Content uploaded by Jennifer E Smith
Author content
All content in this area was uploaded by Jennifer E Smith on Dec 06, 2017
Content may be subject to copyright.
Fitness and hormonal correlates of social and ecological stressors of
female yellow-bellied marmots
Daniel T. Blumstein
a
,
b
,
*
, Kathryn N. Keeley
a
, Jennifer E. Smith
c
a
Department of Ecology and Evolutionary Biology, University of California Los Angeles, Los Angeles, CA, U.S.A.
b
The Rocky Mountain Biological Laboratory, Crested Butte, CO, U.S.A.
c
Department of Biology, Mills College, Oakland, CA, U.S.A.
article info
Article history:
Received 22 February 2015
Initial acceptance 20 March 2015
Final acceptance 30 September 2015
Available online
MS. number: A15-00144R2
Keywords:
ecological stressor
glucocorticoid
reproductive success
reproductive suppression
social rank
social stressor
yellow-bellied marmot
The effects of social and ecological stressors on female reproductive success vary among species and, in
mammals, previous reviews have identified no clear patterns. However, few studies have simultaneously
examined the relation between social rank and stressors and the relationships among rank, stressors and
reproductive success. We used a long-term data set to study free-living facultatively social yellow-bellied
marmots, Marmota flaviventris, to isolate the relationship between female social dominance rank and
faecal glucocorticoid metabolite (FGM) levels (our measure of basal stress) in adult females. In addition,
we examined whether rank and FGM levels were associated with reproductive success by quantifying
the probability of an individual successfully weaning a litter and, for those who did, litter size. High-
ranking females had lower FGM levels and larger litters. However, females with the highest FGM
levels were significantly more likely to wean a litter. Importantly, body condition (as measured by
previous year's mass) was also positively associated with breeding and with weaning larger litters. Thus,
although low-ranking females probably experienced more social stressors than high-ranking females
and although adult females often delayed their first reproduction until they were of a sufficient mass, our
results suggest that elevated baseline FGM levels failed to mediate reproductive suppression. Perhaps, in
species like marmots that have only a single chance per year to reproduce, reproductive suppression
should be rare. These results highlight the importance of social status, body condition and predator
abundance on determining reproductive success in highly seasonal breeders.
©2015 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
In vertebrates, the hypothalamicepituitaryeadrenal (HPA) axis
modulates reactions to stressors by producing glucocorticoid hor-
mones and restoring homeostasis (Reeder &Kramer, 2005).
Glucocorticoid production upon encountering a stressor varies
among species and among individuals and may be a repeatable trait
within individuals (Martínez-Mota, Valdespino, Rebolledo, &
Palme, 2008; Ramamoorthy &Cidlowski, 2013; Rensel &Schoech,
2011; Smith, Monclús, Wantuck, Florant, &Blumstein, 2012;
Tilbrook, Turner, &Clarke, 2000). Glucocorticoid responses may
be influenced by a number of factors, including social factors (e.g.
there are documented relationships between social rank and
glucocorticoid levels; discussed below), factors associated with
body condition (Williams, Kitaysky, Kettle, &Buck, 2008) and
ecological factors (e.g. predators and food availability; Creel,
Dantzer, Goymann, &Rubenstein, 2013).
The relationship between dominance and glucocorticoids is
particularly complex, varying enormously with respect to the
species involved, the breeding system, ecological contexts and the
means by which rank is achieved and maintained (Creel et al.,
2013). It is generally accepted that in relatively closed societies,
socially dominant individuals have better access to resources and
mates than do subordinates (Appleby, 1980) and that dominants
may direct aggressive behaviour (including aggressive threats) to-
wards subordinates to maintain their rank, discourage retaliatory
attacks or cause eviction (Ellis, 1995; Stockley &Bro-Jørgensen,
2011). However, even though socially dominant individuals
benefit from their increased priority of access to resources due to
their social status and, often as a result, reproductive dominance
(defined here as increased reproductive success), high-ranking in-
dividuals may experience high costs associated with the mainte-
nance of social rank (Creel, 2001; Gesquiere et al., 2011; Muller &
Wrangham, 2004). For instance, reproductively dominant females
*Correspondence: D. T. Blumstein, Department of Ecology and Evolutionary
Biology, University of California, 621 Young Drive South, Los Angeles, CA 90025-
1606, U.S.A.
E-mail address: marmots@ucla.edu (D. T. Blumstein).
Contents lists available at ScienceDirect
Animal Behaviour
journal homepage: www.elsevier.com/locate/anbehav
http://dx.doi.org/10.1016/j.anbehav.2015.11.002
0003-3472/©2015 The Association for the Study of Animal Behaviour. Published by Elsevier Ltd. All rights reserved.
Animal Behaviour 112 (2016) 1e11
among cooperatively breeding mammals, such as in African wild
dogs, Lycaon pictus (Creel, 2001), meerkats, Suricata suricatta
(Carlson et al., 2004), ringtailed lemurs, Lemur catta (Cavigelli,
1999; Cavigelli, Dubovick, Levash, Jolly, &Pitts, 2003), wolves,
Canis lupus (Sands &Creel, 2004; but see McLeod, Moger, Ryon,
Gadbois, &Fentress, 1996), and common marmosets, Callithrix
jacchus (Saltzman, Schultz-Darken, Scheffler, Wegner, &Abbott,
1994) have higher glucocorticoid levels than do conspecific sub-
ordinates. This is probably because reproductively dominant fe-
males are involved in more agonistic interactions to maintain their
high rank than are subordinates. Conversely, other studies have
found that socially subordinate individuals may experience
elevated glucocorticoid levels compared to their dominant coun-
terparts, perhaps due to reduced access to food and social support,
frequent harassment and reproductive suppression (Cavigelli &
Chaudhry, 2012). Such a pattern is seen in some primates
(reviewed by Abbott et al., 2003), alpine marmots, Marmota mar-
mota (Hackl€
ander, M€
ostl, &Arnold, 2003), African elephants, Lox-
odonta africana (Foley, Papageorge, &Wasser, 2001), and spotted
hyaenas, Crocuta crocuta (Goymann et al., 2001). In addition to
these opposite patterns, other studies have found no relationships
between female social rank and glucocorticoid levels (ringtailed
lemurs: Starling, Charpentier, Fitzpatrick, Scordato, &Drea, 2010;
meerkats: Barrette, Monfort, Festa-Bianchet, Clutton-Brock, &
Russell, 2012; common marmoset monkeys: Abbott, 1984; golden
lion tamarins, Leontopithecus rosalia:Bales, French, Hostetler, &
Dietz, 2005; Syrian hamsters, Mesocricetus auratus:Chelini,
Palme, &Otta, 2011; baboons, Papio hamadryas ursinus:
Crockford, Wittig, Whitten, Seyfarth, &Cheney, 2008; white rhi-
noceros, Ceratotherium simum simum:Metrione &Harder, 2011;
mandrills, Mandrillus sphinx:Setchell, Smith, Wickings, &Knapp,
2008; Ethiopian wolves, Canis simensis:van Kesteren et al., 2013).
Notably, we have listed multiple studies that documented
different patterns for the same species, such as in ringtailed lemurs,
meerkats and common marmosets. Given this variation and the
multitude of studies suggesting different patterns, generalizations
about social rank and glucocorticoid levels are not obvious.
Furthermore, because high glucocorticoid levels are a plausible
mechanism of reproductive failure (Bonier, Martin, Moore, &
Wingfield, 2009; Johnson, Kamilaris, Chrousos, &Gold, 1992;
Moberg, 1985; Munck, Guyre, &Holbrook, 1984; Sapolsky, 1992;
Scarlata et al., 2012; Welsh &Johnson, 1981), it is meaningful to
look beyond the rankestress relationship to better understand the
direct relationships between stressors and reproductive success. In
particular, female reproductive success can be meaningful to
examine because of the frequently high rates of reproductive failure
in female mammals (Wasser &Barash, 1983). In many group-living
animals, not all individuals breed or produce surviving offspring,
and reproductive suppression is a common reason for this (Abbott,
1987; Arnold &Dittami, 1997; Kaplan, Adams, Koritnik, Rose, &
Manuck, 1986; Rood, 1980; Wasser &Barash, 1983). In some sys-
tems, the socially dominant female is also the reproductively
dominant female. Furthermore, because socially subordinate fe-
males are usually the ones that are reproductively suppressed, it is
often inferred that suppression is a direct result of social stressors
caused by the unpredictable harassment, intimidation, evictions and
violence by the reproductively dominant female(s) (Abbott et al.,
2003; Hackl€
ander et al., 2003; Louch &Higginbotham, 1967;
Mendl, Zanella, &Broom, 1992; Young et al., 2006). Reduced ac-
cess to food resources, although indirect, could be another mecha-
nism of suppression (Ellis, 1995), especially since dominant females
have better body condition than their subordinates due to increased
access to food (Huang, Wey, &Blumstein, 2011).
Whereas social subordinates may have reduced reproductive
success because of the actions of more dominant individuals,
suppressing others may itself be costly and may therefore not be
always be favoured by natural selection. Indeed, Bell, Nichols,
Gilchrist, Cant, and Hodge (2012) found that when socially domi-
nant banded mongooses, Mungos mungo, suppressed subordinates,
the offspring born to dominant females weighed less than those
born to dominant females that did not suppress others. To avoid
such costs, female reproductive suppression is absent altogether in
some carnivores, such as in the African lions, Panthera leo, living in
egalitarian societies, because aggression is hypothesized to be
disadvantageous and cub survival depends on cooperative defence
against infanticidal males (Packer, Pusey, &Eberly, 2001).
There is therefore substantial variation among and within spe-
cies in the relationships between stress and reproduction. Some
studies have reported links between low social rank, high stress
and suppression (Hackl€
ander et al., 2003; Louch &Higginbotham,
1967; Wasser &Barash, 1983), while others have found no rela-
tionship between reproductive success and a female's glucocorti-
coid levels (Beehner, Nguyen, Wango, Alberts, &Altmann, 2006;
Sapolsky, Romero, &Munck, 2000; Setchell et al., 2008;
Weingrill, Gray, Barrett, &Henzi, 2004). Given this variation
among species, it is difficult to make generalizations among taxa.
Empirical studies that simultaneously track multiple social and
ecological variables in free-living wild mammals are therefore
warranted to tease apart the mechanisms and fitness correlates of
femaleefemale competition.
We studied the relationship between social rank and faecal
glucocorticoid metabolite (FGM) levels and the relationship be-
tween glucocorticoid levels and reproduction in plural breeding
yellow-bellied marmots, Marmota flaviventris. Yellow-bellied mar-
mots are an interesting system in which to study this because they
are facultatively social and hibernate throughout the winter, which
creates an important constraint on fitness (Armitage, 1998, 2010,
2014). Marmot colonies contain one to multiple matrilines of
adult females and their offspring. Cooperation and competition
within colonies is concentrated within family groups. Close kin
exchange the highest rates of affiliative and agonistic interactions;
these patterns emerge early in ontogeny at the pup stage and
persist into adulthood (Smith, Chung, &Blumstein, 2013). On
average, fewer than 50% of females wean a litter in a given year (e.g.
Blumstein &Armitage, 1998). Prior research suggests that younger
and less dominant adult females may be reproductively suppressed
(e.g. 2-year-old females reproduce only 34% of the time), and may
thus reproduce at later ages, especially if they remain in the same
groups as their reproductively dominant mothers (Armitage, 1998,
2010). Interestingly, delaying the age of first reproduction beyond
the age of 2 years, however, is rare, and body mass, rather than age,
per se, is a strong predictor of reproductive success among adults
(Lenihan &Vuren, 1996). Previous studies have also shown that
environmental variation (Schwartz, Armitage, &Vuren, 1998), age
(Ozgul, Oli, Olson, Blumstein, &Armitage, 2007), predation pres-
sure (Monclús, Tiulim, &Blumstein, 2011) and parity (e.g. if the
animal has previously produced offspring or not; see Oli &
Armitage, 2003) can influence reproductive success. Thus, we
therefore employed a holistic approach here to integrate these
variables into a single study based on almost a decade's worth of
field data to isolate the effects of rank and stress on female
reproduction.
Our study builds upon earlier research on this species to address
fundamental and highly debated questions. First, after controlling
for variation in environmental factors, age classes, predation pres-
sure, colony size and breeding status, we aimed to understand the
specific relationship between social rank and FGM level (our vali-
dated measure of baseline stress; Smith et al., 2012) in adult female
marmots. Second, we aimed to understand whether FGM levels
were associated with reproductive success by investigating each
D. T. Blumstein et al. / Animal Behaviour 112 (2016) 1e112
individual's probability of breeding as well their annual reproduc-
tive success, which we measured as the number of emergent
young.
Based on what is known about the social system of yellow-
bellied marmots, we predicted the following. Because individuals
are facultatively social and live in groups that can be described by
dominance hierarchies, we expected a relationship between social
rank and the degree to which individuals experienced stressors.
Thus, if femaleefemale competition is a stressor, then we expected
variation in FGM levels as a function of female social rank, and we
predicted that (socially) low-ranking reproductively mature fe-
males in this plural breeding species should have higher gluco-
corticoid levels than those of socially dominant females. Huang
et al. (2011) previously found no effect of an individual's overall
social dominance rank (based on outcomes of agonistic interactions
with all adult male and female group-mates) on reproductive
success of adult female marmots. Nevertheless, theory predicts that
female reproductive success should be most strongly shaped by
intrasexual competition among female mammals in social systems
in which females compete within matrilines (Stockley &Bro-
Jørgensen, 2011).
Here we therefore inquired whether femaleefemale competi-
tion (measured as adult female agonistic rank, after controlling for
group size; see Huang et al., 2011, and methods below) in combi-
nation with interindividual variation in responses to stressors may
together explain previously documented patterns of reproductive
suppression observed among adult female yellow-bellied marmots
(e.g. Armitage, 1998, 2014). Stressors triggered by femaleefemale
competition in combination with ecological stressors may together
explain reproductive failure and/or skew among adult female
marmots if reproductive success is reduced as a consequence of low
social rank and/or elevated FGM levels. We therefore predicted that
females with elevated FGM levels (i.e. young, low-ranking females),
low body mass and increased exposure to predators would have the
lowest reproductive success. Because we expected larger colonies
to attract more predators within an active season, we also included
colony size and its interaction with predator abundance as pre-
dictor variables in our models. While larger females are dominant
to smaller females, there was substantial variation in our data set to
permit us to isolate the effect of these variables statistically (none of
the correlations among mass, rank and predators exceeded 0.25).
METHODS
Field Methods
To determine the relationships among social rank, stress and
reproductive suppression, we studied the social interactions of
yellow-bellied marmots in and around the Rocky Mountain Bio-
logical Laboratory (RMBL, 38
57´N, 106
59´W) in Gunnison County,
Colorado, U.S.A. (Armitage, 2010; Blumstein, 2013). For these ana-
lyses, we focus on data collected on wild individuals over 9 years
(2003e2011) at five colony sites (Bench, Gothic Townsite, Picnic,
Marmot Meadow, and River) located in and around the RMBL.
Colony sites are geographical locations that contain one or more
marmot social groups; social groups are defined by space use
overlap and contain one or more breeding age females (who are
related), typically a single male, young of the year, and predispersal
yearlings of both sexes.
We live-trapped marmots in Tomahawk traps that we placed
around marmot burrow entrances for a few consecutive days every
other week. At each trapping event, we recorded an individual's
mass, sex, age, reproductive status (by scoring nipple development
as ‘present’,‘swollen’or ‘lactating’;Armitage &Wynne-Edwards,
2002; we conservatively estimated that 9% of females that failed
to wean a litter were scored as ‘lactating’) and collected faeces
(when present). The first time a subject was trapped, we collected a
hair sample for subsequent parentage assignment (Blumstein, Wey,
&Tang, 2009). Marmots were also marked permanently with ear-
tags the first time they were captured and their dorsal pelage was
dyed (and redyed when required) with a unique fur mark for visual
recognition from afar (see Armitage, 1982).
Social Factors Affecting Glucocorticoid Levels
Glucocorticoid metabolites can be reliably extracted from
marmot faeces (Smith et al., 2012). We opportunistically collected
faeces at trapping events and stored them on ice in plastic bags.
Samples were frozen within 2 h of collection at 20
C and steroid
hormones were extracted within 6 months of collection at the
University of California Los Angeles (details in Monclús et al., 2011;
Smith et al., 2012). Details of sample preparation and glucocorticoid
assay are in Appendix A of Blumstein, Patton, and Saltzman (2006).
We collected faecal samples each time an individual defecated in a
trap or in a bag. To be conservative, we only used the first faecal
sample collected from each weekly trapping session to assess basal
FGM levels (N¼278; mean ±SD ¼30.8 ±15.79 samples/year). We
did this because repeated capture can result in trap stress and
elevated FGM levels in marmots (Smith et al., 2012). For each in-
dividual (N¼60; mean ±SD ¼14.1 ±6.33 individuals/year), we
calculated the annual glucocorticoid measure by averaging the
combination of each of these single samples collected from each
biweekly trapping session.
To determine an individual's social rank, we observed marmots
both in the morning (0700e1000 hours) and in the afternoon
(1600e1900 hours), when marmots are most active (Armitage,
1962). We observed marmots at distances of 20e150 m (Huang
et al., 2011) to avoid influencing the marmots' behaviour. We
observed marmots in all five colonies for a total of 8762 h over the 9
years (mean ±SD ¼973.6 ±246.30 h/year). Because social groups
and colonies vary in their composition annually through births,
deaths and dispersal, we analysed the data for each colony sepa-
rately for each year. Colonies contained one or more social groups,
defined by space use overlap, and also varied annually. Following
Huang et al. (2011), we calculated rank based on the subset of in-
dividuals in a colony that were observed to interact; for females,
these almost always were restricted to other social group members.
During observations, we recorded both affiliative and agonistic
interactions, although for this study, we focused only on agonistic
interactions because we were interested in the link between
aggressive interactions and stress response. We defined agonistic
interactions as those that were aggressive and accompanied by
biting, chasing or fighting (see ethogram in Wey &Blumstein,
2012). For each social interaction, we recorded the instigator and
the victim, as well as the winner and the loser of the interaction.
These data allowed us to calculate social dominance hierarchies for
adult females using the Clutton-Brock index (CBI).
For a given subject, the CBI considers the total number of wins
and losses against each opponent (Clutton-Brock, Albon, Gibson, &
Guinness, 1979). The CBI is more applicable to the study of yellow-
bellied marmots than other metrics of dominance, like David's
score (DS), or the frequency-based dominance index (FDI), because
it is better suited for species with relatively few observed in-
teractions and does not use the rate of interaction in its calculation
(Bang, Deshpande, Sumana, &Gadagkar, 2010). The index is defined
as CBI ¼(Bþbþ1)/(Lþlþ1), where Bis the total number of
‘losers’(individuals that previously lost an interaction) that an in-
dividual has ‘beaten’,bis the total number of individuals that the
‘losers’(B)have‘beaten’,Lis the total number of ‘winners’(in-
dividuals that previously won an interaction) that have ‘beaten’the
D. T. Blumstein et al. / Animal Behaviour 112 (2016) 1e11 3
individual, and lis the total number of individuals that have
‘beaten’the ‘winners’(L).
Using the CBI, we then calculated relative rank, a measure that
accounts for differences in the number of individuals in a hierarchy
by standardizing each rank with respect to the total number of
individuals (Huang et al., 2011). We ordered the CBI values from
lowest to highest and then divided absolute rank by the total
number of individuals. In the relative ranking data, the lowest rank
was always zero and the highest rank was always one for each
hierarchy.
We determined social rank based solely the two-way in-
teractions between females that were of reproductive age (2 years
or older; Oli &Armitage, 2003). In our calculations we excluded
male interactions because our objective was to understand
reproductive suppression in females only. For the 9 years and the
45 colony-years (each of five colonies studied over 9 years) ana-
lysed, all had sufficient two-way interactions to allow us to
calculate ranks for one or more social groups based on our pa-
rameters of age and sex. However, there were some dominance
hierarchies that could not be used because there were no observed
interactions to base rank upon because breeding-age females in the
group were not observed to interact aggressively with each other.
For our analyses we assume that rank relationships were stable
within years. However, social rank very well could change from
year to year given differences in group composition and agonistic
interactions.
The Relationship between Glucocorticoid Levels and Reproductive
Success
Our second goal was to understand what factors influenced
breeding, and specifically whether stress (as measured by FGM
level) was a likely mechanism underlying reproductive success. We
examined this in two ways: first, by testing whether glucocorticoid
metabolite levels affected the probability of breeding and second,
by testing whether glucocorticoid metabolite levels and/or female
social rank best predicted the number of young weaned (a measure
of annual reproductive success). Because litters from different fe-
males may be mixed (Armitage, 1989; Olson, Blumstein, Pollinger,
&Wayne, 2012), we assigned offspring to their mother's molecu-
larly (details in Blumstein et al., 2009). Briefly, we extracted DNA
from hair and used 12 microsatellite markers to assign maternity
using Cervus 3.0 (Kalinowski, Taper, &Marshall, 2007) with 95%
confidence (see Blumstein, Lea, Olson, &Martin, 2010). Our pedi-
gree included 60 adult females (ca. 14 individuals were studied
each year) and their 306 confirmed offspring.
Statistical Analysis
We fitted generalized linear mixed-effects models (GLMMs) to
evaluate our questions. Following Smith et al. (2012), FGM levels
were ln transformed. To study the relationship between social rank
and stress, our dependent variable was the annual FGM level. For
this model, we entered each set of predictor variables associated
with each trapping event (time of day, day of the year; see below)
for which FGMs were collected to assess their effects on average
annual FGM values for each subject based on a Gaussian distribu-
tion. To study the relationship between social stress and the
probability of breeding, our dependent variable was whether a fe-
male successfully weaned a litter; these effects were modelled
using the binomial distribution. To study the relationship between
social stress and litter size, our dependent variable was the number
of offspring weaned; these effects were modelled using the Poisson
distribution.
Each model contained nine fixed effects, each of which was
entered as a separate predictor variable (rather than as an annual
measure so as to capture seasonal factors that might influence
variation in FGM values), as follows: (1) whether or not the indi-
vidual reproduced in the previous year; (2) the hour the gluco-
corticoid sample was collected (there is seasonal and daily variation
in FGM; Smith et al., 2012); (3) an individual's mass at the current
trapping event (i.e. when the particular faeces were collected for
FGM extraction); (4) the date the glucocorticoid sample was
collected; (5) an individual's body mass as estimated on 15 August
the previous year (because previous year's mass is correlated with
reproductive success: Lenihan &Vuren, 1996; for a description of
the linear mixed modelling approach used to estimate body mass,
see Ozgul et al., 2010); (6) the number of predators seen during
observations between April and July that year (the number of
predators encountered can influence FCM levels: Boonstra, Hik,
Singleton, &Tinnikov, 1998; Monclús et al., 2011); (7) colony size;
(8) female relative rank and/or (9) female rank and FGM levels,
depending on the model tested.
In addition, because female marmots can reproduce at 2 years of
age but may be reproductively suppressed (Blumstein &Armitage,
1999), we tested for the effects of age by creating a binary variable
that separated potentially reproductive females into two cate-
gories: 2 years old versus >2 years old. We tested for all two-way
interactions and only included these terms when it increased the
fit of our model using the Akaike Information Criterion (AIC).
Each model also contained three random effects: the colony that
an individual belonged to, female identity and year. These were
included to explain location, individual and seasonal differences
that may account for variation in FGM levels. We tested for two-
way interactions and retained interaction terms only when doing
so increased the fit of our model (within 2 AIC values of our best
model). Values for excluded terms were based on adding each term
to our final models. We set our alpha to 0.05 and report the results
of two-tailed tests.
We elected to not analyse our data with a formal path analysis
because to do so we would have had to remove about 15% of our
data and risk overfitting a structural equation model. Given the
relatively small effect sizes we report from the GLMM (see below),
we wished to retain as much power as possible and thus conducted
the GLMM.
Ethical Note
All procedures were approved under research protocol ARC
2001-191-01 as well as permits issued by the Colorado Division
of Wildlife. The research protocol was approved by the University
of California Los Angeles Animal Care Committee on 13 May 2002
and renewed annually. After trapping, individuals were released
immediately at the trap location. Marmots were in traps no
longer than 2e3 h, and typically for much less time than that.
Traps were shaded with vegetation on warm days. Marmot
handling was brief (typically 5e15 min depending upon the data
to be collected) and marmots were not injured during handling.
All marmots were handled while inside of a conical cloth-
handling bag to reduce stress. We swabbed ears with alcohol
before tagging individuals to reduce the chance of infection.
Observations were conducted at distances chosen to not overtly
affect marmot behaviour.
RESULTS
Our final data set contained 147 breeding events from 60 unique
adult female marmots, collected over 9 years and from our five
colonies (mean ±SD ¼15.8 ±10.34 breeding events/year). There
D. T. Blumstein et al. / Animal Behaviour 112 (2016) 1e114
were a total of 485 pups weaned over the 9 years from all females
(mean ±SD ¼53.9 ±40.37 pups weaned/year). Main predators
that visited the sites included black bears, Ursus americanus, coy-
otes, Canis latrans, mountain lions, Puma concolor, red foxes, Vulpes
vulpes, and various raptors. On average (±SD), we detected and
identified predators at these colony sites on 19 ±12.87 days/year,
but this varied by colony.
Factors Affecting Glucocorticoid Levels
After controlling for variation accounted for by our random
variables, six of our fixed effects and one interaction between them
significantly explained variation in average annual faecal gluco-
corticoid metabolite levels (Table 1). When considered together,
FGM levels significantly increased when females were heavier in
the previous year (Fig. 1a), when they were 2 years old (rather than
>2 years; Fig.1a), when they were of lower rank (Fig. 2), when there
were more predators in the area (Fig. 1b) and when they were
found in larger colonies (Table 1,Fig. 1b). This was the case after
controlling for the day of sample collection; FGM levels declined
across the season (Table 1). Colony size modulated the effect of
predator abundance (Table 1). The addition of other interactions
did not significantly explain variation in FGM levels. All random
effects were significant (year:
c
21
¼224.7, P<0.0001; colony:
c
21
¼13.7, P¼0.0002; marmot identity:
c
21
¼26.5, P<0.0001).
Table 1
Factors affecting faecal glucocorticoid metabolite levels in female yellow-bellied
marmots
Fixed effects Estimate±SD P
Reproduced in previous year 0.0617±0.4616 0.8935
Hour of faecal sample collection 0.0006±0.0034 0.8401
Mass at current trapping event 0.0001±0.0001 0.1305
Day of year of faecal sample collection 0.0018±0.0009 0.0480
Mass in previous year 0.0002±0.0001 <0.0001
2 years of age 0.2323±0.0409 <0.0001
Relative female social rank ¡0.1269±0.0524 0.0162
Predator abundance 0.0277±0.0079 0.0005
Colony size 0.0129±0.0033 <0.0001
Predator abundance)colony size ¡0.0005±0.0001 0.0001
Significant outcomes are shown in bold.
120
130
140
150
160
170
180
190
(a)
(b)
200
2 years,
light before
>2 years,
light before
2 years,
heavy before
>2 years,
heavy before
Faecal glucocorticoid metabolites (ng/g faeces)
N = 50
N = 80
N = 58
N = 85
120
130
140
150
160
170
180
190
200
Small colony,
low
p
redation
Large colony,
low
p
redation
Small colony,
hi
g
h
p
redation
Large colony,
hi
g
h
p
redation
N = 60
N = 62 N = 29
N = 122
Figure 1. Mean faecal glucocorticoid metabolite concentration in female yellow-
bellied marmots as a function of (a) age (2 years old vs >2 years old) and previous-
year body mass (light before: 2500e3500 g; heavy before: 3500e4500 g) and (b)
colony size (small: <50 marmots; large: >50 marmots) and predator abundance (low:
<22 predators; high: >22 predators). We entered body mass, colony size and predator
abundance as continuous variables directly into our statistical model (Table 1). Sample
sizes above bars indicate the number of faecal samples for females in each category.
Error bars represent raw standard errors around the mean for each category (for
illustrative purposes).
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
80
100
120
140
160
180
200
220
240
260
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Relative social rank of adult females (from low to hi
g
h)
Number of
p
u
p
s survivin
g
to weanin
g
Faecal glucocorticoid metabolites (ng/g faeces)
Figure 2. Effects of relative social rank on faecal glucocorticoid metabolite concen-
tration (open circles) and number of pups surviving to weaning (black circles) in fe-
male yellow-bellied marmots ranked from lowest (0) to highest (1). Error bars
represent raw standard errors around the mean for each category (for illustrative
purposes).
Table 2
Factors affecting the probability that adult female yellow-bellied marmots suc-
cessfully weaned a litter
Fixed effects Estimate±SE P
Reproduced in previous year 136.8±63.940 0.0324
Hour of faecal sample collection 0.020±0.073 0.7817
Mass at current trapping event 0.002±0.001 0.2650
Day of year of faecal sample collection 0.001±0.022 0.9709
Mass in previous year 0.005±0.001 <0.00001
2 years of age ¡3.058±0.943 0.0012
Relative female social rank 0.481±1.346 0.7207
Predator abundance 0.109±0.046 0.0177
Colony size 0.658±0.224 0.0033
Faecal glucocorticoid metabolite (FGM) level 6.474±3.155 0.0402
FGM levels)colony size ¡0.116±0.045 0.0105
FGM levels)reproduced in previous year ¡30.450±13.740 0.0267
Significant outcomes are shown in bold.
D. T. Blumstein et al. / Animal Behaviour 112 (2016) 1e11 5
The Relationships Between Glucocorticoid Levels and Reproductive
Success
After controlling for a variety of variables that might influence
the probability of breeding, six of our fixed effects and two in-
teractions between them explained significant variation in proba-
bility that adult female marmots reproduced (Table 2). The
variability of these data was consistent with that expected for the
binomial distribution (e.g. the dispersion parameter of 0.311 was
less than 1). Females were more likely to wean a litter if they were
heavier in the previous August (Fig. 3), if they lived in larger col-
onies (Figs. 4b and 5a), if they had higher FGM levels (Fig. 5a, b) and
if they reproduced in previous years (Table 2,Fig. 5b). Adult females
were also significantly less likely to wean a litter if they were 2
years of age (rather than >2 years; Fig. 3), and less likely to wean a
litter when more predators were present (Table 2,Fig. 4a). There
were significant negative interactions between FGM levels and
both colony size and whether or not a female weaned a litter in the
previous year (Table 2). However, reproductive failure could not be
explained by social rank, a variable that failed to explain whetheran
adult female weaned a litter (Table 2). All random effects were
significant (year:
c
21
¼13.3, P¼0.0003; colony:
c
21
¼24.2,
P<0.0001; marmot identity:
c
21
¼48.6, P<0.0001).
In contrast, the number of pups weaned was influenced by four
measured main effects, including social rank, and three interactions
(Table 3). The variability of these data was consistent with that
expected for the Poisson distribution (e.g. the dispersion parameter
of 0.784 was less than 1). Weaning litter sizes were larger when
females were of higher social rank (Fig. 2), they had higher FGM
levels (Table 3,Fig. 6a), they lived in larger colonies (Fig. 6a), they
reproduced in the previous year (Fig. 6b) and they were heavier in
the previous year (Fig. 7a). The effect of FGM levels on litter size,
however, was modulated by colony size and whether or not a fe-
male reproduced in the previous year (Table 3). For marmots that
successfully reproduced the previous year, litters were larger if
females were heavier when trapped in the previous August prior to
hibernation, and if they were of higher relative social rank within
their colonies (Table 3). Similarly, for 2-year-olds, litters were
smaller as predator abundance increased (Table 3,Fig. 7b). As with
the tendency to wean a litter or not, all random effects were sig-
nificant (year:
c
21
¼39.5, P<0.0001; colony:
c
21
¼61.0,
P<0.0001; marmot identity:
c
21
¼128.0, P<0.0001).
DISCUSSION
To better understand the influences of social stressors and their
associations with reproductive success, we investigated the prob-
ability of weaning a litter and the annual reproductive success of 60
female yellow-bellied marmots studied over 9 years. Importantly,
0.2
0.3
0.4
0.5
0.6
0.7
0.8
2 years,
li
g
ht before
>2 years,
li
g
ht before
2 years,
heav
y
before
>2 years,
heav
y
before
Proportion of adult females that reproduced
N = 50
N = 80
N = 58
N = 85
Figure 3. Effects of previous-year body mass (light before: 2500e3500 g; heavy
before: 3500e4500 g) and age (2 years old vs >2 years old) on the probability that
female yellow-bellied marmots would wean a litter in the current year. We entered
body mass as a continuous variable directly into our statistical model (Table 2). Sample
sizes above bars indicate the number of females sampled in each category. Error bars
represent raw standard errors around the mean for each category (for illustrative
purposes).
0.46
0.48
0.5
0.52
0.54
0.56
0.58
0.6
0.62 (a)
(b)
Low predation High predation
N = 122
N = 151
0.46
0.48
0.5
0.52
0.54
0.56
0.58
0.6
0.62
Small colony Large colony
Proportion of adult females that reproduced
N = 89
N = 184
Figure 4. Effects of (a) predator abundance (low: <22 predators; high: >22 predators)
and (b) colony size (small: <50 marmots; large: >50 marmots) on the probability that
adult female yellow-bellied marmots would wean a litter in the current year. We
entered colony size and predator abundance as continuous variables directly into our
statistical model (Table 2). Sample sizes above bars indicate the number of females
sampled in each category. Error bars represent raw standard errors around the mean
for each category (for illustrative purposes).
D. T. Blumstein et al. / Animal Behaviour 112 (2016) 1e116
we found that as a female's social rank increased, her FGM levels
significantly decreased, but this was a relatively small effect that
was only detected after controlling for other variables that
explained variation in FGM levels. This is notable because similar
patterns have also been documented in many cooperative breeders
(Cavigelli &Chaudhry, 2012). From this finding, we infer that low-
ranking females experience more stressors than high-ranking fe-
males. Indeed, social competition among female kin is intense
within matrilines; social competition emerges early in ontogeny
and persists into adulthood (Smith et al., 2013).
Social rank also influenced the number of offspring successfully
weaned in a season (but not reproductive failure), with high-
ranking producing the most offspring. Our finding that females
with high social status, based on agonistic interactions, produced
more offspring than did low-ranking females has similarly been
documented in ungulates (e.g. mountain goats, Oreamnos
americanus:C^
ot
e&Festa-Bianchet, 2001), carnivores (e.g. spotted
hyaenas: Holekamp, Smale, &Szykman, 1996) and primates (e.g.
gelada baboons, Theropithecus gelada:Dunbar &Dunbar, 1977).
Low-ranking females may produce fewer offspring because of dif-
ferences in access to resources, amount of social support, body
condition, group size and/or increased agonistic interactions
(Abbott et al., 2003; Appleby, 1980; Cabezas, Blas, Marchant, &
Moreno, 2007; Foley et al., 2001; Hackl€
ander et al., 2003).
Taken together, our findings about ‘stress’levels and reproduc-
tive output are therefore in accordance with classic captive studies
on aggresion, rank and physiological stress responses that support
the ‘stress of subordination’hypothesis (reviewed by Creel et al.,
2013). However, high FGM levels were also correlated with
increased reproductive success: higher FGM levels were correlated
with increased likelihood of successful reproduction as well as litter
size. This finding is inconsistent with many findings of no rela-
tionship between stress levels and reproductive success (reviewed
in: Boonstra, 2013; Creel, 2001; Sapolsky et al., 2000). Our findings
are important because they support the emerging view, such as
that expressed recently by Creel et al. (2013), that the effects of
social stressors on the HPA axis are complex. Our empirical findings
therefore contribute to important data from wild populations of
mammals necessarily to advance this area of science forward.
Social rank was correlated with our measure of physiological
stress, but it failed to predict the probability of breeding. The re-
lationships we identified between FGM and social rank and FGM
and reproductive success may at first appear contradictory (high-
ranking females had lower FGM levels and the probability of
breeding was positively correlated with FGM, suggesting that high-
ranking females have a lower probability of breeding). However,
our finding that females living in large colonies had elevated annual
FGM levels and produced the most offspring is likely attributed to a
surge in glucocorticoids during reproduction. In adult female
marmots, FGM levels vary with reproductive state and are highest
during pregnancy and lactation (Smith et al., 2012). Ebensperger
et al. (2011) found similar results in degus, Octodon degus.In
particular, adult female cortisol levels appear to rise as offspring are
weaned, which seems to be associated with an increased pro-
pensity to produce alarm calls (Blumstein, Steinmetz, Armitage, &
Daniel, 1997), but not with other factors such as group size or
number of females. Perhaps the measure of probability of breeding
(whether or not the female weaned pups) should include pups lost
before the pups' emergence. Unfortunately, such assignments are
nearly impossible for obligate hibernators, such as marmots, under
0.35
0.4
0.45
0.5
0.55
0.6
0.65
0.7
0.75 (b)
Low FGMs,
no re
p
rod.
High FGMs,
no re
p
rod.
Low FGMs,
re
p
roduced
High FGMs,
re
p
roduced
Proportion of adult females that reproduced
N = 83
N = 120
N = 30
N = 40
0.35
0.4
0.45
0.5
0.55
0.6
0.65
0.7
(a)
0.75
Low FGMs,
large colony
Low FGMs,
small colony
High FGMs,
large colony
High FGMs,
small colony
N = 68
N = 45 N = 116
N = 44
Figure 5. Effects of faecal glucocorticoid metabolite concentration (low FGMs:
<128 ng/g faeces; high FGMs: 128 ng/g faeces), (a) colony size (large: >50 marmots;
small: <50 marmots) and (b) reproductive success in the previous year on the prob-
ability that female yellow-bellied marmots would wean a litter in the current year. We
entered FGM and colony size as continuous variables directly into our statistical model
(Table 2). Sample sizes above bars indicate the number of females sampled in each
category. Error bars represent raw standard errors around the mean for each category
(for illustrative purposes).
Table 3
Factors affecting the number of yellow-bellied marmot pups surviving to weaning
Fixed effects Estimate±SE P
Reproduced in previous year 34.97±13.32 0.0087
Hour of faecal sample collection 0.0102±0.0132 0.4372
Mass at current trapping event 0.0002±0.0002 0.1366
Day of year of faecal sample collection 0.0031±0.0035 0.3872
Mass in previous year 0.0008±0.0002 0.0002
Relative female social rank 0.5259±0.2539 0.0384
2 years of age 0.2652±0.3363 0.4303
Predator abundance 0.0078±0.0081 0.3341
Colony size 0.1964±0.0444 <0.00001
Faecal glucocorticoid metabolite
(FGM) level
2.218±0.6253 0.0004
FGM levels)colony size ¡0.0355±0.0091 <0.00001
FGM levels)reproduced in
previous year
¡9.0480±2.875 0.0016
Mass at current trapping event
)reproduced in previous year
0.0021±0.0009 0.0146
2 years of age)predator abundance ¡0.0788±0.0181 <0.00001
Significant outcomes are shown in bold.
D. T. Blumstein et al. / Animal Behaviour 112 (2016) 1e11 7
natural conditions. This would require live-trapping of marmots
with bait that would prematurely prime metabolic processes prior
to snow-melt and the start of the growing season in their high-
elevation habits. Such an effort would not only confound the
reproductive outputs, but would probably be dangerous for the
marmots. Given some uncertainty about the cause of preweaning
failure (litter reabsorption versus preweaning predation), we
elected to define breeders as those who successfully weaned a
litter, but we recognize that at least 9% of females that failed to
wean a litter were scored as lactating. We therefore call for future
research in a more controlled setting than is possible under field
conditions, such as in a laboratory, to unravel these complex pat-
terns. Additional studies may therefore be able to disentangle the
relative effects of preweaning predation and reabsorption to
explain these seemingly contradictory results.
These inconsistent patterns may also suggest that female
reproductive outcomes in yellow-bellied marmots are more
strongly mediated by ecological factors than by stressors associated
with social status. Indeed, an individual's mass from the previous
year was correlated with increased FGM levels and both metrics of
reproductive success. Our finding here that FGM levels increased
with the number of predator visits is consistent with a growing
body of evidence revealing the potential sublethal effects of pred-
ators on prey (e.g. Brooks, Gaskell, &Maltby, 2009; Creel &
Christianson, 2008; Pangle, Peacor, &Johannsson, 2007). Our
finding that the probability of weaning a litter decreased as the
observed number of predators increased is consistent with studies
showing that high predator-induced ‘stress’directly impairs
reproductive function in snowshoe hares, Lepus americanus
(Boonstra et al., 1998; Sheriff, Krebs, &Boonstra, 2009). Given that
marmots consistently reside at the same burrows, marmots may
similarly suffer from sublethal effects from predator-induced
stressors. In contrast, Creel, Winnie, and Christianson (2009)
found that wolf presence limits reproduction in elk, Cervus
0
0.5
1
1.5
2
2.5
3
3.5 (b)
2 years,
low
p
redation
>2 years,
low
p
redation
2 years,
hi
g
h
p
redation
>2 years,
hi
g
h
p
redation
N = 60 N = 62
N = 48
N = 103
0
0.5
1
1.5
2
2.5
3
3.5 (a)
Light now,
no reprod.
Heavy now,
no reprod.
Light now,
reproduced
Heavy now,
reproduced
Number of pups surviving to weaning
N = 126
N = 77
N = 43
N = 27
Figure 7. Effects of (a) current-year body mass (light now: 1000e3000 g; heavy now:
3001e5000 g) and previous-year reproductive success and (b) predator abundance
(low: <22 predators; high: >22 predators) and female age (2 years old vs >2 years old)
on the number of offspring that female yellow-bellied marmots weaned in the current
year. We entered body mass as a continuous variable directly into our statistical model
(Table 3). Sample sizes above bars indicate the number of females sampled in each
category. Error bars represent raw standard errors around the mean for each category
(for illustrative purposes).
1
1.2
1.4
1.6
1.8
2
2.2
2.4
2.6
2.8
3
Low FGMs,
small colony
High FGMs,
small colony
Low FGMs,
large colony
High FGMs,
large colony
Number of pups surviving to weaning
N = 45
(a)
(b)
N = 44
N = 68
N = 116
1
1.2
1.4
1.6
1.8
2
2.2
2.4
2.6
2.8
3
Low FGMs,
no re
p
rod.
High FGMs,
no re
p
rod.
Low FGMs,
re
p
roduced
High FGMs,
re
p
roduced
N = 83
N = 120 N = 30
N = 40
Figure 6. Effects of faecal glucocorticoid metabolite concentration (low FGMs:
<128 ng/g faeces; high FGMs: 128 ng/g faeces), (a) colony size (large: >50 marmots;
small: <50 marmots) and (b) previous-year reproductive success on the number of
offspring that female yellow-bellied marmots weaned in the current year. We entered
FGM and colony size as continuous variables directly into our statistical model
(Table 3). Sample sizes above bars indicate the number of females sampled in each
category. Error bars represent raw standard errors around the mean for each category
(for illustrative purposes).
D. T. Blumstein et al. / Animal Behaviour 112 (2016) 1e118
elaphus; wolves drive elk to forage on suboptimal diets. Perhaps
ecological factors associated with foraging decisions also influence
body mass, which, in turn, influences FGM levels and reproductive
success in marmots. Moreover, because predator abundance may
also contribute directly to pup mortality via predation, carefully
designed field experiments are needed to investigate the mecha-
nisms responsible for this observed pattern in marmots and other
rodents.
Rodent reproductive success often increases with age in wild
mammal populations (King &Allain
e, 2002; King, Festa-Bianchet,
&Hatfield, 1991). Accordingly, 2-year-old marmots (the age of
reproductive maturity; Oli &Armitage, 2003) in the current study
had higher FGM levels and were less likely to reproduce than older
females. Thus, 2-year-old females, which are often socially and
reproductively subordinate to their mothers, might delay their age
of first reproduction. Armitage (1998) suggested that older adult
female marmots may reproductively suppress 2-year-olds. How-
ever, because we found no interactions between being a 2-year-old
and FGM levels, it is unlikely that FGM mediates reproductive
failure. Instead, we found an interesting, nonadditive effect of
predator abundance on the number of offspring weaned by 2-year-
olds and by older females. That is, 2-year-olds appeared to be
particularly sensitive to the risks of high predator abundance,
rearing disproportionally small litters under the risk of high pre-
dation. This raises the possibility that 2-year-olds are most
vulnerable to the nonlethal effects of predators and/or the least
prepared to successfully protect their offspring from predators.
Because adult females belonging to both age categories apparently
similarly perceived large numbers of predators as intense stressors
(additive effect of predators on FGM levels of females that were 2
years old and >2 years old), our findings are instead most consis-
tent with the notion that the loss of offspring born to inexperienced
mothers is due to direct predation rather than to nonlethal effects
of mothers being ‘stressed’by high predator abundance.
Moreover, the significant interaction between body mass and
whether an individual reproduced the previous year highlights the
importance of condition-related factors affecting reproductive
success. Thus, the reproductive skew that Armitage (1998) pro-
posed may be due to factors other than social rank, such as female
body condition and predator abundance found to be important in
our current study. By contrast, for males, Huang et al. (2011)
identified a correlation between rank and reproductive success.
Further study of maleemale competition is clearly warranted.
As reviewed in the Introduction, differences in stress level,
dominance rank and reproductive success are seen among taxa,
and even within species. These connections are clearly complex,
and research has found both positive and negative relationships
between social stress and reproductive success (Bonier et al., 2009;
Brann &Mahesh, 1991; Crespi, Williams, Jessop, &Delehanty,
2013). There are multiple potential causes for these variable re-
lationships. For example, Brann and Mahesh (1991) noted that
reproductive outcome is in part a function of whether the stress is
acute or chronic because acute stressors may facilitate corticoste-
roid production and enhance fertility. By contrast, chronic stressors
may have a more inhibitory effect on reproduction, by preventing
sexual maturation and impairing pregnancies.
Although it may be premature to make generalizations about
the effects of stressors and social rank on reproductive success, our
study highlights the value of longitudinal empirical studies relating
three features of social living in relevant ecological contexts: social
rank, baseline stress and reproductive success. Whether socially
induced or induced through predator threat, females that enter
hibernation at low body mass may reduce their investment in
young to conserve valuable resources. Perhaps in species that only
have a single chance per year to reproduce, like marmots, socially
mediated reproductive suppression should be rare, and thus, social
rank or other social stressors may only reduce offspring number
rather than restrict annual reproductive bouts altogether. Regard-
less, these results highlight the importance of body condition and
other environmental factors, such as predator abundance, on the
reproductive success of highly seasonal breeders.
Acknowledgments
We thank all the marmoteers since 2001 who helped collect
these data. D.T.B was supported by the National Geographic Society
(grant number 8140-06), the University of California Los Angeles
(Faculty Senate and the Division of Life Sciences), a Rocky Mountain
Biological Laboratory research fellowship, and by the National
Science Foundation (IDBR-0754247 and DEB-1119660 to D.T.B., as
well as DBI 0242960 and 0731346 to the Rocky Mountain Biological
Laboratory). J.E.S. was supported by funds from Faculty Develop-
ment Funds and from the Provost's Office at Mills College as well as
by fellowships from the American Association of University
Women, the Institute for Society and Genetics at the University of
California Los Angeles and the American Philosophical Society. We
also thank several anonymous referees whose astute comments
helped us improve previous versions of this manuscript.
References
Abbott, D. H. (1984). Behavioral and physiological suppression of fertility in sub-
ordinate marmoset monkeys. American Journal of Primatology, 6,169e186.
Abbott, D. H. (1987). Behaviourally mediated suppression of reproduction in female
primates. Journal of Zoology, 213, 455e47 0.
Abbott, D. H., Keverne, E. B., Bercovitch, F. B., Shively, C. A., Mendoza, S. P.,
Saltzman, W., et al. (2003). Are subordinates always stressed? A comparative
analysis of rank differences in cortisol levels among primates. Hormones and
Behavior, 43,67e82.
Appleby, M. C. (1980). Social rank and food access in red deer stags. Behaviour, 74,
294e309.
Armitage, K. B. (1962). Social behaviour of a colony of the yellow-bellied marmot
(Marmota flaviventris). Animal Behaviour, 10,319e331.
Armitage, K. B. (1982). Yellow-bellied marmot. In D. E. Davis (Ed.), CRC handbook of
census methods for terrestrial vertebrates (pp. 148e149). Boca Raton, FL: CRC
Press.
Armitage, K. B. (1989). The function of kin discrimination. Ethology Ecology &
Evolution, 1,111e121.
Armitage, K. B. (1998). Reproductive strategies of yellow-bellied marmots: energy
conservation and differences between the sexes. Journal of Mammalogy, 79,
385e393.
Armitage, K. B. (2010). Individual fitness, social behavior, and population dynamics
of yellow-bellied marmots. In I. Billick, & M. V. Price (Eds.), The ecology of place:
Contributions of place-based research to ecological understanding (pp. 134e154).
Chicago, IL: University of Chicago Press.
Armitage, K. B. (2014). Marmot biology: Sociality, individual fitness, and population
dynamics. Cambridge, U.K.: Cambridge Univesity Press.
Armitage, K. B., & Wynne-Edwards, K. E. (2002). Progesterone concentrations of
wild-caught yellow-bellied marmots. In K. B. Armitage, & V. Y. Rumiantsev
(Eds.), Holarctic marmots as a factor of biodiversity: Proceedings of the Third In-
ternational Conference on Marmots, Cheboksary, Russia, 25e30 August 1997 (pp.
41e47). Moscow: ABF.
Arnold, W., & Dittami, J. (1997). Reproductive suppression in male alpine marmots.
Animal Behaviour, 53,53e66.
Bales, K. L., French, J. A., Hostetler, C. M., & Dietz, J. M. (2005). Social and repro-
ductive factors affecting cortisol levels in wild female golden lion tamarins
(Leontopithecus rosalia). American Journal of Primatology, 67,25e35.
Bang, A., Deshpande, S., Sumana, A., & Gadagkar, R. (2010). Choosing an appropriate
index to construct dominance hierarchies in animal societies: a comparison of
three indices. Animal Behaviour, 79,631e636.
Barrette, M.-F., Monfort, S. L., Festa-Bianchet, M., Clutton-Brock, T. H., & Russell, A. F.
(2012). Reproductive rate, not dominance status, affects fecal glucocorticoid
levels in breeding female meerkats. Hormones and Behavior, 61, 463e471.
Beehner, J. C., Nguyen, N., Wango, E. O., Alberts, S. C., & Altmann, J. (20 06). The
endocrinology of pregnancy and fetal loss in wild baboons. Hormones and
Behavior, 49,688e699.
Bell, M. B., Nichols, H. J., Gilchrist, J., Cant, M. A., & Hodge, S. J. (2012). The cost of
dominance: suppressing subordinate reproduction affects the reproductive
success of dominant female banded mongooses. Proceedings of the Royal Society
B: Biological Sciences, 279,619e624.
D. T. Blumstein et al. / Animal Behaviour 112 (2016) 1e11 9
Blumstein, D. T. (2013). Yellow-bellied marmots: insights from an emergent view of
sociality. Philosophical Transactions of the Royal Society B: Biological Sciences,
368, 20120349.
Blumstein, D. T., & Armitage, K. B. (1998). Life history consequences of social
complexity: a comparative study of ground-dwelling sciurids. Behavioral Ecol-
ogy, 9,8e19.
Blumstein, D. T., & Armitage, K. B. (1999). Cooperative breeding in marmots. Oikos,
84, 369e382.
Blumstein, D., Lea, A., Olson, L., & Martin, J. (2010). Heritability of anti-predatory
traits: vigilance and locomotor performance in marmots. Journal of Evolu-
tionary Biology, 23, 879e887.
Blumstein, D. T., Patton, M. L., & Saltzman, W. (2006). Faecal glucocorticoid me-
tabolites and alarm calling in free-living yellow-bellied marmots. Biology Let-
ters, 2,29e32.
Blumstein, D. T., Steinmetz, J., Armitage, K. B., & Daniel, J. C. (1997). Alarm calling in
yellow-bellied marmots: II. The importance of direct fitness. Animal Behaviour,
53,173e184.
Blumstein, D. T., Wey, T. W., & Tang, K. (2009). A test of the social cohesion hy-
pothesis: interactive female marmots remain at home. Proceedings of the Royal
Society B: Biological Sciences, 276,3007e3012.
Bonier, F., Martin, P. R., Moore, I. T., & Wingfield, J. C. (2009). Do baseline gluco-
corticoids predict fitness? Trends in Ecology &Evolution, 24, 634e642.
Boonstra, R. (2013). Reality as the leading cause of stress: rethinking the impact of
chronic stress in nature. Functional Ecology, 27,11e23.
Boonstra, R., Hik, D., Singleton, G. R., & Tinnikov, A. (1998). The impact of predator-
induced stress on the snowshoe hare cycle. Ecological Monographs, 68, 371e394.
Brann, D. W., & Mahesh, V. B. (1991). Role of corticosteroids in female reproduction.
FASEB Journal, 5, 2691e2698.
Brooks, A. C., Gaskell, P. N., & Maltby, L. L. (2009). Sublethal effects and preda-
toreprey interactions: implications for ecological risk assessment. Environ-
mental Toxicology and Chemistry, 28, 2449e2457.
Cabezas, S., Blas, J., Marchant, T. A., & Moreno, S. (2007). Physiological stress levels
predict survival probabilities in wild rabbits. Hormones and Behavior, 51,
313e320.
Carlson, A. A., Young, A. J., Russell, A. F., Bennett, N. C., McNeilly, A. S., & Clutton-
Brock, T. (2004). Hormonal correlates of dominance in meerkats (Suricata sur-
icatta). Hormones and Behavior, 46,141e150.
Cavigelli, S. A. (1999). Behavioural patterns associated with faecal cortisol levels in
free-ranging female ring-tailed lemurs, Lemur catta.Animal Behaviour, 57,
935e944.
Cavigelli, S. A., & Chaudhry, H. S. (2012). Social status, glucocorticoids, immune
function, and health: can animal studies help us understand human
socioeconomic-status-related health disparities? Hormones and Behavior, 62,
295e313.
Cavigelli, S. A., Dubovick, T., Levash, W., Jolly, A., & Pitts, A. (2003). Female domi-
nance status and fecal corticoids in a cooperative breeder with low reproduc-
tive skew: ring-tailed lemurs (Lemur catta). Hormones and Behavior, 43,
166 e179 .
Chelini, M. O. M., Palme, R., & Otta, E. (2011). Social stress and reproductive success
in the female Syrian hamster: endocrine and behavioral correlates. Physiology &
Behavior, 104, 948e954.
Clutton-Brock, T. H., Albon, S., Gibson, R., & Guinness, F. E. (1979). The logical stag:
adaptive aspects of fighting in red deer (Cervus elaphus L.). Animal Behaviour, 27,
211e225.
C^
ot
e, S. D., & Festa-Bianchet, M. (2001). Reproductive success in female mountain
goats: the influence of age and social rank. Animal Behaviour, 62,173e181.
Creel, S. (2001). Social dominance and stress hormones. Trends in Ecology &Evo-
lution, 16,491e497.
Creel, S., & Christianson, D. (2008). Relationships between direct predation and risk
effects. Trends in Ecology &Evolution, 23,194e201.
Creel, S., Dantzer, B., Goymann, W., & Rubenstein, D. R. (2013). The ecology of stress:
effects of the social environment. Functional Ecology, 27,66e80.
Creel, S., Winnie, J. A., & Christianson, D. (2009). Glucocorticoid stress hormones
and the effect of predation risk on elk reproduction. Proceedings of the National
Academy of Sciences of the United States of America, 106, 12388e12393.
Crespi, E. J., Williams, T. D., Jessop, T. S., & Delehanty, B. (2013). Life history and the
ecology of stress: how do glucocorticoid hormones influence life-history vari-
ation in animals? Functional Ecology, 27,93e10 6.
Crockford, C., Wittig, R. M., Whitten, P. L., Seyfarth, R. M., & Cheney, D. L. (2008).
Social stressors and coping mechanisms in wild female baboons (Papio hama-
dryas ursinus). Hormones and Behavior, 53, 254e265.
Dunbar, R. I. M., & Dunbar, E. P. (1977). Dominance and reproductive success among
female gelada baboons. Nature, 266,351e352.
Ebensperger, L. A., Ramírez-Estrada, J., Le
on, C., Castro, R. A., Tolhuysen, L. O.,
Sobrero, R., et al. (2011). Sociality, glucocorticoids and direct fitness in the
communally rearing rodent, Octodon degus.Hormones and Behavior, 60,
346e352.
Ellis, L. (1995). Dominance and reproductive success among nonhuman animals: a
cross-species comparison. Ethology and Sociobiology, 16,257e333.
Foley, C., Papageorge, S., & Wasser, S. (2001). Noninvasive stress and reproductive
measures of social and ecological pressures in free-ranging African elephants.
Conservation Biology, 15,1134e1142.
Gesquiere, L. R., Learn, N. H., Simao, M. C. M., Onyango, P. O., Alberts, S. C., &
Altmann, J. (2011). Life at the top: rank and stress in wild male baboons. Science,
333,357e360.
Goymann, W., East, M. L., Wachter, B., H€
oner, O. P., M€
ostl, E., Van't Holf, T. J., et al.
(2001). Social, state-dependent and environmental modulation of faecal corti-
costeroid levels in free-ranging female spotted hyenas. Proceedings of the Royal
Society B: Biological Sciences, 268, 2453e2459.
Hackl€
ander, K., M€
ostl, E., & Arnold, W. (2003). Reproductive suppression in female
alpine marmots, Marmota marmota.Animal Behaviour, 65,1133e114 0.
Holekamp, K. E., Smale, L., & Szykman, M. (1996). Rank and reproduction in the
female spotted hyaena. Journal of Reproduction and Fertility, 108, 229e237.
Huang, B., Wey, T. W., & Blumstein, D. T. (2011). Correlates and consequences of
dominance in a social rodent. Ethology, 117,573e585.
Johnson, E. O., Kamilaris, T. C., Chrousos, G. P., & Gold, P. W. (1992). Mechanisms of
stress: a dynamic overview of hormonal and behavioral homeostasis. Neuro-
science &Biobehavioral Reviews, 16,115e130.
Kalinowski, S. T., Taper, M. L., & Marshall, T. C. (2007). Revising how the computer
program CERVUS accommodates genotyping error increases success in pater-
nity assignment. Molecular Ecology, 16, 1099e110 6.
Kaplan, J. R., Adams, M. R., Koritnik, D. R., Rose, J. C., & Manuck, S. B. (1986). Adrenal
responsiveness and social status in intact and ovariectomized Macaca fas-
cicularis.American Journal of Primatology, 11,181e193.
van Kesteren, F., Paris, M., Macdonald, D. W., Millar, R., Argaw, K., Johnson, P. J., et al.
(2013). The physiology of cooperative breeding in a rare social canid; sex,
suppression and pseudopregnancy in female Ethiopian wolves. Physiology &
Behavior, 122,39e45.
King, W. J., & Allain
e, D. (2002). Social, maternal, and environmental influences on
reproductive success in female alpine marmots (Marmota marmota). Canadian
Journal of Zoology, 80,2137e2143.
King, W. J., Festa-Bianchet, M., & Hatfield, S. E. (1991). Determinants of reproductive
success in female Columbian ground squirrels. Oecologia, 86, 528e534.
Lenihan, C., & Vuren, D. V. (1996). Growth and survival of juvenile yellow-bellied
marmots (Marmota flaviventris). Canadian Journal of Zoology, 74,297e302.
Louch, C. D., & Higginbotham, M. (1967). The relation between social rank and
plasma corticosterone levels in mice. General and Comparative Endocrinology, 8,
441e444.
Martínez-Mota, R., Valdespino, C., Rebolledo, J. A. R., & Palme, R. (2008). Determi-
nation of fecal glucocorticoid metabolites to evaluate stress response in
Alouatta pigra.International Journal of Primatology, 29, 1365e1373.
McLeod, P. J., Moger, W. H., Ryon, J., Gadbois, S., & Fentress, J. (1996). The relation
between urinary cortisol levels and social behaviour in captive timber wolves.
Canadian Journal of Zoology, 74, 209e216.
Mendl, M., Zanella, A. J., & Broom, D. M. (1992). Physiological and reproductive
correlates of behavioural strategies in female domestic pigs. Animal Behaviour,
44,1107e1121.
Metrione, L. C., & Harder, J. D. (2011). Fecal corticosterone concentrations and
reproductive success in captive female southern white rhinoceros. General and
Comparative Endocrinology, 171, 283e292.
Moberg, G. P. (1985). Influence of stress on reproduction: measure of well-being. In
G. P. Moberg (Ed.), Animal stress (pp. 245e267). Bethesda, MD: American
Physiological Society.
Monclús, R., Tiulim, J., & Blumstein, D. T. (2011). Older mothers follow conservative
strategies under predator pressure: the adaptive role of maternal glucocorti-
coids in yellow-bellied marmots. Hormones and Behavior, 60, 660e665.
Muller, M. N., & Wrangham, R. W. (2004). Dominance, cortisol and stress in wild
chimpanzees (Pan troglodytes schweinfurthii). Behavioral Ecology and Sociobio-
logy, 55, 332e340.
Munck, A., Guyre, P. M., & Holbrook, N. J. (1984). Physiological functions of gluco-
corticoids in stress and their relation to pharmacological actions. Endocrine
Reviews, 5,25e44.
Oli, M. K., & Armitage, K. B. (2003). Sociality and individual fitness in yellow-bellied
marmots: insights from a long-term study (1962e2001). Oecologia, 136,
543e550.
Olson, L., Blumstein, D., Pollinger, J., & Wayne, R. (2012). No evidence of inbreeding
avoidance despite demonstrated survival costs in a polygynous rodent. Molec-
ular Ecology, 21, 562e571.
Ozgul, A., Oli, M. K., Olson, L. E., Blumstein, D. T., & Armitage, K. B. (2007). Spatio-
temporal variation in reproductive parameters of yellow-bellied marmots.
Oecologia, 154,95e106.
Packer, C., Pusey, A. E., & Eberly, L. E. (2001). Egalitarianism in female African lions.
Science, 293, 690e693.
Pangle, K. L., Peacor, S. D., & Johannsson, O. E. (2007). Large nonlethal effects of an
invasive invertebrate predator on zooplankton population growth rate. Ecology,
88(2), 402e412.
Ramamoorthy, S., & Cidlowski, J. A. (2013). Exploring the molecular mechanisms of
glucocorticoid receptor action from sensitivity to resistance. Endocrine Devel-
opment, 24,41e56.
Reeder, D. M., & Kramer, K. M. (20 05). Stress in free-ranging mammals: inte-
grating physiology, ecology, and natural history. Journal of Ma mmalogy, 86,
225e235.
Rensel, M. A., & Schoech, S. J. (2011). Repeatability of baseline and stress-induced
corticosterone levels across early life stages in the Florida scrub-jay (Aphe-
locoma coerulescens). Hormones and Behavior, 59,497e502.
Rood, J. P. (1980). Mating relationships and breeding suppression in the dwarf
mongoose. Animal Behaviour, 28,143e150.
Saltzman, W., Schultz-Darken, N. J., Scheffler, G., Wegner, F. H., & Abbott, D. H.
(1994). Social and reproductive influences on plasma cortisol in female
marmoset monkeys. Physiology &Behavior, 56,801e810.
D. T. Blumstein et al. / Animal Behaviour 112 (2016) 1e1110
Sands, J., & Creel, S. (2004). Social dominance, aggression and faecal glucocorticoid
levels in a wild population of wolves, Canis lupus.Animal Behaviour, 67,
387e396.
Sapolsky, R. M. (1992). Neuroendocrinology of the stress response. In J. B. Becker,
S. M. Breedlove, & D. Crews (Eds.), Behavioral endocrinology (pp. 287e324).
Cambridge, MA: MIT Press.
Sapolsky, R. M., Romero, L. M., & Munck, A. U. (2000). How do glucocorticoids in-
fluence stress responses? Integrating permissive, suppressive, stimulatory, and
preparative actions. Endocrine Reviews, 21,55e89.
Scarlata, C. D., Elias, B. A., Godwin, J. R., Powell, R. A., Shepherdson, D., Shipley, L. A.,
et al. (2012). Relationship between fecal hormone concentrations and repro-
ductive success in captive pygmy rabbits (Brachylagus idahoensis). Journal of
Mammalogy, 93, 759e770.
Schwartz, O. A., Armitage, K. B., & Vuren, D. (1998). A 32-year demography of
yellow-bellied marmots (Marmota flaviventris). Journal of Zoology, 246,
337e346.
Setchell, J. M., Smith, T., Wickings, E. J., & Knapp, L. A. (2008). Factors affecting fecal
glucocorticoid levels in semi-free-ranging female mandrills (Mandrillus sphinx).
American Journal of Primatology, 70, 1023e1032.
Sheriff, M. J., Krebs, C. J., & Boonstra, R. (2009). The sensitive hare: sublethal effects
of predator stress on reproduction in snowshoe hares. Journal of Animal Ecology,
78, 1249e1258.
Smith, J. E., Chung, L. K., & Blumstein, D. T. (2013). Ontogeny and symmetry of social
partner choice among free-living yellow-bellied marmots. Animal Behaviour, 85,
715e725.
Smith, J. E., Monclús, R., Wantuck, D., Florant, G. L., & Blumstein, D. T. (2012). Fecal
glucocorticoid metabolites in wild yellow-bellied marmots: experimental
validation, individual differences and ecological correlates. General and
Comparative Endocrinology, 178,417e426.
Starling, A. P., Charpentier, M. J., Fitzpatrick, C., Scordato, E. S., & Drea, C. M. (2010).
Seasonality, sociality, and reproduction: long-term stressors of ring-tailed le-
murs (Lemur catta). Hormones and Behavior, 57,76e85.
Stockley, P., & Bro-Jørgensen, J. (2011). Female competition and its evolutionary
consequences in mammals. Biological Reviews, 86,341e366.
Tilbrook, A., Turner, A., & Clarke, I. (2000). Effects of stress on reproduction in non-
rodent mammals: the role of glucocorticoids and sex differences. Reviews of
Reproduction, 5,105e113.
Wasser, S. K., & Barash, D. P. (1983). Reproductive suppression among female
mammals: implications for biomedicine and sexual selection theory. Quarterly
Review of Biology, 58,513e538.
Weingrill, T., Gray, D. A., Barrett, L., & Henzi, S. P. (2004). Fecal cortisol levels in free-
ranging female chacma baboons: relationship to dominance, reproductive state
and environmental factors. Hormones and Behavior, 45, 259e269.
Welsh, T. H., Jr., & Johnson, B. H. (1981). Stress-induced alterations in secretion of
corticosteroids, progesterone, luteinizing hormone, and testosterone in bulls.
Endocrinology, 109,185e190 .
Wey, T. W., & Blumstein, D. T. (2012). Social attributes and associated performance
measures in marmots: bigger male bullies and weakly affiliating females have
higher annual reproductive success. Behavioral Ecology and Sociobiology, 66,
1075e1085.
Williams, C. T., Kitaysky, A. S., Kettle, A. B., & Buck, C. L. (2008). Corticosterone levels
of tufted puffins vary with breeding stage, body condition index, and repro-
ductive performance. General and Comparative Endocrinology, 158,29e35.
Young, A. J., Carlson, A. A., Monfort, S. L., Russell, A. F., Bennett, N. C., & Clutton-
Brock, T. (2006). Stress and the suppression of subordinate reproduction in
cooperatively breeding meerkats. Proceedings of the National Academy of Sci-
ences of the United States of America, 103, 12005e12010.
D. T. Blumstein et al. / Animal Behaviour 112 (2016) 1e11 11