Crop size is more important than neighborhood fruitavailability for fruit removal of Eugenia uniflora(Myrtaceae) by bird seed dispersers

Pedro G. Blendinger • Mariana Villegas

Received: 7 July 2010 / Accepted: 23 November 2010 / Published online: 8 December 2010

� Springer Science+Business Media B.V. 2010

Abstract For a plant with bird-dispersed seeds, the

effectiveness of seed dispersal can change with fruit

availability at scales ranging from individual plants to

neighborhoods, and the scale at which frugivory

patterns emerge may be specific for frugivorous

species differing in their life-history and behavior.

The authors explore the influence of multispecies fruit

availability at two local spatial scales on fruit

consumption of Eugenia uniflora trees for two func-

tional groups of birds. The authors related visitation

and fruit removal by fruit gulpers and pulp mashers to

crop size and conspecific and heterospecific fruit

abundance to assess the potential roles that facilitative

or competitive interactions play on seed dispersal. The

same fruiting scenario influenced fruit gulpers (legit-

imate seed dispersers) and pulp mashers (inefficient

dispersers) in different ways. Visits and fruit removal

by legitimate seed dispersers were positively related to

crop size and slightly related to conspecific, but not to

heterospecific fruit neighborhoods. Visits and fruit

consumption by pulp mashers was not related to crop

size and decreased with heterospecific fruit availabil-

ity in neighborhoods; however, this might not result in

competition for dispersers. The weak evidence for

facilitative or competitive processes suggest that

interaction of E. uniflora with seed dispersers may

depend primarily on crop size or other plant’s

attributes susceptible to selection. The results give

limited support to the hypothesis that spatial patterns

of fruit availability influence fruit consumption by

birds, and highlight the importance of considering

separately legitimate and inefficient dispersers to

explain the mechanisms that lie behind spatial patterns

of seed dispersal.

Keywords Frugivorous birds � Heterospecific

neighborhood � Legitimate seed dispersers �Montane forest � Seed dispersal � Spatial scale

Introduction

Frugivorous birds affect seed dispersal through the

number of fruits they remove, and through where and

in what condition seeds are dropped (Loiselle and

Blake 1999; Jordano and Schupp 2000; Wenny 2000).

Dispersal effectiveness can be understood as the

contribution that seed dispersal makes to plant fitness

Electronic supplementary material The online version ofthis article (doi:10.1007/s11258-010-9873-z) containssupplementary material, which is available to authorized users.

P. G. Blendinger (&)

CONICET—Instituto de Ecologıa Regional, Universidad

Nacional de Tucuman, CC 34, 4107 Yerba Buena,

Tucuman, Argentina

e-mail: blendinger@birdecology.com.ar;

blendingerp@umsl.edu

M. Villegas

Department of Biology, University of Missouri—Saint

Louis, One University Boulevard, Saint Louis, MO

63121, USA

123

Plant Ecol (2011) 212:889–899

DOI 10.1007/s11258-010-9873-z

(Schupp 1993). Quantitatively, dispersal effectiveness

can be defined as the number of seeds dispersed away

from the parent plant. This component of seed

dispersal effectiveness depends on the foraging

behavior of the frugivores and on a number of factors

that influence disperser activity, both intrinsic to the

plant (e.g., plant size, phenology, crop size, fruit size,

and seed size) and factors extrinsic to the plant (e.g.,

surrounding habitat structure or fruit production in the

neighborhood) (Herrera et al. 1994; Jordano and

Schupp 2000; Parciak 2002; Carlo 2005; Blendinger

et al. 2008).

The contributions of frugivorous birds to the

effectiveness of seed dispersal can change among

species according to fruit-handling behavior. Fruit

consumption could lead to successful seed dispersal if

done by fruit gulpers (legitimate seed dispersers), or to

unsuccessful or poor dispersal if done by pulp mashers

(ineffective seed dispersers) that drop the seeds

beneath the parent canopy (Foster 1990; Herrera

et al. 1994; Jordano and Schupp 2000). Although pulp

removal by pulp-mashers may enhance germination

and prevent microbial damage to seeds in certain plant

species (Traveset et al. 2007), pulp-mashers could be

mutualism exploiters if they consume fruits but fail to

provide high-quality services in return; this interaction

can result in significant costs to the plant if the

exploiter inhibits associations with legitimate dispers-

ers (Stanton 2003). Accordingly, a change in the

activity level of seed dispersers and pulp-mashers may

alter spatial patterns of seed dispersal, and variations

in visitation rate or fruit consumption relative to fruit

availability (e.g., crop size; neighborhood of fruiting

plants) between these functional groups could promote

individual plant differences in seed dispersal and

affect the degree to which parental plants are success-

ful in producing offspring.

Frugivore activity and movement are related to the

distribution of plants with fleshy fruits that they

consume (Schaefer et al. 2002; Burns 2004; Garcıa

and Ortiz-Pulido 2004; Saracco et al. 2004) and, thus,

spatial patterns of fruit removal and seed deposition

may vary among spatial scales. At a local scale, the

attributes and physical structure of each plant and its

neighbors (e.g., species present, quantity, distance

and array) likely influence the movement patterns and

feeding decisions of frugivorous birds (Levin et al.

2003; Garcıa and Ortiz-Pulido 2004; Carlo and

Morales 2008). Neighborhood effects may be

particularly important because the quantity and types

of available fruits around a plant might influence

foraging behavior of frugivorous birds (Sargent 1990;

Saracco et al. 2005; Carlo et al. 2007) and, conse-

quently, the effectiveness of seed dispersal. Neigh-

borhoods with higher fruit density might attract more

birds (Sargent 1990; Levey et al. 1994; Carlo and

Morales 2008), which in turn might increase the

number of fruits removed per plant, suggesting

facilitation among plants that share dispersers (Kwit

et al. 2004; Carlo 2005; von Zeipel and Eriksson

2007; Blendinger et al. 2008), or it might decrease

suggesting competition among plants (Saracco et al.

2005; Carlo and Morales 2008). Thus, to understand

the processes of fruit removal and seed dispersal, it is

essential to determine how bird behavior responds to

spatial trends of plant distribution, such as distance,

accessibility, or aggregation (Moermond and Denslow

1983; Levey et al. 1984). Despite the broad accep-

tance that plant–seed disperser interactions are a

multi-species mutualism, less emphasis has been

given to indirect species interactions within this

mutualism and how they change with the spatial

context (Stanton 2003; Carlo et al. 2007).

Given that frugivorous birds have preference for

certain fruit species over others and they respond to

spatial fruit distribution, fruit removal in an individ-

ual plant should be influenced by the identity and

spatial patterns of the surrounding fruit offer; per-

ception of the offering may be specific to the

frugivorous species. In this article, the influence of

fruit availability at several spatial scales on fruit

consumption by two functional groups of frugivore

(fruit gulpers and pulp mashers) was explored,

because differences in their foraging behavior could

have strong consequences on dispersal effectiveness.

The outcomes of seed dispersal depend on the scale at

which frugivory patterns emerge (Burns 2004; Garcıa

and Ortiz-Pulido 2004). To evaluate the scale at

which fruit availability affects removal in individual

plants, the authors investigated the importance of

fruit abundance on the amount of consumed fruits, by

considering: (a) the crop size effect, a very local

measure of fruit availability and (b) the effect of

conspecific and heterospecific fruit abundance, at two

neighborhood scales. Fruit removal in Eugenia unifl-

ora L. (Myrtaceae), locally known as arrayan, an

abundant bird-dispersed tree in lower strata of

subtropical forests in the Neotropics was examined.

890 Plant Ecol (2011) 212:889–899

123

If fruit abundance in the neighborhood acts as a

facilitator, then it is predicted that fruit removal in

E. uniflora will be higher in neighborhoods with

higher conspecific and heterospecific fruit availabil-

ity. On the other hand, if local fruit abundance leads

to competition for dispersers, then an inverse rela-

tionship between fruit removal and heterospecific or

conspecific fruit abundance is expected. Understand-

ing the influences of fruit availability factors (Jordano

and Schupp 2000) and their context-dependence is

central to understanding the role of frugivores on

dispersal success.

Methods

Study area

The study was conducted in December 2008 in Parque

Sierra de San Javier, Tucuman, Argentina. The area is

covered mostly by southern Yungas, a latitudinal

extension of the tropical cloud forests distributed along

the eastern slopes of the Andes. The study site (27�300

S, 65�400W) was located at 970 m asl on the upper part

of a slope with southwestern exposure, within the

elevational belt known as ‘‘Selva Montana’’ or lower

montane forest. The weather is highly seasonal, with a

warm and rainy period between November and March

when more than 75% of the rains fall. The mean annual

rainfall is 1,300–1,500 mm and the mean annual

temperature is 18�C (Grau 2002).

The mature forest in the study area is multi-

stratified, with treefall and landslide gaps within which

there is second-growth vegetation (Grau 2002; Malizia

and Grau 2008). The plot has emergent trees of

25–30 m in height represented by Blepharocalyx

salicifolius (Myrtaceae) and Cinnamomum porphyri-

um (Lauraceae), and an upper canopy comprised

mainly of Myrcianthes pungens (Myrtaceae), Para-

piptadenia excelsa (Fabaceae), Pisonia zapallo

(Nyctaginaceae), and Terminalia triflora (Combretaceae).

The lower stratum has crowns of 5–12 m in height,

dominated by E. uniflora (Myrtaceae), Piper tucuma-

num (Piperaceae), Allophylus edulis (Sapindaceae),

and Solanum riparium (Solanaceae). There is a dense

shrubby understorey of up to 4-m height dominated by

Psychotria carthagenensis (Rubiaceae). The forest in

the Parque Sierra de San Javier possesses a good

conservation condition and hosts a typical

complement of frugivorous birds of the southern

Yungas (Blendinger and Giannini 2010).

Focal tree selection

Eugenia uniflora reaches 6–10 m in height, excep-

tionally up to 15 m. It is distributed throughout the

subtropical and tropical forests of Argentina, Brasil,

Paraguay and Uruguay. Within the southern Yungas, it

is locally abundant between 500 and 1,200 m asl. Its

propagation is through vegetative growth and by seeds

dispersed by frugivorous birds and mammals. The

fruits are depressed-globose berries, 1.02 ± 0.45 g

(mean ± SD) of fresh weight and 1.29 ± 0.24 cm in

diameter, with 6–8 longitudinal ribs. The color ranges

from dark-red to black when ripe; they have one single

sub-globose seed of 6–8 mm in diameter.

The fruiting period extended from late November

to late December of 2008, but most of the fruits

ripened and were removed from a plant in 2 weeks.

We worked in a 6-ha permanent plot divided into 150

cells of 20 9 20 m. Within this plot, 17 focal plants

were selected wherein the authors made observations

of fruit consumption by birds and estimated the

number of ripe fruits (henceforth crop size). Trees

were selected to be representative of the crop size

range observed within the whole plot. We selected a

single focal plant per cell, with 33 m as the average

shortest distance between two focal trees (range:

25–67 m). Cells located on the edge of the plot were

avoided in order to measure fruit availability in

neighbor cells (see below). All focal trees had a crop

size C20 ripe fruits, except in two cases where a

group of E. uniflora trees as a single focal plant was

considered because of the low individual crop sizes

and because their crowns were intertwined.

Fruit abundance

The authors quantified the crop size of each focal tree,

immediately after observations of fruit consumption

by birds. In order to measure the fruit availability for

E. uniflora seed dispersers within the neighborhood of

each focal tree, the grid of 150 cells of 20 9 20 m was

used. The purpose behind this was to examine how

fruit availability of plants that were fruiting at the

same time as E. uniflora influenced fruit removal by

seed dispersers. Two neighborhood sizes were con-

sidered: 400 m2 (i.e., the cell in which the focal tree

Plant Ecol (2011) 212:889–899 891

123

was located), and 1,600 m2 (i.e., the cell in which the

focal tree was located, plus the three closest cells to the

focal tree). For all the 17 focal trees selected in each

cell, the crop size of every E. uniflora tree and those of

all other species with fruits (epiphytes, woody vines,

trees, and shrubs) were estimated using the following

scale of fruit abundance: 1–5, 6–10, 11–50, 51–100,

101–500, 501–1,000, and 1,001–5,000 fruits. Two

measures of fruit availability in the neighborhood

were used. The total number of ripe fruits was

estimated, calculated as the sum of all the individual

crop sizes registered in the neighborhood of each focal

tree, where crop size was the central value within each

range of fruit abundance. Since this measure might be

influenced by plants with very large crop sizes, a more

conservative measure, by calculating a Fruit Abun-

dance Index (FAI) that follows a logarithmic scale was

also used: 1 = 1–10 ripe fruits, 2 = 11–100 ripe

fruits, 3 = 101–1,000 ripe fruits, and 4 = C1,001 ripe

fruits (Saracco et al. 2004).

In order to determine which plant species are to be

included in the heterospecific neighborhood, coinci-

dently with the beginning of observations on E. unifl-

ora focal trees, the authors performed 100 h of

observations evenly distributed over the entire plot,

40 min per cell. Five independent observers sampled

the 150 cells for a week. Each cell was sampled twice

(for 20 min each time) in different days and times of

the day; all the observations were performed between

7:30 and 11:00 h, with proper meteorological condi-

tions (without rain or wind). During this time, the

authors registered every frugivory event by birds

(Blendinger et al., pers. obs.), which enabled us to

determine the plant species that were being consumed

by each frugivorous species. This information was

fundamental to make bird species-specific estimates of

fruit availability. Also, the authors checked the

observational data with the literature on the diet of

frugivorous birds in the Sierra de San Javier (Giannini

1999; Blendinger and Giannini 2010).

Bird observations

The authors performed direct observations of fruit

consumption by frugivorous birds in each focal plant,

5 h per plant, in two periods: 2.5 h in the period with

greater bird activity between 7:30 and 11:00 h; and

2.5 h in the period with lower activity from 11:00 h

until 15:00 h. These two observation periods were

conducted on different days for each focal plant,

consecutively when climatic conditions allowed it.

The authors recorded the following data: frugivorous

bird species and number of individuals that arrived to

the plant, fruit consumption and feeding technique

used, and number of fruits removed per bird (fruits

torn off from its peduncle). To determine their role as

dispersers, the authors classified birds as ‘‘legitimate

seed dispersers’’ if they swallowed the whole fruit,

and as ‘‘pulp mashers’’ if they pecked the fruit to

obtain pulp pieces without detaching it from its

peduncle, or after plucking, if they manipulated the

fruit while it was in the bill to tear off the pulp

discarding the seed without swallowing it. The birds

classified in the first group are considered efficient

dispersers because they defecate or regurgitate viable

seeds away from the parent plant, in contrast to birds

in the second group which are considered inefficient

seed dispersers.

Data analysis

The authors used simple linear regressions to analyze

the relationship between number of visits (by legiti-

mate dispersers and pulp mashers) and focal trees’

crop sizes and fruit availability in the neighborhood.

Similarly, the number of fruits consumed was related

to crop sizes and fruit availability. Independent

variables for the analyses at the neighborhood scale

were: fruit abundance of E. uniflora, sum of fruit

abundances of the other consumed species, and

fruit abundance of each one of the most abundant

fruit species consumed by the legitimate dispersers

and pulp mashers. The authors used multiple regres-

sion controlling for crop size, to explore the impor-

tance of fruit availability at the neighborhood scale.

The authors constructed forward stepwise models

using crop size of E. uniflora and conspecific and

heterospecific neighborhood variables. Number of

visits and fruit consumption of E. uniflora by legiti-

mate dispersers and pulp mashers were not spatially

autocorrelated at equal or shorter distances than the

neighborhood sizes that were used in this study

(Moran’s I, P [ 0.05) and in only few cases, did the

number of visits show spatial structure at longer

distance intervals. None of the four Moran’s I corre-

lograms of frugivore activity was significant

(P [ 0.05 after Bonferroni corrections). Exploratory

spatial analyses using modified t-test for correlations

892 Plant Ecol (2011) 212:889–899

123

with Dutilleul corrections of degrees of freedom gave

very similar results to the simple regressions models

reported here.

Results

Fruit availability

The authors registered eight plant species with ripe

fruits, in addition to E. uniflora, that made up the diet of

birds that consumed E. uniflora. These were two

epiphytes Aechmea distichantha (Bromeliaceae) and

Rhipsalis floccosa (Cactaceae), an hemiparasite (Pho-

radendron spp., Viscaceae) represented by three species

(P. falcifrons, P. tucumanense, and P. dipterum, ordered

by decreasing abundance) with similar morphology and

fruit disposition that were considered a single species

for the purposes of the study, an understory shrub

(P. carthagenensis, Rubiaceae) and the following trees:

C. porphyrium (Lauraceae), Cupania vernalis (Sapind-

aceae), P. tucumanum (Piperaceae) and S. riparium

(Solanaceae) (Table 1). C. porphyrium and C. vernalis

were represented by B3 plants, and were excluded from

the analysis. Epiphyte and hemiparasite species did not

grow on the focal trees; moreover, they are very

uncommon on E. uniflora trees.

Crop size in E. uniflora ranged from 20 to 200

fruits (mean ± SD = 57.9 ± 50.0, N = 17). Con-

specific fruit abundance ranged from 25 to 302 fruits

in neighborhoods of 400 m2, and from 46 to 719

fruits in neighborhoods of 1,600 m2 (Table 1). Crop

size in E. uniflora was correlated with conspecific

fruit abundance at 400 m2 (r = 0.70, P = 0.002) but

not at 1,600 m2 (r = 0.42, P = 0.10). Fruit abun-

dance and FAI of E. uniflora, were significantly

correlated between the two neighborhood scales, but

not heterospecific fruit abundance and FAI (Table

S1). FAI was significantly and positively correlated

with fruit abundance (r [ 0.75, P \ 0.017 for the

seven species analyzed). P. carthagenensis had

the highest heterospecific fruit abundance, followed

by Phoradendron spp. and A. distichantha (Table 1).

As a group, heterospecifics (all six species) were

present in the 17 neighborhoods; however, individual

species had ripe fruits in 1–13 of the 400-m2

neighborhoods. P. carthagenensis and A. distichantha

had ripe fruits in neighborhoods of 1,600 m2 of all 17

focal plants, and the other four species had ripe fruits

in 5–16 neighborhoods.

Visits and fruit consumption by legitimate seed

dispersers

In 85 h of observation, the authors recorded seven

frugivorous bird species that visited the focal plants

and consumed fruits of E. uniflora. These species

were Catharus ustulatus, Turdus rufiventris, Turdus

nigriceps, Thraupis sayaca, Chlorospingus ophthal-

micus, Parula pitiayumi, and Euphonia chlorotica. Of

those seven species, the three turdids (C. ustulatus,

T. rufiventris, and T. nigriceps) were legitimate seed

dispersers of E. uniflora, swallowing whole fruits

including seeds, whereas the remaining four species

fed on the pulp without swallowing the seed.

Legitimate seed dispersers accounted for 72% of

the visits and consumed 83% of the fruits (Table 2).

The number of visits and fruits consumed were

Table 1 Conspecific and heterospecific fruit availability around E. uniflora focal plants, at two different neighborhood sizes

Neighborhood at 400 m2 Neighborhood at 1,600 m2

Na No. fruits (mean ± SD) FAI Na No. fruits (mean ± SD) FAI

E. uniflora 17 136.6 ± 90.1 12.35 ± 7.0 17 288.9 ± 183.4 31.65 ± 18.8

A. distichantha 8 91.1 ± 122.1 3.4 ± 4.6 17 574.9 ± 409.4 21.5 ± 15.3

Phoradendron spp. 4 217.7 ± 484.7 3.2 ± 7.1 16 1124.8 ± 1066.0 16.4 ± 15.6

P. tucumanum 8 3.1 ± 4.9 0.7 ± 0.8 12 18.1 ± 21.1 2.5 ± 2.5

P. carthagenensis 11 1425.3 ± 2175.2 2.3 ± 1.8 17 4830.0 ± 5495.3 9.4 ± 3.7

R. floccosa 13 31.4 ± 24.4 4.8 ± 3.8 15 127.7 ± 90.8 19.7 ± 14.0

S. riparium 1 3.5 ± 14.6 0.2 ± 1.0 5 23.2 ± 56.1 1.1 ± 2.3

FAI Fruit Abundance Indexa Number of neighborhoods with C1 plants with ripe fruits

Plant Ecol (2011) 212:889–899 893

123

positively related to crop size (Fig. 1a, b; visits,

R2 = 0.58; fruit consumption, R2 = 0.59). However,

the percentages of total frugivore visits or fruits

consumed by the seed dispersers were not related

with crop size (P [ 0.37).

Seed dispersers consumed fruits from three heter-

ospecific plants (Phoradendron spp., P. carthagen-

ensis, and S. riparium). The authors did not record

fruit consumption from A. distichantha, P. tucuma-

num, or R. floccosa. Neither number of visits

(Table 3) nor number of fruits consumed (Table 4)

by seed dispersers were related to conspecific or fruit

abundance of heterospecific plants (the three species

combined) consumed in any of the neighborhood

sizes. However, fruit consumption and visits were

marginally related (P \ 0.07) with conspecific fruit

abundance in neighborhoods at the 400 m2 scale.

Multiple regression models controlling for crop size

(not reported) gave identical results for neighborhood

effects. There was not a significant relationship

between number of E. uniflora fruits consumed and

neighborhood fruit abundance for any of the species

from which fruits were consumed by seed dispersers

(P C 0.18 in the six cases).

Visits and fruit consumption by pulp mashers

Pulp mashers (T. sayaca, C. ophthalmicus, P. pitiay-

umi, and E. chlorotica) consumed the fruit in the

same plant or in nearby plants and, in most of the

cases (23 of 25 records), did not disperse the seed

Table 2 Number of visits, fruits consumed and focal trees of E. uniflora visited by seed dispersers and pulp mashers

Visits

(%)aFruits

consumed (%)bFocal trees

visited (%)cFocal trees where fruits

were consumed (%)c

Seed dispersers (T. rufiventris, C. ustulatus, T. nigriceps) 99 (71.7) 169 (82.4) 16 (94.1) 14 (82.3)

Pulp mashers (T. sayaca, P. pitiayumi, E. chlorotica,

C. ophthalmicus)

39 (28.3) 36 (17.6) 14 (82.3) 11 (64.7)

a Numbers in parentheses refer to percentage of the 138 visits by all frugivoresb Numbers in parentheses refer to the percentage of the 205 fruits consumed by all frugivoresc Numbers in parentheses refer to the percentage of the 17 focal trees visited

Num

ber

of fr

uits

rem

oved

0

10

20

30

40

50

Num

ber

of fr

uits

con

sum

ed

0

2

4

6

8

10

12

SEED DISPERSERS PULP MASHERS

b = 0.77P < 0.001

b = 0.41P = 0.10

Crop size0 30 60 90 120 150 180 210

Num

ber

of v

isits

0

3

6

9

12

15

18

21

b = 0.76P < 0.001

Crop size0 30 60 90 120 150 180 210

Num

ber

of v

isits

0

1

2

3

4

5

6

b = 0.28P = 0.28

a

b

c

d

Fig. 1 Simple regression

of crop size of E. uniflora in

relation to number of fruits

consumed and number of

visits by seed dispersers

(a–b) and pulp mashers

(c–d)

894 Plant Ecol (2011) 212:889–899

123

away from the parent plant, justifying their classifi-

cation as inefficient dispersers. These four species

performed 28% of visits to E. uniflora plants and

consumed 18% of the fruits (Table 2). Neither

number of visits nor number of fruits consumed

was significantly correlated with crop size (Fig. 1c,

d). The two trees with largest crop size had opposite

extreme values of fruit consumption (Fig. 1c); how-

ever, the authors did not find arguments to consider

them outliers, because other attributes of these trees

and their surrounding did not differ in comparison to

the other samples.

Pulp mashers consumed six additional fruit species

(A. distichantha, P. carthagenesis, P. tucumanum,

Phoradendron spp., R. floccosa, and S. riparium).

Number of visits was significantly related only with

total heterospecific fruit abundance in neighborhoods

of 400 m2 (Table 3), but in a direction opposite to

that which was expected (Fig. S1a, b; standardized

regression coefficients b = -0.50 and -0.58 for total

fruit abundance and FAI, respectively). Fruit con-

sumption in E. uniflora plants was only significantly

related to heterospecific fruit abundance in neighbor-

hoods of 400 m2 (Table 4) and also in a direction

opposite to expected (Fig. S1c; FAI, b = -0.51).

Accordingly, identical results of neighborhood effects

as found with simple regressions were obtained with

stepwise multiple regressions controlling for crop

size of E. uniflora (not reported). Fruit consumption

by pulp mashers was significantly related to fruit

abundance of Phoradendron spp. in neighborhoods of

1,600 m2 (R2 = 0.28, b = 0.53, P = 0.029). For the

remaining five species, the relationship between fruit

abundance in the neighborhoods (N = 17) and fruit

consumption of E. uniflora was not significant

(P C 0.16 for 11 cases).

Table 3 Simple regression analyses between number of visits to focal plants of E. uniflora by seed dispersers and pulp mashers in

relation to total fruit abundance of conspecific and heterospecific trees at two different neighborhood sizes

Fruit abundance Number of visits by seed dispersers Number of visits by pulp mashers

Neighborhood

at 400 m2Neighborhood

at 1,600 m2Neighborhood

at 400 m2Neighborhood

at 1,600 m2

R2 P R2 P R2 P R2 P

Conspecific fruit abundance 0.21 0.06 0.12 0.18 0.00 0.91 0.00 0.78

Conspecific FAIa 0.01 0.70 0.03 0.49 0.01 0.65 0.03 0.49

Heterospecific fruit abundance 0.00 0.97 0.03 0.50 0.25 0.04 0.02 0.64

Heterospecific FAIa 0.03 0.51 0.06 0.33 0.37 0.01 0.03 0.54

Only heterospecific plant species consumed by each group of frugivores were included in the analysis. N = 17 in all casesa FAI = Fruit Abundance Index: 1 1–10 ripe fruits, 2 11–100 ripe fruits, 3 101–1,000 ripe fruits, 4 C1,001 ripe fruits

Table 4 Simple regression analyses between number of E. uniflora fruits consumed by frugivores in relation to total fruit abundance

of conspecific and heterospecific trees at two different neighborhood sizes

Fruit abundance Fruits removed by seed dispersers Fruits consumed by pulp mashers

Neighborhood

at 400 m2Neighborhood

at 1,600 m2Neighborhood

at 400 m2Neighborhood

at 1,600 m2

R2 P R2 P R2 P R2 P

Conspecific fruit abundance 0.21 0.07 0.01 0.22 0.08 0.28 0.05 0.38

Conspecific FAI 0.02 0.62 0.02 0.64 0.04 0.47 0.00 0.98

Heterospecific fruit abundancea 0.00 0.94 0.03 0.53 0.08 0.26 0.00 0.93

Heterospecific FAIa 0.03 0.51 0.12 0.16 0.26 0.04 0.05 0.37

Only heterospecific plant species consumed by each group of frugivores were included in the analysis. N = 17 in all casesa Including three fruit species (Phoradendron spp., P. carthagenensis, and S. riparium) for seed dispersers and six fruit species

(A. distichantha, P. carthagenesis, P. tucumanum, Phoradendron spp., R. floccosa, and S. riparium) for pulp mashers

Plant Ecol (2011) 212:889–899 895

123

Discussion

The authors found limited support for the hypothesis

that predicts an influence of fruiting neighborhood on

fruit consumption by birds at individual plants

(Manasse and Howe 1983; Sargent 1990; Carlo

2005; Saracco et al. 2005). Legitimate seed dispersers

that consumed the fruits of E. uniflora responded to

fruit availability at very local scales, such as

individual plants, but this trend diluted at wider

scales including nearby plants that fruited synchro-

nously. Differences in frugivore life-history and

behavior could influence their ability to respond to

scale-dependent heterogeneity in fruit availability

(Garcıa and Ortiz-Pulido 2004). Consumption of E.

uniflora fruits by pulp mashers was negatively related

to fruit availability in close local neighborhoods,

probably because of their preferences for other fruit

species. However, as pulp mashers fail to return high-

quality dispersal services, indirect interactions among

E. uniflora and their heterospecific neighbors might

not result in competition by seed dispersers.

The different responses of seed dispersers and pulp

mashers to fruit availability in the neighborhood could

have been because they track fruit abundance at

different spatial scales and because of interspecific

differences in fruit choices. Although frugivorous birds

may track local abundance of ripe fruits (Loiselle and

Blake 1993; Rey 1995; Garcıa and Ortiz-Pulido 2004;

Hampe 2008), coexisting species that exploit similar

resources can differ in the spatial scales at which they

respond to fruit abundance (Saracco et al. 2004). The

results show that even at a small scale it is possible to

observe a differential response of frugivorous birds to

spatial variation of fruit abundance. Seed dispersers

(represented by T. rufiventris in about 73% of cases)

responded to fruit abundance at a scale of individual

plants, whereas pulp mashers (mainly T. sayaca, 81%

of consumption) appeared to respond to fruit abun-

dance on larger spatial scales (B400 m2). These results

highlight the fact that the response of each frugivore

group was context-dependent.

The presence of nearby plants that fruit synchro-

nously and which share the same dispersers might

lead to increases or decreases in the number of

dispersed seeds as a consequence of facilitation or

competition for dispersers, which are attracted to

habitat patches with higher fruit availability (Jordano

and Schupp 2000; Carlo 2005; Carlo and Morales

2008). However, the authors found little evidence of

competitive or facilitative processes among neighbor

plants in the E. uniflora-frugivores system. A plant’s

quantitative component of seed dispersal effective-

ness is strongly influenced by its crop size (Jordano

and Schupp 2000). When not all fruits are removed,

the disperser’s response is usually skewed to indi-

vidual plants with higher fruit production (e.g., Howe

and Estabrook 1977; Davidar and Morton 1986;

Ortiz-Pulido and Rico-Gray 2000; Takahashi and

Kamitani 2004; Blendinger et al. 2008), thus increas-

ing dispersal effectiveness of plants with large crops.

Crop size of E. uniflora positively influenced the

absolute number of dispersed seeds by attracting

more legitimate seed dispersers, but was not related

to the relative consumption of fruits by seed dispers-

ers, in comparison with pulp mashers. Thus, the

contribution of seed dispersers to the success of a

plant’s reproductive investment (Izhaki 2002) may

not have been influenced by the activity of pulp

mashers. Production of large crop sizes might repre-

sent an adaptive response to increase the chance of

seed dispersal when frugivore availability is limiting

(Parciak 2002). However, the lack of evidence of

competitive interactions with neighbor plants sug-

gests that at the local scale disperser availability was

not a limiting factor for the dispersal of E. uniflora.

Moreover, a larger crop size could be the conse-

quence of the repetitive production of new structural

modules added onto the older parts during growth

(Barlow 1989) and not subjected to selection by seed

dispersal. Thus, long-term studies that address inter-

annual variability in fruit production and its covari-

ation with frugivore abundance are necessary to know

what the evolutionary consequences of different crop

sizes can be.

More effective seed dispersal is expected for

E. uniflora when fruits are consumed by seed

dispersers such as turdids than when consumed by

pulp mashers such as thraupids. Birds that ingest the

whole fruit will retain the seed in their digestive tract

taking it away from the parent plant below which the

survival probability is usually lower (Janzen 1970;

Howe and Smallwood 1982; Jordano and Schupp

2000). Depositing seeds in faeces without the pulp

can reduce the attack risk of pathogenic fungus attack

(Leal and Oliveira 1998; Cazetta et al. 2008), and

possibly reduce the time and increase the frequency

of germination (Traveset 1998; Traveset et al. 2007).

896 Plant Ecol (2011) 212:889–899

123

Although pulp mashers may provide some benefit by

removing the pulp, they typically drop E. uniflora

seeds partially covered by pulp underneath the parent

plant, with expected negative consequences on the

plant fitness in terms of restricted dispersal and high

mortality risk to the offspring. Pulp mashers

responded to fruit availability of other plant species

located in close neighborhoods (at 400 m2), resulting

in lower frequency of visits and consumption of focal

plants of E. uniflora. This lower consumption of E.

uniflora fruits would not necessarily imply a negative

effect on the plant; a wider variety of fruit for

frugivores in heterospecific neighborhoods decreased

fruit consumption by inefficient dispersers, thereby

leaving fruits available for potential removal by

legitimate seed dispersers.

Turdus rufiventris and T. sayaca are the main

frugivorous species at the study site, where they

consume most available ornithochoric species over the

year (Giannini 1999). A simultaneous community-

level survey conducted in the study site (Blendinger

et al., pers. obs.), allowed the authors to interpret their

findings in terms of quantitative and qualitative

differences of those two species’ diets. In quantitative

terms, T. sayaca was 2.04 times more abundant than

T. rufiventris and it was the most important frugivore

(52% of all the frugivory events, against 18% for

T. rufiventris), but its consumption of E. uniflora was

much lower. In qualitative terms, the main difference

was the consumption of epiphyte species (A. disti-

chantha, Phoradendron spp., R. floccosa) by T. say-

aca, which were not used by T. rufiventris. Of these

epiphyte species, Phoradendron spp. was the most

frequently consumed by T. sayaca (34% of its

frugivorous diet; Blendinger et al., pers. obs.) and it

was the only plant species whose fruit abundance was

negatively related to fruit consumption of E. uniflora

by pulp mashers. The factors determining why fruit-

resource tracking is sometimes only expressed at

particular scales are poorly understood (Garcıa and

Ortiz-Pulido 2004). The authors suggest that the

different responses in consumption of E. uniflora by

these two functional groups of birds might be

explained by preferences for distinct fruit species.

Additionally, it is noteworthy that T. sayaca habitually

swallows small-sized seeds of the coexisting epiphytic

species (potentially enhancing their fitness). This

shows how the same frugivore species could mostly

benefit or be detrimental to different fruit partners.

In summary, the authors show how two functional

groups of frugivorous birds, seed dispersers that ingest

the seeds and pulp mashers that drop seeds, respond in

different ways to fruit availability on E. uniflora plants

and their neighbors. Frequency of visits and number of

fruits consumed by seed dispersers were related to

crop size but only slightly to fruit abundance in the

neighborhood, showing a response at the scale of

individual plants. Frequency of visits by pulp mashers

decreased with increasing availability of fruits of other

species, which might reflect greater preference for

epiphyte species such as Phroradendron spp., more

than of a perception of fruit availability at a wider

spatial scale as compared with seed dispersers.

Nonetheless, this study shows that even slight differ-

ences in frugivore response can be shown at very small

spatial scales. The weak evidence for interactions

among plants at a neighborhood scale implies that,

from the plant’s perspective, the result of interactions

with dispersers could depend mostly on its own

attributes, such as crop size, or characteristics associ-

ated with fruit removal which can be susceptible to

selection. Although crop size could influence plant

fitness, it may not influence the success of the plant’s

reproductive investment, because the relative fruit

removal (a reflection of dispersal efficiency, Izhaki

2002; Blendinger et al. 2008) by legitimate seed

dispersers did not change with crop size.

Acknowledgments The authors thank the members of the

Laboratorio de Ecologıa de Aves of the Instituto de Ecologıa

Regional (IER) and collaborators of the Project ‘‘Local spatial

structure of the interaction between plants and frugivorous birds

in the southern Yungas’’ for their help in the quantification of the

abundance of fruits dispersed by birds. Ricardo Grau allowed us

to use the permanent tree plot where this study was conducted.

Our colleagues at IER and University of Missouri—St. Louis

contributed with their comments to improve a first version of the

manuscript—the authors especially thank Bette Loiselle, Eliot

Miller, Roman Ruggera, and John Blake. This study complies

with the current laws of Argentina; permission for conducting

research in Parque Sierra de San Javier was granted by the

Universidad Nacional de Tucuman. The study was partially

funded by PIP 2008-1025 of CONICET, Argentina.

References

Barlow PW (1989) Meristems, metamers and modules and the

development of shoot and root systems. Bot J Linn Soc

100:255–279

Blendinger PG, Giannini NP (2010) Interacciones planta-ani-

mal en la sierra de San Javier. In: Grau HR (ed) Ecologıa

Plant Ecol (2011) 212:889–899 897

123

de una interfase natural – urbana: la sierra de San Javier y

el Gran San Miguel de Tucuman. EDUNT, Tucuman,

pp 93–117

Blendinger PG, Loiselle BA, Blake JG (2008) Crop size, plant

aggregation, and microhabitat type affect fruit removal by

birds from individual melastome plants in the Upper

Amazon. Oecologia 158:273–283

Burns KC (2004) Scale and macroecological patterns in seed

dispersal mutualisms. Glob Ecol Biogeogr 13:289–293

Carlo TA (2005) Interspecific neighbors change seed dispersal

pattern of an avian-dispersed plant. Ecology

86:2440–2449

Carlo TA, Morales JM (2008) Inequalities in fruit-removal and

seed dispersal: consequences of bird behaviour, neigh-

bourhood density and landscape aggregation. J Ecol

96:609–618

Carlo TA, Aukema JE, Morales JM (2007) Plant-frugivore

interactions as spatially explicit networks: integrating

frugivore foraging with fruiting plant spatial patterns. In:

Dennis AJ, Schupp EW, Green RJ, Westcott DA (eds)

Seed dispersal: theory and its application in a changing

world. CABI, Wallingford, pp 369–390

Cazetta E, Schaefer HM, Galetti M (2008) Does attraction to

frugivores or defense against pathogens shape fruit pulp

composition? Oecologia 155:277–286

Davidar P, Morton MS (1986) The relationship between fruit

crop sizes and fruit removal rates by birds. Ecology

67:262–265

Foster MS (1990) Factors influencing bird foraging preferences

among conspecific fruit trees. Condor 92:844–854

Garcıa D, Ortiz-Pulido R (2004) Patterns of resource tracking by

avian frugivores at multiple spatial scales: two case studies

on discordance among scales. Ecography 27:187–196

Giannini NP (1999) La interaccion de aves-murcielagos-plan-

tas en el sistema de frugivorıa y dispersion de semillas en

San Javier, Tucuman, Argentina. Disertacion doctoral,

Universidad Nacional de Tucuman

Grau HR (2002) Scale-dependent relationships between tree-

falls and species richness in a neotropical montane forest.

Ecology 83:2591–2601

Hampe A (2008) Fruit tracking, frugivore satiation, and their

consequences for seed dispersal. Oecologia 156:137–145

Herrera CM, Jordano P, Lopez-Soria L, Amat JA (1994)

Recruitment of a mast-fruiting, bird-dispersed tree:

bridging frugivore activity and seedling establishment.

Ecol Monogr 64:315–344

Howe HF, Estabrook GF (1977) Intraspecific competition for

avian dispersers in tropical trees. Am Nat 111:817–832

Howe HF, Smallwood J (1982) Ecology of seed dispersal.

Annu Rev Ecol Syst 13:201–228

Izhaki I (2002) The role of fruit traits in determining fruit

removal in east Mediterranean ecosystems. In: Levey DJ,

Silva WR, Galetti M (eds) Seed dispersal and frugivory:

ecology, evolution and conservation. CABI, Cambridge,

pp 161–175

Janzen DH (1970) Herbivores and the number of tree species in

tropical forests. Am Nat 104:501–528

Jordano P, Schupp EW (2000) Seed disperser effectiveness: the

quantity component and patterns of seed rain for Prunusmahaleb. Ecol Monogr 70:591–615

Kwit C et al (2004) Fruit abundance and local distribution of

wintering Hermit Thrushes (Catharus guttatus) and Yel-

low-rumped Warblers (Dendroica coronata) in South

Carolina. Auk 121:46–57

Leal IR, Oliveira PS (1998) Interactions between fungus-

growing ants (Attini), fruits and seeds in Cerrado vege-

tation in southeast Brazil. Biotropica 30:170–178

Levey DJ, Moermond TC, Denslow JS (1984) Fruit choice in

neotropical birds: the effect of distance between fruits on

preference patterns. Ecology 65:844–850

Levey DJ, Moermond TC, Denslow JS (1994) Frugivory: an

overview. In: McDade LA, Bawa KS, Hespenheide HA,

Hartshorn GS (eds) La Selva: ecology and natural history

of a neotropical rain forest. University of Chicago Press,

Chicago, pp 282–294

Levin SA, Muller-Landau HC, Nathan R, Chave J (2003) The

ecology and evolution of seed dispersal: a theoretical

perspective. Annu Rev Ecol Evol Syst 34:575–604

Loiselle BA, Blake JG (1993) Spatial distribution of understory

fruit-eating birds and fruiting plants in a neotropical

lowland wet forest. Vegetatio 107(108):177–189

Loiselle BA, Blake JG (1999) Dispersal of melastome seeds by

fruit-eating birds of tropical forest understory. Ecology

80:330–336

Malizia A, Grau HR (2008) Landcape context and microenv-

iroment influences on liana communities within treefall

gaps. J Veg Sci 19:597–604

Manasse RS, Howe HF (1983) Competition for dispersal

agents among tropical trees: influences of neighbors.

Oecologia 59:185–190

Moermond TC, Denslow JS (1983) Fruit choice in neotropical

birds: effects of fruit type and accessibility on selectivity.

J Anim Ecol 52:407–420

Ortiz-Pulido R, Rico-Gray V (2000) The effect of spatio-

temporal variation in understanding the fruit crop size

hypothesis. Oikos 91:523–527

Parciak W (2002) Environmental variation in seed number,

size, and dispersal of a fleshy-fruited plant. Ecology

83:780–793

Rey PJ (1995) Spatio-temporal variation in fruit and frugivo-

rous bird abundance in olive orchards. Ecology

76:1625–1635

Saracco JF, Collazo JA, Groom MJ (2004) How do frugivores

track resources? Insights from spatial analyses of bird

foraging in a tropical forest. Oecologia 139:235–245

Saracco JF, Collazo JA, Groom MJ, Carlo TA (2005) Crop size

and fruit neighborhood effects on bird visitation to fruit-

ing Schefflera morototoni trees in Puerto Rico. Biotropica

37:81–87

Sargent S (1990) Neighborhood effects on fruit removal by

birds: a field experiment with Viburnum dentatum (Cap-

rifoliaceae). Ecology 71:1289–1298

Schaefer HM, Schmidt V, Wesenberg J (2002) Vertical strat-

ification and caloric content of the standing fruit crop in a

tropical lowland forest. Biotropica 34:244–253

Schupp EW (1993) Quantity, quality and the effectiveness of

seed dispersal by animals. Vegetatio 107(108):15–29

Stanton ML (2003) Interacting guilds: moving beyond the

pairwise perspective on mutualisms. Am Nat 162:S10–

S23

898 Plant Ecol (2011) 212:889–899

123

Takahashi K, Kamitani T (2004) Factors affecting seed rain

beneath fleshy-fruited plants. Plant Ecol 174:247–256

Traveset A (1998) Effect of seed passage through vertebrate

frugivore’s guts on germination: a review. Perspect Plant

Ecol Evol Syst 1:151–190

Traveset A, Robertson AW, Rodrıguez-Perez J (2007) A

review of the role of endozoochory on seed germination.

In: Dennis AJ, Schupp EW, Green RJ, Westcott DA (eds)

Seed dispersal: theory and its application in a changing

world. CABI, Wallingford, pp 78–103

von Zeipel H, Eriksson O (2007) Fruit removal in the forest

herb Actaea spicata depends on local context of fruits

sharing the same dispersers. Int J Plant Sci 168:855–860

Wenny DG (2000) Seed dispersal, seed predation, and seedling

recruitment of a neotropical montane tree. Ecol Monogr

70:331–351

Plant Ecol (2011) 212:889–899 899

123