may result in an increase in the numbers of insects on which
the blackbirds feed. Most young are in the nest during the relatively dry summer period and
fledge before the fall rains begin in September or October.
Pairs form 6-10 weeks before egg laying within groups of birds that visit nesting areas of
previous years. During the mate acquisition period males display at old nest sites, which the
females visit. Males follow and guard females. The extended period of affiliation before
mating may be advantageous to the female if it assures male loyalty, as male parental care is
important in a species with such widely separated feeding and nesting sites. All males and
females studied during a 2-year period were monogamous.
The average clutch is three eggs. The incubation period is 12.5 days, and hatching is asyn-
chronous. Only females incubate and brood, but some males deliver food to their incubating
mates. Males guard the nest when the female is away. Both sexes deliver food at the same
rate. Delivery rates are like those of North Temperate icterids. The growth pattern of A. xan-
thomus is the same as that of Red-winged Blackbirds (A. phoeniceus). After fledging, young
follow adults for at least 24 days.
Nest success (proportion of nests producing at least one young) was 46% (54 nests). Nests
on offshore cays had higher success than those in adjacent mainland areas, because they were
parasitized less by Shiny Cowbirds (Molothrus bonariensis). Pairs nesting in cavities fledged
about three times as many young per nest as did those nesting in the open. Mortality during
the egg stage outweighed that during the nestling stage, mainly because of disruption by
cowbirds.
The average distance between nests in colonies was 16 m. Two pairs occasionally nested in
the same tree. Nesting aggregations seemed to result from active attraction among birds,
rather than common response to limited nest sites. Communal mobbing was a conspicuous
behavior of the nesting groups. The space around the nest consistently defended against con-
specifics and other species was small, extending 3 m in any direction. Males defended the nest
vicinity more often than females.
The species forages mainly in the upper strata of trees, obtaining arthropods by probing
and gleaning epiphytes, leaf clusters, and the surfaces of branches and trunks. This use of
foraging behaviors more typical of orioles and woodpeckers may reflect the scarcity of com-
petitors in habitats where the blackbirds are common, in combination with lack of suitable
ground-foraging sites (marshes) in Puerto Rico. Most food brought to nestlings is gathered up
to 2 km from nests, but commuting pairs do not follow each other to scattered foraging sites.
'The author is an Adjunct Associate Curator of Ornithology at the Florida State Museum, University of Florida. Gaines-
ville, Florida 32611.
POST, WILLIAM. 1981. Biology of the Yellow-shouldered Blackbird-Agelaius on a tropi-
cal island. Bull. Florida State Mus., Biol. Sci. 26(3):125-202.
BULLETIN FLORIDA STATE MUSEUM
The Yellow-shouldered Blackbird feeds on the ground as do other Agelaius, but unlike its
marsh-dwelling congeners, when ground-feeding it gathers mostly vegetable material rather
than arthropods. When delivering to nests, adults carry more than one item in their beaks.
They also regurgitate both plant material and arthropods to young.
The sex ratio is equal, although winter feeding flocks often held significantly larger pro-
portions of one sex. This difference is due to the greater feeding site specialization of females,
perhaps related to male dominance at some feeding stations.
Annual survival of adults was 82.4%, and of juveniles 65.5%. No difference was found
between the annual survival of males and females. The individuals considered were perma-
nent residents in the study area, although they occupied large home ranges, with an average
distance of 1.5 km between successive recaptures.
The species uses at least 26 distinct displays and vocalizations, a larger repertoire than
described for other monogamous icterids. Unlike polygamous icterids, A. xanthomus has few
intersexual differences in the quantity or quality of its displays. Each sex has only one unique
display, and the remaining displays and vocalizations are shared by both. Structural
characteristics of the action patterns and sounds of A. xanthomus make them useful for
transmitting information over short distances, in competition with the displays of close
neighbors.
SINOPSIS: La mariquita o tordo de espaldas amarillas (Agelaius xanthomus) anida en
manglares, arboles aislados, en pastizales y en zonas suburbanas. Los manglares son su
habitat preferido y ahi estos pijaros utilizan cayos aislados dispersos en bandejones de sal. Los
nidos pueden ser copas abiertas colocadas sobre ramas o pueden ser contenidos dentro
cavidades.
La estaci6n de reproducci6n comienza con las lluvias de primavera en Abril y Mayo. Estas
lluvias estimulan nuevo crecimiento vegetative, que puede resultar en un aumento en el
numero de insects que constituyen alimento para los tordos. La mayoria de los jovenes per-
manecen en el nido durante el period relativamente seco del verano y obtienen su plumaje
adulto antes de las lluvias de otofo en Septiembre u Octubre.
Las parejas se forman entire 6 a 10 semanas antes de la puesta de huevos, entire grupos de
pajaros que visitan las areas de anidaje de afios anteriores. Durante el period de adquisici6n
de parejas los machos se exhiben en los antiguos sitios de anidaci6n, que las hembras visitaron.
Los machos siguen y cuidan de las hembras. El extenso period de afiliaci6n antes del apareja-
miento podria ser ventajoso para la hembra si se asegura la lealtad del macho, ya que el
cuidado por parte del macho parental as important en species con lugares de alimentaci6n y
de anidaci6n tan ampliamente separados. Todos los machos y hembras estudiados en un
period de 2 afos fueron mon6gamos.
La camada promedio es de 3 huevos. El period de incubaci6nes 12.5 dias y el rompimi-
ento del cascar6n es asincr6nico. Solo hembras incuban y cuidan la camada pero algunos
machos llevan alimento a su pareja mientras esta incuba. Los machos cuidan el nido cuando
la hembra no estr. Ambos sexos acarrean alimento en la misma proporci6n. La taza de en-
trega de alimento es como en ict6ridos de Zonas Temperadas del Norte. El patron de creci-
miento de A. xanthomus es el mismo que el del tordo de alas rojas (A. phoeniceus). Despu6s
de abandonar el nido los jovenes siguen a los adults a lo menos 24 dias.
El 6xito del nido (proporci6n de nidos que produce por lo menos un joven) fue 46% (54
nidos). Los nidos en los cayos cercanos a la playa tuvieron un 6xito mayor que aquellos de tier-
ras firmes adyacentes ya que fueron menos parasitados por el pijaro vaquero (Molothrus
bonariensis). Parejas anidando en cavidades produjeron cerca de 3 veces mas jovenes por nido
que aquellos anidando en areas abiertas. La mortalidad durante el period de huevo fue
mayor que la durante el period de cria, principalmente a interferencia producida por el pi-
jaro vaquero.
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 127
La distancia promedio entire los nidos en las colonies fue 16 m. Ocasionalmente dos pare-
jas anidaron en el mismo arbol. El agrupamiento de nidos parece ser el resultado de atracci6n
active entire los pAjaros ms que una respuesta a un n6mero limitado de sitios de anidaci6n.
La conduct ataque o rechazo en forma colectiva fue un comportamiento conspicuo en los
grupos de anidaci6n. El espacio cercano al nido, que fue constantemente defendido contra
conespecificos y otras species, fue pequeflo, extendi6ndose 3 m en toda direcci6n. Los
machos defendieron las cercanias del nido mis a menudo que las hembras.
La especie se alimenta principalmente en la estrata superior de los Arboles, obteniendo
artr6podos a partir de un cuidadoso sondeo de epifitas, grupos de hojas y superficies de ramas
y troncos. Esta conduct de forrageo, representative de orioles y pajaros carpinteros, podria
reflejar la falta de competidores en un habitat donde los tordos son mas comunes, en com-
binaci6n con la carencia de adecuados sitios de forrageo en el suelo (pantanos) en Puerto Rico.
La mayoria del alimento llevado a los polluelos es colectada hasta 2 km de los nidos. Parejas
cercanas no se siguen hasta los esparcidos lugares de forrageo. La mariquita se alimenta en el
suelo como lo hacen otros Agelaius, pero a diferencia de sus cong6neres habitando pantanos,
cuando se alimenta en el suelo su alimento consiste principalmente de material vegetal en vez
de artr6podos. Cuando acarrean alimento a los nidos, los adults llevan mas de un item en sus
picos. Ellos ademis regurgitan para el joven tanto material vegetal como artr6podos.
La raz6n de sexos es igual, a pesar que en el invierno las bandadas representan a menudo
proporciones significativamente mayores de un solo sexo. Esta diferencia es debida a la mayor
especializaci6n de las hembras por lugares de alimentaci6n, tal vez relacionada a una
dominancia del macho en algunos sitios de alimentaci6n.
La sobrevivencia annual fue de 82.4% en adults y 65.5% en jovenes. No se encontr6
diferencia en la sobrevivencia annual entire machos y hembras. Los individuos considerados
fueron residents permanentes en el area de studio, aunque ellos ocupaban extensos Ambitos
de hogar, con una distancia promedio de 1.5 Km entire recapturas sucesivas.
La especie utiliza por lo menos 26 distintas conductas de exhibici6n y vocalizaci6n, un
repertorio mayor que el descrito para otros ict6ridos mon6gamos. A diferencia de ict6ridos
poligamos, A. xanthomus tiene pocas diferencias intersexuales en la cantidad y calidad de sus
exhibiciones. Cada sexo tiene sblo una conduct de exhibici6n siendo las restantes conductas
de exhibici6n y vocalizaci6n compartidas por ambos. Las caracteristicas estructurales de los
patrons de acci6n y sonidos de A. xanthomus hace que ellos sean muy tiles para la
transmisi6n de informaci6n a distancias cortas, especialmente cuando compite con sus vecinos
mas cercanos utilizando conductas exhibicionistas.
TABLE OF CONTENTS
INTRODUCTION ............. ............................................. 128
ACKNOW LEDGMENTS .................................... .. ............ 129
METHODS OFSTUDY ............ ......................................... 129
DESCRIPTION OF STUDY AREA ................. ......................... 130
RESULTS AND DISCUSSION ............... .............................. 132
BREEDING BIOLOGY ........................ .... ...................... 132
Nesting Habitats and Nest Characteristics ........................... 132
Breeding Season ................ .............................. 136
Mate Acquisition................................................ 139
Number of Mates Acquired and Persistence of Pairbond ............... 142
Age at First Breeding ................. ......................... 142
Pattern of Parental Care ......................................... 143
BULLETIN FLORIDA STATE MUSEUM
Spatial Pattern of Nests and Agonistic Behavior Related
to Reproduction ............................................ 150
D discussion .......................................... ......... 155
POPULATION DYNAMICS ............. ... ................ ................. 162
Reproductive Success ............................................ 162
Sex Ratio ..................................... ................ 166
Survival ............................................. 169
D discussion ..................................................... 172
B EH AVIOR ......... ................................................... 173
H om e Range ...................................... .......... 173
Food and Foraging .............................. ........... 176
Vocalizations ................................................ 182
Action Patterns ..................................... ........... 187
D discussion ..................................... ............... 193
CONCLUSIONS ...................................................... .. 197
LITERATURE CITED ..................................... ................ 198
INTRODUCTION
The recognition that social systems are integrated complexes of adap-
tations sensitive to variations in environmental conditions (Orians 1961;
Crook 1964, 1965) has generated interest in understanding the evolu-
tionary, ecological, and behavioral diversity of these systems. Theories
relating the social organizations of animals to ecological factors may be
examined by comparing the behavior of the same or related species ex-
isting under different conditions. The blackbird genus Agelaius, with nine
species, is well suited to this approach. Seven of the nine breed mainly in
the emergent vegetation of freshwater marshes. Two North Temperate
species, the Red-winged Blackbird (A. phoeniceus) and the Tricolored
Blackbird (A. tricolor), have been studied intensively in most of their
ranges. A tropical species, the Yellow-hooded Blackbird (A.
icterocephalus) has been studied recently (Wiley and Wiley 1980; T.
Manolis, pers. comm.). A fourth marsh dweller, the Yellow-winged
Blackbird (A. thilius), a South Temperate Zone species, has been studied
recently (Orians 1980). The biology of the other three marsh Agelaius, all
South American species, is still poorly known.
The remaining species, the Yellow-shouldered Blackbird (A. xan-
thomus) and the Tawny-shouldered Blackbird (A. humeralis), both of the
Greater Antilles, depart from the prevalent pattern of the genus in being
occupants of savannah and mangrove habitats rather than marshes. This
paper examines the biology of one of these insular forms, the Yellow-
shouldered Blackbird, to examine a hypothesis based on socioecological
theory, that the social system of a species may be explained in large part as
a product of two selective forces: (1) the nature and distribution of its
food, and (2) the pattern and intensity of predation. These selective forces
are believed mainly to account for the response in sociality of the species
in question.
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 129
A. xanthomus includes two subspecies. A. x. xanthomus is restricted to
Puerto Rico and has not been recorded on any of the nearby islands except
Vieques. A. x. monensis occurs on Mona and Monita Islands only. Once
common throughout lowland Puerto Rico (Wetmore 1916), the Yellow-
shouldered Blackbird is now mainly confined to three restricted coastal
regions: Mona Island, E Puerto Rico in the vicinity of Ceiba, and SW
Puerto Rico in a narrow coastal zone from Guinica to Boquer6n. The
total population is probably less than 3000, of which about 2000 are in
SW Puerto Rico, and the remainder equally divided between Mona Island
and E Puerto Rico, including Vieques. Details of the species' status and
distribution are given elsewhere (Post and Wiley 1976).
The plumage of males and females is similar. Except for the humeral
patch, which is close to spectrum yellow (Smithe 1974), the feathers of
adults are predominately blackish neutral gray (Smithe 1974), and in the
sunlight have a faint bluish green reflection (Ridgway 1902). The
humeral patch is usually edged with a narrow white margin, and under
the wing the humeral feathers are occasionally tinged with orange. When
they appear in nestlings, the feathers of the humeral tract are buff-yellow
to spectrum yellow, tipped with dark grayish brown. Most juveniles re-
tain their dark tipped humeral feathers until the first postnuptial molt (in
the fall of the year after hatching). Other regions of the nestling and
juvenal plumage are dark grayish brown. Males have wings 1.1 times as
long as those of females (102.8+[S.D.] 2.0 mm for 366 males and
93.32.0 mm for 216 females), and weigh 1.17 as much (41.4[S.D.]
2.3 g for 270 males and 35.5 2.8 g for 157 females).
ACKNOWLEDGMENTS
The personnel of the Research Division, North Carolina Department of Mental Health,
especially J. G. Vandenbergh, P. N. Witt, J. Bayles, and A. Sterling supported this study in
many ways. J. G. Vandenbergh provided vegetation data from La Parguera. I was aided in
the field by M. M. Browne and K. W. Post. D. Stephan, North Carolina State University,
identified insect specimens. The Caribbean Primate Research Center at La Parquera pro-
vided logistic support. I am grateful to the following, who criticized the manuscript: J. S.
Greenlaw, F. C. James, G. H. Orians, R. B. Payne, and E. O. Willis. During part of the
study period I was supported by U.S. Health Service Grant DA-0070. I am especially grate-
ful to E. A. Post, who supported this work in many ways.
METHODS OF STUDY
The study ran from December 1972 to February 1976. I averaged about 2 days per week
in the field during the blackbirds' nonbreeding season (October-April), and 5 days per week
during their breeding season.
I captured birds either with mist nets placed around feeding stations where birds gathered
to eat the food of domestic animals, or with manually operated wire drop traps or treadle
traps placed on feeding shelves baited with rice, sugar, or bananas. The birds were marked
with three color bands and one numbered aluminum band. During the breeding season I
BULLETIN FLORIDA STATE MUSEUM
determined the sex of captured birds by looking for the female's brood patch, or for the male's
cloacal protuberance. After I accumulated a sufficient sample of wing measurements of birds
sexed by these criteria, I sexed the birds by wing length alone.
To gather information on marked birds, I made monthly censuses of flocks feeding at
seven stations around La Parguera. During the three years of study I caught 505 Yellow-
shoulders, recaptured 114, and made 4034 sightings of color-marked individuals.
I watched low nests from burlap blinds, usually from 5 m away. I watched higher nests
from blinds placed on platforms so that my eyes were about the same level as the nest, and
observed birds that nested in tall palms through a telescope from a distance. I watched
feeding flocks from towers, houses, or other vantage points that allowed examination of all
flock members through a telescope.
To study nesting activities, I divided the day into six equal-length intervals and appor-
tioned observations equally through the day. This method also permitted comparisons be-
tween samples collected on days of different length (Verner 1965). I visited nests at least once
every other day, examining inaccessible nests by means of a mirror attached to a pole.
Calculations of nesting success are not corrected for exposure (Mayfield 1961). I collected
weight data for nestlings between 0900 and 1000. Before banding at about six days of age,
young were nail-clipped for identification. I collected food from young birds by placing a
tight-fitting pipe cleaner around their necks (Orians 1966) and extracting one hour's ac-
cumulation of food with forceps. Young that were used for food samples were not weighed.
To measure foraging behavior, I walked slowly through habitats used as feeding sites and
recorded the following information on an individual's foraging behavior: height, position in
vegetational strata (e.g. inner canopy, outer subcanopy), and the nature of five consecutive
foraging maneuvers (e.g. probing epiphyte, flycatching). After these data were collected for
one bird, I searched for the next subject. As most individuals were color-marked, I was able to
avoid studying the same bird more than once per day. I determined the rates and durations of
displays by analyzing film sequences.
The statistical measure of dispersion used in this paper is standard deviation, unless other-
wise indicated.
DESCRIPTION OF STUDY AREA
The study site was in southwestern Puerto Rico along the Caribbean
coast near La Parguera (17058'N). It included and extended 5 km SW
from the town of La Parguera (Fig. 1). Most of this area is composed of
mangrove forest, with several scrub-covered islands, mangrove cays,
pastures with scattered deciduous trees, and small settlements. With the
exception of some older red mangrove (Rhizophora mangle) stands, most
of the area at one time was cut over or burned. The principal vegetation
of this coastal zone is red mangrove, bordered on its landward edge by ex-
tensive stands of black mangrove (Avicennia germinans). The black
mangroves are often intermingled with open mud flats and salinas.
Slightly farther inland the mangrove edge grades into dry coastal scrub,
with trees such as corcho (Pisonia albida), mesquite (Prosopis juliflora),
gumbo-limbo (Bursera simaruba), pigeon-berry (Bourreria succulenta),
such shrubs as Lantana involucrata, Bumelia spp., Comocladium
dodonea, and cacti such as Cephalocereus royenii and Opuntia rubescens.
Farther inland the mangroves or coastal scrub border lowland pastures,
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY
SAL.AFfaLAY ;
S MoATITAS
0 CABALLO BLANCO
IEVA OC.VIEQUES
oPELOTA
BCOLLAO C.ENRIQUE
CARIBBEAN SEA
1km
FIGURE 1.-Main study site in SW Puerto Rico, extending WSW from La Parguera to Salinas
Arcelay. Abbreviations: Sal. = salinas: C. = cayo. Arrows indicate roost locations.
dotted with large deciduous trees such as black olive (Bucida buceras).
These savannah-like pastures, intermingled with tracts of scrub, rise
gradually toward the Sierra Bermeja, a system of hills (maximum altitude
about 92 m) that parallels the coast. Further details of the vegetation
cover of the region are given by Gleason and Cook (1926).
Offshore lie two islands (Fig. 1), El Guayacan (ca. 35 ha) and La
Cueva (ca. 20 ha), where I did much of my work. These islands are
facilities of the Caribbean Primate Research Center and harbor about
1000 free-ranging rhesus macaques Macaca mulatta. The vegetation is
mainly red and black mangrove and stunted coastal scrub, described
elsewhere (Post 1978). Many small mangrove cays, usually less than 0.5
ha, lie between El Guayacin and La Parguera (Fig. 1). The smaller cays
are composed entirely of red mangroves, while larger ones have black
mangroves on their drier interiors. At the eastern edge of the study tract,
the town of La Parguera, with a population of some 2000, covers about 75
ha.
The climate of this coastal zone is semi-arid, with an annual rainfall
ranging from 56 cm on La Cueva Island (12 year average, 1962-1973) to
68 cm in La Parguera (6 year average, 1970-1975). Rainfall has two
yearly peaks, most rain falling from August through October, with a
lesser and more variable spring peak around May.
I also studied blackbirds at two inland localities in southwestern
Puerto Rico. In the Lajas Valley I visited the agricultural experimental
station of the University of Puerto Rico, where blackbirds fed around
cattle barns and in cultivated fields. This site, 6.5 km N of La Cueva
Island, has an annual rainfall of about 107 cm (6 year average,
BULLETIN FLORIDA STATE MUSEUM
1970-1975), falling mainly during August-October. Such trees as mangos
(Mangifera indica) or royal palms (Roystonea borinquena) and irrigated
crops such as sugar cane are widespread. The vegetation of this area is
described by Danforth (1926). Still farther inland I visited the campus of
Inter American University at San German where Yellow-shouldered
Blackbirds nested in royal palms planted around the university buildings.
This locality is 12 km N of La Parguera, and its climate is humid tropical,
with an annual rainfall of 182 cm (2-year average, 1975-1977), concen-
trated during July-November. The San German site was characterized by
large ornamental and fruiting trees, such as mangos, avocados, and royal
palms, and by lawns, parking lots, stadia, and other typical university
structures.
RESULTS AND DISCUSSION
BREEDING BIOLOGY
NESTING HABITATS AND NEST CHARACTERISTICS. -Nesting blackbirds
used a variety of habitats. In the 1975 breeding season I searched as
equally as possible those parts of the main study area where I had seen
blackbirds during the previous two years. I found 36 nests in mangrove
salinas, 7 in groves of trees in pastures, 6 on islands and 1 in a town. This
does not include five nests found in suburban habitat at San German in
1975.
The most important habitat, salinas (salt pans), lies 3-5 km SW of La
Parguera, near La Cueva and El Guayacan Islands (photographs in Post
and Wiley 1976). Natural salinas are created in mangrove forests when
tidal circulation is impeded, and salt concentrations build up as the water
evaporates. High concentrations of salt kill the trees, usually black
mangroves, over tracts as large as 4 ha (Holdridge 1940). Salt manufac-
turers also create salinas by diking mangrove areas and removing trees. At
La Parguera salt processing was practiced until about 1960, but since then
most salinas have been abandoned. With the dikes broken and normal
water circulation restored, the salinas are recolonized by mangroves.
Salinas Arcelay, W of El Guayacan (Fig. 1), is such a site. Natural salinas,
and the edges of the man-made salinas, often have scattered dead trees
and stumps, intermingled with recolonizing shrubs and small tree-sized
mangroves. The water level in the salinas is slight, usually less than 10 cm
at high tide, and the area may dry out entirely at low tide.
Yellow-shoulders nested in scattered mangroves as well as in cavities
in the dead trees and stumps (photographs in Post and Wiley 1976). Birds
built open nests on main branches or on crotches of the mangroves. The
average height of 10 of these nests in the salinas near La Parguera was
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 133
0.9-0.7 m, and the average height of eight trees that held nests was
1.8 0.9 m. Birds nested in red mangroves most often (13 cases), followed
by black mangrove (10) and white mangrove (Laguncularia
racemosa)(2).
Yellow-shoulders used two types of cavities, holes in the sides of dead
trees and holes in the tops of stumps or stubs. Two nests were in cavities in
the sides of trees. The entrance of one of these was 4 x 8 cm, and the cavity
depth was 17 cm, while the eggs were 15 cm below the entrance. The
other was in a cavity whose circular entrance was 8 cm in diameter. The
eggs were also 15 cm below the entrance. The five other cavity nests were
in the tops of hollow stumps or dead tree stubs. The openings to these nest
sites ranged in size from 8 to 19 cm in diameter, and the distance of the
eggs below the entrance ranged from 5 to 30 cm. For all seven cavity
nests, the average above-water height of the eggs was 0.610.32 m,
while the average substrate height was 1.6 1.2 m.
Yellow-shoulders also nested in large (11-14 m) deciduous trees,
mainly black olives, scattered through pastures along the edge of the
mangroves (Fig. 2). Nine nests were placed on main branches or crotches
FIGURE 2.-Pasture with scattered trees near La Parguera. Yellow-shouldered Blackbirds
nested in the outer branches of these trees, mainly black olives.
^ \'
BULLETIN FLORIDA STATE MUSEUM
FIGURE 3.-Mangrove cay (W Matita), offshore La Cueva Island. Groups of Yellow-
shouldered Blackbirds nested in the red mangroves composing the cay.
at an average height of 5.6 m, range 4.3-7.6 m. Three nests were 3, 4, and
1.5 m from the main trunk. Nests were usually screened from above by
surrounding leaves, but were visible from below. Yellow-shoulders do not
build pensile (oriole-like) nests when they nest in trees, as do Red-winged
Blackbirds in these circumstances (Bent 1958). Danforth (1926) also
reported A. xanthomus nesting in deciduous trees around Cartagena
lagoon, and this habitat resembles the pastures around La Parguera.
The mangrove cays where the birds nested ranged in size from 10 m2
to 1.3 ha (Fig. 3). The average distance to the mainland mangroves of 21
offshore nests was 1153269 m, range 880-1460 m, and the average
distance of these same nests to feeding grounds on La Cueva was 813 393
m, range 380-1340 m. The minimum distance that some birds, namely
those nesting on E Matita, had to fly to mainland (including La Cueva)
feeding grounds was 1340 m.
The characteristics shared in common by the nesting cays were small
size, total coverage by red mangroves, and freedom from rats. The
relatively small cays used by A. xanthomus had only red mangroves. As
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 135
sediments accumulate under the red mangroves, the cays become drier
and black mangroves begin to colonize the interiors, while the red
mangroves there die. These older cays are larger, harbor rats, and usually
lack nesting passerines.
On the islands Yellow-shoulder usually placed their nests on main
branches or crotches. One nest was built on a red mangrove root 0.8 m
above the water. Fifteen nests averaged 2.14 1.2 m above the water.
The average height of eight trees used for nesting was 4.3 2.7 m. Tides
around La Parguera are usually less than 0.6 m, and no nest was flooded.
Most offshore nests were at or near the edges of their islands, but several
on W Matita were in the interiors. The average distance of 10 nests from
the edge of the cays, as defined by the last mangrove root, was 4.3
3.6 m.
Yellow-shoulders nested fairly high in royal palms planted on the
grounds of Inter American University in San German. The estimated
mean height of four nests was 15.3, range 10.7-19.8 m. The nests were on
the midribs of palm leaves, 1-2 m from the axil. The nest material was
woven under the leaf rib and around the leaflets that extended up on
either side of the nest. One nest was woven into the top of a young palm
leaf growing vertically from the top of a palm, and the nest was com-
pletely surrounded and partly covered over by leaflets. Palm nests were
buffeted around by the wind, and in a strong wind the movements of one
particular nest described an arc of about 1.5 m. In spite of their firm an-
choring to these leaves, nests were occasionally blown down. I found such
a nest with one dead young on 24 May 1975.
The open cup nests of A. xanthomus resemble those described for
other species of Agelaius. Nest dimensions were as follows: inside width:
8.11.1 cm (N=9); outside width: 5.91.6 cm (9); inside depth:
5.62.5 cm (8); outside depth of cup only: 11.63.1 cm (10); outside
depth of nest, including material hanging below or draped above the nest:
25.0 6.7 cm (8). All the nests I examined lacked domes, and the birds did
not appear to manipulate the vegetation to improve cover over the nest.
Surrounding foliage provided shade.
On the mainland birds nesting in the open made nest platforms of
leaves, grass, cotton, and occasionally paper, string, plastic bags, and
twine. Nest cups were made of grass leaves and stems and cotton, and
were usually lined with fine grass leaves and stems. As I worked in the
mainland nesting grounds only during the nesting season of 1975, I have
no information on the extent to which nests or nest sites there were reused
in succeeding years. Some cavity nests were used more than once during
the same nesting season, and also repeatedly from year to year, as
evidenced by the layers of old nest material beneath new nests. All seven
cavity nests that I examined were lined with grass, but material from
BULLETIN FLORIDA STATE MUSEUM
previous nests formed the platforms. All the hole nests were heavily in-
fested with mites (Post 1981).
Nests on Cays were more bulky than those on the mainland, because
they contained large quantities of sargassum. Composition of five nests
averaged 70% sargassum (range 30-90%), 12% turtle grass (Thalassia)
(5-30%), 13.6% miscellaneous wrack (plastic bags, nylon twine, bamboo
roots, pelican feathers, burlap, 3-40%), and 4% fine grass (1 nest only,
20%). The offshore nests often had material hanging under them, and
two nests had sargassum hanging on a branch above the nest, where the
females had probably tangled the material during nest construction. The
weight' of one freshly collected nest was 123.4 g. The same nest weighed
99.3 g when dry. The dry weight of another nest was 118.5 g.
Sargassum was abundant on the mangrove roots of the Cays in the
spring, but at other times nesting material may have been in short supply,
and I saw females stealing material from each other's nests. Also, on two
occasions pairs took over the deserted nests of other birds shortly after
they were abandoned. Nests remaining from previous years were used
again by the same or different pairs. In three cases the same nest site was
reoccupied, although no nest material remained from the previous year.
BREEDING SEASON. At La Parguera the start of pairing coincided with
the spring rains. The timing of the spring rains themselves may be too
variable to act as reliable cues for initiation of gonadal growth, but these
rains stimulate vegetation growth as much as the heavier fall rains
(Vandenbergh and Vessey 1968; Fig. 4), and proximate factors such as
resurgence of vegetation and associated arthropod populations could
serve as cues.
The spring growth of vegetation provides food for foliage-feeding in-
sects, such as moths and crickets, which are the main arthropod foods of
A. xanthomus, and more available protein may be necessary for females
to form eggs (Jones and Ward 1976). Several studies (Fogden 1972; Wolda
1978) have correlated the growth of arthropod populations and new
vegetation in tropical sites. The rate of leaf growth in red mangroves is
highest during spring-summer (Onuf et al. 1977; J. G. Vandenbergh,
unpubl. data). The larvae of a moth, Ecdytolopha, which feeds on the
leaf buds of red mangroves, appear at the time of new leaf growth. This
moth is a main food source for nestling Yellow-shoulders (see below,
under Food and Foraging).
Nestlings were found throughout the summer dry period before the
September rains began and stimulated vegetative growth again. The peak
in new vegetation growth, and presumably in insect availability, was in
October (Fig. 4), the period when young were becoming independent of
their parents (Fig. 5). The peak in the fruit crop also occurred at this time,
Vol. 26, No. 3
oc
w -
O
0
-J
16 60 uL
0
14 NEW LEAVES .50
12 .. ... 40
0 .. .40 C
I-~
LL 8 .. -30
z -"- ...-.---- < M
S6 FLOWERS 20
4 z
10 0
J F M A M J J A S O N D o
MONTH
FIGURE 4.-Phenology of vegetation in relation to rainfall at La Parguera September 1963-August 1974. The left ordinate refers to amount of rainfall
(cm) recorded per month, and indicated by heights of histograms. The right ordinate refers to the percentage of sample trees having the categories of
foliage referred to. The vegetation data are from Vandenbergh and Vessey (1966) and J. G. Vandenbergh (unpubl. data).
MOULT
C PAIRING
WL --------- EGG-LAYING 00
< 4 --------- ----------
'- 24- D. EP_ EN DE.NT. YOUNG....
1--
z
W 16 .. '..
0
LL
0 12
0 8
4
< J F M A M J J A S O N D
cc
MONTH z
FIGURE 5.-Annual cycle of the Yellow-shouldered Blackbird in relation to rainfall, La Parguera, 1973-75. Rainfall: histograms; dotted line: number
of nests with first egg.
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 139
and the young ate fruit after they left the nest. I saw adults feed bananas
and granulated sugar to young that were following them.
Post-nuptial molt of Yellow-shouldered Blackbirds at La Parguera oc-
curred after the fall rains started. I first saw replacement of remiges on 3
August, and the last molt of wing feathers on 20 December. Unfortu-
nately, I was unable to follow the sequence of molt in my study popula-
tions in detail, but the few data from captures and from museum
specimens show that molt is heaviest during October and November.
Some birds that I caught in August were reproductively active and were
also undergoing molt. For example, a female captured on El Guayacan on
9 August 1973 had a hard egg in her oviduct and was starting to replace
her primaries (her two number one primaries were half grown). Another
female caught on 16 August 1973 had a vascularized brood patch, while
her primaries 1-2 were new.
MATE ACQUISITION. -Pairs established themselves in the nesting areas
of previous years. Individuals and small groups of birds visited these
localities all year, and some males began singing and sporadically defend-
ing sections of nest trees as early as late December. I saw no definite signs
of heterosexual affiliation until late March (1975), about six weeks before
eggs were first laid that year. In late March in scattered pasture trees at
Lanilla groups of up to 15 birds gathered mornings after leaving the roost
and stayed for about an hour every day before leaving for feeding grounds
on El Guayacin. Similar groups gathered again in the evening on their
way to roost. Birds arrived on the nesting grounds singly or in small
groups, and not apparently as pairs. They often joined to form larger, but
poorly integrated groups whose members alternated singing with bouts of
foraging or resting. Pairs probably formed within these assemblages.
Pairing apparently depended on two behaviors: males displayed at
specific sites and followed females persistently. Males visited the nest sites
of previous years, usually standing in the nest or at its edge, alternately
singing and jabbing or pulling at old nest material (nest advertisement
display; see below). Nearby birds of either sex often sang within a short
interval, probably in response to the first bird's displays. During this
period of group visits to nesting grounds, the activities of individual males
became more localized, and the male seemed to make the initial choice of
a future nest site, which females then visited. Besides localizing their ac-
tivities, males also began to follow females persistently around the nesting
areas, and defended the females by singing, supplanting, and giving bill-
up displays.
Persistent following and guarding lasted while birds were on the
nesting grounds, but after leaving the males stopped following the
females. For example in early April 1975 at Lanilla I watched a morning
gathering of 15 birds composed of 4 pairs and 7 unassociated birds. I later
TABLE 1. -Histories of pairs nesting near La Parguera, 1974-1975.
Distance Between (m):
Nest and
nearest Nest and
communal nearest
Date Nest Successive feeding communal
Male Female Nest locality Started Nest Fate Nests area roost Comments
W Matita 24May74 deserted before
completion
W Matita 8Jun74 deserted; con-
tained 3 eggs
W Matita 11Aug74 deserted 24 Aug;
contained 1 egg
E Matita 2Jun75 depredated 6 Jul;
contained 1 young
W Matita 19Jun74 depredated, con-
tained 2 eggs
E Matita 26Jul74 deserted when
partially built
Central
Matita
Collao
29Jul74 nest completed,
but deserted 7Aug
14Jul75 failure due to
faulty nest con-
struction; eggs
in nest
12 1260
9 1260
275 1420
2200
1600 male disappeared after
nest failure; female moved
to Collao, mated with AOOB
280 1420
125 1340
male:1950
female: 1945
480 3300 the former mate of
AOOB disappeared after
23Jun75
GABR XARG
AOOB RAYR
AOOB XARG
RARG A-
OAYY BABX
YARB ARBO
AYGY RYRA
W Matita 5Jul74
Collao 1Sep74
Collao 4Jun75
Salinas 3Jun75
Carlos
25Jul75
Salinas 31May75
Carlos
Salinas 6Jul75
Carlos
Salinas 12Aug75
Carlos
W Matita 21Aug74
Cayito Vieques 9Jul75
Collao Jul75
young fledged
6 Aug
young fledged
1 Oct
young fledged
5 Jul
young fledged
3 Jul
depredated 13Aug;
contained eggs
depredated, con-
tained young
fledged young
5 Aug
nest deserted,
contained eggs
deserted?
depredated; con-
tained eggs
?
1260
480
480
720
720
720
720
720
1260
700
480
young depredated after
fledging?
after failure, pair
moves to Collao
nest not found, but pair
defending a territory
BULLETIN FLORIDA STATE MUSEUM
checked flocks containing these same color-marked individuals that were
feeding on El Guayacin, and I saw no evidence of continued pair affilia-
tion in the foraging flocks. In late April when the large winter feeding
flocks began to break up, individuals associated as pairs more persistently,
but even at the height of the nest-building period, males rarely defended
the area around the female when the pair was away from the nesting
grounds. Rather, the sexes maintained affiliation primarily by following
each other.
NUMBER OF MATES ACQUIRED AND PERSISTENCE OF PAIR BOND. -The
Yellow-shouldered Blackbirds I studied in SW Puerto Rico were
monogamous. During 1974 and 1975 I watched 25 marked pairs involved
in 43 nest attempts. I found no cases of polygamy, and only one instance
of mate-switching during the breeding season, a case in which the mates
of both birds had disappeared. I recorded 17 cases of renesting within the
same year, 16 of which involved retention of the previous mate: 9 cases of
one renesting and 4 cases of two renestings.
I have no evidence that Yellow-shouldered pairs are maintained dur-
ing the nonbreeding season, but the birds' returning to their old nesting
areas and nest sites increases the chances that they will pair again with
their previous mates. The four pairs that I studied in 1974 all reformed
in 1975 (Table 1), even though in two cases their last nestings of 1974
were failures.
Within the same nesting season pairs stayed together in spite of
repeated failures, and in several instances the pair moved together to a
new nesting locality, such as in the case of one pair that moved 1250 m
from Cayito Vieques to Collao (AYGY and RYRA: Table 1).
AGE AT FIRST BREEDING. Although I have no information on the age
of first breeding of males, I found no evidence of a surplus population of
unmated males, such as is characteristic of polygynous icterids (Orians
1969). Moreover the sex ratios among groups of birds that I captured at
different times and places during the breeding season were close to unity.
I found no males nesting the first summer after their hatching year;
some females bred the first summer. I color-marked 11 males and 29
females in juvenal plumage, most of them during their first summer and
fall. For the first summer after hatching year I have information on the
status of only two of these, both females. The first, ARYO, I caught 3
August 1974 on El Guayacin. In June 1975 she mated with a male I had
banded as an adult in 1973. This pair nested successfully in a cavity on the
mainland opposite El Guayacan. The second juvenile female was ARRO,
captured on 6 July 1974 on El Guayacin. On 13 July 1975 I found her
defending a nest in Salinas Carlos containing three Yellow-shoulder and
four cowbird eggs. Her mate was an unmarked adult. The nest and eggs
were deserted 22 July.
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 143
PATTERN OF PARENTAL CARE.- The average clutch of 30 undisturbed
nests at La Parguera was 3.03 0.32. There was no significant difference
in the average clutch sizes of birds nesting on islands and on the
mainland, nor between those using cavity or open nests. The distribution
of clutch sizes was: 2 eggs, 1 nest; 3 eggs, 27 nests; 4 eggs, 2 nests. Because
of the small variation in this sample, and because of the presence of
cowbird young, I have no way of judging whether one blackbird clutch
size was more successful than another. The average combined clutch size
of 22 nests with cowbird and blackbird eggs was 5.59 1.33. The most
cowbird eggs contained in any Yellow-shoulder nest was six, in a nest with
three blackbird eggs. All nine of these eggs hatched, the nest fledging two
blackbirds and three cowbirds.
Incubation began after the second egg was laid, and hatching was
asynchronous. I have information on the length of the incubation period
for only one nest, which contained three blackbird and two cowbird eggs.
One blackbird egg hatched 13 days after it was laid, or 12 days after in-
cubation began, and another hatched 13 days after it was laid and after
its incubation began.
Only females incubated and brooded. Females also stayed on the eggs
at night while males usually left the nest sites to join communal roosts.
One female at San German spent 77% of the daylight period incubating,
while another nesting offshore La Parguera incubated 72% of the
daylight period (Tables 2-3). The average duration of 37 daytime incuba-
tion bouts was 23.516.7 min, range, 2.5-99.5 min (two nests). The
average time that the two females spent off their nests during these same
periods was 9.68 min, range 1-36 min (N = 32). The longest continuous
incubation time, 99.5 min, was recorded for a female on 19 July 1974 be-
tween 11:16 hrs and 12:55 hrs. Her eggs hatched on 21 July. I detected no
diurnal variation in time spent on the nest nor in the length of incubation
bouts. At San German the maximum percentage of any diurnal interval
that was spent incubating was 86% (diurnal interval 5; 1433-1644 hrs),
while the minimum percentage was 57% (diurnal interval 3; 1011-1222
hrs). There may have been a slight reduction in time spent incubating
during midday, but this probably depended on the degree of shading sur-
rounding vegetation afforded the eggs.
At one nest in San German the female brooded her young (less than
three days old) 45% of the daylight period, while at another nest offshore
La Parguera the female brooded young of the same age 16% of the
daylight period. The difference in these brooding times may be related to
the fact that the San German nest was more exposed to sunlight, and I
noted that the female in San German sat on the nest more in the middle of
the day than at other times.
I saw females regularly brooding young through day five (counting
BULLETIN FLORIDA STATE MUSEUM
TABLE 2.- Time budgets of a pair of Yellow-shouldered Blackbirds in the nest vicinity at
San German, Puerto Rico.'
Percentage of day2 spent in various activities during:
Incubation period3 by: Nestling period4 by:
Activity Male Female Male Female
At nest 1.1 0.6 6.0 10.0
Incubate or brood 0 76.6 0 45.1
Forage 10.5 1.6 11.1 0.3
Rest 20.3 2.8 3.3 0.8
Body maintenance 11.5 0.9 0.1 0
Aggression 1.6 0.3 4.6 0.1
Flight 1.7 0.5 3.2 2.9
Gone 53.3 16.7 71.7 40.8
'Nest No. 175, located in royal palm on campus of Inter-American University.
'Samples, each at least 1 hr long, were spread equally through day.
'Observations conducted 20-27 May 1975. Male was watched 11.0 hr; female 10.3 hr.
'Observations conducted 2-6 June 1975, when young were less than 5 days old. Male and female were watched 5.7 hr each.
TABLE 3. -Time budgets of a pair of Yellow-shouldered Blackbirds in nest vicinity in La
Parguera, Puerto Rico.'
Percentage of day2 spent in various activities during:
Incubation3 Early nestling4 Late nestling
period by: period by: period5 by:
Activity Male Female Male Female Male Female
At nest 0 0.2 3.1 10.4 2.6 8.2
Incubate or brood 0 71.7 0 15.5 0 5.5
Forage 0 0 1.8 0 2.9 2.6
Rest 20.0 2.0 4.0 0.9 12.8 6.8
Body Maintenance 6.5 2.8 1.1 0.1 0.8 1.5
Aggression 0 0.7 0.9 0 1.1 3.0
Flight 5.5 1.3 12.8 4.9 11.7 6.5
Gone 68.0 21.3 76.3 68.2 68.1 65.9
-Nest 1B74, located on an offshore island (W Matita).
'Samples were spread equally through day.
'Observations conducted 14-19 July 1974. Male watched 2.2 hr; female 12.2 hr.
'Observations conducted 23-25 July 1974, when nest contained 2-3 young, 1-4 days old. Male watched 1.7 hr; female 5.3 hr.
'Observations conducted 26 July-5 August 1974, when nest contained 2 young, 6-15 days old. Male watched 7.7 hr; female
7.5 hr.
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 145
the day of hatching as day 0). For example, on 27 June 1975 a female on
Collao brooded her three young (five days old) six times in 1 hr 16 min, for
a total of 11.5 min. The next day, during the same diurnal interval and
under the same weather conditions, she did not brood during a 1 hr obser-
vation period.
At about the same time that females stopped brooding regularly dur-
ing the day, they began leaving the nest sites at night, flying with the
males to communal roosts. A pair nesting on W Matita in 1974 left their
single (7-8 day old) young to go to the communal roost on Caballo Blanco
at 3 min before sunset on 25 June, and at 22 min before sunset on 26 June.
Both sexes delivered food at equal rates (Table 4). The average rate of
delivery per nest per hr for all nests that I watched (11 nests; 65 samples,
73 hr) was 12.7 5.8. The average delivery rate per young per hr was
5.4 2.0. These delivery rates are about the same as reported for
temperate zone blackbirds in which both males and females feed the
young equally (cf. Willson 1966). I found no significant diurnal variation
in food delivery rates (Table 5), but I did detect a difference between the
rate of delivery to mainland and to offshore nests. This difference was
possibly related to the fewer Yellow-shoulder and cowbird young in the
offshore nests, as the number of deliveries per young per hr was the same
for the two habitats.
Males and females both cleaned the nest cup. When standing on the
nest rims, adults often pecked at their legs and nest surface, probably to
remove mites. I saw both sexes preen feathers of the young, and they were
apparently effective in keeping them free of ectoparasites. I found no
warble flies (Philornis) on any of the 17 young (from seven nests) that I ex-
amined closely. In Puerto Rico, larvae of warble flies burrow under the
skin of other species, such as the Pearly-eyed Thrasher (N. Snyder, pers.
comm.). Both males and females removed fecal sacs (Table 4), carrying
them off or eating them.
The average nestling period was 14.6 1.3 days, range 13-16 (N = 10
young in five nests). The initial growth pattern of nestlings of both sexes is
similar to that of comparable Red-winged Blackbird nestlings, and the
growth pattern of both species is best described by the logistic curve. The
growth rate (K; Ricklefs 1967) of the Yellow-shoulder young that I
weighed was 0.458, while the time required for the young to grow from
10% to 90% of asymptotic weight (28.5 g) was 9.6 days. In comparison,
the growth rate of a sample of Red-winged Blackbirds (Holcomb and
Twiest 1971) was slightly higher (K=0.484; tlo,=9.1). Red-winged
Blackbirds leave the nest sooner than do Yellow-shoulders; the average
nestling period for Red-winged Blackbirds is 10.5 days (Beer and Tibbitts
1950). The main difference between the growth of Yellow-shouldered and
Red-winged Blackbirds is in the upper portion of their curves (Fig. 6).
TABLE 4. -Comparative reproductive efforts of males and females when young were in the nest.
Total No. of food No. of No. of
Observation deliveries No. of distant nest in- fecal sac
time (hr) to nest feeding flights spections removals
Nest No. Male Female Male Female Male Female Male Female
1B74' 13.2 46 60 15 8 5 17 7 5
3741 12.1 32 35 9 11 12 22 0 0
1742 5.1 34 35 1 1
2752 2.2 10 9 3 4 2 2 0 2
775 6.7 51 58 52 57 0 5 6 14
875 6.0 53 57 52 56 0 2 3 8
975 6.6 73 44 54 25 2 0 7 4
15751 8.1 62 78 6 3 2 2 12 18
33B75 3.0 18 11 1 1 0 1 2 0
4575 4.2 25 28 1 2 0 1 1 2
Total 67.2 404 415 193 167 23 52 39 54
'Nests on offshore islands, La Parguera.
'Nests at San German. All others in salinas at La Parguera.
TABLE 5.-Food delivery rates to nests, 1974-1975.
Diurnal Interval Average'
1 2 3 4 5 6 for day
Deliveries per nest per hr
offshore nests (N2 = 32; 39hr)
mainland nests (N= 33; 34hr)
Deliveries per young per hr
offshore nests (N =32; 39hr)
mainland nests (N = 32; 33hr)
14.6 12.6 13.6 10.9
8.3 7.9 11.3
18.9 15.1 14.8 13.5 11.7 17.3
5.1 5.0 5.1
5.6 4.1 5.2
'Each diurnal interval was given equal weight in calculating the average for the dal
N is number of sampling periods.
' = P<.05 that averages were different due to chance alone (Student's t test).
RWBB
3
..... YSBB
5 6 7 8 9 10 11
AGE (DAYS)
0
3
c.
12
'P
o
FIGURE 6.-Growth patterns of Yellow-shouldered and Red-winged Blackbirds. Solid circle is average daily weight of Yellow-shoulder nestlings.
Number by each point is sample size. Dotted line connecting circles is growth curve of Yellow-shoulder, presented as percentage of asymptotic weight
(28.5 g.). Dashed line is Redwing's growth pattern, as percentage of its asymptotic weight (37.0 g), based on data in Holcomb and Twiest (1971).
Method for calculating asymptotic weight follows Ricklefs (1968).
100.
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 149
Red-winged Blackbirds leave the nest soon after attaining asymptotic
weight, while Yellow-shoulders remain 4-5 days after the growth rate
falls off (days 7-8). The more sustained growth of Red-winged Blackbird
nestlings when they are larger is perhaps made possible by the high pro-
ductivity of temperate zone marshes, and has been selected for in the con-
text of higher predation rates.
Species of the humid tropics have lower average growth rates than
temperate zone species, and Ricklefs (1976) discussed the possible reasons
for this difference. In comparison to other tropical species, the Yellow-
shoulder has a higher growth rate. For example, 30 species of the humid
tropics that weigh less than 100 g have an average growth rate of 0.387
(Ricklefs 1976). The growth rate of Yellow-shoulders is thus closer to that
of temperate zone species of comparable size (K = 0.502; N = 51 species;
Ricklefs 1976). No data are available on growth rates of passerines in the
arid or semiarid tropics, but it may be found that other species nesting in
such environments also exhibit growth characteristics more similar to
temperate zone species than to species of the humid tropics.
After fledging, young remained in the nest vicinity for extended
periods, and both parents continued to deliver food to them. Two young
in one nest on W Matita fledged on 6 August 1975; both were still 5-6 m
from the nest on 15 August, and one was still on the island within 50 m of
the nest on 20 August. Both young had left the nesting island by 21
August.
After they left the nest vicinities, young followed their parents, giving
begging vocalizations and a contact call, pink, while flying. The adults
also used the pink call (see below). In the late summer and early fall I
often saw one or two juveniles flying after adults as they visited monkey
feeders. Adults fed these young monkey chow as well as other car-
bohydrate material such as bananas. Adults continued to feed the young
for relatively long periods after they left the nest. Two young birds that
fledged on 1 October 1974 were still following and being fed by their
father, RARG, on 25 October.
As in most other monogamous passerines, males spend considerable
time in parental care, either directly (feeding young and female, nest
sanitation) or indirectly (territory defense, warning for predators). Dur-
ing the initial stages of nesting, females invest more time in reproduction,
but after the young hatch, males probably spend about the same amount
of time as females. (1) Females build the nest alone, but they are occa-
sionally accompanied by males on trips to gather material. For one nest at
San German, the female made 18 trips with nest material in 0.5 hr. In this
case the male did not accompany her, but at the time he was engaged in
mobbing. The one-way distances flown by this female were 60 m (16
trips), 140 m (1 trip), 240 m (1 trip), for a total of 2680 m/hr. (2) Only
BULLETIN FLORIDA STATE MUSEUM
females incubate, and at two nests the females spent 72-77% of their time
in this activity (Tables 2-3). Also, females alone brood, and at the same
two nests they devoted 16-45% of their time to brooding. While females
are on the nest, males may deliver food to them, although this behavior
was variable, occurring in only two of the four pairs that I watched
closely during the incubation period. At San GermAn the male of one nest
began delivering food to his female when she started incubating, about 12
days before the first egg hatched. This male delivered food to her at
regular intervals, but infrequently: 18 deliveries in 12 hr. I did not see
males feed incubating or brooding females at the nests I watched around
La Parguera: in 17.5 hr of watching nest 1B74 during the incubation and
early'nestling period, I never saw the male deliver food to the female.
Similarly, at 274, during 4.4 hr of observation, the male did not deliver
food to the incubating female. (3) During the period the female was on
the nest, males defended the nest area and engaged in mobbing predators
(Table 6). When females were off the nest area, males were often present
and they apparently guarded the nest. Also, during the rest of the nesting
cycle males spent more time in nest defense than did females. (4) Males
began feeding young at the same time as females, and the frequency of the
male's visits to the nest in some cases exceeded that of the female's (Table
4). For two nests on W Matita I measured the minimum distances flown
by the parents on foraging trips to the mainland opposite their nesting
island: during the 2.5 hr observation period the females made 17 trips
totalling 32 km while the males travelled 45 km (24 trips). (5) Both males
and females feed the fledged young.
The male's investment in reproduction is high: although he neither in-
cubates nor broods, he expends considerable time in nest defense. Males
also feed females, which may be crucial in allowing them to stay on the
nest when eggs or small young are threatened by weather or predators.
After the young hatch, males make as many food deliveries as females. As
some feeding sites are 2 km or more from the nests, the greater energy ex-
penditure by males may be significant because of the high metabolic cost
of flapping flight. Perhaps more significantly, males continued to feed
young for long periods (at least 24 days) after they had left the nest. In this
period the male's role may be not only important in providing food, but in
protecting the naive young. Parental division of labor may be most crucial
after the brood has dispersed (Smith 1978).
SPATIAL PATTERN OF NESTS AND AGONISTIC BEHAVIOR RELATED TO
REPRODUCTION. -Yellow-shouldered Blackbirds usually nested in colo-
nies. The average distance between 15 occupied nests in salinas habitat
was 15.9 7.6 m, range 5-35 m. The average distance between three off-
shore nests was 11.5 m, range 11-12.5 m. Pairs sometimes nested in the
same tree: two nests in a pasture tree were 3 m apart, and two others in a
Vol. 26, No. 3
TABLE 6. Nest attentiveness.
Nest Nest
number contents
1B742 3 eggs
2-3 young,
1-4 days old
2 young, 6-15
days old
3742 1 young, 6-9
days old
7753 3-4 young,
6-11 days old
8754 2 young, 5-6
days old
9754 3 young, 9-10
days old
1Sample periods spread equally through day.
5Open nests on offshore island (Matita).
aOpen nest in salinas.
.Cavity nest in salinas.
Duration
of sample
(hr)
12.2
5.3
7.7
11.9
6.8
6.1
4.4
Male was away
from nesting
area
58.9
75.5
60.7
67.6
91.6
80.6
87.7
Percentage of day' that:
Female was away
from nesting
area
37.2
68.1
53.0
69.9
87.3
86.4
73.8
Nest was
unguarded
12.2
50.6
43.7
56.3
77.7
69.3
60.0
--
---
BULLETIN FLORIDA STATE MUSEUM
palm, 3.5 m apart. In 1977 two pairs nested simultaneously in different
holes in the same dead tree at Salinas Arcelay (J. Wiley, pers. comm.).
Adults defended only the immediate vicinity of the nest (Table 7) and
spent long periods of time feeding away from the nesting area. Without
experimentation it is difficult to determine to what extent the spacing of
populations is influenced by social interaction or by common response to
environmental factors. Birds settling to breed may use the presence of
conspecifics as a cue to the suitability of a potential nesting area (Orians
1961), and such behavior would be advantageous to naive or unsuccessful
breeders, especially if the displays of others and the presence of old nests
indicated good nesting grounds. Alternatively, if suitable nest sites are
limited, birds may be forced to settle close together. Such a patchy
distribution of nest sites may be in part responsible for spatial contagion
of Yellow-shoulders as well as for some populations of Brewer's Blackbirds
(Horn 1968).
The following observations suggest that nesting aggregations of
Yellow-shoulders resulted from active attraction among birds. (1) On dif-
ferent days groups composed of the same individuals gathered in the
nesting areas and left together. (2) When disturbances occurred in a
nesting area, all the nesting pairs deserted together, some relocating
together at distant sites. In April 1975 my putting up blinds and markers
on W Matita apparently disturbed the birds that were gathering there in
the morning visiting old nesting sites. Members of this group, the same
pairs that nested on W Matita in 1974, did not settle to nest there in 1975,
but moved to other nearby islands (Table 1). (3) Birds visiting the nesting
areas in early spring moved around together in loose groups composed
mainly of pairs. Individuals and pairs displayed and answered each other
frequently by singing. (4) Pairs whose nests failed or who were delayed in
breeding stayed around the nesting colonies for long periods. (5) Although
I did not sample vegetation, apparently suitable habitat around nesting
groups remained unoccupied. Nests were close together and could have
been spaced out more. All nesting pairs that I studied flew long distances
to obtain food for themselves and their nestlings. For 750 deliveries to the
nest, at least 48% of the food was obtained on distant feeding flights. In
some years groups used certain nesting cays successfully but deserted
others, while the same pairs used apparently identical cays nearby.
In salinas habitat shortage of suitable nest sites, particularly for hole
nests, may have contributed to aggregation. But nests in pastures were
close to each other, while surrounding apparently identical habitat was
unoccupied. In San German royal palms were common, but pairs nested
in adjacent trees or in the same palm, leaving many sites with palms unoc-
cupied.
Birds in nesting groups responded in concert when mobbing such
Vol. 26, No. 3
O
TABLE 7.- Intraspecific nest defense by pairs of Yellow-shouldered Blackbirds.'
Distance from Nest (m) O
Category of Behavior 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 Total
Number of Intrusions 25 20 15 12 16 11 7 0 4 110 O
Number of Intrusions
at which Male was present 2 6 7 0 2 2 6 3 38
Number of Intrusions
Challenged by Male 1 4 7 0 0 1 0 2 15 (54%)2
Number of Intrusions at
which Female was present 10 6 15 11 13 9 7 4 75
Number of Intrusions
Challenged by Female 1 1 3 4 3 0 0 0 12 (16.0%)2
'Data recorded for 3 adjacent nests on W Matita Island. La Parguera. 1975. Total observation time was 41 hr.
PA significantly larger number of intrusions was challenged by the male (x =20.6; P< .005; d.f. = 1).
0
0
BULLETIN FLORIDA STATE MUSEUM
predators as humans, Little Blue Herons (Egretta caerulea), and Smooth-
billed Anis (Crotophaga ani). The mobbing response of individuals varied
according to the stage of their nest. When colony members first spotted a
predator, incubating females often stayed on the nest and gave occasional
check calls. When the predator approached, they usually quietly flew off
the nest, perched 10 m or more from it, and gave checks, chwips, or re-
mained silent. The intensity of mobbing by both males and females in-
creased as the young became older. I never saw adults diving at predators
when their nests contained eggs, but they did so commonly when older
young were in the nest. Non-nesters that frequented the nesting grounds
also joined in communal mobbing. They, as well as neighboring nesting
pairs,'joined in mobbing within the defended area of the pair whose nest
was threatened. In salinas habitat Yellow-shoulders may gain some
predator protection from the proximity of Black-necked Stilts: I was often
distracted by the constant calling and movements of the stilts when I was
searching for blackbird nests. Interspecific nesting aggregation has been
demonstrated to be advantageous in defense against predators (Goransson
et al. 1975).
In March and April when pairs first formed, males showed rudimen-
tary defense of the nest sites they visited as members of pairs or groups.
About the same time, the males began defending females as they traveled
around with them on the nesting grounds. But males did not defend
females when they were away from the nesting sites, nor the females later
in the nest cycle when they were away from the nest site itself. Such per-
sistent defense would have been pointless: I noted that females were sex-
ually receptive only in the vicinity of the nest and then only in the early
stages of the nest cycle. As activities became more localized, the males'
defense of females coincided increasingly with defense of nest site areas. If
an old nest remained from the previous nesting season, the female showed
signs of defending it from other Yellow-shoulders, but I most often saw
site defense by females only after they had actually constructed or rebuilt
a nest, and site-related intraspecific aggressiveness by females decreased
after the young were 5-6 days old, about the time females stopped
brooding.
Throughout the nesting cycle intraspecific defense of the nest site was
primarily the males' role. The area regularly defended around the nest
was small, extending about 3 m in any direction (Table 7). At the three
offshore nests I watched, breeding and non-breeding Yellow-shoulders
frequently entered the defended areas of nesting birds, and these intru-
sions were often left unchallenged, especially by the female (Table 7).
Defense was most intense during the nest construction and egg-laying
periods, probably because at these times neighboring pairs often stole
each others' nest material. As only females incubated and brooded, they
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 155
spent more time guarding the nest, but when they were off territory,
males were usually near the nest. One nest containing eggs was left
unguarded only 12% of the time, although the female was away from it
37% of her time (Table 6). After the eggs hatched, the nests were left
unguarded 44-78% of the day. During periods when nests were un-
guarded, other Yellow-shoulders often visited them, inspecting their con-
tents and sometimes even standing on their rims. Other than stealing of
material, I saw no cases of interference with nests, nor did I see any
assistance in the rearing of the young, as has been reported for some other
icterids (Orians and Orians 1977).
Both sexes defended nest sites against the intrusions of other species,
and as in intraspecific defense, males engaged in this activity significantly
more often than females. For the island nests that I watched, the species
that were challenged near the nest, Green Heron (Butorides striatus) and
Yellow Warbler (Dendroica petechia), were the only others occupying the
islands. GABR, a male nesting on W Matita in 1974, attacked Green
Herons whenever they moved under his nest. Green Herons may have
been predators of nests, but a pair on Collao in 1975 built their nest only 3
m from an occupied Green Heron nest. Yellow-shoulders at San German
seldom invaded each other's defended areas, and I saw no intraspecific
nest defense around the two nests that I watched. But other species com-
monly invaded the defended areas of the Yellow-shoulder nests, and in 14
hr at one nest I saw the Yellow-shoulders challenge Canary-winged
Parakeets (Brotogeris versicolurus) 17 times (14 challenges by the male, 3
by the female), Shiny Cowbirds 13 times (all by the male), and Greater
Antillean Grackles (Quiscalus niger) 5 times (4 by the male, 1 by the
female). The pattern of interspecific nest defense (Table 8) reveals that
many intrusions, particularly by Shiny Cowbirds, were ignored. While
the male challenged only 28% of 46 Shiny Cowbird intrusions, he
challenged 78% of the 18 Canary-winged Parakeet intrusions when he
was present, a significant difference (Table 8). The different response of
the male to two intruding species may have been due to the more surrep-
titious behavior of the cowbirds.
DiscussioN. -The Yellow-shouldered Blackbird shows behavioral
plasticity in its use of breeding habitats and nest sites. In addition to those
situations enumerated above, the species also utilizes black mangrove
forest in E Puerto Rico and cliffs on Mona Island (Post and Wiley 1976).
Also on Mona, Barn6s (1946) reported A. xanthomus nesting in cacti
(Selenicerus) in scrub habitat.
At La Parguera I could detect no consistent pattern between the use of
nesting habitats and feeding areas. Each habitat used for nesting was
quite different, but the birds shared the same communal feeding sites
(e.g. monkey feeders) or foraged independently in the same feeding
TABLE 8.-Response of a male Yellow-shouldered Blackbird to intruders near its nest.'
Distance from Nest (m)
Category of Behavior 0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 Total
Intrusions by Shiny Cowbirds 9 7 11 6 19 15 3 4 1 10 85
Number of Intrusions at
which Male was Present 3 6 3 4 17 7 3 2 0 1 46
Number of Intrusions
Challenged by male 3 2 1 3 2 1 0 0 0 1 13 (28%)2
Number of Intrusions by
Canary-winged Parakeets 0 0 5 1 6 11 1 0 0 2 26
Number of Intrusions at
which Male was present 4 1 4 8 1 0 18
Number of Intrusions
Challenged by male 1 1 4 7 1 14 (78%)2
'Observations made in San German in 1975. totalled 14 hr. The nest contained eggs or voung less than 5 da old.
AA significantly larger proportion of intrusions by Canary-winged Parakeets were challenged (x = 13.0: P<.005: d.f. 1).
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 157
habitats (e.g. pastures, mangrove-scrub edge). Only birds that nested in
the savannah-like pastures commonly fed within the same habitat, but
the pasture-nesting birds also flew long distances outside this habitat to
gather food for nestlings at garbage dumps and monkey feeders.
The long distances that nesting birds flew to foraging sites may be a
consequence of their preference for nest sites inaccessible to predators on
small islands or isolated mangrove salinas. They probably expended
relatively large amounts of energy on flights to feeding areas (see below,
under Food and Foraging).
Birds that nested over water placed their nests lower than pairs
nesting in terrestrial habitats such as pastures or suburbs, where they
chose inaccessible sites, usually the high tips of branches or palm fronds.
Other than those placed in cavities, nests in all habitats were readily visi-
ble. The relative scarcity of avian predators in coastal Puerto Rico may
have encouraged this lack of crypticity. Thus the main selection pressure
in molding habitat and nest site choice in A. xanthomus appears to come
from ground predators, all of which are recent immigrants to Puerto
Rico. It is difficult to envision what nesting habitats A. xanthomus might
have used before the arrival of Europeans and the mammalian predators
associated with their arrival. Possibly blackbirds nested closer to the
ground in terrestrial habitats, such as scrub, or in the emergent vegetation
of freshwater marshes, such as the extensive marsh that once covered the
floor of Boquer6n Valley (Danforth 1926).
The occupation of such a variety of habitats and nest sites by the
Yellow-shoulder is probably possible only under the conditions of lowered
competition for nest sites. On islands species often exhibit ecological
release, one manifestation of which is an increase in the average number
of habitats occupied per species (Cox and Ricklefs 1977). I found A. xan-
thomus using all available cavities in salinas habitat. In continental
regions, icterids rarely nest in cavities (cf. Nero 1956b, Maxwell et al.
1976, Orians, Erckmann, and Schultz 1977), perhaps because of the
presence of other species competing for these limited sites. The only other
birds that I found using cavities in the Puerto Rico salinas were Stolid
Flycatchers (Myiarchus stolidus) and their smaller size probably enabled
A. xanthomus to outcompete them for holes.
The cues that tropical birds use to time their breeding vary with
species and region: photoperiod, rainfall, and changes in vegetation and
insect populations may be stimuli (Keast and Marshall 1954, Marchant
1959, Crook 1964). Breeding may be timed so that the most food is
available for dependent young (Lack 1954). Alternately, breeding may
only commence when females have attained proper body condition (Jones
and Ward 1976, Greenlaw 1978). For A. xanthomus the spring growth of
vegetation and insect populations most likely act as the ultimate cues for
BULLETIN FLORIDA STATE MUSEUM
the initiation of breeding, by allowing females to accumulate sufficient
energy reserves.
The long delay between pairing and egg-laying in the Yellow-shoulder
may reflect the delay in the availability of suitable food. At La Parguera
pairing began in late March in 1975 but egg-laying did not start until late
May. Similarly for Yellow-shoulders on Mona Island, Barn6s (1946) noted
a long delay between the start of gonadal growth and actual deposition of
eggs. Five males and two females he collected in late March and early
April had enlarged gonads, but he found no eggs that year until early
July.
While egg-laying coincides with a peak in new vegetation, most
dependent young are in the nest during the summer dry period, when
there are few new leaves (Figs. 4-5). This may have several advantages.
First, the nestling period of some tropical species may be timed to occur
outside the rainy season because of the difficulty parents have in obtain-
ing food in wet weather. Insects are less active during rain because of im-
peded flight and lowered temperatures (Foster 1974), but as A. xan-
thomus seldom catches insects from the air, and most of the arthropods it
takes are inactive during the day regardless of weather conditions, this
may not be an important factor. Second, as Morton (1971) proposed for
the Clay-colored Robin (Turdus grayi), it may be advantageous for
nesting to occur during periods when predators are less active, such as a
dry season. Third, it may be advantageous for breeding activity to be
timed so that young leave the nest when food is abundant. Most Yellow-
shoulder young fledge just before or during the September rainy period.
The peak of new vegetation growth and fruit, and probably of associated
arthropod populations, occurs at this time. Postnuptial molt also begins
then, and abundant food would be available to partially dependent
young and to molting adults. Similarly the monkey breeding season at La
Parguera appears to be timed so that females are lactating during this
autumn period of new vegetation and fruit growth (Vandenbergh and
Vessey 1968). Other insectivorous birds, such as Yellow Warblers and
Gray Kingbirds (Tyrannus dominicensis), also have young in the nest dur-
ing the dry summer period, while granivores and nectivores, such as
Ground Doves (Columbina passerina) and Bananaquits, begin most of
their nesting activity during the September rainy period.
In species such as the Yellow-shouldered Blackbird, which do not de-
fend a territory, mate choice may be based primarily on the
characteristics of the individual. But as males also display at prospective
nest sites, a female may base her choice, at least in part, on the suitability
of the site. Characteristics of the male that might be reliable predictors of
its fitness or future commitment to offspring are frequency and intensity
of displaying, any behavior that might indicate readiness to defend the
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 159
female or nest site from competitors or predators, and persistence in
following or guarding the female when she is away from the nest site.
Predictors not relating to male characteristics per se relate to the suitabil-
ity of the nest site at which the male displays. Even if these sites were not
used ultimately for egg deposition, the male would presumably select or
defend a similar site. Criteria for a good nest site would be safety from
predators, stability and shelter afforded to nest structure, proximity to
other nesting pairs (as group mobbing may be an important anti-predator
behavior in the Yellow-shoulder; see below), and distance to feeding
grounds.
In species such as the Red-winged Blackbird and Yellow-headed
Blackbird where males establish territories several weeks before females
arrive, territory features may be the main influence on female choice
(Orians and Christman 1968). Red-winged Blackbird females may assess
male suitability by estimating the vulnerability of the nest site (male's ter-
ritory) to predation and by estimating food availability on the territory,
but they probably do not choose a male on the basis of its future nest
defense or its future investment in reproduction (Searcy 1979).
In species that pair away from their nesting grounds, such as the
Brewer's Blackbird and Common Crackle, the choice of mates must be
based more on the characteristics of the individual. In the Brewer's
Blackbird pairing occurs in foraging flocks (Williams 1952), and pairs ar-
rive on the prospective colony already mated. The paired birds stay
together, and the female chooses and defends a nest site and then builds a
nest (Horn 1970). This species shows a high degree of heterosexual coor-
dination through contagion of precopulatory displays and mate guarding
(Horn 1970). Such coordination is characteristic of species that pair away
from their nest sites and also have large, overlapping home ranges (Wiley
1976b). The Common Crackle is similar to the Brewer's Blackbird in that
the pair bond also emerges from interactions of the sexes in group ac-
tivities away from the nesting grounds. Mate guarding by the male and
antiphonal singing between paired individuals are common means of
coordinating activities between the sexes (Wiley 1976a, b, c).
In the Tricolored Blackbird the sexes arrive on the nesting grounds in
mixed flocks, and territory establishment and nest building are completed
in several days (Orians and Christman 1968; Payne 1969). The very small
territory of the male and the rapidity of pairing may mean that mate
choice is based more on male characteristics than in the Red-winged
Blackbird (Orians and Christman 1968).
The mate acquisition pattern of the Yellow-shouldered Blackbird is
similar to that of other nonterritorial icterids, whose pairing is not con-
tingent upon the male's establishing and defending a fixed site over a
period of time. Initial maintenance of the pair bond depends on two proc-
BULLETIN FLORIDA STATE MUSEUM
esses. First, the sexes meet in the nesting area of previous years, and
repeated contact at these sites may facilitate individual recognition.
Wing-raise displays (see below, under action patterns) are exchanged be-
tween individuals, providing opportunities for individual identities to be
learned. Second, the association of individuals with specific sites may pro-
vide additional information on the identity of the partner. After repeated
encounters on the nesting grounds, birds begin to associate as pairs per-
sistently and independently of site.
Yellow-shouldered Blackbirds pair long before (6-10 weeks) they
begin to breed. An extended period of affiliation may be advantageous to
a female if it reduces the changes of male desertion once insemination has
occurred. The female's ability to force loyalty by a long period of court-
ship is possible if the male considers prior investment when deciding
whether to desert (Trivers 1972; Weatherhead 1979). Long periods of
association before breeding should be found in species like the Yellow-
shouldered Blackbird and Common Grackle (Wiley 1976b) in which the
male contributes significantly to parental care.
In the Yellow-shoulder it is also possible that a long period of affilia-
tion before mating is a means of assuring rapid initiation of breeding in
response to irregular events. In A. xanthomus breeding may be delayed
until the spring rains, when sufficient protein for egg formation becomes
available (see above, under breeding season).
In being monogamous, the Yellow-shoulder fits the general pattern of
Icteridae, where 45 of 57 species have been determined to have this
mating arrangement (modified from Orians 1972). Monogamy may be
maintained for a variety of reasons, and the specific ecological condition
for each species must be analyzed separately, although in general
monogamy is favored if the individual's fitness is increased by aiding its
mate in caring for the young. The individual male's choice whether to in-
vest additional time and energy in the progeny of one mate is weighed
against the chance of success of this brood, with and without the in-
dividual's aid, as opposed to switching aid, in whole or part, to the prog-
eny of another mate (Dawkins and Carlisle 1976; Maynard Smith 1977).
If it is in the interests of a female A. xanthomus to assure her mate's fidel-
ity, the possible means to this end are: (1) a long courtship before mating,
(2) aggressiveness toward intruders near the male or nest, (3) rejection of
cuckoldery when she is away from the male (Gladstone 1979), (4) spatial
coordination of intrapair activities (Horn 1970; Wiley 1976c), and (5)
mutual displaying and vocalizing with the male (Wiley 1976a).
In Icteridae the evolution of polygamy is associated with two
ecological conditions: (1) extreme disparity in resources of defensible ac-
tivity spaces (Verner 1964, Orians 1969), and (2) limited nest sites. Both
of these routes for the evolution of polygamy assume that the conditions
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 161
for male emancipation exist, i.e. the male's fitness will not be reduced if
he reduces aid to his first brood. But the requirements of male aid are not
alone sufficient to explain the maintenance of monogamy in situations
where food is plentiful and the breeding season is extended over a long
period. Under these conditions, for example, if male aid was still re-
quired, serial polygamy might evolve (Martin 1974). In addition, the con-
ditions necessary for polygamy to evolve may not occur under some en-
vironmental regimes, such as when breeding takes place during a short
period (Weatherhead 1979). Factors that reduce the chances of finding
more than one mate are: (1) most potential mates in the population are
already paired, (2) many other competitors for mates are still unmated,
(3) availability of resources necessary to raise young may decrease with
time, (4) most resources needed to attract a mate are already claimed, and
(5) receptivity of potential mates drops with time (Boucher 1977). A
number of species that group their nests, among them the Yellow-
shouldered Blackbird, the Greater Antillean Grackle (Post, unpublished
data) and the Common Grackle (Howe 1979), are not usually found to
have more than one mate simultaneously. In the Yellow-shouldered
Blackbird and Great Antillean Grackle, scarcity of predator-free nest sites
may force pairs to build their nests in places that are not near feeding
grounds. From a consideration of the distribution of the Yellow-
shouldered Blackbird's food and the fact that it does not engage in
cooperative foraging, one would predict that nests should be regularly
distributed (Horn 1968; Waser and Wiley 1979), as in another icterid that
feeds on arboreal arthropods, the Jamaican Blackbird (Nesopsar niger-
rimus) (Wiley and Cruz 1981). For A. xanthomus I suggest that the best
nest distribution in relation to food is not possible because of a scarcity of
suitable nest sites. Having to nest in inaccessible places (mangrove islands
and salinas) forces feeding parents to commute long distances, and in such
situations male emancipation may not be possible.
Some of the primarily monogamous species are polygamous under
conditions of limited nest sites. Williams (1952) found a population of
Brewer's Blackbirds in which some pairings were polygamous, but the
more general situation was described by Horn (1968), who found that
Brewer's Blackbirds nesting in sagebrush, where nest sites were plentiful,
were all monogamous. The Yellow-shouldered Blackbirds that I studied
were monogamous in all the habitats where I examined them: 25 marked
pairs nesting on mangrove islands, pastures with scattered trees,
mangrove salinas, and palms in suburban areas.
Birds nested in groups on isolated islands from which they flew as far
as 2 km to reach feeding grounds. Birds also aggregated to nest on the
fronds of tall palms, the outer canopies of large trees in pastures, and in
clusters of isolated mangroves growing in wet salinas. As in open colonies
BULLETIN FLORIDA STATE MUSEUM
of Brewer's Blackbirds (Horn 1968), group defense may be an important
antipredator strategy for the Yellow-shoulder, and the large number of
displays used by mobbing Yellow-shoulders is correlated with the impor-
tance of joint defensive behavior.
Delayed breeding has been reported for other icterids, but the
mechanism whereby breeding is delayed is not fully understood, as ex-
perimental removal of older male Red-winged Blackbirds demonstrates
that first year males are capable of holding territories and breeding
(Orians 1961). Possibly older Red-wings have the advantage over younger
individuals of prior tenancy as well as earlier seasonal development of ag-
gressiveness (Wiley and Hartnett 1976). In the Yellow-shoulder competi-
tion for limited nest sites, in combination with lowered breeding success
for young birds of both sexes, may select against their breeding activity.
Thus the factors that contribute to delayed breeding in Red-wing males
may operate for both sexes in the Yellow-shoulder. As the species is
monogamous and the adult sex ratio is equal, females have the same
chance of breeding as males. Further, in a sedentary species such as the
Yellow-shoulder, dominance related to prior occupancy could be crucial,
especially as both sexes prospect for mates on their old nesting grounds
and use their old nests as display sites. Flocks that visit the nesting areas in
early spring are composed of breeders of previous years. Already
established social units may play a role in allowing access to mates.
Possibly delayed maturation in both sexes is more widespread in Icteridae
than has been assumed; recent studies of the nesting of first-year female
Red-winged Blackbirds show that they breed later within the same season
that do older females (Crawford 1977).
POPULATION DYNAMICS
REPRODUCTIVE SUCCESS. -Pairs nesting on offshore mangrove cays had
higher reproductive success than those nesting in mainland habitats
(Table 9), even though they often flew longer distances to gather food for
their young. Cay-nesting blackbirds had lower fledgling/nestling success
than the mainland nesters at La Parguera, and their higher overall success
is due to the greater proportion of eggs that they hatched (Table 9).
Yellow-shouldered Blackbirds that nested in cavities fledged over
three times as many young per nest as did pairs using open nests in the
same habitat (Table 10). Higher nesting success for cavity- or niche-
nesting species is usual (Nice 1957, Ricklefs 1969). Most of the increased
success of Yellow-shouldered Blackbirds using cavities was due to their
significantly greater hatching success: 74% vs. 28% for open nests. The
proportions of nestlings that fledged were not significantly different be-
tween the two nest types.
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 163
TABLE 9. Reproductive success of Yellow-shouldered Blackbirds in Coastal Puerto Rico.
Offshore nests, Mainland nests
Variable La Parguera La Parguera Ceiba
Nest success 0.58 (11/19) 0.40 (14/35) 0.17 (3/18)
Eggs/nest 2.58 (49/19) 2.60 (91/35) 1.61 (29/18)
Fledglings/nest 0.89 (17/19) 0.77 (27/35) 0.17 (3/18)
Egg hatched/egg laid 0.49 (24/49) 0.37 (34/91) 0.21 (6/29)
Fledgling/nestling 0.67 (16/24) 0.79 (27/34) 0.50 (3/6)
Fledgling/egg 0.33 (16/49) 0.30 (27/91) 0.10 (3/29)
Mortality was much greater during the egg stage than during the
nestling stage. About half the egg losses were within broods: 58% of 81
blackbird egg losses and 56% of 52 cowbird egg losses fell in this category
(Table 11). Predation accounted for about 35% of the combined egg
losses of blackbirds and cowbirds (Table 11). The remaining egg mortality
was for the most part caused directly or indirectly by cowbird brood
parasitism.
All 35 mainland nests, but only 3 of 14 island nests were parasitized by
Shiny Cowbirds. The reproductive interactions of the Shiny Cowbird and
Yellow-shouldered Blackbird have been treated in detail elsewhere (Post
and Wiley 1977) and are outlined here for clarity. The nesting success of
cowbird-parasitized and unparasitized nests may be compared to assess
the impact of the cowbirds, and this comparison is tenable because nest
losses due to causes other than cowbird parasitism, such as predation,
were the same in mainland and cay habitats. Of 35 mainland nests, 31%
were destroyed by predators, in comparison to 26% of 19 cay nests. A
contrast between the nesting success of open nests that were parasitized
TABLE 10. Reproductive success of Yellow-shouldered Blackbirds placing their nests in the
open or in cavities.
Variable Nests in the open Nests in cavities
Nest success 0.27 (6/22) *, 0.86 (6/7)
Eggs/nest 2.59 (57/22) 2.71 (19/7)
Fledglings/nest 0.50 (11/22) 1.71 (12/7)
Egg hatched/egg laid 0.28 (16/57) 0.74 (14/19)
Fledgling/nestling 0.69 (11/16) 0.86 (12/14)
Fledgling/egg 0.19 (11/57) 0.63 (12/19)
'All nests s ere in mainland mangrove habitat near La Parguera. 1975
'= P< .05 that difference between two values is due to chance alone.
TABLE 11.- Causes of mortality in Yellow-shouldered Blackbird nests at La Parguera 1973-1975.
Eggs Nestlings
Yellow-shouldered Blackbird
% composed
of partial
Shiny Cowbird
% composed
of partial
Yellow-shouldered Blackbird Shiny Cowbird
% of composed % composed
of partial of partial
Cause Total % losses' Total % losses Total % losses Total % losses
Failed to hatch or
live 13 16.0 100.0 12 23.1 100.0 11 68.7 100.00 11 84.6 100.0
Disappearance 19 23.4 73.7 1 1.9 100.0 0 0
Predation by
Shiny Cowbirds 2 2.5 100.0 3 5.8 100.0 0 0
Faulty nest
construction 2 2.5 100.0 1 1.9 100.0 3 18.8 0 0
Desertion 20 24.7 50.0 14 26.9 42.9 0 0
Predation 25 30.9 4.0 21 40.4 28.6 2 12.5 100.0 2 15.4 100.0
Total 81 100.0 51.9 52 100.0 55.8 16 100.0 81.3 13 100.0 100.0
'Losses due to various causes are partitioned as to whether they were composed of partial (within brood) mortality or whole brood mortality.
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 165
and not parasitized, making adjustment for the slightly larger number of
eggs in unparasitized nests, indicates that brood parasitism reduced
Yellow-shouldered Blackbird production by about 0.39 fledglings per nest
(Post and Wiley 1977). The success of unparasitized nests was signifi-
cantly higher than that of parasitized nests. Again, this comparison shows
that brood parasitism had its main effect during the egg stage, with a
reduction in nestlings from unparasitized to parasitized nests (Post and
Wiley 1977).
Hatching failure, which accounted for 18.8% of the combined egg
losses of blackbirds and cowbirds (Table 11), may have been because of
the large number of eggs found in parasitized nests. Female blackbirds
may have been unable to turn, incubate, or shade an excessive number of
eggs properly, and some nests had as many as nine eggs. Hatching failure
was high: 33% of 70 blackbird and 33% of 57 cowbird eggs did not
hatch. In contrast, Caccamise (1976) found a hatching failure of only
7.4% for 497 Red-winged Blackbird eggs in New Jersey.
Desertion, which accounted for a combined loss of 25.6% of the eggs
of both species, was also often related to cowbird visitations. Blackbirds
may have deserted nests that contained large numbers of eggs, that had
cowbird-damaged eggs, or that had eggs removed.
Egg disappearance was also an important source of egg mortality, ac-
counting for 23.4% of the blackbird's egg losses, but only 1.9% of the
cowbird's. This difference suggests that cowbirds were doing most of the
egg removing, at least that proportion (73.7%) that were only partial
losses, as predators usually remove all the nest contents.
Mortality in the nestling stage was relatively low and was due mainly
to death in the nest. In most cases the nestling disappeared from an intact
brood and lost weight before its disappearance. I infer that most of these
young died of starvation and were removed by their parents. Young
usually disappeared or starved when they were less than five days old.
The average age of disappearance of 12 young was 4.33.0 days. Cac-
camise (1976) also found that most deaths in Red-winged Blackbird nests
occurred at about this age, but in comparison to the Red-winged
Blackbird, a much lower proportion of nestling Yellow-shouldered
Blackbirds died in the nest. The more variable food supply of temperate-
zone, marsh-dwelling icterids probably is the main cause of their higher
nestling mortality (Ricklefs 1969).
The lower fledgling/nestling success of cay-nesting Yellow-shouldered
Blackbirds was at least in part due to starvation. Pairs nesting offshore
delivered food at a lower rate than those on the mainland (Table 5). The
number of young of both blackbirds and cowbirds fledged per successful
island nest was 1.55 (17/11, Post and Wiley 1977), and the corresponding
figure for open nests on the opposite mainland was 2.56 (28/11). The
BULLETIN FLORIDA STATE MUSEUM
number of young fledged per successful cavity nest was higher: 3.29
(23/7), and the maximum number fledged from any nest, five (three
cowbirds and two blackbirds), was from a mainland cavity nest. In con-
trast, the maximum number from any cay nest was only two Yellow-
shoulders.
SEX RATIO. -In the populations sampled at La Parguera and among
museum specimens that I examined (Table 12), males outnumbered
females. The unbalanced sex ratio prevailed during the non-breeding
season, but during the summer breeding period the sex ratio was close to
unity. For reasons enumerated below, I believe that the samples taken
during the non-breeding period may be biased toward males. I recognize
five possible sources of error in determining the sex ratio:
1) Sampling error due to finite size of population: the large number of
birds captured at several sites throughout the year reduces the chance of
error from this source.
2) Most birds were captured during the winter and possibly a seasonal
bias exists. Unlike the Red-winged Blackbird, the Yellow-shoulder is not
migratory, nor do the sexes use different wintering ranges. Neither does
the Yellow-shoulder show the extreme sexual segregation exhibited by
Red-winged Blackbirds in winter. My data indicate a tendency for dif-
ferent proportions of the sexes to use different winter feeding grounds
locally. This is demonstrated by comparing the sex ratio of groups of birds
captured in La Parguera with those taken on La Cueva and El Guayacin
(Table 12). In subsequent samples of marked birds, the proportions of
males and females feeding on La Cueva and in La Parguera differed
significantly from the ratios in the marked populations at the time of the
samples (Table 13). Because of male dominance in intersexual competi-
tion during the winter, females may avoid some feeding areas. Similarly,
the sex ratio of Evening Grosbeaks (Hesperiphona vespertina) competing
at small feeding trays has been shown to favor the more aggressive male
(Balph and Balph 1976). In contrast to the winter, the numbers of male
and female Yellow-shoulders captured and seen at feeding sites in the
summer were about the same (Table 12).
3) Female Yellow-shoulders show greater feeding site specialization
than do males (Table 14), and if males use different feeding stations more
frequently, they might be more susceptible to capture than females. This
supposition is supported by the fact that marked males were more con-
spicuous than females: the average number of times that 202 marked
males were seen was 12.6 0.9 each, whereas 118 females were seen an
average of 9.2 1.0 times each, a significant difference (P < .05, Student's
t test).
4) Male and female Yellow-shoulders have different parental roles
which might make males more conspicuous during the nesting period.
Vol. 26, No. 3
TABLE 12. -Sex of newly-captured adult Yellow-shouldered Blackbirds in SW Puerto Rico and of museum specimens.
La Cueva and
El Guayacdn Islands
Female Male
La Parguera
Female Male
152
(X2= 33.9)
200
66 50
43 10
109 60
(X'= 28.8)
An asterisk between 2 values indicates a significant difference (P<.05).
Winter
(Nov-Apr)
Summer
Total
Museum
Female
(X2' 5.5)
37 58
(X= 6.3)
TABLE 13. Sex composition of groups of color-marked Yellow-shouldered Blackbirds at La Parguera.
Total No. Test of Sig-
Duration of No. of No. of birds Test of Sig- of birds nificance2
sampling Marked Ratio marked locally Ratio, nificance' marked at Ratio,
Location Date period (hr) birds M:F at time of sample M:F x2 P time of sample M:F x2 P
Feeder no. 1,
La Cueva 24Oct74 1 22 10.00 317 1.59 6.4 <.05 422 1.54 6.9 <.01
Feeders no.
5-8, La Cueva 15Feb75 2 71 3.73 324 1.63 6.5 <.05 432 1.57 7.6 <.01
19Apr75 2 34 5.80 328 1.62 6.4 <.05 436 1.56 7.0 <.01
Feeders no.
19-22,
El Guayacin 16Feb76 3 53 3.42 328 1.62 4.1 <.05 444 1.57 4.1 <.05
13Feb76 2 62 1.95 328 1.62 0.4 NS 444 1.57 0.2 NS
E La Parguera 26Jan75 8 19 8.50 106 1.30 7.4 <.01 429 1.57 6.3 <.05
'The numbers of each sex in the sample are compared to numbers of each sex that had been color-marked in the local area where the sample was made.
'The numbers of each sex in the sample are compared to number of each sex that had been color-marked in the entire study area.
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 169
TABLE 14. -Number of individuals seen at only one feeding site and at more than one site.
Seen at: Test for differences
One site More than between sex and age
Sex and Age Category only one site categories
Adult males 31 (16%) 167 (84%)
X2=4.2; P<.05
Adult females 24 (27%) 66 (73%)
Total adults 55 (19%) 233 (81%)
x2= 1.2; N.S.
Total juveniles 5 (11%) 40 (89%)
Females incubate and brood, while males engage in defense of the nest
area to a greater degree than do females. At San German males were often
trapped while females were on the nest, and the sex ratio of captured
birds there was 7 males:3 females.
5) The 36 mm nets used to catch the blackbirds probably did not
favor one sex; if anything, the smaller females were captured more readily
because of the greater tendency for smaller birds to become entangled.
However, males mobbed more frequently than females. Mobbing birds
sometimes landed on net shelves in response to the screams of netted birds,
perhaps increasing the chances of males' being captured.
Male Yellow-shouldered Blackbirds usually dominated females in
competition for food. I have data for 20 aggressive two-way interactions
between marked males and marked females at winter feeding stations: in
17 cases males dominated females, while in three the reverse occurred. As
males are larger, their dominance is to be expected, at least during the
nonbreeding season. Male dominance may affect the social structure of
Yellow-shoulder populations during the nonbreeding season. Some winter
flocks had significantly more males than females (Table 13).
SURVIVAL. -The annual survival of 250 adults was 82.4% (Table 15).
This calculation is based on a composite population (Hickey 1952) of
color-marked individuals, captured from December 1972 to May 1974.
Recaptures were made either by trapping or by sighting during censuses
at feeding and nesting sites. I continued censuses through February 1976.
Birds that were not seen or captured again were considered dead, and sur-
viving up to the time of their last appearance (von Haartman 1971). These
250 individuals (Table 15) were classified as residents of the La Parguera-
El Guayacin area; birds that were not seen or captured at least once after
their first capture were excluded from this analysis. This exclusion is war-
ranted by the fact that a few of the birds captured in winter may have
been transients, stopping to feed on El Guayacin while they were on their
way to the roost at Collao (Fig. 1). Most of these individuals were prob-
TABLE 15. Annual survival of Yellow-shouldered Blackbirds.
Numbers marked
Marking Period
Dec. 72-Jan. 73
Feb.-Mar. 73
Apr.-May 73
Subtotal, Dec. 72-May 73
Jul.-Aug. 73
Feb.-Mar. 74
Apr.-May 74
Subtotal, Jul. 73-May 74
Total, excluding individuals
not seen at least once after
original capture
Total, including all birds
captured
All
Males
53
12
8
73
All
Females
21
7
2
30
Numbers surviving at least one year
Juveniles Males Females Juveniles
48 21
11 7
7 2
66(90%) 30(100%)
35
36
78(78%)
173 77
10
18
32(68%)
29 144(83%) 62(81%)
196 95 32 144(74%) 62(65%)
Numbers surviving
at least two years
Males Females
41 15
8 4
4 1
53(73%) 20(66%)
11
8
19(65.5%)
19(65.5%)
19(59.4%)
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 171
ably residents in the Pitahaya-Bahia Sucia area, west of the study tract. In
fact most of the birds that were residents in the La Parguera-El
Guayacin area were seen much more often than once: 320 marked birds
were seen an average of 11.4 times each. I believe 82.4% to be an ac-
curate estimate of survival, as the proportion of birds that died shortly
after being marked probably was small, and undoubtedly balanced by the
proportion that emigrated from the study area.
Including all the birds captured between December 1972 and May
1974 in the analysis would increase the size of the sample to 291. The cor-
responding numbers surviving to one year would remain the same, but
the annual survival rate estimate would decrease to 70.8% (Table 15).
For the population captured between December 1972 and May 1973,
I was able to compute the survival rate to two years: 70.9% of 103 in-
dividuals survived at least two years after capture. The weighted annual
survival rate (Farner 1945) of this group is 69.8%.
My data do not allow the computation of time-specific survival, but as
the adult birds were captured throughout the year and were presumably a
cross section of the total population, their annual survival represents the
average for the adult population. Such data must be used with caution in
computing parameters such as longevity (Botkin and Miller 1974).
The data do not show a sex difference in annual survival for the first or
second year after capture. Comparing the numbers of each sex surviving
after one year (group captured December 1972-May 1973; Table 15)
shows that 53 of 66 males (80.3%) and 20 of 30 females (66.7%) survived
for two years, and this difference is not significant (X2 = 1.4;
0.25
0.1, ld.f.)
The survival to adulthood of 29 individuals that were marked as
juveniles was 65.5% (Table 15). Again birds that were not seen at least
once after capture were excluded from the analysis, but if all the juveniles
captured during this period are included, the survival rate is 59.4 %. The
juveniles considered here were caught in their first winter or spring, and
the survival value is for one year from capture, i.e. through their second
year.
During the same period that I caught these 32 juveniles (February-
May 1974; Table 15), I also took 158 adults. This gives a juvenile:adult
ratio of 0.203/1.000, which agrees fairly closely with the estimated an-
nual adult mortality of 17.6%. This ratio is also close to the 0.210/1.000
ratio among the 92 birds collected throughout Puerto Rico that I exam-
ined in museums.
In the nesting seasons of 1974 and 1975 I color-marked 29 fledglings.
Eight of these were later seen flying about, independent of their parents.
Although the sample is small, it gives an estimated fledgling to juvenile
survival rate of 27.6%. The product of the survival rates of fledglings to
BULLETIN FLORIDA STATE MUSEUM
juvenilehood and of juveniles to adulthood is 18.1%, which again is close
to the average annual adult mortality rate of 17.6%.
Determining survival by sighting color-marked individuals is more ac-
curate than any system dependent on recaptures, which are often biased
by the vagaries of trapping (Grosskopf 1964; Coulson and Wooller 1976).
This system is particularly effective with Yellow-shoulders because they
are sedentary and conspicuous, and the census stations, widely spaced in
nesting and feeding areas, were manned throughout the year.
DIscussioN. -The success of open, unparasitized nests at La Parguera
is similar to that Marchant (1960) reported for nine open-nesting
passerines in an arid tropical region of Ecuador, where nest success was
51% and overall egg success was 44% (versus 63% and 35% for Yellow-
shoulders).
Open, cowbird-parasitized nests at La Parguera had a lower success
than the above, and their success was similar to that of some species that
occupy cleared edges of humid tropical forests, where partial losses
related to such factors as brood parasitism are high. For example, the
Clay-colored Robin had nest success of 25% and egg success of 16%,
while, as with parasitized Yellow-shoulder nests, its fledgling/nestling
success was relatively high: 53% (Skutch 1966).
No information is available on the reproductive success of cavity-
nesting birds in the arid tropics, but the success of cavity-nesting Yellow-
shoulders at La Parguera exceeds the average success of 16 species of hole
nesters in the humid tropics (Ricklefs 1969). These species had a nest suc-
cess of 54 %, and 44 % of their eggs produced fledglings, in comparison to
86% and 63% for the Yellow-shoulder cavity nests.
In comparison to icterid nests in the temperate zone of North America,
Yellow-shouldered Blackbird nests not parasitized by cowbirds have
lower success in the percent hatched of eggs laid, but higher fledg-
ling/nestling success. In addition, overall nest success was higher in the
Yellow-shoulder. Robertson (1972) found that Red-winged Blackbirds
nesting in marshes (where there was no cowbird parasitism) had nest suc-
cess of 53 %, fledgling/nestling success of 65 % and fledgling/egg success of
44 %. In continental icterid populations, nestlings appear to be subjected
to more predation, and because of larger clutch size may experience
greater starvation, both of these effects increasing with age of nestlings
(Smith 1943, Young 1963, Robertson 1972). Egg loss in the Yellow-
shoulder greatly outweighs loss of nestlings, while in temperate zone Red-
winged Blackbird populations mortality is fairly constant throughout the
nest period.
Red-winged Blackbirds nesting in fresh water marshes in Costa Rica
had a nest success of 21.5% (Orians 1973), in comparison to 27% for
Yellow-shouldered Blackbird open nests (Table 10). As with Costa Rican
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 173
Red-wings, most of the Yellow-shoulder losses during the nestling stage
were due to starvation. Whole nest losses, occurring during both egg and
nestling stages, and usually by predation, were more important in Costa
Rica than in Puerto Rico, but partial losses in Puerto Rico, directly or in-
directly attributable to brood parasitism, nearly equaled whole-nest loss
in Costa Rican Red-wings. As expected, cavity-nesting Yellow-shouldered
Blackbirds at La Parguera had much higher nest success than open-
nesting Red-winged Blackbirds in Costa Rica.
In comparison to estimates of survival rates based on similarly
gathered data for Red-winged Blackbirds, Yellow-shoulders have a higher
rate of survival. Data for 325 Red-wings marked and retrapped in coastal
Massachusetts gave an annual (weighted) adult survival rate of 53.4%
(Fankhauser 1967). Significantly lower than my weighted estimate of
69.8% (x2 = 9.6; p<.005, 1 d.f.) for the Yellow-shoulder. Another
estimate of Red-wing survival generated from shootings of banded birds
throughout North America gave a figure of 51.1% (Fankhauser 1967).
High survival rates appear to be characteristic of some tropical bird
species, although further research, particularly in disturbed neotropical
areas, needs to be done. On Sarawak, Fogden (1972) found a minimum
annual survival rate of 86 % for a composite sample of species occupying
forest habitats. Snow and Lill (1974) estimated minimum annual survival
for the White-bearded Manakin (Manacus manacus), a species occupying
tropical forests in Trinidad, to be 89%. Although my estimate is for a
tropical species occupying open, disturbed habitat, for which category
there appear to be no other survival estimates, some of the factors con-
tributing to the increased longevity of tropical forest birds must apply to
species in disturbed habitats, i.e. stable climate, buffered food supply,
and lack of migration.
BEHAVIOR
HOME RANGE. The marked birds that I studied seem to be permanent
residents around La Parguera, and I detected no seasonal movements by
any significant proportion of the population. Some individuals that I had
marked on the coast were seen inland during the non-breeding season; 11
birds (6 males and 5 females) appeared occasionally in flocks of 50-70
Yellow-shoulders that fed around the cattle barns at the Lajas Experimen-
tal Station, 7.3 m N of La Parguera. During this time several of these
birds were also seen on the coast, where they apparently returned nightly
to roosts.
Individual birds ranged widely (Table 16; Fig. 7), but I detected no
seasonal differences in size or usage pattern of their home ranges. This is
probably due to (1) the year-round use of communal roosts, which were
usually distant from nesting and feeding areas (Fig. 1); (2) year-round use
TABLE 16.-Home ranges and distances between captures.
Males Females
N x Range N x Range
Home ranges (km2) 26 2.56+0.502 0.27-13.34 12 1.73+ 0.36 0.39-3.70
Distances between
recaptures (m)1 200 1325 95.22 25-5440 200 1630 131.5 25-7260
LOnly distances between sightings or recaptures at successively different localities counted.
2Standard error of the mean.
2 2
C o o
31 22 0 2
CO
0
0
a
FIGURE 7.-Home ranges of Yellow-shouldered Blackbirds. For localities, refer to Fig. 1. Black dots indicate sighting or capture points. Single dots
mean one instance; multiple instances are indicated by numbers. Open arrows indicate nesting sites, with year; solid arrow is roost site. (a) male-,i
GYG-A; (b) male BA-RG; (c) female YA-XY; (d) female RYR-A.
0 0
FIGURE 7.-Home ranges of Yellow-shouldered Blackbirds. For localities, refer to Fig. 1. Black dots indicate sighting or capture points. Single dots
mean one instance; multiple instances are indicated by numbers. Open arrows indicate nesting sites, with year; solid arrow is roost site. (a) male
GYG-A; (b) male BA-RG; (c) female YA-XY; (d) female RYR-A.
BULLETIN FLORIDA STATE MUSEUM
of communal feeding grounds such as those on La Cueva and El
Guayacin; and (3) distant feeding behavior.
The large sizes of the home ranges are reflected in the distance be-
tween successive recaptures. Distances between recaptures (Table 16) are
reasonable estimates of daily movements: birds using the La Cueva and El
Guayacin feeding grounds often flew to one of the La Parguera roosts
(Fig. 1), making daily round-trip flights of 10-12 km. Other individuals
used feeding station in La Parguera and flew nightly 5.5 km to a roost in
Bahia Montalva E of the study area.
Most of the marked birds whose nests I located used winter feeding
grounds near their nest sites (Fig. 7). The main nesting places, Salinas
Arcelay and Salinas Carlos, were 1600 m and 800 m from the nearest
monkey feeders on El Guayacin and La Cueva. At least seven birds
marked as nestlings used one of these feeding grounds the winter after
they hatched. The average distance between their hatching places and
winter feeding grounds was 1549 m, range 980-2680 m.
FOOD AND FORAGING.- During the nesting season Yellow-shouldered
Blackbirds brought two types of food to dependent young; the bulk of the
food was arthropods, the remainder vegetable matter (Table 17). Arth-
ropods were mainly gathered in the canopy and subcanopy layers of trees,
while vegetable matter was obtained by birds visiting feeding sites of
domestic animals. The main bulk of the 25 food samples that I took from
young birds consisted of large items such as wood-boring beetles
(Buprestidae), with an average length of 11 mm, tree crickets (Gryllidae),
averaging 20.6 mm, larvae and pupae of moths (Olethreutidae and Noc-
tuidae), the larvae averaging 11.9 mm and pupae averaging 14.1 mm,
and arboreal spiders (Anyphaenidae) with a mean length of 9.1 mm.
Observations of foraging sites and tactics of the blackbirds (Table 18) and
the behavior of the arthropods showed that the blackbirds get most of
their food by probing into crevices in trunks and branches, epiphytes, and
leaf clusters. Most of the forms brought to the young are either always
hidden (Buprestidae and immature Olethreutidae) or are nocturnal and
hide during the day in crevices, leaf clusters, or epiphytes (Noctuidae,
Gryllidae, and Anyphaenidae). Among the noctuids collected from nest-
ling A. xanthomus, Melipotis sp. were common. Larvae of these moths
feed in the tree canopy at night and during the day move down the trunk
to hide in crevices in the bark, which is also screened by surrounding
shrubs. They also pupate in these sites (D. H. Habeck, pers. comm.). Im-
mature Olethreutidae, the most common food delivered to young
blackbirds, infested leaf buds of red mangroves, where I often saw
Yellow-shoulders probing.
Small arthropods such as leafhoppers (Fulgoridae), about 2 mm, were
occasionally taken (Table 17). These insects infested patches of scrub such
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 177
as Bumelia spp. One leafhopper, Petrusa epilepsis, was utilized by groups
of warblers and was so abundant that foraging birds flushed 5-10 insects
with each move (Post 1978). I saw Yellow-shoulders take these only four
times. The long distances that nesting blackbirds often flew to gather food
for their young may have precluded their bringing such small items to the
nests (Emlen 1966), and they may have been used mainly by the adults
that encountered them while searching for larger items. Yellow-
shouldered Blackbirds may have delivered small items more frequently
than my samples indicate because small pieces of food may slip past neck
collars (Orians 1966).
When delivering to the young, adults usually carried more than one
arthropod at a time. They also regurgitated food from their gullets,
mainly plant material and occasional small fragments of arthropods. I
determined this by collecting food from the throats of some collared
young immediately after I saw a parent regurgitate material to them.
Vegetable matter appeared in 15 of the 25 food samples I collected (Table
17): rice or grain (7 out of 25 samples), bread or flour products (5 of 25),
and monkey chow (4 of 25). The average representation by volume of
vegetable material in the 15 samples was 29%, range 5-100%.
Birds nesting offshore usually foraged in trees in pastures along the
mangroves, flying up to 2 km between these sites and their nests. I also
saw Cay-nesting birds foraging in red mangroves near their nests. Birds
nesting in red mangroves occasionally fed on the roots and trunks, but a
cursory examination of these sites revealed few arthropods, possibly
because of the high tannin content of red mangrove bark.
In salinas the birds did not forage on the open mud around their nests.
Birds nesting in salinas occasionally foraged on the trunks and branches of
dead mangroves, although they most often flew relatively long distances
(0.5-1 km) to forage in trees in pastures. These birds, as well as those
nesting in the pastures themselves, did not forage for arthropods on the
ground there either.
The foraging behavior of Yellow-shoulders varied. On trunks and
branches they used their bills to probe crevices, flake off pieces of bark,
and enlarge holes. Yellow-shoulders gaped inside crevices, fruit, buds,
and cocoons. Individuals clung to the sides of vertical trunks or to the
undersides of branches to probe or to glean surfaces. Birds gleaned leaves
and twigs as they walked along branches, but more often they flew be-
tween leaf clusters. They commonly probed epiphytes (Tillandsia recur-
vata), usually standing on top of the tangled balls of vegetation and
inserting their beaks up to the base of their skulls in the clumps, but
sometimes they hung under the clumps to probe the undersurfaces. Other
species I saw probing the Tillandsia recurvata clumps were Adelaide's
Warblers (Dendroica adelaide) and Black-cowled Orioles (Icterus
TABLE 17.- Food brought to young in nests at La Parguera, 1974-1975.
Order or Group
Family and stage
Lepidoptera
Olethreutidae, larvae
pupae
Noctuidae, larvae
pupae
Unidentified pupa
Araneae
Anyphaenidae
Salticidae
Araneidae
Unidentified
Orthoptera
Gryllidae
Blattidae
Homoptera
Fulgoridae
Tropiduchidae
Coleoptera
Buprestidae
Elateridae
Unidentified
Number of samples'
in which found
14 (56)2
4 (16)
4 (16)
3 (12)
1 (4)
10 (40)
5 (20)
2 (8)
5 (20)
7 (28)
1 (4)
2 (8)
1 (4)
2 (8)
1 (4)
1 (4)
Total No.
found
46 (28.2)
4 (2.5)
10 (6.1)
31 (19.0)
1 (0.6)
16 (9.8)
1 (0.6)
17 (10.4)
1 (0.6)
3 (1.8)
1 (0.6)
1 (0.6)
'Length (mm)
Average Range
10.43.6
7.0
18.71.6
15.00
7.0
9.11.5
7.4
7.5
7.6
20.67.0
6.0
2.0
piece
5-22
7
15-20
15
6-12
5-12
5-10
6-10
5-30
2
10-13
11.0
10.0
piece
Hymenoptera
Unidentified
Arachnida
Unidentified
Vegetable
Matter3
1 (0.6) piece
1 (0.6)
Total 25
'One sample is the food from all young of nest during 1 h period.
2Values in parentheses are percentages of total.
3Bread, rice and monkey chow.
1 (4)
1 (4)
15 (60)
TABLE 18. -Foraging behavior of the Yellow-shouldered Blackbird during breeding season (May-September).'
Foraging Zone
Foraging Tactic
Gleaning Leaf
Gleaning Twig
Gleaning or Probing
Branch
Flycatching
and Hovering
Chasing
Probing Flower
Probing Fruit
Probing Terminal
Bud
Probing Epiphyte
Pecking Ground
Total and
(Percent)
Outer
282
3
15
Canopy
Middle
3
1
Inner
2
0
118 (18.2) 7 (1.1) 4 (0.6)
129 (19.9)
Outer
25
17
Subcanopy
Middle
28
17
29
186 (28.7) 114 (17.6)
419 (64.7)
Inner
9
5
Herb
Layer
5
0
Ground
Layer
0
0
0
119 (18.4)
43 (6.6) 57 (8.8)
'Does not include foraging at human food sources. See methods section for procedures.
2Number of times behavior was performed in each foraging zone.
5Number in parentheses are percent of total observations.
Total and
(Percent)
100 (15.4)3
43 (6.6)
112 (17.3)
55 (8.5)
17 (2.6)
39 (6.0)
27 (4.2)
47 (7.3)
177 (27.3)
31 (4.8)
648(100.0)
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 181
dominicensis), both of which were rare in the pastures where these
epiphytes were common. I saw Yellow-shoulders flycatch, most often
after they had flushed insects as they moved, but occasionally they sat on
exposed perches and sallied.
When foraging on the ground, Yellow-shoulders walked, occasionally
scratching with one foot. They gaped under monkey biscuits to turn them
over, thereby pushing the food away from the body. Orians (1961) saw
Red-winged Blackbirds turn rocks by this method. Yellow-shoulders also
turned the food over with their closed mandibles. Individuals also in-
serted the lower mandible under monkey biscuits and lifted them toward
the body. Yellow-shoulders used the sides as well as the tip of the beak to
break off material from large pieces of food, the sides of the beak being
moved back and forth in a shearing motion.
The birds probed for nectar commonly in January and February when
Aloe vulgaris was blooming, and during this period small groups were in
constant attendance at patches of aloe. They probed flowers by inserting
their beaks up to the bases of their skulls, gaping to enlarge the entrances.
On one occasion I noted three Yellow-shoulders probing Yucca flowers.
Wetmore (1916) saw Yellow-shoulders probe bucare (Erythrina sp.)
blossoms, and Danforth (1926) reported Yellow-shoulders taking nectar
from guama (Inga laurina), as well as fruits of several species of cacti, e.g.
Selenicerus sp. and Cephalocereus royenii.
During the nesting period, adult A. xanthomus foraged mainly in the
subcanopy (64.7% of 648 observations; Table 18) and canopy (9.9%)
layers of trees. Within these layers most foraging maneuvers were per-
formed in the outer zones. Probing epiphytes, the most common foraging
tactic, composed 27.3% (177 of 648) of the maneuvers observed. Of the
total, 21.9% were preformed in the outer canopies and subcanopies.
Similarly leaf gleaning, the third most common foraging tactic, occurred
mainly in the outer zones of the canopy and subcanopy (8.2% out of a
total of 15.4%). Gleaning and probing of branches occurred mainly in
the subcanopy (14.7% out of 17.3%), the zone of vegetation which has
the greatest mass of branches. Flycatching was a relatively important ac-
tivity, and it occurred mainly (4.9 % out of 8.5 %) in the inner subcanopy,
which was relatively open in vegetation such as red mangroves, and
where lower turbulence may have allowed easier capture of flying insects.
Flycatching and hovering also took place in the outer canopy and sub-
canopy (2.3% of total). Probing of terminal buds was accomplished
mainly in the subcanopy (6.6% of total), and most of this activity was
seen in red mangroves. Most flower probing, a relatively uncommon
behavior in the breeding season (6%), involved aloe, and therefore oc-
curred in the herb zone.
BULLETIN FLORIDA STATE MUSEUM
kHz
8-
A B
0 0.25 0.50 0.75 100
SECONDS
FIGURE 8.-Sound spectrograms of Yellow-shouldered Blackbird vocalizations. (a) growl,
wide band filter; (b) same as preceding, but with narrow band filter.
VOCALIZATIONS
GROWL. Lasting about 1 sec, this sound is composed of a short in-
troductory note and a buzzy trill. The click-like introduction extends
beyond 16 kHz, and resembles a chwip or check call (below). The in-
troductory figure is probably imperceptible to humans, as it is overlapped
temporally by the more audible buzz component, which has a frequency
range of 1.5-5.0 kHz, with energy concentrated at 4-5 kHz (Fig. 8). I
noted some individual variation in the structure of the growl, such as in-
terruption in the trill, abbreviated and lengthened trills, as well as incor-
poration of other vocalizations such as gueea in the beginning of the
vocalization.
The growl is probably homologous with those vocalizations given dur-
ing "song-spread" (Nero 1956a) by other icterids. The structure of the
growl resembles the trill part of the buzzing song of the Yellow-headed
Blackbird Xanthocephalus xanthocephalus (Orians and Christman 1968;
Figure 14b).
Both sexes utter growls during the song-spread and during wing-raise.
With song-spread the growl is given as the wings reach their full height,
the sound continuing as the wings are lowered. No head movements are
associated with the utterance of growl, but the beak is opened (see Action
Patterns). The rate of singing varied from 1.2 to 2.3/min by some birds
perched near their nests, while others in communal roosts wing-raised
(with growls) 12 to 17 times/min. Juveniles sometimes rapidly uttered a
short, muted growl accompanied by a wing-raise. These may have been
examples of subsong (Thorpe 1961).
On nesting areas growls as well as pee-puus (see below) were ex-
changed by neighboring birds that were out of each other's sight. In April
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 183
TABLE 19. Contagion of growls on nesting grounds.
No. of growls initiated per
10 consecutive sec interval': 0 1 2 3 or more
Observed 267 83 51 17
Expected2 238 134 38 7
'Observation periods, totalling 70 min. were each 300 sec long.
2Calculated from the Poisson series. The observed number of calls in each category is significantly different from the
random distribution (2 41.7: P< .001 d.f.= 3).
1975 I recorded the incidence of growling by individuals in groups of 6-8
that were visiting nesting sites in the red mangroves on W. Matita. I
found a significant contagion between growls: a growl from one bird was
usually followed within an unexpectedly short interval by one or more
growls from other individuals (Table 19). Birds sitting next to each other
and engaging in song spread occasionally initiated growls and pee-puus
before neighbors had terminated their growl, and in several instances 3 or
4 individuals overlapped their growls to form a continuous sound lasting
about 4 sec.
As discussed under wing-raise, growls as well as gueeas were
sometimes repeated erradically in mass by flocks of birds that were mob-
bing humans, or at least reacting to their presence. These may have been
instances of group performance of antipredator strategy (Smith 1977), as
the "indefinably confusing, all-pervading sound" (Grinnell 1903) made
localizing any one individual difficult.
RASP. -This call, rendered vvvt, is structurally similar to growl,
although its fundamental is lower, at about 1.5 kHz, with discernible
overtones at 3 and 4.5 kHz. To humans the call sounds thinner and less
resonant than a growl. The rasp was heard only during agonistic en-
counters, and it was uttered by flying birds or by ones about to fly, as
when preparing to dive at predators. Yellow-shoulders also gave the call
when supplanting and being supplanted from feeding positions.
GREEAH: This call, which sounds to me like the scold of the Red-eyed
Vireo (Vireo olivacus), has a complaining, nasal quality. I heard it on
numerous occasions, but only in the vicinity of nests, and most often from
parents around the time young were fledging. It was associated with a
tendency to fly or move. With disturbances near the nest, both sexes gave
the call as they moved around, often accompanying each other. In addi-
tion, I once heard greeahs given by a fledgling that was being fed near its
nest. The female that was feeding it also gave greeahs. On another occa-
sion a female that was mobbing me wing-trailed while giving greeahs.
The nearby young gave pink calls (below) and moved toward the greeah-
calling female. Individuals often uttered greeah when they engaged in
BULLETIN FLORIDA STATE MUSEUM
kHz
SA B C
6- -
4 $1^ 1 \ \
i, I
0 0 25 0.50 075 100
SECONDS
FIGURE 9.-Sound spectrograms of cut-zee type vocalizations. (a) cut-zee, made with wide
band filter; (b) same as preceding, but with narrow band filter; (c) a tslink variants, wide
band filter.
moth flight. On one occasion I noted that a male's greeahs merged into
growls, given with song-spread.
SCREAM. -This harsh noise is similar in structure to screams given by
other icterids (Orians and Christman 1968, Figure 21d). The scream was
most often uttered by birds that were being handled, but also by those
confined in nets or traps.
CUT-ZEE. A common alarm call with much individual variation, cut-
zee is composed of two parts: a short introduction followed by a slightly
longer, falling element. There is temporal overlap in the two figures, in-
dicating the use of two or more syringeal membranes (Fig. 9).
Both sexes utter cut-zee when mobbing of predators around nest sites,
and I also heard it throughout the year in feeding and roosting areas. Cut-
zee is given with or without wing-raise and tail flip. Wing-trailing birds
also give cut-zee. It is occasionally combined with other vocalizations: on
25 May 1974 female XAGR, 4 m from her nest, in 3 min gave eight cut-
zees that intergraded into growls. Some of these combinations were
repeated rapidly with only a 5-sec lapse.
During mobbing cut-zee is often associated with checks and queeas.
The information provided by cut-zee is probably similar to that made
available by queea, but cut-zee appears to be used in more intense mob-
bing, as indicated by the closer approach of cut-zee calling birds to
predators. Nestling Yellow-shoulders may recognize cut-zee as an alarm.
On 25 June 1974 a female perched under her nest repeatedly gave chwips,
while the young continued to beg loudly. The female then began uttering
cut-zees, and the young immediately became quiet. Occasionally only
part of cut-zee is given; the second note, zee, or a close variant sounding
like tslink, was recorded (Fig. 9). On other occasions this sound was
repeated rapidly in flight, sounding like zeenk zeenk zeenk.
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 185
QUEEA.- Queea is composed of two figures, the fundamental of the
first being about 3.5 kHz, with two discernible harmonics. The second
note, which temporally overlaps the first, falls slightly in pitch, and has
its fundamental at about 5 kHz (Fig. 10). This variable call is occasionally
repeated in series with wing-raises, and is often combined with other
vocalizations such as growl and chwip. Queea is usually given with tail
flip.
Queea is frequently given by birds mobbing predators, such as
humans, monkeys, or cats, or by birds that are disturbed in any way. The
call is also given by females that are being pursued persistently during sex-
ual chases.
Average rate of calling was 18/min, range 7-28/min (602 calls from 13
birds timed for 30 min). Queea is given in association with other sounds
such as check. For example, on 18 February 1975 male RARG, sitting
with two other Yellow-shoulders, gave 13 queeas, 20 checks, and 1 growl
as I approached, switching from check to queea each time I moved.
Although queea-type calls appeared to be used most often as alarms, I
also noted that they were occasionally used when birds seemed to be un-
disturbed. At these times the call was often given rapidly with wing-
raises, 2-6 times in succession, to form what sounded like a chatter. When
repeated rapidly as a chatter, queea may function as a social or contact
signal between members of feeding flocks or nesting groups. This chatter
may be homologous with the female song of the Red-winged Blackbird
(Nero 1956a).
CHWIP. -Relatively simple in structure (Fig. 10), chwip resembles
check of the Yellow-headed Blackbird (Orians and Christman 1968; Fig.
20p-r). Young in the nest begin to give chwip when they are 6-7 days old,
kHz
8- A B C D E
O
0 0.25 050 075 100
SECONDS
FIGURE 10.-Sound spectrograms of Yellow-shouldered Blackbird vocalizations, all made
with wide band filter; (a) queea. variant 1; (b) queea. variant 2: (c) check: (d) chi ip: (e) pee
(first part of pee-puu).
BULLETIN FLORIDA STATE MUSEUM
so after begging calls and pink (below), it is one of the first vocalizations
young Yellow-shoulders give. Chwip is used in practically all situations,
by birds in flocks or alone, flying or sitting, and may indicate that a bird is
seeking association, may fly, escape, or is indecisive. Chwip is usually ac-
companied by tail-flip. Average rate of calling was 14/min, range
1-41/min (36 birds timed for 75 min).
CHECK. -Somewhat more structured than chwip, check shows a
distinct harmonic, and the call lasts about twice as long. The energy of
check is centered around 4 kHz, but the call extends over a wide fre-
quency (Fig. 10). Check is given by birds that are scolding predators, are
engaged in agonistic encounters, or possibly are disturbed in some way.
Check probably provides information about the individual's readiness to
attack, escape, or fly, or indicates that it is acting indecisively. It is often
uttered in association with other vocalizations that function as alarms,
such as queea, cut-zee, and chwip. The associated action pattern is the
tail-flip. Average rate of calling was 13/min, range 1-47/min (16 birds
timed for 30 min).
PEE-PUU. The pee-puu is composed of two components: the introduc-
tory pee is a clear, slightly rising note, while puu, also unslurred, falls in
pitch. This vocalization, for which I have a spectrograph for only the pee
component (Fig. 10), probably functions mainly in communication be-
tween paired individuals. I frequently heard it on nesting grounds, and
pee-puus from females were usually answered by growls from their males.
On 5 August 1974 male GABR inspected his nest. His female, XAGR,
perched 4 m away, gave a pee-puu, which was immediately followed by
the male's song spread. The female repeated her pee-puu 30 sec later. On
11 April 1975 male GYGA, perched next to a 1974 nest, in 3 min gave nine
song-spreads, each immediately followed by a pee-puu from the female
with whom he was associated. These pee-puus were also returned by
neighboring birds.
PINK. -To my ear, pink sounds remarkably like the flight note of the
Bobolink. It was given as a contact call between young and their parents.
I first heard the call given by nestlings when they were six days old. Free-
flying young gave pinks as they followed their parents about, either singly
or as a double call. Adults leading their young in flight also gave the call.
Rate of utterance by one young just fledged on 1 July 1974 was 122 in 3
min.
FLIGHT SERIES. -A flight series, usually uttered when birds were first
air-borne, appeared to have no unique components, but rather was a
composite of different calls, such as cut-zee, queea, pee-puu, and chwip. I
noted a large amount of individual variation in the composition of the
call. The call was given all year, in the winter from birds leaving feeding
flocks or secondary roosts, but most commonly during the nesting period
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 187
from birds that were leaving their nests on distant feeding flights. The
flight vocalization may be important as a means of contact between
mated birds traveling between nesting and feeding areas.
ACTION PATTERNS
SONG-SPREAD.- Initially the head may be pointed upward (Fig. 11).
After about 2 sec the bird lowers its beak and begins to raise its wings, the
tail becoming increasingly fanned and the body plumage fluffed. The
wing elevation phase takes about 1 sec, at the end of which the beak
touches the breast and the ventral plumage is ruffled (Fig. 11). The wings
are held up for about 2-3 secs, and lowered in about 1 sec, then the head is
again pointed up. During wing elevation the carpus is rotated forward,
providing maximum frontal exposure of the epaulets. I noted much in-
dividual variation in song-spread, mainly in asymmetry of wing and leg
positions and in angle of body tilt. The bill-up at the start of the song-
spread was sometimes omitted, but the terminal bill-up was nearly always
present.
Both sexes sing in many situations, but most commonly in (1) close
quarter agonistic encounters with conspecifics, usually near the nest, but
also in feeding flocks and roosts; (2) the presence of predators in the nest
vicinity; (3) instances when mated birds meet near their nest; (4) agonistic
encounters with other bird species; and (5) nest-site advertisement by the
male (see below).
Representative rates of singing are (1) 9 times in 3 min by a male near
his nest and mate; (2) 18 times in 15 min by a male near his nest; and (3)
17 times in 7.5 min by a female near her nest. The only vocal signal that I
heard given with song-spread, and one that accompanied it invariably,
was the growl, which lasted about 1 sec, and was initiated as the wings
reached their full elevation.
WING-RAISE. Less complex than song spread, wing-raise may corre-
spond to a less ritualized stage of the former display, in which some com-
ponents are lacking (bill-up and bill-down) and others are less exag-
gerated (wings and body plumage not as elevated). In the less extreme
form, the carpus may be rotated upward only slightly, with the tips of the
remiges remaining in contact with the body. In the more extreme form,
the wings may be spread and raised more fully, but the tips are seldom
raised more than 300 above the horizontal.
Wing-raise is usually repeated rapidly, every 3-5 sec; in April 1975 a
juvenile Yellow-shoulder sitting alone in the mangroves wing-raised 55
times in 4 min, giving an abbreviated growl with each wing-raise. This
may have been an instance of practice singing (see p. 182).
Wing-raise occurred in the following situations: (1) during close-
quarter intraspecific and interspecific agonistic encounters on and off
BULLETIN FLORIDA STATE MUSEUM
FIGURE 11.-Song-spread of the Yellow-shouldered Blackbird, showing lateral (a) and frontal
(b) aspects.
nesting grounds; (2) during mobbing of predators; and (3) when a bird
was alone, either on or off nesting areas. Rapidly repeated wing-raises,
given with growls and queeas, were occasionally uttered in mass by birds
sitting close together in roosts. These massed displays seemed to be elicited
by disturbances and may have been instances of group mobbing or confu-
sion chorus (Grinnell 1903).
Vocal signals given with the wing-raise were growl, pee-puu, queea,
chatter (queea in series), and cut-zee. When repeated rapidly, the wings
may be lowered and raised with each utterance, as usually occurred with
growls, or they may be held up continuously and fluctuated only slightly
with the repetition of each vocalization, as usually occurred with chatter.
BILL-UP.- In this posture the head is rotated upward, near the ver-
tical. The neck may be extended, normal, or contracted. The body
plumage is normal or fluffed, while the wings are folded normally or
drooped slightly. The displaying bird may gape. I saw bill-up throughout
the year, but only in close-quarter interactions, in feeding, roosting, and
nesting areas. Yellow-shoulders gave bill-ups to other Yellow-shoulders as
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 189
BULLETIN FLORIDA STATE MUSEUM
well as to grackles, cowbirds, and Troupials. Mated birds also gave bill-
ups when near each other. Bill-up appears to have become incorporated
into the beginning and ending of the song-spread. Bill-up displays are
widespread among icterids and other passerines (Nero 1956a, Andrew
1961).
BILL-DOWN. The head is lowered and the bill is pointed toward the
abdomen. The head plumage may be ruffled. Yellow-shoulders give the
bill-down when near conspecifics in situations similar to those in which
they gave bill-ups, but the posture may indicate a greater tendency to
escape or a greater conflict between escape and attack than does bill-up.
Bill-down is also often given by birds that are moving or about to move
while in the presence of conspecifics. Unlike bill-up, the display is given
by birds that are alone. Andrew (1961) suggested that bill-down ("bill
lowering") is an example of a "reverse movement" to bill-up ("bill
raising") in species in which the latter has a threat function, e.g. it may
act as an appeasement gesture. In this regard it is interesting that bill-
down has been incorporated in song-spread and wing-raise.
HEAD-FORWARD.- In this posture, similar to that described for other
passerines (Andrew 1961), the head is extended toward an opponent,
while the legs may be flexed and plumage normal, fluffed, or sleeked (see
illustrations in Post and Wiley 1976). The wings may be raised slightly at
the shoulders and the bird may gape. Head-forward was seen most often
in agonistic encounters at feeding sites, and it was frequently preceded by
bill-up.
HEAD-IN. -The bird usually crouches, bringing its head in to the
body. The body plumage may be fluffed or ruffled. The beak is often
directed at the nearby individual, and perhaps gaped. This posture is seen
most often during agonistic encounters in the feeding areas, but occa-
sionally near nests by birds that are challenged upon entering the activity
spaces of others. Males also occasionally give the head-in when they ap-
proach females that are giving wing flutter.
WING-FLUTTER. -The body feathers are fluffed, and belly feathers
ruffled, while the wings are held out from the body and vibrated. The legs
are flexed. The tail is spread, and may be elevated or held normal. Wings
are occasionally raised asymmetrically when they are fluttered.
I saw this display from females that were begging food from their
mates or soliciting, usually near their nests. Once a male gave wing-
flutter repeatedly to a female that had just been released from a trap. The
female was trying to remove her bands and exhibited no special behavior.
Another time a male wing-fluttered in the presence of a female that was
nest-building. In many species wing vibration in male courtship is prob-
ably homologous with female precopulatory display (Andrew 1961).
Wing-flutter of both male and female Yellow-shoulders resembles the
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 191
precopulatory displays of Red-winged Blackbirds and other icterids (Nero
1956a).
On those few occasions when I saw copulation, prior to it the female's
tail was above the horizontal, while on other occasions males were re-
pulsed by wing-fluttering females whose tails were held normally. The
head-up tail-up variant is homologous with the female precopulatory
display of other female passerines (Andrew 1961), but other than the posi-
tion of the tail, there is little difference between this form of the display
and that given when a female begs for food. The wing-flutter probably
functions as a distance decreasing mechanism when mated birds are at-
tempting to approach one another. Some males brought food to brooding
or incubating females, and the females wing-fluttered and gaped for food
only when they both were at the nest. In the case of males that did not
regularly deliver food to females, their mates often food-begged and
wing-fluttered when the males arrived at the nest with food for the young
and the females were present. The females took the food, sometimes man-
dibulated it, and then gave it to the nestlings.
WING-TRAIL. As the bird walks slowly, the wings are lowered at the
carpals and the remiges are spread so that the feathers may touch the
ground. Body feathers are raised to varying degrees, but the rump
feathers are usually ruffled. The rectrices are spread and may also drag
the ground. The beak is horizontal or pointed slightly down. Wing-trail
was given by both sexes when I approached nests containing young. On
several occasions I handled young without their parents' wing-trailing,
but when the young screamed, the adults began to wing-trail and also
utter screams. Other vocalizations given with wing-trail were queea,
chwip, greeah, cut-zee, and check. When used by parents whose young
are threatened, the wing-trail may be a relatively unritualized form of
distraction display, as described for many other passerines, including
Red-winged Blackbirds (Orians 1973) and Bobolinks (Dolichonyx
oryzivorous) (Nero 1955).
MOTH FLIGHT. -This display takes the form of short flights in which
the wings are moved slowly and with small amplitude. I saw this peculiar
flight pattern given only by females around their nests. Females gave
moth flights when either approaching or leaving the nest and the male
was near it. Once moth flight preceded the female's wing-trailing, and
several times it preceded wing-fluttering. Females gave the display fre-
quently around the time the young fledged, especially when a human was
near. At these times many short moth flights were accompanied by
vocalizations such as greeah and cut-zee. As pointed out by Andrew
(1956), moth ("impeded") flight may occasionally form part of distraction
display, and in general appears in situations of strong conflict between
escape and approach tendencies. On the other hand, it may also be used
BULLETIN FLORIDA STATE MUSEUM
during courtship, as by male Red-winged Blackbirds during symbolic
nest-site selection (Nero 1956a).
MALE NEST ADVERTISEMENT.-Although females actually construct
nests, during the period of early pairing males stand in the cups of old
nests and pull or jab at nest material. They also try to form the nest cup by
crouching and pushing against its edge with their breasts. Sometimes
males carried nest material away from the nest, but usually dropped it
nearby. All of these activities are associated with the presence of a female
with which the male is attempting to pair. Symbolic nest-building by
males of species in which only the female constructs the nest has been
recorded for several other icterids (Nero 1956a). According to Nero's
descriptions, in the Red-winged Blackbird this behavior seems to be more
fully developed than in the Yellow-shoulder, and seems to be associated
with the period of egg-laying rather than that of pair establishment.
TAIL-FLIP. The tail is moved rapidly up and down (duration about
1.3 sec). During tail-flip other body parts are usually positioned nor-
mally. Tail-flip is given in all situations, with nearly all vocalizations.
Typically, a call such as chwip or check was uttered as the tail reached its
highest point or was just beginning to be lowered. Rate of tail-flipping
varied; for example, a bird that was mobbing humans gave 54/min, while
a preening bird gave 4/min.
SLEEK. The body plumage is compressed, while the bird may flex its
legs, indicating a tendency to fly. Occasionally the bird assumes an erect
posture and sleeks the neck and breast feathers, in which case sleek may
indicate readiness to fly at an opponent. Sleek is probably homologous
with similar postures of other icterids (Orians and Christman 1968).
FOOD-BEGGING. Begging postures of young Yellow-shoulders are
similar to those described for other icterids. The head is held in, the beak
is pointed up and gaped. Wing tips are held into the body, while the wing
is vibrated at the carpus. I saw food-begging by young that had been out
of the nest 24 days.
BREAST-TO-BREAST FIGHTING. -Fighting occurs frequently in ter-
ritorial disputes, and also at feeding locations. Opponents rise in the air,
clawing at each others' breasts and beating their wings against one
another. Occasionally protagonists fall in the water while fighting.
Breast-to-breast fighting is sometimes preceded by a form of pouncing, in
which an attacking individual hops with outstretched feet on the oppo-
nent's back.
SEXUAL CHASING. -Infrequently I saw group sexual chasing, such as
described for Red-winged Blackbirds (Nero 1956a). On the nesting cays
2-4 birds occasionally chased a female, ending in one bird's pursuing her
down into the mangrove roots, and even into the water. These chases
were accompanied by loud calls of queea, check, and cut-zee.
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 193
BILL-WIPING. Under relevant conditions, bill-wiping serves to clean
the bill, usually after feeding or preening. The bird simply wipes one side
of the beak from base to tip on a branch, then perhaps repeats on the
other side. These movements also occurred often in seemingly irrelevant
circumstances, and more frequently than body maintenance alone seemed
to warrant. On 15 February 1973 a bird waiting to feed inside a monkey
cage bill-wiped 24 times in 105 secs. It was probably showing strong con-
flict between approach and escape tendencies, as during the same period
it gave 33 queeas, each with a tail-flip. Bill-wiping was also common dur-
ing mobbing.
DISCUSSION
The distances that nesting Yellow-shoulders commute to gather food
suggest that food distribution is not the determining factor in nest place-
ment. Considering: (1) the energy costs of flying to foraging sites, (2) that
birds do not seem to exchange information about the location of food, and
(3) that food is not highly concentrated within the large areas (e.g. groves
of trees in pastures, scattered trees in savannahs, mangroves) foraging
blackbirds visited, one would expect nests to be spaced uniformly within
the feeding areas (Horn 1968; Waser and Wiley 1979). Wiley and Wiley
(1980) found a similar discrepancy between food distribution and ideal
nest locations in their study of the Yellow-hooded Blackbird in Venezuela.
These birds nested in small patches of suitable marsh vegetation and flew,
independently of each other, long distances outside the nesting marshes.
Similar examples of limited nest sites and widespread feeding grounds
occur in populations of Clay-colored Sparrows (Spizella pallida) (Knap-
ton 1979) and Seaside Sparrows (Ammospiza maritima) (Post 1974). In
these cases the critical factor may be finding secure nest sites, and the
energy budget model (Horn 1968), which considers nest distribution a
function of food supply, is not appropriate, or food may not be limiting in
any of these situations. Adult Yellow-shouldered Blackbirds usually
delivered 3-6 items per trip to the nest, indicating that food was readily
available. In contrast, female Red-winged Blackbirds in Costa Rica
(Orians 1973) and Yellow-hooded Blackbirds in Venezuela (Wiley and
Wiley 1980) usually delivered one item per nest visit.
The varied foraging behavior of the Yellow-shoulder may be at-
tributable to ecological release (Crowell 1961, 1962). Reduced competi-
tion on islands may allow a species to use niches occupied by other forms
on continents. The Yellow-shoulder uses some foraging maneuvers and
foraging sites that in adjacent continents are preempted by woodcreepers
(Dendrocolaptidae), woodpeckers (Picidae), and wrens (Troglodytidae):
probing and gleaning the branches and trunks in the inner zones of trees.
Only one member of these families, the Puerto Rican Woodpecker
BULLETIN FLORIDA STATE MUSEUM
(Melanerpes portoricensis) is sympatric with A. xanthomus. In the SW
coastal zone where the Yellow-shoulder was most common, the Puerto
Rican Woodpecker was rare. Another niche A. xanthomus occupied is
that of orioles: gleaning leaves and probing leaf clusters and terminal
buds in the outer zones of trees, and probing fruit and flowers in all
strata. Only one species of oriole, Icterus dominicensis, is native to Puerto
Rico, and it is uncommon in the habitats A. xanthomus occupies. An in-
troduced oriole, Icterus icterus, was fairly common in the SW coastal
zone, but it foraged in the herb layer and was mainly frugivorous (Post,
unpubl. data).
The phenomenon of expanded niche use in the absence of competition
is well illustrated by the behavior of the Jamaican Blackbird (Nesopsar
nigerrimus), closely related to Agelaius (Bond 1950, Cruz 1978, Wiley
and Cruz 1980). Like A. xanthomus, N. nigerrimus is arboreal, and a
prober of epiphytes (58% of its observed foraging activity, Cruz 1978)
and branches and trunks (10%). The Jamaican Blackbird forages more on
the trunks and inner branches and less in the outer zone of trees than does
the Yellow-shouldered Blackbird. In the Jamaican habitats Cruz (1978)
studied, the Jamaican Oriole (Icterus leucopteryx) foraged mainly in the
outer parts of trees.
The foraging behavior of the Yellow-shoulder has diverged from that
of marsh-dwelling Agelaius, most species of which are ground feeders in
the breeding season. This difference may be a result of reduced competi-
tion combined with scarcity of marsh habitat on Puerto Rico. In North
America Red-winged Blackbirds occasionally feed in trees (pers. obs.).
Another West Indian Agelaius, the Tawny-shouldered Blackbird, is prob-
ably arboreal in the breeding season (Lack 1976) and also nests in palms
(Barbour 1923).
That the two insular species, which presumably evolved from a
marsh-dwelling form similar to A. phoeniceus, are partly arboreal may be
attributable to the elimination of Cuban marshes during the post-
Pleistocene rise in sea levels, though Bond (1950) considers it more likely
that the ancestor of A. humeralis and A. xanthomus evolved arboreal
foraging behavior during a period of competition with the Red-winged
Blackbird (A. p. assimilis), which occupies marshes on Cuba. Whether
the ancestor of xanthomus was arboreal when it arrived on Puerto Rico is
conjectural, but there may have been some large marshes such as those
that existed until recently in the Yauco-Boquer6n Valley, covering an area
of about 150 km2 (Danforth 1926).
The Yellow-shoulder uses at least 26 distinct displays and vocaliza-
tions. This repertoire appears to be larger than that of other monogamous
icterids that have been studied (Orians and Christman 1968), and its size
is closer to that of polygynous species such as the Tricolored and Red-
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 195
winged Blackbirds. Male Red-wings, for instance, have 18 vocalizations
and 12 action patterns, females 6 and 9 (Orians and Christman 1968).
Estimates of the number of displays of various species vary according to
authors' procedures and preferences (Smith 1977). As I lack adequate
spectrographic material in some cases, I reserve judgement on variants of
some vocal signals. More detailed study of the vocal repertoire of the
Yellow-shoulder may reveal it to be larger.
Unlike Red-winged and Tricolored Blackbirds, the Yellow-shoulder
has few intersexual differences in its display repertoire. Each sex appears
to have only one unique visual display, and all the remaining displays,
visual and vocal, are shared by both. Orians and Christman (1968) note a
tendency for males of polygynous species to have a larger number of
displays than females, this disparity being correlated with their advertis-
ing for and holding more mates than monogamous species. The sexual
dimorphism reflected in display repertoire is also found in body size and
plumage of polygamous species. Hamilton (1961) suggested that as those
species breeding farthest from the equator are generally migratory, and
on returning to their breeding grounds have little time for pairing,
dimorphism is also advantageous in reducing intersexual strife and allow-
ing rapid pair formation. Intersexual differences in display repertoire
facilitate rapid mating in these northern species, but species such as the
Yellow-shoulder, that pair well in advance of mating, may be under little
selective pressure to segregate displays sexually. The ancestral condition
was presumably one of intersexual equality in display repertoire size, but
with the evolution of non-monogamous mating systems, sexual segrega-
tion of displays was accentuated, and eventually each sex appropriated
certain displays.
Some differential use of certain displays may occur in the Yellow-
shoulder, such as has been shown for the Black-capped Chickadee, Parus
atricapillus (Ficken et al., 1978). Examples of different degree of usage by
each sex probably occur in pee-puu and queea in series (chatter), both
used more often by females, and (single) queeas used more often by males
in mobbing.
Among species of North Temperate icterids that are not cryptic, those
having large territories usually have more displays than others that defend
only a small area around the nest (Orians and Christman 1968). Yellow-
headed and Red-winged Blackbirds have more displays than Brewer's
Blackbirds or Common Crackles. This difference is related to the larger
number of displays transmitting messages over long distances that the first
two species have, such as flight displays and song.
More needs to be learned about the influence of winter social
organization on size of display repertoire, but Orians and Christman
(1968) suggested that the large number of displays of the Tricolored
BULLETIN FLORIDA STATE MUSEUM
Blackbird may be due in part to year-round association of both sexes. A
similar effect may occur in the Yellow-shoulder, whose display repertoire
is about the same size as that of the Tricolor.
As Wiley (1976a) found for the Common Grackle, few of the vocal
signals of the Yellow-shoulder transmit unequivocal information about
the internal state of the transmitter, but most vocalizations are given in
close-quarter interactions, whether in nesting or feeding places, and more
precise information is available to the receiver by reference to context or
to accompanying visual display. Most of the Yellow-shoulder sounds are
abrupt, broad spectrum signals that are easily located in noisy en-
vironments such as exist in nesting colonies, feeding flocks, or roosts.
The growl vocalization given with song-spread has a simpler structure
than similar sounds of other icterids. There has probably been little selec-
tive pressure for further elaboration of song, because of the small territory
size of the Yellow-shoulder, and of relaxation of selection for specific
distinctiveness in the absence of closely related species (Grant 1972).
In species occupying large territories, song must be transmitted
without distortion over relatively long distance. The song (growl) of the
Yellow-shoulder is highly modulated, and it covers a wide frequency
range, so it is probably of little use in long-range communication. Rather,
the growl is used repetitively in close-quarter interactions, and coupled
with action patterns such as song-spread and wing-raise. Song-answering,
a form of antiphonal singing that Wiley (1976c) considers to be important
in the vocal coordination of Common Grackle pairs, was also
characteristic of Yellow-shoulders during their period of pairing and nest
site selection. The timing of song-answering is thus similar to that of
grackles.
In the Yellow-shoulder, song-spread is always accompanied by song
(growl), and growl is invariably accompanied by some form of wing
elevation.
In the Red-winged Blackbird, song-spread and song are not always
coupled (Peek 1972). Red-wing song is given most often with incipient
song-spreads, and also without any wing elevation. The Red-wing's song-
spread is used mainly in close-quarter interactions, while its song is used
for both long and short range communication. In the Red-wing, the vocal
component of song-spread is perhaps diverging in function from the visual
component, because of its importance in long-range communication.
Most of the vocal signals of the Yellow-shoulder are emphasized at
high frequencies, have wide frequency ranges and are highly modulated.
These characteristics make them easily localized and identified, but
mainly by nearby individuals in the open, as these characteristics also lead
to rapid attenuation and distortion over distance or through dense vegeta-
tion such as mangroves (Morton 1975). One vocalization, pee-puu, ap-
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 197
pears to be composed of pure tones, and was used frequently for long-
range communication by birds nesting in red mangroves.
Correlated with the historic rarity of predators on Puerto Rico, the
Yellow-shoulder has no specialized predator ("hawk") alarm calls as do
several continental icterids. Instead, the Yellow-shoulder uses general
alarm calls such as queea, that appear to have functions besides predator
warning. Cut-zee seems to be the one vocalization most closely associated
with predator mobbing response, and its use is largely restricted to the
nest vicinity. Even this degree of specialization is interesting, and possibly
cut-zee has evolved primarily in response to aerial predators such as gulls
and frigatebirds that have probably been present in Puerto Rico since the
blackbird's ancestors arrived.
Unlike some open-country icterids, such as the Red-winged and
Yellow-headed Blackbirds (Orians and Christman 1968), the Yellow-
shoulder does not have stereotyped flight displays, and its flight vocaliza-
tion is composed of a recombinable series of other signals. Displays related
to bonding are also few: wing-flutter, probably homologous with
precopulatory displays of other passerines; male nest-building display;
and pink call, given between parents and young. Displays relating to
establishment of pair bonds may be more important for polygynous,
usually highly dimorphic species, in which male dominance is well
developed, but mollified by the use of these displays to facilitate pairing.
In the presence of predators Yellow-shoulders use a variety of visual
displays, but as with vocal displays, no single one has become specialized
as a predator warning. As with cut-zee, one visual display, wing-trail, oc-
curs mainly in the presence of predators, and it may be homologous with
distraction displays given by other passerines in the vicinity of their nests.
CONCLUSIONS
The Yellow-shouldered Blackbird, probably derived from marsh-
dwelling and ground-feeding blackbirds of North America, has diverged
from the behavior of its closest mainland relatives. Its foraging behavior is
similar to that of orioles and woodpeckers. This may be a result of re-
duced competition, and also lack of suitable feeding sites on the ground,
particularly marshes, in Puerto Rico.
The species does not engage in coactive feeding as do some relatives,
nor exhibit cooperative breeding, though it aggregates when nesting. As
colony sites are isolated from feeding grounds, and the types of food
delivered to the young are widely distributed within relatively large
patches, it would be predicted, on the basis of maximum foraging effi-
ciency, that nests should be uniformly distributed in the food field. As this
is not the case, nest dispersion is probably a result of selective pressure ex-
BULLETIN FLORIDA STATE MUSEUM
erted by predation. This conclusion is supported by the frequency with
which colony members engage in communal mobbing, the large number
of displays and vocalizations used in the presence of predators, the
inaccessible sites used for nesting, and the relatively high prevalence of
nest predation.
Yellow-shouldered Blackbirds were monogamous in all the habitats in
which they were studied. Among other icterid species that nest in colonies
and visit remote feeding grounds, polygamy is seen at least occasionally.
In the Yellow-shouldered Blackbird, I conclude that the conditions for
male emancipation are not realized because of the crucial role of the male
in parental care. The mate acquisition and pair maintenance behavior of
the species, in which birds affiliate long before breeding, and in which
pairing takes place around nest sites of previous years, assures high levels
of intra- and interseasonal mate loyalty. A long period of affiliation
before breeding may also be advantageous if it allows pairs to respond
rapidly to unpredictable events (spring rains). The size and complexity of
the display repertoire has probably evolved as a response to pressure for a
closely coordinated monogamous pair bond of relatively long duration.
The species' social behavior, in the context of its high reproductive suc-
cess, lack of dispersal, high annual survival, and especially delayed
breeding of both sexes implies that a conservative reproductive strategy
has been evolved.
LITERATURE CITED
Andrew, R. J. 1956. Fear responses in Emberiza spp. Anim. Behav. 4:125-132.
1961. The displays given by passerines in courtship and reproductive fighting:
A review. Ibis 103a:315-348, 549-579.
Balph, M. H., and D. F. Balph. 1976. Some factors influencing observed sex ratios in a
population of Evening Grosbeaks. Bird-Banding 47:340-344.
Barbour, T. 1923. The birds of Cuba. Memoir no. 6, Nuttall Ornithol. Club.
Barnis, V., Jr. 1946. The birds of Mona Island, Puerto Rico. Auk 63:318-327.
Beer, J. R., and D. Tibbitts. 1950. Nesting behavior of the Red-winged Blackbird. Flicker
22:61-77.
Bent, A. C. 1958. Life histories of North American blackbirds, orioles, tanagers, and allies.
U.S. Nat. Mus. Bull. 211.
Bond, J. 1950. Some remarks on West Indian Icteridae. Wilson Bull. 62:216-217.
Botkin, D. B., and R. S. Miller. 1974. Mortality rates and survival of birds. Am. Natur.
108:181-192.
Boucher, D. H. 1977. On wasting parental investment. Am. Natur. 111:786-788.
Caccamise, D. F. 1976. Nesting mortality in the Red-winged Blackbird. Auk 93:517-534.
Coulson, J. D., and R. D. Wooller. 1976. Differential survival rates among breeding Kitti-
wake Gulls Rissa tridactyla (L.). J. Anim. Ecol. 45:205-213.
Cox, G. W., and R. E. Ricklefs. 1977. Species diversity and ecological release in Caribbean
land bird faunas. Oikos 28:113-122.
Crawford, R. D. 1977. Breeding biology of year-old and older female Red-winged and
Yellow-headed Blackbirds. Wilson Bull. 89:73-80.
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 199
Crook, J. H. 1964. The evolution of social organization and visual communication in the
weaver birds (Ploceinae). Behaviour, Suppl. No. 10.
1965. The adaptive significance of avian social organizations. Symp. Zool.
Soc. Lond. 14:181-218.
Crowell, K. L. 1961. The effects of reduced competition in birds. Proc. Natl. Acad. Sci.
47:240-243.
1962. Reduced interspecific competition among the birds of Bermuda. Ecology
43:75-88.
Cruz, A. 1978. Adaptive evolution in the Jamacian blackbird Nesopsar nigerrimnis. Ornis
Scand. 9:130-137.
Danforth, S. T. 1926. An ecological study of Cartagena Lagoon, Porto Rico, with special
reference to the birds. J. Dept. Agric. Porto Rico 10:1-36.
Dawkins, R., and T. R. Carlisle. 1976. Parental investment, mate desertion and a fallacy.
Nature 262:131-133.
Emlen, J. M. 1966. The role of time and energy in food preference. Am. Natur. 100:611-
617.
Fankhauser, D. P. 1967. Survival rates in Red-winged Blackbirds. Bird-Banding 38:139-142.
Farner, D. S. 1945. Age groups and longevity in the American Robin. Wilson Bull. 57:56-74.
Ficken, M. S., R. W. Ficken, and S. R. Witkin. 1978. Vocal repertoire of the Black-capped
Chickadee. Auk 95:34-48.
Fogden, M. P. L. 1972. The seasonality and population dynamics of equatorial forest birds
in Sarawak. Ibis 114:307-343.
Foster, M. S. 1974. Rain, feeding behavior, and clutch size in tropical birds. Auk 91:722-726.
Gladstone, D. E. 1979. Promiscuity in monogamous colonial birds. Am. Natur. 114:545-
557.
Gleason, H. A., and M. T. Cook. 1926. Plant ecology of Porto Rico. N.Y. Acad. Sci. Scien-
tific Survey of Porto Rico and the Virgin Islands. Vol. 7, part 1.
Giiransson, G., J. Karlsson, S. G. Nilsson, and S. Ulfstrand. 1975. Predation on birds' nests
in relation to antipredator aggression and nest density: An experimental study. Oikos
26:117-120.
Grant, P. R. 1972. Convergent and divergent character displacement. Biol. J. Linn. Soc.
4:39-68.
Greenlaw, J. S. 1978. The relation of breeding schedule and clutch size to food supply in the
Rufous-sided Towhee. Condor 80:24-33.
Grinnell, J. 1903. Call notes of the Bush-tit. Condor 5:85-87.
Grosskopf, G. 1964. Sterblichkeit und Purchschnittsalter einiger Kustenvogel. J. Orn. 105:
427-449.
Haartman, L. von. 1971. Population dynamics. In Avaian biology 1 (D. S. Farner and
J. R. King, Eds.). Academic Press, New York.
Hamilton, T. H. 1961. On the functions and causes of sexual dimorphism in breeding
plumage characters of North American species of warblers and orioles. Am. Natur.
95:121-123.
Hickey, J. J. 1952. Survival studies of banded birds. Special scientific report: Wildlife No. 15.
Bureau of Sport Fisheries and Wildlife.
Holcomb, L. C., and G. Twiest. 1971. Growth and calculation of age for Red-winged
Blackbird nestlings. Bird-Banding 42:1-17.
Holdridge, L. R. 1940. Some notes on the mangrove swamps of Puerto Rico. Caribbean
Forester 1:19-29.
Horn, H. S. 1968. The adaptive significance of colonial nesting in the Brewer's Blackbird
(Euphagus cyanocephalus). Ecology 49:682:694.
.1970. Social behavior of nesting Brewer's Blackbirds. Condor 72:15-23.
Howe, H. F. 1979. Evolutionary aspects of parental care in the Common Grackle. Quiiscalus
quiscula L. Evolution 33:41-51.
BULLETIN FLORIDA STATE MUSEUM
Jones, P. J. and P. Ward. 1976. The level of reserve protein as the proximate factor control-
ling the timing of breeding and clutch-size in the Red-billed Quelea Quelea quelea. Ibis
118:547-574.
Keast, J. A., and A. J. Marshall. 1954. The influence of drought and rainfall on reproduction
in Australian desert birds. Proc. Zool. Soc., London 124:493-499.
Knapton, R. W. 1979. Optimal size of territory in the Clay-colored Sparrow. Canadian J.
Zool. 57:1358-1370.
Lack, D. 1954. The natural regulation of animal numbers. Oxford Univ. Press, Oxford.
1976. Island biology, illustrated by the land birds of Jamaica. Univ. California
Press, Berkeley.
Marchant, S. 1959. The breeding season in S.W. Ecuador. Ibis 101:137-152.
1960. The breeding of some S.W. Ecuadorian birds. Ibis 102:349-382, 584-599.
Martin, S. G. 1974. Adaptations for polygynous breeding in the Bobolink Dolichonyx
oryzivorous. Am. Zool. 14:109-119.
Maxwell, G. R. II, J. M. Nocilly, and R. I. Shearer. 1976. Observations at a cavity nest of the
Common Crackle and an analysis of grackle nest sites. Wilson Bull. 88:505-507.
Mayfield, H. 1961. Nesting success calculated from exposure. Wilson Bull. 73:255-261.
Maynard Smith, J. 1977. Parental investment: A prospective analysis. Anim. Behav. 25:1-9.
Morton, E. S. 1971. Nest predation affecting the breeding season of the Clay-colored Robin, a
tropical song bird. Science 171:920-921.
1975. Ecological sources of selection on avian sound. Am. Natur. 108:17-34.
Nero, R. W. 1955. Distraction display in the Bobolink. Passenger Pigeon 17:34.
1956a. A behavior study of the Red-winged Blackbird. Wilson Bull. 66:5-37,
129-150.
1956b. Red-wing nesting in bird house. Auk 73:284.
Nice, M. M. 1957. Nesting success in altricial birds. Auk 74:305-321.
Onuf, C. P., J. M. Teal, and I. Valiela. 1977. Interactions of nutrients, plant growth and her-
bivory in a mangrove ecosystem. Ecology 58:514-526.
Orians, G. H. 1961. The ecology of blackbird (Agelaius) social systems. Ecol. Monogr.
31:283-312.
.1966. Food of nestling Yellow-headed Blackbirds, Cariboo Parklands, British
Columbia. Condor 68: 321-337.
1969. On the evolution of mating systems in birds and mammals. Am. Natur.
103:589-603.
1972. The adaptive significance of mating systems in the Icteridae. Proc. 15 Int.
Orn. Congr.:389-398.
1973. The Red-winged Blackbird in tropical marshes. Condor 75:28-42.
1980. Some adaptations of marsh-nesting blackbirds. Princeton Univ. Press,
Princeton, N.J.
and G. M. Christman. 1968. A comparative study of the behavior of Red-
winged, Tricolored and Yellow-headed Blackbirds. Univ. California Publs. Zool.
84:1-81.
., L. Erckmann, and J. C. Schultz. 1977. Nesting and other habits of the Boli-
vian Blackbird (Oreopsar bolivianus). Condor 79:250-256.
., C. E. Orians, and K. J. Orians. 1977. Helpers at the nest in some Argentine
blackbirds. Pp. 137-151 In Evolutionary Ecology. (B. Stonehouse and C. Perrins, Eds.)
University Park Press, Baltimore.
Payne, R. B. 1969. Breeding seasons and reproductive physiology of Tricolored Blackbirds
and Red-winged Blackbirds. Univ. California Publs. Zool. 90:1-137.
Peek, F. W. 1972. An experimental study of the territorial function of vocal and visual display
in the male Red-winged Blackbird (Agelaius phoeniceus). Anim. Behav. 20:112-118.
Post, W. 1974. Functional analysis of space-related behavior in the Seaside Sparrow. Ecology
55:564-575.
Vol. 26, No. 3
1981 POST: YELLOW-SHOULDERED BLACKBIRD BIOLOGY 201
1978. Social and foraging behavior of warblers wintering in Puerto Rican coastal
scrub. Wilson Bull. 90:197-214.
1981. The prevalence of some ectoparasites, diseases, and abnormalities in the
Yellow-shouldered Blackbird. J. Field Ornith. 52:16-22.
and J. W. Wiley. 1976. The Yellow-shouldered Blackbird-present and future.
Am. Birds 30:13-20.
1977. Reproductive interactions of the Shiny Cowbird and the Yellow-shoul-
dered Blackbird. Condor 79:176-184.
Ricklefs, R. E. 1967. A graphical method of fitting equations to growth curves. Ecology
48:978-983.
1969. An analysis of nesting mortality in birds. Smithsonian Contrib. Zool.
9:1-48.
.1976. Growth rates of birds in the humid New World tropics. Ibis 118:179-207.
Ridgway, R. 1902. The birds of North and Middle America. Part II. U.S. Natl. Mus. Bull. 50.
Robertson, R. J. 1972. Optimal niche space of the Red-winged Blackbird (Agelaiu.s phoeni-
ceus). I. Nesting success in marsh and upland habitat. Canadian J. Zool. 50:247-263.
Search, W. A. 1979. Female choice of mates: A general model for birds and its application
to Red-winged Blackbirds (Agelaius phoeniceus). Am. Natur. 114:77-100.
Skutch, A. F. 1966. A breeding bird census and nesting success in Central America. Ibis
108:1-16.
Smith, H. M. 1943. Size of breeding populations in relation to egg-laying and reproductive
success in the Eastern Red-wing (Agelainis p. phocniceuts). Ecology 24:183-207.
Smith, J. N. M. 1978. Division of labour by Song Sparrows feeding fledged young. Canadian
J. Zool. 56:187-191.
Smith, W. J. 1977. The behavior of communicating. An ethological approach. Harvard
Univ. Press, Cambridge.
Smithe, F. B. 1974. Naturalist's color guide supplement. American Museum Natural History,
New York.
Snow, D. W., and A. Lill. 1974. Longevity records for some neotropical land birds. Condor
76:262-267.
Thorpe, W. H. 1961. Bird-song. Cambridge Univ. Press, Cambridge.
Trivers, R. 1972. Parental investment and sexual selection. In Sexual selection and the des-
cent of man. B. Campbell, Ed. Aldine Publ. Co., Chicago.
Vandenbergh, J. G., and S. Vessey. 1968. Seasonal Breeding of free-ranging rhesus monkeys
and related ecological factors. J. Reprod. Fert. 15:71-79.
Verner, J. 1964. Evolution of polygamy in the Long-billed Marsh Wren. Evolution 18:
252-261.
S1965. Time budgets of the male Long-billed Marsh Wren during the breeding
season. Condor 67:125-139.
Waser, P. M., and R. H. Wiley. 1979. Mechanisms and evolution of spacing in animals.
In Handbook of Behavioral Neurobiology 3 (P. Marler and T. Vandenbergh, Eds.).
Weatherhead, P. J. 1979. Do Savannah Sparrows commit the concorde fallacy? Behav.
Ecol. Sociobiol. 5:373-381.
Wetmore, A. 1916. Birds of Porto Rico. U.S. Dept. Agric. Bull. 326:1-140.
Wiley, R. H. 1976a. Communication and spatial relationships in a colony of Common
Crackles. Anim. Behav. 24:570-584.
1976b. Affiliation between the sexes in Common Crackles. I. Specificity and
seasonal progression. Z. Tierpsychol. 40:59-79.
S1976c. Affiliation between the sexes in Common Crackles. II: Spatial and vocal
coordination. Z. Tierpsychol. 40:244-264.
and A. Cruz. 1980. The Jamaican Blackbird: A "natural experiment" for hypo-
theses in socioecology. Evol. Biol. 13:261-293.
202 BULLETIN FLORIDA STATE MUSEUM Vol. 26, No. 3
and S. A. Hartnett. 1976. Effects of interactions with older males on behavior
and reproductive development in first-year male Red-winged Blackbirds Agelaids
phoeniceus. J. Exper. Zoo. 196:231-242.
and M. S. Wiley. 1980. Spacing and timing in the nesting ecology of a tropical
blackbird: comparison of populations in different environments. Ecol. Monogr. 50:
153-178.
Williams, L. 1952. Breeding behavior of the Brewer Blackbird. Condor 54:3-47.
Willson, M. F. 1966. Breeding ecology of the Yellow-headed Blackbird. Ecol. Monogr. 36:
51-77.
Wolda, M. 1978. Seasonal fluctuations in rainfall, food and abundance of tropical insects.
J. Anim. Ecol. 47:369-381.
Young, H. 1963. Age specific mortality in the eggs and nestlings of blackbirds. Auk 80:
145-155.
Contributions to the BULLETIN OF THE FLORIDA STATE MUSEUM, BIOLOGICAL SCI-
ENCES SERIES, may be in any field of biology. Manuscripts dealing with natural history of
systematic problems involving the southeastern United States or the New World tropics are
solicited especially. Manuscripts should be of medium length-circa 35 to 200 pages (10,500-
16,000 words). Examination for suitability is made by an Editorial Board.
The BULLETIN is distributed worldwide through institutional subscriptions and exchanges.
It is considered the responsibility of the author to distribute his paper to all interested individ-
uals. To aid in this the authors) receives) 50 copies free, and he may purchase additional
separates at cost if ordered when page proof is returned. The author is also responsible for any
charges incurred for alterations made by him on galley or page proofs. The Museum will send
an invoice to the author for this amount upon completion of publication.
PREPARATION OF MANUSCRIPT
Contributors should consult recent numbers of the BULLETIN for preferred style and format.
Highly recommended as a guide is the CBE Style Manual, 3rd Edition, 1972 (Washington, D.C.,
Amer. Inst. Biol. Sci.).
MSS must be submitted in duplicate (please no onionskin) and satisfy the following minimal re-
quirements: They should be typewritten, double-spaced (especially tables, figure captions, and
"Literature Cited"), on one side of numbered sheets of standard (8-1/2 x 11 in.) bond paper,
with at least one-inch margins all around. Figure legends and tables should be typed on separate
sheets. All illustrations are referred to as figures. They must comply with the following stan-
dards: Photographs should be sharp, with good contrast, and printed on glossy paper. Drawings
should be made with dense black waterproof ink on quality paper or illustration board. All il-
lustrations should have a cover sheet. All lettering will be medium weight, sans-serif type (e.g.
Futura Medium, News Gothic) in cutout, dry transfer, or lettering guide letters. Make allow-
ance so that after reduction no lowercase letter will be less than 1 mm high (2 mm is preferred)
nor any capital letter greater than 5 mm high. The maximum size for illustrations is 8-5/8 x 14 in.
(twice typepage size); illustrations should not be less than typepage width (4-5/16 in.). Designate
the top of each illustration and identify on the back with soft pencil by author's name, MS title,
and figure number.
All manuscripts not submitted in BULLETIN format will be returned to the author for retyping.
Manuscripts and all editorial matters should be addressed to:
Managing Editor of the BULLETIN
Florida State Museum
Museum Road
University of Florida
Gainesville, FL 32611
