Neotropical Avian Frugivores Patterns of Behavior Morphology and Nutrition With Consequences For Fruit Selection

Neotropical Avian Frugivores: Patterns of Behavior, Morphology, and Nutrition, with
Consequences for Fruit Selection

Author(s): Timothy C. Moermond and Julie Sloan Denslow
Reviewed work(s):
Source: Ornithological Monographs, No. 36, Neotropical Ornithology (1985), pp. 865-897
Published by: University of California Press for the American Ornithologists' Union
Stable URL: http://www.jstor.org/stable/40168322 .
Accessed: 20/09/2012 16:10
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
University of California Press and American Ornithologists' Union are collaborating with JSTOR to digitize,
preserve and extend access to Ornithological Monographs.
http://www.jstor.org
NEOTROPICALAVIAN FRUGIVORES:PATTERNS OF
BEHAVIOR,MORPHOLOGY,AND NUTRITION, WITH
CONSEQUENCESFOR FRUIT SELECTION
Timothy C. Moermond1 and Julie Sloan Denslow2
department of Zoology, and 2Department of Botany, University of Wisconsin-Madison,
Madison, Wisconsin 53706 USA
ABSTRACT. A large number of neotropical bird species in many families reg-

ularly eat fruit. We discuss the physiological, morphological, and behavioral adap-
tations associated with eating fruits. Fruits present a number of difficulties for the
frugivores, such as the low protein to calorie ratio, the watery bulk, and the undi-
gestible seed mass. Although we find no consistent morphological adaptations for
digesting fruits, we show that important adaptations in digestive physiology are to
be expected. Both large and small birds have similarly diverse diets, but small birds,
such as tanagers and manakins, eat mainly small, carbohydrate-rich fruits, while large
species, such as cotingas and toucans, eat a wider range of fruit sizes including both
small, carbohydrate-rich fruits and large, lipid-rich fruits.
Most fruit-eaters have large gapes, and many birds swallow fruits whole, although
a number of bird species, both small and large, may facultatively eat some large fruits
piecemeal. Tanagers and some emberizid finches generally eat fruits by mashing the
pulp, swallowing the pulp, juices and small seeds, but dropping medium and large
seeds. Birds that take fruits on the wing usually have short, wide, flat bills, whereas
those that reach fruits from perches often have long and narrow bills. Differences in
bill structure affect the manner of fruit handling and may influence fruit selection.
Most species take fruits either primarily on the wing or primarily from perches.
We divide these two general foraging methods into four foraging maneuvers on the
wing (hovering, stalling, swooping, and snatching) and three foraging maneuvers from
a perch (picking, reaching, and hanging). These foraging techniques differentially affect
access to fruits. Differences in flight maneuvers are associated with differences in
wing-loading, wing aspect ratio, and degree of slotting. Likewise differences in ability
to reach fruit from a perch are correlated with differences in tarsus shape and leg
musculature. Very few birds are able to reach fruits well and also pluck them expertly
on the wing.
The dichotomy of frugivorous birds into specialists and generalists, with only a
couple exceptions, is not supported by comparison of diet, behavior, morphology, or
function as dispersers. The important differences among fruit-eating birds depend on
the constraints of morphology on behavior. Differences in digestive systems, bill
strength and shape, wing and leg morphology, and body weight differentially affect
the accessibility of fruits, the ease with which they are handled, the rate at which they
are eaten, the efficiency with which they are digested, and thus influence which fruits
are selected. The presence of fruit-eating birds in so many families is not evidence
for lack of requirements for eating fruits but rather represents adaptive specialization
to exploit fruits displayed in different ways.
RESUMEN. En las zonas neotropicales se encuentran un gran numero de especies

de aves pertenecientes a numerosas familias cuya dieta consiste de frutas. Presentamos
una discusi6n sobre las adaptaciones fisiol6gicas, morfol6gicas y de comportamiento
que estan asociadas con la costumbre de comer frutas. El comer frutas le presenta
una serie de problemas a aves frugivoras como por ejemplo: una relaci6n baja entre
proteinas y calorias, un gran volumen de materia acuosa y la presencia de semillas
indigeribles. Aunque no encontramos adaptaciones digestivas morfologicas generates
para comer frutas, demostramos que adaptaciones digestivas fisiol6gicas en aves
frugivoras son de esperarse. Las aves frugivoras, tanto las grandes como las pequenas,
tienen dietas similares en cuanto a la diversidad de frutas. Sin embargo, las aves
pequenas, como las tangaras y los saltarines, se alimentan mayormente de frutas
pequenas que son ricas en carbohidratos; mientras que las aves de mayor tamano,
como las cotingas y los tucanes, se alimentan de frutas de diversos tamanos; tanto
de frutas pequenas ricas en carbohidratos, como de frutas grandes ricas en lipidos.
La mayoria de las especies frugivoras tienen un rictus ancho y muchas ingieren la
fruta entera sin despedazarla. Sin embargo, algunas especies, tanto grandes como
pequenas, tienen la capacidad facultativa de comerze la fruta trozo a trozo. Las
865
866 ORNITHOLOGICAL MONOGRAPHS NO. 36
tangaras y algunos de los pinzones usualmente se comen la fruta despues de aplastarla

con el pico. Usando este metodo ellos ingieren la pulpa, el jugo y las semillas pequenas,
pero rechazan las semillas grandes y medianas. Las especies que obtienen la fruta
volando, generalmente tienen picos cortos, anchos y de poca profundidad. Mientras
que las que obtienen la fruta desde una percha comunmente tienen picos largos y
estrechos. Estas diferencias en la estructura del pico afectan la manera en que el ave
manipula la fruta y pueden influenciar el proceso de selecci6n entre diferentes frutas.
La mayoria de las especies usan uno de dos m6todos principales para obtener las
frutas, las toman volando o las toman desde una percha. Para las especies que obtienen
las frutas volando, dividimos las maniobras de forrajeo en cuatro categorias: "hov-
ering," "stalling," "snatching" y "swooping." Para las especies que obtienen la fruta
desde una percha, dividimos las maniobras de forrajeo en tres categorias: "pick-
ing," "reaching" y "hanging." Todos estos metodos de forrajeo afectan de manera
diferente el acceso que un ave pueda tener a una fruta. Las diferencias en maniobras
de vuelo estan asociadas con diferencias aerodinamicas como el "wing-loading,"
"wing aspect ratio" y el grado de "slotting." De igual manera, las diferencias entre
las especies en la abilidad de alcanzar frutas desde una percha estan asociadas con
diferencias en la forma del tarso y la musculatura de las patas. Son muy pocas las
especies que tienen la capacidad de obtener frutas desde una percha facUmente y a
la misma vez poder tomar frutas en vuelo con destreza.
La dicotomia existente entre aves frugivoras "especialistas" y aves frugivoras "ge-
neralistas" no esta apoyada por nuestras comparasiones de las dietas, comporta-
mientos, morfologias, o la funci6n de las especies como dispersoras de semillas. Las
diferencias mas importantes entre las aves frugivoras se basan en las restricciones
impuestas por la morfologia y el comportamiento del ave. Diferencias en el sistema
digestivo, la fuerza y connguraci6n del pico, la morfologia de las patas y alas y el
peso del ave afectan de la manera diferente la accesibilidad de las frutas, la facilidad
con que se manipulan, la rapidez con que se ingieren y la eficiencia con que se digieren
y por lo tanto, influyen de manera muy importante en el proceso de selecci6n
entre diferentes frutas. La presencia de tantas aves frugivoras pertenencientes a nu-
merosas familias no representa una evidencia de que no se necesitan adaptaciones
para comer frutas, si no demuestra numerosas especializaciones adaptivas para ex-
plotar frutas que estan presentadas en maneras diferentes.
Fruit-eating among tropical birds is so wide-spread as to make definition of "frugivorous"

difficult. Few species are so restricted to fruits that their nestlings are completely reared on
fruits (Morton 1973). Oilbirds (Steatornis caripensis, Snow 196 1, 1962c) and Bearded Bellbirds
(Procnias averano, Snow 1970) are in this category. At the other extreme are the many primarily
insectivorous birds that exploit the abundant fruits of some trees (e.g., Eisenmann 196 1; Olson
and Blum 1968; Morton 1971; Leek 1972a, 1973; Howe and DeSteven 1979) or take an
occasional fruit while foraging for insects. Traylor and Fitzpatrick (1982) noted for flycatchers
(Tyrannidae) that while fruit features heavily in the diets of only a few species, nearly all eat
some fruit occasionally. The same is probably true of woodpeckers, and even rails have been
observed taking fruits in Costa Rica (Aramides cajanea feeding on low growing Melastoma-
taceae berries, D. J. Levey, pers. comm.; Porphyrula feeding on Heliconia fruits, Skutch 1933).
Between these extremes are birds whose diets consist of some mixture of fruits and insects
and occasionally vertebrates, snails, or leaves (Jenkins 1969; Morton 1978; Stiles 1983).
Here we use the word "frugivore" for birds whose diets include a substantial portion of
fruit at least during some seasons, rather than confining it to birds that eat solely fruits. Nearly
all of the so-called frugivorous birds eat other types of foods as well (usually insects or other
invertebrates) and should be properly referred to as omnivores. However, the term "frugivore"
has been commonly used for such frequent but not exclusive fruit-eaters as tanagers. Here we
use the terms "frugivore" and "fruit-eater" interchangeably.
Adaptations of fruits that enhance seed dispersal by birds have been described in detail
(Ridley 1930; van der Pijl 1969). The most obvious common characteristic is an energy or
nutrient reward in the form of a soft, fleshy pulp surrounding the seed. In addition, bright
colors and/or morphological modifications presumably advertise the presence of the reward
(Morden-Moore and Willson 1982; Willson and Thompson 1982; Janson 1983). The seed
itself is frequently provided with some form of protection that reduces the probability of
destruction by the bird or discourages invertebrate seed-eaters, e.g., hard seed coat or toxins
in the seed, seed coat, or fruit pulp (Herrera 1982). Given that fruits, especially small-seeded,
NEOTROPICALORNITHOLOGY 867
sweet, berry-like fruits, are at times abundant, visible, and easily accessible, it is not surprising
that many birds occasionally exploit this ready source of energy.
Morphological adaptations for fruit-eating in birds have not been so clearly identified. The
most commonly cited examples concern a relatively large gape (Snow 1973) and lack of crop
or muscular gizzard. Even these modifications of the gut may not be typical (see Morphology
of the Digestive Tract below). Some species exhibit behavioral adaptations, such as regional
or elevational migrations that track local fruit abundances (e.g., Fogden 1972; Crome 1975a,
c; Morton 1977; Karr et al. 1982; Wheelwright 1983; F. G. Stiles, pers. comm.). Most fruit-
eating birds do not specialize on one or a few species of fruits and dietary overlap among
many fruit-eating birds is broad (e.g., Snow and Snow 1971; Leek 1971b; Crome 1975a),
especially with respect to small-seeded fruits (Eisenmann 1961; Land 1963; Willis 1966; Olson
and Blum 1968; Leek 1973; Howe 1981, 1982).
In contrast to this general picture is the persistent notion of frugivores arrayed on a con-
tinuum between specialists and generalists (McKey 1975; Howe and Estabrook 1977; Snow
1980). The idea is based on the assumption that reliable attraction of dispersers to large-
seeded fruits entails provision of a high quality reward to the bird. Hence, large-seeded fruits
are seen to be nutritious and closely linked evolutionarily with highly frugivorous birds that
depend on fruits for most of their protein and energy needs (hence specialists) and whose
gapes are large enough to handle large seeds (Snow 1971a; McKey 1975). These birds are also
seen to provide high quality dispersal, i.e., they preferentially feed on these large fruits and
disperse the "seedsaway from the parent plant. The seeds are not destroyed. Most small, fruit-
eating birds are seen as opportunists (generalists) that take fruits nonselectively as they are
encountered and are willing to settle for small, juicy, low-reward fruits (McKey 1975; Howe
and Estabrook 1977).
However, there is no reason to expect that fruit specialists provide a higher quality dispersal
than generalists (Wheelwright and Orians 1982), and we shall show that the diets of specialists
and generalists often have much in common. Our own studies of Costa Rican fruit-eating
birds have suggested that fruits displayed in different ways are differentially accessible to birds
that forage in different ways (Moermond and Denslow 1983; Denslow and Moermond, in
press). Here we compare frugivores with respect to fruit choices, fruit handling, and foraging
behavior and suggest that these factors form a more functional basis on which to categorize
this phylogenetically and morphologically diverse group of birds.
WHO EATS FRUITS
Neotropical fruit-eating birds are found among a broad variety of bird families. Very few
eat only fruits as adults and nestlings. Most supplement their fruit diets with different, relatively
protein-rich foods, and those that do feed almost exclusively on fruits as adults usually feed
insects to their young. The nature of these alternative protein sources- seeds on one hand
and insects or vertebrates on the other- provides a convenient basis for classifying frugivores.
Seed-eating fruvigores.- That relatively few species combine a diet of seeds and fruit pulp
is not surprising in view of the different digestive apparati necessary for processing the two
foods. Granivorous birds are characterized by large crops, muscular gizzards, long small
intestines, and slow passage rates through the digestive tract in contrast to frugivores (Ziswiler
and Farner 1972). Many fruit-eaters in this category are, in fact, feeding on both seed and
pulp of fleshy fruits. They are often seed destroyers, rather than dispersers, and generally
prefer green (unripe) or partially ripe fruits to fully ripe fruits in which the seed coats are likely
to be fully hardened. Most also supplement their diets with young leaves, shoot tips, buds,
and occasional insects.
Columba pigeons swallow seeds and berries whole; large seeds especially are undoubtedly
destroyed in the large, muscular gizzard, although very small seeds, such as those of Cecropia,
Ficus, or Miconia, may pass through intact. The bill is relatively weak and only very soft
fruits can be eaten by biting or tearing out chunks (Goodwin 1970). New World parrots are
also primarily seed-eaters that break up seeds with their bills before swallowing (Forshaw
1978; Howe 1981; Janzen 1981). The woody husks surrounding arillate seeds of many canopy
tree fruits may offer some measure of protection from such seed predators.
The Cracidae also subsist largely on fruits and seeds. Curassows (Crax, Nothocrax) are
primarily terrestrial foragers that feed on fallen or low growing, unripe fruits (Delacour and
Amadon 1973). Crop and gizzard are well-developed, so many seeds are probably digested
in addition to the fruit pulp. Chachalacas (Ortalis) and guans (Penelope) are primarily arboreal
868 ORNITHOLOGICALMONOGRAPHSNO. 36
and forage on ripe fruits (Delacour and Amadon 1973). Long legs and neck facilitate reaching
of fruits on small diameter, pendent twigs. The gizzard is thin- walled and the crop is poorly
developed, so most seeds probably pass through the gut intact.
Finches in the subfamilies Emberizinae and Cardinalinae feed primarily on seeds, but some
cardinalines, especially Saltator spp. and the grosbeaks (e.g., Caryothraustes, Pytilus, and
Passerina (Cyanocompsa) cyanoides), eat a substantial amount of fruit as well. All include
insects in their diets, and saltators are also known to feed on young leaves (Jenkins 1969;
pers. observ.). The size and hardness of a seed probably influence how it is handled. Seeds of
Heliconia spp. (Musaceae) are too hard for even the heavy-billed Passerina cyanoides to break
(Stiles 1979, pers. observ.), but both P. cyanoides and Caryothraustes poliogaster will crack
the smaller seeds of Psychotria brachiata (Rubiaceae). Saltators appear to digest very few
seeds and supplement their diets primarily with insects (Jenkins 1969). The fruit-eaters in
this group crush the fruit pulp before swallowing (Moermond 1983). Large seeds are frequently
dropped in place if not eaten and, therefore, poorly dispersed, whereas small seeds may be
swallowed intact (Moermond and Denslow, unpubl. data; D. J. Levey, pers. comm.).
Insect-eating frugivores.- Most fruit-eating birds eat some insects and provide them to
young nestlings. Like frugivores, primarily insectivorous birds have thin-walled gizzards and
small crops (Cvitanic 1970). Passage rates through the digestive tracts of the two groups are
comparable (Ziswiler and Farner 1972; Herrera 1984). Incorporation of fruits into the diet of
an insectivore is apparently done as readily as is the reverse. Frugivores frequently take
advantage of locally superabundant insect populations (Eisenmann 1961), and some migrants,
e.g., Dendroica castanea, D. pensylvanica, and Vermivora peregrina are insectivorous on
their temperate breeding grounds but often highly frugivorous in the tropics (articles in Keast
and Morton 1980). In fact north temperate migrants may be important dispersers of seeds of
some neotropical trees (Leek 1972a; Howe and DeSteven 1979; Howe and Vande Kerckhove
1980; Greenberg 1981). Except during seasons of low fruit abundance when partially ripe
fruits are taken (Foster 1977), insect-eating fruvigores prefer fully ripe fruits to partially ripe
fruits (Moermond and Denslow 1983). Most take both watery, sweet fruits and dry or oily
fruits. Fruit-eating birds in the following families supplement their fruit diets with insects
and, occasionally, small vertebrates: Cotingidae, Pipridae, Tyrannidae, Muscicapidae (Tur-
dinae), and Emberizidae (including Parulinae, Icterinae, Thraupinae, and Cardinalinae).
NUTRITION AND DIGESTION
Fruit as a Food Resource
Pulp quality varies widely with respect to amount, carloric and nutrient contents, digestibility
of the pulp (fiber, water content), seed size and total bulk of seeds, toughness of fruit coat,
and presence of secondary compounds (e.g., phenols and alkaloids, Herrera 1982). Seeds
ingested with the pulp may add considerably to the non-digestible bulk of the fruit (e.g.,
commonly more than 50 percent of the total fruit weight for few-seeded fruits but much less
for fruits with many small seeds, White 1974; Snow 1981; Wheelwright et al. 1984). In
contrast to both seeds and insects, fruits are generally low in protein and in lipid and may be
high in carbohydrate (Table 1). A survey of families producing bird-dispersed seeds (Table
1) also reveals that in general fruits are either rich in lipid or in carbohydrate. Only in a few
instances does a fruit contain more than 20 percent each of lipid and carbohydrate. The
median protein content for 29 tropical families of bird-dispersed fruits is 8.4 percent of the
dry weight of the fruit pulp (Table 1) compared to averages of approximately 12 percent for
12 species of cereal seeds (Jenkins 1969) and 66 percent for 19 species of insects (White 1974).
Among these fruits, there is no correlation between lipid and protein content of fruit pulp
(Spearman R = 0.182, n.s.).
The water content of fruits is generally very high, contributing to the impression that fruits
are an extremely low quality food source. Experimental diets fed to granivores are low in
moisture (about 1 1%) in comparison to fruits (50% to 94%, Table 1). On a dry weight basis,
the protein contents of many fruits are only slightly less than those of seeds, and nutrients
are easily assimilated from an aqueous solution. Fruit pulp is also generally low in fiber,
requiring little mechanical reduction to render the cell contents accessible to digestive pro-
cesses. Fruits thus are characterized by a bulky, dilute pulp coupled with a substantial portion
of undigestible seed mass that must also be handled. It is apparent that efficient handling of
a bulky food is one of the major constraints on a bird dependent on fruits for a large portion
of its energetic and nutritional needs. One solution for the birds is a rapid gut passage time
TABLE 1
Energetic and Nutrient Contents of Fruits in Twenty-nine Plant Families that Use
Birds as Seed-dispersers1
No.
species
sampled Water% Protein2 Lipid Carbohydrate kcal/g3 P/C4 Source5
Annonaceae 1 86 4.8 3.3 21.0 1.40 .034 8

Araliaceae 1 68 12.0 33.0 55.0 5.83 .020 7
Apocynaceae 2 73 11.0 17.3 4.5 2.35 .047 6
Boraginaceae 2 80 16.8 5.2 68.1 3.96 .042 3,7
Burseraceae 4 85 11.0 3.9 65.0 3.53 .031 1,7
Connaraceae 1 79 8.4 15.9 70.3 4.72 .018 7
Flacourtiaceae 1 90 5.7 .9 37.8 1.89 .030 8
Lauraceae 24 68 6.2 25.3 9.0 3.03 .020 1,2,8
Leguminosae 1 56 1.9 22.7 73.2 5.14 .004 7
Loranthaceae 1 55 5.8 52.7 38.1 6.73 .009 7
Malvaceae 1 56 3.1 0.3 35.2 1.60 .019 8
Melastomataceae 2 66 4.2 6.8 84.0 4.21 .010 7
Meliaceae 1 54 15.1 59.7 22.3 7.23 .020 9
Moraceae 4 78 6.4 4.3 79.0 3.89 .016 3,7
Musaceae 1 74 10.6 39.2 40.4 5.81 .018 7
Myrsinaceae 2 87 6.0 2.7 53.8 2.71 .022 5,8
Myristicaceae 1 55 2.5 63.0 9.0 6.35 .004 4
Myrtaceae 2 81 4.5 1.6 53.8 2.53 .018 7,8
Ochnaceae 1 57 4.9 11.0 81.0 4.52 .011 7
Olacaceae 1 68 17.5 17.6 7
Palmae 4 85 9.5 20.8 66.6 4.88 .027 1,7
Phytolaccaceae 1 80 1.9 2.4 7
Rosaceae 3 80 6.5 1.5 38.0 2.00 .032 3,8
Rubiaceae 2 87 13.2 5.2 74.6 4.14 .032 7
Rutaceae 1 77 8.8 1.2 8
Sapindaceae 1 83 9.9 3.2 82.5 4.11 .024 7
Solanaceae 1 89 9.2 1.3 15.6 1.22 .075 8
Verbenaceae 3 80 2.2 2.2 38.4 1.86 .012 7,8
Zingiberaceae 1 92 17.9 1.8 60.9 3.54 .051 7
Medians 79 6.5 5.2 53.8 3.92 .020
(n = 29) (n = 29) (n = 29) (n = 26) (n = 26) (n = 26)
Medianvaluesare providedwhendata for more than one speciesper familyare available.Exceptfor water,all percentsare basedon
dry weights.
2Proteinwas calculatedas 4.4 x percentnitrogen(Miltonand Dintzis 1981).
3The followingconversionfactorswereusedto estimatecoloriccontent:protein5.2 kcal/g;lipid 9.3 kcal/g;carbohydrate
4.0 kcal/g(Watt
and Merrill1963).
4 Protein: calorieratiois calculatedas %protein/kcalkg1 food.
5Sources:1 = Snow 1962c;2 = Snow 1981; 3 = Munzellet al. 1949;4 = Howe and VandeKerckhove1980; 5 = M. Foster 1977;6 =
McDiarmidet al. 1977;7 - White 1974;8 - Wheelwrightet al. 1984;9 - Fosterand McDiarmid1983.
(Groebbels 1932; Walsberg 1975; Herrera 1984). Another solution is for the birds to select
fruits with low proportions of undigestible matter. Studies by Howe and Vande Kerckhove
(1980) and Herrera (1981) have shown that birds prefer fruits with a large pulp to seed ratio
(at least among large-seeded fruits), and Sorensen (1984) correlated fruit preference in Turdus
merula with fruits whose seeds are regurgitated rather than defecated. Regurgitated seeds have
significantly less residence time in the gut.
Although yearly extremes of fruit abundance may commonly differ only two-fold (Fogden
1972) compared to many-fold in temperate habitats, seasons of fruit abundance may never-
theless be well-defined even in aseasonal wet tropical climates (Snow 1965; Janzen 1967;
Smythe 1970; Frankie et al. 1974; Hilty 1980; Foster 1982a). Fruiting seasons may show
marked year-to-year variation (Ward 1969a; Fogden 1972; Foster 1982a, b). Trees may fruit
on a supraannual periodicity (Frankie et al. 1974), crops of some species may fail completely
in some years (Crome 1975a; Foster 1982b; Faaborg and Terborgh 1980), and some species
may not fruit consistently at the same time each year (Foster 1982a). Crop sizes, fruit char-
acteristics, spatial distributions, and phenologies of fruit production differ between canopy
and understory plants. Understory trees and shrubs are more likely to ripen few, small fruits
over an extended season than are canopy trees (Frankie et al. 1974; Hilty 1980; Opler et al.
1980).
Thus, even though fruits appear to be common year round in rain forest, suitability of fruits
for a particular frugivore may differ, and favored fruits may be patchily distributed in space
(Snow 1962c; Fogden 1972) and poorly predictable in time. The tracking and exploitation of
a locally variable but sometimes abundant food source (for example by flocking, Leek 1971a;
Buskirk 1976) is then another important attribute of a frugivorous habit.
Nutritional Constraints on Frugivores
Unfortunately, studies on the nutritional requirements of frugivorous birds are scarce. We
are often limited to extrapolations from studies on granivores- chickens, domestic pigeons,
and sparrows- the results of which may be of limited value in understanding digestive phys-
iology of frugivores. Nevertheless, some general patterns are suggested.
Studies of temperate bird species suggest that ingestion is limited primarily by energetic
requirements (Fisher 1972) and (to some extent) volumetric constraints (chickens, Hill and
Dansky 1954; Tree Sparrows, Spizella arborea, Martin 1968). For primarily carnivorous,
insectivorous, or granivorous birds, other nutritional requirements such as protein and min-
erals are likely to be largely satisfied at the same time that energetic requirements are met if
there is variety in the diet.
Studies of adult birds suggest that a diet of 4 percent to 8 percent protein (fresh weight) is
necessary for maintenance (chicken = 7%, Leveille and Fisher 1958; Spizella = 8%, Martin
1968; and papers reviewed by Berthold 1976a). Martin (1968) found that although a diet of
4 percent protein was minimal for survival of Spizella arborea, assimilation efficiencies of
nitrogen and carbohydrate are significantly greater in diets containing at least 8 percent
protein. Initial indications are, however, that protein requirements (or ability to extract suf-
ficient protein from fruits) differ importantly among species. Berthold (1976a) showed clearly
that several European passerines (Sylvia atricapilla, S. borin, Turdus merula, and Erithacus
rubecula) were unable to maintain weight on a diet consisting solely of fruits. The birds
consumed adequate amounts of calories but not of protein. In contrast, the waxwing (Bom-
bycilia garrulus) maintained weight for 10 to 18 days on a diet consisting solely of berries
(Berthold 1976b). In another study, the heavily frugivorous Phainopepla nitens maintained
weight on a diet of mistletoe berries, but the house finch (Carpodacus mexicanus) did not
(Walsberg 1975). These studies suggest that there are important constraints on the digestive
physiology of birds dependent on fruits for the majority of their calorie and nutrient require-
ments. Although many species may exploit fruits on an occasional basis, dependence on fruits
for longer periods of time may not be so easily accomplished.
For adults of many frugivores, protein and calorie levels of many fruits are adequate
providing that sufficient quantity can be processed daily. We have kept tropical birds of families
Emberizidae (tanagers and finches), Pipridae, Trogonidae, and Ramphastidae, and temperate
birds in the Turdinae, Bombycillidae, and Mimidae on a synthetic diet for several months
during which time their weights stayed close to or above capture weight. The content of this
diet was richer than that of many tropical fruits: 86 percent water; 13 percent protein; 6 percent
lipid; 78 percent carbohydrate by dry weight. Consistent with Walsberg's study, the three
exceptions that required an additional protein supplement to maintain weight were all seed-
eating finches- Arremon aurantiirostris, Caryothraustes poliogaster, and Passerina (Cyano-
compsa) cyanoides (Moermond and Denslow, unpubl. data; D. J. Levey, pers. comm.). The
other species did well on the diet and, although we do not know whether they could have
reproduced, they were able to molt. Caged temperate frugivores maintained solely on our
synthetic diet ingested between 1.8 and 2.2 g (fresh weight) of food per gram of body weight
per day in comparison to 0.2 g food per gram of body weight per day in the case of the
granivorous Spizella (Martin 1968). A 30 g frugivore must consume approximately 60 g of
our artificial diet daily -the volume equivalent to 17 1 Ardisia revoluta fruits (6.0 mm diameter)
or 20 Cinnamomum laubautii fruits (23.3 x 13.2 mm), including their seeds (data from Foster
1978). In caloric equivalents the numbers of fruits are 287 and 12.2 fruits respectively. A
similar pattern has been found for a frugivorous neotropical bat (Morrison 1980).
It is apparent from these estimates that processing time (as opposed to searching or handling
time) is likely to be an important component of the foraging equation for frugivorous birds.
For birds handling large-seeded fruits with firmly attached pulp (e.g., Lauraceae and Palmae),
separation of the pulp from the seed may be an important limitation to processing rate.
NEOTROPICALORNITHOLOGY 87 1
Resplendent Quetzals (Pharomachrus mocinno) may take an hour to regurgitate the seed from
a single large Beilschmiedia (Lauraceae) fruit (N. T. Wheelwright, pers. comm.), and Collared
Aracaris (Pteroglossus torquatus) may regurgitate a single Ocotea (Lauraceae) fruit and then
reswallow it three times over a period of almost an hour before the pulp is removed from the
seed (E. Santana C, pers. comm.). For birds feeding on small-seeded, sugary fruits, physio-
logical specializations that permit rapid absorption of nutrients from dilute aqueous solution
are likely to be more important. Studies on starlings (Sturnus vulgaris) suggest that unusually
high rates of kidney function may be one component of the physiological adaptations of
frugivores to their diets (B. Wentworth, pers. comm.).
Digestive adaptations for efficiently processing watery, sweet fruits may constrain the facility
with which some frugivores are able to incorporate dry or oily fruits into their diets and at
least partially explain the general preference of frugivores for ripe over partially ripe fruits
(Foster 1977; Moermond and Denslow 1983). Preliminary data suggest that the watery texture
of fully ripe fruits allows them to be more efficiently assimilated although unripe fruits may
also contain higher levels of digestion inhibitors such as tannins. Caged Collared Aracaris
thoroughly digested nearly all fully ripe Hamelia patens (Rubiaceae) fruits offered but digested
a significantly lower proportion of partially ripe fruits (Santana C, Moermond, and Denslow,
unpubl. data).
Differences in physiological treatments of proteins and carbohydrates impose further con-
straints on the nutrition of frugivores. Assimilated protein can be converted to fat, stored as
protein in muscles (Fogden and Fogden 1979), or burned as an immediate source of energy.
Nitrogen in excess of that required for body maintenance, feather production, or reproduction
can be excreted. Assimilated lipid and carbohydrate may be burned for energy or stored as
fat or glycogen but cannot be excreted. Unless frugivores are unusual in possessing some
mechanism by which protein can be differentially assimilated relative to carbohydrate, protein
and carbohydrate are probably assimilated with similar efficiencies (about 75%, White 1974).
For these reasons, the protein : calorie ratio (here reported as percent protein : Kcal/100 g
food, White 1974) is an important indication of the quality of the diet. The few bird species
for which dietary requirements have been determined appear to do best on dietary protein :
calorie ratios of approximately .02 to .05 for adults (.039 for laying quail, Nestler et al. 1944;
.038 to .052 for laying and breeding chickens, National Research Council 1960; .028 for
Spizella, Martin 1968). Protein : calorie ratios have been calculated for several tropical fruits
(Table 1). Some are considerably lower than the predicted dietary requirements, e.g., lipid-
rich Myristicaceae .004; at best the protein : calorie requirements of adult birds can be met
only minimally by a random selection of tropical fruits if frugivores have similar protein
assimilation and utilization efficiencies as granivores.
The problem of lack of protein is compounded for nestlings and for reproductive and molting
adults, which require protein for both maintenance and for new tissue production. During
times in which protein demands are high, calories are also likely to be in excess of metabolic
needs (Foster 1978). Thus, birds feeding predominantly on fruits must possess some mech-
anism to deal with excess calories in their diets. For example, young Oilbirds deposit large
quantities of fat before leaving the nest even though dietary protein is so low as to prolong
the fledgling period much beyond that usual for birds of that size (Snow 1961, 1962c; White
1974). Other neotropical frugivores (e.g., Cotingidae, Pipridae, Thraupinae) do not appear to
store large quantities of oil and duration of the nestling period appears to be no longer (on
average) than for non-frugivores (Ricklefs 1974).
Walsberg (1975) suggested that frugivores may be less efficient at assimilating calories than
granivores, or, alternatively, that the daily energy budgets of birds feeding mainly on fruits
may be higher than those for birds supplementing their diets with insects. This appears to be
the pattern in bats (D. W. Thomas, pers. comm.), and White (1974) presented evidence that
the efficiency of conversion of protein to tissue is about 63 percent in Oilbird nestlings in
contrast to 6.8 percent for the insectivore nestlings he studied. Protein is used to meet only
14.2 percent of the metabolic needs of Oilbird nestlings in comparison to 75.5 percent of the
needs of insectivorous nestlings. For most nestlings and many adult frugivorous birds, some
protein is supplied from insects (Skutch 1954, 1960, 1969; Snow 1962a, b; Snow and Snow
1971; Morton 1980), and fruits are then used primarily as a source of calories. Thus, most
birds that are highly or totally frugivorous as adults feed insects to very young nestlings; the
proportion of fruits in the diet increases as the nestlings mature, and the ratio of protein to
calories metabolized approaches that of adult birds (Morton 1973; White 1974; Skutch 1976;
Wheelwright 1983). White (1974) noted that the ratios of protein to calories metabolized by
nestlings of five tropical species were all initially higher than that obtainable from fruits alone,
but that this ratio was lower for birds feeding both fruits and insects to nestlings (Elaenia
flavogaster, Turdusgrayi, Zarhynchuswagleri)than for a total insectivore (Iridoprocnealbi-
linea). Oilbirds only bring fruit to nestlings, but the nestlings receive food from the parents
with a higher protein : calorie ratio than fruits provide. White (1974) suggested that adults
provision the young nestlings with a high protein regurgitated liquid (Snow 1961) analagous
to that produced by pigeons.
Nevertheless, for many species, fruits are, at times, an important energy source. In this
sense even those species that feed primarily on insects may be obligate frugivores, because
fruits provide critical calories, perhaps freeing insect proteins for egg or feather production.
The demand for fruits is particularly noticeable during seasons in which few ripe fruits are
available. During these times, birds may take green (unripe) fruits (Leek 1972b), greatly
increasing the foraging time necessary to acquire sufficient food (Foster 1974, 1977). In a
Costa Rican rain forest, we have recorded 15 species of birds eating green or partially ripe
fruits during one season of particularly low fruit availability (Moermond and Denslow, unpubl.
data).
Reproduction in an adult bird often depends on the level of protein in the diet (Ward 1969a;
Fogden 1972; Crome 1975a, b; Fogden and Fogden 1979). Among tropical frugivores, breeding
and molting seasons may be closely tied to the quality and abundance of the resource base
both for protein and for energetic requirements. In Trinidad the breeding seasons of Oilbirds
(Snow 1961) and Bearded Bellbirds (Snow 1970) coincide with the fruiting seasons of Lau-
raceae, Burseraceae, and Palmae (all of which produce particularly nutritious fruits). In Sin-
gapore and Sarawak, Yellow- vented Bulbuls (Pycnonotus goiavief) breed during seasons of
fruit and insect abundance (Ward 1969b; Fogden 1972).
Fogden (1972) suggested that frugivores may not be sufficiently good at hunting insects to
meet both daily metabolic protein requirements and to store sufficient protein for reproduction
or feather production on a totally insect diet. Reproduction and molting are, therefore, closely
dependent on energy obtained from fruits. Under such circumstances, all protein acquired
from insects can then be used to meet protein rather than energetic demands. This thesis is
supported in the breeding of Phainopepla in the deserts of the southwestern United States.
Nesting coincides with abundance of both insects and mistletoe fruits. If either resource fails,
nesting is abandoned or not initiated (Walsberg 1977). The breeding seasons of the Black-
and-white and the Golden-headed Manakins (M. manacus and Pipra erythrocephala) in
Trinidad coincide with fruiting peaks of Melastomataceae and Rubiaceae, neither of which
have notable concentrations of either proteins or lipids (Snow 1962a, b, 1965).
Morphology of the Digestive Tract
We expect that the digestive tract of primarily frugivorous birds should reflect a diet of
bulky, watery, non-fibrous foods that are generally low in protein, high in soluble carbohy-
drate, and variable in lipid. As noted above, it would be advantageous for frugivores to
process fruit pulp quickly, absorbing only the easily available nutrients from large volumes
of fruits. Thus, we expect to find little provision for food storage, short intestines (Pulliainen
et al. 1981), non-muscular gizzards (because fruits require little mechanical reduction), and
large livers relative to body size (detoxification of the secondary compounds frequently present
in fruits (Herrera 1982) occurs primarily in the liver in vertebrates (Brattsten 1979)).
Comparative data on the morphology of the digestive apparatus of predominantly frugivo-
rous birds are not extensive. Moreover, they are complicated by seasonal variation in diets
(especially among temperate frugivores), differences in fruit and seed handling behavior, and
the nature of alternative sources of protein (e.g., seeds, insects, vertebrates). As Herrera (1984)
noted, functional and behavior modifications may be more important components of the
evolution of frugivory than less-flexible morphological adaptations. Although some general-
izations are apparent, there is considerable variation in the morphology of the digestive tract
among fruit-eaters, and some of the generalizations suggested above are not supported. We
must await considerably larger sample sizes that take into account diverse feeding ecologies
of frugivores.
In most species there appears to be little provision for short-term storage of fruits (but see
Ziswiler and Farner 1972). The crop (if present), esophagus, and/or proventriculus are small
(Wetmore 1914; Cadow 1933; Jenkins 1969 and papers described therein; Walsberg 1975).
However, the proventriculus is enlarged in Euphonia, and the esophagus is extensible in
swallow-tanagers (Schaefer 1953). Wheelwright (1983) also reported the presence of an ex-
pandable proventriculus in Resplendent Quetzals that presumably accommodates large laur-
aceous fruits. These birds have no crop, and the esophagus is not used to store food.
The soft, non-fibrous nature of fruit pulp is reflected by the non-muscular gizzard in many
frugivores (Cadow 1933; Jenkins 1969 and papers described therein; Walsberg 1975). The
quetzal has rings of muscles around the esophagus, presumably to facilitate regurgitation of
large seeds, and also a muscular gizzard (Wheelwright 1983). Herrera (1984) found no dif-
ference in gizzard size among seed dispersers (all Muscicapidae) and non-frugivores in a
Mediterranean woodland. In addition, Herrera (1984) reported no morphological differences
between seed dispersers and non-frugivores with respect to liver size or intestine length in
relation to body size.
Other studies also have failed to find a consistent pattern of intestine length in relation to
body weight of frugivores. Although some authors have suggested that the intestine length in
frugivores is short (e.g., Desselberger 1931; Docters van Leeuwen 1954; Ziswiler and Farner
1972; Walsberg 1975), waxwings apparently have long intestines (Cvitanic 1970; but see
Walsberg 1977), and Jenkins (1969) reported no consistent pattern among species of Costa
Rican frugivores he examined. Intestinal caecae are not common in frugivores, presumably
reflecting the non-fibrous nature of their food; however, Wheelwright (1983) described paired
caecal sacs packed with fruit skins in the Resplendent Quetzal.
It is apparent that some of the most important adaptations to a fruit-eating habit involve
a digestive physiology that can accommodate the processing of a bulky, dilute food that is
relatively high in calories in comparison to protein. To some degree these may be reflected
in the morphology of the digestive tract, but this structure is also likely to retain elements
that maintain the capability of processing other foods and that reflect the phylogenetic affinities
of the frugivores, obscuring patterns associated primarily with frugivory. Moreover, fruits do
not represent a homogeneous food resource. Important differences among them in source of
calories (primarily lipid or carbohydrate), seed number and size, and the firmness with which
the pulp is attached to the seed are likely to be reflected in the morphology of the digestive
tract and the digestive physiology of birds feeding on fruits.
Diet Breadth and Fruit Selection
With a few notable exceptions tropical frugivores exploit many species of fruits (Snow 198 1;
Stiles, in press; Wheelwright et al. 1984). Three species of Costa Rican saltators together
included 189 fruit species in their diets (Jenkins 1969). Worthington (1982) recorded 38 species
in the diet of Manacus vitellinns on Orchid Island, Panama, Snow (1962a, b) observed 66
and 43 species in the diets of two species of manakins in Trinidad, and N. T. Wheelwright
(pers. comm.) recorded more than 96 species of fruits in the diets of Emerald Toucanets
(Aulacorhynchus prasinus). Even Oilbirds, which take few if any insects, use fruits from 37
species of plants during their nesting season (Snow 1962c).
Such data, together with observations of many diverse bird species exploiting abundant
crops of some fruiting trees, have been the basis for the hypothesis that some frugivores,
especially the small birds that feed on small, berry-like fruits, are not highly selective but
choose fruits opportunistically as they are encountered (Howe and Estabrook 1977; Fleming
1979). The hypothesis is supported by studies such as those of Sorensen (1981), who was
unable to find a basis for selection of fruits by British thrushes and titmice, and Worthington
(1982, pers. comm.) who concluded that manakins likely take fruits randomly as they are
encountered.
The notion that some of these species are specialists on a few fruits while others are
indiscriminate generalists on many fruits derives in part from the great abundance and diversity
of small, sweet, watery, low quality fruits in comparison to the relatively fewer large, lipid-
rich fruits. The apparent richness of the large fruits and their sheer size, which restricts the
number of bird species that are able to eat them, offer a plausible basis for assuming that such
fruits could provide a staple diet to the few bird species able to eat them and hence allow diet
specialization (Snow 1971a; McKey 1975).
However, data on diets of a range of small and large frugivores reveal no particular evidence
for marked differences in fruit specialization at the level of the plant family (Table 2). Fruits
from only three or four families make up large percentages of the diets of all birds studied.
H
9B30BUIIO X X X X X X X *
3B30B3qi X 2
3B30BUBIOS X X I
N
3B30Bpuides X S
§ 3B90Biqn^ XX X *
O 3BUIIBJ X II
2 ^
5UJ 3B90BJOÎ X XXXXXXXX * £
g
£ 9B30Bipp^ XX I
fc 9B30BJBUIOlSBpj^ XXX XXXXXXXXX X
I
U 3B90BAlBp^ X I
O 8
£ dBdOBqiuBJoq X e
£ 3B3Dwnei XXXX XXXX i
S 3BJ9ji;mo X |
5u, 3B30BiynooBu X XXX*-
I
0 $
3B3DBiqjoqdn3 X X X X X X ^
|
" XX £
y 3B9DB9Uinia
n] I f
PQ H 3B30BUBiBqosAjq3 X •«
h
^
s iII
23 3B30BllOJUdB3 X
5 T.
S 3B30BJDB0 X «
^ a
H 3B9DBJSSJna XX X X
|
1 SBQOBipUIOJg XXX I
Z 9B30BIIBJV X
XXXXXX^
I 3B30BJV XXX X I
q r
Q 3B30Biiqjinbv X |
0MO^^-O0N00N00^vOMa\00'H00boO<N^(N(N0\^ '
^ SSqiUIBJ lUBUIUIOp p-£
w > I "c>.
S p ^; p vo vo ^ <N r- <n "^ r»t o\ «o rs -^ so § 3
i s is§ I 11
1 I 1| j«3 -l| Is, Jiff 1^*1
ilsiiii
NEOTROPICAL ORNITHOLOGY 875
Fig. 1. Fruit sizes in the diets of neotropical frugivorous birds grouped according to body weight.
Data summarized from sources listed in Table 2.
Althoughsome largefrugivoresare closely dependenton a few species of fruitingtrees, most

take a diversity of fruits and concentrateon three or four plant families.
Even dependenceat the family level is not rigid, as can be seen by comparingthe diets of
two congenericpairs of manakins(consideredspecialistsby Snow 1981). Pipra mentalisand
P. erythrocephalaare morphologicallyand ecologicallyvery similar, and Manacus manacus
and M. vitellinusare so similarthat Snow (1979) describedthem as subspeciesof M. manacus.
On the largeisland of Trinidad,berriesof the diverse family Melastomataceaefigureprom-
inantly in the diets of both species (P. erythrocephala-63 percent of fruits eaten; M. man-
acus- Al percent, Snow 1962a, b). On the very small Orchid Island, melastomes are much
less diverse and apparentlyless common (Worthington1982). In any case they are not par-
ticularlyimportantto the two manakinsthere, both of which, however, include a high pro-
portion of fruits of epiphytes in the family Araceae in their diets (P. mentalis- 22 percent,
M. vitellinus- IS percent).
Small birds are restrictedin the diametersof fruits they are able to swallow by the size of
their gapes. Data on the sizes of fruits in the diets of large birds shows that they take both
largeand small fruits, includingthe small fruits taken by small species (Fig. 1; Wheelwright,
in press).Althougha numberof largespeciesdo feed extensivelyon the largefruitsof lauraceous
plants (Snow 1981), many of these take a variety of other fruits (Table 2; Wheelwright1983;
Wheelwrightet al. 1984). Even when we consider cases where the degree of dependence
betweenthe birdand plantis assumedto be very strong,as with mistletoesand theirdispersers
(e.g., Dicaeum spp., Docters Van Leeuwen 1954; Salomonsen 1964; Euphonias,Snow 1981;
Phainopepla,Walsberg1975; Zimmerius(Tyranniscus),G. Stiles, pers. comm.; pers. observ.;
Ampelion (Zaratornis) stresemanni, Parker 1981), we find that the bird species span a wide
range of size and families and that none of the birds feeds exclusively on the mistletoe berries.
With respect to the fruits for which data are available, both tanagers and manakins feed
mainly on small, juicy fruits that are rich in carbohydrate but relatively low in lipid (Tables
1, 2). Larger birds (5 species of cotingas, a trogon, the Oilbird, and a toucanet) include both
lipid rich and carbohydrate rich fruits in their diets (Tables 1, 2). But, with the exception of
the Oilbird, median protein contents of fruits in families in the diets of frugivores were not
different from the median for all fruits analyzed.
Also, no evidence suggests that highly frugivorous birds provide higher quality dispersal or
that plants and their dispersers are tightly co-evolved (Wheelwright and Orians 1982). The
same tree species is likely to be visited by different frugivores in different parts of its range
(Howe and Primak 1975; Howe 1977), and frugivores may exploit different plant families in
different parts of their ranges (Snow 1962a, b; Worthington 1982).
With regard to the selectivity of fruit-eating birds, experimental work with caged birds
indicates strong feeding preferences among a wide variety of fruit-eating birds. Although all
of the bases for these choices are not yet known, some patterns are becoming clear. Costa
Rican tanagers, saltators, and manakins each show hierarchical preferences among fruits of
different species, ripeness, size, and accessibility (Moermond and Denslow 1983). Preference
for fruits with large pulp/seed ratios have been demonstrated for toucans (Howe and Vande
Kerckhove 1980), Sylvia atricapilla (Herrera 1981), and aracaris and trogons (Santana C,
Moermond, and Denslow, unpubl. data).
Responses of the birds to particular characteristics of the fruits themselves are not too
surprising, but experimental aviary and field studies have also shown that the accessibility of
fruits strongly influences fruit choice (Denslow and Moermond 1982, in press; Moermond
and Denslow 1983). Numerous observations of responses of birds in the field to fruits differing
in accessibility (e.g., D. W. Snow 1971b; Snow and Snow 1971; Howe 1977; B. K. Snow 1977;
Simms 1978; Kantak 1979; Sorensen 1981; Denslow and Moermond 1982, in press) under-
score the likely importance of this aspect of fruit selection. This may be particularly critical
with regard to adaptations for frugivory, since it is the single characteristic of fruits most likely
to influence the morphology of the bird.
BEHAVIOR AND MORPHOLOGY

Methods of Handling Fruit
Fruits may be swallowed whole, eaten piecemeal, or mashed. Many birds actively handled
fruits in their bills, rotating or moving the fruit with rapid movements of the mandibles. We
refer to this action as mandibulation to distinguish this action on the item from the effect on
the item (cf. Kear 1962; Snow and Snow 1971; Moermond 1983). Cases in which a fruit is
mandibulated to such an extent that it is partially flattened or crushed we have termed
"mashing" (Moermond 1983). When fruits are swallowed whole, the amount of mandibulation
is usually minimal. Eating fruits in piecemeal fashion refers to cases in which birds bite or
tear our chunks of pulp. These differences in fruit handling are potentially of great importance
to the plants and are likely to have differing consequences to the birds' feeding and digestion
rates.
The size of fruit swallowed by a bird is limited by the breadth of its gape, where fruit is
defined as the unit of seed(s) and surrounding pulp that is ingested (Snow 1973; Wheelwright,
in press). Small birds such as Manacus spp. can only handle fruits smaller than about 15 mm
diameter; seeds of large fruits, such as those of many Lauraceae and Palmae, can be dispersed
only by birds that have large gapes such as the larger cotingas and a number of the non-
passeriform frugivores. Many fruits, however, consist of removable units containing one or
a few seeds with the surrounding pulp within a woody capsule (e.g., Guarea spp., Meliaceae;
Glusia spp., Guttiferae). As with berries and drupes, the frugivore is limited by the diameter
of the unit to be swallowed. Seeds may be either defecated if small or regurgitated if large
(Snow 1971a; Worthington 1982; Moermond and Denslow, unpubl. data). In both cases the
seed is carried within the bird from several minutes to an hour or more before being dropped
(pers. observ.; D. J. Levey, pers. comm.).
Mashing of fruits appears to be nearly completely restricted to the tanagers and emberizid
finches (Moermond 1983). Other birds, such as manakins, certainly mandibulate their fruits
to some extent, but such actions appear to serve primarily for positioning the fruit in prep-
aration for swallowing and only on rare occasion is the fruit actually crushed (Moermond and
Denslow, unpubl. data). All 20 species of tanagers we have observed crush the fruit pulp,
often thoroughly, squeezing out and swallowing much of the fruit pulp and juice. In the case
of very small fruits (<3 mm diameter), the mashing may be very rapid (< 1 sec), and Snow
and Snow (1971) reported that some species of Tangara swallow small fruits whole without
noticeable mandibulation. Moermond (1983) has suggested that the "suction-drinking" ca-
pability of tanagers may be directly related to efficient swallowing of juice crushed out of
watery fruits. In addition, large seeds, as well as the tough outer skin of some fruits, are
frequently separated from the pulp and dropped during the mashing process. This reduces the
bulk of indigestible material to be processed but also results in little dispersal of the seeds of
fruits with even moderately large seeds (>3.5 mm diam., Moermond and Denslow, unpubl.
data). Whether or not a seed is dropped depends not only on the size of the seed and the size
of the bill but also on the bill shape and the handling technique of the bird. This was recently
demonstrated in systematic experiments with captive birds of five tanager species and four
other frugivorous species in Costa Rica (D. J. Levey, unpubl. data).
Small seeds, such as those found in Phytolacca (Phytolaccaceae), many Psychotria, and
Melastomataceae, are usually swallowed whole by tanagers along with the crushed pulp and
the juices. In some epiphytic plant species with small fruits, small seeds, and a tough fruit
coat such as Rhipsalis (Cactaceae) and many species of Anthurium (Araceae), the fruit coat
is nearly always dropped by tanagers, but the seeds are forced out of the fruit with the first
drops of pulp and juice as the fruits are mashed and all are swallowed together before the
seed coat is dropped (Moermond and Denslow, unpubl. data). This is likely an adaptation of
the plant to insure ingestion of the seeds by fruit mashers such as Euphonia tanagers that are
often the principal consumers of these fruits.
Eating fruits piecemeal is not the speciality of any particular species of frugivore but rather
appears to be a facultative response to fruits that are large relative to the bird's gape. Birds
that have strong bills such as toucans, barbets, tanagers, and finches are the most likely to eat
piecemeal fruits too large to be easily swallowed whole (Snow and Snow 1971; Skutch 1981;
pers. observ.); however, even relatively weak-billed birds such as the manakins occasionally
eat pieces out of large soft fruits such as those of Coussarea (Rubiaceae; pers. observ.; D. J.
Levey, pers. comm.) and Henriettea (Melastomataceae; Snow 1962b) or out of catkins or
compound fruits (Snow 1962a). The ability of birds to eat fruits piecemeal increases the size
range of exploitable fruits beyond those that are swallowed whole.
The fate of seeds from fruits eaten piecemeal depends on how and where they are eaten.
Large cultivated fruits such as papaya (Carica papaya), banana (Musa sapiens), and the
Pejiballe palm (Bactris gasipaes) are regularly eaten in piecemeal fashion. Pejiballe fruits attract
many species of birds and mammals, but among the birds, only a few very large species, such
as oropendolas (Gymnostinops montezuma), are actually able to carry the intact fruit from
the tree (pers. observ.). Other fruits, such as Cecropia (Moraceae), Clusia, Drymonia (Ges-
neriaceae), Piper (Piperaceae), Stemmadenia (Apocynaceae), and Xylopia (Annonaceae), ap-
pear to be adapted to being eaten piecemeal in that the seeds are small and imbedded in the
matrix of pulp which is eaten.
In other cases fruits eaten piecemeal are first carried away from the plant, and the seeds
thus dispersed to some degree, e.g., Tersina viridis eating large fruits (Schaefer 1953), Trogon
massena eating large lauraceous fruits in captivity (Santana C, Moermond, and Denslow,
unpubl. data), Ramphastos sulfuratus and Pteroglossus torquatus eating Swartzia cubensis
fruits, and Euphonia gouldi eating Asterogyne martiana fruits (pers. observ.). Whether or not
the seeds of fruits eaten piecemeal are dispersed effectively bears on the coevolution between
the plants and their dispersers and hence may influence the nature of food resources available
to fruit-eating birds.
Bill Morphology
Given the diverse set of species that regularly take fruits, the absence of common morpho-
logical adaptations for fruit-eating per se is not surprising (see Karr and James 1975; Ricklefs
1977). Even bills, the structures used directly for "capturing" and handling fruits, exhibit no
common pattern, as can be seen in part by examining the diversity of bill shapes among
tanagers (Storer 1969) and frugivorous tyrannids (Traylor and Fitzpatrick 1982). Nevertheless,
if differences in the methods of taking and handling fruits are considered, a few general adaptive
features of bill size and shape can be seen. A statistical comparison of a large set of temperate
Fig. 2. Variation in bill size and shape associated with different methods of removing fruits. Gape/
oilmen depth is used to separate wide flat bills from narrow deep ones. Measurements are from skins in
the University of Wisconsin Zoology Museum representing 28 species in 1 1 families or subfamilies of
neotropical frugivorous birds. Culmen depth was taken at the base of the nares and bill length from the
juncture with the skull.
species feeding on fruits in southernSpain demonstratedthat those species (all muscicapids)

thatwereactualseed dispersers(swallowingfruitwhole) had flatterand widerbills than species
that were seed and pulp predators(Hen-era 1984). We compared bill shape to method of
capturingfruits by plotting the ratio of gape width to culmen depth against relative bill size
(bill length/wingchord) for a diverse set of species (Fig. 2). Species that predominantlyuse
flight to take fruits have relatively short, wide, flat bills which separatethem sharplyfrom
those picking or reachingfruit while perched. A similar patternappearsalso among insect-
eaters (Partridge1976).
Frugivoresfrom several families that have relativelywide, flat bills commonly take fruits
on the wing (see next section). Even within families in which most species predominantly
take fruitsfrom perches,those species that take many fruitson the wing againhave the widest,
flattestbills, e.g., Tersinaviridis(Thraupinae)and Myadestesmelanops(Turdinae).
Although a wide bill is in part associated with aerial feeding per se, a wide bill and gape
are also advantageousfor handlingand swallowinglargefruits. Among Africandrongos(Di-
cruridae),a family of birds that take most of their food items on the wing, the species that
are most heavily frugivoroushave shorter, flatter bills (Karr and James 1975). Similarly,
among the flycatchers,many of which take their food on the wing, some of the most heavily
frugivorousspecies (e.g., Myiozetetesspp., Megarhynchuspitangua, Myiodynastesspp., Le-
gatus leucophaius)have wider and flatterbills than average(Traylorand Fitzpatrick1982).
Other highly frugivorousflycatcherssuch as Mionectes spp., Elaenia spp., and Zimmerius
(Tyranniscus)spp., that feed on relativelysmall fruits, have short, somewhatterete bills, but
neverthelessstill have relatively largegapes (F. G. Stiles, pers. comm.). A parallelpatternis
found among cotingas,most of which are highly frugivorousand take most food on the wing.
Those species such as Procnias and Xipholenathat feed on relatively large fruits have very
wide and short bills, whereasthose species such as Capornisand Pipreolathat eat relatively
small fruits have short but narrowerand less specialized bills (Snow 1982). The Sharpbill,
Oxyruncuscristatus(recentlyplaced in the Cotingidae,Sibley et al. 1984) which has a longer,
narrowerbill than most cotingas, takes fruits by reaching and hanging (F. G. Stiles, pers.
comm.).
The advantageto a frugivoreof a largegape is clear since gape size limits the size of fruit
that can be swallowed (Snow 1973; Wheelwright,in press), but the bill length and strength
may be associated with other aspects of fruit handling. Virtually all the aerial feeders we
considered (Fig. 2) swallow fruits whole; however, the type of fruit handling employed by
specieswith longer,deeperbills appearsto depend on the presenceof other structuralfeatures
of the bills. For example, although wide bills need not also be long, a long bill may be
advantageousfor reaching fruits from perches. Woodpeckers,barbets, and toucans, which
often swallow fruits whole, have long to extremely long, deep bills, which they use to reach
fruits from perches.The toucans, in particular,reach much fartherwith their long bills than
any other frugivoreand are so dexterousthat they are able to select and pick individualsmall
fruits with the tip of the bill (Santana C, Moermond,and Denslow, unpubl. data).
Tanagersand finches,which also pick most of their fruitsfrom perches,mash most of their
fruits. These birds have short to relatively long beaks that are markedlynarrower,deeper,
and frequentlymuch strongerthan those of aerialfruit-eaters.Their bills have well-developed
lateral ridges on the horny palate (Beecher 1951), which may facilitate ingestion of crushed
pulp and juice as they mash fruits. Some species, such as the Ramphocelustanagers,have
expandedramphothecae,which may act to preventfruitjuices from foulingthe feathers(Storer
1969).
Within each of the groups discussed, modificationsin bill shape and form may influence
the method of fruit handling and the types of fruits taken. Although few appropriatedata
exist, two within-familycomparisonswill serve to illustrateconsequencesof relativelyminor
differencesin bill shape. Within the Trogonidae,both Trogon massena and Pharomachrus
mocinnoare highly frugivorous,and both frequentlyeat largelauraceousfruits(Skutch 1972;
Wheelwright1983). Differencesin bill size and strength,however, affect the ways in which
the two species handle fruits. The bill of T. massena is narrower,deeper, relatively longer,
and noticeablyheavierthanthatof Pharomachrus(Fig. 3), even thoughT. massenais markedly
smallerthan the P. mocinno(141 g vs 206 g). In captivity, T. massena has been observed to
eat large lauraceousfruits piecemeal (SantanaC, Moermond, and Denslow, unpubl. data),
whereasin a detailedfieldstudyWheelwrightneverobservedP. mocinnoto eat fruitspiecemeal
(N. T. Wheelwright,pers. comm.). Similarly, among paleotropic fruit-eatingpigeons, the
strong-billedTreronspp. areable to bite chunksout of fruitswhereasthe weakerbilled Ducula
and Ptilinopusspp. swallow fruits whole (Goodwin 1970).
Such examples furtherreinforce the idea that bill shape and size are likely to influence
which fruitscan be taken most efficiently.We contend that bill featuresare more likely to be
associatedwith particularfruit handlingbehavior than with the frugivoroushabit in general.
Methods of Taking Fruits
Snow and Snow (1971) suggestedthat fruit can be pluckedby birds in only a few different
ways. We agree,but suggestthat differencesamong those ways influencewhich fruitsare most
accessibleand the benefit:cost ratio of takingthose fruits.Althoughfew quantitativedata are
available, we shall offer descriptionsof techniquesbirds use to "capture"fruits and, with a
few examples,proposehow the capturetechniquesmay influencefruitselection.Examination
of both fruit capturemethods and their morphologicalcorrelates(see next section) provides
a basis for understandingadaptive divergenceand specializationamong frugivores.
Fruits may be taken on the wing or from a perchedposition (Herreraand Jordano 1981;
Denslow and Moermond,in press).A comparisonof the use frequenciesof these two general
techniquesfor 10 bird species feeding on the same tree (Fig. 4) illustratesa typical pattern:
most species predominantlyuse only one of the two techniques, taking fruits either on the
wing or from a perch. Only two of the species in this example, the Tityraand the Catharus
thrush,commonly used both techniques.Despite the strongimplicationof stereotypyin fruit
capturetechniquesamong these species, nearlyall of them occasionallyused the other tech-
nique. If other fruit species were included in the comparison, the versatility in foraging
Fig. 3. Bill shapes (lateral [left] and dorsal [right] views) of male (A) Pharomachrus mocinno and (B)
Trogon massena. The bills were drawn from specimens in the University of Wisconsin Zoology Museum.
techniques used by several bird species would be more obvious. For example, even though
Pteroglossus torquatus rarely takes fruit on the wing, captive Pteroglossus were induced to
take fruits on the wing repeatedly under exceptional circumstances in which highly desirable
food items were inaccessible by any other capture method (Santana C, Moermond, and
Denslow, unpubl. data). The Querula which clearly took most Miconia multispicata fruits on
the wing, as is typical for this cotinga, was observed to regularly perch "to pluck the fruits of
Didymopanax morototoni and Guarea trichilioides, both of which bear their fruits in large
bunches on strong stalks that afford a foothold" (Snow 1971b: 7). Perissocephalus tricolor is
another larger cotinga that takes most fruits exclusively on the wing, except when feeding on
Didymopanax fruits which it picks while perched (Snow 1972).
These two fruit capture categories can be further divided to reveal additional functional
differences among frugivores. Unfortunately, few studies report fruit capture techniques, and
those that do often use general categories that are inconsistent from study to study. The subtle
functional differentiation of prey capture methods recently applied to some insect-eaters (Da-
vies and Green 1976; Fitzpatrick 1980) has not yet been developed for fruit-eaters. The
following descriptions are based on our own observations of over 80 species of neotropical
birds in the field in numerous localities and on experiments with 23 species of six families in
captivity.
We have observed four distinct flight maneuvers used to pluck fruit while on the wing:
hovering, stalling, swooping, and snatching. A hovering bird pauses in front of the fruit while
flapping its wings so as to maintain zero air speed. This method is used commonly by manakins,
tyrannid flycatchers, and many small tanagers, especially those that weigh less than 20 g, such
as Tangara and Euphonia spp. A stalling bird pauses briefly in front of the fruit by using a
very steep wing attack angle allowing the bird to slow down and stall just in front of or below
the fruit. Both hovering and stalling are functionally equivalent to "hover-gleaning" of Fitz-
patrick (1980), but the flight motions employed and the morphological features associated
Fig. 4. Feedingbehavior of the most common visitors to Miconia multispicata(Melastomataceae).

Histogramsshow percentagesof total fruitremovals(Denslow and Moermond,in press,with additions).
with each technique differ (see next section). Trogon massena was observed and photographed
while stalling, and morphological examination of trogon species of three genera (Trogon,
Pharomachrus, and Harpactes) suggest this technique may be generally employed by the
Trogonidae and possibly Phoenicircus spp. (Cotingidae; Moermond and Santana C, unpubl.
data). Swooping and snatching both involve continuous movement past the fruit as it is taken.
In swooping, the wings are held out and the bird glides up to the fruit, whereas in snatching,
the wings are flapped throughout. In both cases, some forward momentum is conserved in
contrast to hovering or stalling in which the bird actually stops at the point of contact with
the fruit. We have observed swooping used by several cotingids (Querula purpurea, Carpo-
dectes nitidus) and by Tityra spp. (Tityrinae).
We suspect that nearly all species may at times snatch fruits, but the first three methods
are used primarily by those species that use flight to capture most of their food items. The
costs and skill of taking fruits by snatching and swooping are likely to differ from those
associated with hovering and stalling. Hovering and possibly stalling are likely to be more
costly than snatching or swooping (Pennycuick 1975; Rayner 1981), but the total distance
flown when capturing a fruit is sometimes shorter for those using hovering vs snatching.
Hovering and stalling may allow a more precise approach to fruits that are less accessible
from a longer flight. Snatching and swooping may result in more "misses" and possibly more
"collisions" with twigs and infructescences.
Birds that take fruits from perches may do so in three different ways, by picking, reaching,
or hanging. While birds using all of these techniques "pick" or "reach" fruit in an ordinary
sense, we have restricted use of these terms to describe the position assumed by the bird when
taking the fruits. "Picking" refers to all cases where the birds take fruits close to their perch
without extending their bodies or assuming special positions. "Reaching" refers to cases in
which the bird extends its body out or down from the perch. Appropriate qualifiers or adjectives
can be applied as necessary to distinguish the degree of reach (e.g., tarsus above or below level
of perch, wings fluttered or not, reach horizontal or vertically down). Hanging refers to cases
in which the bird's entire body and legs are under the perch with the ventral side up. Picking
and reaching are often used by a wide variety of birds, but we have seen only woodpeckers
(e.g., Melanerpes pucherani and Campephilus guatemalensis) regularly hanging below perches
when eating fruits.
Although virtually all birds are able to pick fruits close to perches, few are able to reach
very far below a perch. Some appear restricted to taking fruits within easy reach of a perch,
while others can reach outward or downward close to the limit of their body length (Denslow
and Moermond, in press). The ability to use reaching or hanging does not appear to be
associated with the ability to take fruits on the wing. Some species that take fruits primarily
on the wing, such as trogons, flycatchers, some manakins (e.g., Manacus), and some cotingids
(e.g., Querula, Cephalopterus), are unable to reach well from a perch (Snow 1982; Denslow,
Moermond, and Santana C, unpubl. data), but other primarily aerial feeders, such as Tityra
spp., and several cotingids (e.g., Procnias, Carpodectes) regularly reach fruits from a perch
(Snow 1977; Santana C. and Milligan 1984; pers. observ.).
Likewise, birds that take most of their fruits from perches also differ considerably in their
ability to reach. Neotropical thrushes (Turdinae) and emberizine ground finches (e.g., Arremon
aurantiirostris and Arremonops conirostris) are poor reachers. These species tend to take all
their fruits by either picking or snatching (e.g., Catharus ustulatus in Fig. 4). Toucans, barbets,
and most tanagers can reach well below perches. The tanagers are particularly interesting in
this respect, showing considerable variation in reaching ability among similar-sized species.
Some tanagers such as Thraupis palmarum (36 g) and Euphonia gouldi (12 g) are so adept at
reaching that they can extend their entire bodies and legs below the perch and recover their
original upright position without the aid of their wings (Denslow and Moermond, unpubl.
data). Others such as Habiafuscicauda and Ramphocelus passerinii are relatively poor reachers
as tested in aviary experiments (Denslow and Moermond, in press, unpubl. data; D. J. Levey,
pers. comm.).
Differences in reaching ability observed in the field were described for tanagers and hon-
eycreepers of Trinidad by Snow and Snow (1971). Their descriptions correspond to our
observations on Costa Rican species and show that marked differences in such abilities may
be found within the same genus. For example, Thraupis palmarum and Tachyphonus delatrii
demonstrated greater reaching abilities than Thraupis episcopus and Tachyphonus rufus, re-
spectively. Although such abilities have been little studied, they seem to be associated with
some readily identifiable morphological traits (see next section) and, as with different flight
techniques, they also likely influence the types of fruits that can be efficiently exploited.
We compared the limits of reach of two small tanagers to that of Manacus candei (Fig. 5).
All three species were able to pick fruits above them at nearly full body lengths, but only the
tanagers were able to reach full body lengths downward. Similar determinations of the distances
fruits can be taken from a perch for Pteroglossus torquatus and Trogon massena demonstrated
a similar pattern. Pteroglossus was able to reach a full body length below its perch, whereas
Fig. 5. Distances of maximum reach in 3 directions from a sturdy perch for 3 small frugivores.
Manacus candei, an aerial feeder, can reach only as far below the perch as the much smaller Euphonia
gouldi, a perching feeder.
the trogon was able to pick fruits slightly below the perch only with difficulty and was unable
to reach (Santana C, Moermond, and Denslow, unpubl. data). Clearly such differences are
likely to restrict the number of fruits that can be taken by the manakin or trogon from a single
perch relative to the number obtainable by the tanagers and toucanet. On the other hand,
both the manakin and the trogon were adept at taking fruits on the wing and in aviary
experiments were shown to select fruits hanging clear from perches (Denslow, Moermond,
and Santana C, unpubl. data). In the field both species commonly take hanging fruits that
other species take only rarely and with difficulty.
Species using different fruit capture techniques can exploit the same species of fruit (Fig.
4); however, detailed data are likely to show that species using different techniques take fruits
from different portions of the tree as has been reported in other studies (Kantak 1979; Herrera
and Jordano 1981; Santana C. and Milligan 1984). All fruits in a tree were not equivalent
resources contrary to such a suggestion by Leek (1 97 lb); fruits on an infructescence differ with
respect to distances to a perch and thus to accessibility and cost of "capture" to different
species of birds. Differences in feeding techniques and ability therefore may affect which fruit
species are incorporated into a bird's diet.
A comparison of trogons, which rarely take fruits from a perch, to toucans which are rarely
observed to take fruits on the wing (Wagner 1944; Skutch 1971, 1972; Bourne 1974; Wheel-
wright 1983; Santana C. and Milligan 1984), illustrates other possible consequences of
differences in feeding technique. Trogons usually take only one fruit per sally whereas toucans
frequently reach and take multiple fruits from one perching position. We suggest that the
likely greater expenditure per fruit by the trogon leads it to use a narrower range of fruit types
compared to toucans. In aviary experiments Trogon massena showed a nearly exclusive
preference for fully ripe Hamelia patens (Rubiaceae) fruit; whereas Pteroglossus torquatus
readily accepted several stages of less than fully ripe Hamelia, although fully ripe fruits were
their first choice (Santana C, Moermond, and Denslow, unpubl. data). As a consequence,
trogons are also expected to include fewer fruit species in their diets than toucans as has been
shown for Pharomachrus mocinno (43 spp.) vs Aulocorhynchus prasinus (96 spp.; Wheelwright
1983).
When additional appropriate species pairs are studied in sufficient detail, these comparisons
may be extended. Our own fruit selection experiments with manakins (aerial fruit feeders)
and tanagers (primarily perching fruit feeders) suggested that manakins were more discrimi-
nating in choices between paired fruits of different fruit species than were tanagers (Moermond
and Denslow 1983), and recently completed experiments by Levey show that manakins accept
fewer stages of ripening berries (i.e., only those fully ripe or nearly fully ripe) than do tanagers
and that the manakins are more precise in their selection among sets of fruits at different
stages of ripeness (D. J. Levey, pers. comm.).
These differences in fruit selection are likely based on differences in the benefit : cost ratio
associated with each technique of fruit capture. Differences in the costs depend not only on
the technique used but also on the ability of each bird species to execute the maneuvers
required to use the technique. Such abilities will be determined in large part by morphology.
Locomotory Morphology
Morphology is assumed to determine the diversity of the food capture techniques employed
by a given bird species, their relative costs, and the efficiencies with which they are used.
Wings.- Basic aerodynamic principles dictate that the size of bird and the size and shape
of its wings strongly influence flight characteristics and options (e.g., Savile 1957; Pennycuick
1969, 1975; Kokshaysky 1973; Greenewalt 1975; Rayner 1981). A few general predictions
can be made with regard to wing length, wing loading, and degree of slotting. For a bird of a
given weight, long wings reduce the costs of flight (Pennycuick 1969, 1975; Hails 1979) and
aid hovering (Norberg 1979), whereas short wings increase stability and maneuverability
(Savile 1957; Kokshaysky 1973; Norberg 1981), increase acceleration due in part as a result
of higher wing beat frequency (Kokshaysky 1973; Norberg 1981), and reduce the inertial forces
on the wing skeleton (Norberg 1979, 1981). Low wing loading reduces the cost of flight
(Pennycuick 1969, 1975; Greenewalt 1975) and, as a consequence, aids hovering and ma-
neuverability (Norberg 1981). Increased slotting of the primaries allows a higher angle of
attack of the wing which may increase lift, delay stalling, and increase acceleration (Savile
1957; Brown 1963; Alexander 1968; Kokshaysky 1973). The suggestion that slotting primarily
serves to reduce the vortex at the wing tip (Savile 1957; Cone 1968) is not supported (Kok-
shaysky 1973). All of these predictions are constrained by body weight of the bird. Among
geometrically similarly shaped birds, a decrease in size is associated with increases in the
margin between power required for flight and the power available (Pennycuick 1969, 1975;
Greenewalt 1975) and improvements in most aspects of aerial performance such as rate of
acceleration, velocity range, and maneuverability (Norberg 1981; DeJong 1983).
Constraints on flight appear far less critical for birds below about 100 g body weight
(Greenewalt 1975; DeJong 1983). Among frugivorous birds heavier than 100 g (Table 3),
birds with wing loadings greater than 0.42 g/cm take fruits primarily by reaching, whereas
those with lower wing loadings take fruits primarily on the wing. Such a correlation does not
hold well for species weighing less than 100 g. Among these forms, some birds that reach well
also appear able to hover easily (e.g., several Tangara spp. [Denslow and Moermond unpubl.
data]), whereas Manacus spp., which take virtually all fruits on the wing, have unusually high
wing loadings (Table 3, Fig. 6).
Fitzpatrick (1978) showed that large, aerial-feeding tyrannid flycatchers have long and often
somewhat slotted wings, while small forest flycatchers have short, rounded wings presumably
giving them greater maneuverability. A similar pattern holds for large sally-gleaning antshrikes
versus perch-gleaning antshrikes (Schulenberg 1983). However, small frugivores, such as tan-
agers (Euphonia and Tangara spp.) and honeycreepers (Cyanerpes, Dacnis, and Chlorophanes
spp.) generally have relatively longer and narrower wings (X aspect ratio of 15 species = 1.76)
Fig. 6. Wing loadingfor manakins(numbers)and trogons(closed symbols)of differentbody weights

(datafrom Hartman1961); 1 = Manacus vitellinus,2 = Pipra mentalis,3 = Chiroxiphialanceolata,4 =
Corapipoleucorrhea.
than small insectivorous tyrannids (19 species, X = 1.62) and formicariids (15 species, X =
1.46) (P < 0.01, median test; data from Hartman 1961). The short, rounded wings of the
insectivores may be important in maneuverability and rapid acceleration in capturing insects.
For the tanagers feeding on fruits, the longer wings may enhance their ability to hover for
fruits and reduce the flight costs of commuting between fruiting plants. A consequence of
longer, narrower wings may be the higher wing loadings of the tanagers compared to similarly
sized tyrannids (.16 g/cm vs .22 g/cm; P < 0.01, median test; data from Hartman 1961).
This first order approach gives an unexpected result when applied to frugivorous manakins
which take nearly all their food on the wing. The four manakin species we considered (Table
3) have wings with a mean aspect ratio (1 .62) similar to that of the small flycatchers discussed
above and distinctly lower than the small tanagers, but manakin wing loadings (X = .22 g/cm)
are higher than that of the flycatchers but equivalent to that of the tanagers. The functional
advantage of this type of wing for the manakins is not clear.
Assessments based on single factors such as wing loading or aspect ratio are unlikely to
explain satisfactorily foraging differences among similar species (e.g., Partridge 1976, Levey
et al. 1984). Wing loading is highly correlated with body weight for some homogeneous groups
of birds such as trogons (8 species in 2 genera), but no clear relationship emerges from a
similar plot for four species of manakins of four genera (Fig. 6). Similarly, wing length with
respect to body weight varies considerably among tyrannids (Fitzpatrick 1978) and cotingids
(Snow 1982), although, in general, larger birds have longer wings. A suite of wing characteristics
must be considered to understand differences in foraging style among ecologically similar
species (Norberg 1979).
A more detailed comparison of the four species of manakins (Fig. 6) reveals distinct dif-
ferences in aspect ratio and degree of slotting. Manacus vitellinus, with the highest wing loading,
has a relatively low aspect ratio and a high degree of slotting, which may compensate for the
low wing area. Pipra mentalis, with intermediate wing loading, has the highest aspect ratio
and virtually no slotting similar to the wings of some Tangara tanagers with similar wing
loadings. Chiroxiphia lanceolata has an intermediate wing loading, relatively low aspect ratio,
and an intermediate degree of slotting with fewer and shallower slots than Manacus. A similar
pattern can be found within the Cotingidae. The Purple-throated Fruit-crow (Querula pur-
pur-ata), (91-112 g) in the Guianas has much longer wings (165-195 mm chord) than the
similar-sized Guianan Red Cotinga (Phoenicircus carnifex, 75-85 g, 90-1 10 mm wing chord;
Snow 1982). Like Manacus vitellinus, Phoenicircus has highly slotted wings as compared to
the little slotted wings of the fruitcrow (Fig. 7). Regardless of the effect of wing characteristics
on courtship displays of these birds, wing morphology should strongly influence feeding be-
havior. Such parallel patterns in wing loading, slotting, and aspect ratios among divergent
TABLE 3
Morphological1 and Behavioral Characteristics of Selected Species of Neotropical
Fruit-eating Birds
Flight Leg Aspect

muscles muscles Wing ratio
Body weight (% body (% body loading (span/
(g) weight) weight) (g/cm2) chord) FCM3 FHM4
Columbidae
Columbaspeciosa 259 29.1 5.0 0.64 2.0 P Pr
C. nigrirostris 144 28.7 7.0 0.46 1.8 P Pr
Psittacidae
Amazona autumnalis 416 18.7 8.0 0.60 2.0 P Pr
Pionis senilis 213 20.8 7.2 0.46 2.2 P Pr
Steatornithidae
Steatomis caripensis2 415 15.6 - 0.29 3.2 A S
Trogonidae
Pharomachrusmocinno 206 22.2 3.4 0.42 1.9 A S
Trogonmassena 141 21.8 3.2 0.34 2.0 A S
T.curucui 50 22.3 2.7 0.28 2.1 A S
Momotidae
Momotusmomotus 133 20.5 6.2 0.34 1.7 A S
Capitonidae
Eubuccobourcierii 33 13.1 9.4 0.33 1.6 P S
Ramphastidae
Aulacorhynchus prasinus 155 144 11.0 0.53 1.7 P S
Pteroglossustorquatus 226 13 9.9 0.52 1.7 P S
Ramphastosswainsonii 480 14.2 13.0 0.56 1.6 P S
Picidae
Phloeoceastesguatemalensis 240 15.3 11.5 0.41 1.7 P S
Melanerpespucherani 54 17.9 7.1 0.28 2.0 P S
Pipridae
Pipra mentalis 15 19.9 5.4 0.23 1.7 A S
Chiroxiphialanceolata 20 18.4 6.5 0.18 1.6 A S
Corapipoleucorrhoa 12 19.2 5.2 0.18 1.6 A S
Manacus vitellinus 18 17.8 10.2 0.34 1.6 A S
Cotingidae
Cotingaridgwayi 57 25.2 6.4 0.31 1.9 A/P S
Lipaugusunirufus 80 21.1 - 0.29 1.9 A S
Querulapurpurata 103 17.1 5.0 0.15 1.7 A S
Tyrannidae
Tityrasemifasciata 79 18.4 9.2 0.33 1.7 A/P S
Tyrannusmelancholichus 40 24.4 3.4 0.21 1.9 A S
Legatusleucophaius 26 22.1 4.3 0.22 2.0 A S
Myiodynastesmaculatus 46 24.7 4.5 0.23 1.8 A S
Megarhynchuspitangua 71 24.0 3.9 0.31 1.7 A S
Myiozetetesgranadensis 29 22.1 3.6 0.20 1.8 A S
Elaeniaflavogaster 24 19.2 6.0 0.21 1.6 A S
E.frantzii 20 22.6 5.5 0.18 1.8 A S
Tyranniscusvilissimus 9 18.9 4.3 0.16 1.7 A S
Mionectesoleaginea 11 24.5 4.2 0.17 1.5 A S
Mimidae
Dumatellacarolinensis 39 14.0 8.4 0.25 1.7 P S
Mimuspolyglottos 51 16.0 9.7 0.24 1.7 P S
TABLE 3
Continued
Flight Leg Aspect

muscles muscles Wing ratio
Body weight (% body (% body loading (span/
(g) weight) weight) (g/cm2) chord) FCMJ FHM4
Muscicapidae,Turdinae
Turdusplebejus 87 22.0 7.6 0.30 1.9 P S
Myadestesmelanops 32 18.7 4.8 0.24 1.7 P S
Catharusustulatus 33 19.6 5.6 0.24 1.9 A/P S
Bombycillidae
Bombycillacedrorum 33 22.6 4.0 0.29 2.1 P S
Vireonidae
Vireoolivaceus 17 18.6 5.9 0.18 1.9 P S
Emberizidae,Parulinae
Vermivoraperegrina 9 17.1 6.6 0.15 1.9 P S
Dendriocapennsylvanica 10 15.2 7.0 0.14 1.9 P S
Icterinae
Zarhynchuswagleri 163 17.2 9.4 0.34 2.0 P S
Cacicusuropygialis 59 15.6 9.9 0.26 1.7 P S
Icterusgalbula 34 18.0 8.1 0.24 1.8 P S
Thraupinae
Chlorophanesspiza 18 18.8 8.1 0.23 1.7 P M
Cyanerpeslucidus 13 17.5 7.0 0.22 1.8 P M
Dacnis venusta 16 22.1 6.3 0.21 1.8 P M
Chlorophoniacallophrys 25 18.3 7.2 0.30 1.8 P M
Euphonialuteicapilla 12 18.2 5.2 0.25 1.8 P M
E. imitans 13 17.6 6.1 0.20 1.7 P M
Tangaralarvata 19 21.5 6.1 0.21 1.7 P M
T.guttata 20 17.8 9.3 0.26 1.7 P M
Thraupisepiscopus 32 20.9 6.9 0.25 1.8 P M
T.palmarum 39 20.8 - 0.26 1.6 P M
Pirangarubra 30 19.1 5.6 0.22 1.9 P M
P.olivacea 31 20.1 5.7 0.19 2.0 P M
Tachyphonusluctuosus 15 16.1 6.0 0.19 1.6 P M
T. rufus 35 17.4 9.9 0.25 1.6 P M
Ramphocelusdimidiatus 30 15.9 7.4 0.25 1.6 P M
R.passerinii 32 18.3 8.4 0.30 1.6 P M
Habiarubica 36 18.8 8.8 0.25 1.5 P M
H.fuscicauda 39 16.4 8.7 0.22 1.6 P M
Chlorospingusophthalmicus 20 15.7 8.9 0.25 1.6 P M
Cardinalinae
Saltatormaximus 48 18.4 8.2 0.26 1.5 P M
Passerina{Cyanocompsa)cyanoides 32 19.0 6.8 0.31 1.5 P M/Pr
Emberizinae
Arremonopsconirostris 40 14.5 13.0 0.34 1.5 P M/Pr
1 Data from Hartman 1961 except as noted. Values presented are averages for males and females where given separately by Hartman.
2 Data for Steatornis caripensis from Snow 1961.
3 FCM = fruit capture mode; A = in air, P = from perch.
4 FHM = fruit handling mode; S = swallow whole; M = mash; Pr = seed predator.
groups of fruit-eating birds encourage the search for general functional relationships between
wing morphology and feeding behavior.
If frugivores can be fit into a functional-morphological space (see James 1982), then we
offer three examples from Costa Rica that represent three extremes of that space and that will
serve to illustrate the morphological adaptations and behavioral consequences that characterize
Fig. 7. Wing outlines of (A) Querula purpurata, (B) Trogon massena, and (C) Pteroglossus torquatus
drawn to approximately same scale from photographs of birds in flight.
frugivores within that space. The Collared Aracari, the Slaty-tailed Trogon, Trogon massena,
and the Purple-throated Fruit-crow represent three divergent types of wing morphology and
feeding behavior of frugivores (Fig. 7). Because of the marked aerodynamic constraints im-
posed by increasing size (Pennycuick 1969; Kokshaysky 1973; Greenewalt 1975; Norberg
1981), these relatively large birds (> 100 g) may better illustrate the limiting cases, at least as
far as the aerial feeders are concerned.
Querula purpurata obtains most of its fruits by swooping, although snatching and picking
are used (Snow 1971b; pers. observ.; Fig. 4). Trogon massena takes its fruits by stalling (we
have never seen it hover); and Pteroglossus torquatus obtains nearly all its food by picking
and reaching (rarely using snatching) (Santana C., Moermond, and Denslow, unpubl. data;
Fig. 4). The morphological differences in the species correspond to differences in feeding
behavior. Querula has the lowest wing loading of any bird of its size for which data are
available (Table 3; Snow 1982: fig. 1). While all three birds have relatively long wings, Querula
has unusually wide wings mainly due to the broad, long secondaries and shows a relatively
small degree of slotting at the wing tips (Fig. 7). This very broad wing allows these birds to
make long, relatively slow, swooping flights to pluck fruits, one at a time. By comparison,
Trogon massena has a narrower wing (higher aspect ratio) with very prominant, deep square
slots. This slotting pattern allows the trogon to fly very slowly before stalling. Its flight pattern
consists of a rise, followed by an abrupt slowing until it stalls and drops. Fruits are plucked
singly at the top of the rise at the point of stalling. This type of flight is likely facilitated by
extreme slotting of the primaries as well as by well-developed flight muscles and heart (Table
3; Hartman 1961; Moermond and Santana C., unpubl. data). The unusually large, broad tail
of the trogon (see Hartman 1961) is spread during stalling and may aid in braking or in
controlling the precise moment of the stall.
In contrast to these two aerial feeders, Pteroglossus torquatus appears to have difficulty
flying slowly. Its wing loading is high, and it has poorly developed V-shaped slots that may
operate effectively only when the wing is under maximum stress, as in take-off (pers. observ.).
These toucanets usually fly from tree to tree in a straight line with rapid wing beats and
relatively high impact landings. In comparison to Querula or Trogon, the toucanet's velocity
range is narrower and its horizontal cruising velocity faster, as expected for a bird with high
wing loading (cf. Pennycuick 1969, 1975). Correlated with better reaching ability, leg mus-
culature of Pteroglossus is considerably heavier than that of the trogon or fruit-crow (Table 3).
The morphological differences between the toucanet and the two aerial species are marked
and correspond well with differences in fruit-taking behavior and fruit choice described in
earlier sections; however, the strong morphological differences between the trogon and fruit-
crow are more surprising given that both species take fruit primarily on the wing. Nonetheless,
the differences in their feeding techniques likely have consequences for fruit selection. The
trogon's flight style is characterized by expensive powered flapping; the fruit-crow's swooping
glides may provide cheaper flight and lower costs of plucking its fruit. Trogons, however,
appear better at negotiating small spaces in the understory and lower parts of tree crowns,
while the fruit-crow forages more often (and presumably more efficiently) among the larger
open spaces of the canopy and takes more fruits from the upper portions of trees.
Field observations of similar birds at Monte Verde, Costa Rica (Snow 1977; Wheelwright
1983; Santana C. and Milligan 1984) provide strong support for the patterns we observed
in the lowland birds listed above. Observations of the Resplendent Quetzel {Pharomachrus
mocinno), and the Three-wattled Bellbird (Procnias tricarunculata), a large, broad-winged
cotinga, feeding in the same trees show that Pharomachrus take fruit primarily from the lower
portions of the trees whereas the bellbirds take significantly more fruits from the upper portions
of the trees (Santana C. and Milligan 1984). Compared to Querula, the bellbird does not
appear to be as specialized for aerial flight, having a shorter wing, likely higher wing loading,
and relatively longer (and stronger?) legs (Snow 1982). The bellbird often is able to pick and
reach some fruit from a perch (Snow 1977; Santana C. and Milligan 1984) but frequently
resorts to taking fruit on the wing if the fruits are not easily accessible or are difficult to remove
(Snow 1977). The two toucans at Monte Verde, Aulacorhynchus prasinus and Ramphastos
sulfuratus, pick or reach virtually all their food items. The Ramphastos, which are much
heavier than the Aulacorhynchus, are "able to pluck every fruit they reach out for and grasp
with their mandibles, so they hardly move at all while feeding" (Snow 1977:628).
The small wing area and deep slotting of both Manacus and Phoenicircus suggest wing
movements and flight behavior similar to those described for Trogon massena above. The
relatively longer wing of Pipra mentalis and other small manakins may improve their hovering
abilities and lower other flight costs (cf. Norberg 198 1). Other examples with relatively minor
but, nevertheless, noticeable, differences in wing shape invite further analysis.
Legs.- Legs are the other major morphological feature influencing manner of capture and
selection of fruits. Although muscle weights are seldom taken, Hartman's data (1961; Table
3) show considerable interspecific variation in the percent of body weight in leg muscles. The
leg musculature of trogons accounts for only about 3 percent of their body weight (2.7-3.2
percent for 7 Trogon spp. and 3.4 percent for Pharomachrus), and their feet are so weak that
the birds may be unable to turn around on a perch without using their wings (Santana C,
Moermond, and Denslow, unpubl. data). Many other species, however, have two to four times
as much of the body weight in leg muscles (e.g., Ramphastidae and tanagers, Table 3). In the
absence of functional analyses, we assume that birds with a greater percentage of body weight
in leg musculature have stronger legs.
Species with strong legs fall into two fairly distinct functional groups: ground feeding birds,
such as most thrushes (e.g., Catharus spp. and Turdus spp.) and emberizines (e.g., Arremon
and Arremonops spp.), and perching birds, such as many tanagers and honeycreepers, that
reach food items from perches on narrow branches (see previous section).
Relative tarsal lengths differ in the two groups of strong-legged species. Those birds that
often perch or feed on the ground have longer tarsi than similar species that feed primarily
above the ground; this pattern has been observed within diverse groups of birds (Dilger 1956;
Newton 1967; Fretwell 1969; Partridge 1976; Fitzpatrick 1978). The tarsi of birds that perch
primarily on branches vary in length and diameter depending on their feeding behavior. Among
tyrannid flycatchers, aerial feeders using perches for resting have short tarsi, while those that
feed by gleaning from perches have long, narrow tarsi (Fitzpatrick 1978). The long legs of
perch-gleaners are presumed to increase the visible scanning area and to facilitate clinging to
angled perches (Fitzpatrick 1978). Few tyrannids, however, reach below their perch level.
Species that reach down from perches tend to have shorter and thicker tarsi (Grant 1965,
1966; Sturman 1968; Moermond, Howe, Bruskewitz, and Rusterholz, unpubl. data; see Par-
tridge 1976 and references therein).
Among frugivores, species with the "ground" type legs are able to reach out or down from
perches only with difficulty. Our aviary experiments with two such species (Catharus minimus
and Arremon aurantiirostris) have shown that these birds would rather jump up to hanging
fruits that reach down when fruits are further than a few centimeters below a perch (Denslow
and Moermond, in press, unpubl. data). Species with strong legs of the "perching" type are
usually agile on small diameter perches and able to reach below their perch to pluck fruits.
Many of the smaller tanagers and honeycreepers appear able to reach well below a perch. As
in hovering ability, small absolute size facilitates such behavior (Newton 1967). Nevertheless,
not all small birds reach equally well. Thraupis palmarum, for example, can reach its full
body length below a perch, while the somewhat smaller Ramphocelus passerini cannot do so
without apparent difficulty (Denslow and Moermond, in press, unpubl. data). The Thraupis,
a heavier bird, has a distinctly shorter but thicker tarsus than the Ramphocelus.
Additional morphological features may enhance the reaching abilities of "perchers" (Os-
terhaus 1962; Leisler and Thaler 1982). Snow and Snow (1971) noted aspects of the foot,
such as grip strength and claw shape, that may enhance the reaching ability of Thraupis
palmarum. They also noted that T. palmarum and Tangara guttata, which frequently foraged
by clinging head-down, had longer tails than their sympatric congeners. We have also noted
in our aviary experiments that the long tail of Saltator maximus appears to aid in balance,
allowing this species to forage in hanging positions on thin stems or leaves.
Reaching ability may be constrained by perch diameter. We have observed that a small
bird such as Euphonia gouldi will reach readily from a 3 mm diameter perch but reluctantly
from a 12 mm diameter perch (Moermond and Denslow 1983). A study of the ability of two
paruline warbler species to reach from a series of perches demonstrated a considerable dif-
ference in distance reached over a relatively small range of perch diameters (3 mm-24 mm)
(Moermond, Howe, Bruskewitz, and Rusterholz, unpubl. data). Similar detailed studies of
closely related species have shown that minor differences in leg dimensions are correlated
with distinct differences in activity patterns and microhabitat (Dilger 1956 with thrushes;
Pearson 1977 with antwrens; Leisler and Thaler 1982 with Kinglets). These studies suggest
that the variations in tarsal length and leg muscle mass described for fruit-eating birds are
associated with differences in ability to negotiate different substrates and to pluck fruits pre-
sented on infructescences of different structure.
Among medium to large tanagers (Table 3), species that forage in the canopy or outer edges
of vegetation (Piranga rubra, P. olivacea, Thraupis episcopus, T palmarum, Tachyphonus
luctuosus) have lower wing loading, higher aspect ratios, a higher percent of body weight in
flight muscles, and a lower percent of body weight in their legs muscles than most species that
frequently forage on or near the ground and in low vegetation (Tachyphonus rufus, Rampho-
celus passerinii, R. icteronotus, Habia rubica, H.fuscicauda). Among the canopy tanagers, the
ones with the best reaching ability (Thraupis spp.) have the highest percent of body weight
in their lower extremities, although the percentage is still below that shown by the "low
vegetation and ground" species.
The suite of morphological features of a particular bird are expected to determine the relative
accessibility of and the costs of taking available fruits and, thus, to influence strongly the bird's
diet. While this is most clearly seen in such divergent birds as trogons and toucans, we believe
it also contributes importantly to fruit choice among all fruit-eaters. Although species of diverse
morphology often feed at the same fruiting tree (Eisenmann 1961; Land 1963; Willis 1966;
Olson and Blum 1968; Leek 1969; Howe 1977; McDiarmid et al. 1977; Fig. 4), a discriminant
function analysis of data from four of these studies was partially successful in separating groups
of species on the basis of body, wing, and leg characteristics, although not of bill dimensions
(Ricklefs 1977). Similarly, a multivariate analysis (Denslow and Moermond, in press) of 15
species of tanagers and honeycreepers studied by Snow and Snow (1971) demonstrated that
these species could be separated by the species of fruits in their diets and that the differences
among the bird groups appeared correlated with aspects of their morphology (bill size, percent
body weight in leg muscles, wing loading). Likewise our aviary choice tests with 20 species
of tanagers, manakins, thrushes, finches, and tyrannids have shown that all those tested exhibit
clear, well-defined preferences for more accessible fruits with the differences in preferences
associated with differences in morphology (Moermond and Denslow 1983; Denslow and
Moermond, in press, unpubl. data; Levey, unpubl. data).
A phylogenetic survey of frugivores shows, not surprisingly, that the members of each family
are usually similar in morphology and similar in the techniques they use to take fruits (Table
3). Where important variation in locomotory morphology exists in these families, it is often
associated with different techniques of fruit capture. The adaptive values of these patterns are
apparent if one assumes that fruits presented in different ways actually constitute different
resources. If certain morphological characteristics allow a bird to take some of these fruits
more cheaply than others, then selection on such morphological adaptations may be enhanced.
The presence of frugivorous species in so many families is not evidence for lack of requirements
for eating fruits but rather represents adaptive "specialization" to exploit fruits displayed in
different ways.
SPECIALISTS, GENERALISTS, AND SEED DISPERSAL

Snow (1980) designated as fruit specialists those species that can survive entirely on fruit
and rear their young mainly on fruit. Such species (Oilbirds, a few cotingids) represent a small
subset of the total range of fruit-eating birds that still may be classed as "legitimate" frugivores,
i.e., species that digest the fruit pulp or juices but not the seed (Snow 1980). A further division
of this large group of "legitimate" frugivores into specialist and generalists would be difficult
and beset with contradictions and inconsistencies. For example, although large, lipid-rich
fruits are taken primarily by large frugivores, these same birds generally take many species of
small fruits also. By virtue of their sizes, the large frugivores (including those usually considered
specialists) are certainly capable of taking the largest size range of fruits and, therefore, the
largest number of fruit species (Wheelwright in press; Table 2, Fig. 1). In addition, morpho-
logical specializations for fruit-eating have been noted for only a few species, but include both
large and small birds (e.g., Oilbirds and euphonias). Specializations in bill, wing, and leg
morphology that affect methods of removal of fruits can be identified in several species among
both traditional specialists and generalists. Finally, the "high quality" seed dispersal said to
be associated with specialists is not necessarily characteristic of all birds traditionally called
specialists (Oilbirds leave many seeds on the floor of their roosting caves), and, in addition,
many traditional generalists appear to be good seed dispersers (e.g., Leek 1972a; Howe and
DeSteven 1979; Greenberg 1981). With the exception of a few species that appear to be total
frugivores, there does not seem to be sufficient basis in behavior, morphology, or function on
which to recognize generalists and specialists among frugivorous birds.
As Wheelwright and Orians (1982) pointed out, the quality of seed dispersal for a given
plant is likely to depend on a variety of factors and is not easily predicted just from the species
of birds feeding on the plant's fruits. Safe sites for seed germination are not likely recognizeable
by the birds and in any case a "disperser" (in contrast to a pollinator) is "rewarded" for taking
the seed but not for delivering it. Adaptive strategies for seed dispersal are more likely to
involve simply moving as many seeds as possible from the immediate vicinity of the parent
tree (Denslow 1980). Co-evolution among fruiting plants and their dispersers has very ob-
viously occurred as evidenced in the characteristics of fruit displays (Van der Pijl 1969), but
a better understanding of the interaction at the species level must take into account the factors
influencing fruit selection by the dispersers.
The birds' choice of fruits is neither confined to specific fruits nor is it dictated only by
chance of encounter. Experiments with a number of species of small and two species of large
frugivores have shown that the birds make systematic decisions and that their choices depend
on such factors as fruit size, pulp-to-seed ratio, ripeness, color, taste, and importantly, acces-
sibility of the fruits from a perch and the distance between fruit clusters (Best 1981; Herrera
1981; Sorensen 1981; Moermond and Denslow 1983; Denslow and Moermond, in press;
Levey et al. 1984; Santana C, Moermond, Denslow, unpubl. data).
Of most interest among the aviary experiments are choices in which a bird is induced to
change an initial fruit preference by a change in one of the conditions under which the preferred
fruit is offered (for example, a decrease in the accessibility of the preferred fruit or an increase
in the flight distance to the preferred fruit). Such changes in preference have been observed
in each of the frugivores we have studied so far, leading us to believe that fruit choice in the
field is a dynamic process, i.e., the species and quantities of fruits taken are contingent upon
the context within which each fruit is found (e.g., Best 1981; Levey et al. 1984).
Differences in fruits taken by different frugivores living in the same habitat may be attributed
to differences in the birds' morphological abilities and physiological requirements that translate
into differences in the benefit : cost balance associated with a given choice. The morphological
and behavioral differences among frugivores are assumed to influence not only number and
species of fruits eaten but also to influence the quality of dispersal of seeds. For example, very
small seeds, such as those of Melastomataceae or Ficus, are easily passed through the digestive
tract of large and small frugivores. Dispersal of larger seeds, however, is increasingly limited
by their size. Not only may they be swallowed only by birds with sufficiently large gapes, but
small and medium-sized birds (tanagers and finches) mash the fruits before swallowing them,
often dropping the seeds at or near the parent plant. Thus, it may be advantageous for a tree
producing medium to large seeds to display fruits in such a way as to lower the probability
of removal by birds in this group. Because these birds are perching birds, display of fruits at
the ends of long peduncles or at the tips of narrow flexible twigs, in effect, lowers their
accessibility (Snow 1977; Denslow and Moermond, in press). Such fruit displays may also
even enhance their probability of removal by aerial-feeding, fruit-swallowing birds (Denslow
and Moermond 1982, in press; Santana C, Moermond, and Denslow, unpubl. data).
We suggest that the key to understanding such differences among fruit-eating birds lies in
the constraints of morphology on behavior. Differences in digestive systems, bill strength and
shape, wing and leg morphology, and body weight differentially affect the accessibility of fruits,
the ease with which they are handled, the rates at which they are eaten, and the efficiency
with which they are digested. Such constraints are expected to influence the birds' choices of
fruits, the benefits and costs of taking those fruits, and ultimately, the fruiting patterns and
display structures of plants competing for birds as dispersers.
ACKNOWLEDGMENTS
We thank D. J. Levey and E. Santana C. for assistance in the field, for unpublished obser-
vations and for valuable discussions. We thank J. G. Blake, E. DeVito, R. L. Knight, D. J.
Levey, B. Loiselle, E. Santana C, D. W. Snow, F. G. Stiles, B. C. Wentworth, N. T. Wheel-
wright, and M. F. Willson for comments on portions of the manuscript. D. E. Stone, D. A.
Clark, D. B. Clark of the Organization for Tropical Studies provided much needed logistic
support at the La Selva Biological Station in Costa Rica. F. G. Stiles, J. R. Karr, and S. Flores
P. made it possible to visit other areas of the neotropics. Researchers at La Selva as well as
many others have added comments, advice, and observations for which we are grateful. The
University of Wisconsin Zoology Museum provided permission to work with its collection.
C. M. Hughes prepared the figures and graphs. The field work was supported in part by grants
from the National Science Foundation, P. and M. Hess, the Nave Fund and the Graduate
School of the University of Wisconsin. Assistance and support was also generously provided
by the Departments of Zoology and Botany of the University of Wisconsin.
LITERATURE CITED
Alexander, R. M. 1968. Animal Mechanics. University Washington Press, Seattle, Washington.
Beecher, W. J. 1951. Convergence in the Coerebidae. Wilson Bull. 63:274-287.
Berthold, P. 1976a. The control and significance of animal and vegetable nutrition in omnivorous
songbirds. Ardea 64:140-154.
Berthold, P. 1976b. Der Seidenschwanz Bombycilla garrulus als frugivorer Ernahrungsspezialist. Ex-
perientia 32:1445.
Best, L. S. 1981. The effect of specific fruit and plant characteristics on seed dispersal. Unpubl. Ph.D.
dissert., University of Washington, Seattle.
Bourne, G. R. 1974. The Red-billed Toucan in Guyana. Living Bird 13:99-126.
Brattsten, L. B. 1979. Biochemical defense mechanisms in herbivores against plant allelochemicals.
Pp. 199-270, In G. A. Rosenthal and D. H. Janzen (eds.), Herbivores: Their Interaction with
Secondary Plant Metabolites. Academic Press, New York.
Brown, R. H. J. 1963. The flight of birds. Biol. Rev. 38:460-489.
Buskirk, W. H. 1976. Social systems in a tropical forest avifauna. Am. Nat. 1 10:293-310.
Cadow, G. 1933. Magen undDarm der Fruchtauben. J. Omithol. 81:236-252.
Cone, C. D., Jr. 1968. The aerodynamics of flapping flight. Virginia Inst. Marine Sci., Spec. Scient.
Rep. No. 52.
Crome, F. H. J. 1975a. The ecology of fruit pigeons in tropical northern Queensland. Aust. Wildl. Res.
2:155-185.
Crome, F. H. J. 1975b. Notes on the breeding of the purple-crowned pigeon. Emu 75:172-174.
Crome, F. H. J. 1975c. Breeding, feeding, and status of the Torres Strait Pigeon on Low Isles, northeastern
Queensland. Emu 75:189-198.
Cvitanic, A. 1970. The relationships between intestine and body length and the nutrition of some bird
species. Larus 21-22:181-190.
Da vies, N. B., and R. E. Green. 1976. The development and ecological significance of feeding techniques
in the reed warbler (Acrocephalus scirpaceus). Animal Behaviour 24:213-229.
DeJong, M. J. 1983. Bounding flight in birds. Unpubl. Ph.D. dissert., University of Wisconsin-Madison,
Madison.
Delacour, J., and D. Amadon. 1973. Curassows and related birds. American Museum Natural History,
New York.
Denslow, J. S. 1980. Gap partitioning among tropical rainforest trees. Biotropica 12 Suppl.:47-55.
Denslow, J. S., and T. C. Moermond. 1982. The effect of accessibility on rates of fruit removal from
neotropical shrubs: an experimental study. Oecologia 54:170-176.
Denslow, J. S., and T. C. Moermond. in press. The interaction of fruit display and the foraging
strategies of small frugivorous birds. In W. D'Arcy and M. D. Correa A. (eds.), The Botany and
Natural History of Panama: la Botanica y Historia Natural de Panama. Missouri Botanical Garden,
St. Louis, Missouri.
Desselberger, H. 1931. Der Verdauungskanal der Dicaeiden nach Gestalt und Funktion. J. Ornithol.
79:353-370.
Dilger, W. C. 1956. Adaptive modifications and ecological isolating mechanisms in the thrush genera
Catharus and Hylocichla. Wilson Bull. 68:171-199.
Docters Van Leeuwen, W. M. 1954. On the biology of some Javanese Loranthaceae and the role birds
play in their life-histories. Beaufortia 4:105-205.
Eisenmann, E. 1961. Favorite foods of neotropical birds: flying termites and Cecropia catkins. Auk 78:
636-637.
Faaborg, J. R., and J. W. Terborgh. 1980. Patterns of migration in the West Indies. Pp. 157-163,
In A. Keast and E. S. Morton (eds.), Migrant Birds in the Neotropics. Smithsonian Institution
Press, Washington, D.C.
Fisher, H. 1972. The nutrition of birds. Pp. 431-469, In D. S. Farner and J. R. King (eds.), Avian
Biology Vol. II. Academic Press, New York.
Fitzpatrick, J. W. 1978. Foraging behavior and adaptive radiation in the avian family Tyrannidae.
Unpubl. Ph.D. dissert., Princeton University, Princeton, New Jersey.
Fitzpatrick, J. W. 1980. Foraging behavior of neotropical tyrant flycatchers. Condor 82:43-57.
Fleming, T. H. 1979. Do tropical frugivores compete for food? Am. Zool. 19:1 157-1 172.
Fogden, M. P. L. 1972. The seasonality and population dynamics of equatorial forest birds in Sarawak.
Ibis 114:307-343.
Fogden, M. P. L., and P. M. Fogden. 1979. The role of fat and protein reserves in the annual cycle
of the Grey-backed Camaroptera in Uganda (Aves:Sylviidae). J. Zool. (Lond.) 189:233-258.
Forshaw, J. M. 1978. Parrots of the World. David and Charles, London.
Foster, M. S. 1974. Rain, feeding behavior, and clutch size in tropical birds. Auk 91:722-726.
Foster, M. S. 1977. Ecological and nutritional effects of food scarcity on a tropical frugivorous bird
and its fruit source. Ecology 58:73-85.
Foster, M. S. 1978. Total frugivory in tropical passerines: a reappraisal. Trop. Ecol. 19:131-154.
Foster, M. S., and R. McDiarmid. 1983. Nutritional value of the aril of Trichilia cuneata, a bird-
dispersed fruit. Biotropica 15:26-31.
Foster, R. B. 1982a. The seasonal rhythm of fruit fall on Barro Colorado Island. Pp. 151-172, In E.
G. Leigh, Jr., A. S. Rand, and D. Windsor (eds.), The Ecology of a Tropical Forest. Smithsonian
Institution Press, Washington, D.C.
Foster, R. B. 1982b. Famine on Barro Colorado Island. Pp. 201-212, In E. G. Leigh, Jr., A. S. Rand,
and D. Windsor (eds.), The Ecology of a Tropical Forest. Smithsonian Institution Press, Wash-
ington, D.C.
Frankie, G. W., H. G. Baker, and P. A. Opler. 1974. Comparative phenological studies of trees in
tropical wet and dry forests in the lowlands of Costa Rica. J. Ecol. 62:881-919.
Fretwell, S. D. 1969. Ecotypic variation in the nonbreeding season m migratory populations: a study
of tarsal length in some Fringillidae. Evolution 23:406-420.
Goodwin, D. 1970. Pigeons and Doves of the World. British Museum (Natural History), London.
Grant, P. R. 1965. The adaptive significance of some size trends in island birds. Evolution 19:355-
367.
Grant, P. R. 1966. Further information on the relative length of the tarsae in land birds. Postilla 98:
1-13.
Greenberg, R. 1981. Frugivory in some migrant tropical forest wood warblers. Biotropica 13:215-
223.
Greenewalt, C. H. 1975. The flight of birds. Trans. Am. Phil. Soc., New Ser. 65:1-67.
Groebbels, F. 1932. Der Vogel. Bau, Funktion, Lebenserscheinung, Einpassung. Vol. 1. Borntraeger,
Berlin.
Hails, C. J. 1979. A comparison of flight energetics in hirundines and other birds. Comp. Biochem.
Physiol. A 63:581-586.
Hartman, F. A. 1961. Locomotor mechanisms of birds. Smithson. Misc. Collect. 143:1-91.
Herrera, CM. 1981. Fruit variation and competition for dispersers in natural populations of Smilax
aspera. Oikos 36:51-58.
Herrera, C. M. 1982. Defense of ripe fruits from pests: its significance in relation to plant-disperser
interactions. Am. Nat. 120:218-241.
Herrera, C. M. 1984. Adaptation to frugivory of Mediterranean avian seed dispersers. Ecology 65:
609-617.
Herrera, C. M., and P. Jordano. 1981. Prunus mahaleb and birds: the high-efficiency seed dispersal
system of a temperate fruiting tree. Ecol. Monogr. 51:203-218.
Hill, F. W., and L. M. Dansky. 1954. Studies of the energy requirements of chickens I. The effect of
dietary energy level on growth and feed consumption. Poult. Sci. 33:1 12-1 19.
Hilty, S. L. 1980. Flowering and fruiting periodicity in a premontane rain forest in Pacific Colombia.
Biotropica 12:292-306.
Howe, H. F. 1977. Bird activity and seed dispersal of a tropical wet forest tree. Ecology 58:539-550.
Howe, H. F. 1981. Dispersal of a neotropical nutmeg ( Virola sebifera) by birds. Auk 98:88-98.
Howe, H. F. 1982. Fruit production and animal activity at two tropical trees. Pp. 189-199, In E. G.
Leigh, Jr., A. S. Rand, and D. Windsor (eds.), The Ecology of a Tropical Forest. Smithsonian
Howe, H. F., and D. DeSteven. 1979. Fruit production, migrant bird visitation, and seed dispersal of
Guarea glabra in Panama. Oecologia 39:185-196.
Howe, H. F., and G. F. Estabrook. 1977. On intraspecific competition for avian dispersers in tropical
trees. Am. Nat. 1 1 1:817-832.
Howe, H. F., and R. B. Primak. 1975. Differential seed dispersal by birds of the tree Casearia nitida
(Flacourtiaceae). Biotropica 7:278-283.
Howe, H. F., and G. A. Vande Kerckhove. 1980. Nutmeg dispersal by tropical birds. Science 210:
925-927.
James, F. C. 1982. The ecological morphology of birds: a review. Ann. Zool. Fennici 19:265-275.
Janson, C. 1983. Adaptation of fruit morphology to dispersal agents in a neotropical forest. Science
219:187-189.
Janzen, D. H. 1967. Synchronization of sexual reproduction of trees within the dry season in Central
America. Evolution 21:620-637.
Janzen, D. H. 1981. Ficus ovalis seed predation by an orange-chinned parakeet (Brotogeris jugularis)
in Costa Rica. Auk 98:841-844.
Jenkins, R. 1969. Ecology of three species of Saltators in Costa Rica with special reference to their
frugivorous diet. Unpubl. Ph.D. dissert., Harvard University, Cambridge, Massachusetts.
Kantak, G. E. 1979. Observations on some fruit-eating birds in Mexico. Auk 96:183-186.
Karr, J. R., and F. C. James. 1975. Eco-morphological configurations and convergent evolution in
species and communities. Pp. 258-291. In M. L. Cody and J. M. Diamond (eds.), Ecology and
Evolution of Communities. Harvard University Press, Cambridge, Massachusetts.
Karr, J. R., D. W. Schemske, and N. Brokaw. 1982. Temporal variation in the undergrowth bird
community of a tropical forest. Pp. 441-453. In E. G. Leigh, Jr., A. S. Rand, and D. Windsor
(eds.), The Ecology of a Tropical Forest. Smithsonian Institution Press, Washington, D.C.
Kear, J. 1962. Food selection in finches with special reference to interspecific differences. Proc. Zool.
Soc. London 138:163-204.
Keast, A., and E. S. Morton (eds.). 1980. Migrant Birds in the Neotropics. Smithsonian Institution
Press, Washington, D.C.
Kokshaysky, N. V. 1973. Functional aspects of some details of bird wing configurations. Syst. Zool.
22:442-450.
Land, H. C. 1963. A tropical feeding tree. Wilson Bull. 75:199-200.
Leck, C. F. 1969. Observations of birds exploiting a Central American fruit tree. Wilson Bull. 8 1:264-
269.
Leck, C. F. 1971a. Measurement of social attractions between tropical passerine birds. Wilson Bull.
83:278-283.
Leck, C. F. 1971b. Overlap in the diet of some neotropical birds. Living Bird 10:89-106.
Leck, C. F. 1972a. The impact of some North American migrants at fruiting trees in Panama. Auk 89:
842-850.
Leck, C. F. 1972b. Seasonal changes in feeding pressures of fruit and nectar eating birds in the neotropics.
Condor 74:54-60.
Leck, C. F. 1973. Observations of birds at Cecropia trees in Puerto Rico. Wilson Bull. 84:498-500.
Leisler, B., and E. Thaler. 1982. Differences in morphology and foraging behaviour in the goldcrest
Regulus regulus and the firecrest R. ignicapillus. Ann. Zool. Fennici 19:277-284.
Levey, D. J., T. C. Moermond, and J. S. Denslow. 1984. Fruit choice in neotropical birds: the effect
of distance between fruits on preference patterns. Ecology 65:844-850.
Leveille, G. A., and H. Fisher. 1958. The amino acid requirements for maintenance in the adult
rooster. 1. Nitrogen and energy requirements in normal and protein-depleted animals receiving
whole egg protein and amino acid diets. J. Nutr. 66:441-453.
Martin, E. W. 1968. The effects of dietary protein on the energy and nitrogen balance of the Tree
Sparrow, (Spizella arborea arborea). Physiol. Zool. 41:313-331.
McDiarmid, R. W., R. E. Ricklefs, and M. S. Foster. 1977. Dispersal of Stemmadenia donnell-
smithii (Apocynaceae) by birds. Biotropica 9:9-25.
McKey, D. 1975. The ecology of coevolved seed dispersal systems. Pp. 159-191, In L. E. Gilbert and
P. H. Raven (eds.), Coevolution of Animals and Plants. University Texas Press, Austin, Texas.
Milton, K., and F. R. Dintzis. 1981. Nitrogen-to-protein conversion factors for tropical plant samples.
Biotropica 13:177-181.
Moermond, T. C. 1983. Suction-drinking in tanagers and its relation to fruit handling. Ibis 125:545-
549.
Moermond, T. C, and J. S. Denslow. 1983. Fruit choice in neotropical birds: effects of fruit type and
accessibility on selectivity. J. Anim. Ecol. 52:407-420.
Morden-Moore, A. L., and M. F. Willson. 1982. On the ecological significance of fruit color in Prunus
serotina and Rubus occidentalis: field experiments. Can. J. Bot. 60:1554-1560.
Morrison, D. W. 1980. Efficiency of food utilization by fruit bats. Oecologia 45:270-273.
Morton, E. S. 1971. Food and migration habits of Eastern Kingbird in Panama. Auk 88:925-926.
Morton, E. S. 1973. On the evolutionary advantages and disadvantages of fruit eatmg in tropical birds.
Am. Nat. 107:8-22.
Morton, E. S. 1977. Intratropical migration in the Yellow-green Vireo and Piratic Flycatcher. Auk 94:
97-106.
Morton, E. S. 1978. Avian arboreal folivores: Why not? Pp. 123-129, In G. G. Montgomery (ed.),
The Ecology of Arboreal Folivores. Smithsonian Institution Press, Washington, D.C.
Morton, E. S. 1980. Adaptations to seasonal changes by migrant land birds in the Panama Canal Zone.
Pp. 437-453, In A. Keast and E. S. Morton (eds.), Migrant Birds in the Neotropics. Smithsonian
Munzell, H. E., L. D. Williams, L. P. Guild, C. B. Troescher, G. Nightingale, and R. S. Harris.
1949. Composition of food plants of Central America I. Honduras. Food Research 14:144-164.
National Research Council. 1960. Nutnent requirements of domestic animals I. Nutrient require-
ments of poultry. Publ. 827.
Nestler, R. B., W. W. Bailey, L. M. Llewellyn, and M. J. Kensberger. 1^44. winter protein
requirements of bobwhite quail. J. Wildl. Manage. 8:218-222.
Newton, I. 1967. The adaptive radiation and feeding ecology ot some British nnches. ibis 1UV:33-y».
Norberg, U. M. 1979. Morphology ot the wings, legs ana tail oi tnree coniierous iorest tits, me goiacresi,
and the treecreeper in relation to locomotor pattern and feeding station selection. Philos. Trans.
R. Soc. Lond. B Biol. Sci. 287:131-165.
Norberg, U. M. 1981. Flight, morphology and the ecological niche in some birds and bats. Symp.
Zool. Soc. Lond. 48:173-197.
Olson, S. L., and K. E. Blum. 1968. Avian dispersal of plants in Panama. Ecology 49:565-566.
Opler, P. A., G. W. Frankie, and H. G. Baker. 1980. Comparative phenological studies of treelet
and shrub species in tropical wet and dry forests in the lowlands of Costa Rica. J. Ecol. 68:167-
188.
Osterhaus, M. B. 1962. Adaptive modifications m the leg structure ot some North American warblers.
Am. Midi. Nat. 68:474-486.
Parker, T. A. 1981. Distribution and biology of the White-cheeked Cotinga Zaratornis stresemanni,
a high Andean frugivore. Bull. Br. Ornithol. Club 101:256-265.
Partridge, L. 1976. Some aspects of the morphology ot Blue I its {Pants caerulens) ana coal iits
(Parus ater) in relation to their behaviour. J. Zool. (Lond.) 179:121-133.
Pearson, D. L. 1977. Ecological relationships of small antbirds m Amazonian bird communities. Auk
94:283-292.
Pennycuick, C. J. 1969. The mechanics of bird migration. Ibis 1 1 1:525-556.
Pennycuick, C. J. 1975. Mechanics of flight. Pp. 1-75, In D. S. Farner and J. R. King (eds.), Avian
Biology, Vol. 5. Academic Press, New York.
Pullainen, E., P. Helle, and P. Tunkkari. 1981. Adaptive radiation ot the digestive system, neart
and wings of Turdus pilaris, Bombycilla garrulus, Sturnus vulgaris, Pyrrhula pyrrhula, Pinicola
enuclieator and Loxia pytyopsittacus. Ornis Fenn. 58:21-28.
Rayner, J. M. V. 1981. Flight adaptations of vertebrates. Symp. Zool. Soc. Lond. 48:137-172.
Ricklefs, R. E. 1974. Energetics of reproduction m birds. Pp. 152-292, In R. A. Paynter (ed.), Avian
Energetics. Publ. Nuttall Ornithol. Club No. 15.
Ricklefs, R. E. 1977. A discriminant function analysis ot assemblages ot truit-eating birds in central
America. Condor 79:228-231.
Ridley, H. N. 1930. The dispersal of plants throughout the world. L. Reeve and Co., Ashford, Kent,
England.
Salomonsen, F. 1964. Flowerpeckers. Pp. 311-313, In A. Landsborough Thomson (ed.), A New Dic-
tionary of Birds. Thomas Nelson, London.
Santana C, E., and B. G. Milligan. 1984. Behavior of toucanets, bellbirds, and quetzals feeding on
Lauraceous fruits. Biotropica 16:1 52-1 54.
Savile, D. B. O. 1957. Adaptive evolution in the avian wing. Evolution 1 1:212-224.
Schaefer, E. 1953. Contribution to the life history of the swallow-tanager. Auk 70:403-460.
Schulenberg, T. S. 1983. Foraging behavior, eco-morphology, and systematics of some antshrikes
(Formicariidae: Thamnomanes). Wilson Bull. 95:505-521.
Sibley, C. G., S. M. Lanyon, and J. E. Alquist. 1984. The relationships of the Sharpbill (Oxyruncus
cristatus). Condor 86:48-52.
Simms, E. 1978. British Thrushes. Collins, London.
Skutch, A. F. 1933. The aquatic flowers of a terrestrial plant, Heliconia bihai L. Am. J. Bot. 20:535-
544.
Skutch, A. F. 1954. Life histories of Central American Birds. Pacific Coast Avifauna No. 31.
Skutch, A. F. 1960. Life histories of Central American Birds Vol. II. Pacific Coast Avifauna No. 34.
Skutch, A. F. 1969. Life histories of Central American Birds Vol. III. Pacific Coast Avifauna No. 35.
Skutch, A. F. 1971. Life history of the Keel-billed Toucan. Auk 88:381-396.
Skutch, A. F. 1972. Studies of tropical American birds. Publ. Nuttall Ormthol. Club No. 10.
Skutch, A. F. 1976. Parent Birds and Their Young. University Texas Press, Austin, Texas.
Skutch, A. F. 1981. New studies of tropical American birds. Publ. Nuttall Ornithol. Club No. 19.
Smythe, N. 1970. Relations between fruiting seasons and seed dispersal methods in a neotropical forest.
Am. Nat. 104:25-35.
Snow, B. K. 1970. A field study of the Bearded Bellbird in Trinidad. Ibis 1 12:299-329.
Snow, B. K. 1972. A field study of the Calfbird, Perissocephalus tricolor. Ibis 1 14:139-162.
Snow, B. K. 1977. Territorial behavior and courtship of the Three-wattled Bellbird. Auk 94:623-645.
Snow, B. K., and D. W. Snow. 1971. The feeding ecology of tanagers and honeycreepers in Trinidad.
Auk 88:291-322.
Snow, D. W. 1961. The natural history of the Oilbird, Steatornis caripensis, in Trinidad, W. I. Part 1.
General behavior and breeding habits. Zoologica (N.Y.) 46:27-48.
Snow, D. W. 1962a. A field study of the Black and White Manakin, Manacus manacus, in Trinidad.
Zoologica (N.Y.) 47:65-104.
Snow, D. W. 1962b. A field study of the Golden-headed Manakin, Pipra erythrocephala, in Trinidad.
Zoologica (N.Y.) 47:183-198.
Snow, D. W. 1962c. The natural history of the Oilbird, Steatornis caripensis, in Trinidad, W. I. Part
2. Population, breeding ecology, food. Zoologica (N.Y.) 47:199-221.
Snow, D. W. 1965. A possible selective factor in the evolution of fruiting seasons in tropical forests.
Oikos 15:274-281.
Snow, D. W. 1971 a. Evolutionary aspects of fruit-eating by birds. Ibis 113:1 94-202.
Snow, D. W. 1971b. Observations on the Purple-throated Fruit-crow in Guyana. Living Bird 10:5-17.
Snow, D. W. 197 lc. Notes on the biology of the Cock-of-the-rock (Rupicola rupicola). J. Ornithol. 1 12:
323-333.
Snow, D. W. 1973. Distribution, ecology and evolution of the bellbirds (Procnias, Cotingidae). Br. Mus.
(Nat. Hist.) Bull. Zool. 25:369-391.
Snow, D. W. 1979. Family Pipridae. Pp. 245-280, In M. A. Traylor, Jr. (ed.), Check-list of Birds of
the World, Vol. 3. Museum Comparative Zoology, Cambridge, Massachusetts.
Snow, D. W. 1980. Regional differences between tropical floras and the evolution of frugivory. Acta
XVII Congr. Int. Ornithol. 2:1 192-1 198.
Snow, D. W. 1981. Tropical frugivorous birds and their food plants: a world survey. Biotropica 13:
1-14.
Snow, D. W. 1982. The Cotingas. Cornell University Press, Ithaca, New York.
Sorensen, A. E. 1981. Interactions between birds and fruits in a temperate woodland. Oecologia 50:
242-249.
Sorensen, A. E. 1984. Nutrition, energy and passage time: experiments with fruit preference in European
blackbirds (Turdus merula). J. Anim. Ecol. 53:545-557.
Stiles, F. G. 1979. Notes on the natural history of Heliconia (Musaceae) in Costa Rica. Brenesia 15:
151-180.
Stiles, F. G. 1983. Birds. Pp. 502-530, In D. H. Janzen (ed.), Costa Rican Natural History. University
Chicago Press, Chicago, Illinois.
Stiles, F. G. In press. On the roles of birds in the dynamics of neotropical forests. In A. W. Diamond
and T. Lovejoy (eds.), Conservation of Tropical Forest Birds. International Council for Bird
Preservation, Cambridge, England.
Storer, R. W. 1969. What is a tanager? Living Bird 8:127-136.
Sturman, W. A. 1968. The foraging ecology of Parus atricapillus and P. rufescens in the breeding
season, with comparisons with other species of Parus. Condor 70:309-322.
Traylor, M. A., Jr., and J. W. Fitzpatrick. 1982. A surveyof the tyrantflycatchers.Living Bird 19:
7-50.
vanderPul, L. 1969. Principlesof Dispersalin HigherPlants. Springer-Verlag,New York.
Wagner, H. O. 1944. Notes on the life history ol the emeraldtoucanet. Wilson Bull. do:o3-/o.
Walsberg, G. E. 1975. Digestive adaptationsof Phainopepla nitens associated with the eating ol
mistletoe berries.Condor77:169-174.
Walsberg, G. E. 1977. Ecologyand energeticsof contrastingsocial systemsin Phainopeplanitens(Aves:
Ptilogonatidae).Univ. Calif. Publ. Zool. 108:1-63.
Ward, P. 1969a. Seasonaland diurnalchangesm the lat content ol an equatorialoira. rnysioi. ôoi.
42:85-95.
Ward, P. 1969b. The annualcycle of the Yellow-ventedBulbulPycnonotusgoiavierm a humid equa-
torial environment.J. Zool. (Lond.) 157:25-45.
Watt, B. K., and A. L. Muriel. 1963. Composition of foods. Agnc. Handbk. No. 8. u.b. uept.
Agriculture,Washington,D.C.
Wetmore, A. 1914. The developmentof the stomach m the euphonias.Auk 3 1:458-40 1.
Wheelwright, N. T. 1983. Fruitsand the ecology ol Kespienaentiûeizais.auk iuu:zoo-ûi.
Wheelwright, N. T. In press. Fruit size, gape width, and the diets ol truit-eatingDims,tcoiogy.
Wheelwright, N. T., W. A. Haber, K. G. Murray, and C. Uuindon. iy»4. iropicai mut-eaung
birds and their food plants:a survey of a Costa Rican lower montane forest. Biotropica 16:173-
192. __. _
Wheelwright, N. T., and G. H. Orians. 1982. Seed dispersal by animals: contrasts witn pouen
dispersal,problemsof terminology,and constraintson convolution.Am. Nat. 119:402-413.
White, S. C. 1974. Ecologicalaspects of growthand nutritionm tropical Iruit-eatingbirds. Unpubl.
Ph.D. dissert.,University of Pennsylvania,Philadelphia.
Willis, E. O. 1966. Competitiveexclusionand birds at truitingtrees m western coiumoia (sic.), auk
83:479^180.
Willson, M. F., and J. N. Thompson. 1982. Phenologyand ecology of color m bird-dispersedfruits,
or why some fruitsare red when they are "green."Can. J. Bot. 60:701-713.
Worthington, A. 1982. Populationsizes and breedingrhythmsof two species ot manakinsm relation
to food supply. Pp. 213-225, In E. G. Leigh, Jr., A. S. Rand; and D. M. Windsor (eds.), The
Ecologyof a TropicalForest. SmithsonianInstitutionPress, Washington,D.C.
Ziswiler, V., and D. W. Farner. 1972. Digestion and the digestive system. Pp. 343-430, in U. b.
Farnerand J. R. King (eds.), Avian Biology, Vol. 2. Academic Press, New York.

Neotropical Avian Frugivores Patterns of Behavior Morphology and Nutrition With Consequences For Fruit Selection

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Neotropical Avian Frugivores Patterns of Behavior Morphology and Nutrition With Consequences For Fruit Selection

Uploaded by

Copyright:

Available Formats

Neotropical Avian Frugivores: Patterns of Behavior, Morphology, and Nutrition, with

Consequences for Fruit Selection

ABSTRACT. A large number of neotropical bird species in many families reg-

RESUMEN. En las zonas neotropicales se encuentran un gran numero de especies

tangaras y algunos de los pinzones usualmente se comen la fruta despues de aplastarla

Fruit-eating among tropical birds is so wide-spread as to make definition of "frugivorous"

Annonaceae 1 86 4.8 3.3 21.0 1.40 .034 8

£ 3B3Dwnei XXXX XXXX i

Althoughsome largefrugivoresare closely dependenton a few species of fruitingtrees, most

BEHAVIOR AND MORPHOLOGY

species feeding on fruits in southernSpain demonstratedthat those species (all muscicapids)

Fig. 4. Feedingbehavior of the most common visitors to Miconia multispicata(Melastomataceae).

Fig. 6. Wing loadingfor manakins(numbers)and trogons(closed symbols)of differentbody weights

Flight Leg Aspect

Flight Leg Aspect

SPECIALISTS, GENERALISTS, AND SEED DISPERSAL

You might also like