g
g
D
, U
te,
elic
ial
ve
hy
. W
thi
rato
mp
n si
at
tive
ign
nta
e a
; H
to visit
tropica
mid-nin
ated a
140 ye
remain
is. Her
butterfl
data as
Com
mals, i
especia
in bees
[8–10]
‘a beha
close p
hours’
from th
[11,12]
been o
repeate
range t
at thei
depart
mates are generally conspecifics, but sometimes different
ing to
atures,
warn-
atism
types,
signals
ch as
sema-
hieved
tterns.
apose-
cts of
mount
ittle is
nd its
osting
al stu-
ion of
ences
ations
lection
etime,
oured
explanations have been narrowed to two major hypotheses:
information sharing and/or anti-predator defence. The
information-sharing hypothesis proposes that new roost
mates, presumably related individuals, follow experienced
 on January 29, 2015http://rspb.royalsocietypublishing.org/Downloaded from * Author for correspondence (sfinkbei@uci.edu).Received
Acceptedthe Amazon, reported communal roosting in the
l Heliconius passion-vine butterflies [3]. Since the
eteenth century, this unusual behaviour has gener-
great deal of scientific and popular interest. After
ars of work on butterfly roosting, however, it still
s unclear what the benefit of being a social butterfly
e, we test the major hypotheses for why Heliconius
ies roost communally, and present experimental
sessing the adaptive function of this behaviour.
munal roosting is observed in many types of ani-
ncluding birds, bats and primates [4–7], and is
lly widespread in insects, having been observed
, wasps, beetles, dragonflies, butterflies and moths
. In butterflies, communal roosting is described as
vior in which individuals aggregate quiescently in
roximity to each other at a site for more than a few
([9] p. 90). This behaviour is known primarily
e heliconiines, acraeines, ithomiines and danaiines
. Many species of Heliconius in particular have
bserved to form communal roosts in which adults
dly gather in a particular location in their home
o roost for the night (figure 1a). Butterflies arrive
r roost sites as early as 3 h before sunset, and
from roosts within the first 2 h after sunrise. Roost
coloured wing patterns and their unpalatability ow
cyanogenic glycosides [18]. Because of these fe
Heliconius butterflies serve as a textbook example of
ing signalling—also known as aposematism. Aposem
is a major theme in the evolution of animal pheno
where its principal function is to provide warning
associated with unprofitability to predators, su
toxicity, unpalatability or capture costs [19–21]. Apo
tism is widespread in invertebrates and is often ac
through visual signalling via conspicuous colour pa
Collective aposematism is a phenomenon in which
matic prey form aggregations to enhance the effe
warning signals [22,23]. Despite the substantial a
that is known about Heliconius natural history, l
known about their communal roosting behaviour a
possible relationship to collective aposematism.
There is a broad literature on Heliconius ro
[12,13,24–29]; however, relatively few experiment
dies have been performed to address the funct
this behaviour. Although the adaptive consequ
of roosting remain unclear, it is unlikely that aggreg
are involved with thermoregulation [10], kin se
[13] or mating (females usually mate once in their lif
within hours or days after eclosion) [30,31]. The favroosting in butterflies [1,2]. Shortly thereafter W. H.
Edwards, the entomologist who inspired Bates and Wallace
long lifespan owing to pollen consumption [16,17].
Heliconius butterflies are well known for their brightlyThe benefit of bein
communal roostin
Susan D. Finkbeiner1,*, Adriana
1Department of Ecology and Evolutionary Biology
2Smithsonian Tropical Research Institu
Aposematic passion-vine butterflies from the genusH
behaviour has been hypothesized to be benefic
predator defence. To better understand the adapti
hypotheses in field studies. The information-sharing
haviour of butterflies departing from natural roosts
another to resources, thus providing no support for
was tested using avian-indiscriminable Heliconius e
sites. A significantly higher number of predation atte
tions ofmodels. This relationship between aggregatio
butterflies enjoy the benefits of both overall decreased
munal roosts probably deter predators through collec
unpalatable prey communicate amore effective repel s
that predation by birds is a key selective pressure mai
Keywords: communal roosting; collectiv
predation
1. INTRODUCTION
In 1867, the naturalist J. A. Allen first described the
spectacular aggregations of migratory monarchs, and was
thus the first to report the phenomenon of communal27 January 2012
29 February 2012 2769a social butterfly:
deters predation
. Briscoe1 and Robert D. Reed1,2
niversity of California, Irvine, CA 92697, USA
Panama City 0843-03092, Panama
onius form communal roosts on a nightly basis. This
in terms of information sharing and/or anti-
value of communal roosting, we tested these two
pothesis was addressed by examining following be-
e found no evidence of roost mates following one
s hypothesis. The anti-predator defence hypothesis
models placed singly and in aggregations at field
ts were observed on solitary models versus aggrega-
ze and attack rate suggests that communally roosting
tack frequency as well as a prey dilution effect. Com-
aposematism inwhich aggregations of conspicuous,
al to predators. On the basis of our results, we propose
ining Heliconius communal roosting behaviour.
posematism; dilution effect; aggregation;
eliconius
species—often Mu¨llerian co-mimics—roost together [13].
The evolution of communal roosting behaviour inHeliconius
is believed to be facilitated by unpalatability, slow repro-
ductive rate [14], limited learned home range [12,15] and
Proc. R. Soc. B (2012) 279, 2769–2776
doi:10.1098/rspb.2012.0203
Published online 21 March 2012This journal is q 2012 The Royal Society
((
o. (
2770 S. D. Finkbeiner et al. Butterfly communal roosting
 on January 29, 2015http://rspb.royalsocietypublishing.org/Downloaded from (c)
Figure 1. (a) Natural and (b) artificial roosts of Heliconius erat(a)members from the roost to food sources [27,32]. This form
of information sharing is a common behaviour in other
communal animals such as birds [33,34]. Conversely,
the anti-predator hypothesis suggests predators avoid
Heliconius aggregations as a result of collective aposematism
or predator confusion [12,13,35]. In other aposematic
insects, gregarious behaviour contributes to collective
enhancement of warning signalling, resulting inmore effec-
tive predator deterrence [22,36–38]. Another potential
anti-predator mechanism is the prey dilution effect, often
known as ‘safety in numbers’, which posits that the prob-
ability of a single individual being attacked in a group is
lower with increasing density [39–42].
In field studies in Panama and Costa Rica, we tested
these hypotheses to determine why Heliconius passion-
vine butterflies assemble in communal roosts. To test
whether Heliconius butterflies rely on roosts as infor-
mation-sharing centres, we examined following behaviour
by butterflies during departures from natural aggregations.
The anti-predator defence hypothesis was tested using
avian-indiscriminable artificial butterfly models placed
singly and in aggregations in the forest. Following the pre-
dation study, we investigated whether naturally occurring
roost sizes correspond with optimal roost sizes inferred
from experimental data.
2. MATERIAL AND METHODS
(a) Field sites
All data collection was completed in Costa Rica and Panama.
Field sites for this work were chosen based on the abundance
mark imprint on wax wing (indicated by arrow). (d) Dorsal view
Proc. R. Soc. B (2012)b)
d)
c) Beak pinch at end of model abdomen, and triangular beakand accessibility of Heliconius butterflies and communal
roosts. The Panama sites were part of the Smithsonian Tro-
pical Research Institute; we used areas in Gamboa and
Soberanı´a National Park along Pipeline Road. Data were col-
lected in Panama from June through September of 2010 and
2011 during the rainy season. The 2010 visit resulted in
natural roost data collected from Heliconius erato, and the
2011 visit resulted in the predation experiment data.
In Costa Rica, we worked at the Organization for Tropical
Studies’ La Selva Tropical Biological Station in Sarapiquı´.
La Selva was visited in April and May of 2011, during the
end of the dry season into the beginning of the rainy
season. This site was used for collecting natural roost data
from Heliconius sara and model predation experiments.
(b) Natural roost observations
To assess the information-sharing hypothesis, which predicts
that butterflies use roosts to learn the locations of foraging
sites from other roost mates, we observed following behaviour
of H. erato and H. sara individuals departing natural roosting
aggregations. Following was confirmed only if a butterfly was
observed departing the roost with another roost mate to sub-
sequently arrive at a flowering plant with that same roost
mate. The roosts were within 30 m of the first visited flowering
plant, and it was feasible to follow butterflies to these plants.
We began all observations approximately 30 min before sun-
rise, before butterflies left to forage, and stayed until only
one individual remained. All H. erato roost members were
given unique identification numbers using a Sharpie marker,
and were sexed and age determined based on wing wear
[43]. Butterflies were captured and marked after departing
of abdominal beak pinch.
single butterflies were used for the first predation experiment.
The models were placed in 80 different forest sites, each con-
taining four roosts and four single individuals. All 80 sites
were at least 250 m apart to control for the home range ter-
ritories of primary Heliconius avian predators. Predator home
range sizes vary between 100 and 250 m, and have been
determined by other researchers through radio tracking
(flycatchers [54]), harmonic distance method and core area
use (tanagers [55]), minimum convex polygon modelling
(flycatchers [54]) and observation (tanagers [56]; jacamars
[57]; C. E. G. Pinheiro 2011, personal communication;
L. E. Gilbert 2011, personal communication). Tree Tanglefoot
0
80(b)
60
40
pe
r c
en
t r
ef
le
ct
an
ce
20
0
80
300 400
natural wing black (forewing) A
A
B
A
B
natural wing black (hindwing) B
natural wing pink (light) A
natural wing pink (dark) B
model black
model pink
500
wavelength (nm)
600 700
(c)
60
40
pe
r c
en
t r
ef
le
ct
an
ce
20
0
Figure 2. Reflectance spectra of natural and artificial butter-
fly wings used in discriminability calculations. Shown are the
mean values and standard errors (each n ¼ 12). Graphs
include a ventral image of Heliconius erato petiverana with
circled regions indicating where wing reflectance measure-
ments were taken. (a) Yellow reflectance spectra. (b) Pink
reflectance spectra, comparisons made using lightest A and
darkest B pink regions on the wing. (c) Black reflectance
spectra, comparisons made using both forewing A and
hindwing B regions.
Butterfly communal roosting S. D. Finkbeiner et al. 2771
 on January 29, 2015http://rspb.royalsocietypublishing.org/Downloaded from their roosts to prevent them from associating the roost with
danger [44]. Average roost sizes were determined by recording
the number of individuals in each roost nightly.
(c) Models
We used artificial butterfly models to test the hypothesis that
Heliconius roosts provide an anti-predatory benefit. Wing
images were designed in Adobe ILLUSTRATOR using high-
resolution scans of ventral Heliconius erato petiverana wings
as a reference. Model butterfly wings were printed on What-
man filter paper, which produces reflectance spectra close
in brightness to actual wings, using an Epson Stylus Pro
4880 printer with UltraChrome K3 ink. A 3-hydroxy-DL-
kynurenine (3-OHK) pigment solution of 1.0 mg 3-OHK
dissolved in 100 ml of methanol was applied to the yellow
bands on the hindwing to provide accurate UV reflectance,
because printed yellows do not accurately mimic Heliconius
yellow in the shape of their reflectance spectra. 3-OHK is
the same yellow wing pigment as found in the wings of the
butterflies themselves [45,46]. Portions of the artificial
wings were dipped in clear wax to allow imprints of beak
and bite marks, then Krylon matte finishing spray was
applied lightly (before spectra measurements were taken) to
coat the 3-OHK with a waterproofing element. Model
abdomens were made of Newplast Plasticine.
Birds, in particular jacamars, flycatchers and tanagers, are
major Heliconius predators [27,36,47–50]. Therefore, artifi-
cial butterfly models were designed and assessed using
tetrachromatic bird colour-vision models in order to ensure
that avian predators would find colour stimuli presented by
the models indiscriminable from actual butterflies. Reflec-
tance spectra of yellow, pink and black from the models
and the ventral surface of natural H. erato petiverana wings
were measured using an Ocean Optics USB2000 fibre optic
spectrometer. A deuterium–halogen tungsten lamp (DH-
2000, Ocean Optics) was used as a standardized light
source, and measurements were taken using a bifurcating
fibre cable (R400-7-UV–vis, Ocean Optics). The axis of
the illuminating and detecting fibre was at an elevation of
458 to the plane of the wing and pointed left with respect
to the body axis for every measurement. A white spectralon
standard (WS-1-SL, Labsphere) was used to calibrate the
spectrometer. Spectra measurements from the fibre optic
spectrometer were processed using MATLAB software (see
Briscoe et al. [46]). The quantum catches for stimuli [51]
were calculated, and discriminability between artificial
models and natural wing reflectance spectra was determined
using tetrachromatic bird-vision models from Vorobyev &
Osorio [52]. The comparisons were made using the blue tit
(Parus caeruleus) and chicken (Gallus gallus) cone sensi-
tivities, which represent the UV- and violet-type avian
visual systems, respectively. Low light intensity and open
habitat irradiance spectra were used [53]. All spectral com-
parisons represented by an average of wing measurements
(n ¼ 12) fell below the threshold of one just noticeable differ-
ence (figure 2 and table 1); therefore, the reflectance spectra
of models and actual butterfly wings were inferred to be
indiscriminable to birds.
(d) Predation experiments
Butterfly models were tied to branches with thread in appro-
priate roosting habitats and in natural roosting postures
[10,13] (figure 1b). At our Costa Rica field site, a total of
320 aggregations containing five butterflies each and 320Proc. R. Soc. B (2012)80(a)
60
40
pe
r c
en
t r
ef
le
ct
an
ce
20
A
natural wing yellow (A)
model yellow
site included as a random effect. The post hoc tests for the
lig
allu
tice
HV
mo
0.7
0.8
2772 S. D. Finkbeiner et al. Butterfly communal roosting
 on January 29, 2015http://rspb.royalsocietypublishing.org/Downloaded from was applied to the base of plant stems containing artificial
butterflies to avoid removal or attack of the models by ants
and other small arthropods, and it was also effective in prevent-
ing small vertebrates such as lizards from reaching the butterfly
models [58].
The models were left at their sites for a total of 96 h
(4 days), and each model was examined daily for predation
evidence and replaced if attacked. None of the 80 sites
were used twice in the study because predator forgetting
time varies between bird species, and is affected by prey con-
spicuousness and distastefulness [59–61], both difficult to
measure for this project. A butterfly was considered attacked
if damage to the abdomen and wings appeared in the form
of beak marks and/or large indentations in the abdomen
(figure 1c,d; see also [62,63]). Small chew-like marks, prob-
ably from mandibular insects such as grasshoppers, were not
considered in the data analyses. If a model was attacked twice
on two separate days, then it was counted only as a single
attack. If multiple butterflies in a roost were attacked, this
was also counted as a single attack (i.e. each roost was treated
as a unit), because in nature when a roost is disturbed most
or all individuals depart from the roosting site [13,29] (S. D.
Finkbeiner 2010, personal observation), thereby reducing
the probability of further predation attempts on individual
roost mates. Predation differences were analysed using a
Wilcoxon signed-rank test with continuity correction, using
each site as a sample.
To assess the per capita attack risk of individual butterflies,
single (focal) individuals were randomly selected from each
roost to compare attack rates under the conservative assump-
tion that one attack leads to the dispersal of the other
roosting butterflies. For roosts of one, this single individual
was the focal individual. The binomial response of attack
(yes, no) was modelled as dependent upon roost size using
generalized linear models. In these analyses, site was
included as a random effect to account for potential
non-independence among replicate roosts within sites.
As a control for our models, we compared the attack rate on
models with real wings (and Plasticine abdomens) to the attack
rate on models with artificial wings, using five forest sites sep-
Table 1. Results from discriminability calculations using low
made using the blue tit (Parus caeruleus) and chicken (Gallus g
Spectra comparisons fell below the threshold of one just no
ventral; L, light; D, dark.
JND
comparisons
FV-L pink versus
model pink
FV-D pink versus
model pink
P. caeruleus 0.597 0.374
G. gallus 0.556 0.424arated 250 m apart, each with four models with real wings and
four models with artificial wings. They were left for 96 h and
checked daily for attacks. We found no difference in attacks
between real-wingmodels and artificial-wingmodels (Wilcoxon
signed-rank test: W¼ 249.5, p ¼ 0.836, n ¼ 20).
The second predation study, conducted in Panama, inves-
tigated the association between roost size and predation
frequency. This was tested using the same artificial butterfly
models described earlier. One hundred forest sites were
chosen, approximately 250 m apart, and each of these sites
contained two roosts of 2, two roosts of 5 and two roosts of
10 butterflies. This totalled 600 artificial roosts: 200 roosts
Proc. R. Soc. B (2012)significance of pairwise comparisons were made using a
Tukey test.
We determined at what time of day roosts are most sus-
ceptible to attack to gain further information about the
roost predators. This study was conducted with 100 artificial
roosts in Panama: 52 roosts of two placed randomly along
two trails in Soberanı´a National Park, with the remaining
48 roosts already being observed for predation data (16
more roosts of 2, 5 and 10 each). The artificial roosts
were checked for attacks every 3 h from 06.00 (15 min
prior to sunrise) to 18.00 (just before sunset) for 9 days.
3. RESULTS
(a) Observations of roost departure following
To assess whether Heliconius butterflies rely on communal
roosts as information-sharing centres, we observed fol-
lowing behaviour during morning roost departures. Of
256 H. erato departures by at least 66 unique individuals
from nine different roosts, only one instance of following
from the roost to a flowering plant was observed
(Wilcoxon signed-rank test: W ¼ 256, p , 0.0001, n ¼
256, based on the null hypothesis that there is no differ-
ence between the number of butterflies that do and do
not follow). Additionally, out of 74 H. sara butterfly
departures from at least 25 unique individuals and threefor each treatment. Roosts were removed after 4 days, each
of the 100 sites was used only once and attacked models
were replaced when necessary. If a roost was attacked more
than once or if more than one butterfly in the roost was
attacked, then it was counted only as one attack; thus each
roost was considered as a single unit in the analysis. These
predation data were analysed using a Kruskal–Wallis mul-
tiple comparison test with a post hoc Bonferroni correction,
using each site as a sample. Individual per capita attack risk
was determined by comparing attack rates between randomly
selected single (focal) individuals in each roost treatment,
and the binomial response of attack was modelled as depen-
dent upon roost size using generalized linear models, with
ht intensity and open habitat irradiance. Comparisons were
s) cone sensitivities, and are based on mean values (n ¼ 12).
able difference (JND). FV, forewing ventral; HV, hindwing
yellow versus
del yellow
FV black versus
model black
HV black versus
model black
36 0.336 0.440
58 0.315 0.533different roosts, we observed no incidence of follow-
ing. When butterflies departed roosts in the morning
occasionally more than one individual would leave at the
same time, but they were never observed to follow one
another. Most of the time the butterflies left individually,
even when there was a disturbance event.
Lantana and Psychotria plants, considered to be key
Heliconius resources, were common at our study sites, and
it is important to note that spatial distribution and density
of nectar and pollen sources may influence whether follow-
ing happens. We observed that many roosting butterflies
shared the same flower resources and followed each other
(b) Observations on roost size and proximity
(c) Effect of roost size on predation frequency
ld
roos
tes
ds
s.
Butterfly communal roosting S. D. Finkbeiner et al. 2773
 on January 29, 2015http://rspb.royalsocietypublishing.org/Downloaded from To determine whether communal roosts provide an
anti-predatory benefit, we first tested whether there is a
difference in predation between single butterfly models
or models placed in aggregations of five. We observed a
very strong difference in attack frequencies between roost-
ing and solitary butterflies. Of 320 artificial H. eratoThe average H. erato roost size was observed to be 4.3
individuals (s.d. ¼ 1.6, n ¼ 233 observations across nine
natural roosts) from roosts in Gamboa, Panama. It was
common to find multiple roosts within 15 m of each
other, some as close as 3 m apart, and in line-of-sight
from one another in a given part of the home range.
We observed this in both H. erato and H. sara. When but-
terflies of H. erato were exercising pre-roosting behaviour,
the butterflies often interacted with one another before
convening at their preferred roosts.between flowering plants, as previously described by
Waller & Gilbert [27]. As well, on multiple occasions, we
observed a new recruit following an established roost
member to the aggregation, suggesting following behaviour
may play a role in roost recruitment.
Table 2. Data representing attacks on butterfly models at fie
size. Roost attack risk was calculated by dividing attacked
Individual attack risk was determined by comparing attack ra
roost treatment, based on the assumption that one attack lea
values between pairwise comparisons are indicated by asterisk
aggregation size total observations roosts attacked
Costa Rica
single individuals 320 68g***roost of five 320 25
Panama
roost of two 200 21g*
roost of five 200 8
roost of 10 200 *f24
*p, 0.05.
**p, 0.001.
***p, 0.0005.roosts, only 25 were attacked compared with 68 attacks
on 320 single individuals (Wilcoxon signed-rank test:
W ¼ 4141.5, p ¼ 0.000262, n ¼ 80 sites; table 2). Indi-
vidual attack risk also differed significantly between
roosts of 1 (single butterflies) and roosts of 5 (F1,559 ¼
36.85, p , 0.0001, n ¼ 640 roosts total), with 21.3 per
cent of focal individuals being attacked in roosts of 1
but only 3.4 per cent of focal individuals being attacked
in roosts of 5 (table 2).
Considering the difference we found in predation
between single butterflies and roosts, we decided to inves-
tigate whether the predator deterrence effect was similar
across roosts of different sizes (table 2). We found signifi-
cantly higher predation on roosts of 2 and 10 than on
roosts of 5 (Kruskal–Wallis test: x2¼ 8.7356, p ¼
0.0127, n ¼ 100 sites). The post hoc Bonferroni correction
showed that roosts of 10 were attacked more than roosts
of 5 (p ¼ 0.016), and roosts of 2 were attacked more
than roosts of 5 (p ¼ 0.028). There was no significant
Proc. R. Soc. B (2012)difference in predation frequency between roosts of 2
and 10 (p ¼ 1.000). Individual attack rate also differed
significantly between the three roost sizes (F2,498 ¼ 5.51,
p ¼ 0.0043, n ¼ 600 roosts total), with attack rates on
focal individuals in roosts of 2, 5 and 10 being 9.5
per cent, 1.5 per cent and 4.5 per cent, respectively
(table 2). Adjusting significance levels to account for mul-
tiple comparisons, Tukey tests showed that a single
butterfly in a roost of 2 has a higher attack risk than a
single butterfly in a roost of 5 (p ¼ 0.00064), but there
is no difference in individual attack risk between roosts
of 5 and 10 butterflies (p ¼ 0.22), or between roosts of
2 and 10 butterflies (p ¼ 0.13).
(d) Timing and nature of predation
We sought to determine at which time of day roosts are
most susceptible to attack. We found that all attacks
occurred during the morning hours, as previously
observed by Mallet [13]. Ten out of 12 attacks, from
100 different artificial roosts, occurred between the
hours of 06.00 and 09.00, and two attacks occurred
between the hours of 09.00 and 12.00. Triangular beak
marks observed in the model abdomens supports pre-
vious findings [48–50] that birds are the primary
predators of Heliconius. This is further supported because
all attacks on models occurred between the hours ofsites in Costa Rica and Panama, categorized by aggregation
ts by the overall number of roosts used in that treatment.
between randomly selected single (focal) individuals in each
to the dispersal of the other roosting butterflies. Probability
attack risk (per roost, %) attack risk (per individual, %)
21.3 21.3g***7.8 3.4
10.5 9.5g**4.0 1.5
12.0 4.506.00 and 12.00 when birds are most active [56,64].
4. DISCUSSION
In this study, we assessed the two major hypotheses for
explaining why Heliconius butterflies participate in com-
munal roosting behaviour. We found no support for the
information-sharing hypothesis, because there was little
evidence of roost mates following each other to resources
upon departure from roosts. These findings are in agree-
ment with Mallet [13], who observed a similar lack of
following and even a predictable tendency for roost
mates to visit different flowers.
In contrast, we found very strong support for the anti-
predator defence hypothesis. Our field experiments in
Costa Rica using avian-indiscriminable butterfly models
showed predation attempts on singly placed models
were nearly three times higher than predation attempts
on roosts of five models, and the predation risk for a
butterflies are aware of other roosts in their home range,
yet still choose to join smaller aggregations, despite
2774 S. D. Finkbeiner et al. Butterfly communal roosting
 on January 29, 2015http://rspb.royalsocietypublishing.org/Downloaded from single butterfly is over six times the per capita preda-
tion risk for an individual butterfly in a roost of 5
(table 2). A second field experiment in Panama showed
the same trend, with attack rates more than twice as high
on roosts of 2 versus 5, and individual risk over six times
higher in a roost of 2 than a roost of 5 (table 2). Surpris-
ingly, however, the Panama experiment also showed that
attacks on roosts of 10 were three times as high as on
roosts of 5 (table 2), thus suggesting the predator deter-
rence effect may be weak or non-existent in large
aggregations. In the Panama experiment, the greatest indi-
vidual fitness benefit was seen in roosts of 5, with an
individual predation risk of 1.5 per cent; however, individ-
uals in roosts of 10 benefited only slightly less (not
significant) than those in roosts of 5 (individual risk of
4.5% in roosts of 10). Therefore, even though roosts of
10 did not enjoy a significantly decreased predation rate
compared with roosts of 2 or 5, a simple prey dilution
effect [39] would still favour large roost sizes.
Our field studies suggest that the most beneficial mini-
mum roost size, with respect to group advantage, may be
around 5 individuals. This is because our experimental
aggregations of five models experienced the lowest overall
attack rates and also offered the lowest per capita attack
risk for individuals. Interestingly, this experimentally
determined minimum roost size corresponds closely to
naturally observed H. erato roost sizes (4.3). This corre-
spondence implies that predator deterrence, coupled
with a prey dilution effect, could help explain roost sizes
observed in natural populations. An optimal or minimum
roost size may be important for predator deterrence when
butterfly densities are too low for assembling in larger
aggregations; however, roost sizes are probably influenced
by foraging and resource availability as well.
Because medium-sized roosts provide an anti-predatory
benefit through collective aposematism, it is unclear why
larger aggregations appear to lose their ability to deter pre-
dators. It is possible that solitary individuals or very small
roosts are too small to communicate an effective warning
signal, whereas very large roosts may be conspicuous
enough to attract naı¨ve predators. In support of this idea,
Salcedo [29] noted that themost frequent predator disturb-
ances on H. sara roosts occurred on the largest aggregation
studied (10–16 individuals), suggesting that oversized
aggregations of Heliconius may increase the frequency of
predator attacks. Although there is much evidence that
predator wariness and aggregation-induced phobias
increase with the size of aposematic prey aggregations
[61], greater aggregation distinctiveness increases detect-
ability costs and, in some cases, larger aggregation sizes of
defended animals result in higher predation [23,65].
There may be other costs to forming larger groups such
that small movements made by other butterflies could pro-
duce incidental disturbances or ‘false alarms’, causing
unexpected or premature roost departures, but this would
not explain higher predation on larger roosts from our
experimental data.
We propose that it is no coincidence that multiple
roosts are found in the same location in a home range
and often in line-of-sight from one another. We have
observed this in H. erato, H. sara, H. melpomene and
H. charithonia, with some roosts neighbouring the roosts
of heterospecifics (in Costa Rica). Others have observed
this Heliconius behaviour of preferentially forming smallerProc. R. Soc. B (2012)tion sizes. This correspondence suggests that predation
may be the key selective pressure maintaining communal
roosting in Heliconius, thus providing insight into the
types of ecological pressures that contribute to the evolution
of social behaviour in historically solitary animals.
We thank Kailen Mooney for aid in project design and
statistical analyses; Jim Mallet for advice and comments;
Christian Salcedo and Daniel Osorio for feedback on the
manuscript; Nancy Burley, Owen McMillan and Larry
Gilbert for advice; Jasmine Velez, Talia Gustafson, Jimena
Golcher and Maranatha Kellinger for aid in field
observations and assisting in the preparation of butterfly
models; Johannes Spaethe, Chris Jiggins and Patricio
Salazar for tips on butterfly models; Dave Krueger and
UCI ImageWorks for aid in designing and printing models;
the Smithsonian Tropical Research Institute (STRI) and
Organization for Tropical Studies (OTS) for use of field
sites; La Autoridad Nacional del Ambiente (ANAM,
Panama) and El Ministerio del Ambiente, Energı´a, y
Telecomunicaciones (MINAET, Costa Rica) for research
permit approval; and our funding sources: the Smithsonian
Tropical Research Institue, the Organization for Tropical
Studies, the U.S. Department of Education GAANN, the
National Geographic Society; this material is based upon
work supported by the National Science Foundation (NSF)
Graduate Research Fellowship under award no. DGE-
0808392 to S.D.F. and NSF grant no. IOS-1025106 to
A.D.B. and R.D.R.
REFERENCES
1 Scudder, S. H. & Allen, J. A. 1869 A preliminary list
of the butterflies of Iowa. Trans. Chicago Acad. Sci. 1,
326–337.
2 Brower, L. P. 1995 Understanding and misunderstanding
the migration of the monarch butterfly (Nymphalidae) in
North America: 1857–1995. J. Lepid. Soc. 49, 304–385.plenty of substrate (i.e. dry branches and twigs) to sustain
much larger aggregations. On the basis of the assumption
that individual fitness should increase with larger aggrega-
tions owing to a reduced per capita predation risk, again it
is unclear why local butterflies do not simply choose to
form very large roosts. There is the possibility that an
interaction among roosts is introduced, so that if individ-
uals in one group are attacked the predator is inhibited
from attacking individuals in other groups [42]. This
repeated warning display could therefore facilitate rapid
learning in naı¨ve birds whose feeding areas may include
multiple Heliconius roosting sites [67,68]. It is also poss-
ible that very large roosts could attract high enough
levels of predation to cancel out the prey dilution benefit,
although experimental work is required to test this idea.
Here, we show that Heliconius butterflies do not rely on
communal roosts as information-sharing centres. Instead,
our field studies indicate that this behaviour confers an
anti-predatory benefit at both the individual level (prey
dilution) and group level (collective aposematism), and
attack risks vary between individuals in different aggrega-aggregations as well [13,66] (C. Boggs 2011, personal
communication). Salcedo [29] proposes these may be
early-stage aggregations made up of individuals who
have not yet located a larger roost. In contrast, however,
our field observations of pre-roosting interactions
between butterflies from different roosts suggest that the
Butterfly communal roosting S. D. Finkbeiner et al. 2775
 on January 29, 2015http://rspb.royalsocietypublishing.org/Downloaded from 3 Edwards, W. H. 1881 On certain habits of Heliconia char-
itonia Linn., a species of butterfly found in Florida.
Papilio 1, 209–215.
4 Ward, P. 1965 Feeding ecology of the black-faced dioch
Quellea quellea in Nigeria. Ibis 107, 173–214. (doi:10.
1111/j.1474-919X.1965.tb07296.x)
5 Soini, P. 1987Ecology of the saddle-back tamarin Saguinus
fuscicollis-illigeri on the Rio Pacaya, northeastern Peru.Folia
Primatol. 49, 11–32. (doi:10.1159/000156305)
6 Lewis, S. E. 1995 Roost fidelity of bats: a review.
J. Mammal. 76, 481–496. (doi:10.2307/1382357)
7 Merkel, F. R. & Mosbech, A. 2008 Diurnal and noctur-
nal feeding strategies in common eiders. Waterbirds 31,
580–586. (doi:10.1675/1524-4695-31.4.580)
8 Pearson, D. L. & Anderson, J. J. 1985 Perching heights
and nocturnal communal roosts of some tiger beetles
(Coleoptera: Cicindelidae) in southeastern Peru.
Biotropica 17, 126–129. (doi:10.2307/2388504)
9 DeVries, P. J., Schull, J. & Greig, N. 1987 Synchro-
nous nocturnal activity and gregarious roosting in the
neotropical skipper butterfly Celaenorrhinus fritzgaertneri
(Lepidoptera: Hesperiidae). Biol. J. Linn. Soc. 89,
89–103. (doi:10.1111/j.1096-3642.1987.tb01345.x)
10 Salcedo, C. 2010 Environmental elements involved in
communal roosting in Heliconius butterflies (Lepidoptera:
Nymphalidae). Environ. Entom. 39, 907–911. (doi:10.
1603/EN09340)
11 Benson, W. W. & Emmel, T. C. 1973 Demography of
gregariously roosting populations of the nymphaline
butterfly Marpesia berania in Costa Rica. Ecology 54,
326–335. (doi:10.2307/1934340)
12 Turner, J. R. G. 1975 Communal roosting in relation to
warning colour in two heliconiine butterflies (Nymphali-
dae). J. Lepid. Soc. 29, 221–226.
13 Mallet, J. 1986 Gregarious roosting and home range in
Heliconius butterflies. Natl. Geogr. Res. 2, 198–215.
14 Erlich, P. R. & Gilbert, L. E. 1973 Population structure
and dynamics of the tropical butterfly Heliconius ethilla.
Biotropica 5, 69–82. (doi:10.2307/2989656)
15 Mallet, J. 1986 Dispersal and gene flow in a butterfly with
home range behavior: Heliconius erato (Lepidoptera:
Nymphalidae). Oecologia 68, 210–217. (doi:10.1007/
BF00384789)
16 Gilbert, L. E. 1972 Pollen feeding and reproductive
biology of Heliconius butterflies. Proc. Natl Acad. Sci.
USA 69, 1403–1407. (doi:10.1073/pnas.69.6.1403)
17 Boggs, C. L., Smiley, J. T. & Gilbert, L. E. 1981 Patterns
of pollen exploitation by Heliconius butterflies. Oecologia
48, 284–289. (doi:10.1007/BF00347978)
18 Engler-Chaouat, H. S. & Gilbert, L. E. 2007 De novo
synthesis versus sequestration: negatively correlated
metabolic traits and the evolution of host plant specializ-
ation in cyanogenic butterflies. J. Chem. Ecol. 33, 25–42.
(doi:10.1007/s10886-006-9207-8)
19 Cott, H. B. 1957 Adaptive coloration in animals. London,
UK: Methuen.
20 Guilford, T. 1990 The evolution of aposematism. In
Insect defenses: adaptive mechanisms and strategies of prey
and predators (eds D. L. Evans & J. O. Schmidt), pp.
23–61. Albany, NY: State University of New York Press.
21 Santos, J. C., Coloma, L. A. & Cannatella, D. C. 2003
Multiple, recurring origins of aposematism and diet
specialization in poison frogs. Proc. Natl Acad. Sci. USA
100, 12 792–12 797. (doi:10.1073/pnas.2133521100)
22 Gamberale, G. & Tullberg, B. S. 1998 Aposematism and
gregariousness: the combined effect of group size and
coloration on signal repellence. Proc. R. Soc. Lond. B
265, 889–894. (doi:10.1098/rspb.1998.0374)
23 Riipi, M., Alatalo, R. V., Lindstro¨m, L. & Mappes, J. 2001
Multiple benefits of gregariousnes cover detectability costsProc. R. Soc. B (2012)in aposematic aggregations. Nature 413, 512–514.
(doi:10.1038/35097061)
24 Jones, F. M. 1930 The sleeping heliconias of Florida.
Nat. Hist. 30, 635–644.
25 Young, A. M. & Thomason, J. H. 1975 Notes on com-
munal roosting of Heliconius charitonius (Nymphalidae)
in Costa Rica. J. Lepid. Soc. 29, 243–255.
26 Mallet, J. L. B. 1980 A laboratory study of gregarious
roosting in the butterfly Heliconius melpomene. MSc
thesis, University of Newcastle-upon-Tyne, Newcastle-
upon-Tyne, UK.
27 Waller, D. A. & Gilbert, L. E. 1982 Roost recruitment
and resource utilization: observations on a Heliconius
charitonia L. roost in Mexico. J. Lepid. Soc. 36, 178–184.
28 Mallet, J. & Gilbert, L. E. 1995 Why are there so many
mimicry rings? Correlations between habitat, behaviour
and mimicry in Heliconius butterflies. Biol. J. Linn. Soc.
55, 159–180. (doi:10.1111/j.1095-8312.1995.tb01057.x)
29 Salcedo, C. 2011 Evidence of predation and disturbance
events at Heliconius (Insecta: Lepidoptera: Nymphalidae)
nocturnal aggregations in Panama and Costa Rica.
J. Nat. Hist. 45, 1715–1721. (doi:10.1080/00222933.
2011.559692)
30 Gilbert, L. E. 1976 Postmating female odor in Heliconius
butterflies: male-contributed anti-aphrodisiac. Science
193, 419–420. (doi:10.1126/science.935877)
31 Schulz, S., Estrada, C., Yildizhan, S., Boppre´, M. &
Gilbert, L. E. 2008 An antiaphrodisiac in Heliconius
melpomene butterflies. J. Chem. Ecol. 34, 82–93.
(doi:10.1007/s10886-007-9393-z)
32 Gilbert, L. E. 1975 Ecological consequences of a coevolved
mutualism between butterflies and plants. In Coevolution
of animals and plants (eds L. E. Gilbert & P. R. Raven),
pp. 210–240. Austin, TX: University of Texas Press.
33 Ward, P. & Zahavi, A. 1973 The importance of certain
assemblages of birds as ‘information centers’ for food
finding. Ibis 115, 517–534. (doi:10.1111/j.1474-919X.
1973.tb01990.x)
34 Dall, S. R. X. 2002 Can information sharing explain
recruitment to food from communal roosts? Behav.
Ecol. 13, 42–51. (doi:10.1093/beheco/13.1.42)
35 Gillett, S. D., Hogarth, P. J. & Noble, F. E. 1979 The
response of predators to varying densities of Gregaria
locust nymphs. Anim. Behav. 27, 592–596. (doi:10.
1016/0003-3472(79)90195-7)
36 Benson, W. W. 1971 Evidence for the evolution of
unpalatability through kin selection in the Heliconiinae
(Lepidoptera). Am. Nat. 105, 213–226. (doi:10.1086/
282719)
37 Edmunds, M. 1974 Defence in animals: a survey of anti-
predator defences. Harlow, UK: Longman.
38 Vulinek, K. 1990 Collective security: aggregation by
insects as a defense. In Insect defenses: adaptive mechanisms
and strategies of prey and predators (eds D. L. Evans &
J. O. Schmidt), pp. 251–288. Albany, NY: State
University of New York Press.
39 Bertram, B. C. R. 1978 Living in groups: predators and
prey. In Behavioural ecology: an evolutionary approach
(eds J. R. Krebs & N. B. Davies), pp. 64–96. Sunderland,
MA: Sinauer Associates.
40 Hamilton, W. D. 1971 Geometry for the selfish herd.
J. Theor. Biol. 31, 295–311. (doi:10.1016/0022-5193
(71)90189-5)
41 Turner, G. F. & Pitcher, T. J. 1986 Attack abatement: a
model for group protection by combined avoidance and
dilution. Am. Nat. 128, 228–240. (doi:10.1086/284556)
42 Sille´n-Tullberg, B. & Leimar, O. 1988 The evolution
of gregariousness in distasteful insects as a defense
against predators. Am. Nat. 132, 723–734. (doi:10.
1086/284884)
43 Karlsson, B. 1987 Variation in egg weight, oviposition
rate and reproductive reserves with female age in a natu-
ral population of the speckled wood butterfly, Pararge
aegeria. Ecol. Entomol. 12, 473–476. (doi:10.1111/j.
1365-2311.1987.tb01029.x)
44 Mallet, J., Longino, J. T., Murawski, D., Murawski, A. &
Simpson de Gamboa, A. 1987 Handling effects in Helico-
nius: where do all the butterflies go? J. Anim. Ecol. 56,
377–386. (doi:10.2307/5054)
45 Reed, R. D., McMillan, W. O. & Nagy, L. M. 2008 Gene
expression underlying adaptive variation in Heliconius
wing patterns: non-modular regulation of overlapping
cinnabar and vermilion prepatterns. Proc. R. Soc. B 275,
37–45. (doi:10.1098/rspb.2007.1115)
46 Briscoe, A. D., Bybee, S. M., Bernard, G. D., Yuan, F.,
Front. Biol. China 2, 345–350. (doi:10.1007/s11515-
007-0051-1)
55 Samuel, M. D., Pierce, D. J. & Garton, E. O. 1985 Iden-
tifying areas of concentrated use within home range.
J. Anim. Ecol. 54, 711–719. (doi:10.2307/4373)
56 Buskirk, W. H., Powell, G. V. N., Wittenberger, J. F.,
Buskirk, R. E. & Powell, T. U. 1972 Interspecific
bird flocks in tropical highland Panama. The Auk 89,
612–624.
57 Pinheiro, C. E. G., Bagno, M. A. & Branda˜o, R. A. 2003
Diet and foraging behavior of the rufous-tailed jacamar
(Galbula ruficauda, Galbulidae) in central Brazil.
Ararajuba 11, 241–243.
58 Dial, R. & Roughgarden, J. 1995 Experimental removal
of insectivores from rain forest canopy: direct and indirect
2776 S. D. Finkbeiner et al. Butterfly communal roosting
 on January 29, 2015http://rspb.royalsocietypublishing.org/Downloaded from ente-Bosquets, J. & Chiao, C.-C. 2010 Positive selection
of a duplicated UV-sensitive visual pigment coincides
with wing pigment evolution in Heliconius butterflies.
Proc. Natl Acad Sci. USA 107, 3628–3633. (doi:10.
1073/pnas.0910085107)
47 Chai, P. 1986 Field observations and feeding experiments
on the responses of rufous-tailed jacamars (Galbula rufi-
cauda) to free-flying butterflies in a tropical rainforest.
Biol. J. Linn. Soc. 29, 161–189. (doi:10.1111/j.1095-
8312.1986.tb01772.x)
48 Pinheiro, C. E. G. 2003 Does Mu¨llerian mimicry work in
nature? Experiments with butterflies and birds (Tyranni-
dae). Biotropica 35, 356–364. (doi:10.1111/j.1744-7429.
2003.tb00589.x)
49 Langham, G. M. 2004 Specialized avian predators
repeatedly attack novel color morphs of Heliconius butter-
flies. Evolution 58, 2783–2787.
50 Pinheiro, C. E. G. 2011 On the evolution of warning
coloration, Batesian and Mu¨llerian mimicry in Neotro-
pical butterflies: the role of jacamars (Galbulidae) and
tyrant-flycatchers (Tyrannidae). J. Avian Biol. 42,
277–281. (doi:10.1111/j.1600-048X.2011.05435.x)
51 Kelber, A., Vorobyev, M. & Osorio, D. 2003 Animal colour
vision: behavioural tests and physiological concepts. Biol.
Rev. 78, 81–118. (doi:10.1017/S146479 3102005985)
52 Vorobyev, M. & Osorio, D. 1998 Receptor noise as a
determinant of colour thresholds. Proc. R. Soc. Lond. B
265, 351–358. (doi:10.1098/rspb.1998.0302)
53 Bybee, S. M., Yuan, F., Ramstetter, M. D., Llorente-
Bousquets, J., Reed, R. D., Osorio, D. & Briscoe, A. D.
2011 UV Photoreceptors and UV-yellow wing pigments
in Heliconius butterflies allow a color signal to serve
both mimicry and intraspecific communication. Am.
Nat. 179, 38–51. (doi:10.1086/663192)
54 Ning, W., Zhang, Y. & Zheng, G. 2007 Home ranges
and habitat vegetation characters in breeding season of
Narcissus Flycatcher and Yellow-rumped Flycatcher.Proc. R. Soc. B (2012)59 Servedio, M. R. 2000 The effects of predator learning,
forgetting, and recognition errors on the evolution of
warning coloration. Evolution 54, 751–763. (doi:10.
1111/j.0014-3820.2000.tb00077.x)
60 Speed, M. P. 2000 Warning signals, receiver psychology
and predator memory. Anim. Behav. 60, 269–278.
(doi:10.1006/anbe.2000.1430)
61 Ruxton, G. D., Sherratt, T. N. & Speed, M. P. 2004
Avoiding attack: the evolutionary ecology of crypsis, warning
signals and mimicry. Oxford, UK: Oxford University
Press.
62 Brodie, E. D. 1993 Differential avoidance of coral snake
banded patterns by free-ranging avian predators in Costa
Rica. Evolution 47, 227–235. (doi:10.2307/2410131)
63 Stobbe, N. & Schaefer, H. M. 2008 Enhancement of
chromatic contrast increases predation risk for striped
butterflies. Proc. R. Soc. B 275, 1535–1541. (doi:10.
1098/rspb.2008.0209)
64 DesLauriers, J. V. & Francis, C. M. 1991 The effect of
time of day on mist-net captures of passerines on spring
migration. J. Field Ornithol. 62, 107–116.
65 Lindstedt, C., Huttunen, H., Kakko, M. & Mappes, J.
2011 Disentangling the evolution of weak warning sig-
nals: high detection risk and low production costs of
chemical defences in gregarious pine sawfly larvae.
Evol. Ecol. 25, 1029–1046. (doi:10.1007/s10682-010-
9456-4)
66 Salcedo, C. 2006 Spatial dynamics of night roosting in
Heliconius erato petiverana (Lepidoptera: Nymphalidae).
MSc thesis, University of Florida, Gainesville, FL.
67 Poulton, E. B. 1931 The gregarious sleeping habits of
Heliconius charitonius L. Proc. R. Entomol. Soc. Lond. 6,
4–10.
68 Tuskes, P. M. & Brower, L. P. 1978 Overwintering ecol-
ogy of the monarch butterfly, Danaus plexippus L., in
California. Ecol. Entomol. 3, 141–153. (doi:10.1111/j.
1365-2311.1978.tb00912.x)Sison-Mangus, M. P., Reed, R. D., Warren, A. D., Llor- effects. Ecology 76, 1821–1834. (doi:10.2307/1940714)