Effects of Native and Non-Native Vertebrate
Mutualists on Plants
CLARE E. ASLAN,∗‡ ERIKA S. ZAVALETA,∗DON CROLL,† AND BERNIE TERSHY†
∗Department of Environmental Studies, University of California-Santa Cruz, Santa Cruz, CA 95064, U.S.A.
†Department of Ecology and Evolutionary Biology, University of California-Santa Cruz, Santa Cruz, CA 95064, U.S.A.
Abstract: Extinctions can leave species without mutualist partners and thus potentially reduce their fitness.
In cases where non-native species function as mutualists, mutualism disruption associated with species’ ex-
tinction may be mitigated. To assess the effectiveness of mutualist species with different origins, we conducted
a meta-analysis in which we compared the effectiveness of pollination and seed-dispersal functions of native
and non-native vertebrates. We used data from 40 studies in which a total of 34 non-native vertebrate mutu-
alists in 20 geographic locations were examined. For each plant species, opportunistic non-native vertebrate
pollinators were generally less effective mutualists than native pollinators. When native mutualists had been
extirpated, however, plant seed set and seedling performance appeared elevated in the presence of non-native
mutualists, although non-native mutualists had a negative overall effect on seed germination. These results
suggest native mutualists may not be easily replaced. In some systems researchers propose taxon substitution
or the deliberate introduction of non-native vertebrate mutualists to reestablish mutualist functions such as
pollination and seed dispersal and to rescue native species from extinction. Our results also suggest that in
places where all native mutualists are extinct, careful taxon substitution may benefit native plants at some
Keywords: islands, meta-analysis, mutualism, non-native species, restoration, taxon substitution
Efectos de Vertebrados Mutualistas Nativos y No Nativos sobre Plantas
Resumen: Las extinciones pueden dejar a especies sin socios mutualistas y por lo tanto potencialmente
reducen su adaptabilidad. En casos en los que especies no nativas funcionan como mutualistas, la disrupci´
del mutualismo asociada con la extinci´
on de la especie puede ser mitigada. Para evaluar la efectividad de
especies mutualistas de or´
ıgenes diferentes, realizamos un meta-an´
alisis en el que comparamos la efectividad
de las funciones de polinizaci´
on y dispersi´
on de semillas por vertebrados nativos y no nativos. Utilizamos
datos de 40 estudios en los que se examin´
o un total de 34 especies de vertebrados mutualistas no nativas en 20
aficas. Para cada especie de planta, los polinizadores vertebrados no nativos oportunistas
generalmente fueron mutualistas menos efectivos que los polinizadores nativos. Sin embargo, cuando los
mutualistas nativos fueron extirpados el funcionamiento de las semillas y pl´
antulas aumentaba en presencia
de mutualistas no nativos, aunque los mutualistas no nativos tuvieron un efecto general negativo sobre la
on de semillas. Estos resultados sugieren que los mutualistas nativos no pueden ser reemplazados
acilmente. En algunos sistemas, los investigadores proponen la sustituci´
on de tax´
on o la introducci´
erada de vertebrados mutualistas no nativos para restablecer las funciones mutualistas como la polinizaci´
y la dispersi´
on de semillas y para rescatar de la extinci´
on a especies nativas. Nuestros resultados tambi´
sugieren que en sitios en los que est´
an extintos todos los mutualistas nativos, la sustituci´
on de tax´
puede beneficiar a las plantas nativas en algunas etapas de su vida.
Palabras Clave: especie no nativa, islas, meta-an´
alisis, mutualismo, restauraci´
on de tax´
Paper submitted August 28, 2011; revised manuscript accepted March 18, 2012.
Conservation Biology, Volume 26, No. 5, 778–789
2012 Society for Conservation Biology
In recent decades, widespread environmental changes
have caused extinction rates to move well beyond normal
background levels (Pimm et al. 1995; Brook et al. 2008;
Barnosky et al. 2011). As species disappear, the ecological
functions they provided may be lost as well, potentially
affecting the taxa with which they interacted (Petchey
& Gaston 2002; Srivastava & Bell 2009; Anderson et al.
2011). Loss of a critical function may increase the prob-
ability of extinction of other species in the community
(Hobbs & Mooney 1998; Luck et al. 2003; Zavaleta et al.
2009). Such cascading effects may arise when lost func-
tions include mutualistic services such as pollination or
seed dispersal (Bond 1994; Rezende et al. 2007; Hansen
et al. 2008).
Because mutualisms provide reciprocal fitness bene-
fits (Foster & Wenseleers 2006) to the species involved
in the relationship, their disruption can reduce the fit-
ness of species (Kiers et al. 2010; Traveset & Richardson
2011). Reproductive declines following extirpation of
mutualists have been observed, for example, for mammal-
dispersed trees in Thailand and central Africa (Brodie
et al. 2009; Vanthomme et al. 2010), bird-dispersed plants
in Australia and Hawaii (Moran et al. 2009; Chimera
& Drake 2010), and bat-pollinated plants in the South
Pacific (Cox et al. 1991). In some systems, non-native
species have been observed acting as pollinators and
seed dispersers for native plants (e.g., Foster & Robinson
2007; Pattemore & Wilcove 2011). Non-native species
have become integrated into many mutualistic networks
(Richardson et al. 2000; Olesen et al. 2002; Traveset &
Richardson 2006), but the degree to which species’ ori-
gins affect their capacity to provide mutualistic functions
is debated (Aizen et al. 2008; Davis et al. 2011; Traveset
& Richardson 2011).
In systems where cascading extinctions resulting from
mutualism disruption appear likely, introduction of non-
native species that are functional analogues of an extir-
pated species has been proposed as a way to restore mu-
tualist processes (Griffiths & Harris 2010; Kaiser-Bunbury
et al. 2010; Parker et al. 2010) and is known as “taxon
substitution” (Atkinson 2001). Recent taxon substitutions
have been implemented under controlled conditions in
Mauritius and in the Gal´
apagos (Griffiths & Harris 2010;
Knafo et al. 2011). However, deliberately introducing
non-native species to a new area has substantial risks.
Invasive species in many regions were originally intro-
duced to provide ecological functions such as erosion
control (Underwood et al. 2007; Pearce & Smith 2008)
and biological control (Howarth 1991; Simberloff & Stil-
ing 1996). Before taxon substitution becomes common
practice, we believe existing patterns of interspecific in-
teractions in invasion biology and community ecology
should be examined to determine under what circum-
stances non-native species become effective mutualists
We conducted a meta-analysis of the ability of non-
native species to provide effective mutualistic services
(pollination or seed dispersal) to native plants. We re-
stricted our analyses to vertebrate-plant mutualists be-
cause in general vertebrates are introduced in taxon sub-
stitutions (Parker et al. 2010; Griffiths et al. 2011; Knafo
et al. 2011) and because declines and extinctions of native
vertebrate mutualists are better-documented than those
of invertebrates (Dunn 2005; IUCN 2011). Vertebrates
are estimated to pollinate 5.6% and disperse the seeds of
well over half of all angiosperm genera (Howe & Small-
wood 1982; Renner & Ricklefs 1995).
We examined seed dispersal and pollination by non-
native vertebrates versus native vertebrates; plant re-
productive success in the presence of non-native seed
dispersers and pollinators and in the absence of seed dis-
persers and pollinators; and the number of species with
which native and non-native mutualists interact.
The species we examined were in most cases intro-
duced accidentally or for purposes other than taxon sub-
stitution. Their mutualistic roles resulted from oppor-
tunistic foraging. Thus, these species may be less well
associated with local plant reproduction than would ver-
tebrates introduced deliberately as seed dispersers or
pollinators. Nevertheless, by synthesizing information on
existing interactions and examining factors underlying ef-
fectiveness of non-native mutualists, our results provide
a baseline that may help prioritize and guide efforts to
We searched ISI Web of Science and Google Scholar
for publications (spanning the years 1899–2011) that
reported non-native vertebrates as reproductive mutu-
alists for native plants. We used 3 groups of search terms:
seed dispersal. Search strings included one word from
each group and all possible strings were used. We used
all strings with and without native.Wesearchedtheref-
erence lists of all relevant papers for additional sources.
All papers included in our meta-analyses reported data
on quantitative mutualism effectiveness, by non-native
vertebrate species, for pollination or seed dispersal. Fur-
thermore, all papers included quantitative information
that allowed us to compare the effectiveness of non-
native mutualists with native mutualists or with the
absence of mutualism. In only 2 quantitative studies
were the effects of non-native pollinators compared with
Volume 26, No. 5, 2012
780 Meta-Analyses of Non-Native Mutualists
reproduction in the absence of all pollinators. Due to this
small sample size, we did not attempt a meta-analysis of
these studies; instead, we summarized the results of these
Metrics of the effectiveness of plant mutualism vary
considerably in the ecological literature (Herrera 1987;
Schupp et al. 2010). Considering each metric in a sepa-
rate meta-analysis would have resulted in a prohibitively
small number of studies that used any given metric. How-
ever, all of the studies we used allowed us to examine
the same metric for within-study comparisons (either the
comparison between non-native and native mutualists or
the comparison between non-native mutualists and mu-
tualist absence). This ensured that our quantitative esti-
mates of effect size were derived only from within-metric
comparisons. On the basis of the results of within-study
comparisons, we used a common effect metric that al-
lowed us to consider various measures of effectiveness
combined. We used the response ratio as a common mea-
sure of effect size (Hedges et al. 1999; Rosenberg et al.
2000). For the analyses of non-native disperser or pol-
linator effectiveness compared with that of natives, the
response ratio Rwas calculated as
XEis the weighted mean of a specific component
of non-native mutualist effectiveness (e.g., average pollen
load on captured pollinators) and ¯
XCis the weighted
mean of the identical component of native mutualist ef-
fectiveness. In our comparison of the effectiveness of
the presence and absence of non-native dispersers, the
response ratio was the natural log of the ratio of the
weighted mean of a given component of non-native mutu-
alist effectiveness to the same component in the absence
To assess the relative contributions of each effective-
ness metric to overall meta-analysis results, we used cate-
gorical analyses to generate separate meta-analysis results
for each effectiveness metric with a sample size of ≥2
in our data (Rosenberg et al. 2000). Other variables for
which we used categorical analyses to explore underly-
ing structure in our data included non-native mutualist
class (mammal vs. bird), land type (island vs. continen-
tal), and taxonomic match (i.e., whether the non-native
mutualist was in the same taxonomic class as the extinct
or extant native mutualist). All pollination records were
for non-native species in the same taxonomic class as
the natives they were compared with, so this categor-
ical variable was examined only for dispersal records.
When a study reported mutualism effectiveness for mul-
tiple native vertebrate species interacting with the same
plant species, we used the mean, weighted by sample
size, of those reports to generate the denominator of the
For approximately one-fifth of the records in our data
set, the only metric reported was the diversity of the
species of fruiting or flowering plants visited by the non-
native mutualist. Although this metric does not provide
information relevant to the effectiveness of the mutualism
between these animals and each plant species, it provides
information on how broadly their effects as introduced
mutualists may propagate across the ecosystem. In cases
of deliberate mutualist introduction, this diffuseness of
effect is a particularly important component of the deci-
sion to introduce such a species. Therefore, we included
diffuseness in an additional meta-analysis in which we
compared the number of plant species with which each
non-native mutualists interacted with the diffuseness of
Because studies with larger sample sizes contribute
more to overall effect-size estimates in meta-analyses, the
reported variances and sample sizes of all included means
are used to generate the overall effect size (Rosenberg
et al. 2000). For a large number of the studies included
in our meta-analyses, however, the variances in effec-
tiveness were not included. This was particularly com-
mon when diffuseness or percentages (e.g., percentage
of all visits recorded to a plant or percentage of seeds
germinated) were reported. These values were gener-
ally provided as direct counts. To allow inclusion of the
results of these studies in our analyses, we performed
unweighted meta-analyses (after Johnson & Curtis 2001).
This method assigns a common weight (1) to all studies
and bases calculations on sample sizes and mean effect
sizes. However, certain species of non-native vertebrate
mutualists occurred in multiple records because they in-
teracted with multiple native plant species within a given
study or because they appeared in multiple studies. To
control for this potential bias, we conducted the overall
meta-analyses at both the record and the species levels
(Roberts et al. 2004). For species-level analyses, we calcu-
lated a single effect size for each species as the weighted
average of all effect sizes for that species in the data.
This reduced the number of records per meta-analysis
and hence the analyses’ power, but enabled examina-
tion of overall trends without potential bias toward more
We calculated effect sizes, total heterogeneity or un-
derlying data structure (QT), and between-group hetero-
geneity (QB) with a random-effects model in MetaWin 2.0
(Rosenberg et al. 2000). We evaluated significance of het-
erogeneity and significant differences between mean ef-
fect sizes for explanatory variables by calculating Bonfer-
roni corrected pvalues. We generated 95% bias-corrected
bootstrap confidence intervals around mean effect sizes
by resampling with 5000 data randomizations because
data randomization provides robust results when data
are not normally distributed and this method is recom-
mended for meta-analyses (Rosenberg et al. 2000). For
ease of interpretation, we back-transformed all means
Volume 26, No. 5, 2012
and confidence intervals to ratios of non-native to native
mutualist effectiveness. After back-transformation, effect-
size means and confidence intervals <1 signified that the
effect of non-native mutualists was significantly less than
that of native mutualists or mutualist absence, whereas
effect-size means and confidence intervals >1 signified
that the effect of non-native mutualists was significantly
greater than native mutualists or mutualist absence. Con-
fidence intervals encompassing 1 indicated there was no
significant difference between non-native mutualists and
native mutualists or mutualist absence.
In addition, we used Rosenberg’s (2005) fail-safe cal-
culations to determine the number of additional studies
with nonsignificant results that would be required to re-
duce the overall significance of each of our meta-analyses
to α=0.05. The fail-safe calculation addresses the po-
tential of a file-drawer problem in a meta-analysis (i.e.,
nonsignificant results unreported and therefore the ef-
fect size may be artificially inflated by considering only
published studies) (Rosenberg et al. 2000). A Rosenthal
fail-safe calculation result of at least 5n+10 (where
nis the original number of records) is an indicator of
reliable meta-analysis results, whereas smaller fail-safe
numbers imply that meta-analyses results are less robust
To consider the environmental effects associated with
non-native mutualist introductions, we examined the lit-
erature for records of negative environmental effects as-
sociated with each of the most common non-native mu-
tualists in our data set (those appearing in >3 studies).
We conducted our literature search in Google Scholar
and Web of Science (1899–2011) and used as search
strings the full scientific names of each non-native mu-
We found 40 publications with enough quantitative in-
formation to include in our meta-analyses (Supporting
Information). These studies examined 34 unique non-
native vertebrates acting as pollinators or seed dispersers
(Supporting Information). Our data contained 74 unique
records, and each non-native species in each study was
a separate record. Pollination mutualisms were repre-
sented by 18 records from 9 studies, and 56 records from
31 studies concerned seed-dispersal mutualisms. Our data
set included 1 non-native reptile, 15 non-native mammals,
and 18 non-native birds (Supporting Information). Stud-
ies we included had been conducted in 20 geographic ar-
eas, 11 of which were islands (Fig. 1). In all, 13 different
metrics of mutualism effectiveness were used, including
5 pretransport metrics (quantitative components of pol-
lination and seed dispersal before propagule transport,
for example, number of visits), 7 post-transport metrics
(quantitative components of pollination and seed disper-
sal following propagule transport, for example, number
of seeds deposited), and number of plants interacting
with pollinators or seed dispersers (Supporting Infor-
mation). One study (Griffiths et al. 2011) reported re-
sults of a deliberate taxon substitution (giant tortoises
Effectiveness of Non-Native Versus Native Mutualists
When the full data set was examined, the overall effective-
ness of non-native dispersers was not significantly differ-
ent from that of native dispersers (Fig. 2a). Heterogeneity
of the data was low (QT=47.67, df =56, p=0.78), which
indicates little structure in the data set. After Bonfer-
roni correction, non-native mutualists were significantly
less effective than native mutualists when the taxonomic
class of the mutualists differed (either bird or mammal)
(Fig. 2a). However, native and non-native mutualists in
the same taxonomic group did not differ significantly in
effectiveness (Fig. 2a). Effect sizes did not differ signifi-
cantly between non-native taxonomic groups (Qb=0.07;
df =1; between-group chi-square p=0.81) or by land
type after Bonferroni correction (island vs. continental)
(Qb=5.89; df =1; between-group chi-square p=0.03).
However, individual effectiveness metrics differed signif-
icantly in pattern (Qb=58.98; df =6; between-group
chi-square p=0.01) (Table 1).
As a group, non-native pollinators were significantly
less effective than native pollinators (Fig. 2b). Hetero-
geneity in this meta-analysis was significant (QT=43.11,
df =15, p<0.001). Fifteen of the records came from
either Hawaii or New Zealand. These 2 geographic re-
gions had different effectiveness trends. Non-native polli-
nators in Hawaii were more effective than native pollina-
tors (Fig. 2b), whereas native pollinators in New Zealand
were more effective than non-native pollinators (Fig. 2b).
Effectiveness among individual metrics did not differ sig-
nificantly (Qb=10.26; df =1; between-group chi-square
p=0.06) (Table 1).
Non-native Mutualists in the Absence of Native Mutualists
When the full data set was examined, there was no signif-
icant difference in plant reproductive metrics between
non-native dispersers and absence of dispersers (mean
effectiveness ratio =0.73; 95% bias-corrected CI 0.57
to 1.02, n=27). However, heterogeneity was signifi-
cant (QT=66.73, df =26, p<0.001), which high-
lights the widely divergent patterns of effectiveness typ-
ical of studies in which measures of seed germination
and seedling survival or growth rate are used to evaluate
effects of mutualisms. In the absence of native seed dis-
persers, passage through the gut of non-native dispersers
had a significant and negative effect on seed germination
(Fig. 2c) compared with seeds that had not experienced
gut passage. By contrast, although the sample size was
Volume 26, No. 5, 2012
782 Meta-Analyses of Non-Native Mutualists
Figure 1. Geographic locations of
published studies identifying
non-native vertebrate mutualists
of native plants. Numbers of
non-native taxa indicate number
of unique species examined in
small, non-native seed dispersers had significant and pos-
itive effects on seedling survival and growth rate follow-
ing dispersal (Table 1 & Fig. 2c). When the taxonomic
group of the native mutualist and the non-native mutu-
alist differed, seed germination was significantly lower
than when mutualism was absent. When the native mu-
tualist and the non-native mutualist belonged to the same
taxonomic group, however, there was no significant dif-
ference between seed germination or seedling survival
in the absence of mutualism and the same metrics in
the presence of non-native mutualists (Fig. 2c). Effect
sizes also differed by land type. On islands germination
was significantly reduced in the presence of non-native
seed dispersers (mean effectiveness ratio =0.58; 95%
bias-corrected CI 0.47 to 0.74, n=15), whereas on con-
tinents germination was enhanced significantly by non-
native seed dispersers (mean effectiveness ratio =2.42;
95% bias-corrected CI 1.27 to 6.29, n=10).
Two studies examined non-native vertebrate polli-
nators in the absence of other pollinators. The only
pollinator bird species visiting the Hawaiian endemic
Freycinetia arborea, whose known native pollinators
are extinct, was the non-native Japanese White-eye
(Zosterops japonica) (Cox 1983). Flowers exposed to
birds set fruit, whereas those that were not exposed to
birds did not, which suggests Z. japonica is the major
pollinator of F. arborea (Cox 1983). On the basis of
seed set and flower damage, Lord (1991) concluded that
New Zealand’s kiekie (Freycinetia baueriana) is likely
pollinated largely by the non-native brush-tailed possum
(Trichosurus vulpecula) in the absence of its native pol-
linator, New Zealand lesser short-tailed bat (Mystacina
Diffuseness and Fail-Safe Analyses
Non-native mutualists visited significantly fewer na-
tive plants than native mutualists (mean effectiveness
ratio =0.35; 95% bias-corrected CI 0.25–0.49, n=27).
This pattern held for both pollinators and dispersers (95%
Volume 26, No. 5, 2012
Figure 2. Mean effectiveness ratios (filled circles) and
bias-corrected confidence intervals (vertical lines) for
studies comparing (a) seed dispersal by non-native
mutualists with seed dispersal by native mutualists
(categorized by whether non-native and native
mutualists were in the same or different taxonomic
group), (b) pollination by non-native mutualists with
bias-corrected CI 0.15–0.26, n=5 and 0.29–0.63 n=22,
respectively). All available records were for islands and
Fail-safe analyses indicated there was a potential for
some of our results to be biased by a tendency to not
report nonsignificant results. For pollinators only 48 ad-
ditional, nonsignificant studies would be necessary to
shift the results to a nonsignificant difference between
native and non-native pollinators. In the diffuseness meta-
analysis, 894 additional nonsignificant studies would be
necessary to shift the results to a nonsignificant differ-
ence between non-native and absent dispersers.
Repeated Occurrences of Non-Native Species
When we used mean effect sizes for each species so
that no species appeared more than once in each meta-
analysis, results differed qualitatively from the results of
the original meta-analyses. The overall comparison of
non-native pollinators with native pollinators, adjusted
for multiple records per species, showed no significant
difference between the 2 groups (mean effectiveness ra-
tio =0.42; 95% bias-corrected CI 0.22 to 1.11, n=6).
Overall, plant dispersal facilitated by non-native species
and native species, adjusted for multiple records per
species, did not differ significantly (mean effectiveness
ratio =0.50; 95% bias-corrected CI 0.30 to 1.03, n=
26). Overall, seed germination and seedling survival and
growth, adjusted for multiple records per species, were
significantly lower when non-native species dispersed
seeds than when dispersers were absent (mean effective-
ness ratio =0.69; 95% bias-corrected CI 0.53 to 0.99, n=
16). Overall comparisons of diffuseness, adjusted for mul-
tiple records per species, showed that non-native species
visited significantly more plant partners than native mutu-
alists (mean effectiveness ratio =1.12; 95% bias-corrected
CI 1.08 to 1.33, n=13).
Ecological Effects of Non-Native Seed Dispersers
Six non-native mutualists appeared more than 3 times
each in our data set (Table 2). Negative ecological effects
were reported in the peer-reviewed literature for 5 of
pollination by native mutualists (categorized by
geographic area), and (c) seed germination after
ingestion by non-native mutualists with germination
in the absence of mutualists (categorized by mutualist
taxonomic group). Effect sizes are significant if their
confidence intervals do not include one. Significant
effect sizes <1 indicate non-native mutualists are
significantly less effective than native mutualists.
Numbers above each confidence interval are sample
Volume 26, No. 5, 2012
784 Meta-Analyses of Non-Native Mutualists
Table 1. Mean effectiveness ratios and bias-corrected conﬁdence intervals for individual metrics of effectiveness of pollination and seed dispersal
in meta-analyses of the effectiveness of native versus non-native mutualists.
Meta-analysis comparison Effectiveness metric (n) Effectiveness ratio Bias-corrected CI
Non-native and native dispersers percent seed germination (16) 0.45 0.31–0.78
no. seeds dispersed/weight of feces (16) 0.41 0.12–1.33
time to germination (3) 1.82 1.11–2.33
no. seeds/individual/day (8) 2.23 0.77–5.39
percent samples with seeds (4) 4.21 1.18–10.30
survival of defecated seeds (2) 1.94 1.25–2.95
percent plants visited (7) 0.13 0.07–0.26
Non-native and native pollinators visits/plant/hour (10) 0.84 0.27–2.03
percent plants visited (5) 0.18 0.10–0.35
No dispersal and non-native dispersers percent seed germination (25) 0.65 0.52–0.85
seedling performance (2) 1.62 1.58–1.63
these 6 generalist species (Table 2). The only exception
was the Chaffinch (Fringilla coelebs),whichactsasa
seed disperser and pollinator of native plants in New
Zealand (Kelly et al. 2006).
Negative effects of introduced species are well-known
(e.g., Traveset & Richardson 2006; Richardson & Pyˇ
2008). In recent years, a few articles have highlighted the
positive effects of non-native species on native species
(e.g., Goodenough 2010; Griffiths et al. 2011; Schlaepfer
et al. 2011). The results of these studies underline a grow-
ing awareness of the multifaceted nature of species re-
assembly driven by species introductions. Overall, we
found that non-native vertebrate pollinators already inter-
acting with native plants were generally less effective pol-
linators than are native vertebrates, although non-native
vertebrate seed dispersers, as a group, did not appear to
differ in effectiveness from native seed dispersers. These
results strongly support prioritizing the conservation of
native vertebrate pollinators because their functions are
unlikely to be replaced by non-native species (Traveset
& Richardson 2011).
Although it is difficult to derive firm conclusions from
our categorical analyses of individual metrics of effec-
tiveness because the sample sizes of each metric were
small, differences in mutualist performance were ap-
parent. With native seed dispersers, seed germination
was greater and faster than with non-native dispersers,
whereas non-native dispersers defecated more seeds and
these seeds demonstrated higher survival. Among both
seed dispersers and pollinators, native mutualists were
associated with higher proportions of visits to individual
native plants, which suggests they may be prone to
greater partner fidelity or to specialize more closely on
individual native plants.
Table 2. Negative effects associated with generalist non-native mutualists.
Chaffinch (Fringilla coelebs) none reported in peer-reviewed literature
Common Mynah (Acridotheres
competition with native fauna
predation of native species
dispersal of invasive non-native plants
wildlife disease or parasite transmission
Peacock et al. 2007
European Blackbird (Turdus merula) dispersal of invasive plants
wildlife disease transmission
Williams 2006; Tompkins &
European rabbit (Oryctolagus
herbivory on native plants
competition with native fauna
increase in abundance of regional predators resulting in
hyperpredation of native prey
Jaksic 1998; Lees & Bell 2008
European Starling (Sturnus vulgaris) dispersal of non-native plants
possible competition with natives
disease transmission to wild animals
Aslan 2011; Koenig 2003; Linz
et al. 2007
Japanese White-eye (Zosterops
dispersal of non-native plants Vitousek & Walker 1989
∗Each of these species occurred more than 3 times in our data set and interacted with a variety of native plants in a variety of geographic
Volume 26, No. 5, 2012
Specialization and generalization may explain several
trends in our results. Overall, non-native mutualist species
tend not to have specialized geographic distributions,
behaviors, or physical features, such as curved bills (McK-
inney & Lockwood 1999; Traveset & Richardson 2011;
Westcott & Fletcher 2011). For example, our data in-
cluded livestock and introduced wild species strongly
associated with humans (e.g., black rats [Rattus rattus]
and House Sparrows [Passer domesticus]). Such species
have been introduced in many regions and have been
successful. They also have highly diverse diets and can
thrive in a variety of land-cover types (Ehrlich 1989). Mu-
tualisms between these species and native plants are op-
portunistic and transitory (Richardson et al. 2000; Aslan
anek 2010). This lack of specialization is likely
responsible for reduced effectiveness demonstrated by
non-native pollinators relative to native mutualists. Spe-
cialization in mutualisms enhances the efficiency of these
relations while diminishing the total number of species in-
volved (Schemske & Horvitz 1984). It seems highly possi-
ble that ecologically specialized species that are carefully
selected for taxon substitution may be more effective
pollinators and seed dispersers than generalists.
Careful selection of species for introduction and their
deliberate introduction is most likely when all native mu-
tualists are extinct. However, the relation between re-
productive success of plants in the presence of known
non-native mutualists and in the absence of mutualisms is
unclear. The large majority of studies that could be used
to compare reproduction in the absence of mutualisms
with reproduction via non-native mutualists focused on
percent seed germination, which is strongly and signifi-
cantly lower after passage through the gut of non-native
mutualists. Results of another meta-analysis showed that
after a seed passes through the gut of an animal seed
germination of half of all plant species is affected by en-
dozoochory (gut passage), and germination is reduced for
one-third of the species affected by endozoochory (Trav-
eset 1998). In our study, because native mutualists in the
relevant studies were extinct, it was impossible to know
whether seeds that passed through the gut of native mutu-
alists would also have reduced germination. Furthermore,
in most studies that we could use in our analyses, seeds
passing through non-native dispersers were compared
with seeds from which the flesh was removed by hand.
This action confounds the effects of pericarp removal and
gut passage. Assessment of germination of whole fruits
is necessary to fully understand the effect of non-native
dispersers on germination rates (Samuels & Levey 2005).
Most non-native dispersers in our analyses (21 out of 25
records) were mammals. The digestive tracts of nonflying
mammals may be more likely to damage seeds than the
digestive tracts of birds (Traveset 1998; but see Verd´
Traveset 2004). Because seeds represent unusable ballast
with the potential to interfere with flight, it is advanta-
geous for birds to defecate seeds rapidly (Sorensen 1984;
Levey & Grajal 1991) and thus disperse largely undam-
aged seeds over a more limited distance. Many mammals
can, by contrast, disperse a very large number of seeds
over very long distances (Willson 1993). Dispersal that re-
moves seeds from their parental neighborhood enhances
their ability to escape natural enemies (Janzen 1970; Con-
nell 1971) and on average generates elevated seedling
survival (Harms et al. 2000; Wotton & Kelly 2011). With
their contrasting dispersal distances and effects on germi-
nation, birds and mammals as seed dispersers may have
different effects on fitness of plants (Howe 1990).
The results of the few studies we found in which mu-
tualism effectiveness in the absence of native mutualists
was examined with metrics other than percent seed ger-
mination (seedling survival, seedling growth, and seed
set) showed that reproductive success for plants part-
nering with non-native species was significantly greater
than when mutualists were absent, but it is not yet pos-
sible to examine global patterns. Germination may have
little to do with mature plant populations, especially be-
cause many long-lived plants are insensitive to variations
in seed survival (Silvertown et al. 1993; but see Brodie
et al. 2009). If our results apply across more systems
than we examined, native plants partnering with non-
natives may have relatively higher seed set and seedling
performance but relatively lower percent germination.
The contribution of these demographic components to
the overall population trajectory of a given plant species
depends on species-specific demography (Morris & Doak
When non-native seed dispersers were from a differ-
ent taxonomic group than extinct or extant native seed
dispersers, their mutualist effectiveness was significantly
less than that of native seed dispersers. However, there
was no significant effectiveness difference when non-
native seed dispersers were from the same taxonomic
group as extinct or extant native seed dispersers. These
results suggest that taxonomic matching may be impor-
tant when evaluating non-native species as potential seed
The Role of Islands
Non-native seed dispersers exerted a positive effect on
seed germination in continental ecosystems. In contrast,
compared with absence of dispersers, non-native seed
dispersers had a negative effect on seed germination on
islands, where many species are specialists. It is unsur-
prising that introduced, generalist species are unable to
match the mutualism effectiveness of remnant or extinct
specialists. All native species we considered specialists
(narrow dietary range, curved or large bills, and, among
frugivorous birds, large body size) are island species
(e.g., Hawaiian honeycreepers [Drepanidinae], New
Zealand kereru [Hemiphaga novaeseelandiae], and
Yellow-crowned Parakeet [Cyanoramphus auriceps]).
Volume 26, No. 5, 2012
786 Meta-Analyses of Non-Native Mutualists
By contrast, both island and continental records con-
tained generalist species (e.g., Japanese White-eye, gua-
naco [Lama guanicoe], and Baird’s tapir [Tapirus
bairdii]). In our data, then, random losses of native
vertebrate species from islands may remove specialists,
whereas losses on continents may not. Among pollina-
tion records, introduced pollinators in New Zealand were
less effective than native pollinators, whereas non-native
species in Hawaii were more effective pollinators than
natives. This again likely reflects the difference between
generalist and specialist species. Remnant native birds in
Hawaii do not include many of the highly specialized pol-
linators that once visited Hawaiian flowers (Cox 1983).
Both native and non-native birds in Hawaii today have less
specialized morphology and behavior than was likely the
case in the past (Freed et al. 1987).
Results of our fail-safe calculation suggested that diffuse-
ness analysis results were robust, but that the significant
difference in effectiveness between native and non-native
pollinators may be biased by a potential tendency not to
report nonsignificant results. The overall effectiveness
difference we found may be an artifact of this publishing
bias. This would imply that non-native pollinators may be
better substitutes than our results show.
Meta-analysis at the species (rather than study) level re-
moved the significant difference between native and non-
native pollinators. However, it introduced a significant
difference between presence of non-native dispersers
and absence of dispersers that increased the importance
of reduced percent seed germination to the effective-
ness of seed dispersal. Furthermore, the species-level
meta-analysis reversed the results of the diffuseness meta-
analysis (which indicated non-native species visited sig-
nificantly more plant species). This suggests that results
with the full set of records, in which native species vis-
ited significantly more plant species than did non-native
species, were driven by a few non-native species that
visited a particularly small number of native plants (e.g.,
Common Mynah [Acridotheres tristis], European Starling
[Sturnus vulgaris], and House Sparrow [Passer domes-
ticus]). These tests highlight the context dependence
of these mutualisms and emphasize the need for careful
study of effectiveness of novel mutualists before taxon
Non-native species can displace or predate native species,
transmit disease, and alter ecological communities (Trav-
eset 1998; Clavero & Garc´
ıa-Berthou 2005; Traveset &
Richardson 2006). Negative effects on native species pop-
ulations or diversity have been attributed to all but one
of the most generalist non-natives in our study.
When all native pollinators or seed dispersers have be-
come extinct, non-native vertebrates may enhance some
native-plant fitness components, such as seed set and
seedling survival. Their current rarity makes effective
restoration management challenging.
As conservation practitioners seek new and innova-
tive ways to conserve biological diversity, introduction
of non-native mutualists to augment functional diversity
may become a common conservation strategy, much like
biological control (Hansen et al. 2010; Parker et al. 2010;
Seddon 2010). We believe the risks of such introductions
need to be minimized. Invasion biology is sufficiently
well-developed to allow development and rigorous exam-
ination of a catalogue of likely risks before introduction
of mutualist species. We suggest that deliberate introduc-
tions of mutualists be considered only where all native
mutualists have been extirpated. Under these conditions,
a carefully selected ecological analogue species might be
able to provide the mutualistic functions that have been
lost (Parker et al. 2010) if thorough screening and rigor-
ous monitoring are applied and demographic studies of
plants suggest that the analogue can facilitate ecosystem
restoration and conservation of biological diversity. Our
results pinpoint factors, such as taxonomic group, taxo-
nomic matching, and specialization, that may influence
the effectiveness of mutualist introductions and that may
be considered when managers are determining whether
to employ taxon substitution.
Control and eradication of non-native species are
overwhelmingly beneficial to global biological diversity
anek & Pitcairn 2002; Howald et al. 2007), and
novel mutualisms could complicate such conservation
measures. As non-native species form mutualistic rela-
tions with native species, the non-native species may in
some instances enhance site-specific species diversity. In
such cases, eradication of the non-natives could threaten
native species that depend on them.
We thank D. Richardson and 2 anonymous reviewers for
feedback on this manuscript. C.E.A. was supported by a
David H. Smith Postdoctoral Fellowship.
The full lists of studies included in our meta-analyses (Ap-
pendix S1), non-native species appearing in our database
(Appendix S2), and effectiveness metrics (Appendix S3)
are available online. The authors are solely responsible for
the content and functionality of these materials. Queries
(other than absence of the material) should be directed
to the corresponding author.
Volume 26, No. 5, 2012
Aizen, M. A., C. L. Morales, and J. M. Morales. 2008. Invasive mutualists
erode native pollination webs. Public Library of Science Biology 6
Anderson, S. H., D. Kelly, J. J. Ladley, S. Molloy, and J. Terry. 2011.
Cascading effects of bird functional extinction reduce pollination
and plant density. Science 331:1068–1071.
Aslan, C. E. 2011. Implications of newly-formed seed-dispersal mutu-
alisms between birds and introduced plants in northern California,
USA. Biological Invasions 13:2829–2845.
Aslan, C. E., and M. Rejm´
anek. 2010. Avian use of introduced plants:
ornithologist records illuminate interspecific associations and re-
search needs. Ecological Applications 20:1005–1020.
Atkinson, I. A. E. 2001. Introduced mammals and models for restoration.
Biological Conservation 99:81–96.
Barnosky, A. D., et al. 2011. Has the Earth’s sixth mass extinction already
arrived? Nature 471:51–57.
Bond, W. J. 1994. Do mutualisms matter? Assessing the impact of pol-
linator and dispersal disruption on plant extinction. Philosophical
Transactions of the Royal Society London B 344:83–90.
Brodie, J. F., O. E. Helmy, W. Y. Brockelman, and J. L. Maron. 2009.
Bushmeat poaching reduces the seed dispersal and population
growth rate of a mammal-dispersed tree. Ecological Applications
Brook, B. W., N. S. Sodhi, and C. A. J. Bradshaw. 2008. Synergies among
extinction drivers under global change. Trends in Ecology & Evolu-
Chimera, C. G., and D. R. Drake. 2010. Patterns of seed dispersal and
dispersal failure in a Hawaiian dry forest having only introduced
birds. Biotropica 42:493–502.
Clavero, M., and E. Garc´
ıa-Berthou. 2005. Invasive species are a leading
cause of animal extinctions. Trends in Ecology & Evolution 20:
Connell, J. H. 1971. On the role of natural enemies in preventing com-
petitive exclusion in some marine mammals and in rainforest trees.
Pages 298–312 in P. J. den Boer and G. R. Gradwell, editors. Dy-
namics of populations. Centre for Agricultural Publishing and Doc-
umentation, Wageningen, The Netherlands.
Cox, P. A. 1983. Extinction of the Hawaiian avifauna resulted in a change
of pollinators for the ieie, Freycinetia arborea.Oikos41:195–
Cox, P. A., T. Elmqvist, E. D. Pierson, and W. E. Rainey. 1991. Flying
foxes as strong interactors in South Pacific Island ecosystems: a
conservation hypothesis. Conservation Biology 5:448–454.
Davis, M. A., et al. 2011. Don’t judge species on their origins. Nature
Dunn, R. R. 2005. Modern insect extinctions, the neglected majority.
Conservation Biology 19:1030–1036.
Ehrlich, P. R. 1989. Attributes of invaders and the invading processes:
vertebrates. Pages 315–338 in D. R. Drake, H. A. Mooney, F. di Cas-
tri, R. H. Groves, F. J. Kruger, and M. Williamson, editors. Biologi-
cal invasions: a global perspective. John Wiley, Chichester, United
Foster, J. T., and S. K. Robinson. 2007. Introduced birds and
the fate of Hawaiian rainforests. Conservation Biology 21:
Foster, K. R., and T. Wenseleers. 2006. A general model for the evo-
lution of mutualisms. Journal of Evolutionary Biology 19:1283–
Freed, L. A., S. Conant, and R. C. Fleischer. 1987. Evolutionary ecology
and radiation of Hawaiian passerine birds. Trends in Ecology &
Goodenough, A. E. 2010. Are the ecological impacts of alien species
misrepresented? A review of the “native good, alien bad” philoso-
phy. Community Ecology 11:13–21.
Griffiths, C. J., and S. Harris. 2010. Prevention of secondary extinctions
through taxon substitution. Conservation Biology 24:645–646.
Griffiths, C. J., D. M. Hansen, C. G. Jones, N. Zu¨
el, and S. Harris. 2011.
Resurrecting extinct interactions with extant substitutes. Current
Hansen, D. M., C. J. Donlan, C. J. Griffiths, and K. J. Campbell. 2010.
Ecological history and latent conservation potential: large and giant
tortoises as a model for taxon substitutions. Ecography 33:272–284.
Hansen, D. M., C. N. Kaiser, and C. B. M¨
uller. 2008. Seed dispersal and
establishment of endangered plants on oceanic islands: the Janzen-
Connell model, and the use of ecological analogues. Public Library
of Science ONE 3DOI: 10.1371/journal.pone.0002111.
Harms, K. E., S. J. Wright, O. Calder´
on, A. Hern´
andez, and E. A. Herre.
2000. Pervasive density-dependent recruitment enhances seedling
diversity in a tropical forest. Nature 404:493–495.
Hedges, L. V., J. Gurevitch, and P. S. Curtis. 1999. The meta-analysis of
response ratios in experimental ecology. Ecology 80:1150–1156.
Herrera, C. M. 1987. Components of pollinator “quality”: comparative
analysis of a diverse insect assemblage. Oikos 50:79–90.
Hobbs, R. J., and H. A. Mooney. 1998. Broadening the extinction de-
bate: population deletions and additions in California and western
Australia. Conservation Biology 12:271–283.
Howald, G. R., et al. 2007. Invasive rodent eradication on islands. Con-
servation Biology 21:1258–1268.
Howarth, F. G. 1991. Environmental impacts of classical biological con-
trol. Annual Review of Entomology 36:485–509.
Howe, H. F. 1990. Seed dispersal by birds and mammals: implica-
tions for seedling demography. Pages 191–218 in K. S. Bawa and
M. Hadley, editors. Reproductive ecology of tropical forest plants.
UNESCO/Parthenon Publishing Group, Paris.
Howe, H. F., and J. Smallwood. 1982. Ecology of seed dispersal. Annual
Review of Ecology and Systematics 13:201–228.
IUCN (International Union for Conservation of Nature). 2011. IUCN red
list of threatened species. Version 2011.2. IUCN, Gland, Switzer-
land. Available from http://www.iucnredlist.org (accessed Novem-
Jaksic, F. M. 1998. Vertebrate invaders and their ecological impacts in
Chile. Biodiversity and Conservation 7:1427–1445.
Janzen, D. H. 1970. Herbivores and the number of tree species in trop-
ical forests. The American Naturalist 104:501–528.
Johnson, D. W., and P. S. Curtis. 2001. Effects of forest management on
soil C and N storage: meta-analysis. Forest Ecology and Management
Kaiser-Bunbury, C. N., A. Traveset, and D. M. Hansen. 2010. Con-
servation and restoration of plant-animal mutualisms on oceanic
islands. Perspectives in Plant Ecology, Evolution and Systematics
Kelly, D., A. W. Robertson, J. J. Ladley, S. H. Anderson, and R. J. McKen-
zie. 2006. Relative (un)importance of introduced animals as pollina-
tors and dispersers of native plants. Pages 227–245 in R. B. Allen and
W. G. Lee, editors. Biological Invasions in New Zealand. Springer,
Kiers, E. T., T. M. Palmer, A. R. Ives, J. F. Bruno, and J. L. Bronstein.
2010. Mutualisms in a changing world: an evolutionary perspective.
Ecology Letters 13:1459–1474.
Knafo, S. E., S. J. Divers, S. Rivera, L. J. Cayot, W. Tapia-Aguilera, and J.
Flanagan. 2011. Sterilisation of hybrid Galapagos tortoises (Geoche-
lone nigra) for island restoration. Part 1: endoscopic oophorectomy
of females under ketamine-medetomidine anaesthesia. Veterinary
Record 168 DOI: 10.1136/vr.c6520.
Koenig, W. D. 2003. European starlings and their effect on native cavity-
nesting birds. Conservation Biology 17:1134–1140.
Lees, A. C., and D. J. Bell. 2008. A conservation paradox for the 21st
century: the European wild rabbit Oryctolagus cuniculus,anin-
vasive alien and an endangered native species. Mammal Review
Volume 26, No. 5, 2012
788 Meta-Analyses of Non-Native Mutualists
Levey, D. J., and A. Grajal. 1991. Evolutionary implications of fruit-
processing limitations in cedar waxwings. The American Naturalist
Linz, G. M., H. J. Homan, S. M. Gaukler, L. B. Penry, and W. J. Bleier.
2007. European starlings: a review of an invasive species with far-
reaching impacts. Pages 378–386 in G. W. Witmer, W. C. Pitt, and
K. A. Fagerstone, editors. Managing vertebrate invasive species: pro-
ceedings of an international symposium. United States Department
of Agriculture/Animal and Plant Health Inspection Service Wildlife
Services, National Wildlife Research Center, Fort Collins, Colorado.
Lord, J. M. 1991. Pollination and seed dispersal in Freycinetia baueriana,
a dioecious liane that has lost its bat pollinator. New Zealand Journal
of Botany 29:83–86.
Luck, G. W., G. C. Daily, and P. R. Ehrlich. 2003. Population diversity
and ecosystem services. Trends in Ecology & Evolution 18:331–336.
McKinney, M. L., and J. L. Lockwood. 1999. Biotic homogenization:
a few winners replacing many losers in the next mass extinction.
Trends in Ecology & Evolution 14:450–453.
Moran, C., C. P. Catterall, and J. Kanowski. 2009. Reduced dispersal of
native plant species as a consequence of the reduced abundance of
frugivore species in fragmented rainforest. Biological Conservation
Morris, W. F., and D. F. Doak. 2002. Quantitative conservation biol-
ogy: theory and practice of population viability analysis. Sinauer
Associates, Sunderland, Massachusetts.
Olesen, J. M., L. I. Eskildsen, and S. Venkatasamy. 2002. Invasion of pol-
lination networks on oceanic islands: importance of invader com-
plexes and endemic super generalists. Diversity and Distributions
Parker, K. A., M. Seabrook-Davison, and J. G. Ewen. 2010. Opportuni-
ties for nonnative ecological replacements in ecosystem restoration.
Restoration Ecology 18:269–273.
Pattemore, D. C., and D. S. Wilcove. 2011. Invasive rats and recent
colonist birds partially compensate for the loss of endemic New
Zealand pollinators. Proceedings of the Royal Society B-Biological
Sciences DOI: 10.1098/rspb.2011.2036.
Peacock, D. S., B. J. van Rensburg, and M. P. Robertson. 2007. The distri-
bution and spread of the invasive alien common myna, Acridotheres
tristis L. (Aves: Sturnidae), in southern Africa. South African Journal
of Science 203:465–473.
Pearce, C. M., and D. G. Smith. 2008. Invasive saltcedar (Tamarix): its
spread from the American southwest to the northern Great Plains.
Physical Geography 28:507–530.
Petchey, O. L., and K. J. Gaston. 2002. Extinction and the loss of func-
tional diversity. Proceedings of the Royal Society of London. Series
B-Biological Sciences 269:1721–1727.
Pimm, S. L., G. J. Russell, J. L. Gittleman, and T. M. Brooks. 1995. The
future of biodiversity. Science 269:347–350.
anek, M., and M. J. Pitcairn. 2002. When is eradication of ex-
otic pest plants a realistic goal? Pages 249–253 in C. R. Veitch and
M. N. Clout, editors. Turning the tide: the eradication of invasive
species. International Union for the Conservation of Nature Species
Survival Commission, Invasive Species Specialist Group, Gland,
Renner, S. S., and R. E. Ricklefs. 1995. Dioecy and its corre-
lates in the flowering plants. American Journal of Botany 82:
Rezende, E. L., J. E. Lavabre, P. Guimar˜
aes, P. Jordano, and J. Bascompte.
2007. Non-random coextinctions in phylogenetically structured mu-
tualistic networks. Nature 448:925–928.
Richardson, D. M., N. Allsopp, C. M. D’Antonio, S. J. Milton, and M.
anek. 2000. Plant invasions: the role of mutualisms. Biological
Richardson, D. M., and P. Pyˇ
sek. 2008. Fifty years of invasion
ecology—the legacy of Charles Elton. Diversity and Distributions 14:
Roberts, M. L., K. L. Buchanan, and M. R. Evans. 2004. Testing the
immunocompetence handicap hypothesis: a review of the evidence.
Animal Behaviour 68:227–239.
Rosenberg, M. S. 2005. The file-drawer problem revisited: a general
weighted method for calculating fail-safe numbers in meta-analysis.
Rosenberg, M. S., D. C. Adams, and J. Gurevitch. 2000. MetaWin: sta-
tistical software for meta-analysis. Version 2.0. Sinauer Associates,
Samuels, I. A., and D. J. Levey. 2005. Effects of gut passage on seed
germination: Do experiments answer the questions they ask? Func-
tional Ecology 19:365–368.
Schemske, D. W., and C. C. Horvitz. 1984. Variation among floral visitors
in pollination ability: a precondition for mutualism specialization.
Schlaepfer, M. A., D. F. Sax, and J. D. Olden. 2011. The potential
conservation value of non-native species. Conservation Biology
Schupp, E. W., P. Jordano, and J. M. G´
omez. 2010. Seed disper-
sal effectiveness revisited: a conceptual review. New Phytologist
Seddon, P. J. 2010. From reintroduction to assisted colonization: moving
along the conservation translocation spectrum. Restoration Ecology
Silvertown, J., M. Franco, I. Pisanty, and A. Mendoza. 1993. Comparative
plant demography—relative importance of life-cycle components
to the finite rate of increase in woody and herbaceous perennials.
Journal of Ecology 81:465–476.
Simberloff, D., and P. Stiling. 1996. Risks of species intro-
duced for biological control. Biological Conservation 78:
Sorensen, A. E. 1984. Nutrition, energy and passage time: experiments
with fruit preference in European blackbirds (Turdus merula). Jour-
nal of Animal Ecology 53:545–557.
Srivastava, D. S., and T. Bell. 2009. Reducing horizontal and vertical
diversity in a foodweb triggers extinctions and impacts functions.
Ecology Letters 12:1016–1028.
Tompkins, D. M., and D. M. Gleeson. 2010. Relationship between
avian malaria distribution and an exotic invasive mosquito in
New Zealand. Journal of the Royal Society of New Zealand 36:
Traveset, A. 1998. Effect of seed passage through vertebrate frugivores’
guts on germination: a review. Perspectives in Plant Ecology, Evo-
lution and Systematics 1/2:151–190.
Traveset, A., and D. M. Richardson. 2006. Biological invasions as dis-
ruptors of plant reproductive mutualisms. Trends in Ecology & Evo-
Traveset, A., and D. M. Richardson. 2011. Mutualisms—key drivers of
invasions...key casualties of invasions. Pages 143–160 in D. M.
Richardson, editor. Fifty years of invasion ecology: the legacy of
Charles Elton. Wiley-Blackwell, Oxford, United Kingdom.
Underwood, E. C., S. L. Ustin, and C. M. Ramirez. 2007. Use of hyper-
spectral data to assess the effects of different nitrogen applications
on a potato crop. Environmental Management 39:225–239.
Vanthomme, H., B. Bell´
e, and P.-M. Forget. 2010. Bushmeat hunting
alters recruitment of large-seeded plant species in Central Africa.
u, M., and A. Traveset. 2004. Bridging meta-analysis and the com-
parative method: a test of seed size effect on germination after
frugivores’ gut passage. Oecologia 138:414–418.
Vitousek, P. M., and L. R. Walker. 1989. Biological invasion by Myrica
faya in Hawai’i: plant demography, nitrogen fixation, ecosystem
effects. Ecological Monographs 59:247–265.
Westcott, D. A., and C. S. Fletcher. 2011. Biological invasions and the
study of vertebrate dispersal of plants: opportunities and integration.
Acta Oecologica 37:650-656.
Volume 26, No. 5, 2012
Williams, P. A. 2006. The role of blackbirds (Turdus merula)in
weed invasion in New Zealand. New Zealand Journal of Ecology
Willson, M. F. 1993. Mammals as seed-dispersal mutualists in North
America. Oikos 67:159–176.
Wotton, D. M., and D. Kelly. 2011. Frugivore loss limits recruitment
of large-seeded trees. Proceedings of the Royal Society of London,
Series B-Biological Sciences DOI: 10.1098/rspb.2011.0185.
Zavaleta, E. S., J. R. Pasari, J. Moore, D. Hernandez, K. B. Suttle, and
C. C. Wilmers. 2009. Ecosystem responses to community disas-
sembly. Annals of the New York Academy of Sciences 1162:311–
Volume 26, No. 5, 2012