Extreme allomaternal care and unequal task participation by unmated females in a cooperatively breeding spider

Abstract

Division of reproductive behaviour and alloparental care are key aspects of many animal societies. In cooperatively breeding species, variation in helping effort and unequal task participation are frequently observed. However, the extent to which the reproductive state of an individual affects the tasks performed during offspring care remains poorly understood. In the social spider Stegodyphus dumicola, approximately 40% of females reproduce, and mothers show extended maternal care including eggsac tending, regurgitation feeding and matriphagy, in which they are consumed by the offspring. We asked whether and to what extent virgin females participate in extreme maternal care and whether they differ from reproducing females in foraging activity. We show that virgin females contributed to all aspects of extended brood care, including regurgitation feeding and matriphagy. This suggests a physiological adaptation in virgin females to cooperative breeding, since in the subsocial Stegodyphus lineatus only mated females provide extended maternal care. Although virgin females and mothers are behaviourally totipotent, we found evidence for task differentiation as virgins engaged less in brood care and more in prey attack than mothers. High relatedness among nestmates and low probability of future reproduction in virgin helpers suggest alignment of reproductive interests between mothers and allomothers. Therefore, extreme allomaternal care by virgin helpers can be considered an adaptation to cooperative breeding in social spiders.

Full text

Extreme allomaternal care and unequal task participation by unmated
females in a cooperatively breeding spider
Anja Junghanns
a
,
*
,
1
, Christina Holm
b
,
1
, Mads Fristrup Schou
b
, Anna Boje Sørensen
b
,
Gabriele Uhl
a
,
2
, Trine Bilde
b
,
2
a
Department of General and Systematic Zoology, Zoological Institute and Museum, Ernst Moritz Arndt University, Greifswald, Germany
b
Section for Genetics, Ecology and Evolution, Department of Bioscience, Aarhus University, Aarhus, Denmark
article info
Article history:
Received 29 December 2016
Initial acceptance 24 January 2017
Final acceptance 17 July 2017
MS. number: 17-00013R
Keywords:
cooperation
reproductive skew
reproductive state
social spiders
sociality
task differentiation
Division of reproductive behaviour and alloparental care are key aspects of many animal societies. In
cooperatively breeding species, variation in helping effort and unequal task participation are frequently
observed. However, the extent to which the reproductive state of an individual affects the tasks per-
formed during offspring care remains poorly understood. In the social spider Stegodyphus dumicola,
approximately 40% of females reproduce, and mothers show extended maternal care including eggsac
tending, regurgitation feeding and matriphagy, in which they are consumed by the offspring. We asked
whether and to what extent virgin females participate in extreme maternal care and whether they differ
from reproducing females in foraging activity. We show that virgin females contributed to all aspects of
extended brood care, including regurgitation feeding and matriphagy. This suggests a physiological
adaptation in virgin females to cooperative breeding, since in the subsocial Stegodyphus lineatus only
mated females provide extended maternal care. Although virgin females and mothers are behaviourally
totipotent, we found evidence for task differentiation as virgins engaged less in brood care and more in
prey attack than mothers. High relatedness among nestmates and low probability of future reproduction
in virgin helpers suggest alignment of reproductive interests between mothers and allomothers.
Therefore, extreme allomaternal care by virgin helpers can be considered an adaptation to cooperative
breeding in social spiders.
©2017 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
Group living in animals varies from anonymous collectives to
highly specialized communities in which individuals cooperate in
nest maintenance, predator defence, foraging and breeding
(Clutton-Brock, 2002; Cockburn, 2006; Keller &Reeve, 1994; Lubin
&Bilde, 2007). Cooperative breeding involves alloparental care, in
which adult individuals in addition to the genetic parents partici-
pate in rearing the offspring. There are two types of cooperative
breeding: in one type there is some degree of shared parentage of
offspring and, more intriguingly, in the other adult nonbreeders
help raise the young. The latter indicates the common occurrence of
division of reproductive behaviour in cooperatively breeding soci-
eties, with one or a few individuals exclusively reproducing, while
others act as helpers at the nest (e.g. in insects: Wilson, 1971; birds:
Cockburn, 2006; mammals: Lukas &Clutton-Brock, 2012;fish:
Taborsky, 2009; arachnids: Avil
es, 1997; Lubin &Bilde, 2007). The
evolutionary causes underlying decisions to forgo one's own
reproduction and take on the helping role is often ascribed to kin
selection (Boomsma, 2009; Hamilton, 1964a, 1964b; Ratnieks,
Foster, &Wenseleers, 2006), which favours helping behaviour by
inclusive fitness benefits. The degree of helping is shaped by
relatedness between helper and recipient, the fitness costs of
helping and the benefit of helping to the recipient. Variation in
these terms between individuals may explain the frequently
observed differences between group members in helping effort
(Cant &Field, 2001; English, Browning, &Raihani, 2015; Huchard
et al., 2014). Furthermore, individuals in cooperatively breeding
societies often show differential task participation, leading to task
differentiation or even to specialized castes as in eusocial insects
(Jeanson &Weidenmüller, 2014; Oster &Wilson, 1978; Trumbo,
2012). In addition to kinship (Boomsma, 2009), behavioural varia-
tion and task differentiation may be influenced by proximate fac-
tors such as body size, age, place in the dominance hierarchy or
*Correspondence: A. Junghanns, Allgemeine und Systematische Zoologie, Anklamer
Straße20,D-17489Greifswald,Germany.
E-mail address: anja.junghanns@arcor.de (A. Junghanns).
1
Shared first authors.
2
Shared senior authors.
Contents lists available at ScienceDirect
Animal Behaviour
journal homepage: www.elsevier.com/locate/anbehav
http://dx.doi.org/10.1016/j.anbehav.2017.08.006
0003-3472/©2017 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
Animal Behaviour 132 (2017) 101e107

behavioural type (Carter, English, &Clutton-Brock, 2014; Clutton-
Brock, 2002; Emlen &Wrege, 1991; Monnin &Peeters, 1999;
Pruitt, Oufiero, Avil
es, &Riechert, 2012; Radnieks &Anderson,
1999). One aspect that is not well understood is whether and/or
how the reproductive state of individuals shapes their propensity
to provide different types of help, that is, whether being mated or
virgin influences their absolute and relative engagement in (1)
alloparental care and (2) other helping tasks within cooperatively
breeding societies.
Social spiders are cooperative breeders that share a communal
nest and collaborate in prey capture, nest defence and brood care
(Avil
es, 1997; Lubin &Bilde, 2007). Sociality has evolved indepen-
dently at least 20 times in seven spider families, and phylogenetic
patterns suggest that it is derived from a subsocial state (Agnarsson,
Avil
es, Coddington, &Maddison, 2006; Settepani, Bechsgaard, &
Bilde, 2016), by elimination of premating dispersal and the forma-
tion of family groups. Mating and reproduction take place in the nest
among related group members, which results in extreme inbreeding
and high genetic relatedness (Agnarsson, Avil
es, &Maddison, 2013;
Settepani, Bechsgaard, &Bilde, 2014; Settepani et al., 2017). Social
spider species show a strong female-biased sex ratio; however, only
a small proportion of females reproduce (Avil
es, 1997; Lubin &Bilde,
2007; Salomon, Mayntz, &Lubin, 2008). As females mature asyn-
chronously and males mature and die early, many females in a nest
remain unmated (Grinsted, Breuker, &Bilde, 2014; Salomon &
Lubin, 2007). These females are assumed to take on the role of
helpers in brood care and other tasks in the nest (Lubin &Bilde,
2007; Salomon &Lubin, 2007). Previous studies showed that fe-
males of uncertain reproductive state direct care to offspring of
other group members (Christenson, 1984; Kraus, 1988; Kullmann,
Sittertz &Zimmermann, 1971; Samuk &Avil
es, 2013). Notably, in
the context of cooperative breeding and alloparental care, the role of
mating status has not been investigated, and we currently do not
know whether virgin females actually provide allomaternal care.
We aimed to fill this gap by assessing the role of virgin females in
cooperative breeding, and to investigate whether mating state
shapes patterns of allomaternal care.
In the genus Stegodyphus, which contains three permanently
social species, females show extreme and suicidal maternal care
that includes eggsac construction, tending and guarding of eggsacs,
regurgitation feeding of the hatched spiderlings and matriphagy
where females are consumed by the offspring (Schneider, 2002;
Seibt &Wickler, 1987, 1988). In the social Stegodyphus dumicola,
brood size and offspring growth rate increase in the presence of
mated but nonreproducing females, suggesting that helpers ac-
quire indirect benefits and promote group productivity (Salomon &
Lubin, 2007). However, since about 60% of females in a nest of
S. dumicola remain unmated (Salomon et al., 2008), it is essential to
investigate the contribution of these females to the various tasks to
understand group organization. To this aim, we asked whether
virgin females perform allomaternal care to the offspring of
reproducing females, and whether the brood care provided in-
cludes all activities from eggsac care and regurgitation feeding to
matriphagy. Interestingly, in the subsocial solitarily breeding
congener Stegodyphus lineatus, only mated females that have pro-
duced an eggsac provide maternal care, suggesting that brood care
and regurgitation feeding are triggered by a preceding reproductive
event (Schneider, 2002). Maternal care behaviours in virgin
S. dumicola females would therefore indicate an adaptation to
cooperative breeding in social species (Jones, Riechert, Dalrymple,
&Parker, 2007; Schneider, 2002).
Accumulating evidence for behavioural specialization in prey
capture, web construction and defence behaviour, as well as brood
care, shows that despite the lack of morphological differentiation,
social or facultative social spiders exhibit some degree of task
differentiation (Pruitt &Riechert, 2011; Settepani, Grinsted,
Granfeldt, Jensen, &Bilde, 2013; Wright, Holbrook, &Pruitt,
2014; but see Ainsworth, Slowtow, Crouch &Lubin, 2002;;
Settepani, Bilde, &Grinsted, 2015). Although there is some evi-
dence for unequal task participation among subordinate and
(presumably) nonreproducing helpers in various animal groups
(Cant, 2003; Emlen &Wrege, 1991; Monnin &Peeters, 1999;
Mooney, Filice, Douglas, &Holmes, 2015), in general we have lit-
tle knowledge about how the reproductive state of individuals
shapes behavioural specialization. Therefore, we asked whether the
relative investment in reproductive and nonreproductive tasks
differs between virgin females and mothers. If there is differential
task participation between virgins and mothers, we expected virgin
females to specialize on prey capture, which is assumed to be a
risky task that exposes individuals to injury, parasitism or preda-
tion (Bradoo, 1980; Griswold &Meikle, 1990; Henschel, 1998;A.
Junghanns &C. Holm, personal observation 2013; V. Settepani,
personal communication 13 March 2017), whereas mothers are
predominantly occupied with brood care.
METHODS
Study Species
The genus Stegodyphus contains at least 18 subsocial species
(Kraus &Kraus, 1988) in which offspring share a nest with the
mother for an extended period of brood care. The mother provides
the spiderlings with a nourishing fluid through regurgitation
feeding for a period of several weeks and eventually is eaten by her
offspring (matriphagy; Kullmann, Nawabi, &Zimmermann, 1971).
After matriphagy, the spiderlings of the subsocial species disperse
and live solitarily. In three species in this genus, including
S. dumicola, spiderlings do not disperse but live socially throughout
their lives (Kraus &Kraus, 1988). Stegodyphus dumicola can be
found in dry and warm habitats of southern Africa (Kraus &Kraus,
1988). The spiders inhabit a dense retreat of silk and plant material
with one or more two-dimensional capture webs attached. New
nests are founded by local budding or fission of nests or by prop-
agule dispersal by a single mated female (Bilde et al., 2007). Nests
grow over several generations to a size of a few tens up to more
than a thousand individuals. Males mature early and mate with the
first females that become adult within their natal nest (Henschel,
Lubin, &Schneider, 1995). Male dispersal is limited to inter-
connected nearby nests (Lubin, Birkhofer, Berger-Tal, &Bilde,
2009), which are usually closely related (Johannesen, Hennig,
Dommermuth, &Schneider, 2002) and at a similar develop-
mental stage (A. Junghanns &C. Holm, personal observation 2013;
Salomon et al., 2008). This highly inbred mating system results in a
high relatedness between members of a nest (Agnarsson et al.,
2013). Males die within a few weeks after maturation while fe-
males mature over a period of several months (Henschel et al.,
1995). This can result in a situation with about 60% of all females
of a nest remaining unmated (Salomon et al., 2008).
Point-sampling Trials
Groups of S. dumicola were composed as specified below. They
were checked daily for eggsac care behaviour and during feeding
bouts prey attack behaviour was recorded. Since the recording rule
resulted in sample points over the course of the observation phase,
we refer to these trials as point-sampling trials.
Collection site, spider maintenance and group composition
Twenty-four nests (called source nests hereafter) of S. dumicola
were collected in South Africa in November 2013 from Shingwedzi
A. Junghanns et al. / Animal Behaviour 132 (2017) 101e107102

(22.98S, 31.30E), Middlefontein (24.68S, 28.55E) and Mokopane
(24.40S, 28.78E) before females reached adulthood. We separated
juvenile and subadult (penultimate instar) females from subadult
and adult males to ensure virginity of the females. Immature in-
dividuals were raised to adulthood on a diet of crickets and house
flies provided twice per week. After the final moult to maturity, a
male and a female from the same nest were allowed to mate
overnight in a separate container. The following day, the female
was checked under a stereoscope for presence of secretions on her
genital area (epigyne), which indicates a previous mating event (A.
Junghanns, personal observation 2014). These females were used as
mothers in the set-ups. Virgin females in our trials had no contact
with males. As maturation times differed within and between
natural nests, groups were established over a period of 9 weeks
between November 2013 and January 2014 and were checked daily
for oviposition events.
To test whether virgin females provide allomaternal care, we
arranged 192 groups, each group including three virgin females and
two mated females from the same nest to increase the chance of at
least one eggsac being produced per group. Spiders were individ-
ually marked with water-based acrylic colours. Virgin females
received a blue, yellow or green marking on the dorsal side of their
opisthosoma, mated females a red or orange marking. Each group
was housed in a transparent plastic container (122 82 mm and
52 mm high) with a plastic ring (diameter 53 mm) in the centre of
the box for silk attachment. Mesh on two opposite sides of the
container ensured sufficient ventilation. The groups were kept in a
climate chamber at 25

C with a 13:11 h (light:dark) photoperiod.
The groups were fed two or three times per week with house flies,
Musca domestica, and crickets, Gryllus bimaculatus, and water was
sprayed in the boxes every second day. Observations on brood care
behaviour were conducted during 18 weeks between January,
when mothers started to oviposit, and May 2014. Feeding trials
were conducted over a period of 10 weeks between January and
April 2014 with a total of 24 observations per group.
Brood care behaviour
Groups included in the analysis (N¼171) produced at least two
eggsacs, and because females produce only one eggsac at a time
(which can be replaced), it is a reasonable assumption that both
mated females reproduced. All groups were checked daily as soon
as oviposition occurred and we noted whether the spider tending
the eggsac was a virgin female or a mother. A spider touching,
handling, transporting or sitting on an eggsac was considered as
tending it. If no spider was tending an eggsac this was likewise
recorded. In 217 cases, more than one female cared for one or more
eggsacs simultaneously. Only groups that were checked on more
than 20 occasions for eggsac tending behaviour were included.
Consequently, the statistical analysis is based on 825 individuals
from 165 groups that led to a total of 3719 observed cases of one or
more females caring for an eggsac.
Attack behaviour
To investigate whether virgin and mated females differ in attack
behaviour we only used those groups in which oviposition occurred
during the observation period (N¼171 groups including 860 in-
dividuals). During feeding events, we recorded the identity and
reproductive state of the first attacking spider for each group. Only
trials with one (N¼2483) or no (N¼1645) first attacker were
considered for the analysis (total N¼4128). No attack was scored
for feeding trials where no attack occurred within 1 h after pre-
senting the prey item. To assess whether attack behaviour of virgin
females and mothers changed in the presence of an eggsac, we
analysed those groups that had been observed at least five times
before and five times after oviposition. Consequently, 113 groups
including 565 individuals were included in the statistical analysis
on eggsac-dependent attack behaviour.
Statistical analysis
We analysed the data on eggsac care and attack behaviour using
generalized linear mixed models (GLMM) in the R package lme4
(Bates, Maechler, Bolker, &Walker, 2015)inR(R Core Team, 2016).
To investigate whether there was an effect of reproductive state on
eggsac care a mixed logistic regression was performed. The
response variable was the number of observations on eggsac care
behaviour out of the total number of observations per individual
spider (probability of caring for an eggsac). The reproductive state
(virgin female or mother) was incorporated as a fixed effect. Source
nest, experimental group and set-up date of the group were
included as random effects. We detected minor but unproblematic
levels of overdispersion in the model (Zuur, Ieno, Walker, Saveliev,
&Smith, 2009). To estimate the Pvalue of reproductive state, we
used a likelihood ratio test to compare the full model with a
reduced model only containing the random effects.
To address the question whether reproductive state affected
attack behaviour, we performed a mixed logistic regression in
which the response variable was the number of attacking first out
of the total number of feeding trials per individual spider (proba-
bility of being the first attacker). As above, reproductive state was
included as a fixed effect, while the source nest, the experimental
group and set-up date of the group were included as random ef-
fects. Since experimental group and the set-up date did not
contribute to explaining any variance of the full model (estimated
variance ¼0), these factors were removed from the model. We
detected minor but unproblematic levels of overdispersion in the
model (Zuur et al., 2009). To estimate the Pvalue of reproductive
state, we used a likelihood ratio test to compare the full model with
a reduced model only containing the random effect of source nest.
We ran another mixed logistic regression to test whether the
presence of an eggsac affected the attack behaviour of virgins and
mothers. As we only used groups that were observed at least five
times before as well as after oviposition, the number of groups was
reduced to 113 and the 24 trials of each group were split into two
unevenly sized sets (one with eggsac presence and one without).
The response variable was the number of attacks out of the total
number of feeding trials per individual spider (probability of being
the first attacker before versus after oviposition for each individ-
ual). The fixed effects were reproductive state and presence of an
eggsac (yes or no) as well as their two-way interaction. Sourcenest,
experimental group, set-up date of the group and individual (as
each individual has two counts) were included as random factors,
but as in the previous model, experimental group and set-up date
did not explain any variance and were removed from the full model.
We did not detect overdispersion in the model. We performed
sequential model reduction and compared models with likelihood
ratio tests to obtain Pvalues for model components related to ef-
fects of eggsac (eggsac)reproductive state and eggsac).
Continuous Observations
Regurgitation behaviour is more difficult to witness than
eggsac care, as the spiders produce a denser web before spider-
lings hatch. Furthermore, females are more susceptible to dis-
turbances in the regurgitation phase than in the eggsac care
phase, and often stop regurgitating when the container is moved
for inspection or feeding. Hence, it was not possible to quantita-
tively and reliably observe regurgitation feeding during point-
sampling trials. We therefore established a second set of groups
to investigate individual participation in regurgitation feeding
and events of matriphagy. Groups were observed continuously
A. Junghanns et al. / Animal Behaviour 132 (2017) 101e107 103

during several time periods after hatching of spiderlings (see
below) and regurgitation feeding events that took place on the
surface or entrances of the nest were recorded. We refer to the
trials that led to a qualitative assessment of regurgitation behav-
iour as continuous observations.
Collection site, spider maintenance and group composition
Four nests were collected in November 2014 in South Africa at
Shingwedzi (22.98S, 31.30E) and Skukuza (24.99S, 31.60E). An
adult female from the field was classified as mated if secretions on
her genital area were detected. Females that matured in the labo-
ratory without males present were used as virgin females. Groups
consisting of one mated female and three virgin females from the
same nest were established. Spiders were marked with an indi-
vidual pattern of acrylic colour on their opisthosomata. Groups
were kept in a climate chamber at 25

C under a 13:11 h (light:dark)
photoperiod in transparent plastic containers (122 82 mm and
52 mm high) with mesh on two opposite sides and a plastic ring
(diameter 53 mm) in the centre of the box. Each group was fed with
three or four house flies and provided with water three times per
week. When the hatching of spiderlings was observed (in a total of
eight groups) feeding was halted, as matriphagy can be delayed in
well-fed groups (A. Junghanns &C. Holm, personal observation
2014). In one group, one virgin female died before the spiderlings
hatched, resulting in a total of 23 virgin females and eight mothers
that were assessed for the probability of matriphagy. Of these fe-
males, we scrutinized 14 virgin females and five mothers for
regurgitation behaviour using video recording.
Observations of regurgitation feeding
Five groups were observed for regurgitation feeding during 8
weeks between March and May 2015 using network cameras
(Vivotek ip8332; frame rate 30 fps; resolution: 1280 800;
infrared lights). The beginning of the regurgitation feeding phase
is indicated by one or more spiderlings circling around and
climbing on the female and finally congregating at her mouth
region. During regurgitation, the female's prosoma shows slight
vertical and horizontal movements and when her mouthparts are
in full view a liquid substance can be observed appearing between
her chelicerae.
Observations of matriphagy
Whether matriphagy occurred was documented for each indi-
vidual of all groups in which spiderlings had hatched (N¼8). After
the regurgitation phase, spiderlings gather around the body of the
female and feed on it for some hours. We recorded matriphagy
when this behaviour resulted in death and a dried-up body of the
female.
Ethical Note
We held permits for collecting nests of S. dumicola in South
Africa (CPM19821, BILDT1008). The spiders were kept under
favourable conditions and allowed to follow their natural life cycle.
RESULTS
Brood Care Behaviour
Groups in which oviposition occurred (N¼171) produced at
least two eggsacs during the observation period, suggesting that
both mated females reproduced. Our point-sampling trials showed
that both virgin females and mothers tended the eggsacs (virgin
females: 98% of groups, N¼161 of 165; mothers: 97% of groups,
N¼160; both virgin females and mothers: 95% of groups, N¼157).
In one case, we observed a virgin female using her own silk to close
an eggsac produced by a mother. Overall, virgin females tended the
eggsacs significantly less than mothers (X
2
1
¼135.94, P<0.001;
Fig. 1a). Excluding observations in which more than one female was
observed at an eggsac (N¼217) did not alter the results
(X
2
1
¼139.33, P<0.001). Regurgitation feeding occurred in all five
groups that were video-observed and was performed by all virgins
(N¼14) and mothers (N¼5). Matriphagy occurred in all eight
groups that were inspected at the end of the regurgitation phase:
35% of virgins (eight of 23) and 25% of mothers (two of eight) were
consumed by the end of the observation period. As to the remaining
females, 12 virgin females died of unknown causes without being
consumed by the spiderlings and three survived the observation
period, whereas two mothers died of unknown causes and four
survived the observation period.
Attack Behaviour
Females of different reproductive states differed in mean attack
probability with virgins being more likely to engage in prey attack
than mothers (X
2
1
¼6.51, P¼0.01; Fig. 1b). Neither the presence of
an eggsac (X
2
1
¼1.08, P¼0.30) nor the interaction between eggsac
and reproductive state (X
2
1
¼1.15, P¼0.28) had a significant effect
on the probability of being the first attacker in virgin females
(before oviposition: 0.120; after oviposition: 0.117) or in mothers
(before oviposition: 0.128; after oviposition: 0.115).
DISCUSSION
We investigated whether virgin females of the social spider
S. dumicola perform brood care and show the same repertoire of
brood care behaviours as mothers. Our results demonstrate that
virgin females show extreme allomaternal care by engaging in
eggsac care and regurgitation feeding of the spiderlings, and that
they are, like mothers, consumed by spiderlings. Even though vir-
gin females engaged in the same tasks as mothers, they were more
likely to engage in prey attack, irrespective of whether an eggsac
was present in the group, while mothers were more likely to tend
eggsacs. Although the degree of differential task participation was
small, it suggests that reproductive state determines some degree
of behavioural specialization between virgin females and mothers.
In many cooperatively breeding societies with alloparental care,
helpers are adult individuals that care for the offspring of other
adults (Clutton-Brock, 2002; Cockburn, 1998). In social spiders,
brood care involves not only tending eggsacs but also costly be-
haviours of food provisioning by means of regurgitation of midgut
contents and liquefied midgut tissue (Nawabi, 1974; Salomon,
Aflalo, Coll, &Lubin, 2015), and matriphagy by the offspring
(Kullmann, Nawabi et al., 1971; Salomon, Schneider, &Lubin, 2005).
Although allomaternal care has been demonstrated for social
Stegodyphus and Anelosimus species (Christenson, 1984; Furey,
1998; Grinsted, Agnarsson, &Bilde, 2012; Kraus, 1988; Kullmann,
Sitterz et al., 1971; Samuk &Avil
es, 2013), it was unknown
whether mating status determines the ability and propensity for
individuals to perform extreme brood care relative to other tasks in
the group. In a fostering study on the solitarily breeding S. lineatus,
only females that had produced their own young would feed
offspring of other females, suggesting that only mothers are in the
right reproductive or physiological state to provide allomaternal
care (Schneider, 2002). Similarly, in Coelotes terrestris, a solitary
spider with extended brood care, foreign spiderlings are accepted
only by females that previously cared for their own offspring
(Bess
ekon &Horel, 1996). In contrast, our study shows that virgin
females of the social S. dumicola provide extensive brood care. This
ability seems to represent an adaptation to cooperative breeding
A. Junghanns et al. / Animal Behaviour 132 (2017) 101e107104

and entails a shift in the trigger for brood care behaviours from
mated to virgin females.
In S. dumicola, a large proportion of females remain unmated
(Salomon et al., 2008), and their engagement in allomaternal care
must enhance the growth and survival of the offspring in the nest.
In return, these virgin females gain considerable inclusive fitness
benefits because of the high genetic relatedness among individuals
in the nest (Lubin &Bilde, 2007; Settepani et al., 2017). The phys-
iological ability of virgins to provide extreme brood care may also
provide assurance against the death of reproductive females
(Schneider &Lubin, 1997). In the subsocial S. lineatus, more than
50% of females with an eggsac die before their spiderlings hatch
(Schneider, 1996), and in S. dumicola up to 90% of single-female
nests go extinct (Bilde et al., 2007; see also ; Henschel, Schneider,
&Meikle, 1996; Henschel, 1998). As the survival of the offspring
relies heavily on intensive brood care, single females that die before
matriphagy will have no or low reproductive output (Schneider,
2002). In contrast, in multifemale nests, allomaternal care by fe-
males of unknown mating status was demonstrated to secure
growth and survival of offspring even when the mother died (Jones
et al., 2007; Kullmann, 1972; Lubin &Bilde, 2007). It remains to be
investigated whether virgin females are flexible in altering their
participation in various tasks in response to the needs of the
breeding group (Mooney et al., 2015).
High relatedness among group members is expected to reduce
the cost of helping in individuals that are unlikely to reproduce
compared to individuals with a high chance of reproduction, which
can ultimately lead to worker specialization (Boomsma, 2009). For
example, nonbreeding and breeding banded mongoose, Mungos
mungo, appear to be totipotent, but subordinate males increase
their contribution to costly babysitting of pups at the time of day
presumed to be the most energetically demanding, and at oestrus
when dominant males and females reduce their babysitting effort
(Cant, 2003). We found that in S. dumicola, females are largely
totipotent but our point-sampling trials revealed a propensity to
engage in different tasks between virgin females and mothers.
Virgin females were less likely to tend eggsacs than mothers, and
more likely to undertake the risky prey capture that requires
abandoning the safe retreat, attack of a potentially dangerous and
defensive prey (Bradoo, 1980; Griswold &Meikle, 1990; Henschel,
1998; A. Junghanns &C. Holm, personal observation 2013; V. Set-
tepani, personal communication 13 March 2017) and the costly use
of venom to subdue the prey (Morgenstern &King, 2013). It is
possible that the participation of virgin females in prey attack en-
hances foraging success by improving the nest's prey capture effi-
ciency (Pasquett &Krafft, 1992) and simultaneously allows
reproductive females to direct more resources into brood care
while reducing their risk of injury during prey attack. This behav-
ioural differentiation, together with recent evidence for individual
specialization (Settepani et al., 2013; Pruitt &Riechert, 2011;
Wright et al., 2014), suggests a more complex social organization
in social spiders than previously recognized. Because relatedness
within the nest is extremely high (Lubin &Bilde, 2007; Settepani
et al., 2017), the nest represents the reproductive unit, and fitness
is defined by nest productivity and not solely by individual repro-
ductive success (Boomsma, 2009; Grinsted &Bilde, 2013; Keller,
1999). High intranest relatedness should reduce conflict over
reproduction because indirect fitness of helpers will be similar to
direct fitness, thereby aligning the interests of individuals. If
reproductive conflict is reduced, and task differentiation is bene-
ficial (Dukas &Visscher, 1994; Julian &Cahan, 1999; Wright et al.,
2014), this may further promote behavioural differentiation.
Moreover, since social spiders are sedentary and dependent on the
prey that arrives in their capture web, they are particularly limited
by resources (Bilde et al., 2007), which has been proposed to favour
a life history strategy of investing in fewer but larger offspring
(Grinsted et al., 2014). Cooperative breeding and behavioural dif-
ferentiation may be adaptations that increase reproductive success
of the group under ecological constraints (Clutton-Brock, 2002;
Emlen, 1995; Grinsted et al., 2014).
The higher engagement by virgins in prey attack may also be
explained by virgins trying to secure resources for a later repro-
ductive event. Our results demonstrate that, even before eggsac
production, virgin females engaged relatively more in foraging
behaviour, and it is known that the first attacking spider gains more
weight (Amir, Whitehouse, &Lubin, 2000; Whitehouse &Lubin,
1999). However, we consider it unlikely that virgin females forage
more to promote their own reproductive success, because only
early maturing females mate. Males in a nest are short lived, there
is little male dispersal activity between nests (Lubin et al., 2009)
and nearby nests are fairly synchronous (Bilde et al., 2007; Salomon
et al., 2008). Hence, we assume that the resources acquired by
virgin females will directly benefit the growth and survival of
offspring in the nest by means of allomaternal care. Furthermore,
by engaging more in prey capture, virgin females may reduce the
energy expended by mated females and at the same time provide
0.06
0.055
0.05
0.045
0.035
0.04
0.03
0.14
0.13
0.12
0.11
0.1
Eggsac care (proportion)
Prey attack (proportion)
Vir
g
in females Mothers Vir
g
in females Mothers
(a) (b)
Figure 1. Task differentiation in S. dumicola. (a) Probability of eggsac care in virgin females and mothers. (b) Probability of virgin females and mothers being the first to attack prey
during feeding trials. The individual probabilities represent the number of observed behaviours out of the total number of observations per spider. Error bars show standard error of
the mean.
A. Junghanns et al. / Animal Behaviour 132 (2017) 101e107 105

them with food. Acquiring sufficient amounts of energy may be
necessary for mated females to overcome a threshold for successful
oviposition (Drent &Daan, 1980) from which virgin females profit
through inclusive fitness.
The mechanisms determining whether a female stays virgin and
becomes an allomother or reproduces remain to be investigated
(Grinsted &Bilde, 2013). It is possible that reproductive ability is
determined by resource availability during early development that
shapes a certain developmental trajectory of an individual
(Beldade, Mateus, &Keller, 2011; English et al., 2015; Mayntz,
Raubenheimer, Salomon, Toft, &Simpson, 2005; Miura, 2005;
Nijhout, 2003; Salomon et al., 2008). Scramble competition in the
nest may imply that some females do not develop beyond a specific
threshold in time (Whitehouse &Lubin, 1999), and, as pointed out
above, late maturing females are unlikely to be able to reproduce
and are therefore destined to become allomothers. However, that
the helper role is not determined solely by mating status is indi-
cated by observations that mated females can also act as allo-
mothers (Kullmann, Nawabi et al., 1971), and even immature
females are reported as helpers (Christenson, 1984; G
omez, Rojas-
Buffet, &Viera, 2015; Viera, Ghione, &Costa, 2005). In general, we
have little understanding of how early life experience influences
‘social trajectories’(English et al., 2015; Z€
ottl, Thorley, Gaynor,
Bennett, &Clutton-Brock, 2016), and to what extent individuals
have control over the flexibility to move between social trajectories
or to specialize (Carter et al., 2014; Huchard et al., 2014; Jeanson &
Weidenmüller, 2014). The degree to which individuals specialize or
remain flexible may further depend on what type of information
they have about the fitness outcomes of their decision (Reale,
Reader, Sol, McDougall, &Dingemanse, 2007; English et al., 2015;
Holman, 2014). When direct and indirect fitness outcomes are
aligned, as in the social spiders, the ability of totipotent helpers to
modify their effort and task strategically may evolve in response to
ecological and social parameters.
Acknowledgments
We dedicate this publication to Otto and Margarete Kraus for
their fundamental and inspiring work on Stegodyphus brood care
and systematics. An enormous thank you to Michelle Greve for
hosting and supporting A.J. and C.H. in Pretoria, South Africa. We
thank Tom Tregenza and Rolando Rodríguez Mu~
noz for help with
setting up video recording and Pierick Mouginot for fruitful sta-
tistical discussions. Yael Lubin provided very helpful suggestions on
the manuscript. We thank the members of The Spider Lab at Aarhus
University and the spider group at the University of Greifswald for a
stimulating work environment. C.H. was supported by ERC StG-
2011-282163 to T.B. and A.J. by a stipend from the federal state of
Mecklenburg-Vorpommern, Germany.
References
Agnarsson, I., Avil
es, L., Coddington, J. A., & Maddison, P. W. (2006). Sociality in
theridiid spiders: Repeated origins of an evolutionary dead end. Evolution, 60,
2342e2351.
Agnarsson, I., Avil
es, L., & Maddison, W. P. (2013). Loss of genetic variability in social
spiders: Genetic and phylogenetic consequences of population subdivision and
inbreeding. Journal of Evolutionary Biology, 26,27e37.
Ainsworth, C., & Slotow, R. (2002). Lack of task differentiation during prey capture
in the group living spider Stegodyphus mimosarum (Araneae, Eresidae). Journal
of Arachnology, 30,39e46.
Amir, N., Whitehouse, M. E. A., & Lubin, Y. (2000). Food consumption rates and
competition in a communally feeding social spider, Stegodyphus dumicola
(Eresidae). Journal of Arachnology, 28,195e200.
Avil 
es, L. (1997).Causes and consequences of cooperation and permanent-sociality in
spiders. In J. C. Choe, & B. J. Crespi (Eds.), The evolution of social behavior in insects
and arachnids (pp. 476e498). Cambridge, U.K.: Cambridge University Press.
Bates, D., Maechler, M., Bolker, B., & Walker, S. (2015). Fitting linear mixed-effects
models using lme4. Journal of Statistical Software, 67,1e48.
Beldade, P., Mateus, A. R. A., & Keller, R. A. (2011). Evolution and molecular mech-
anisms of adaptive developmental plasticity. Molecular Ecology, 20, 1347e1363.
Bess
ekon, D. A., & Horel, A. (1996). Social-maternal relations in Colotes terrestris
(Araneae, Agelenidae): Influence of the female reproductive state on its toler-
ance towards conspecific spiderlings. Behavioural Processes, 36,19e26.
Bilde, T., Coates, K. S., Birkhofer, K., Bird, T., Maklakov, A. A., Lubin, Y., et al. (2007).
Survival benefits select for group living in a social spider despite reproductive
costs. Journal of Evolutionary Biology, 20,2412e2426.
Boomsma, J. J. (2009). Lifetime monogamy and the evolution of eusociality. Philo-
sophical Transactions of the Royal Society B, 364,3191e3207.
Bradoo, B. L. (1980). Feeding behaviour and recruitment display in the social spider
Stegodyphus sarasinorum Karsch (Aranae, Eresidae). Tijdschrift voor Entomologie,
123,89
e104.
Cant, M. A. (2003). Patterns of helping effort in co-operatively breeding banded
mongooses (Mungos mungo). Journal of the Zoological Society London, 259,
115 e121.
Cant, M. A., & Field, J. (2001). Helping effort and the future fitness in cooperative
animal societies. Proceedings of the Royal Society B, 268, 1959e1964.
Carter, A. J., English, T. H., & Clutton-Brock, T. H. (2014). Cooperative personalities
and social niche specialization in female meerkats. Journal of Evolutionary
Biology, 27,815e825.
Christenson, T. (1984). Behaviour of colonial and solitary spiders of the theridiid
species Anelosimus eximus.Animal Behaviour, 32,725e734.
Clutton-Brock, T. (2002). Breeding together: Kin selection and mutualism in
cooperative vertebrates. Science, 296,69e72.
Cockburn, A. (1998). Evolution of helping behaviour in cooperatively breeding birds.
Annual Review of Ecology and Systematics, 29,141e177.
Cockburn, A. (2006). Prevalence of different models of parental care in birds. Pro-
ceedings of the Royal Society B, 273,1375e1383.
Drent, R. H., & Daan, S. (1980). The prudent parent: Energetic adjustments in avian
breeding. Ardea, 68, 225e252.
Dukas, R., & Visscher, K. (1994). Lifetime learning by foraging honey bees. Animal
Behaviour, 48,1007e1012 .
Emlen, S. T. (1995). An evolutionary theory of the family. Proceedings of the National
Academy of Sciences, 92, 8092e8099.
Emlen, S. T., & Wrege, P. H. (1991). Breeding biology of white-fronted bee-eaters at
Nakuru: The influence of helpers on breeder fitness. Journal of Animal Ecology,
60, 309e326.
English, S., Browning, L. E., & Raihani, N. J. (2015). Developmental plasticity and
social specialization in cooperative societies. Animal Behaviour, 106,37e42.
Furey, R. E. (1998). Two cooperatively social populations of the theridiid spider
Anelosimus studiosus in a temperate region. Animal Behviour, 55,727e735.
G
omez, D., Rojas-Buffet, C., & Viera, C. (2015). The influence of regurgitation by
sisters on male growth in the subsocial spider Anelosimus vierae (Araneae,
Theridiiade). Arachnology, 16, 252e254.
Grinsted, L., Agnarsson, I., & Bilde, T. (2012). Subsocial behaviour and brood adop-
tion in mixed-species colonies of two theridiid spiders. Naturwissenschaften, 99,
1021 e1030.
Grinsted, L., & Bilde, T. (2013). Effects of within-colony competition on body size
asymmetries and reproductive skew in a social spider. Journal of Evolutionary
Biology, 26, 553e561.
Grinsted, L., Breuker, C. J., & Bilde, T. (2014). Cooperative breeding favours maternal
investment in size over number of eggs in spiders. Evolution, 68, 1961e1973.
Griswold, C. E., & Meikle, T. C. (1990). Social life in a web.Natural History, 3/90,6e10.
Hamilton, W. D. (1964a). Genetical evolution of social behaviour. I. Journal of
Theoretical Biology, 7,1e16.
Hamilton, W. D. (1964b). Genetical evolution of social behaviour. II. Journal of
Theoretical Biology, 7,17e52.
Henschel, J. R. (1998). Predation on social and solitary individuals of the spider
Stegodyphus dumicola (Araneae, Eresidae). Journal of Arachnology, 26,61e69.
Henschel, J. R., Lubin, Y. D., & Schneider, J. (1995). Sexual competition in an
inbreeding social spider, Stegodyphus dumicola (Araneae, Eresidae). Insectes
Sociaux, 42,419e426.
Henschel, J. R., Schneider, J., & Meikle, T. (1996). Does group-living or aggregation of
spiders of the genus Stegodyphus affect parasitism by pompiid wasps? Bulletin
of the British Arachnological Society, 10,138e14 0 .
Holman, L. (2014). Conditional helping and evolutionary transitions to eusociality
and cooperative breeding. Behavioural Ecology, 25,1173e1182 .
Huchard, E., Charmantier, A., English, S., Bateman, A., Nielsen, J. F., & Clutton-
Brock, T. (2014). Additive genetic variance and developmental plasticity in
growth trajectories in a wild cooperative mammal. Journal of Evolutionary
Biology, 27, 1893e1904.
Jeanson, R., & Weidenmüller, A. (2014). Interindividual variability in social insects e
proximate causes and ultimate consequences. Biological Reviews, 89,671e687.
Johannesen, J., Hennig, A., Dommermuth, B., & Schneider, J. M. (2002). Mitochon-
drial DNA distributions indicate colony propagation by single matri-lineages in
the social spider Stegodyphus dumicola (Eresidae). Biological Journal of the
Linnean Society, 76,591e600.
Jones, T. C., Riechert, S. E., Dalrymple, S. E., & Parker, P. G. (2007). Fostering model
explains variation in levels of sociality in a spider system. Animal Behaviour, 73,
195e204.
Julian, G. E., & Cahan, S. (1999). Undertaking specialization in the desert leaf-cutter
ant Acromyrmex versicolor.Animal Behaviour, 58,437e442.
Keller, L. (1999). Levels of selection in evolution. Princeton, NJ: Princeton University
Press.
A. Junghanns et al. / Animal Behaviour 132 (2017) 101e107106

Keller, L., & Reeve, H. K. (1994). Partitioning of reproduction in animal societies.
Trends in Ecology &Evolution, 9,98e102 .
Kraus, M. (1988). Cocoon-spinning behaviour in the social spider Stegodyphus
dumicola (Arachnida, Araneae): Cooperating females as ‘helpers’.Verhandlungen
des Naturwissenschaftlichen Vereins in Hamburg (NF), 30, 305e309.
Kraus, O., & Kraus, M. (1988). The genus Stegodyphus (Arachnida, Araneae). Sibling
species, species groups and parallel origin of social living. Verhandlungen des
Naturwissenschaftlichen Vereins in Hamburg (NF), 30,151e254.
Kullmann, E. J. (1972). Evolution of social behavior in spiders (Araneae eEresidae
and Theridiidae). American Zoologist, 12,419e426.
Kullmann, E. J., Nawabi, S., & Zimmermann, W. (1971). Neue Ergebnisse zur Brut-
biologie cribellater Spinnen aus Afghanistan und der Serengeti (Araneae, Ere-
sidae). [New results from the reproductive biology of cribellate spiders from
Afghanistan and the Serengeti (Araneae, Eresidae)]. Zeitschrift des K€
olner Zoos,
14,87e108.
Kullmann, E. J., Sitterz, H., & Zimmermann, W. (1971). Erster Nachweis von
Regurgitationsfütterung bei einer cribellaten Spinne (Stegodyphus lineatus
Latreille, 1871, Eresidae). [First proof of regurgitation feeding in a cribellate
spider (Stegodyphus lineatus Latreille, 1871, Eresidae)]. Bonner Zoologische
Beitr€
age, 22,175e188.
Lubin, Y., & Bilde, T. (2007). The evolution of sociality in spiders. Advances in the
Study of Behavior, 37,83e145 .
Lubin, Y., Birkhofer, K., Berger-Tal, R., & Bilde, T. (2009). Limited male dispersal in a
social spider with extreme inbreeding. Biological Journal of the Linnean Society,
97,227e234.
Lukas, D., & Clutton-Brock, T. (2012). Cooperative breeding and monogamy in
mammalian societies. Proceedings of the Royal Society B, 279, 2151e2156 .
Mayntz, D., Raubenheimer, D., Salomon, M., Toft, S., & Simpson, S. J. (2005).
Nutrient-specific foraging in invertebrate predators. Science, 307,111e113.
Miura, T. (2005). Developmental regulation of caste-specific characters in social-
insect polyphenism. Evolution &Development, 7,122e129.
Monnin, T., & Peeters, C. (1999). Dominance hierarchy and reproductive conflicts
among subordinates in a monogynous queenles ant. Behavioural Ecology, 10,
323e332.
Mooney, S. J., Filice, D. C. S., Douglas, N. R., & Holmes, M. M. (2015). Task speciali-
zation and task switching in eusocial mammals. Animal Behaviour, 109,
227e233.
Morgenstern, D., & King, G. F. (2013). The venom optimization hypothesis revisited.
Toxicon, 63,120e128.
Nawabi, S. (1974). Histologische Untersuchungen an der Mitteldarmdrüse von Steg-
odyphus pacificus (Pocock 1900) (Araneae, Eresidae) (Doctoral dissertation).
Bonn, Germany: University of Bonn.
Nijhout, H. F. (2003). Development and evolution of adaptive polyphenisms. Evo-
lution &Development, 5,9e18.
Oster, G. F., & Wilson, E. O. (1978). Caste and ecology in the social insects. Princeton,
NJ: Princeton University Press.
Pasquett, A., & Krafft, B. (1992). Cooperation and prey capture efficiency in a social
spider, Anelosimus eximius (Araneae, Theridiidae). Ethology, 90, 121e133.
Pruitt, J. N., Oufiero, C. E., Avil
es, L., & Riechert, S. E. (2012). Iterative evolution of
increased behavioural variation characterizes the transition to sociality in spi-
ders and proves advantageous. American Naturalist, 180, 496e510.
Pruitt, J. N., & Riechert, S. E. (2011). Within-group behavioral variation promotes
biased task performance and the emergence of a defensive caste in a social
spider. Behavioural Ecology and Sociobiology, 65, 1055e1060 .
R Core Team. (2016). R: A language and environment for statistical computing. Vienna,
Austria: R Foundation for Statistical Computing. http://www.R-project.org/.
Radnieks, F. L. W., & Anderson, C. (1999). Task partitioning in insect societies.
Insectes Socieaux, 46,95e108.
Ratnieks, F. L. W., Foster, K. R., & Wenseleers, T. (2006). Conflict resolution in insect
societies. Annual Review of Entomology, 51, 581e608.
Reale, D., Reader, S. M., Sol, D., McDougall, P. T., & Dingemanse, N. J. (2007). Inte-
grating animal temperament within ecology and evolution. Biological Reviews,
82,291e318.
Salomon, M., Aflalo, E. D., Coll, M., & Lubin, Y. (2015). Dramatic histological changes
preceeding suicidal maternal care in the subsocial spider Stegdoyphus lineatus
(Araneae: Eresidae). Journal of Arachnology, 43,77e85.
Salomon, M., & Lubin, Y. (2007). Cooperative breeding increases reproductive suc-
cess in the social spider Stegodyphus dumicola (Araneae, Eresidae). Behavioral
Ecology and Sociobiology, 61,1743e1750.
Salomon, M., Mayntz, D., & Lubin, Y. (2008). Colony nutrition skews reproduction in
a social spider. Behavioral Ecology, 19, 605e611.
Salomon, M., Schneider, J. M., & Lubin, Y. (2005). Maternal investment in a spider
with suicidal maternal care, Stegodyphus lineatus (Araneae, Eresidae). Oikos, 109,
614e622.
Samuk, K., & Avil
es, L. (2013). Indiscriminate care of offspring predates the evolu-
tion of sociality in alloparenting spiders. Behavioral Ecology and Sociobiology, 67,
1275e1282.
Schneider, J. M. (1996). Differential mortality and relative maternal investment in
different life stages in Stegodyphus lineatus (Araneae, Eresidae). Journal of
Arachnology, 24,148e154 .
Schneider, J. M. (2002). Reproductive state and care giving in Stegodyphus (Araneae:
Eresidae) and the implications for the evolution of sociality. Animal Behaviour,
63,649e658.
Schneider, J. M., & Lubin, Y. (1997). Does high adult mortality explain semelparity in
the spider Stegodyphus lineatus (Eresidae)? Oikos, 79,92e100 .
Seibt, U., & Wickler, W. (1987). Gerontophagy versus Cannibalism in the social
spiders Stegodyphus mimosarum Pavesi and Stegodyphus dumicola Pocock. An-
imal Behaviour, 35, 1903e1905.
Seibt, U., & Wickler, W. (1988). Why do “familiy spiders”,Stegodyphus (Eresidae),
live in colonies? Journal of Arachnology, 16,193e198.
Settepani, V., Bechsgaard, J., & Bilde, T. (2014). Low genetic diversity and strong but
shallow population differentiation suggests genetic homogenization by meta-
population dynamics in a social spider. Journal of Evolutionary Biology, 27,
2850e2855.
Settepani, V., Bechsgaard, J., & Bilde, T. (2016). Phylogenetic analysis suggests that
sociality is associated with reduced effectiveness of selection. Ecology and
Evolution, 6, 469e477.
Settepani, V., Bilde, T., & Grinsted, L. (2015). Temporarily social spiders do not show
personality-based task differentiation. Animal Behaviour, 105,95e102 .
Settepani, V., Grinsted, L., Granfeldt, J., Jensen, J. L., & Bilde, T. (2013). Task
specialization in two social spiders, Stegodyphus sarasinorum (Eresidae) and
Anelosimus eximius (Theridiidae). Journal of Evolutionary Biology, 26,51e62.
Settepani, V., Schou, M. F., Greve, M., Grinsted, L., Bechsgaard, J., & Bilde, T. (2017).
Evolution of sociality in spiders leads to depleted genomic diversity at both
population and species levels. Molecular Ecology, 2017, 00,X1eX14.
Taborsky, M. (2009). Reproductive skew in cooperative fish groups: Virtue and
limitations of alternative modelling approaches. In R. Hager, & C. B. Jones (Eds.),
Reproductive skew in vertebrates: Proximate and ultimate causes (pp. 265e304).
Cambridge, U.K.: Cambridge University Press.
Trumbo, S. T. (2012). Patterns of parental care in invertebrates. In N. J. Royle,
P. T. Smiseth, & M. Kolliker (Eds.), The evolution of parental care (pp. 81e100).
Oxford, U.K.: Oxford University Press.
Viera, C., Ghione, S., & Costa, F. G. (2005). Regurgitation among penultimate juve-
niles in the subsocial spider Anelosimus cf. studiosus (Theridiidae): Are males
favoured? Journal of Arachnology, 34, 258e260.
Whitehouse, M. E. A., & Lubin, Y. (1999). Competitive foraging in the social spider
Stegodyphus dumicola.Animal Behaviour, 58,677e688.
Wilson, E. O. (1971). The insect societies. Cambridge, MA: Belknap Press.
Wright, C. M., Holbrook, C. T., & Pruitt, J. N. (2014). Animal personality aligns task
specialization and task proficiency in a spider society. Proceedings of the Na-
tional Academy of Sciences, 111, 9533e9537.
Z€
ottl, M., Thorley, J., Gaynor, D., Bennett, N. C., & Clutton-Brock, T. (2016). Variation
in growth of Damaraland mole-rats is explained by competition rather than by
functional specialization for different tasks. Biology Letters, 12, 20160820.
Zuur, A. F., Ieno, E. N., Walker, N. J., Saveliev, A. A., & Smith, G. M. (2009). Mixed
effects models and extensions in ecology with R. New York, NY: Springer.
A. Junghanns et al. / Animal Behaviour 132 (2017) 101e107 107